Acknowledgements
We thank B. Diamond, J. Warner, M. Goodman and R. Laskov for critical review of the
manuscript, and A. Bothwell for providing the P1-5 hybridoma. This work was supported
by grants from the National Institutes of Health to P.D.B., to C.J.W. and to M.D.S., who is
also supported by the Harry Eagle chair provided by the National Women’s Division of the
Albert Einstein College of Medicine. A.M. is a recipient of Cancer Research Institute and
Harry Eagle fellowships.
Competing interests statement
The authors declare that they have no competing financial interests.
Correspondence and requests for materials should be addressed to M.D.S.
(e-mail: scharff@aecom.yu.edu). Accession numbers for mutated sequences of Ramos clones 6
and 7 are AF385858–AF385893, and those of Ramos clones A.2 and A.5, P1-5 clone A.1 and N114
clone A.3 are AF439565 –AF439624, and those of P1-5 clone A.2 are AF455739–AF455745.
..............................................................
Lateral relocation of auxin efflux
regulator PIN3 mediates tropism
in Arabidopsis
Jir
ˇ
ı
´
Friml*†, Justyna Wis
´
niewska*‡, Eva Benkova
´
*, Kurt Mendgen§
& Klaus Palme*k
* Max-Delbru
¨
ck-Laboratorium in der Max-Planck-Gesellschaft, 50829 Ko
¨
ln,
Germany
† Zentrum fu
¨
r Molekularbiologie der Pflanzen, Universita
¨
tTu
¨
bingen, 72076
Tu
¨
bingen, Germany
‡ Department of Biotechnology, Institute of General and Molecular Biology,
87–100 Torun, Poland
§ Lehrstuhl Phytopathologie, Universita
¨
t Konstanz, 78457 Konstanz, Germany
k Institut fu
¨
r Biologie II, Universita
¨
t Freiburg, 79104 Freiberg, Germany
.............................................................................................................................................................................
Long-standing models propose that plant growth responses to
light or gravity are mediated by asymmetric distribution of the
phytohormone auxin
1–3
. Physiological studies implicated a
specific transport system that relocates auxin laterally, thereby
effecting differential growth
4
; however, neither the molecular
components of this system nor the cellular mechanism of auxin
redistribution on light or gravity perception have been identified.
Here, we show that auxin accumulates asymmetrically during
differential growth in an efflux-dependent manner. Mutations in
the Arabidopsis gene PIN3, a regulator of auxin efflux, alter
differential growth. PIN3 is expressed in gravity-sensing tissues,
with PIN3 protein accumulating predominantly at the lateral cell
surface. PIN3 localizes to the plasma membrane and to vesicles
that cycle in an actin-dependent manner. In the root columella,
PIN3 is positioned symmetrically at the plasma membrane but
rapidly relocalizes laterally on gravity stimulation. Our data
indicate that PIN3 is a component of the lateral auxin transport
system regulating tropic growth. In addition, actin-dependent
relocalization of PIN3 in response to gravity provides a mech-
anism for redirecting auxin flux to trigger asymmetric growth.
Plants orientate their growth with respect to the direction of light
(phototropism) or gravity (gravitropism)
1
. As early as 1926 a widely
accepted model for plant tropisms, the Cholodny–Went hypothesis,
was presented
2
. It proposes differential distribution of the plant
hormone auxin in lateral direction on gravity or light stimulation.
Subsequently, different auxin levels elicit differential growth rates,
which ultimately lead to bending of the shoot or root
3
. Visualization
of asymmetrically distributed auxin response in gravistimulated
tobacco stems
5
and Arabidopsis roots
6
experimentally supported
this hypothesis. Polar auxin transport represent a plausible means of
lateral auxin distribution, as its chemical inhibition affects differ-
ential growth responses such as tropisms and apical hook for-
mation
7,8
. Physiologically characterized components of polar
auxin transport are cellular efflux carriers, whose polar localization
within cells is thought to determine the direction of auxin flux
9
. The
recently identified PIN genes of Arabidopsis appear to encode
essential components of these carriers
7
. A role of PIN2 in regulation
of basipetal auxin transport and gravitropism in root
6,10,11
as well as
a role of PIN1 in basipetal auxin transport in the stem have been
reported
12
; however, so far the molecular basis of shoot tropic
responses remains elusive. Lateral auxin transport with a specific,
laterally localized auxin efflux carrier was proposed
4
to explain the
exchange of auxin between vasculature, where the main basipetal
auxin stream occurs
13
, and peripheral tissues controlling
elongation
14
. Nevertheless the lack of any molecular data support-
ing this concept still leaves the existence of such a system in
question.
We analysed the relationship between auxin efflux, auxin redis-
tribution and differential growth responses in Arabidopsis hypo-
cotyls. Growth responses were examined and auxin levels indirectly
visualized using the synthetic DR5::GUS auxin reporter
15
, whose
activity correlates with direct auxin measurements
16,17
. Seedlings
grown under standard conditions displayed normal phototropic
and gravitropic curvature accompanied by an asymmetric
DR5::GUS expression suggesting elevated auxin levels on the more
elongated side of the hypocotyl (Fig. 1a, c). In contrast, seedlings
that did not undergo light or gravity stimulation showed lower and
uniform expression of the DR5::GUS reporter (data not shown).
Seedlings subjected to light and gravity stimulation with simul-
taneous auxin efflux inhibition failed to show any differential
DR5::GUS expression as well as tropic curvature (Fig. 1b, d).
Differential DR5::GUS expression was also detected during the
process of apical hook formation (Fig. 1e), and seedlings grown
on auxin efflux inhibitors failed to show asymmetric auxin distri-
bution and apical hook formation (data not shown). These results
show that asymmetric growth of the shoot correlates with auxin
efflux-dependent asymmetric auxin distribution, and implicate the
existence of auxin efflux components involved in asymmetric
growth responses.
The auxin efflux carrier candidates are PIN proteins, although
their function in the regulation of differential growth has not been
reported
7,8
. We screened for new members of the PIN gene family
and analysed corresponding knockout mutants for differential
growth defects. One of the genes, PIN3, was isolated from genomic
and complementary DNA libraries with probes derived from the
conserved region of PIN1. The deduced PIN3 protein shares 67%
Figure 1 Auxin response during differential hypocotyl growth. a –d, Expression of the
DR5::GUS reporter in hypocotyl of untreated (a, c) or auxin efflux inhibitor (NPA)-treated
(b, d) wild-type seedlings upon stimulation by light (a, b) or gravity (c, d). Insets show
details of DR5::GUS expression. Scale bars, 400
m
m. e, Asymmetric expression of
DR5::GUS reporter in the apical hook of an etiolated seedling. Scale bar, 50
m
m.
letters to nature
NATURE | VOL 415 | 14 FEBRUARY 2002 | www.nature.com806
© 2002 Macmillan Magazines Ltd
First publ. in: Nature 415 (2002), pp.806-809
Konstanzer Online-Publikations-System (KOPS)
URL: http://www.ub.uni-konstanz.de/kops/volltexte/2007/3737/
URN: http://nbn-resolving.de/urn:nbn:de:bsz:352-opus-37379
identity to PIN1, displays the characteristic three-domain topology
with about 10 transmembrane domains characteristic of all PIN
proteins
7,8
, and shows similarity to bacterial transporters and
detoxification carriers, suggesting the same biochemical function
for PIN proteins in auxin efflux
7,8
. By a reverse genetic strategy
18
two mutant alleles designated as pin3-1 and pin3-2 were identified
(Fig. 2a). From the progeny of pin3-2 a stable, footprint mutant
allele exhibiting a deletion of two base pairs, designated pin3-3, was
isolated. PIN3 expression (data not shown) and immunolocaliza-
tion (Fig. 4e) studies demonstrated that all three alleles were null.
pin3 mutants display defects in differential growth. Gravitropic
and phototropic responses as well as apical hook maintenance are
reduced in pin3 mutants (Fig. 2b–d and Table 1). pin3 seedlings also
display shorter hypocotyls and roots than wild-type seedlings when
grown in light (Fig. 2f and Table 1). However, no such differences
were observed in dark-grown seedlings (Fig. 2e and Table 1).
Shorter hypocotyl epidermis cells in pin3 suggested that this
phenotype was caused by a defect in cell elongation (Fig. 2g, h
and Table 1). The specificity of the pin3 phenotype was further
confirmed by identification of two revertants displaying wild-type
phenotype among homozygous pin3-2 seedlings. Sequencing of the
PIN3 locus in both plants confirmed the excision of the En-1
transposon and restoration of the wild-type PIN3 coding sequence.
All defects in pin3 mutants can be mimicked by growing wild-type
plants on auxin efflux inhibitors (Fig. 2i and Table 1)
7,19,20
, further
supporting a role for PIN3 in auxin efflux. The pin3 mutation does
not abolish differential growth to the same extent that auxin efflux
inhibitors at increased concentrations do. This suggests some level
of functional redundancy between PIN3 and other PIN protein(s).
Northern blot analysis revealed the presence of PIN3 transcripts
Figure 2 pin3 mutants. a, Exon/intron organization of PIN3 on chromosome 1 showing
positions of En-1 elements in pin3-1 and -2 alleles. b, c, pin3 hypocotyls are defective in
gravitropic (b) as well as phototropic (c) responses. Scale bars, 500
m
m. d, pin3 mutants
are defective in root gravitropism. Each gravistimulated root was assigned to one of twelve
308 sectors. The length of each bar represents the percentage of seedlings showing
direction of root growth within that sector. e, f, Dark-grown pin3 seedlings (e) have the
same length of hypocotyls and roots as wild type, but display defects in apical hook
maintenance. Light-grown pin3 seedlings (f) display shorter hypocotyls and roots
compared with wild type. Scale bars, 500
m
m. g–i, Hypocotyl epidermis cells of pin3 (h)
and NPA-grown wild-type (i) seedlings show reduced length compared with untreated
wild type (g). Arrowheads indicate cell boundaries. Scale bars, 10
m
m. The following pin3
alleles are depicted: pin3-1 (b), pin3-2 (c, d) and pin3-3 (e, f, h).
Figure 3 Expression of the PIN3 gene. a – d, PIN3::GUS transgenic seedlings show GUS
staining in the apical hook (b), around the hypocotyl vasculature (c), and in the root
pericycle (arrow) and columella (arrowhead) (d). Scale bars, 50
m
m. e –g, PIN3 mRNA
was detected by in situ hybridization in hypocotyl (e), stem (f) starch sheath (arrows) and
root columella (arrowheads) (g) cells. Insets show sense controls (e–g). Scale bars:
200
m
m(e), 100
m
m(f) and 20
m
m(g). Longitudinal sections (e, f) and whole-mount
preparations (a – d, g) are shown.
letters to nature
NATURE | VOL 415 | 14 FEBRUARY 2002 | www.nature.com 807
© 2002 Macmillan Magazines Ltd
predominantly in the shoot and root (data not shown). The staining
of transgenic seedlings carrying the PIN3 promoter fused to the
b
-
glucuronidase (GUS) gene revealed PIN3::GUS expression associ-
ated with vasculature (Fig. 3a). Detailed inspection revealed GUS
staining in the apical hook of the etiolated seedling (Fig. 3b), in the
shoot endodermis (starch sheath) around the vasculature (Fig. 3c),
and in the root pericycle and columella (Fig. 3d). To localize the
PIN3 protein, PIN3-specific antibodies were raised and immuno-
localization experiments performed. The PIN3 protein was found at
the periphery of shoot starch sheath as well as root pericycle and
columella cells (Fig. 4a–e). Longitudinal sections revealed that
PIN3 is localized in a polar way predominantly at the lateral,
inner side of the starch sheath as well as root pericycle cells (Fig.
4c, d). In a minor portion of the cells an additional basal or uniform
localization of PIN3 was observed (data not shown). In contrast,
only uniform distribution of PIN3 in root columella cells was
detected (Fig. 4e). No PIN3 signal was detected in pin3 knockout
mutants (Fig. 4e, inset). PIN3 messenger RNA localization using in
situ hybridization confirms the GUS and immunolocalization
staining patterns (Fig. 3e–g).
Anti-PIN3 immunogold labelling revealed signals at the plasma
membrane (Fig. 5a) and frequently in vesicles of about 70 nm
diameter, similar in size to exocytotic vesicles (Fig. 5b), suggesting
a dynamic subcellular movement of PIN3. The PIN3 protein
appears to cycle rapidly between the plasma membrane and unde-
fined endosomal compartments, since the incubation of seedlings
with the exocytosis inhibitor Brefeldin A (BFA) led to internaliz-
ation of PIN3 (Fig. 5d, compare with c), also in the presence of the
protein synthesis inhibitor cycloheximide (Fig. 5f, compare with e).
Pericycle cells show an accumulation of the PIN3 label in peri-
nuclear compartments (Fig. 5d), very similar to previously
described BFA compartments
21
. In contrast, in columella cells
PIN3 internalizes in smaller compartments without any regular
Table 1 Quantitative effects of pin3 mutations and auxin efflux inhibition
Measurement Col-0 (n) pin3 (n) NPA* (n)
...................................................................................................................................................................................................................................................................................................................................................................
Hypocotyl length in light (mm) 2.21 ^ 0.16 (58) 1.68 ^ 0.15 (66) 1.43 ^ 0.14 (56)
Root length in light (cm) 2.41 ^ 0.22 (58) 1.70 ^ 0.16 (66) 1.67 ^ 0.16 (56)
Hypocotyl length in dark (cm) 1.32 ^ 0.18 (54) 1.35 ^ 0.12 (62) 1.41 ^ 0.15 (62)
Root length in dark (cm) 0.77 ^ 0.07 (54) 0.78 ^ 0.14 (62) 0.85 ^ 0.11 (62)
Hypocotyl cell length (
m
m) 101.8 ^ 10.2 (63) 75.5 ^ 8.40 (54) 72.2 ^ 7.90 (61)
Hypocotyl gravitropism (deg) 78.1 ^ 16.9 (78) 31.8 ^ 15.0 (81) – –
Hypocotyl phototropism (deg) 72.5 ^ 15.5 (65) 40.0 ^ 9.70 (72) – –
Opened apical hooks (%) 16.2 ^ 9.67 (81) 65.4 ^ 8.02 (97) – –
...................................................................................................................................................................................................................................................................................................................................................................
Wild-type plants treated with NPA.
Figure 4 Localization of the PIN3 protein. a, b, PIN3 protein localizes to the surface of
hypocotyl (a) and stem (b) starch sheath cells. Scale bars, 50
m
m(a) and 100
m
m(b).
c, d, Longitudinal view of stem starch sheath (c) and root pericycle (d) cells shows lateral
localization of PIN3 at the inner cell surface. Scale bars: 10
m
m(c) and 5
m
m(d). e, In the
root tip, PIN3 is found uniformly in the columella cell boundaries. Inset shows no PIN3
staining in the pin3 mutant. Scale bar, 20
m
m. Transversal (a, b) and longitudinal (c)
sections, and whole-mount (d, e) preparations are shown. The PIN3 protein signal is
visualized by green coloration.
Figure 5 Subcellular localization, actin-dependent cycling and gravity-dependent
asymmetric relocation of the PIN3 protein. a, b, Immunogold electron microscopy shows
PIN3 at the plasma membrane (a) and in vesicles (b). cw, cell wall; mv, membrane vesicle.
Scale bars, 50 nm. c–g, Actin-dependent cycling of PIN3. Localization of PIN3 in
columella (e –g) and pericycle (c, d) cells. Internalization of PIN3 after BFA (d),
cycloheximide/BFA (f) and latrunculin B (g) treatments. Untreated controls (c, e) are
shown. Perinuclear compartments are indicated by arrows in d. Scale bars, 5
m
m(c, d)
and 10
m
m(e–g). h–j, Relocation of PIN3 in columella cells after a change in the gravity
vector after 2 min (h) and 10 min (i). After 1 h the PIN3 localization domain expands in
lateral root cap (indicated by arrows) and in columella initials (j). Scale bars, 10
m
m. The
apparent gravity vectors are indicated by arrowheads.
letters to nature
NATURE | VOL 415 | 14 FEBRUARY 2002 | www.nature.com808
© 2002 Macmillan Magazines Ltd
positioning (Fig. 5f). A similar pattern of PIN3 localization was
observed when the actin cytoskeleton was disrupted by latrunculin
B (Fig. 5g) or cytochalasin D (data not shown) treatment,
suggesting an actin-dependent cycling of PIN3. A similar cycling
pattern has been demonstrated for the homologous PIN1 protein;
however, the biological meaning of this process remains unclear
21
.
PIN3, in contrast to PIN1, was frequently detected in vesicles,
suggesting that PIN3 cycles more rapidly or that the equilibrium
of intracellular PIN3 pools is shifted more in favour of the inter-
nalized PIN3 pool.
The rapid PIN3 cycling in gravity-sensing cells (starch sheath,
columella)
22
raised the possibility that this cycling may provide the
mechanism for rapid relocalization of PIN3 after perception of an
environmental stimulus. We examined the localization of the PIN3
protein in columella cells after a change in the gravity vector.
Already at 2 min after root rearrangement, the originally symmetric
PIN3 distribution (Fig. 5e) becomes asymmetric, with most of the
PIN3 protein at the lateral side of columella cells (Fig. 5h). After
5 min the relocalization of PIN3 protein appeared to be complete
(Fig. 5i), and was maintained up to 20 min afterwards (data not
shown). One hour after gravity stimulation the asymmetry of the
PIN3 localization was observed at the tissue level. PIN3 protein was
detected in an enlarged, asymmetric localization domain including
columella initial and lateral root cap cells (Fig. 5j, compare with e
and Fig. 4e).
Our findings provide the molecular evidence in support of the
classical Cholodny–Went model for tropic growth
2
. We show that
asymmetric growth correlates with auxin efflux-dependent asym-
metric auxin distribution and requires laterally localized PIN3
protein. Developmental requirements for PIN3, its localization, as
well as the proposed biochemical function for PIN proteins make
PIN3 a likely candidate for an efflux component of the lateral auxin
transport system. The actin-dependent cycling and rapid relocaliza-
tion of PIN3 in cells containing gravity-sensing statoliths
22
, taken
together with previous findings supporting a direct role of the actin
cytoskeleton in gravity perception
23,24
, imply a cellular mechanism
for the regulation of asymmetric auxin distribution and differential
growth. After a change of the gravity vector, actin-enmeshed
statoliths
25
sediment, causing reorganization of the actin cytoske-
leton. Thus PIN3 is redirected towards one side of the columella
cells and determines direction of auxin flux, which leads to asym-
metric auxin accumulation and differential growth. It is possible
that tropic responses of shoots are regulated by a similar mechan-
ism, as the loss of PIN3 or the PIN3-containing starch sheath
correlate with the shoot agravitropic phenotype
26
. A
Methods
Materials used
The PIN3 gene was identified in a bacterial artificial chromosome genomic library (http://
www.mpimp-golm.mpg.de/mpi-mp-map/) with PIN1 probe (nucleotides 1–385;
GenBank accession number AF089084). The full-length PIN3 cDNA was isolated from a
stem cDNA library (GenBank accession number AF087818). The PIN3::GUS construct
was generated by fusion of a PCR-amplified fragment (nucleotides -1,764 to -1) upstream
of the ATG codon and the GUS gene. We used the following probes and primers:
nucleotides 999–1449 of the PIN3 cDNA, 5
0
-TCCTCTCACTTCTTCTTCTTCCTC-3
0
;
5
0
-TTTATTT ATCTTTTCTTTGTCTCG-3
0
.
Phenotype analyses
Seedlings were grown as described previously
10
. For hypocotyl bending, 3-day-old
etiolated DR5::GUS, Col-0 (Columbia ecotype) and pin3 seedlings were transferred and
orientated on vertical plates with or without 10
m
M1-N-naphthylphthalmic acid (NPA),
immediately subjected to the gravity or unilateral light stimulation for 20 h stained for
GUS activity, and photographed. The root gravitropism was scored in the dark in 3-day-
old, light-grown seedlings, 6 h after turning the roots 1358. The quantification of responses
was performed with Adobe Illustrator software. Apical hook opening was scored 50 h after
germination in vertically grown, etiolated seedlings. The described defects were observed
in all alleles examined (pin3-1, -2 and -3).
Expression and localization analysis
Histochemical staining for GUS activity, northern blot analysis, in situ hybridization,
immunolocalization and immunogold electron microscopy were performed as
described
10,12
. PIN3-specific antibodies were generated using a recombinant protein
corresponding to amino acids 334–483. Affinity-purified primary anti-PIN3, fluorescein
isothiocyanate-conjugated and CY3-conjugated anti-rabbit secondary antibodies
(Dianova) were diluted 1:40, 1:200 and 1:600, respectively. Brefeldin A (Molecular Probes)
and latrunculin B (Duchefa) treatments were performed with corresponding controls
before immunolocalization as described
21
. For the gravitropism treatment, before
immunolocalizations, plates with vertically grown seedlings were placed horizontally, and
samples were fixed directly on plates at specific time points.
Received 17 October; accepted 10 December 2001.
1. Darwin, C. & Darwin, F. in Darwins Gesammelte Werke Bd. 13 (Schweizer-bart’sche
Verlagsbuchhandlung, Stuttgart, 1881).
2. Went, F. W. Reflections and speculations. Annu. Rev. Plant Physiol. 25, 1–26 (1974).
3. Hart, J. W. Plant Tropism and Other Movements (Unwin Hyman, London, 1990).
4. Epel, B. L., Warmbrodt, R. P. & Bandurski, R. S. Studies on the longitudinal and lateral transport of
IAA in the shoots of etiolated corn seedlings. J. Plant Physiol. 140, 310 –318 (1992).
5. Li, Y., Hagen, G. & Guilfoyle, T. J. An auxin-responsive promoter is differentially induced by auxin
gradients during tropisms. Plant Cell 3, 116 –1176 (1991).
6. Luschnig, C., Gaxiola, R. A., Grisafi, P. & Fink, G. R. EIR1, a root specific protein involved in auxin
transport, is required for gravitropism in Arabidopsis thaliana. Genes Dev. 12, 2175 –2187 (1998).
7. Friml, J. & Palme, K. Polar auxin transport
—
old questions and new concepts. Plant Mol. Biol. (in the
press).
8. Palme, K. & Ga
¨
lweiler, L. PIN-pointing the molecular basis of auxin transport. Curr. Opin. Plant Biol.
2, 375–381 (1999).
9. Rubery, P. H. & Sheldrake, A. R. Carrier-mediated auxin transport. Planta 118, 101–121 (1974).
10. Mu
¨
ller, A. et al. AtPIN2 defines a locus of Arabidopsis for root gravitropism control. EMBO J. 17,
6903–6911 (1998).
11. Rashotte, A. et al. Basipetal auxin transport is required for gravitropism in roots of Arabidopsis. Plant
Physiol. 122, 481–490 (2000).
12. Ga
¨
lweiler, L. et al. Regulation of polar auxin transport by AtPIN1 in Arabidopsis vascular tissue.
Science 282, 2226–2230 (1998).
13. Morris, D. A. & Thomas, A. A microautoradiographic study of auxin transport in the stem of intact
pea seedlings (Pisum sativum L.) J. Exp. Bot. 29, 147– 157 (1978).
14. Bjorkman, T. & Cleland, R. E. The role of the epidermis and cortex in gravitropic curvature of maize
roots. Planta 176, 513– 518 (1988).
15. Ulmasov, T., Murfett, J., Hagen, G. & Guilfoyle, T. J. Aux/1AA proteins repress expression of reporter
genes containing natural and highly active synthetic auxin response elements. Plant Cell 9, 1963–1971
(1997).
16. Casimiro, I. et al. Auxin transport promotes Arabidopsis lateral root initiation. Plant Cell 13, 843–852
(2001).
17. Friml, J. et al. AtPIN4 mediates sink driven auxin gradients and patterning in Arabidopsis roots. Cell
(in the press).
18. Wisman, E., Cardon, G. H., Fransz, P. & Saedler, H. The behaviour of the autonomous maize
transposable element En/Spm in Arabidopsis thaliana allows efficient mutagenesis. Plant Mol. Biol. 37,
989–999 (1998).
19. Ruegger, M. et al. Reduced naphthylphthalamic acid binding in the tir3 mutant of Arabidopsis is
associated with a reduction in polar auxin transport and diverse morphological defects. Plant Cell 9,
745–757 (1997).
20. Jensen, P. J., Hangarter, R. P. & Estelle, M. Auxin transport is required for hypocotyl elongation in
light-grown Arabidopsis. Plant Physiol. 16, 455 –462 (1998).
21. Geldner, N. et al. Polar auxin transport inhibitors block PIN1 cycling and vesicle trafficking. Nature
413, 425–428 (2001).
22. Hasenstein, K. H. Gravisensing in plants and fungi. Adv. Space Res. 24, 677–685 (1999).
23. Sievers, A., Sondag, C., Trebacz, K. & Hejnowicz, Z. Gravity induced changes in intracellular
potentials in statocytes of cress roots. Planta 197, 392–394 (1995).
24. Monshausen, G. B., Zieschang, H. E. & Sievers, A. Differential proton secretion in the apical
elongation zone caused by gravistimulation is induced by a signal from the root cap. Plant Cell
Environ. 19, 1408– 1414 (1996).
25. Collings, D. A., Zsuppan, G., Allen, N. S. & Blancaflor, E. B. Demonstration of prominent actin
filaments in the root columella. Planta 212, 392–403 (2001).
26. Fukaki, H. et al. Genetic evidence that the endodermis is essential for shoot gravitropism in
Arabidopsis thaliana. Plant J. 14, 425– 430 (1998).
Acknowledgements
We thank G. Ju
¨
rgens for enabling J.F. to accomplish part of this work in his laboratory;
P. Ta
¨
nzler and M. Sauer for technical assistance; H. Vahlenkamp for technical assistance in
immunocytochemistry; M. Estelle for providing material and suggestions; T. Altman for
BAC filter sets; the ADIS (Automated DNA Isolation and Sequencing) service group for
DNA sequencing; ZIGIA (Center for Functional Genomics in Arabidopsis) for the En lines;
and N. Geldner, T. Hamann, G. Ju
¨
rgens, K. Schrick and C. Schwechheimer for comments
and critical reading of the manuscript. This work was supported by a fellowship of the
DAAD (J.F.), the DFG (Schwerpunktprogramm Phytohormone), the Fonds der
chemischen Industrie, the European Communities Biotechnology Programs, the INCO-
Copernicus Program and the European Space Agency MAP-Biotechnology Programme.
Competing interests statement
The authors declare that they have no competing financial interests.
Correspondence and requests for materials should be addressed to J.F.
(e-mail: jiri.friml@zmbp.uni-tuebingen.de) or K.P. (e-mail: palme@mpiz-koeln.mpg.de).
letters to nature
NATURE | VOL 415 | 14 FEBRUARY 2002 | www.nature.com 809
© 2002 Macmillan Magazines Ltd
