Directory UMM :Data Elmu:jurnal:P:PlantScience:PlantScience_Elsevier:Vol158.Issue1-2.Sept2000:

(1)

A new

Arabidopsis thaliana

root gravitropism and chirality mutant

S. Ferrari, S. Piconese, G. Tronelli, F. Migliaccio*

Consiglio Nazionale delle Ricerche,Institute of Plant Biochemistry and Ecophysiology,Via Salaria Km29.300,00016Monterotondo,

Rome,Italy

Received 23 March 2000; received in revised form 25 May 2000; accepted 2 June 2000

Abstract

A new Arabidopsis thaliana (L.) Heynh. root mutant, named clg1, was isolated from the Feldmann – Du Pont T-DNA insertional mutagenesis collection. It is characterized by primary roots that make early right-handed coils, show increased right-handed slanting, and reduced positive gravitropism. The mutant roots however did not exhibit increased resistance to IAA, but only a modest increment of resistance to the auxins 2,4 D, and NAA, and to the auxin transport inhibitors TIBA and NPA. By contrast, the mutant roots showed a notable resistance to plant hormone ethylene (given as ACC). clg1 appears to be new, since it complements the most known auxin and gravitropism mutants, maps to chromosome 5, and shows a phenotype largely different from that of the known ethylene mutants. The increased right-handed slanting (chirality) can possibly be a consequence of the reduced gravitropic response, since gravitropism and slanting are competitive growth-direction leading forces. The increased resistance to ethylene, seems to indicate that this phythormone plays a role in the gravitropic response of roots (as already proposed for shoots), and possibly in the regulation of the connected signal transduction pathway. The gene involved in theclg1 mutation, which is recessive, was mapped, as above reported, to chromosome 5, close to the visible markertt3. © 2000 Elsevier Science Ireland Ltd. All rights reserved.

Keywords:Arabidopsis; Roots; Gravitropism; Right-hand slanting

www.elsevier.com/locate/plantsci

1. Introduction

As is the case for many arguments in plant biology, in the last few years the research on root growth and development was notably fos-tered and renewed by the use of the model sys-tem Arabidopsis. In particular, it has recently been discovered that the roots of this plant do not grow straight along the gravitational vector, but instead follow a complex pattern that seems to be the result of forces of a different nature. These are essentially circumnutation, positive gravitropism, and thigmotropism [1 – 5].

Many scientists contributed to this research: Maher and Martindale [6] and Mirza [7] first

reported the coiling habit of Arabidopsis roots; Okada and Shimura [2,8] showed the waving habit of the roots down tilted hard-agar plates; Simmons et al. [3,4] characterized the slanting and the right-handedness of the roots; Ruther-ford and Masson [9] elucidated many fine as-pects of the process.

Fundamental to this research was the recur-rent discovery of interesting mutants, the first of which, aux1, showing resistance to auxin and no gravitropic response, was described by Maher and Martindale [6]. It was followed by the de-scription by Estelle and Somerville [10], and Lin-coln et al. [11] by a full set of new mutants, the most significant of which are axr1, axr2, and

axr3, showing increased root resistance to auxin and reduced gravitropism. The description of these mutants was followed by the isolation of

agr1 by Bell and Maher [12], also characterized

* Corresponding author.Tel.: +39-06-90672530; fax: + 39-06-9064492.

E-mail address:[email protected] (F. Migliaccio).

(2)

by reduced root gravitropism, but not by resis-tance to auxin, and by the mutants named wa6

by Okada and Shimura [2,8], because of the anomalies shown in the root waving pattern. Later wa65 was determined to be allelic with

aux1, and wa66 with agr1, eir1, and pin2; auxin

and ethylene mutants. Simmons et al. [3,4] there-after isolated rgr1, also agravitropic in the roots, which was later shown to be allelic with axr4 [13]. Rutherford and Masson [9] isolated

sku1,2,3,5. When its roots were grown down a tilted hard-agar surface, they showed modifica-tions in the wavy pattern, and more recently, Marinelli et al. [14] isolated 1 – 6C, a mutant that shows inverted (left-handed) root slanting. In the last few years some of the above genes were also cloned, and resulted in the encoding of elements involved in the transport and metabolism of auxin [15 – 20]. The ensemble of the data gath-ered in these years, led at the same time to the formulation of different theories regarding the root growth patterns, which are somewhat dis-cordant. In particular, Okada and Shimura [2] hypothesized that the wavy pattern was the re-sult of the cumulative effects of positive gravit-ropism and thigmotgravit-ropism, the root tip moving from side to side on an inclined plate after the impact with the agar surface. Simmons et al. [3], and Migliaccio et al. [21] saw instead the process as the resultant of circumnutation and positive gravitropism, connecting the waving with the well know tendency of roots of many plants to grow circumnutating to one hand [1,5,22]. Rutherford and Masson [9] substantially backed the Okada and Shimura hypothesis, even ac-knowledging that circumnutation should be in-volved in the process.

The nature of the root pattern seems therefore still unclear, and more research, mutant analysis, and the cloning of the involved genes are surely necessary. With this in mind we started our search for new Arabidopsis mutations of the root growth pattern, that led us to the isolation of

clg1, a mutant that does not appear to have been already described, that shows reduced grav-itropism, and probably, as a consequence, high tendency to produce coils and increased slanting to the right-hand. This paper is about the char-acterization of this mutant, that we hope can become a new tool to investigate growth and tropic movements in Arabidopsis roots.

2. Materials and methods

2.1. Plant material and growth conditions

The T-DNA mutagenized seeds (collection Feldmann – Du Pont), as well as the mutants

aux1-7, axr1-3, axr2 were provided by the Not-tingham Arabidopsis Stock Center (Nottingham, UK). Seeds of A. thaliana (L.) Heynh., ecotype Wassilevskjia, as well as the rgr1 and wa66

mu-tants were obtained from D. So¨ll laboratory (Yale University, New Haven CT, USA).

Before plating, the seeds were sterilized by treating them for ten min with a solution made up of 50% commercial bleach (brand name Clorex) and 0.01% (w/v) SDS, followed by four washes with sterile distilled water. The seeds were then plated in horizontal rows, in petri dishes, on a medium made up (except in some cases where 0.8 and 1% agar were used) of 1.5% agar, 1% sucrose, and 0.5×MS basal medium, enriched with Gamborg’s vitamins (from Sigma, n. M0404), adjusted to pH 5.7 with KOH. To synchronize germination, dishes were left in a cold room (4°C) for 2 days before moving them to the growth room.

Plants were at first grown on dishes for 10 – 14 days in a growth room (in white fluorescent light at 80 mmol m−2 s−1 and 25°C), and then

trans-ferred in Arasystem pots (5 cm) on universal soil (Einheitserde-Humuswerke, type VM; Gebr. Patzer, Sinntal-Jossa, Germany), in a Conviron climatic chamber, in white fluorescent light at 200 mmol m−2 s−1. The plants were grown

keeping them at first in short days (8/16 h light – dark periods) to promote rosette growth, and, after 3 weeks, in long days (16/8 h light – dark periods) to promote flowering. Humidity in the Conviron was 65%, temperature 23°C. The pots were watered three times a week: two with tap water, and one with a nutrient solution (0.5 mg/l Mannalin-A). At the time of seed maturation the plants were moved to a greenhouse, where they were kept until seed collection.

2.2. Mutant selection

Of the 49 pools of 100 lines that constitute the Feldmann collection, 48, for a total of 20 758 single seedlings, were submitted to the first

(3)

screening. This consisted of the visual examination of the growth direction (slanting and gravit-ropism) and of the general behavior of the pri-mary root, looking for modification of the above characters.

Seeds were germinated and grown in petri dishes, on agar medium containing kanamycin (50 mg/l). The dishes were kept vertical in the light for 3 days, then part of them were rotated 90° with respect to the vertical to check for the gravitropic response, and part were inclined 60° on the hori-zontal plane to check for slanting. The dishes thereafter were kept in the above settings for about 1 week. Plants resistant to kanamycin, showing an anomalous root growth pattern, were selected, grown to maturity and selfed. The second generation was then submitted to a new screening on a medium deprived of kanamycin, together with the wild-type as a control. As a result of the screening we made, a few interesting root mutants were isolated, of which we started to study the most promising one. This was namedclg1 for the characteristic of making very early coils, even on vertically oriented agar plates, and for showing increased right-handed slanting.

2.3. Gra6itropism and slanting determination

To quantify the gravitropic response of the roots, clg1 seeds were grown in square dishes between two 1% agar layers to avoid the produc-tion of slanting, which can interfere with gravit-ropism. The dishes were kept vertical throughout the experiments, for 3 days in the light, then moved to dark for 7 days, and rotated by 90° in a clockwise direction. The angle made by the roots with the first direction of growth was measured at 2 h intervals over an 8 h period, and after 24 h of gravistimulation. Angle determination was made by a protractor on line marks traced on the bot-tom of the dishes.

Root slanting was measured similarly by a pro-tractor, making reference to a vertical line parallel to the gravitational vector traced on the bottom of the plates. Dishes were kept vertical for 3 days, and then inclined 60° over 7 additional days, in light conditions.

Root elongation was also evaluated in light conditions, after the 3rd day from germination, by measuring under a stereomicroscope, the root growth over a 5 day period.

2.4. Response to hormones

Seeds were germinated and grown for 5 days in the light, in vertically set petri dishes. Thereafter, same size seedlings were transferred to a new medium containing the hormones, and grown for 7 days in the dark. Root elongation was measured over the last 7 days, and from the measurements the growth relative to controls was evaluated. Data are mean values of at least 10 samples9SE. Hormonal substances, e.g. IAA (indole-3-acetic acid), 2,4D (2,4 dicloro phenoxyacetic acid), NAA (naphtalene acetic acid), TIBA (2,3,5 triiodoben-zoic acid), NPA (N-(1-naphtyl) phtalamic acid)), were administered by dissolving them in ethyl alcohol, and by dispersing a minimal amount of the substance in the media (alcohol less than 0.1%, w/v). Ethylene was administered as ACC (1-aminocyclopropane-1-carboxylic acid) dissolved in water.

The response of shoots to ethylene and auxins (hook formation) was measured on plants germi-nated and grown totally in the dark on 0.8% agar, visually determining the hook shape and assigning it to one of three classes: regular, exaggerated, or absent.

The environment temperature was kept in all the above experiments constantly at 25°C.

2.5. Genetic analyses

Crosses were made on the apical part of a flowering stalk carrying the immature buds follow-ing standard techniques [4].

All data were submitted to the x2, or the Stu-dent’s t-tests, to check for confidence with the expected segregation ratio.

Pictures were taken by means of a Nikon cam-era on Ektachrome 160 T Kodak film.

3. Results

3.1. Genetics of clg1

clg1 was selected through a new screening of the 4900 T-taggedArabidopsislines of the Feldmann – Du Pont collection as reported in the previous section. This mutant is characterized by a ten-dency to make early right-handed coils, by in-creased right-handed slanting, and by slow

(4)

Fig. 1. Growth habit ofclg1 primary roots compared with wild-type Ws (ecotype Wassilevskjia). Plants grown on hard-agar plates (1.5%), set 60° inclined with respect to the horizontal plane.

Table 1

Chromosome mapping ofclg1a,b

F2 total Marker phenotype of F2clg1 individuals

er1 gl-1 cer2-2 tt3-1 Marker phenotype clg1 ch1-1

1 2 3 4

Chromosome location 5 5

110

110 110

110

605 110

Number screened

22 30 21 0

No _{marker phenotype} ₁₁₀ ₂₀

20.0 27.2 19.0 0

% 18.1 18.1

2.73 1.47 0.31 2.05

X2_{per 3:1 ratio} _36.66

a_{A chromosome mapping line with visible markers} _ch_1, _er_1, _gl_1, _cer_2, _tt_{3 (ABRC seed stock CS3078) was crossed to} _clg₁

(pollen donor) and grown to generate the F2 progeny for mapping.

b_{Low values result from deliberate conservative}_clg_{1 scoring.}

gravitropic response (Fig. 1). The first two charac-teristics, in contrast with the wild type, are visible not only on inclined, but frequently also on verti-cally oriented hard-agar plates. This set of root characteristics persisted after three generations produced through selfing and two crossings with the wild type, indicating the monogenetic nature of the traits. Moreover, the cross with the wild-type showed, through the suppression of the muta-tion in the F1, and the segregamuta-tion in the F2, the recessivity of the characters, and the Mendelian transmission of them. The mutation did not show however, cosegregation with the resistance to kanamycin in the F2 population originated from the kanamycin resistant F1 individuals. A segrega-tion consistent with a ratio 9:3:3:1 was seen. Con-sequently, the involved gene did not result tagged.

Tests of allelism were made with most known and available mutants1 _{Through crosses with}

aux1 – 7,axr1 – 3, axr2,wa66 (agr1,eir1,pin2),rgr1

(axr4), allelism was primarily excluded on the basis of the complementation seen in the F1 or F2. On the other hand, the map position of clg1 excluded also allelism with the above mutants, with the exception of agr1, that maps on chromo-some 5. The complementation test we run however contradicts this possibility.

1_clg₁_{, as reported, was complemented with the most common and} available mutants that show a modified response to gravitropism or auxin concentration. In spite of this, we cannot exclude that it could be allelic with one of the many already described mutants. We did not think, however, it is useful to furthermore extend complementation tests that are time consuming and not always productive.

(5)

The mapping of clg1 was made by crossing it with the visible markers mapping line CS 3078 (originally DP 23, Patton and Meinke line), from the ABRC of Ohio State University. As reported in Table 1, the mutation showed linkage with the marker tt3, and thus it is located on chro-mosome 5, close to that marker. No recombi-nants were recovered among the 110 F2 clg1 individuals of the reported screen.

A control of other root characteristics, which frequently appear modified in similar mutants

Fig. 3. clg1 and wild-type Ws root slanting from seedlings grown on 1.5% agar plates, 60° inclined with respect to the horizontal plane. Values represent the distribution of mea-sured root angles of clg1 and wild-type for each of the 15° sectors showed;n=33 for both genotypes. Total mean values (9SE) are 22.992.2° for wild-type, and 48.295.7° forclg1.

Fig. 2. ((A) Gravitropic curvature of clg1 and wild-type Ws roots. Seedlings grown at first for three days in the light, and then for seven days in the dark, in vertically set petri dishes, between two agar layers. Dishes 90° reoriented in the clock-wise direction at the beginning of the dark incubation. Values represent mean curvatures9SE ((n=16 for wt, and 21 for

clg1). (B) Elongation of clg1 and wild-type Ws roots from seedlings grown on 1.5% agar medium in the light. Primary root length was measured every 24 h from the third day after germination. Values represent mean root length9SE. (n=16 for wt and 21 forclg1). In this graph, only the value relative to 72 h is slightly significant at the Student’s t-test, for

PB0.05.

such as axr1, rgr1, showed that clg1 roots are wild-type as concerns the number of lateral roots, root hairs, and the number of statolyts in the root cap cells.

3.2. Gra6itropism, slanting and elongation

The gravitropic response of clg1, reported in Fig. 2 (top), was clearly reduced, since about 30° of difference were observed with respect to the wild-type after an eight hour period. After 24 h, however, both the wild type and the mutant were for the most part 90° bent (data not shown). Therefore, in the mutant the gravitropic response appears only slowed down, but not re-duced in absolute values.

The slanting to the right-hand, reported in Fig. 3, was increased in clg1, even though a widened distribution was observed with respect to the wild-type. The absolute mean values were 22.9° in wild-type and 48.2° in the mutant.

Root elongation (Fig. 2, bottom) was only slightly increased with respect to the wild-type

(6)

Fig. 4.clg1 and wild-type Ws root elongation from seedlings grown in presence of different concentration of the auxins 2,4 D, IAA, NAA and of the ethylene precursor ACC. Elongation expressed relative to the average root growth of the same genotype on a medium without the added substances. Each value represents the mean of measurements of at least ten seedlings9SE. Absolute mean values (9SE) of root elongation on hormone free media were: 2,4 D plot, wild-type=15.6891.40 mm,

clg1=12.7791.10 mm; IAA plot, wild-type=7.2590.26,clg1=12.7590.77; NAA plot, wild-type=2.0390.06,clg1=2.259

0.08; ACC plot, wild-type=15.1590.96 mm; clg1=19.8891.99 mm.a=differences significant at the Student’s t-test for

PB0.05. In the case of ACC the ttest was highly significant at all the tested concentrations.

4. Discussion

clg1 has been isolated as a root chirality2 _and gravitropism mutant, by means of a screening that was principally directed to find new mutants show-ing variations in the degree of right-handed slant-ing of the primary roots, in comparison with the wild-type. In the case of clg1 the variation in the chirality is observable in the clearly increased right-handed coiling and slanting attitude of the after 72 h (the Student’s t is barely significant for

PB0.05).

3.3. Effect of hormones

As can be seen from Fig. 4, clg1 roots did not show increase of resistance to IAA at all the tested concentrations, showed only a small resistance to 2,4 D (at conc. 10−7_{and 10}−6 _{M), and a} moder-ate resistance to NAA at all the tested concentra-tions, but 10−6_{. By contrast, a high resistance to} the plant hormone ethylene, given as ACC, was observed at all the tested concentrations.

In addition, from Fig. 5 it can be seen that the increase of resistance to the auxin transport in-hibitors NPA and TIBA, at concentrations be-tween 10−6₋₅_×₁₀−5_{, of} _clg_{1 roots was very} modest, and resulted slightly significant at the Student’s t-test.

2_{We use the expression chirality when referring to the} characteris-tic of Arabidopsis roots of slanting and producing coils to the right-hand, which apparently correspond to right-handed spirals flat-tened on an agar plate. The right side of the plant has been deter-mined by observing, from the top of the plants, the coiling direction in the wild-type roots with respect to the direction of growth. This coiling hand corresponds to that of a common screw, which in physics is considered right-handed (3,9,21).

(7)

roots. The mutant primary roots produce coils very early (after 4 – 5 days) on inclined, or even on vertically set plates (Fig. 1), whereas these are seen sometimes only in old roots (2 – 3 weeks) in the wild type, and, in addition, the roots show an increased slanting to the right-hand (Fig. 3). These characteristics can be the consequence of a re-duced gravitropic response, since gravitropism and slanting appear to be competitive forces, that lead the growth of the root toward different directions. And because reduced gravitropism was demon-strated in roots growing inside agar layers, e.g. in a condition in which any slanting and coiling are abolished, a primary reduction of gravitropism, and a dependent increase of slanting can be con-sidered as a reason to explainclg1 root character-istics. In addition, observations made on roots growing on horizontally set plates (data not re-ported), a setting that should abolish gravitropism, did not show any clear difference in the coiling habit ofclg1 with respect to the wild type. We can thus preliminarily conclude that the gain in slant-ing in clg1 probably depends on a reduction of

gravitropism, and not necessarily on an increased absolute right-hand deviation from the vertical of the roots. Nevertheless, this point can be settled only by abolishing the force of gravity that pushes also in the case of horizontally set plates the root tip down, and this can be obtained with the use of a clinostat or in microgravity. We can anticipate that we are already working and planning experi-ments in this direction.

The slow gravitropic response, on the other hand, could indicate allelism of clg1 with some of the already isolated mutants that show a similar reduction of gravitropism. The complementation tests we ran (Table 2), however, and the mapping of clg1 to chromosome 5, close to the visible marker tt3, ruled out this possibility.

clg1 thus seems a new mutant, showing reduced gravitropic response, but only a moderate resis-tance to the auxins (mainly NAA). This is a condition that makes this mutant possibly more interesting, because, with the exception of agr1, which, however, was recently shown to encode for an auxin efflux carrier [16,18 – 20,23], and arg1

Fig. 5.clg1 and wild-type Ws root elongation from seedlings grown in presence of the auxin transport inhibitors TIBA and NPA. Elongation is expressed relative to the average root growth of the same genotype on medium without the inhibitors. Each value represents the mean of measurements of at least ten seedlings9SE. Absolute mean values (9SE) of root elongation on hormone free media were: TIBA plot, wild-type=1.3790.04, clg1=1.7590.05; NPA plot, wild-type 8.8690.34, clg1=10.2590.54;

a=differences significant at the Student’st-test forPB0.05.

Table 2

Effect of ACC and NPA on the apical hook of etiolated wtWs andclg1 seedlingsa

Control _{ACC (10}−4 _M) _{NPA (5×10}−5_M) NPA (10−5_M)

+++ + − +++ + − +++ + − +++ + −

wt 0 13 0 18 4 0 0 0 11 0 0 7

7 1 0 0 14 0 0 13

15 0

clg1 0 15

a_{Seedlings were grown in darkness for three days on 0.8% agar medium containing ACC, or NPA. Values represent the number}

(8)

[24], all the other gravitropic mutants show a notable resistance to the auxinic substances and inhibitors of auxin transport. NAA however, is not a natural auxin and it is a easier permeant than IAA and 2,4 D. There is thus the possibil-ity that clg1 could be impaired in a step of the gravitropic response pathway not or weakly con-nected with the action of auxin, possibly located downstream of the known auxin sensitive genes (AUX1, AGR1, AXR1, AXR2, AXR3).

clg1, in addition, showed clearly increased resistance to ethylene, a fact that was already observed in axr1,axr2,aux1 and agr1(eir1,

Atpin2,wa66) [25], even though these mutants are

also highly resistant to auxin, and thus the resis-tance to ethylene could be a consequence of that to auxin. Pickett et al. [26] affirm that they could not find out the real nature of the ethylene resistance, even though they suspect it can be in some way connected with auxin physiology. Eth-ylene resistance in clg1 supports a role for this hormone in the control of gravitropism not nec-essarily connected with auxin physiology. A role that has not been sufficiently clarified so far, es-pecially as concerns roots. clg1, on the other hand, does not show the most common charac-teristics of ethylene insensitive mutants, such as variation in the production of lateral roots, ab-sence of root hairs, and a modification in the seedling apical hook shape [25]. All these charac-teristics in fact appear wild-type in clg1 (Table 2).

Under the genetic point of view the gene in-volved in the clg1 mutation was recessive, trans-mitted through a Mendelian mechanism, and mapped to chromosome 5, close to the visible marker tt3 (Table 1). Close to this marker among the mutants that show characteristics similar to clg1 there is only agr1 on the map. The complementation test we made nevertheless indicated that these two mutants are not allelic. A loss-of-function mutation is thus likely to be involved. The gene concerned, however, did not result tagged, and consequently the only way to isolate it is by means of positional techniques. We can anticipate that this part of the research will be attempted soon. In the meantime, we hope clg1 could be used as a new tool in investi-gating gravitropism and growth movements in plant roots.

Acknowledgements

Thanks are given to the NASC Center for providing the Feldmann – Du Pont collection, as well as the mutants aux1 – 7, axr1 – 3, axr2, the ABRC for providing the line CS 3078, and to Dr D. So¨ll from Yale University (Connecticut, USA) for furnishing rgr1 and wa66, and for

use-ful discussion of the data. This research was sup-ported by a grant from the Italian Space Agency (ASI).

References

[1] C. Darwin, The Power of Movement in Plants, Murray, London, 1880.

[2] K. Okada, Y. Shimura, Reversible root tip rotation in

Arabidopsisseedlings induced by obstacle-touching stim-ulus, Science 250 (1990) 274 – 276.

[3] C. Simmons, D. Soll, F. Migliaccio, Circumnutation and gravitropism cause root waving inArabidopsis thaliana, J. Exp. Bot. 46 (1995a) 143 – 150.

[4] C. Simmons, F. Migliaccio, P. Masson, T. Caspar, D. Soll, A novel root gravitropism mutant of Arabidopsis thaliana exhibiting altered auxin physiology, Physiol. Plant. 93 (1995b) 790 – 798.

[5] A. Johnsson, Circumnutation: results from recent exper-iments on earth and in space, Planta 203 (1997) S147 – 158.

[6] E.P. Maher, S.J.B. Martindale, Mutants of Arabidopsis thaliana with altered responses to auxin and growth, Biochem. Gen. 18 (1980) 1041 – 1053.

[7] J.I. Mirza, The effect of light and gravity in the horizon-tal curvature of roots of gravitropic and agravitropic

Arabidopsis thaliana, Plant Physiol. 83 (1987) 118 – 120. [8] K. Okada, Y. Shimura, Aspects of recent developments

in nutational studies of plant signaling pathways, Cell 70 (1992) 369 – 372.

[9] R. Rutherford, P. Masson,Arabidopsis thaliana sku mu-tant seedlings show exaggerated surface dependent alter-ation in root growth vector, Plant Physiol. 111 (1996) 987 – 998.

[10] M.A. Estelle, C.R. Somerville, Auxin resistant mutants of Arabidopsis with an altered morphology, Mol. Gen. Gen. 206 (1987) 200 – 206.

[11] C. Lincoln, J.H. Britton, M. Estelle, Growth and devel-opment of the axr1 – 3 mutants ofArabidopsis, Plant Cell 2 (1990) 1071 – 1080.

[12] C.J. Bell, E.P. Maher, Mutants of Arabidopsis with abnormal gravitropic responses, Mol. Gen. Gen. 220 (1990) (1990) 289 – 293.

[13] L Hobbie, M. Estelle, The axr4 auxin resistant mutant ofArabidopsis thalianadefine a gene important for root gravitropism and lateral root initiation, Plant J. 7 (1995) 211 – 220.

(9)

[14] B. Marinelli, S. Gomarasca, C. Soave, A pleiotropic

Arabidopsis thaliana mutant with inverted root chiral-ity, Planta 202 (1997) 196 – 205.

[15] M.J. Bennet, A. Marchant, H.G. Green, S.T. May, S.P. Ward, P.A. Millner, A.R. Walker, B. Schulz, K.A. Feldman, Arabidopsis AUX1 gene: A permease-like regulator of root gravitropism, Science 273 (1997) 948 – 950.

[16] R. Chen, P. Hilson, J. Sedbrook, E. Rosen, T. Caspar, P.H. Masson, The Arabidopsis thaliana AGRAVIT-ROPIC1 gene encdes a component of the polar auxin-transport efflux carrier, Proc. Natl. Acad. Sci. USA 95 (1998) 15112 – 15117.

[17] H. Leyser, C. Lincoln, C. Timpte, D. Lammer, J. Turner, M. Estelle, Arabidopsis auxin resistance gene AXR1 encodes a protein related to ubiquitin-activating enzyme E1, Nature 364 (1993) 161 – 164.

[18] C. Luschnig, R. Gaxiola, P. Grisafi, G.R. Fink, EIR1, a root specific protein involved in auxin transport is required for gravitropim in Arabidopsis thaliana, Gene Dev. 12 (1998) 2175 – 2187.

[19] A. Mueller, C. Guan, L. Gaelweiler, P. Taezler, P. Huijser, A. Marchant, G. Parry, M. Bennet, E. Wis-man, K. Palme, AtPIN2 defines a locus ofArabidopsis

for root gravitropism control, EMBO J. 17 (1998) 6903 – 6911.

[20] K. Utsuno, T. Shikanai, Y. Yamada, T. Hashimoto, AGR, an agravitropic locus of Arabidopsis thaliana, encodes a novel membrane protein family member, Plant Cell Physiol. 39 (1998).

[21] F. Migliaccio, O. Micciulla, S. Ferrari, Waving and coiling in Arabidopsis roots, Plant Physiol. Bioch. 19 (1996) S02 – 04.

[22] L. Baillaud, Les Mouvements d’Exploration et d’En-roulement des Plant Volubiles, in: E. Buenning (Ed.), Encyclopedia of Plant Physiol. New Series, vol. 17, Springer Verlag, Berlin, 1962, pp. 637 – 715.

[23] M. Estelle, Polar auxin transport, new support for an old model, Plant Cell 10 (1998) 1775 – 1778.

[24] J.C. Sedbrook, R. Chen, P.H. Masson, Altered re-sponse to gravity (ARG1) encodes a DnaJ-like protein that potentially interacts with the cytoskeleton, Proc. Natl. Acad. Sci. USA 96 (1999) 1140 – 1145.

[25] J.R. Ecker, The ethylene signal transduction pathway in plants, Science 268 (1995) 667 – 678.

[26] F.B. Pickett, A.K. Wilson, M. Estelle, The aux1 muta-tion of Arabidopsis confers both auxin and ethylene resistance, Plant Physiol. 94 (1990) 1462 – 1466.

(1)

Fig. 1. Growth habit ofclg1 primary roots compared with wild-type Ws (ecotype Wassilevskjia). Plants grown on hard-agar plates (1.5%), set 60° inclined with respect to the horizontal plane.

Table 1

Chromosome mapping ofclg1a,b

F2 total Marker phenotype of F2clg1 individuals

er1 gl-1 cer2-2 tt3-1

Marker phenotype clg1 ch1-1

1 2 3 4

Chromosome location 5 5

110

110 110

110

605 110

Number screened

22 30 21 0

No _{marker phenotype} ₁₁₀ ₂₀

20.0 27.2 19.0 0

% 18.1 18.1

2.73 1.47 0.31 2.05

X2_{per 3:1 ratio} _36.66

a_{A chromosome mapping line with visible markers} _ch_1, _er_1, _gl_1, _cer_2, _tt_{3 (ABRC seed stock CS3078) was crossed to} _clg₁

(pollen donor) and grown to generate the F2 progeny for mapping.

b_{Low values result from deliberate conservative}_clg_{1 scoring.} gravitropic response (Fig. 1). The first two charac-teristics, in contrast with the wild type, are visible not only on inclined, but frequently also on verti-cally oriented hard-agar plates. This set of root characteristics persisted after three generations produced through selfing and two crossings with the wild type, indicating the monogenetic nature of the traits. Moreover, the cross with the wild-type showed, through the suppression of the muta-tion in the F1, and the segregamuta-tion in the F2, the recessivity of the characters, and the Mendelian transmission of them. The mutation did not show however, cosegregation with the resistance to kanamycin in the F2 population originated from the kanamycin resistant F1 individuals. A segrega-tion consistent with a ratio 9:3:3:1 was seen. Con-sequently, the involved gene did not result tagged.

Tests of allelism were made with most known and available mutants1 _{Through crosses with}

aux1 – 7,axr1 – 3, axr2,wa66 (agr1,eir1,pin2),rgr1

(axr4), allelism was primarily excluded on the basis of the complementation seen in the F1 or F2. On the other hand, the map position of clg1 excluded also allelism with the above mutants, with the exception of agr1, that maps on chromo-some 5. The complementation test we run however contradicts this possibility.

1_clg_{1, as reported, was complemented with the most common and}

available mutants that show a modified response to gravitropism or auxin concentration. In spite of this, we cannot exclude that it could be allelic with one of the many already described mutants. We did not think, however, it is useful to furthermore extend complementation tests that are time consuming and not always productive.

(2)

S.Ferrari et al./Plant Science158 (2000) 77 – 85 81 The mapping of clg1 was made by crossing it

with the visible markers mapping line CS 3078 (originally DP 23, Patton and Meinke line), from the ABRC of Ohio State University. As reported in Table 1, the mutation showed linkage with the marker tt3, and thus it is located on chro-mosome 5, close to that marker. No recombi-nants were recovered among the 110 F2 clg1 individuals of the reported screen.

A control of other root characteristics, which frequently appear modified in similar mutants

PB0.05.

such as axr1, rgr1, showed that clg1 roots are wild-type as concerns the number of lateral roots, root hairs, and the number of statolyts in the root cap cells.

3.2. Gra6itropism, slanting and elongation

The gravitropic response of clg1, reported in Fig. 2 (top), was clearly reduced, since about 30° of difference were observed with respect to the wild-type after an eight hour period. After 24 h, however, both the wild type and the mutant were for the most part 90° bent (data not shown). Therefore, in the mutant the gravitropic response appears only slowed down, but not re-duced in absolute values.

The slanting to the right-hand, reported in Fig. 3, was increased in clg1, even though a widened distribution was observed with respect to the wild-type. The absolute mean values were 22.9° in wild-type and 48.2° in the mutant.

Root elongation (Fig. 2, bottom) was only slightly increased with respect to the wild-type

(3)

clg1=12.7791.10 mm; IAA plot, wild-type=7.2590.26,clg1=12.7590.77; NAA plot, wild-type=2.0390.06,clg1=2.259

0.08; ACC plot, wild-type=15.1590.96 mm; clg1=19.8891.99 mm.a=differences significant at the Student’s t-test for

PB0.05. In the case of ACC the ttest was highly significant at all the tested concentrations.

4. Discussion

clg1 has been isolated as a root chirality2 _and

gravitropism mutant, by means of a screening that was principally directed to find new mutants show-ing variations in the degree of right-handed slant-ing of the primary roots, in comparison with the wild-type. In the case of clg1 the variation in the chirality is observable in the clearly increased right-handed coiling and slanting attitude of the after 72 h (the Student’s t is barely significant for

PB0.05).

3.3. Effect of hormones

As can be seen from Fig. 4, clg1 roots did not show increase of resistance to IAA at all the tested concentrations, showed only a small resistance to 2,4 D (at conc. 10−7_{and 10}−6 _{M), and a}

moder-ate resistance to NAA at all the tested concentra-tions, but 10−6_{. By contrast, a high resistance to}

the plant hormone ethylene, given as ACC, was observed at all the tested concentrations.

In addition, from Fig. 5 it can be seen that the increase of resistance to the auxin transport in-hibitors NPA and TIBA, at concentrations be-tween 10−6₋₅_×₁₀−5_{, of} _{clg1 roots was very}

modest, and resulted slightly significant at the Student’s t-test.

2_{We use the expression chirality when referring to the}

characteris-tic of Arabidopsis roots of slanting and producing coils to the right-hand, which apparently correspond to right-handed spirals flat-tened on an agar plate. The right side of the plant has been deter-mined by observing, from the top of the plants, the coiling direction in the wild-type roots with respect to the direction of growth. This coiling hand corresponds to that of a common screw, which in physics is considered right-handed (3,9,21).

(4)

S.Ferrari et al./Plant Science158 (2000) 77 – 85 83 roots. The mutant primary roots produce coils

very early (after 4 – 5 days) on inclined, or even on vertically set plates (Fig. 1), whereas these are seen sometimes only in old roots (2 – 3 weeks) in the wild type, and, in addition, the roots show an increased slanting to the right-hand (Fig. 3). These characteristics can be the consequence of a re-duced gravitropic response, since gravitropism and slanting appear to be competitive forces, that lead the growth of the root toward different directions. And because reduced gravitropism was demon-strated in roots growing inside agar layers, e.g. in a condition in which any slanting and coiling are abolished, a primary reduction of gravitropism, and a dependent increase of slanting can be con-sidered as a reason to explainclg1 root character-istics. In addition, observations made on roots growing on horizontally set plates (data not re-ported), a setting that should abolish gravitropism, did not show any clear difference in the coiling habit ofclg1 with respect to the wild type. We can thus preliminarily conclude that the gain in slant-ing in clg1 probably depends on a reduction of

The slow gravitropic response, on the other hand, could indicate allelism of clg1 with some of the already isolated mutants that show a similar reduction of gravitropism. The complementation tests we ran (Table 2), however, and the mapping of clg1 to chromosome 5, close to the visible marker tt3, ruled out this possibility.

clg1 thus seems a new mutant, showing reduced gravitropic response, but only a moderate resis-tance to the auxins (mainly NAA). This is a condition that makes this mutant possibly more interesting, because, with the exception of agr1, which, however, was recently shown to encode for an auxin efflux carrier [16,18 – 20,23], and arg1

a=differences significant at the Student’st-test forPB0.05. Table 2

Effect of ACC and NPA on the apical hook of etiolated wtWs andclg1 seedlingsa

Control _{ACC (10}−4 _M) _{NPA (5}_×₁₀−5_M) NPA (10−5_M)

+++ + − +++ + − +++ + − +++ + −

wt 0 13 0 18 4 0 0 0 11 0 0 7

7 1 0 0 14 0 0 13

15 0

clg1 0 15

a_{Seedlings were grown in darkness for three days on 0.8% agar medium containing ACC, or NPA. Values represent the number}

(5)

[24], all the other gravitropic mutants show a notable resistance to the auxinic substances and inhibitors of auxin transport. NAA however, is not a natural auxin and it is a easier permeant than IAA and 2,4 D. There is thus the possibil-ity that clg1 could be impaired in a step of the gravitropic response pathway not or weakly con-nected with the action of auxin, possibly located downstream of the known auxin sensitive genes (AUX1, AGR1, AXR1, AXR2, AXR3).

clg1, in addition, showed clearly increased resistance to ethylene, a fact that was already observed in axr1,axr2,aux1 and agr1(eir1, Atpin2,wa66) [25], even though these mutants are

also highly resistant to auxin, and thus the resis-tance to ethylene could be a consequence of that to auxin. Pickett et al. [26] affirm that they could not find out the real nature of the ethylene resistance, even though they suspect it can be in some way connected with auxin physiology. Eth-ylene resistance in clg1 supports a role for this hormone in the control of gravitropism not nec-essarily connected with auxin physiology. A role that has not been sufficiently clarified so far, es-pecially as concerns roots. clg1, on the other hand, does not show the most common charac-teristics of ethylene insensitive mutants, such as variation in the production of lateral roots, ab-sence of root hairs, and a modification in the seedling apical hook shape [25]. All these charac-teristics in fact appear wild-type in clg1 (Table 2).

Under the genetic point of view the gene in-volved in the clg1 mutation was recessive, trans-mitted through a Mendelian mechanism, and mapped to chromosome 5, close to the visible marker tt3 (Table 1). Close to this marker among the mutants that show characteristics similar to clg1 there is only agr1 on the map. The complementation test we made nevertheless indicated that these two mutants are not allelic. A loss-of-function mutation is thus likely to be involved. The gene concerned, however, did not result tagged, and consequently the only way to isolate it is by means of positional techniques. We can anticipate that this part of the research will be attempted soon. In the meantime, we hope clg1 could be used as a new tool in investi-gating gravitropism and growth movements in plant roots.

Acknowledgements

Thanks are given to the NASC Center for providing the Feldmann – Du Pont collection, as well as the mutants aux1 – 7, axr1 – 3, axr2, the ABRC for providing the line CS 3078, and to Dr D. So¨ll from Yale University (Connecticut, USA) for furnishing rgr1 and wa66, and for