Journal of Experimental Botany, Vol. 55, No. 397, pp. 685±693, March 2004
DOI: 10.1093/jxb/erh068 Advance Access publication 30 January, 2004
RESEARCH PAPER
Developmental anatomy and auxin response of lateral root
formation in Ceratopteris richardii
Guichuan Hou1, Jeffrey P. Hill2 and Elison B. Blanca¯or1,*
1
Plant Biology Division, The Samuel Roberts Noble Foundation Inc., 2510 Sam Noble Parkway, Ardmore, OK
73401, USA
2
Department of Biological Sciences, Idaho State University, Pocatello, ID 83209, USA
Received 29 July 2003; Accepted 14 November 2003
Abstract
The homosporous fern Ceratopteris richardii exhibits a homorhizic root system where roots originate from the shoot system. These shoot-borne
roots form lateral roots (LRs) that arise from the
endodermis adjacent to the xylem poles, which is
in contrast to ¯owering plants where LR formation
arises from cell division in the pericycle. A
detailed study of the ®fth shoot-borne root showed
that one lateral root mother cell (LRMC) develops
in each two out of three successive merophytes.
As a result, LRs emerge alternately in two ranks
from opposite positions on a parent root. From
LRMC initiation to LR emergence, three developmental stages were identi®ed based on anatomical
criteria. The addition of auxins (either indole-3acetic acid or indole-3-butyric acid) to the growth
media did not induce additional LR formation, but
exogenous applications of both auxins inhibited
parent root growth rate. Application of the polar
auxin-transport inhibitor N-(1-naphthyl)phthalamic
acid (NPA) also inhibited parent root growth without changing the LR initiation pattern. The results
suggest that LR formation does not depend on
root growth rate per se. The result that exogenous
auxins do not promote LR formation in C. richardii
is similar to reports for certain species of ¯owering plants, in which there is an acropetal LR population and the formation of the LRs is insensitive
to the application of auxins. It also may indicate
that different mechanisms control LR development
in non-seed vascular plants compared with angiosperms, taking into consideration the long and independent evolutionary history of the two groups.
Key words: Auxin, Ceratopteris, development, fern, lateral
root.
Introduction
Lateral root (LR) formation in ¯owering plants has drawn
a great deal of attention for a number of reasons. First, LRs
form a major component of the root system (Lloret and
Casero, 2002). Second, LR development involves de novo
organ production from pericycle cells in the parent roots,
affording ample opportunities for studying plant organogenesis (Malamy and Benfey, 1997). Third, LR formation
can often be induced by the application of auxin, allowing
the generation of large populations of developmentally
synchronized meristems (Laskowski et al., 1995). However, relatively little attention has been paid to LR
development in ferns, despite potentially interesting anatomical and morphological differences between ferns and
¯owering plants (Clowes, 1961; Groff and Kaplan, 1988;
Steeves and Sussex, 1989; Charlton, 1996; Hou and Hill,
2002).
In most ¯owering plants, LR formation occurs in the
pericycle cells of the maturation zone of the parent roots
(Esau, 1965). Developmental studies have demonstrated
that the formation of LR meristems in Raphanus sativus
and Arabidopsis thaliana is a two-stage process, i.e. there
is a transition of the meristem from a primordial stage in
which there was no anatomically de®ned meristem to a
stage with a de®ned functioning meristem (Laskowski
et al., 1995). Further histological investigation has indicated that there can be as many as eight developmental
stages in Arabidopsis LR development (Malamy and
Benfey, 1997). There is also evidence at the genetic level
that LR development in Arabidopsis involves several
* To whom correspondence should be addressed. Fax: +1 580 224 6692. E-mail: eblanca¯[email protected]
Journal of Experimental Botany, Vol. 55, No. 397, ã Society for Experimental Biology 2004; all rights reserved
686 Hou et al.
developmental phases (Celenza et al., 1995). These studies
have helped to pave the way for further molecular genetic
dissection of LR formation in ¯owering plants (Rost and
Bryant, 1996; van den Berg et al., 1998; Casimiro et al.,
2003).
It has been known for some time that auxin (indole-3acetic acid, IAA) plays a pivotal role in LR development in
¯owering plants (Torrey, 1950). It is likely that acropetal
auxin transport (i.e. toward the root apex) in the parent root
of Arabidopsis controls LR development (Reed et al.,
1998). Basipetal auxin transport (i.e. toward the root base)
also may be required for LR initiation (Casimiro et al.,
2001). Application of exogenous auxin can greatly
increase the number of LRs formed on parent roots in R.
sativus and Arabidopsis (Laskowski et al., 1995). Further
evidence supporting the notion that auxin is involved in LR
formation in angiosperms comes from genetic studies in
Arabidopsis. For instance, certain mutants of Arabidopsis
with signi®cantly higher levels of endogenous IAA
concentration produce more LRs (Boerjan et al., 1995;
King et al., 1995). Transgenic plants that overexpress
auxin biosynthesis genes also produce more LRs (Klee
et al., 1987). Auxin transport inhibitors (e.g. naphthylphthalamic acid, NPA) or mutations that cause a defect in
auxin transport can block LR production (Ruegger et al.,
1997). Despite overwhelming evidence supporting a role
for auxin in LR production in ¯owering plants, it is not
known whether auxin regulates LR development in ferns in
a similar manner.
Recent advances in the study of root development in the
angiosperm model Arabidopsis have been based, in part,
on detailed information about organ ontogeny at the
cellular level. The relative simplicity and regularity of
cellular pattern in Arabidopsis has been regarded as
advantageous (Dolan et al., 1993). This point of view is
noteworthy because the developmental anatomy of fern
roots generally lends itself to an even more accurate
analysis of the cellular basis of organ development using
routine histological methods. In ferns, root development
typically involves the initiation and subsequent division
activities of a single root apical cell (RAC), which
typically is tetrahedral in shape (Gunning et al., 1978;
Gifford, 1983; Hou and Hill, 2002). Merophytes, which are
packets of cells clonally related by cell division from a
common mother cell, can be traced back to individual
cutting faces of the RAC. Merophytes are produced from
the three proximal faces of the RAC and contribute to the
root proper. Merophytes arising from the distal face form
the root cap. Cell divisions along the three proximal faces
of the root apical cell occur in a strict sequence (Hou and
Hill, 2004), producing three merophyte `orthostichies' in
the root body (Gunning et al., 1978; Gifford, 1983; Hou,
2001; Hou and Hill, 2002). Formative divisions, which
refer to cell divisions occurring in the zone of the root tip
where the initial cells for various tissue zones are formed,
also take place in a fairly regular and predictable sequence.
These divisions give rise to a characteristic number of cell
®le initials in each merophyte during its ontogeny. Thus,
the relatively precise developmental and anatomical
features that make roots in species like Arabidopsis an
attractive model system are even more convenient in ferns
like Azolla and Ceratopteris. Previous experience with C.
richardii suggested that this non-seed plant model (Hickok
et al., 1995) would be an excellent candidate for the
investigation of lateral root development.
The objectives of this research on Ceratopteris are (1) to
determine if there are any characteristic stages evident in
LR development, based on anatomical data from the ®fth
shoot-borne root of young sporophytes and (2) to test the
effects of exogenous auxin and an auxin transport inhibitor
on the production of LRs from this ®fth root, providing the
information necessary for a comparison with LR development in ¯owering plants. To the authors' knowledge, this
is the ®rst experimental study on the effect of exogenous
auxin on LR production in a representative non-seed
vascular plant species.
Materials and methods
Plant material and growth conditions
Ceratopteris richardii (Hn-n) (Scott and Hickok, 1987) spores were
used in this study. Spores were axenically cultured on half-strength
basal salt medium (Gamborg's B-5 BSM: G-5768, Sigma Chemical
Co., St Louis, MO) in Petri dishes sealed with para®lm at 25 °C with
constant light (160 mmol photons m±2 s±1) in a walk-in growth
chamber. Under these conditions, male and hermaphroditic
gametophytes became morphologically distinct in about 7 d.
Sterilized water was added to the Petri dishes once a week.
Gametophytes became sexually mature in 2 weeks and the ®rst
leaves of the sporophytes became visible in about 4 weeks.
One hundred and eighty sporophytes (for the different experiments described below) at the 2-leaf stage were transferred to GA7
growth vessels (Magenta Co., Chicago, IL), one plant per container
with 300 ml of the BSM liquid medium. Plants were grown in the
same chamber with the same culture conditions described above.
Light microscopic analysis of lateral root development on the
®fth root
Roots (including LRs) of different lengths at node 5 from 16 plants
were ®xed in 2.5% glutaraldehyde in 25 mM potassium phosphate
buffer, pH 7.1, for 1±4 d. After ®xation, all roots were washed with
buffer, dehydrated through a graded series of ethanol, and embedded
in JB-4â methacrylate plastic (Polysciences, INC., Warrinton, PA).
Eight roots were sectioned longitudinally and another eight roots
were sectioned transversely. Serial 3 mm sections were stained with
1% aqueous acid fuchsin, and counter-stained with 0.05% toluidine
blue. Images were captured on a Nikon Microphot-FX microscope
(Nikon Inc., Melville, NY) equipped with a SPOT RT digital camera
(Diagnostic Instruments, Sterling Heights, MI). To illustrate the
morphological pattern of LR development, a root from position 5
was cleared in 70% ethanol, rehydrated and stained in dilute
toluidine blue (0.05%), and photographed with a Nikon CoolPix 990
digital camera. All digital images were labelled and printed using
Adobe Photoshop 5.0LE (Adobe Systems Inc. Mountain View, CA).
Lateral root development in Ceratopteris 687
Auxin treatments
The effects of exogenous indole-3-acetic acid (IAA) on LR
development were tested on 60 sporophytes that were growing in
the liquid BSM media for 3 weeks. A stock solution of 100 mM IAA
(Sigma Chemical Co., St Louis, MO) was prepared in 95% ethanol
as described by Rogg et al. (2001) and stored at ±20 °C. Different
concentrations of IAA (5, 10, 20, 40, and 60 mM) were prepared by
adding the appropriate amount of the stock solution to the liquid
BSM. Two control groups consisting of ten plants in each group
were included initially. One set of controls was without the addition
of ethanol and the other involved the addition of 95% ethanol
corresponding to the volume of ethanol in the 60 mM IAA treatment
(solvent controls). Five groups, each consisting of eight sporophytes,
were treated with different concentrations of IAA. Since treatment
with IAA for 3 d in pea gives the maximum effect on LR production
such that no additional increase in LR number is observed beyond
this time (Torrey, 1950), young C. richardii sporophytes were
treated with IAA for 3 d prior to making LR counts and light
microscopic preparation. The length of the ®fth root and LR number
were measured under a dissecting microscope before and after the
IAA treatments (72 h interval). The measurement and addition of
IAA were done axenically in a clean bench. LR production was
de®ned as the difference in LR numbers before and after auxin
treatment (i.e. the number of new LRs formed during the 3 d period
of IAA treatment). One-way analysis of variance (ANOVA) and
Tukey HSD analyses for the ®fth root's growth and LR production
were performed using SPSSâ11.5 (SPSS Inc., Chicago, IL). After
the measurements, half of the plants were transferred to IAA-free
media and the other half returned to IAA solution so it could be
observed whether new LRs emerge after longer auxin treatments.
In addition to IAA, the effects of 5 mM indole-3-butyric acid
(IBA) were tested on root growth and LR production. IBA is also a
naturally occurring auxin and more ef®cient than IAA in promoting
LR initiation in ¯owering plants (Rogg et al., 2001). Twenty-two
plants were included in this study. The experimental procedure for
testing IBA was similar to the protocol for testing IAA. Differences
among ®fth root growth rates and LR production were evaluated by
t-tests performed on SPSSâ11.5. The effects of varying concentrations (5, 10, 20, 40, and 60 mM) of the polar auxin transport inhibitor
N-(1-naphthyl)phthalamic acid (NPA) on ®fth root growth and LR
production were also performed. Experimental procedures and data
analyses were similar to those described for the IAA experiments.
At the end of each experiment, representative roots at position 5
were photographed using a SZX12 stereomicroscope (Olympus,
Melville, NY) equipped with a SPOT RT digital camera (Diagnostic
Instruments, Sterling Heights, MI). These roots were then ®xed for
light microscopy as described above.
Results
Lateral root primordia formation on the ®fth root
In the ®fth root of Ceratopteris, new LRs were initiated in
an acropetal direction, such that longer LRs were located at
the proximal end of the parent root and progressively
shorter LRs were located toward the root tip (Fig. 1A). LRs
emerged alternately on opposite sides along the parent root
longitudinal axis, with long and short segments of parent
root tissue between successive LRs (Fig. 1A).
Anatomically, the ®rst sign of LR initiation was cell
expansion in the endodermal layer at the base of a
particular merophyte (Fig. 1B, C). This enlarged cell was
designated as the lateral root mother cell (LRMC) and was
relatively easy to identify based on its size and characteristic position (Hou and Hill, 2002). Only two out of every
three successive merophytes produced by the three
proximal division faces of the parent root apical cell
(RAC) formed LRMCs (Figs 1B, C, 2A, B).
The early expansion of the LRMC prior to any cell
division was designated as stage 1 of LR development.
During stage 1, each LRMC became much larger than the
adjacent endodermal initials (Fig. 2A, B). The next stage
(designated stage 2) involved four cell divisions. The ®rst
division of a LRMC was asymmetrical and oblique
(Fig. 2C). It produced one founder cell and a daughter
merophyte (Fig. 2C). The founder cell divided asymmetrically two additional times and produced two more
daughter merophytes (Fig. 2D). The three merophytes
were produced one after another and around the founder
cell that was now roughly tetrahedral in shape, with its
vertex oriented towards the stele (Fig. 2D). The fourth
division of the founder cell was approximately periclinal
with respect to the surface of the parent root (arrowhead in
Fig. 2D) and produced the ®rst root cap initial. Upon
completion of the fourth division, the former founder cell
was designated as the lateral root apical cell (LRAC). The
LRAC was completely surrounded by its four daughter
merophytes so there was no longer any direct cell-to-cell
contact with the parent root. In stage 2, the ®rst three
divisions of the LRMC established a speci®c division
sequence (either clockwise or counter clockwise) for the
proximal cutting faces of the newly forming LRAC that
was maintained in subsequent LR development. Stage 3
was de®ned as a LR that had a well-de®ned LRAC and had
begun the addition of more merophytes from all four of its
cutting faces (Fig. 2E±G).
Exogenous auxin does not affect lateral root
production
Since auxin is a key regulator of LR development in
¯owering plants, it was tested whether the application
exogenous IAA or IBA produces a similar effect on LR
production in C. richardii. At all IAA concentrations
tested, the growth rate of the ®fth root of C. richardii was
signi®cantly inhibited (P <0.05) (Fig. 3A). The average
growth rate of control roots was 5.97 mm d±1 and the
addition of 5 mM IAA to the media reduced root growth
rate to 1.35 mm d±1. The growth rates of roots among the
®ve IAA treatments were not signi®cantly different from
each other (P >0.05) (Fig. 3A). Despite the strong
inhibitory effect of IAA on the growth rate of the ®fth
root, IAA did not signi®cantly alter LR production on the
®fth root of C. richardii (P >0.05) (Fig. 3B). After longer
exposure to IAA for up to 9 d, there was no evidence of
additional LR initiation (data not shown).
Application of 5 mM IBA to young C. richardii
sporophytes yielded similar effects on growth rate and
LR production to that of 5 mM IAA treatment. Mean
688 Hou et al.
Fig. 1. Representative ®fth root morphology and anatomy in C. richardii showing early lateral root (LR) development. (A) LRs are initiated in an
acropetal sequence and emerge in two ®les from the parent root. There are alternating long (L) and short (S) segments of parent root tissue
between successive LRs. (B) The LR developmental pattern is evident microscopically in the root tip. One lateral root mother cell (LRMC, white
stars) is produced per merophyte in every two out of three successive merophytes. (C) A line drawing depicting merophyte boundaries in the
median longitudinal section shown in (B). The numbers represent successive merophytes derived from the root apical cell (RAC), with the
youngest one designated as merophyte 1. The arrow indicates cells that constitute the third column of merophytes that are mostly out of the plane
of the section. LRMCs were identi®ed in merophytes 10, 11, 13, and 14. The predicted sites of LR production were in merophytes 8, 7, 5, 4, 2,
and 1. Merophytes 3, 6, 9, and 12 were not involved in LR production in this root. a, Aerenchyma chamber; RAC, root apical cell; PD,
protoderm; RC, root cap. Bar=5 mm (A), 20 mm (B).
growth rate deceased signi®cantly from 5.15 mm d±1 in
control to 1.08 mm d±1 in the IBA treatment (P <0.05)
(Fig. 3C). Although LR production decreased from an
average of 4.83 in the control to 3.70 in the IBA treatment,
this difference was not statistically signi®cant (P >0.05)
(Fig. 3D).
Aspects of ®fth root morphology generally related to
organ growth changed in response to auxin treatment. The
alternating long and short intervals between successive
LRs in auxin treated roots (Fig. 4B, C) were shorter
compared with the control roots (Figs 1A, 4A). However,
exogenous IAA and IBA did not change the overall
morphological pattern of LR development. LRs still
emerged in two ranks along the longitudinal axis of the
parent root and the pattern of long and short intervals
between successive LRs was still evident (Fig. 4).
Anatomical analysis indicated that treating the ®fth root
for 3 d with 5 mM IAA (Fig. 5A, B) or 9 mM IAA (Fig. 5C)
did not change its cellular developmental pattern.
Merophyte boundaries in the parent root could still be
identi®ed and there was still a single LRMC or LRAC
visible in every two out of three successive merophytes
(Fig. 5A, C). The pericycle and endodermis were both
present, each as a single sheath of cells located between the
cortex and stele. There was no evidence that additional
endodermal or pericycle cells near the root tip or in the
more mature regions of the root had responded to the
exogenous auxin to initiate the LR developmental
programme (Fig. 5B, C). Moreover, exogenous IAA did
not substantially alter the development of LRs themselves.
For instance, a spectrum of LR development was still
evident along the parent root longitudinal axis, with earlier
stages of LR development being progressively closer to the
parent root apical cell (Fig. 5A, C).
The auxin transport inhibitor N-1-naphthylphthalamic
acid (NPA) does not affect lateral root production
In Arabidopsis, the auxin transport inhibitor NPA arrests
LR development by blocking primordium initiation
(Casimiro et al., 2001). This prompted an investigation
as to whether NPA causes a similar effect on LR initiation
in C. richardii. Like auxin treatments, addition of NPA to
the growth medium signi®cantly inhibited the growth rate
of the ®fth root in C. richardii (P <0.05) (Fig. 6A). The
average growth rate was 6.67 mm d±1 for control plants and
the application of 5 mM NPA decreased the growth rate to
2.93 mm d±1. Increasing the NPA concentration did not
cause further inhibition of root growth (Fig. 6A) nor did it
inhibit LR production in the ®fth root of C. richardii (P
>0.05) (Fig. 6B). Longer exposures (e.g. 7 d) to NPA in the
growth media did not change the LR initiation pattern, but
inhibited growth in both the parent and the LR roots (data
not shown).
Discussion
Developmental anatomy of lateral root formation
Anatomical analysis of the ®fth root of C. richardii
indicated that LR formation can be divided into three
major developmental stages: (1) LRMC expansion, (2)
ordered LRMC divisions to generate a distinct LRAC, and
(3) LR growth and emergence via the activities of the
Lateral root development in Ceratopteris 689
Fig. 2. Transverse (A±C, F) and longitudinal (D, E, G) sections of the ®fth root showing the development of LRs of C. richardii. (A) The ®rst
sign of LR initiation was cell expansion of a lateral root mother cell (LRMC) located in the proximal position in the endodermal layer of the
parent root. (B) The LRMC enlarged 3±4-fold without cell division. (C) The ®rst division of the LRMC was oblique and asymmetrical. (D) Three
additional and successive asymmetrical divisions occurred and generated a tetrahedral cell with its vertex pointing toward the longitudinal axis of
the parent root. This tetrahedral cell was the lateral root apical cell (LRAC) of the newly formed LR. The arrowhead indicates the fourth division
(telophase) of the LRMC as it produced the ®rst root cap merophyte of the LR. (E±G) After the LRAC was formed, the subsequent development
of the LR involved more LRAC divisions from all four cutting faces. (F, G) Adjacent pericycle cells also divided and contributed to tissues that
connected the LR and parent root stele. (A), (B), (C), and (F) were transverse sections from the same root and photographed at the same
magni®cation (except C). (D), (E), and (G) were photographed from different roots at position 5. In (C) and (D), d represents daughter merophyte
of the founder cell (f). A, Aerenchyma chamber; P, pericycle; C, cortex, E, epidermis, RC, root cap, EN, endodermis. Bars=20 mm (A, B, C, F);
30 mm (D, E, G).
LRAC and its developing merophytes. These events
correspond well with earlier descriptions of LR development in C. thalictroides reported nearly a century ago by
Lachmann (discussed in Barlow, 1996). The ®rst stage (i.e.
LRMC expansion) is different from that described for LR
formation in Arabidopsis (Laskowski et al., 1995; Malamy
and Benfey 1997; Dubrovsky et al., 2001) since the latter
occurs at a greater distance from the apical meristem and
involves cell divisions by pericycle cells. In C. richardii,
LRMC expansion occurs in the meristematic zone, very
close to the parent root apical cell and before the
endodermis has fully differentiated. During LRMC expansion, adjacent endodermal initials continue to undergo
further cell divisions (Figs 1B, 2). The enlargement of the
LRMC is not simply based on a failure of these cells to
divide relative to adjacent endodermal cells, since the
direction of cell expansion in LRMCs are also different
from normal endodermal cells.
Regulation of lateral root initiation in the ®fth root
Other studies suggest that the instructive signal(s) for the
LR formation (i.e. establishing the sites for stage 1 LRMC
development) might come from the vascular tissues in the
parent root due to the fact that LR initiation sites are
adjacent to the vascular poles in the roots investigated
(Esau, 1965; Mallory et al., 1970; McCully, 1975). The
development of LRMCs in C. richardii progresses into
stage 2 before full differentiation of the adjacent xylem
(Fig. 2C, D). Although the speci®cation of LRMCs must
occur even earlier, stage 1 does not become evident until
after the parent root has already completed the formative
cell divisions that establish the locations of xylem and
phloem within the organ (Hou, 2001; Hou and Hill, 2004).
Thus, there was a developmental window when factors that
control the 2-fold symmetry of the diarch stele within the
root could also be in¯uential in establishing the corresponding placement of LRMCs in C. richardii. This
690 Hou et al.
Fig. 3. Effects of exogenous indole-3-acetic acid (IAA) and indole-3-butyric acid (IBA) on growth rate and LR production of the ®fth root in
young sporophytes of C. richardii. Young sporophytes were treated for 3 d in IAA. (A) IAA at all concentrations tested signi®cantly inhibited
growth of the ®fth root (P <0.05). (B) Application of IAA to the growth medium did not signi®cantly affect the LR production in the ®fth root,
although average LR numbers were lower in all groups of IAA concentrations compared with the control group (P >0.05). Bars are means 6SD.
N=8 for each group of IAA treatment and 19 for the control. Effects of 5 mM IBA on parent root growth and LR production of roots at position 5
in young sporophytes of C. richardii. Young sporophytes were treated for 3 d in IBA. (C) IBA signi®cantly (P <0.05) inhibited the growth of
roots at position 5. (D) LR production was not signi®cantly different (P >0.05) between controls and IBA-treated roots. N=12 for the control and
10 for the group of IBA treatment.
Fig. 4. Representative ®fth root morphology of C. richardii after
auxin treatment. (A) In control roots, LRs emerged in two ranks along
the longitudinal axis of the parent root producing a pattern of long (L)
and short (S) intervals between successive LRs. Treatment for 3 d
with 5 mM IAA (B) or 5 mM IBA (C) did not change the pattern of
lateral root (LR) formation on the root, i.e. LRs (arrows) still emerged
alternatively along the longitudinal axis on the parental root. Short (S)
and long (L) segments of root tissue remain between successive LRs,
as seen in the ®fth root, although the overall lengths are reduced
(compare with Fig. 1A). The three roots were photographed at the
same magni®cation. Scale bar=2 mm.
potential in¯uence of the stele in the radial plane does not
address the positioning of LRMCs in the longitudinal
plane. However, LRs always form from the most proximal
cell within the endodermal cell lineage in two out of every
three successive merophytes (Fig. 1B) (Hou and Hill,
2004). This precise placement of LRs along the main root
axis contrasts with many ¯owering plants where LR
position may not follow a strict pattern (Dubrovsky et al.,
2000; Charlton, 1996).
The pattern of long and short segments between
successive LRs (Fig. 1A) can be attributed to the fact
that only two out of every three successive merophytes
initiate LRMCs (Fig. 1B, C). The short segment seen in
mature regions of the main root is the distance between the
bases of two successive merophytes that both produce
LRMCs. The long segment between LRs results from
intercalating a merophyte that does not initiate a LRMC in
between two merophytes that do. Hypothetically, if a
mutation exists in which the third merophyte also initiates
a LRMC, a third rank of LRs should be produced such that
all the segments between successive LRs become equal.
Mutations that alter the pattern of LR initiation would be
also invaluable in deciphering the relationship between LR
and xylem pattern formation.
The initiation of LRs may be related very precisely to
the cell division pattern of the root apical cell and its
derivatives in the parent root of C. thalictroides (Chiang
and Gifford, 1971) and Marsilea coromandelica (Charlton,
1996, and references therein). This view is reminiscent of
root development in Azolla, a heterosporous fern, which
also exhibits relatively precise cell division and differentiation patterns (Gunning et al., 1978). However, the short
determinate roots of Azolla do not produce any LRs. In the
®fth shoot-borne roots of C. richardii, the future sites of
LR initiation are predictable within merophytes in the
meristematic zone (Fig. 1B). Consequently, the pattern is
amenable to further molecular genetic analysis of LR
initiation in relation to precise formative divisions that
yield the endodermal cell lineage in the parent root tip in
this model fern species (Hou, 2001; Hou and Hill, 2004).
Lateral root development in Ceratopteris 691
Fig. 5. Anatomy of roots at position 5 in C. richardii after treatment with 5 mM IAA (A, B) or 9 mM IAA (C) for 3 d. Cell division was active in
the root tip. The endodermis (EN) and pericycle (P) remained as a single sheath of cells (A, C). There was no anatomical evidence showing the
initiation of additional lateral roots after a 3 d treatment with IAA (A±C). The section in (B) depicts a representative ®fth root treated with 5 mM
IAA 2 mm behind the tip. Arrowheads (in A) indicate merophyte boundaries. LRAC, lateral root apical cell; LRMC, lateral root mother cell;
RAC, root apical cell. Bars=50 mm.
Fig. 6. Effects of the auxin transport inhibitor N-1-naphthylphthalamic acid (NPA) on root growth rate and LR production of the
®fth root in C. richardii. All NPA concentrations tested signi®cantly
inhibited (P <0.05) root growth (A), but LR production among the
experimental groups was not signi®cantly different (P >0.05) (B).
Each bar represents the mean 6SD. N = 7 for each group.
Identi®cation of mutations that disrupt formative divisions
in the root, analogous to ones that alter radial patterning in
Arabidopsis (Scheres et al., 1995; Laurenzio et al., 1996;
Nakajima et al., 2001), will help to clarify relationships
between parent root cell lineage and LR initiation in C.
richardii.
Although there may be some important differences
between the early development of the embryonic root and
lateral roots in Arabidopsis, there are also many similarities in the overall ontogeny of these two root types
(Laskowski et al., 1995; Malamy and Benfey, 1997; van
den Berg et al., 1998). In C. richardii, there are also some
noteworthy developmental similarities among the different
root types formed by the sporophyte. The differentiation of
root apical mother cells (RAMCs) leading to the formation
of shoot-borne and leaf-borne roots (Hou and Hill, 2002)
essentially conforms to the same three developmental
stages seen in the present study of LRMC formation.
Shoot-borne roots (including the ®fth root studied here)
develop from an enlarged RAMC recently derived from
the shoot apex. A second population of leaf-borne RAMCs
develops from hypodermal parenchyma cells that initiate
cell expansion followed by division (i.e. stages 1 and 2),
forming tetrahedral root apical cells at the base of leaf. The
fact that de novo meristem organization in shoot-borne,
leaf-borne, and lateral roots in C. richardii has so much in
common suggests that highly regulated cell division
patterns that sometimes precede fern root meristem
formation are not necessarily critical to the process of
RAMC formation. In addition, shoot-borne roots produced
before root 4 in this species form a normal endodermis, but
typically lack lateral roots (Hou and Hill, 2002). This
provides another line of evidence that factors other than the
regulation of cell division per se is likely to be relevant to
692 Hou et al.
accomplish LRMC speci®cation in C. richardii. This
perspective does not diminish the value of having a
tractable study system where LRMC formation is so highly
predictable in time and space.
Initiation of LRMCs in C. richardii occurs very close to
the parent root apical cell (Fig. 1B) where other cells of the
root including endodermal initials are still actively dividing (Fig. 2). Differentiated endodermal cells in more
mature regions of the root are characterized by having
Casparian strips on their anticlinal walls (Hou, 2001; Hou
and Hill, 2004). The early development of LRMCs from
cells in the endodermal layer may eliminate the need for
complex cell differentiation and dedifferentiation processes during LR initiation in C. richardii. Cellular
dedifferentiation processes were once generally thought
to be one important early step of LR development from the
pericycle in typical ¯owering plants, although the accuracy
of this assumption has recently been called into question
(Casimiro et al., 2003). In Arabidopsis, at least some LR
primordia may form from pericycle cells that have
undergone little or no prior differentiation and have
continued to progress through the cell cycle (Dubrovsky
et al., 2000; Beeckman et al., 2001). Regardless, cellular
dedifferentiation of endodermal cells is not an issue in C.
richardii because LRMCs are speci®ed early in root
development and late-forming LRs do not arise.
between the sites of acropetal lateral roots. It is possible
that only the adventitious laterals are inducible from the
pericycle by auxin, while the initiation pattern of acropetal
laterals does not rely on auxin signals and is thus unaltered
by the application of exogenous auxins or auxin transport
inhibitors. The results from the present experiments with
C. richardii ®t this model, if it is assumed that the fern
simply lacks an adventitious LR population.
Given the fact that investigations of fern root and lateral
root development fall far behind those in ¯owering plants,
it is hoped that the ®ndings in this study will stimulate new
interest in the developmental biology of roots in non-seed
vascular plants. Such studies will no doubt shed more light
on the present understanding of plant root evolution
(Raven and Edwards, 2001) and the full diversity of
mechanisms underlying this aspect of development in land
plants (Pryer et al., 2001).
Acknowledgements
We thank two anonymous reviewers for their valuable comments
and suggestions and Dr Xin Shun Ding (Noble Foundation) for
careful reading of an earlier draft of this study. This work was
supported by the National Aeronautics and Space Administration
(NASA grant number NAG2-1518) and the Noble Foundation to
EBB.
Auxin does not induce additional lateral roots in C.
richardii
References
An important ®nding reported in this study is that
exogenous IAA or IBA does not induce additional LR
formation in C. richardii under the experimental conditions provided. IAA or IBA signi®cantly inhibited root
growth rate, suggesting that the primary impact of the
auxin in roots of C. richardii was on processes related to
general cell and organ expansion. Exogenous auxins
signi®cantly inhibited parent root growth (Fig. 3A, C),
but did not preclude LR emergence (Fig. 4B, C), implying
that LR emergence is independent of growth rate per se.
Like the parent root, however, LR growth was also
inhibited by auxin treatment (data not shown). These
experiments with the auxin transport inhibitor NPA, which
signi®cantly inhibited growth of the ®fth root (Fig. 6A) and
its LRs (data not shown), further supports the notion that
auxin is mainly involved in the regulation of organ growth
in C. richardii.
Previous reports vary as to whether exogenous auxins do
or do not promote LR production in ¯owering plants
(Blakely et al., 1982; Lloret and Pulgarin, 1992).
Particularly in eudicots, there may be two LR populations:
acropetal and adventitious (Charlton, 1996, and references
therein). Acropetal lateral roots form in the meristematic
zone of the parent root and emerge in an acropetal
sequence. In some species, these may be the only lateral
roots that form. Adventitious lateral roots form later and in
Barlow PW. 1996. Cellular patterning in root meristems. In: Waisel
Y, Eshel A, Kafka® U, eds. Plant roots: the hidden half, 2nd edn.
New York: Marcel Dekker Inc., 77±109.
Beeckman T, Burssens S, Inze D. 2001. The peri-cell-cycle in
Arabidopsis. Journal of Experimental Botany 52, 403±411.
Blakely LM, Durham M, Evans TA, Blakely RM. 1982.
Experimental studies on lateral root formation in radish
seedling roots. I. General methods, developmental stages, and
spontaneous formation of laterals. Botanical Gazette 143, 341±
352.
Boerjan W, Cervera MT, Delarue M, Beeckman T, Dewitte W,
Bellini C, Caboche M, Onckelen HV, Montagu MV, Inze D.
1995. superroot, a recessive mutation in Arabidopsis, confers
auxin overproduction. The Plant Cell 7, 1405±1419.
Casimiro I, Marchant A, Bhalerao RP, et al. 2001. Auxin
transport promotes Arabidopsis lateral root initiation. The Plant
Cell 13, 843±852.
Casimiro I, Beeckman T, Graham N, Bhalerao R, Zhang H,
Casero P, Sandberg G, Bennett MJ. 2003. Dissecting
Arabidopsis lateral root development. Trends in Plant Science
8, 165±171.
Celenza Jr JL, Grisa® PL, Fink GR. 1995. A pathway for lateral
root formation in Arabidopsis thaliana. Genes and Development
9, 2131±2142.
Charlton WA. 1996. Lateral root initiation. In: Waisel Y, Eshel A,
Kafka® U, eds. Plant roots: the hidden half, 2nd edn. New York:
Marcel Dekker Inc., 149±173.
Chiang SH, Gifford Jr EM. 1971. Development of the root of
Ceratopteris thalictroides with special reference to apical
segmentation. Journal of the Indian Botanical Society 50A, 96±
106.
Lateral root development in Ceratopteris 693
Clowes FAL. 1961. Apical meristems. Oxford: Blackwell Scienti®c
Publications.
Dolan L, Janmaat K, Willemsen V, Linstead P, Poethig S,
Roberts K, Scheres B. 1993. Cellular organization of the
Arabidopsis thaliana root. Development 119, 71±84.
Dubrovsky JG, Doerner PW, Colon-Carmona A, Rost TL. 2000.
Pericycle cell proliferation and lateral root initiation in
Arabidopsis. Plant Physiology 124, 1648±1657.
Dubrovsky JG, Rost TL, Colon-Carmona A, Doerner PW. 2001.
Early primordium morphogenesis during lateral root initiation in
Arabidopsis thaliana. Planta 214, 30±36.
Esau K. 1965. Plant anatomy, 2nd edn. New York: John Wiley and
Sons Inc.
Gifford Jr EM. 1983. Concept of apical cells in bryophytes and
pteridophytes. Annual Review of Plant Physiology 34, 419±440.
Groff PA, Kaplan DR. 1988. The relation of root systems to shoot
systems in vascular plants. Botanical Reviews 54, 387±422.
Gunning BES, Hughes JE, Hardham AR. 1978. Formative and
proliferative cell divisions, cell differentiation, and
developmental changes in the meristem of Azolla roots. Planta
143, 121±144.
Hickok LG, Warne RR, Fribourg RS. 1995. The biology of the
fern Ceratopteris and its use as a model system. International
Journal of Plant Sciences 156, 332±345.
Hou GC. 2001. Developmental and molecular genetic studies of
roots in the fern Ceratopteris richardii. PhD thesis, Idaho State
University, USA.
Hou GC, Hill JP. 2002. Heteroblastic root development in
Ceratopteris richardii (Parkeriaceae). International Journal of
Plant Sciences 163, 341±351.
Hou GC, Hill JP. 2004. Developmental anatomy of the ®fth shootborne root in young sporaphytes of Ceratopteris richardii.
Planta. (in press).
King JJ, Stimart DP, Fisher RH, Bleecker AB. 1995. A mutation
altering auxin homeostasis and plant morphology in Arabidopsis.
The Plant Cell 7, 2023±2037.
Klee JH, Horsch RB, Hinchee MA, Hein MB, Hoffman NL.
1987. The effects of overproduction of two Agrobacterium
tumefaciens T-DNA auxin biosynthetic gene products in
transgenic petunia plants. Genes and Development 1, 86±96.
Laskowski MJ, Williams ME, Nusbaum HC, Sussex IM. 1995.
Formation of lateral root meristem is a two-stage process.
Development 121, 3303±3310.
Laurenzio D, Wysocka-Diller LJ, Malamy JE, Pysh L,
Helariutta Y, Freshour G, Hahn MG, Feldmann KA,
Benfey PN. 1996. The SCARECROW gene regulates an
asymmetric cell division that is essential for generating the
radial organization of the Arabidopsis root. Cell 86, 423±433.
Lloret PG, Casero PJ. 2002. Lateral root initiation. In: Waisel Y,
Eshel A, Kafka® U, eds. Plant roots: the hidden half, 3rd edn.
New York: Marcel Dekker Inc., 127±155.
Lloret PG, Pulgarin A. 1992. Effect of naphthalene acetic acid on
the formation of lateral roots in the adventitious roots of Allium
cepa: number and arrangement of laterals along the parent root.
Canadian Journal of Botany 70, 1891±1896.
Malamy JE, Benfey PN. 1997. Organization and cell
differentiation in lateral roots of Arabidopsis thaliana.
Development 124, 33±44.
Mallory TE, Chiang SH, Cutter EG, Gifford Jr EM. 1970.
Sequence and pattern of lateral root formation in ®ve selected
species. American Journal of Botany 57, 800±809.
McCully ME. 1975. The development of lateral root. In: Torrey
JG, Clarkson DT, eds. The development and function of root. New
York: Academic Press, 105±124.
Nakajima K, Sena G, Nawy T, Benfey PN. 2001. Intercellular
movement of the putative transcription factor SHR in root
patterning. Nature 413, 307±311.
Pryer KM, Schneider H, Smith AR, Cran®ll R, Wolf PG, Hunt
JS, Sipes SD. 2001. Horsetails and ferns are a monophyletic
group and the closest living relative to seed plants. Nature 409,
618±622.
Raven JA, Edwards D. 2001. Roots: evolutionary origins and
biogeochemical signi®cance. Journal of Experimental Botany 52,
381±401.
Reed RC, Brady SR, Muday GK. 1998. Inhibition of auxin
movement from the shoot into the root inhibits lateral root
development in Arabidopsis. Plant Physiology 118, 1369±1378.
Rogg LE, Lasswell J, Bartel B. 2001. A gain-of-function mutation
in IAA28 suppresses lateral root development. The Plant Cell 13,
465±480.
Rost TL, Bryant JA. 1996. Root organization and gene expression
patterns. Journal of Experimental Botany 47, 1613±1628.
Ruegger M, Dewey E, Hobbie L, Brown D, Bernasconi P,
Turner J, Muday G, Estelle M. 1997. Reduced
naphthylphthalamic acid binding in the tir3 mutant of
Arabidopsis is associated with a reduction in polar auxin
transport and diverse morphological defects. The Plant Cell 9,
745±757.
Scheres BL, Laurenzio D, Willemsen V, Hauser MT, Janmaat
K, Weisbeek P, Benfey PN. 1995. Mutations affecting the radial
organization of the Arabidopsis root display speci®c defects
throughout the embryonic axis. Development 121, 53±62.
Scott RJ, Hickok LG. 1987. Genetic analysis of antheridiogen
sensitivity in Ceratopteris richardii. American Journal of Botany
74, 1872±1877.
Steeves TA, Sussex IM. 1989. Patterns in plant development.
Cambridge: Cambridge University Press.
Torrey JG. 1950. The induction of lateral roots by indoleacetic
acid and root decapitation. American Journal of Botany 37, 257±
264.
van den Berg C, Weisbeek P, Scheres B. 1998. Cell fate and cell
differentiation status in the Arabidopsis root. Planta 205, 483±
491.