Skip to main content
NCBI home page
Annals of Botany logoLink to Annals of Botany
. 2007 Apr;99(4):765–776. doi: 10.1093/aob/mcm013

Abnormal Root and Nodule Vasculature in R50 (sym16), a Pea Nodulation Mutant which Accumulates Cytokinins

Alicia N Pepper 1, Andrew P Morse 1, Frédérique C Guinel 1,*
PMCID: PMC2802945  PMID: 17383989

Abstract

Background and Aims

R50 (sym16) is a pea nodulation mutant with fewer and shorter lateral roots (LR), fewer nodules and high levels of cytokinins (CK). Because a link exists between CK imbalance and abnormal vasculature, the vasculature of the primary root (PR) and LR of R50 was studied and it was compared with that of the wild-type ‘Sparkle’. Also nodule vasculature was investigated to correlate R50 low nodulation phenotype with CK accumulation.

Methods

PR and first-order LR were hand-sectioned transversely in different locations and at different ages. Vascular poles were counted and root and stele diameters measured. To evaluate LR primordia number, roots were cleared. Nodules obtained from inoculated plants were either fixed and sectioned or cleared; numbers of vascular strands and of tracheary elements in the strands were counted.

Key Results

‘Sparkle’ PR is triarch, whereas that of R50 can be triarch, tetrarch or pentarch. Furthermore, as the R50 roots developed, supernumerary vascular strands appeared but, as they aged, the new growth of more roots displayed the triarch pattern. LR vasculature differed from that of PR: whereas ‘Sparkle’ LR had three or four poles, those of R50 had two or three. No differences in PR or PR stele diameters existed between the two lines. Whereas ‘Sparkle’ nodules had two vascular strands, most R50 nodules possessed three; however, because R50 nodules were variable in size, their vasculature was highly diverse in terms of strand length. A strong correlation was found between nodule length and number of tracheary elements in strands.

Conclusions

R50 displays an additional number of vascular poles in its PR, a smaller number of vascular poles in its first-order LR and an altered vasculature in its nodules. It appears that these three characteristics are linked to the high levels of CKs that the mutant accumulates over its development.

Key words: Cytokinin, heterorhizy, nodulation mutant, nodule vasculature, pea mutant, Pisum sativum, pleiotropic mutant, R50 (sym16), root vasculature, vascular tissue

INTRODUCTION

Mutants are great tools for studying the physiology behind growth and development, and they are especially useful in understanding the control exercised by hormones during these processes. One nodulation mutant from pea (Pisum sativum) proves to be extremely powerful as it displays several pleiotropic traits. R50 (sym16) was first described as a low nodulator by Kneen et al. (1994); it was further characterized by Guinel and Sloetjes (2000) and Ferguson et al. (2005). It is shorter than the near-isogenic wild-type (WT) line ‘Sparkle’, with pale leaves caused by a lower total chlorophyll content and a reduced root system (Guinel and Sloetjes, 2000). Three weeks after inoculation with Rhizobium, R50 displays a low number of nodules which are pale and small. Mostly arrested in the root inner cortex, the infection threads are convoluted and generally not associated with a nodule primordium (Guinel and Sloetjes, 2000). R50 seedlings have a total cytokinin (CK) content higher than that of WT; as the plants mature, the CK levels decrease in WT shoots whereas they remain high in those of R50 (Ferguson et al., 2005). This CK accumulation was partly explained by Held et al. (Wilfrid Laurier University, Canada, unpubl. res.), who provided evidence for an R50 deficiency (biochemical activity and RNA expression) in cytokinin oxidase/dehydrogenase (CKX), the enzyme responsible for CK degradation. In addition, the mutant exhibits a partial de-etiolation phenotype and its shoots and roots display a differential sensitivity towards ethylene (Ferguson et al., 2005). Whereas R50 shoots are less sensitive to exogenous ethylene than WT shoots, R50 and WT roots exhibit a similar sensitivity to the gas.

The high levels of CKs in R50 prompted us to study the vasculature of the mutant; although the mutant shoot was not altered, the root exhibited a tetrarch or pentarch pattern instead of the usual triarch seen in WT pea root (Ferguson, 2002). The arrangement of primary vascular tissue is determined very early in the development of a plant. A mature embryo is organized in such a way that it possesses two meristems which give rise to either the shoot or the root procambium. As the meristematic cells are continuously dividing, procambial cells are continuously laid out and will follow a tight differentiation programme to become cells of either xylem or phloem tissues. In the lifetime of a plant, vascular development is a constant and predictable process which must, however, remain flexible as plants are incessantly adapting to new conditions. For example, mature vascular parenchyma cells are able to dedifferentiate in response to wounding (e.g. Hardham and McCully, 1982) and the different vascular tissues will alter their anatomy in response to environmental stimuli [e.g. temperature (Lu et al., 1991) or high salinity (Junghans et al., 2006)]. The differentiation of the procambial cells into mature cells is a complex process that involves many players. Thus, recently, brassinosteroids and an arabinogalactan protein, known as xylogen, have been implicated (reviewed in Fukuda, 2004). In contrast, for numerous decades now, two major hormones, auxins and cytokinins, have been known to play key roles in this developmental step (Aloni, 1995). While polar auxin transport has been recognized as crucial for the continuous formation of the vascular bundles, cytokinins have proved to be essential in the formation and maintenance of the procambial cells (Fukuda, 2004). In addition, different types of cells in the vascular bundles are apparently active sites of CK biosynthesis (Miyawaki et al., 2004). Thus AtIPT, the arabidopsis gene encoding isopentenyltransferase, the first enzyme in the CK biosynthesis pathway, is differentially expressed: AtIPT1 is expressed in the xylem progenitor cells of the root tip and AtIPT3 in the phloem throughout the plant (Miyawaki et al., 2004).

That cytokinins are important signalling molecules in the differentiation of vascular tissues (Fukuda, 2004) is supported by the study of both transgenic plants and mutants. However, because non-physiological interactions are likely to occur in the heterologous transgenics (Heyl and Schmülling, 2003), results using mutants are easier to analyse. Thus, among many arabidopsis mutants described by Scheres et al. (1995), woodenleg (wol) stands out because it exhibits only protoxylem in the stele of its primary root (Mähönen et al., 2000). Furthermore, its procambial cells are unable to divide periclinally, resulting in only a few cell files in the stele (Scheres et al., 1995; Mähönen et al., 2000). The WOL gene, also identified as CRE1 (Cytokinin Response 1; Inoue et al., 2001), encodes a histidine kinase which is part of the two-component signal transduction system involved in CK perception.

Here, the data obtained by Ferguson (2002) on the abnormal root vasculature observed in R50, a CK-accumulating nodulation mutant, are expanded. Because the CK levels increase with ageing of the mutant, a developmental study of the vasculature was undertaken not only in the primary root but also in the lateral roots. We were interested in determining the origin of the additional vascular poles of the primary roots and whether or not the lateral roots exhibited an altered vasculature. Because we were unaware of any study on the hormonal regulation of nodule vasculature, we were particularly interested in investigating the vascular arrangement in R50 nodules to potentially link the low nodulation phenotype to the high levels of root cytokinins.

MATERIALS AND METHODS

Growth conditions

Seeds of Pisum sativum L. ‘Sparkle’ and of the mutant R50 (sym16) were surface-sterilized in 8 % household bleach (5·25 % sodium hypochlorite) for 5 min and rinsed three times with sterile distilled water. They were left to imbibe overnight in the dark in sterile water and then planted, four to a pot, in round pots (155 mm diameter) filled with sterile Holiday vermiculite (Vil Vermiculite Inc., Toronto, ON, Canada). Pots were placed in a tray partially filled with deionized water. Plants were grown in a controlled growth-chamber (Lab-line Biotronette; Lab-line Instruments Inc., Melrose Park, IL, USA), under cool white fluorescent lights (average light intensity of 380 µmol m−2 s−1) and 58 % humidity, with a 23/18 °C, 16/8 h, light/dark regime.

Plants grown for the root vasculature study were watered, as needed, by adding deionized water to the tray. Those for the nodule vasculature study were inoculated 4 d after planting with 5 mL of either 2 % or 5 % (v/v) Rhizobium leguminosarum bv. viciae 128C53K (Lipha Tech Inc., Milwaukee, WI, USA) grown in a yeast–mannitol broth to an optical density of 0·5–0·6 at 600 nm. Inoculated plants were watered daily for 1 week after planting and then every second day, alternating between deionized water and a low nitrogen nutrient solution (2 mm KH2PO4, 0·5 mm Ca(NO3)2·4H2O, 2 mm K2SO4, 1 mm MgSO4·7H2O, 0·2 mm FeIII EDTA, 0·05 mm KCl, 0·025 mm H3BO3, 2 × 10−3 mm ZnSO4·7H2O, 2 × 10−3 mm MnSO4·H2O, 5 × 10−4 mm CuSO4·5H2O, 5 × 10−4 mm Na2MoO4·2H2O).

Tissue preparation and microscopy

Fresh tissue

Primary roots were harvested 5, 9, 13 and 17 d after planting (DAP). Transverse hand-sections were made at three locations on the primary root: 0·5 cm from the root tip (i.e. meristematic zone or MeZ), at the root hair initiation zone (RhZ) and 1·0 cm from the cotyledons (i.e. maturation zone or MaZ). Two first-order lateral roots, one being the closest to the cotyledons and the other located 3 cm from the cotyledons, were also harvested 13 and 17 DAP. These were hand-sectioned transversely 1·0 cm distal to the site of branching from the primary root.

Some sections were stained with 0·05 % (w/v) toluidine blue O (TBO; Fisher Scientific Company, Fair Lawn, NJ, USA) in benzoate buffer (pH 4·4), rinsed and mounted in water. These sections were observed with a Reichert Microstar IV microscope using ×10 (NA 0·25) and ×40 (NA 0·66) objective lenses to determine the number of vascular poles. For all sections, except those made in the MeZ, the number of xylem poles was determined based on lignified vessel elements (VEs) and that of phloem poles on lignified phloem fibres because such cells were easily identified with TBO. For sections taken in the MeZ, enlarged immature VEs or the first lignified VEs were used to determine the number of xylem poles. Photographs were taken on Pan F (ASA 50) film using a Yashica 108 Multi Program camera (Kyocera Corporation, Japan) connected to a Reichert-Jung Photostar camera system. Negatives were scanned using a CanoScan 4200F scanner (Canon Inc., Mississauga, ON, Canada) and images were edited using ArcSoft PhotoStudio® 5·5 (ArcSoft Inc., Fremont, CA, USA).

Additional sections, prepared for fluorescence microscopy, were stained with TBO to quench autofluorescence and then with 0·01 % (w/v) ethidium bromide (BioShop Canada Inc., Burlington, ON, Canada) to stain specifically lignified tissues (O'Brien and McCully, 1981). Sections were rinsed in water before being mounted in 0·05 % (w/v) aniline blue (Sigma Chemical Company, St Louis, MO, USA) in phosphate buffer (pH 8·5) to stain callose for the localization of phloem sieve plates (Smith and McCully, 1978). A quick verification of the arrangement of the vascular tissues was done using a Jenalumar (Jena Zeiss, Germany) fluorescence microscope with a blue excitation filter set (two KP490 filters and a B229 filter) and a G47 barrier filter. The lignified cell walls fluoresce red and the sieve tubes with their sieve plates yellowish white.

Unstained sections were used to measure the diameters of the primary roots and of the first-order lateral roots, and their stele diameters. For consistency, the longest diameter was measured since the roots were not always perfectly round. Root and stele diameter measurements were made using a graticule (Jena Zeiss, Germany) and the aforementioned light microscope (objectives ×4 and ×10, respectively).

Fixed tissue

Nodules were harvested from plants 17 d after inoculation (DAI) in such a way that each nodule studied was still attached to the root bearing it. Samples were fixed in 3 % (v/v) glutaraldehyde (Canemco & Marivac Inc., Canton de Gore, QC, Canada) in 0·05 m potassium phosphate buffer (pH 6·8) under a continuous vacuum for 3·5 h. They were rinsed twice (15 min each) in 0·025 m potassium phosphate buffer (pH 6·8). An increasing ethanol gradient (3, 10, 30, 50, 70, 90, 95 and 100 %) was used to dehydrate the tissue which was then infiltrated with an increasing gradient of LR White® hard-grade acrylic resin (London Resin Company Ltd, distributed by Canemco & Marivac Inc.) (1 : 3, 2 : 2, 3 : 1 LR white resin : 100 % ethanol). Samples were infiltrated with the resin for a minimum of 1 week, changing it once a day, before being embedded into fresh resin and cured at 60 °C for 24 h in an oxygen-free environment.

Fixed nodules were sectioned both longitudinally and transversely using a Sorvall® Porter-Blum MT-2 ultramicrotome. Sections were heat-fixed to slides and then stained with TBO (pH 4·4); they were observed and photographed as previously described for the root sections. To determine the number of vascular poles in the nodules, serial transverse sections were performed. Ten consecutive thin sections (approx. 450 nm thick each) were viewed but the next ten thick sections (approx. 1500 nm thick each) were discarded. This process was repeated until the entire nodule was sectioned from its apex to the connection with the root vasculature. This allowed for an approximate three-dimensional reconstruction of the nodule vasculature to be made.

Cleared tissue

Entire root systems were excised 7 and 12 DAP. After the number of emerged lateral roots was counted, the root system was cleared according to the protocol of Luschnig et al. (1998). In short, individual root systems were placed in a methanol : glacial acetic acid (3 : 1, v/v) solution, which was replaced 30 min later by a solution made of 130 mm NaCl, 10 mm sodium phosphate (pH 7) and 0·1 % Tween 20. After two short baths (5 min) in that solution, roots were placed within the clearing solution (0·16 g chloral hydrate per millilitre of 20 % glycerol). The primordia corresponding to the initiation of first-order lateral roots were counted using a dissecting microscope.

As an additional means to understand the three-dimensional arrangement of the nodule vasculature, nodules at 17 DAI were cleared according to a protocol modified from Eskew et al. (1993). The samples were placed in undiluted household bleach for 2 h, the first of which was under continuous vacuum; they were then rinsed twice (5 min each) with water before being placed under a vacuum in 30 % glycerol (1 h), then in 60 % glycerol (1 h). Nodules were mounted in 60 % glycerol for microscopic analysis and were examined and photographed as previously described. Several parameters listed below were recorded: nodule length (i.e. the distance taken perpendicularly from the centre of the root vasculature to the nodule apex), number of vascular traces linked to the root vasculature, width of the vasculature at the base of the nodule (i.e. distance along the root vasculature between the nodular traces the furthest apart), and the number of tracheary elements (TEs) across that base. Nodule height and vasculature width were measured using a graticule. In addition, the longest vascular strand was chosen in individual nodules and its length assessed by counting the number of TEs from the root vasculature to the trace tip.

Statistical analysis

In all experiments, the mutant R50 was compared with the wild-type ‘Sparkle’. Statistical analysis of data was done using SigmaStat® software (Jandel Scientific, San Rafael, CA, USA). Student's t-tests were used to determine significant line-related differences between all parameters, but in the event of non-normally distributed samples, the Mann–Whitney Rank Sum test was performed instead. An ANOVA was used to compare the mean diameters when measurements were grouped according to vascular patterns. Correlation was determined using a linear regression to generate the R values.

RESULTS

Primary roots

The wild-type ‘Sparkle’

The tissues making up the primary root of a young wild-type seedling (5 DAP) were seen to differentiate as expected for a dicotyledon species (e.g. Popham, 1955; Torrey, 1967; Esau, 1977), especially for pea (Rost et al., 1988). The root tissues, including the vascular tissues, differentiated in a basipetal manner, and the maturation process occurred centripetally. In the MeZ (Fig. 1A), while the phloem tissue was difficult to distinguish, non-lignified primary VEs of the xylem were not because of their obvious enlarged size (Fig. 1J). At the RhZ (Fig. 1A), all tissues were easily recognized. The inner cortex was delimited by an endodermis while the stele was enclosed within the pericycle; as noted by Bond (1948) and reiterated by Rost et al. (1988), the pericycle in pea is complex, composed of a monolayer over the phloem but of two or three cell layers opposite the xylem. In the RhZ, some VEs had secondary walls; however, the central region of the stele was still occupied by non-lignified parenchyma cells (Fig. 1F). Phloem fibres with their secondary walls were also present; the phloem was easily located in aniline blue-mounted sections since the sieve plates appeared fluorescent (data not shown). In the MaZ (Fig. 1A), lignified VEs and parenchyma cells occupied the central region of the vascular cylinder, and lignified phloem fibres were obvious (Fig. 1B).

Fig. 1.

Fig. 1.

Open in a new tab

Schematic representation of the root system of a young pea plant (A) showing the relative position of the hand sections taken along the primary root and the arrangement of the vascular tissue seen in the photomicrographs on the right. Transverse hand sections of 9-d-old ‘Sparkle’ (B, F and J) and R50 (C–E, G–I, K and L) primary roots were taken from the maturation zone (MaZ, 1·0 cm from the cotyledons) (B–E), from the root hair initiation zone (RhZ) (F–I), and from the meristematic zone (MeZ, 0·5 cm from the root tip) (J–L). All sections were stained with TBO. For ‘Sparkle’, pf = phloem fibres; non-lignified primary vessel elements are indicated by an asterisk. Scale bars in B–L = 50 µm.

Xylem is unable to differentiate in the absence of phloem (Aloni, 1995); thus, it is usual for a root to have as many xylem poles as phloem poles. Although in the stele of a root, these poles of different tissues have separate locations, it is conventional to group them together in what is known as a vascular pole; the number of vascular poles is equal to either that of the xylem poles or that of the phloem poles. In this study, the conventional term of vascular poles was used for convenience. In a young ‘Sparkle’ seedling as well as in older plants, generally three vascular poles were present at all locations (e.g. Fig. 1B, F, J); thus, as reported in the literature for pea (e.g. Torrey, 1967), ‘Sparkle’ had a triarch pattern as the norm. Occasionally, however, four poles could be seen (Fig. 2); this occurred most often in the root MaZ (e.g. four out of 36 in Fig. 2A; one out of 36 in Fig. 2D), as if one vascular pole were added along the root length as the tissues differentiated basipetally. The additional vascular pole was initially composed exclusively of primary tissue since it developed prior to the formation of a vascular cambium (data not shown). Surprisingly, as the plants grew older, fewer of them (as a percentage) displayed a tetrarch pattern in their vascular tissue (e.g. compare A with D in Fig. 2); however, the numbers of vascular poles in ‘Sparkle’ plants of different ages were not significantly different.

Fig. 2.

Fig. 2.

Open in a new tab

Percentages of ‘Sparkle’ (black bars) and R50 (open bars) plants having three, four or five vascular poles in their primary roots for day 5 (A), day 9 (B), day 13 (C) and day 17 (D). MaZ represents the maturation zone (1·0 cm from the cotyledons), RhZ the root hair initiation zone and MeZ the meristematic zone (0·5 cm from the root tip). Percentages were chosen for simplicity of presentation; however, statistical tests were done on the raw data, and significant differences were found between the two pea lines at all locations and all ages, except for the MeZ at 9, 13 and 17 d (n = 30–36).

The mutant R50

Primary roots of R50 developed in a fashion similar to that of ‘Sparkle’ (i.e. basipetal cell differentiation and centripetal cell maturation); however, they showed a significant amount of variability in their vascular patterning (Fig. 1). For example 9 DAP, whereas roots displayed mostly a triarch vasculature at the MeZ, they exhibited either a triarch, tetrarch or pentarch vasculature at the RhZ and MaZ (Fig. 2B). At all days analysed, as roots of individual plants grew in length, a trend towards increased vasculature in basal regions was observed with > 20 % of roots having a pentarch stele in the MaZ (Fig. 2), suggesting that the supernumerary poles are the results of a secondary stage of primary vascularization. Statistically significant differences were also seen across the ages, but only in the RhZ and MeZ. For example, at the RhZ, as the plants grew older, an increasing percentage of primary roots displayed a triarch stele (Fig. 2A–D), as if the additional vascular poles observed in the seedling 5 DAP were disappearing during temporal growth. As in ‘Sparkle’, the additional poles were composed initially only of primary tissues (e.g. Fig 1H, I).

First-order lateral roots

Because cytokinins have been known to affect the development of lateral roots, especially their emergence (Li et al., 2006), we sought to determine using cleared root systems if there were any line-related differences between primordium initiation and emergence of first-order lateral roots. At the two ages studied, ‘Sparkle’ had a significantly greater number of total lateral root events (i.e. sum of primordia, emerging and emerged roots) on their primary roots than R50 (Table 1). ‘Sparkle’ also had significantly more first-order lateral roots (i.e. emerging and emerged) than R50; this was caused partially by a failure of the R50 primordia to emerge in the young seedlings (Table 1).

Table 1.

Number (mean±s.e.) of root primordia and of lateral roots of first-order (emerging and emerged) counted on the primary roots of 7 and 12 DAP plants (n = 14–17)

Primordia First-order lateral roots
‘Sparkle’ R50 ‘Sparkle’ R50
7 d 12·29 ± 1·08a 19·14 ± 1·37b 37·43 ± 1·30c 12·64 ± 1·14d
12 d 13·31 ± 1·66e 13·29 ± 1·16e 53·44 ± 2·03f 44·07 ± 2·49g
Open in a new tab

For a specific age, means with different letters indicate significant differences (P < 0·001) between ‘Sparkle’ and R50.

The number of vascular poles in the first-order lateral roots of ‘Sparkle’ was often increased in comparison to that seen in the primary roots; on the contrary, in R50, these roots frequently had a decreased number of poles (Fig. 3). The majority of the ‘Sparkle’ first-order lateral roots located in the MaZ were either tetrarch (Fig. 3A) or triarch (Fig. 3B), with the tetrarch pattern being the most common 13 and 17 DAP (Table 2). In the same region and at the same age, R50 first-order lateral roots were either diarch (Fig. 3C) or triarch (Fig. 3D), with the diarch pattern observed in most cases (Table 2). Similar vascular patterns were observed in the first-order lateral roots located 3 cm from the cotyledons (data not shown).

Fig. 3.

Fig. 3.

Open in a new tab

Transverse hand-sections of ‘Sparkle’ (A and B) and R50 (C and D) first-order lateral roots located the closest to the cotyledons. These lateral roots were from 17-d-old plants and sections were made 1·0 cm from the site of branching from the primary root. Whereas most first-order lateral roots were tetrarch (A) and diarch (C) for ‘Sparkle’ and R50 respectively, there were triarch roots (B and D) in both lines. All sections were stained with TBO. For ‘Sparkle’, VE = vessel elements; pf = phloem fibres. Scale bars = 20 µm.

Table 2.

Percentage of plants having different vascular patterns in the first-order lateral roots located in the MaZ (n = 27–29)

13 DAP 17 DAP
Vascular pattern ‘Sparkle’ R50 ‘Sparkle’ R50
Diarch 7·1 63·0 0·0 72·4
Triarch 32·1 37·0 44·8 24·1
Tetrarch 57·1 0·0 55·2 3·4
Pentarch 3·6 0·0 0·0 0·0
Open in a new tab

All comparisons are statistically significant (P < 0·001) between the two pea lines on both days.

Root and stele diameter

There was no significant difference between the two pea lines in either the primary root diameter or the stele diameter of that root at all locations and all ages studied; the data for MaZ at 17 DAP are shown in Table 3. When the root or stele diameters of R50 plants with a triarch vasculature were compared with those of R50 roots having increased vasculature (four or five poles), the differences were not significant at most locations and for plants of different ages (data not shown), suggesting that the increased vasculature was not related to a larger vascular cylinder. This contrasts with the first-order lateral roots where the root and stele diameters of ‘Sparkle’ were significantly greater than those of R50, as shown in Table 3 for the oldest first-order lateral roots. This increase in lateral root and stele diameters was correlated with a higher number of vascular poles (as determined through an ANOVA).

Table 3.

Root diameter and stele diameter of the primary root in the MaZ (n = 18) and of the oldest first-order lateral root (n = 12) of ‘Sparkle’ and R50 plants 17 DAP

Primary root First-order lateral root
Poles Root diameter (mm) Stele diameter (mm) Poles Root diameter (mm) Stele diameter (mm)
‘Sparkle’ 3 1·90 ± 0·18a 0·65 ± 0·07b 3–4 0·67 ± 0·02c 0·23 ± 0·01e
R50 3–5 2·01 ± 0·13a 0·69 ± 0·08b 2–3 0·54 ± 0·03d 0·18 ± 0·01f
Open in a new tab

Different lower-case letters indicate significant line-related differences (P < 0·001) between ‘Sparkle’ and R50.

The nodule

Gross anatomy

‘Sparkle’ mature nodules appeared as described in the literature (e.g. Bond, 1948; Newcomb et al., 1979). In short, as seen in a median longitudinal section (data not shown) as well as in a cleared mature nodule (Fig. 4A), the central infected zone comprising infected and non-infected cells was surrounded by a cortical zone, which was divided by the common nodule endodermis into outer and inner cortex. Two main vascular strands connected to the root vasculature were seen running within the inner cortex, parallel to the nodule epidermis (Fig. 4A). The differentiation of these strands was acropetal, i.e. from the base of the nodule to its apex, in contrast to the basipetal differentiation of the root vasculature.

Fig. 4.

Fig. 4.

Open in a new tab

Cleared nodules harvested from 17 DAI ‘Sparkle’ (A) and R50 (B–E) plants. (A) is considered to be a representative of the ‘Sparkle’ nodule; this nodule is used here to describe the organ gross anatomy. The cortex protects the central zone (CZ) which comprises infected and non-infected cells; it is traversed longitudinally by the common endodermis (En) which separates outer (triangle) from inner (diamond) cortex. Each vascular strand (white arrows) is connected to the root vasculature (RV); it possesses its own pericycle and vascular endodermis (not indicated). RC = Root cortex; MZ = meristematic zone. Nodules depicted in (B) to (E) illustrate the large diversity of size and shape observed in R50. The vascular traces (white arrows) were highly variable in R50 nodules with numbers of tracheary elements (TEs) ranging from 1 (E) to 18 (B). Refer to Materials and methods for assessment of the number of TEs. Scale bars = 250 µm.

When a reconstruction of the ‘Sparkle’ nodule was accomplished through serial transverse sectioning (Fig. 5A), the two main vascular strands appeared to branch dichotomously; thus, in cross-sections of a ‘Sparkle’ nodule, there were two vascular bundles at the base of the nodule, but four at mid-height and above (right of Fig. 5A). There were never more than four bundles observed at the apex of each nodule, suggesting that there was no more than one dichotomous branching per bundle. In a cross-section, the individual bundles were simple in their making (as described by Bond, 1948); each had its own vascular endodermis enclosing a pericycle layer surrounding a few small TEs and a few phloem cells (data not shown).

Fig. 5.

Fig. 5.

Open in a new tab

Schematic 3D-reconstruction of the vasculature of mature nodules of ‘Sparkle’ (A) and R50 (B) plants obtained via serial transverse sectioning. In the centre of the figure are schematics displaying the arrangement of the vascular bundles in the cross-sections taken at the relative positions indicated by the dashed lines. Only the position and the shape of the vascular bundles are comparable between ‘Sparkle’ and R50, not their sizes. N = nodule; NV = nodule vasculature; RV = root vasculature; R = root.

At 17 DAI, R50 nodules were smaller than those of ‘Sparkle’; for a meaningful comparison between pea lines, care had to be taken to section nodules of similar size. When the nodule size was thus taken into consideration, R50 nodules had an anatomy similar to those of ‘Sparkle’, with an infected central zone and a cortex including vascular traces in its inner region. However, R50 vasculature was more complex than that of ‘Sparkle’; the number of bundles within nodules was highly variable. The variability was most obvious in the three-dimensional reconstruction of the nodules where, most of the time, two main bundles existed but one very quickly bifurcated to give an additional main bundle (Fig. 5B). Furthermore, these main bundles did branch in a dichotomous manner but not necessarily at the same height (Fig. 5). Thus, overall, the vasculature of the nodule was not uniform in contrast to what was seen in the WT (compare B with A in Fig. 5); this lack of symmetry was evident in cross-sections where three to five vascular bundles were observed in the inner cortex of most nodules (left of Fig. 5B).

Morphology of cleared nodules

To confirm and to complement the previous findings, and to increase the sample size, cleared nodules were observed. WT at 17 DAI displayed a homogeneous population of nodules; these were mostly mature and nitrogen-fixing (indicated by their pink hue). Generally, cleared nodules of ‘Sparkle’ possessed two main strands (Fig. 4A); a few had three or four main vascular strands (Table 4) which had been missed via thin sectioning. By continually re-adjusting the fine focusing on the cleared nodule and by following the longest vascular strand along the nodule length, the number of TEs making that trace could be counted. The strands comprised, on average, 16 TEs positioned end-to-end (Table 5); however, this number was quite variable ranging from nine to 32. There was a positive correlation between the number of TEs and the length of the nodule (Fig. 6); longer nodules had strands with a greater number of TEs.

Table 4.

Percentage of nodules having 1, 2, 3, 4 or 5 traces of vasculature in ‘Sparkle’ (n = 53) and R50 (n = 71)

Trace ‘Sparkle’ R50
1 0 12·7
2 94·3 26·8
3 3·8 53·5
4 1·9 5·6
5 0 1·4
Open in a new tab
Table 5.

Mean measurements ( ± s.e.) from cleared nodules of ‘Sparkle’ (n = 53) and R50 (n = 71)

Parameter ‘Sparkle’ R50
Length (μm) 1039·5 ± 37·2 651·9 ± 15·5
Width (μm) 481·4 ± 14·1 401·3 ± 10·6
Number of TEs across nodule base 18·5 ± 0·6 14·6 ± 0·4
Number of TEs along one trace 16·5 ± 0·7 6 ± 0·3
Number of traces in one nodule 2·1 ± 0·04 2·6 ± 0·1
Open in a new tab

There is a statistically significant difference between ‘Sparkle’ and R50 for all parameters (P < 0·001).

Fig. 6.

Fig. 6.

Open in a new tab

Correlation between the number of TEs along a single nodule vascular strand and the nodule length, and lines of best fit for ‘Sparkle’ (open circles; n = 53) and R50 (closed circles; n = 71).

R50 nodules borne by 17 DAI plants were significantly smaller than ‘Sparkle’ nodules (Table 5) with a greater variability in size and shape than previously reported (Guinel and Sloetjes, 2000); few of these nodules were pink. With the clearing technique, the vasculature appeared especially variable (Table 4): some nodules had only a single strand while a few had up to five different main strands. However, most had three strands and further variation was observed within this population (Fig. 4B–E). Some of the strands appeared vestigial because they were made of very few TEs (Fig. 4E); others were well developed with up to 18 TEs lengthwise (Fig. 4B). However, on average, they were only six TEs long (Fig. 4C). In R50 too, there was a strong correlation between length of the nodule and number of TEs within a strand (Fig. 6), but because R50 nodules rarely reached sizes as large as those of ‘Sparkle’ nodules, they never displayed as many TEs. To determine whether or not there was a correlation between nodule length and nodule width, the width of the nodule was measured at its base, and the number of TEs which were in direct contact with the root vasculature was counted (Table 5). Both parameters were significantly different between WT and R50, but no correlation existed between these two parameters (data not show).

DISCUSSION

Variability in the vasculature of the underground organs of ‘Sparkle’ and the mutant R50

Primary root

The establishment of the vascular tissue of a primary root occurs under genetic control during embryogenesis; in an older plant, the pre-existing pattern is continuously perpetuated by the root meristematic activity (e.g. Raghavan, 1999). Thus, in a ‘Sparkle’ seedling, the radicle possesses three vascular poles ‘inherited’ from the embryo. As it develops into a primary root, the triarch pattern, described elsewhere for pea roots (e.g. Popham, 1955; Torrey, 1967), is propagated in the stele. Variations in this triarch pattern were seldom seen in ‘Sparkle’ primary roots; some variability was also reported by Bond (1948) and Popham (1955) in the pea cultivars they observed but not by Torrey and Wallace (1975). In contrast, variability was high in R50 primary roots which often had an increased vasculature in the MaZ. As R50 primary roots were similar to those of ‘Sparkle’ in the MeZ and displayed a triarch pattern, the increase of vasculature is likely to be a post-embryonic event. These alterations from the embryonic template are likely to be occurring via either the spontaneous appearance or the loss of a vascular pole, as had been observed previously in pea by Barlow (1984).

First-order lateral root

Contrary to what is traditionally reported in the literature (i.e. the primordium of a lateral root displays a pattern similar to that of the embryonic root; e.g. Raghavan, 1999), the stelar arrangement in the first-order lateral roots of both pea lines was different from that of their primary roots. In contrast to ‘Sparkle’ lateral roots which often had an increased vasculature compared with the primary root, those of R50 exhibited a decreased vasculature. The first-order lateral roots of ‘Sparkle’ exhibited more extensive variations in their vascular pattern than the primary roots because their number of vascular poles could range from two to five. ‘Sparkle’ variability was reminiscent of that observed by Torrey and Wallace (1975) who studied hundreds of pea root systems obtained through sub-cultures of a single root. These authors reported no obvious order in the arrangement of the vascular pattern seen in lateral roots which emerged at different locations along the main axis; accordingly, a young lateral root could display a diarch pattern whereas an older one could exhibit a pentarch pattern or vice versa (Torrey and Wallace, 1975). Additionally, they found significant differences in the stelar arrangement within a single lateral root through the addition or deletion of a strand. The first-order lateral roots of R50 were more homogeneous in their vascular pattern than those of ‘Sparkle’ as they displayed steles with two or three vascular poles (omitting one plant out of 29 which had four poles).

Diameter of root stele

Torrey (1957) demonstrated that the number of vascular poles could be altered through the manipulation of the growth conditions of excised pea root tips. By adding auxin, the subsequent new growth of those roots which previously displayed a triarch pattern had changed to exhibit a hexarch stele. When these roots were transferred back to an auxin-free medium, the subsequent new growth displayed the normal triarch pattern (Torrey, 1957). Torrey proposed that auxin influenced the procambium cylinder by making it larger; and to accommodate this enlargement the number of vascular poles increased. From these experiments, a linear correlation between the stele diameter and the number of vascular poles was proposed and is still cited in the literature (e.g. Aloni et al., 2006). In hindsight (see below), this relationship should perhaps not be applied indiscriminately to any type of root as it was drawn from excised roots. In support of this point, the first-order lateral roots of ‘Sparkle’ had a tetrarch arrangement compared with the primary roots, despite having smaller diameters (Table 3). Consequently, in the present analysis, we decided to separate the two types of roots (i.e. primary and first-order lateral roots). For the primary roots of ‘Sparkle’, a link between stele diameter and number of poles could not be verified because only a few plants displayed variations in their vasculature. On the contrary, the large variability in the vascular pattern of R50 primary roots allowed such a correlation to be tested; however, there was no linear relationship between stele diameter and number of poles. This suggests that the increase in poles was not in response to the presence of a larger stele. To test a correlation between stele diameter and number of poles in first-order lateral roots, the data for the two lines had to be grouped. When this was done, a relationship between the two parameters was evident; there were a larger number of poles in the roots with a larger stele diameter.

Nodule

‘Sparkle’ nodule vasculature with its two main strands was consistent with what has been described in the literature (Bond, 1948; Dart, 1977). However, contrarily to what is reported in the literature whereby up to ten strands could be counted in the apical region of a mature pea nodule (e.g. Bond, 1948), ‘Sparkle’ displayed only one dichotomy for each of its strands. It is possible that the difference lies in the age of the nodule, as Bond (1948) did not indicate a precise time for her observations. R50 nodules were unique in that they commonly had three vascular strands which were linked to the root vasculature; as far as is known, such abnormal nodule vasculature has never been reported in legume mutants. In the literature, two mutants of Lotus japonicus [alb1, Imaizumi-Anraku et al. (2000); crinkle, Tansengco et al. (2003)] have been shown to exhibit incompletely developed vascular bundles. These mutants differ from R50 in that their nodule vasculature is uniform; in contrast R50 displays large variations in the number of TEs present along an individual strand. In R50 as in ‘Sparkle’, the TEs number (along the strand length) and nodule length were strongly correlated (Fig. 6). However, because R50 nodules were heterogeneous in their size, their strand length was highly variable. This correlation could be a reflection of the nodule function, i.e. larger nodules would need longer vascular traces to accommodate the export of nitrogenous compounds. However, the present data do not agree with this explanation for a nodule such as that seen in Fig. 4C had a large central infected zone but little differentiation of vascular strands.

R50 root system and cytokinins

The R50 root system differs from that of ‘Sparkle’ by a shorter primary root, shorter lateral roots (Guinel and Sloetjes, 2000), a reduced number of first-order lateral root primordia and an abnormal vasculature (this study). Furthermore, R50 roots have a higher total CK concentration than those of ‘Sparkle’ (Ferguson et al., 2005). We postulate that these traits are linked to the accumulation of CKs. It could be a direct effect or more likely the result of an altered auxin : CK ratio (Aloni et al., 2006). Although it is likely to be a simplistic view (e.g. Nordström et al., 2004), it is thought that there are two opposite hormonal gradients along the root length. The auxin gradient is acropetal as a continuous polar flow of auxin moves from the aerial parts of the plant (e.g. Sachs, 1968) to the root tip which acts as a sink; the CK gradient is basipetal since CKs are synthesized in the root tip (Miyawaki et al., 2004) and travel apoplastically towards the root base (e.g. Aloni, 1995). Accordingly, it is the ratio of auxin : CK which determines the fate of the cells and influences the differentiation of the tissues imprinted in the embryo.

CK and the initiation of lateral root primordium

Some of the target cells responsive to the hormonal gradient are the progenitor cells of the lateral roots (i.e. the pericycle cells; Li et al., 2006); when triggered by the appropriate signals, these cells divide first anticlinally then periclinally to initiate the formation of a lateral root primordium (Raghavan, 1999). The spatial location where the lateral roots develop (Torrey, 1957; Stirk and van Staden, 2001), the primordium initiation (Li et al., 2006) and the root emergence (Hinchee and Rost, 1986; Raghavan, 1999) are all controlled tightly by auxin and CK. Specifically, CKs have been shown to inhibit lateral root growth. For example, CKX over-expressing plants (i.e. CK-deficient) exhibit an increased number of lateral roots (Werner et al., 2003; Lohar et al., 2004) whereas exogenous treatment of root systems by CKs reduces that number (Lorteau et al., 2001; Li et al., 2006). The lower number is caused by an inhibition of the primordium formation whereby CKs act through the regulation of the cell cycle as shown in arabidopsis (Li et al., 2006). In addition, a link between the number of primordia and CK perception (via CRE1 in combination with the shoot CK-receptor AHK3) has been drawn in arabidopsis (Li et al., 2006). Consequently, the accumulation of CKs could be the cause of the reduced number of lateral root primordia in R50.

CK and vasculature morphogenesis

CKs are also known to affect the development of the root vascular tissue; they are thought to increase the sensitivity of the tissue to auxin (Aloni, 1995) and thus these two hormones act together to specify the root vascular pattern (Raghavan, 1999; Aloni et al., 2006). That auxin is not the sole actor playing in the differentiation of vascular tissue and that it acts with CKs have been demonstrated elegantly using in vitro xylogenic cultures of Zinnia cells; if one of the two hormones is omitted from the medium, no TEs differentiate from the mesophyll cells (e.g. Pesquet et al., 2005). Other examples of the synergistic action of auxin and CK in the vasculature formation are found in several mutants. For example, the rice mutant ral1 exhibits fewer vascular poles in its roots than the WT seedlings and metaxylem elements with smaller diameters (Scarpella et al., 2003). In this mutant, it appears difficult to dissociate a hypersensitivity to CK and a defective auxin response (Scarpella et al., 2003). Another mutant, the pls arabidopsis mutant (Casson et al., 2002), displays a short root phenotype and a surprisingly reduced vascularization in its leaf. In this mutant, Casson et al. (2002) were unable to uncouple the effects of auxin from those of CK.

Recently, CKs alone have been placed at the centre stage of root vasculature morphogenesis. Our interpretation of the recent literature is that CK perception plays a larger role in this process than CK homeostasis. In arabidopsis transgenic plants over-expressing the CKX gene, there was no alteration in the root vascular pattern despite low levels of CK (Werner et al., 2003). However, other effects on root anatomy, such as an enlarged meristem, increased number of cells in some files, and occasional increase in cell size, were observed (Werner et al., 2003). In contrast, as demonstrated by Helariutta's group with arabidopsis mutants, appropriate CK signalling is essential for the correct maintenance of the root procambium. The WOL/CRE1/AHK4 root CK-receptor was shown to be required for controlling the number of cells composing the procambial root cylinder (Mahönen et al., 2000). The receptor is required for the specification of the protoxylem cells because in the roots of the wol mutant, all procambial cells differentiate into protoxylem cells (Mahönen et al., 2000). Recently, Mahönen et al. (2006) demonstrated that this differentiation event was the result of the effect of AHP6, a pseudo-phosphotransfer protein inhibiting the action of CK.

R50 is unique in that the alteration of its root vasculature is post-embryonic. The high CK levels measured in the R50 root system (Ferguson et al., 2005) must create an imbalance in the auxin : CK ratio. The effects of such an imbalance are likely to be observed anywhere along the length of the root. An optimal ratio of auxin : CK will provoke xylogenesis which requires the re-entry of some parenchyma cells of the mature vascular tissue into the cell cycle.

CK and nodules

It is known that CK regulates nodule number; however, this effect appears to be concentration-dependent with the existence of an optimum (Lorteau et al., 2001). This could explain seemingly contradictory results where exogenous CK-treatment (Lorteau et al., 2001) and CK-deficiency obtained through the over-expression of CKX (Lohar et al., 2004) both lead to a reduction in nodule formation. As far as is known, anomalous vasculature of nodules as seen in R50 has never been described. In view of what has been written above, regarding a relationship between abnormal vasculature and CK influences, we suggest that the increased vasculature observed in R50 nodules is also a result of the high levels of CK found in R50 (Ferguson et al., 2005).

Existence of heterorhizy in pea

The discrepancy of the vascular pattern in primary and first-order lateral roots of ‘Sparkle’ and R50 further supports the need for questioning the ‘a root is a root’ concept (Zobel, 2003). The present data add further evidence to the existence of heterorhizy demonstrated by Hochholdinger et al. (2004a, b) with mutants of Zea mays and Oryza sativa. If we are correct in our assumptions that R50 abnormal vasculature is caused by high levels of CKs, then R50 primary roots with their increased vasculature and R50 first-order lateral roots with their decreased vasculature are regulated differently, suggesting the existence of multiple components in the complex organ which is the root. Further evidence for separate controls for the two types of roots can be seen in the recent work of Riefler et al. (2006), whose results indicate that primary root growth and lateral root branching are under different regulation. Whereas the former process appears to be under the control of the root CK-receptor WOL/CRE1/AHK4, the latter process is regulated by a combination of the two shoot CK-receptors AHK2 and AHK3 (Riefler et al., 2006). The concept of ‘a root is a root’ has been long-lived, but it is obviously too simplistic to match the complexity of the hormonal interactions and the intricacy of molecular control. We hope that researchers in the future will take into account the sophistication of the root system and carefully distinguish the different types of roots.

Conclusions

In summary, in this study, the R50 phenotype has been refined by demonstrating that the mutant generally displays an additional number of vascular poles in its primary root, a smaller number of vascular poles in its first-order lateral roots, and an altered vasculature in its nodules. We trust that these three characteristics are linked to the high levels of CKs that the mutant accumulates over its development (Ferguson et al., 2005). In view of the literature, altered root vasculature is not caused simply by CK but rather by abnormal auxin/CK levels; furthermore, it appears to be more an effect of CK perception and signalling (Mahönen et al., 2000, 2006) than of CK metabolism (Werner et al., 2003; Takei et al., 2004). Because of the surprisingly strong similarity existing between the phenotype of R50 and those of the CK receptor-double and -triple mutants of arabidopsis (Riefler et al., 2006), we hypothesize that R50 is a CK-receptor mutant; however, at present the product of the gene sym16 remains unknown. Regardless of the sym16 identity, R50 should be considered as a useful tool for studying vasculature regulation, especially in the nodule.

ACKNOWLEDGEMENTS

The research was supported by the Natural Sciences and Engineering Research Council of Canada operating grants to FCG, and three Natural Sciences and Engineering Research Council of Canada Undergraduate Student Research Awards to A.N.P. The authors wish to thank Dr T. A. LaRue for his gift of R50 seeds, and Dr M. E. McCully for her critical reading of the manuscript.

LITERATURE CITED

  1. Aloni R. The induction of vascular tissues by auxin and cytokinin. In: Davies PJ, editor. Plant hormones: physiology, biochemistry and molecular biology. 2nd edn. Dordrecht: Kluwer; 1995. pp. 531–546. [Google Scholar]
  2. Aloni R, Aloni E, Langhans M, Ullrich CI. Role of cytokinin and auxin in shaping root architecture: regulating vascular differentiation, lateral root initiation, root apical dominance and root gravitropism. Annals of Botany. 2006;97:883–893. doi: 10.1093/aob/mcl027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Barlow PW. Positional controls in root development. In: Barlow PW, Carr DJ, editors. Positional controls in plant development. Cambridge: Cambridge University Press; 1984. pp. 281–318. [Google Scholar]
  4. Bond L. Origin and developmental morphology of root nodules of Pisum sativum. Botanical Gazette. 1948;109:411–434. [Google Scholar]
  5. Casson SA, Chilley PM, Topping JF, Evans IM, Souter MA, Lindsey K. The POLARIS gene of Arabidopsis encodes a predicted peptide required for correct root growth and leaf vascular patterning. The Plant Cell. 2002;14:1705–1721. doi: 10.1105/tpc.002618. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Dart P. Infection and development of leguminous nodules. In: Hardy RWF, Silver WS, editors. A treatise of dinitrogen fixation. Section III. Biology. New York, NY: John Wiley & Sons; 1977. pp. 367–471. [Google Scholar]
  7. Esau K. Anatomy of seed plants. 2nd edn. New York, NY: John Wiley; 1977. [Google Scholar]
  8. Eskew DL, Jiang Q, Caetano-Anollés G, Gresshoff PM. Kinetics of nodule development in Glycine soja. Plant Physiology. 1993;103:1139–1145. doi: 10.1104/pp.103.4.1139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Ferguson BJ. MSc Thesis. Waterloo, ON: University of Waterloo; 2002. The roles of cytokinins in nodulation and in R50, a putative cytokinin-overproducer of Pisum sativum cv. Sparkle. [Google Scholar]
  10. Ferguson BJ, Wiebe EM, Emery RJN, Guinel FC. Cytokinin accumulation and an altered ethylene response mediate the pleiotropic phenotype of the pea nodulation mutant R50 (sym16) Canadian Journal of Botany. 2005;83:989–1000. [Google Scholar]
  11. Fukuda H. Signals that control plant vascular cell differentiation. Nature Reviews – Molecular Cell Biology. 2004;5:379–391. doi: 10.1038/nrm1364. [DOI] [PubMed] [Google Scholar]
  12. Guinel FC, Sloetjes LL. Ethylene is involved in the nodulation phenotype of Pisum sativum R50 (sym16), a pleiotropic mutant that nodulates poorly and has pale green leaves. Journal of Experimental Botany. 2000;51:885–894. [PubMed] [Google Scholar]
  13. Hardham AR, McCully ME. Reprogramming of cells following wounding in pea (Pisum sativum L.) roots. I. Cell division and differentiation of new vascular elements. Protoplasma. 1982;112:143–151. [Google Scholar]
  14. Heyl A, Schmülling T. Cytokinin signal perception and transduction. Current Opinion in Plant Biology. 2003;6:480–488. doi: 10.1016/s1369-5266(03)00087-6. [DOI] [PubMed] [Google Scholar]
  15. Hinchee MAW, Rost TL. The control of lateral root development in cultured pea seedlings. I. The role of seedling organs and plant growth regulators. Botanical Gazette. 1986;147:137–147. [Google Scholar]
  16. Hochholdinger F, Park WJ, Sauer M, Woll K. From weeds to crops: genetic analysis of root development in cereals. Trends in Plant Science. 2004a;9:42–48. doi: 10.1016/j.tplants.2003.11.003. [DOI] [PubMed] [Google Scholar]
  17. Hochholdinger F, Woll K, Sauer M, Dembinsky D. Genetic dissection of root formation in maize (Zea mays) reveals root-type specific developmental programmes. Annals of Botany. 2004b;93:359–368. doi: 10.1093/aob/mch056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Imaizumi-Anraku H, Kouchi H, Syono K, Akao S, Kawaguchi M. Analysis of ENOD40 expression in alb1, a symbiotic mutant of Lotus japonicus that forms empty nodules with incompletely developed nodule vascular bundles. Molecular and General Genetics. 2000;264:402–410. doi: 10.1007/s004380000330. [DOI] [PubMed] [Google Scholar]
  19. Inoue T, Higuchi M, Hashimoto Y, Seki M, Kobayashi M, Kato T, et al. Identification of CRE1 as a cytokinin receptor from Arabidopsis. Nature. 2001;409:1060–1063. doi: 10.1038/35059117. [DOI] [PubMed] [Google Scholar]
  20. Junghans U, Polle A, Düchting P, Weiler E, Kuhlman B, Gruber F, et al. Adaptation to high salinity in poplar involves changes in xylem anatomy and auxin physiology. Plant, Cell and Environment. 2006;29:1519–1531. doi: 10.1111/j.1365-3040.2006.01529.x. [DOI] [PubMed] [Google Scholar]
  21. Kneen BE, Weeden NF, LaRue TA. Non-nodulating mutants of Pisum sativum (L.) cv. Sparkle. Journal of Heredity. 1994;85:129–133. [Google Scholar]
  22. Li X, Mo X, Shou H, Wu P. Cytokinin-mediated cell cycling arrest of pericycle founder cells in lateral root initiation of Arabidopsis. Plant Cell Physiology. 2006;47:1112–1123. doi: 10.1093/pcp/pcj082. [DOI] [PubMed] [Google Scholar]
  23. Lohar DP, Schaff JE, Laskey JG, Kieber JJ, Bilyeu KD, Bird DMcK. Cytokinins play opposite roles in lateral root formation, and nematode and rhizobial symbioses. The Plant Journal. 2004;38:203–214. doi: 10.1111/j.1365-313X.2004.02038.x. [DOI] [PubMed] [Google Scholar]
  24. Lorteau MA, Ferguson BJ, Guinel FC. Effects of cytokinin on ethylene production and nodulation in pea (Pisum sativum) cv. Sparkle. Physiologia Plantarum. 2001;112:421–428. doi: 10.1034/j.1399-3054.2001.1120316.x. [DOI] [PubMed] [Google Scholar]
  25. Lu P, Gladish D, Rost TL. Temperature-induced cavities and specialized parenchyma cells in the vascular cylinder of pea roots. American Journal of Botany. 1991;78:729–739. [Google Scholar]
  26. Luschnig C, Gaxiola RA, Grisafi P, Fink GR. EIR1, a root-specific protein involved in auxin transport, is required for gravitropism in Arabidopsis thaliana. Genes and Development. 1998;12:2175–2187. doi: 10.1101/gad.12.14.2175. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Mähönen AP, Bonke M, Kauppinen L, Riikonen M, Benfey PN, Helariutta Y. A novel two-component hybrid molecule regulates vascular morphogenesis of the Arabidopsis root. Genes and Development. 2000;14:2938–2943. doi: 10.1101/gad.189200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Mähönen AP, Bishopp A, Higuchi M, Nieminen KM, Kinoshita K, Törmäkangas K, et al. Cytokinin signaling and its inhibitor AHP6 regulate cell fate during vascular development. Science. 2006;311:94–98. doi: 10.1126/science.1118875. [DOI] [PubMed] [Google Scholar]
  29. Miyawaki K, Matsumoto-Kitano M, Kakimoto T. Expression of cytokinin biosynthetic isopentenyltransferase genes in Arabidopsis: tissue specificity and regulation by auxin, cytokinin and nitrate. The Plant Journal. 2004;37:128–138. doi: 10.1046/j.1365-313x.2003.01945.x. [DOI] [PubMed] [Google Scholar]
  30. Newcomb W, Sippel D, Peterson RL. The early morphogenesis of Glycine max and Pisum sativum root nodules. Canadian Journal of Botany. 1979;57:2603–2616. [Google Scholar]
  31. Nordström A, Tarkowski P, Tarkowska D, Norbaek R, Åstot C, Dolezal K, et al. Auxin regulation of cytokinin biosynthesis in Arabidopsis thaliana: a factor of potential importance for auxin-cytokinin-regulated development. Proceedings of the National Academy of Sciences of the USA. 2004;101:8039–8044. doi: 10.1073/pnas.0402504101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. O'Brien TP, McCully ME. Melbourne: Termarcarphi; 1981. The study of plant structure principles and selected methods. [Google Scholar]
  33. Pesquet E, Ranocha P, Legay S, Digonnet C, Barbier O, Pichon M, et al. Novel markers of xylogenesis in Zinnia are differentially regulated by auxin and cytokinin. Plant Physiology. 2005;139:1821–1839. doi: 10.1104/pp.105.064337. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Popham RA. Levels of tissue differentiation in primary roots of Pisum sativum. American Journal of Botany. 1955;42:529–540. [Google Scholar]
  35. Raghavan V. Root growth and development. In: Raghavan V, editor. Developmental biology of flowering plants. New York, NY: Springer-Verlag; 1999. pp. 105–141. [Google Scholar]
  36. Riefler M, Novak O, Strnad M, Schmülling T. Arabidopsis cytokinin receptor mutants reveal functions in shoot growth, leaf senescence, seed size, germination, root development, and cytokinin metabolism. The Plant Cell. 2006;18:40–54. doi: 10.1105/tpc.105.037796. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Rost TL, Jones TJ, Falk RH. Distribution and relationship of cell division and maturation events in Pisum sativum (Fabaceae) seedling roots. American Journal of Botany. 1988;75:1571–1583. [Google Scholar]
  38. Sachs T. The role of the root in the induction of xylem differentiation in peas. Annals of Botany. 1968;32:391–399. [Google Scholar]
  39. Scarpella E, Rueb S, Meijer AH. The RADICLELESS1 gene is required for vascular pattern formation in rice. Development. 2003;130:645–658. doi: 10.1242/dev.00243. [DOI] [PubMed] [Google Scholar]
  40. Scheres B, Di Laurenzio L, Willemsen V, Hauser MT, Janmaat K, Weisbeek P, et al. Mutations affecting the radial organization of the Arabidopsis root display specific defects throughout the embryonic axis. Development. 1995;121:53–62. [Google Scholar]
  41. Smith MM, McCully ME. A critical evaluation of the specificity of aniline blue induced fluorescence. Protoplasma. 1978;95:229–254. [Google Scholar]
  42. Stirk WA, van Staden J. The role of cytokinins in root development of pea seedlings. Plant Growth Regulation. 2001;34:173–180. [Google Scholar]
  43. Takei K, Ueda N, Aoki K, Kuromori T, Hirayama T, Shinozaki K, et al. AtIPT3 is a key determinant of nitrate-dependent cytokinin biosynthesis in Arabidopsis. Plant Cell Physiology. 2004;45:1053–1062. doi: 10.1093/pcp/pch119. [DOI] [PubMed] [Google Scholar]
  44. Tansengco ML, Hayashi M, Kawaguchi M, Imaizumi-Anraku H, Murooka Y. crinkle, a novel symbiotic mutant that affects the infection thread growth and alters the root hair, trichome, and seed development in Lotus japonicus. Plant Physiology. 2003;131:1054–1063. doi: 10.1104/pp.102.017020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Torrey JG. Auxin control of vascular pattern formation in regenerating pea root meristems grown in vitro. American Journal of Botany. 1957;44:859–870. [Google Scholar]
  46. Torrey JG. Root development. In: Torrey JG, editor. Development in flowering plants. New York, NY: MacMillan Press; 1967. pp. 78–93. [Google Scholar]
  47. Torrey JG, Wallace WD. Further studies on primary vascular tissue pattern formation in roots. In: Torrey JG, Clarkson DT, editors. The development and function of roots. London: Academic Press; 1975. pp. 91–103. [Google Scholar]
  48. Werner T, Motyka V, Laucou V, Smets R, Van Onckelen H, Schmülling T. Cytokinin-deficient transgenic Arabidopsis plants show multiple developmental alterations indicating opposite functions of cytokinins in the regulation of shoot and root meristem activity. The Plant Cell. 2003;15:2532–2550. doi: 10.1105/tpc.014928. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Zobel R. Fine roots – discarding flawed assumptions. New Phytologist. 2003;160:273–278. doi: 10.1046/j.1469-8137.2003.00893.x. [DOI] [PubMed] [Google Scholar]

Articles from Annals of Botany are provided here courtesy of Oxford University Press

RESOURCES