Abstract

R50 is characterized as a pleiotropic pea mutant; it forms few nodules and has short lateral roots, short stature and pale leaves. Using grafting techniques, R50 paleness was found to be controlled by the shoot of the mutant whereas the nodulation phenotype was regulated by its root. The paleness of R50 is due to a lower than normal total chlorophyll content in its young leaves. The defect appears to be overcome with age because, as the plant matures, the chlorophyll levels increase in the older leaves. The reduction in stature is attributed to shorter internodes, and the oldest internodes are thicker than those of the parent Sparkle. Upon rhizobial inoculation, R50 forms as many infection threads as Sparkle. However, most of these are arrested in the inner cortex. The threads appear to have lost their directional growth towards the stele, and they coil around within enlarged cortical cells. In addition, very few infection threads are associated with divisions of the inner cortical cells. These aborted nodule primordia are abnormal, flat and mainly composed of cells which have divided anticlinally only. Nodulation of R50 was restored by treating the roots with ethylene inhibitors. The R50 mutant further supports the postulated role of ethylene in regulating rhizobial infection with a probable role in the control of the primordium development.

Introduction

R50 is a low nodulation mutant of pea (Pisum sativum L. cv. Sparkle) obtained by γ radiation (Kneen and LaRue, 1988) and genetic analysis showed that its abnormal nodulation was conditioned by a single recessive gene. This gene, designated sym16, is located on chromosome 5 (Kneen et al., 1994). The mutant was characterized as forming few or no nodules. In addition, it had a shorter and thicker epicotyl than that of its parent Sparkle, as well as pale green leaves. Its few laterals were described as being thicker than those of the parent (Kneen et al., 1994).

Several of the characteristics of R50 can be attributed to an overproduction of, or over‐sensitivity to, ethylene. Short stature and thick internodes are part of the triple response demonstrated in peas treated with exogenous ethylene by Neljubow (Ainscough et al., 1992). Exogenous ethylene also causes low nodulation (Lee and LaRue, 1992). Several pea mutants, such as the temperature‐sensitive mutant E2 (Fearn and LaRue, 1991; Guinel and LaRue, 1991) and E107, a mutant that accumulates large amounts of metal ions in its shoots and leaves (Guinel and LaRue, 1992), have had their nodulation restored by treating their roots with ethylene inhibitors.

In this study, the development, morphology and pigmentation of R50 are compared to those of the near‐isogenic normal parent Sparkle. In addition, the effects of grafting on the pigmentation and nodulation of R50 are reported. The developmental step at which nodulation is blocked and the effects of ethylene inhibitors on R50 nodulation phenotype are also described.

Materials and methods

Plant growth conditions

Seeds were surface‐sterilized with 8% household bleach (5.25% NaOCl) for 10 min, rinsed four times with sterile distilled water, left imbibing in the dark on wet filter paper in Petri plates for 2 d, and then transferred where appropriate (a to d below). Plants, except those used for nodulation studies (see below), were watered twice a week with a nutrient solution made of 2 mM KH2PO4, 2.5 mM Ca(NO3)2, 2 mM K2SO4, 1 mM MgSO4·H2O, FeIII EDTA (16.4 g l−1), with 1 ml l−1 micronutrient solution (50 mM KCl, 25 mM H3BO3, 2 mM ZnSO4·7H2O, 2 mM MnSO4·H2O, 0.5 mM CuSO4·5H2O, 0.5 mM Na2MoO4·2H2O), pH 6.8. If the plants required it, additional deionized water was given to them. The plants were grown in a light‐room under cool white fluorescent lights (intensity of minimum 145 to maximum 172 μE m−2 s−1) with a 22/18 °C, 16/8 h, light/dark regime. Plants were used in four different studies:

  • (a) For the study of early development, the seedlings were left in the dark in the Petri plates for an additional 6 d.

  • (b) For grafting, single seedlings were transferred to square (110 mm side, 650 ml volume) pots filled with a 1:1 mixture of sterile peat (Greenworld Garden Products, Pointe Sapin, NB) and Turface® (Applied Industrial Materials Corp., Buffalo Grove, IL).

  • (c) For the study of morphology and leaf pigmentation, seedlings were transferred to round (155 mm diameter, 1550 ml volume) pots filled with a substrate similar to that described above. There were 4 seedlings per pot.

  • (d) For the study of nodulation, seedlings were individually transferred to Dispobottles® (180 ml capacity, American Scientific Products, Rochester, NY) or to ‘conetainers’ (66 ml capacity, Stuewe and Sons, Inc., Corvallis, OR) filled with Holiday vermiculite (Vil Vermiculite Inc., Toronto, ON). All pots were wrapped in aluminium foil to minimize light in the rhizosphere.

Morphology

For early development, 7–10 seedlings (two trials) were grown and the following were recorded: time of the seed coat breakage, primary root length, epicotyl height, lateral root number, and length of the longest lateral root.

For the development of older plants, the first measurements were made on 8‐d‐old plants and thereafter every 3 or 4 d. At each time, at least 4 plants were harvested; the experiment was done twice. The length and dry weight (DW) of their primary root and shoot epicotyl, as well as the total number and DW of their lateral roots, were recorded. Epicotyl diameters of 8, 14 and 22‐d‐old plants [n=6 (1 trial), n=4 (3 trials), n=8 (3 trials), respectively; n is the sample size per trial] were estimated by taking transverse hand‐sections in the middle of each internode present at those ages. The diameter of each internode was estimated from prints made of these sections mounted on coverslips, set in a photographic enlarger, and directly printed onto photographic paper. To be consistent, the longest dimension of each section was measured because the internodes were not perfect circles. For internodes 1 and 2, hand‐sections adjacent to those used above were made, stained with toluidine blue O (pH 4.4), and photographed under 4× and 10× objectives (bright‐field) for estimating the number of cell layers and cell diameter, respectively. For cell diameter analysis, to ensure randomness of measurements, only the cells located below ten dots drawn randomly on an acetate sheet were measured. For each experiment, the measurement on cortical cells was done on three plants (totalling 30 cells); the experiment was done twice.

Shoot grafting experiments

Because of the difference in the thickness of their epicotyls, pea plants of different ages were used; R50 was 10‐d‐old, and Sparkle 6‐d‐old. The excision, straight and perpendicular to the epicotyl, was made with a sharp single‐edge sterile razor blade approximately 1–1.5 cm above the cotyledons. For Sparkle, this meant half‐way between the cotyledons and the first node, whereas for R50 it was half‐way between the first and second nodes. Reciprocal =Sparkle/R50 (SR) and R50/Sparkle (RS)] and control [R50/R50 (RR) and Sparkle/Sparkle (SS)] grafts were held together with the tips of disposable pipettes tailored to the appropriate stem diameter. The grafted plants were placed in small hothouses in the growth room. Each hothouse consisted of a plastic basin (29.5×26.5 cm) at the four corners of which tailored coat hangers were adjusted to hold a clear plastic sheath. Turface® saturated with sterile deionized water was laid at the bottom of the basin to ensure adequate moisture. Seven days after grafting, the pots were transferred from the hothouses to trays, set in the growth room, where they stayed for an additional 2 weeks. At all times, when present, side shoots were excised. Grafted plants were used either for evaluation of grafting effects (study b, n=6, 1 trial), for pigment extraction (study c, n=13, 3 trials), or for the nodulation study (study d, n=9, 3 trials). The effects of grafting were determined by comparing the total DW (shoot DW and root DW) of non‐grafted plants to that of control grafts.

Pigment extraction, qualitative and quantitative study

Leaves (0.5 g fresh weight) from non‐grafted (4th node at 14 d and 4th, 6th and 7th nodes at 22 d) and grafted (4th and 6th nodes at 22 d after grafting (dag)) plants were homogenized and their pigments extracted using 80% acetone (Arnon, 1949). For qualitative analysis, extracts were applied onto Whatman No.1 filter paper and chromatographed with petroleum ether:acetone (27:3, v:v) for 1 h. For quantitative analysis, absorbances of the samples were read at 470, 647, 663, and 710 nm wavelengths with a Novaspec II spectrophotometer (Pharmacia LKB). Three samples were taken and their average calculated; the average absorbance at 710 nm was then subtracted from that value to account for particle scattering. The final value was inserted into Lichtenthaler's equations (Lichtenthaler, 1987) to obtain chlorophyll a (chl a), chlorophyll b (chl b), and carotenoid concentrations, which were then converted to mg g−1 of fresh weight (FW). The pigment analysis experiment was done three times (total of at least 25 plants for the 4th and 6th nodes and at least 8 plants for the 7th node).

Nodulation experiments

Upon transfer to conetainers, seedlings were inoculated with 5 ml of a 2% dilution of Rhizobiumleguminosarum bv. viciae 128C53K (Nitragin® Inoculants, Liphatech Inc., Milwaukee, WI) grown in yeast–mannitol broth (absorbance of the bacterial culture read 0.550 at a wavelength of 600 nm). To determine the stage at which nodulation is blocked, 5 plants were harvested at 12, 14 and 18 d after inoculation (dai) and their third and 14th lateral roots hand‐sectioned. Fourteen dai was the best age to study nodulation arrest, therefore, the experiment was repeated twice at that age. Stages of nodule development were scored according to Guinel and LaRue (Guinel and LaRue, 1991). The number of infections at each stage was recorded, and the total number of infections was calculated per cm of lateral root. Some plants were harvested at 28 dai and nodules counted.

Ethylene inhibitors (10 μM of either aminoethoxy vinyl glycine (AVG), CoSO4·7H2O, or Ag2SO4) in low nitrogen nutrient solution (i.e., as above but with only 0.5 mM Ca(NO3)2) were applied in 20 ml aliquots onto the substrate 7 dai, and every 3–4 d after, until harvest at 28 dai. Nodules were counted on the treated root systems (n=6, 3 trials) at that time. In addition, at least five AVG‐ and silver‐treated plants had their third lateral root hand‐sectioned 14 dai, and stages of infection were recorded.

Grafted plants were inoculated on the day of grafting. Nodule number and nodule mass were measured after 28 d.

Statistical analysis

In all cases, Sparkle was used as a control. In the grafting experiments, two types of controls were used, non‐grafted plants and control grafts. t‐Values obtained from the data were compared to the critical t‐value using the t‐test calculated by the SigmaStat program (Jandel Scientific, San Rafael, CA). All plots were done on SigmaPlot® (Jandel Scientific, San Rafael, CA). In the tables, data were analysed at the 95% confidence level. In the figures, data were analysed at three different levels of confidence. Symbols, for example asterisks, were used to indicate the appropriate confidence level, either 99.9% (***), 99% (**) or 90% (*).

Results

Morphology

Early plant development:

The primary roots and epicotyls of R50 and Sparkle emerged through the seed coat 2 d after imbibing (d). The length of the primary roots was not significantly different (about 70 mm) over an 8 d period although R50 was always slightly smaller. The heights of their epicotyls were significantly different from emergence onwards, with R50 height 25% smaller than that of Sparkle at 8 d (20 mm versus 27 mm). The first lateral roots on Sparkle were visible at 6 d whereas on R50 they appeared 1 d later. From time of emergence onwards, R50 lateral roots were fewer compared to Sparkle (at 8 d, 8 and 19 laterals, respectively), and they were significantly shorter (data not shown). At 8 d, the longest lateral root on Sparkle was of an average length of 100 mm whereas that of R50 was 30 mm.

Late plant development:

The differences observed in the roots' early development remained the same throughout most of the development of the two pea lines (Fig. 1). The R50 root system was smaller with a significantly shorter primary root (Fig. 1A), a significantly smaller number of lateral roots (Fig. 1C) and, on average, a shorter longest lateral root (Fig. 1D). However, the primary root of R50 had a slightly greater DW than that of Sparkle (Fig. 1B), reflecting a larger diameter (data not shown).

At all ages, the epicotyl of R50 was significantly shorter than Sparkle (data not shown); at day 22, it was on average 35% shorter than that of Sparkle. The difference in shoot DW, although significant, was not as pronounced as for the root DW. The mutant thus appears stubby with a short but thick stature. The reduction in height was due to a length reduction of all internodes because the number of nodes was similar in plants of the same age. Thus, at 14 d, both peas had 4–5 nodes whereas at 22 d they had 7 or 8. At 22 d, internodes 1 and 2 of R50 were the only ones significantly thicker than those of Sparkle; internodes 3–5 were thicker, but not significantly so, and the younger internodes were thinner (data not shown). Internode 1 of R50 was found to be thicker at all ages measured with, on average, a diameter of 3.2 mm compared to that of 2.6 mm for Sparkle. The increase in R50 thickness was mainly attributed to a difference in the number of cortical cell layers; thus, at internode 1, R50 had on average 15 cell layers whereas Sparkle had only 12. The difference between the two pea types was not as pronounced at internode 2 (13 and 12 cell layers, respectively); however, it was still significant. The differences between the diameter of the cortical cells were not significant; on average, for both pea types, the cortical cells had a cross‐sectional diameter of 85 μm.

Fig. 1.

Late development: morphological parameters of R50 and Sparkle root systems. Primary root length (A) and DW (B) are shown, along with the number of lateral roots (C) and the length of the longest lateral root (D). Values are means ±SE (n=8, 2 trials). The asterisks indicate confidence levels, either 99.9% (***), 99% (**) or 90% (*).

Fig. 1.

Late development: morphological parameters of R50 and Sparkle root systems. Primary root length (A) and DW (B) are shown, along with the number of lateral roots (C) and the length of the longest lateral root (D). Values are means ±SE (n=8, 2 trials). The asterisks indicate confidence levels, either 99.9% (***), 99% (**) or 90% (*).

Pigmentation

In non‐grafted plants:

When young, R50 leaves were much paler than those of its parent. As the leaves mature, however, they became darker and greener. Thus, at 22 d, both pale (younger) and dark (mature) leaves were present on an R50 plant; although the basal leaves were indistinguishable from those of Sparkle, the leaves of young nodes were still very pale (Fig. 2A). The leaves borne by nodes 4 and 7, at 14 d and 22 d, were chosen as representatives of the two populations.

The chromatograms of both types of plants were similar (data not shown); no qualitative differences in the pigmentation of the two pea lines were discernible. Quantitatively, at 14 d, there was significantly less total chlorophyll (Tchl) in R50 leaves than in those of Sparkle (1.25 mg g−1 FW and 1.95 mg g−1 FW, respectively); at that age, node 4 was one of the youngest and it bore pale leaves. At 22 d, whereas Sparkle leaves on average had a Tchl content of 2.6 mg g−1 FW, those of R50 varied depending on the age of the leaves. While older leaves on nodes 4 and 6 had reached normal chl contents (2.45 and 2.7 mg g−1 FW, respectively), those from node 7 (younger) had significantly less Tchl (1.8 mg g−1 FW; Fig. 2B).

Fig. 2.

(A) R50 and Sparkle epicotyls (22 dap) shown next to each other to demonstrate the paleness of the young leaves of the mutant compared to either the younger leaves of the parent or the older leaves of the mutant. Note also the smaller size of the mutant. (B) Similar pigment extracts obtained from R50 and Sparkle to show the mutant's lower content of total chlorophyll.

Fig. 2.

(A) R50 and Sparkle epicotyls (22 dap) shown next to each other to demonstrate the paleness of the young leaves of the mutant compared to either the younger leaves of the parent or the older leaves of the mutant. Note also the smaller size of the mutant. (B) Similar pigment extracts obtained from R50 and Sparkle to show the mutant's lower content of total chlorophyll.

In grafted plants:

The grafting procedure impaired growth, resulting in a decrease of total DW. R50 was much more affected than Sparkle as the DW of the R50 control graft (RR) was 40% that of the non‐grafted plant whereas that of SS was more than 70%. It was thus difficult to compare the different plants at different ages and, as a consequence, only the Tchl of leaves from nodes 4 and 6 located above the graft line was measured. At 22 dag, the leaves borne by node 4 had matured and were green; Tchl was similar in all types of grafts (Fig. 3A) with an average concentration of 2.75 mg g−1 FW. There were differences in Tchl content in the leaves borne by node 6 (Fig. 3B); although all grafts having Sparkle for scions had greener leaves at that specific node (with an average Tchl concentration of 3.4 mg g−1 FW), grafts with R50 scions had pale leaves on similar nodes (Tchl concentration of about 2.1 mg g−1 FW). The pale phenotype of R50 is therefore shoot‐controlled.

Fig. 3.

Tchl (a+b) content of grafted plants (control grafts RR and SS, and reciprocal grafts RS and SR). Extractions were done from the 4th and 6th node above the graft line (A and B, respectively) to compare older and younger tissue, respectively. Values are means ±SE (n=13, 3 trials). Values with similar symbols were significantly different; the number of symbols indicate the confidence level [3 (***) means a 99.9% confidence level, 2 (** and oo) a 99% confidence level and 1 (*, o and +) a 90% confidence level.

Fig. 3.

Tchl (a+b) content of grafted plants (control grafts RR and SS, and reciprocal grafts RS and SR). Extractions were done from the 4th and 6th node above the graft line (A and B, respectively) to compare older and younger tissue, respectively. Values are means ±SE (n=13, 3 trials). Values with similar symbols were significantly different; the number of symbols indicate the confidence level [3 (***) means a 99.9% confidence level, 2 (** and oo) a 99% confidence level and 1 (*, o and +) a 90% confidence level.

Nodulation in non‐grafted plants

Nodule development:

At 28 dai, nodules were observed only rarely on R50 (Table 1, column 4). When they formed, nodules were small (Table 1, column 5) and white (i.e. not fixing nitrogen; data not shown). By that age, Sparkle had about 240 large pink nodules (Table 1, columns 4 and 5), mostly on the lateral roots. There were no significant differences between Sparkle and the mutant R50 in the total number of infections per cm of lateral roots (Table 1, column 3).

At 14 dai, infections were observed at all stages on the Sparkle root (Fig. 4A). In contrast, although numerous infection threads were found in the cortex of R50 roots (stage b, Fig. 4B), very few were associated with cortical cell divisions (stage c, Fig. 4B). Sparkle infection threads have already been described in the literature (Guinel and LaRue, 1991). Here, only the differences found between the mutant and the parent are emphasized. The infection threads in R50 roots were variable in diameter (Fig. 5A). Often, these threads were irregularly shaped with dark patches on their surfaces (Fig. 5B). A tuberculated surface was also observed on Sparkle, but only on threads that had aborted. One obvious difference was the intricate appearance of the threads formed in the mutant. These threads formed numerous loops in the cortex and the cells that were invaded seemed to have enlarged significantly (Fig. 5B, C). The threads, instead of growing towards the endodermis as in Sparkle, extended into the outer and inner cortical cells parallel to the surface of the root (Fig. 5B, C). A few inner cortical cells had divided forming primordia which appeared to be flat (Fig. 5D) unlike the dome‐shaped nodule meristems typical of pea. This peculiar shape is likely to be related to a predominance of anticlinal divisions in the inner cortex associated with an obvious lack of periclinal divisions (Fig. 5D).

Fig. 4.

Number of infections at 14 dai in control lateral root 3 for Sparkle (A) and R50 (B), and in AVG‐treated lateral 3 for Sparkle (C) and R50 (D). The infections were classified into six different categories: a, infection thread arrested either in root hair or in epidermal cell; b, infection thread arrested in cortex; c, infection thread in cortex associated with cell divisions; d, nodule primordium; e, emerging nodule; and f, mature nodule. The total number of infections is found in Table 1. Each panel shows the results obtained from at least five plants. Values are means ±SE of infections at each stage.

Number of infections at 14 dai in control lateral root 3 for Sparkle (A) and R50 (B), and in AVG‐treated lateral 3 for Sparkle (C) and R50 (D). The infections were classified into six different categories: a, infection thread arrested either in root hair or in epidermal cell; b, infection thread arrested in cortex; c, infection thread in cortex associated with cell divisions; d, nodule primordium; e, emerging nodule; and f, mature nodule. The total number of infections is found in Table 1. Each panel shows the results obtained from at least five plants. Values are means ±SE of infections at each stage.

Fig. 5.

Micrographs of nodulation events observed in R50; longitudinal sections of lateral root 3 stained with toluidine blue. Unless stated otherwise, all are from plants harvested 14 dai. On these micrographs, the root epidermis is always located on the left side. Note the larger size than usual of the cells (asterisks) containing the infection threads. (A) An infection thread in the inner cortex is seen invading two large neighbouring cells and branching abnormally. The diameter of the thread is not uniform, showing in some places constrictions (arrowheads) and in others expansions (arrows) (bar=50 μm). (B) An infection thread (28 dai) is seen forming loops in the cortical cells; note that the major direction of its growth is parallel to the surface of the root. Small dark patches, associated with aborted infection threads in Sparkle, are present on the surface of the thread (bar=50 μm). (C) Another infection thread in the cortex, growing parallel to the long axis of the cell and branching abnormally. The younger portions (arrows) of the thread do not stain with toluidine blue; this was also commonly seen in Sparkle. The cell containing the infection thread, or the neighbouring cell, has green‐stained protuberances (arrowheads) on its periclinal walls (bar=50 μm). (D) An elongated primordium (P) is forming; most of the divisions occurred anticlinally (arrows), i.e. perpendicular to the surface of the epidermis (Ep). The infection threads are spreading in the opposite direction. The domed shape associated usually with nodule primordium is not observed here (bar=200 μm).

Fig. 5.

Micrographs of nodulation events observed in R50; longitudinal sections of lateral root 3 stained with toluidine blue. Unless stated otherwise, all are from plants harvested 14 dai. On these micrographs, the root epidermis is always located on the left side. Note the larger size than usual of the cells (asterisks) containing the infection threads. (A) An infection thread in the inner cortex is seen invading two large neighbouring cells and branching abnormally. The diameter of the thread is not uniform, showing in some places constrictions (arrowheads) and in others expansions (arrows) (bar=50 μm). (B) An infection thread (28 dai) is seen forming loops in the cortical cells; note that the major direction of its growth is parallel to the surface of the root. Small dark patches, associated with aborted infection threads in Sparkle, are present on the surface of the thread (bar=50 μm). (C) Another infection thread in the cortex, growing parallel to the long axis of the cell and branching abnormally. The younger portions (arrows) of the thread do not stain with toluidine blue; this was also commonly seen in Sparkle. The cell containing the infection thread, or the neighbouring cell, has green‐stained protuberances (arrowheads) on its periclinal walls (bar=50 μm). (D) An elongated primordium (P) is forming; most of the divisions occurred anticlinally (arrows), i.e. perpendicular to the surface of the epidermis (Ep). The infection threads are spreading in the opposite direction. The domed shape associated usually with nodule primordium is not observed here (bar=200 μm).

Table 1.

Total number of infections cm−1 of the third lateral root, number of nodules and individual nodule mass on root systems non‐treated or treated with either ethylene inhibitors (AVG or cobalt), or ethylene antagonists (silver) at 28 dai

Each value represents a mean±SE (n=6, 3 trials). Although data from each pea line were compared between treatments, the statistics are not reported here because they are not as relevant as those concerning the comparison between non‐treated (control) and treated.

Treatment Pea line Total number of Nodule number Individual nodule 
  infections cm−1 per plant mass (mg) 

 

 
of lateral root
 

 

 
Control R50 10.6±1.7 11.9±2.3 c 0.38±0.07 c 
(non‐treated) Sparkle 14.7±3.1 239.5±26.10 c 1.93±0.19 c 
AVG R50 6.9±1.4b 144.5±10.2 a, b 1.42±0.20a, c 
 Sparkle 10.1±1.8 187.1±16.0 b 3.38±0.37 a, c 
Cobalt R50 ND d 63.3±7.0a, b, c 1.31±0.08 a, b, c 
 Sparkle ND d 169.4±11.5a, c 3.92±0.34 a, c 
Silver R50 5.2±1.6a, b 139.2±8.2 a, b 3.06±0.24 a, b 
 Sparkle 8.2±2.5 159.4±18.9 a 3.26±0.20 a 
Treatment Pea line Total number of Nodule number Individual nodule 
  infections cm−1 per plant mass (mg) 

 

 
of lateral root
 

 

 
Control R50 10.6±1.7 11.9±2.3 c 0.38±0.07 c 
(non‐treated) Sparkle 14.7±3.1 239.5±26.10 c 1.93±0.19 c 
AVG R50 6.9±1.4b 144.5±10.2 a, b 1.42±0.20a, c 
 Sparkle 10.1±1.8 187.1±16.0 b 3.38±0.37 a, c 
Cobalt R50 ND d 63.3±7.0a, b, c 1.31±0.08 a, b, c 
 Sparkle ND d 169.4±11.5a, c 3.92±0.34 a, c 
Silver R50 5.2±1.6a, b 139.2±8.2 a, b 3.06±0.24 a, b 
 Sparkle 8.2±2.5 159.4±18.9 a 3.26±0.20 a 

a Significantly different from control of the same pea line.

b Significantly different from control of the opposite pea line.

c Significantly different either between non‐treated lines or between lines for a single treatment.

d ND: not determined.

Effects of ethylene inhibitors on R50 nodulation:

Ethylene inhibitors did not alter total number of infections (Table 1, column 3), but at 28 dai they all significantly increased nodulation on R50 (Table 1, column 4). The most effective treatments were AVG and silver. These two chemicals restored nodulation of the mutant to numbers similar to those found in the wild‐type Sparkle. The treated plants (either R50 or Sparkle) had, however, heavier nodules than the respective non‐treated plants (Table 1, column 5). When individual nodule mass was measured, silver was the most effective antagonist. The nodules obtained after silver treatment of R50 roots were comparable in weight to those of treated Sparkle, whereas those obtained with either AVG or cobalt were less than half the size (Table 1, column 5).

Sections of AVG‐treated plants showed that the inner cortical cells that divide normally in the parent in preparation to the formation of the nodule primordium (stage d) are sensitive to ethylene because they are dividing after AVG treatment in 14‐d‐old R50 (compare stages c and d in Fig. 4B, D). In addition, in the AVG‐treated roots of R50, more of the infection threads progressed past the outer cortex (stage b).

Nodulation in grafted plants

Grafted plants have fewer nodules than ungrafted controls and these were smaller than those found on non‐grafted plants (Table 2). Nevertheless, nodules were only formed on grafts with Sparkle stocks; nodulation was inhibited on any grafts that had R50 stocks (Table 2). Thus the nodulation phenotype is root‐controlled.

Table 2.

Number and mass of nodules on control and grafted plants at 28 dai

Values are means ±SE (n=9, 3 trials except control with 2 trials). Control grafted plants were not compared to control non‐grafted plants of the opposite line (e.g. RR was not compared to Sparkle). The two reciprocal grafts were compared and were found to be significantly different in their number of nodules.


 
Number of nodules/plant
 
Individual nodule mass (mg)
 
R50 1.0±0.5 b 0 b 
Sparkle 70.3±5.2 a 2.09±0.13a 
RR 0.5±0.4 d 0 d 
SS 39.9±4.6 b, c 0.69±0.20 b, c 
RS 40.3±7.9 a, b, c 0.99±0.16 a, b, c 
SR 0.5±0.2 b, d 0b, d 

 
Number of nodules/plant
 
Individual nodule mass (mg)
 
R50 1.0±0.5 b 0 b 
Sparkle 70.3±5.2 a 2.09±0.13a 
RR 0.5±0.4 d 0 d 
SS 39.9±4.6 b, c 0.69±0.20 b, c 
RS 40.3±7.9 a, b, c 0.99±0.16 a, b, c 
SR 0.5±0.2 b, d 0b, d 

a Significantly different from R50.

b Significantly different from Sparkle.

c Significantly different from RR.

d Significantly different from SS.

Discussion

Characterization of R50 (sym 16)

Here, the pleiotropic pea mutant R50 (sym 16) with its few and thick laterals, short internodes, pale leaves and rare nodules was partially characterized. Pleiotropic mutants with altered nodulation have been described before (Fearn and LaRue, 1991; Markwei and LaRue, 1997). However, this is the first time, to our knowledge, that a nodulation mutant affected by a single recessive gene (Kneen et al., 1994) also has a defect in chlorophyll content. Whereas R50 nodulation is root‐controlled, its chlorophyll content and paleness are regulated by its shoot. Noteworthy is the fact that the pale phenotype of R50 is expressed even in plants grown with ample fixed nitrogen (data not shown); thus, the paleness is not due to nitrogen deficiency. Pale phenotype mutants, i.e. chlorophyll‐deficient mutants, have been described in several species. Such a phenotype is often associated with environmental stresses, especially with light and/or nitrogen availability (Lexa and Cheeseman, 1997), and with improper hormonal biosynthesis or action, especially cytokinins (Chory et al., 1994).

R50 nodulation phenotype is unique among those of low nodulators. The number of infections is normal, and the nodulation block occurs at stage d in our developmental classification (Guinel and LaRue, 1991), i.e. the stage at which cellular divisions initiate the nodule primordium. Because of an abnormal ratio of periclinal to anticlinal divisions, the primordium appears flat and elongated.

The cluster of inner cortical cells which marks the origin of the primordium appears to be regulated tigthly during nodule development. Its activation is controlled by gene expression (Bauer et al., 1996; Vernoud et al., 1999) and also by phytohormones (Bauer et al., 1996). Recently, new molecular markers for nodule primordium initiation have been found; these are specifically expressed in the cluster when the cortical cells begin to divide. In transgenic alfalfa, MsEnod 12A, a gene encoding proline‐rich putative cell wall protein, was found to be induced in the clusters after inoculation with Rhizobium meliloti (Bauer et al., 1996). This gene also responded to cytokinin treatment when plants were grown in the absence of nitrate (Bauer et al., 1996). MtENOD 20 is another marker of the dividing cortical cells (Vernoud et al., 1999); upon rhizobial inoculation, this gene which encodes an extracellular protein is first expressed in the cluster and later in infected root hairs.

In the mutant R50, sym 16 may be involved in the control of this cell cluster, with the nodule primordium unable to progress from anticlinal to periclinal divisions. Moreover, the aborted primordia may not be providing the signal to attract an infection thread. In this mutant, the threads seem to lack a ‘sense of direction’, and do not grow towards the stele.

Low nodulation mutants

In pea, several nodulation mutants have been isolated. For E107 (brz), a pleiotropic mutant that accumulates large amounts of metals in its shoot (Kneen et al., 1990), the nodulation block is found to be at the epidermis. Infection threads very rarely penetrate the outermost cortex (Guinel and LaRue, 1992). For E2 (sym 5), a temperature‐sensitive mutant (Fearn and LaRue, 1991), there are two blocks, the major one just before the inner cortical cells divide to form the nodule primordium (Guinel and LaRue, 1991). Another pleiotropic mutant, E132 (sym 21) is blocked later; nodule meristems form but the nodules rarely emerge (Markwei and LaRue, 1997).

If the development of a nodule were to be dissected and all the symbiosis genes of pea studied so far placed on a spectrum spanning from bacterial recognition (Markwei and LaRue, 1992), through bacterial entry and infection thread formation (stage a of Guinel and LaRue, 1991) to mature nodule (stage f of Guinel and LaRue, 1991), R50 (sym 16) would be found just after E2 (sym 5), but much before E132 (sym 21) (Fig. 6).

Fig. 6.

Schematic representation of the nodulation pea mutants in our collection and the location of their putative defect.

Fig. 6.

Schematic representation of the nodulation pea mutants in our collection and the location of their putative defect.

Ethylene, a negative regulator of nodulation

What is interesting about most of the mutants mentioned above, including R50, is their response to ethylene inhibitors and antagonists. When these mutants are treated with ethylene inhibitors, nodulation is partially or entirely restored. From the data on R50, and those on E2 (Guinel and LaRue, 1991), it appears that stage d in nodule development is the most sensitive to ethylene inhibitors; mutants treated with AVG or silver produce numbers of nodules close to those of the parent. Stage a, especially when the thread has already reached the basal epidermal cells, is another step of nodule organogenesis quite sensitive to ethylene inhibitors because R50 (this study; compare stages a in Fig. 4B, 4D), E2 (Guinel and LaRue, 1991) and E107 (Guinel and LaRue, 1992), when treated with AVG or silver, show much less infection thread arrest in the epidermis. However, the number of nodules obtained on treated E107 was much less than that of the control Sparkle (Guinel and LaRue, 1992). Lee and LaRue were also able to demonstrate that this step in nodule development was quite sensitive to this hormone by treating Sparkle, a wild‐type plant, with exogenous ethylene (Lee and LaRue, 1992). On the contrary, the very early and late stages of the nodule development do not appear to be very sensitive to either inhibitors or antagonists of ethylene. Neither R25 nor R72 (described in Markwei and LaRue, 1992) had their nodulation restored when treated with silver (T LaRue, personal communication). The number of nodules on E132 increased slightly when the roots were treated with cobalt or silver, but nodulation was not improved by AVG treatment (Markwei and LaRue, 1997).

The data from this study, i.e. that nodule formation is sensitive to ethylene, are in agreement with those of other authors (Grobbelaar et al., 1971; Goodlass and Smith, 1979; Peters and Crist‐Estes, 1989; Lee and LaRue, 1992). Whereas several authors (Grobbelaar et al., 1971; Goodlass and Smith, 1979; Lee and LaRue, 1992) demonstrated that exogenous levels reduced nodule number and inhibited growth of the infection threads in the cortex in pea, Peters and Crist‐Estes showed that AVG treatments of alfalfa roots resulted in the stimulation of nodulation (Peters and Crist‐Estes, 1989). This inhibition may not occur for all legume species however, because soybean nodulation does not appear to be inhibited by ethylene (Lee and LaRue, 1992; Schmidt et al., 1999).

Numerous studies published recently suggest that ethylene can play several roles in the regulation of the development of indeterminate nodules, but with no obvious major roles. Thus, it has been postulated that, in pea, ethylene is possibly a negative regulator of primordium formation (Heidstra et al., 1997). In a study on sickle, a Medicago truncatula mutant with a hyperinfection phenotype and an insensitivity to ethylene, it was proposed that ethylene could be important in controlling the persistence of rhizobial infection and act subsequently to the initiation of the infection (Penmetsa and Cook, 1997). Finally, in a study performed on Sesbania rostrata, evidence was given to show that ethylene could be involved in the type (determinate versus indeterminate) of nodule a plant root forms (Fernández‐López et al., 1998). Sesbania rostrata shows plasticity in nodule development depending on growth conditions (Fernández‐López et al., 1998); ethylene appears to play a role in the switch of its nodule type by influencing the nodule meristem persistence. Ethylene thus is a major actor in nodule regulation in several legumes; in others, as in soybean, it may not play a predominant role (Schmidt et al., 1999). The number of ethylene‐sensitive pea mutants now available shows that the role of ethylene is much more complex than originally thought, and that ethylene inhibition occurs after infection but before the thread has reached the nodule primordium.

Conclusions

R50 is a mutant with a novel phenotype. R50 pleiotropy is evident in that the nodulation phenotype is controlled by the root whereas the shoot phenotype, characterized by a lower chlorophyll content, is regulated by the shoot. In addition, its short stature, poor root system and low nodulation, which can be restored by AVG treatment, are all evidence for an involvement of ethylene in its phenotype. The location of the nodulation block (cortical cell activation, to use the term of Vernoud et al., 1999) makes R50 unique and useful in the continued study of nodule development. R50 strengthens the case of using pea as a model system since mutants of this species have now been obtained for several steps of nodule morphogenesis (Fig. 6) with new ones still being uncovered.

This research was supported by a NSERC operating grant to FCG. Additional funding was provided by Wilfrid Laurier University. We would like to thank Drs TA LaRue and AM Hirsch for critical and helpful comments, and S Hyam for initiating the nodulation study of R50. Thanks also to R Geil for his help with the colour photographs.

References

Ainscough EW, Brodie AM, Wallace AL.
1992
. Ethylene—an unusual plant hormone.
Journal of Chemical Education
 
69
,
315
–318.
Arnon DI.
1949
. Copper enzymes in isolated chloroplasts: polyphenoloxidase in Beta vulgaris.
Plant Physiology
 
24
,
1
–15.
Bauer P, Ratet P, Crespi MD, Schultze M, Kondorosi A.
1996
. Nod factors and cytokinins induce similar cortical cell division, amyloplast deposition and MsEnod 12A expression patterns in alfalfa roots.
The Plant Journal
 
10
,
91
–105.
Chory J, Reinecke D, Sim S, Washburn T, Brenner M.
1994
. A role for cytokinins in de‐etiolation in Arabidopsis. det mutants have an altered response to cytokinins.
Plant Physiology
 
104
,
339
–347.
Fearn JC, LaRue TA.
1991
. Ethylene inhibitors restore nodulation to sym5 mutants of Pisum sativum L. cv. Sparkle.
Plant Physiology
 
96
,
239
–244.
Fernández‐López M, Goormachtig S, Gao M, D'Haeze W, Van Montagu M, Holsters M.
1998
. Ethylene‐mediated phenotypic plasticity in root nodule development on Sesbania rostrata.
Proceedings of the National Academy of Sciences, USA
 
95
,
12724
–12728.
Goodlass G, Smith KA.
1979
. Effect of ethylene on root extension and nodulation of pea (Pisum sativum L.) and white clover (Trifolium repens L.).
Plant and Soil
 
51
,
387
–395.
Grobbelaar N, Clarke B, Hough MC.
1971
. The nodulation and nitrogen fixation of isolated roots of Phaseolus vulgaris L. III. The effect of carbon dioxide and ethylene.
Plant and Soil
  (Special volume)
215
–223.
Guinel FC, LaRue TA.
1991
. Light microscopy study of nodule initiation in Pisum sativum L. cv. Sparkle and its low‐nodulating mutant E2 (sym 5).
Plant Physiology
 
97
,
1206
–1211.
Guinel FC, LaRue TA.
1992
. Ethylene inhibitors partly restore nodulation to pea mutant E107 (brz).
Plant Physiology
 
99
,
515
–518.
Heidstra R, Yang WC, Yalcin Y, Peck S, Emons AM, van Kammen A, Bisseling T.
1997
. Ethylene provides positional information on cortical cell division but is not involved in Nod factor‐induced root hair tip growth in Rhizobium–legume interaction.
Development
 
124
,
1781
–1787.
Kneen BE, LaRue TA.
1988
. Induced symbiosis mutants of pea (Pisum sativum) and sweetclover (Melilotus alba annua).
Plant Science
 
58
,
177
–182.
Kneen BE, LaRue TA, Welch RM, Weeden NF.
1990
. Pleiotropic effects of brz. A mutation in Pisum sativum (L.) cv ‘Sparkle’ conditioning decreased nodulation and increased iron uptake and leaf necrosis.
Plant Physiology
 
93
,
717
–722.
Kneen BE, Weeden NF, LaRue TA.
1994
. Non‐nodulating mutants of Pisum sativum (L.) cv. Sparkle.
Journal of Heredity
 
85
,
129
–133.
Lee KH, LaRue TA.
1992
. Exogenous ethylene inhibits nodulation of Pisum sativum L. cv Sparkle.
Plant Physiology
 
100
,
1759
–1763.
Lexa M, Cheeseman JM.
1997
. Growth and nitrogen relations in reciprocal grafts of wild‐type and nitrate reductase‐deficient mutants of pea (Pisum sativum L. var. Juneau).
Journal of Experimental Botany
 
48
,
1241
–1250.
Lichtenthaler HK.
1987
. Chlorophylls and carotenoids: pigments of photosynthetic biomembranes.
Methods of Enzymology
 
148
,
350
–382.
Markwei CM, LaRue TA.
1992
. Phenotypic characterization of sym 8 and sym 9, two genes conditioning non‐nodulation in Pisum sativum ‘Sparkle’.
Canadian Journal of Botany
 
38
,
548
–554.
Markwei CM, LaRue TA.
1997
. Phenotypic characterization of sym 21, a gene conditioning shoot‐controlled inhibition of nodulation in Pisum sativum cv. Sparkle.
Physiology Plantarum
 
100
,
927
–932.
Penmetsa RV, Cook DR.
1997
. A legume ethylene‐insensitive mutant hyperinfected by its rhizobial symbiont.
Science
 
275
,
527
–530.
Peters NK, Crist‐Estes DK.
1989
. Nodule formation is stimulated by the ethylene inhibitor aminoethoxyvinylglycine.
Plant Physiology
 
91
,
690
–693.
Schmidt JS, Harper JE, Hoffman TK, Bent AF.
1999
. Regulation of soybean nodulation independent of ethylene signaling.
Plant Physiology
 
119
,
951
–959.
Vernoud V, Journet EP, Barker DG.
1999
. MtENOD 20, a Nod factor‐inducible molecular marker for root cortical cell activation.
Molecular Plant–Microbe Interactions
 
12
,
604
–614.

Comments

0 Comments