SEARCH

SEARCH BY CITATION

Keywords:

  • Lupinus albus;
  • citrate exudation;
  • citric acid;
  • phosphate uptake;
  • phosphorus;
  • proteoid roots;
  • root exudates;
  • white lupin

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

We investigated (1) the effect of constant and altered inorganic phosphate (Pi) supply (1–100 mmol m–3) on proteoid root production by white lupin (Lupinus albus L.); and (2) the variation in citrate efflux, enzyme activity and phosphate uptake along the proteoid root axis in solution culture. Proteoid root formation was greatest at Pi solution concentrations of 1–10 mmol m–3 and was suppressed at 25 mmol m–3 Pi and higher. Except at 1 mmol m–3 Pi, the formation of proteoid roots did not affect plant dry matter yields or shoot to root dry matter ratios, indicating that proteoid roots can form under conditions of adequate P supply and not at the expense of dry matter production. Plants with over 50% of the root system as proteoid roots had tissue P concentrations considered adequate for maximum growth, providing additional evidence that proteoid roots can form on P-sufficient plants. There was an inverse relationship between the Pi concentration in the youngest mature leaf and proteoid root formation. Citrate efflux and the activities of enzymes associated with citric acid synthesis (phosphoenolpyruvate carboxylase and malate dehydrogenase) varied along the proteoid root axis, being greatest in young proteoid rootlets of the 1–3 cm region from the root tip. Citrate release from the 0–1 and 5–9 cm regions of the proteoid root was only 7% (per unit root length) of that from the 1–3 cm segment. Electrical potential and 32Pi uptake measurements showed that Pi uptake was more uniform along the proteoid root than citrate efflux.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Proteoid roots are bottlebrush-like clusters of rootlets covered with a dense mat of root hairs that develop on lateral roots (Purnell 1960; Gardner, Parbery & Barber 1982a; Dinkelaker, Hengeler & Marschner 1995; Johnson, Vance & Allan 1996b). The formation of proteoid roots in response to P deficiency is well documented for white lupin (e.g. Gardner et al. 1982a; Marschner, Römheld & Cakmak 1987; Johnson et al. 1996b). The ability of proteoid roots of white lupin to exude large quantities of citrate or citric acid has been reported by several authors (Gardner, Barber & Parbery 1983; Dinkelaker, Römheld & Marschner 1989; Gerke, Römer & Jungk 1994; Johnson et al. 1996a). In soil, the major limiting steps in the acquisition of P by plants are the mobilization and diffusion of P to the roots (Barber, Walker & Vasey 1963; Ernst, Römheld & Marschner 1989). Citrate exuded into the rhizosphere enhances P availability to the plant by solubilizing P in the soil that is bound to Fe, Al or Ca (Gardner et al. 1983; Hoffland, Findenegg & Nelemans 1989; Fox, Comerford & McFee 1990), and this may explain the high productivity of white lupin on soils that have low plant-available P. For example, Gerke et al. (1994) found that the concentration of soluble phosphate (Pi) in the rhizosphere of proteoid roots of white lupin grown in two oxisols was twice that in the bulk soil despite depletion due to Pi uptake by the plant, and this was attributed to citrate exudation by proteoid roots. According to Marschner et al. (1987), the internal P concentration of the plant determines the initiation of proteoid roots rather than the Pi concentration in the substrate, and it has generally been assumed that proteoid roots form as a response to severe P deficiency. However, previous studies of proteoid root formation by white lupin have only compared plants with a luxury supply of P with plants that were severely P deficient, and did not assess proteoid root formation at Pi concentrations normally found in soils (Dinkelaker et al. 1995).

The exudation of citrate by proteoid roots of white lupin indicates that they possess a metabolic pathway for synthesizing large quantities of citrate from suitable precursors, as well as an efficient mechanism to transport citrate out of rootlets into the rhizosphere. This is supported by Johnson, Allan & Vance (1994) and Johnson et al. (1996b) who reported that the in vitro specific activity of phosphoenolpyruvate carboxylase (PEPC) was greater in proteoid roots than in non-proteoid roots. However, little is known about the mechanism of citrate exudation from proteoid roots. Agar techniques with pH indicators have shown strong acidification of the rhizosphere around proteoid roots of white lupin, suggesting that H+ is extruded with the citrate (Dinkelaker et al. 1989), but it is unclear whether other cations are also involved. Although previous studies of white lupin have measured citrate efflux from whole proteoid roots (Gardner et al. 1983; Dinkelaker et al. 1989; Johnson et al. 1996a), it is not known whether there is spatial variation in citrate efflux along the proteoid root axis. Such information is important for understanding the physiology of proteoid roots, and for identifying the region of the proteoid root most likely to express genes whose products enhance citrate efflux.

Studies on Banksia grandis Willd. (Malajczuk & Bowen 1974), Protea compacta R. Br. (Vorster & Jooste 1986a) and Leucadendron uliginosum R. Br. (Vorster & Jooste 1986b) showed that proteoid roots have enhanced Pi uptake compared with non-proteoid roots. This suggests that proteoid roots, in addition to exuding organic acids, have a greater capacity than non-proteoid roots for taking up Pi. However, we are not aware of any information in the literature comparing Pi uptake by proteoid and non-proteoid roots of white lupin, or whether Pi uptake varies along the proteoid root axis.

The objectives of this study with white lupin were to determine: (1) the effect of both constant and altered Pi supply on proteoid root formation in relation to dry matter production and P concentration in the plant; and (2) whether citrate efflux, the activity of enzymes associated with citric acid synthesis [PEPC and malate dehydrogenase (MDH)], and Pi uptake vary along the proteoid root axis.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Plant culture with constant and altered Pi supply

Constant Pi supply

Seeds of white lupin (Lupinus albus L. cv. Kiev Mutant) were sown in washed river sand in a glasshouse maintained at 22/15 °C on a 12/12 h day/night cycle. Seedlings were selected for uniformity at the two-leaf stage and transferred to large plastic vessels containing 50 dm3 of a continuously aerated minus-P basal nutrient solution with the following composition: 0·63 mol m–3 KNO3, 0·25 mol m–3 CaCl2, 0·25 mol m–3 MgSO4, 6 mmol m–3 FeCl3, 6 mmol m–3 Na2EDTA, 11 mmol m–3 H3BO3, 2 mmol m–3 MnCl2, 0·35 mmol m–3 ZnCl2, and 0·2 mmol m–3 CuCl2. Phosphate was added as KH2PO4 to provide 1, 5, 7·5, 10, 25, 50, 75, and 100 mmol m–3 Pi, and the pH adjusted to 6·0. Phosphate concentrations in the nutrient solutions were measured every second day over the first week after transplanting the seedlings, and thereafter once or twice daily (depending on the size of the plants) to maintain the nominated Pi concentrations. There were six replicate plants for each Pi treatment. All nutrient solutions were made up in deionized water, and were changed weekly to replenish the supply of basal nutrients.

Cotyledons were removed 7 d after transplanting to reduce the transfer of seed P reserves to seedlings. Preliminary work indicated that this hastened the formation of proteoid roots but had little effect on seedling development. Plants were harvested 50 d after starting the Pi treatments, by which time small pods were developing on the main stem inflorescence. Roots were rinsed in 0·1 mol m–3 K2SO4 followed by deionized water to remove Pi from the nutrient solution, and the plants were then separated into proteoid roots, non-proteoid roots, stem plus petioles, and leaflets of leaves. Proteoid roots were defined as those portions of secondary lateral roots bearing bottlebrush-like root clusters with a density of 10 or more rootlets per cm (Johnson et al. 1996a). Plant material was dried at 70 °C, weighed, ground to a fine powder and analysed for total P by X-ray fluorescence spectrometry (XRFS, Norrish & Hutton 1977).

Altered Pi supply

In this experiment, white lupin plants were grown under similar conditions as for plants with the constant Pi supply. Seedlings were transferred to the 50 dm3 containers of basal nutrient solution at the two-leaf stage, and KH2PO4 was added to provide either 1 or 100 mmol m–3 Pi. The nutrient solutions were monitored once or twice daily to maintain the nominated Pi concentrations, and all nutrient solutions were renewed weekly. There were six containers per Pi concentration, each supporting six uniform seedlings.

At 21 d after transplanting, the Pi concentration of three containers of plants grown at 1 mmol m–3 was increased to 100 (1[RIGHTWARDS ARROW]100) mmol m–3, and the Pi concentration of a further three containers of plants was decreased from 100 to 1 (100[RIGHTWARDS ARROW]1) mmol m–3. The remaining containers of plants were maintained at their previous Pi concentrations of either 1 or 100 mmol m–3 for the duration of the experiment. Thus, there were three containers of plants of each Pi treatment after the changes were made, giving a total of 18 plants per treatment. One plant was harvested from each container of the two initial Pi concentrations just before altering the solutions. Thereafter, one plant from each container of a particular treatment was harvested at 7, 15, 22, 29, and 37 d after the Pi concentrations had been altered. At the final harvest, plants had developing pods on the main stem inflorescence, and had commenced flowering on the first-order lateral branches.

At each harvest, the roots were rinsed as described above, and the plants separated into proteoid roots, non-proteoid roots, youngest mature leaf (YML; excluding the petiole), and rest of shoot. Any leaflets shed by a plant before it was harvested were collected and subsequently included with the shoot. The YML was defined as the youngest leaf which had fully unfolded leaflets; when the plants developed lateral branches, the YML was taken from the longest branch of each plant. The YML from each plant was placed in a small plastic bag immediately after removal and stored on ice until analysed for Pi. Pi in the YML was determined after extracting subsamples (50–65 mg) of fresh leaflet tissue with 5 kmol m–3 H2SO4, as described by Dowling & Bouma (1985). All other plant material was dried, weighed, ground and analysed for total P by XRFS.

Citrate efflux measurements

White lupin seeds were germinated in washed river sand, and two groups of uniform plants were transferred to either a plus-Pi or minus-Pi (+ Pi or – Pi) nutrient solution at the two-leaf stage. The composition of the basal nutrient solution was as described above, with KH2PO4 added to provide 250 mmol m–3 for the + Pi treatment. Each pot contained one plant in 2·5 dm3 of continuously aerated solution. Nutrient solutions were changed weekly and the pH adjusted to 6·0. The temperature in the glasshouse was maintained at 22/15 °C on a 16/8 h day/night cycle. Citrate efflux measurements were initiated when the plants were 42 d old.

To measure citrate efflux, plants were transferred to Perspex™ containers (23 × 36 × 14 cm) with two side chambers (13 × 9 × 5 cm) attached at a height of 7 cm from the base (Fig. 1). A nutrient solution of the same composition as used for growing the plants was added to each container up to the base level of the side chambers. A plant was placed near each side chamber so that all the roots were immersed in the nutrient solution. Two representative proteoid roots from the plant which conformed to a standard length (≈ 12 cm), and in which the youngest proteoid rootlets were in the 1–3 cm region (Fig. 2), were positioned horizontally in the side chambers. Small Perspex™ rings (internal diameter 20 mm) were placed at selected positions 0–1, 1–3, 3–5, 5–7, and 7–9 cm from the root tip along the proteoid root axis to create individual chambers (Fig. 1). Small notches on opposite sides of the rings enabled them to be placed directly over the roots without damaging the tissue, and the rings were sealed on to the roots with silicon vacuum grease. The nutrient solution (2 cm3) was added inside the rings, and then added to the container so that the level was just below the tops of the rings. At hourly intervals over 6 h, all the solution from each ring was removed for citrate assay and the wells were refilled with fresh nutrient solution for further measurements. Only root segments that showed approximately constant citrate efflux throughout the sampling period were selected to calculate the average citrate efflux. The calculation of citrate efflux from the proteoid roots took into account the length of the rootlets; this was measured by placing the proteoid roots on an illuminated mm scale using a magnification of 20 times. The same sampling technique was used to measure citrate efflux from non-proteoid roots (lateral roots without proteoid rootlets) of the + Pi and the – Pi treatments. Measurements for each root segment were carried out on 30 different roots (15 plants, two roots from each plant).

image

Figure 1. . Container with side chambers used for citrate efflux measurements. Proteoid root segments were enclosed in Perspex™ rings filled with solution to create chambers used for citrate efflux measurements.

Download figure to PowerPoint

image

Figure 2. . Schematic diagram of a representative experimental proteoid root of white lupin grown in – Pi nutrient solution. The 1–3 cm segment is the region of the youngest proteoid rootlets. Arrows show qualitatively the spatial variation in proteoid root function with respect to citrate efflux and Pi uptake, and the shading shows the relative enzyme [phosphoenolpyruvate carboxylase (PEPC) and malate dehydrogenase (MDH) combined] activity (2, 4see Tables 2, 4 and 5 for data).

Download figure to PowerPoint

To test whether microbial degradation of citrate had occurred during the sampling period, varying amounts of citric acid (1–10 mg dm–3) were added to rings enclosing proteoid root segments for which the average citrate efflux had been determined previously. The same amount of citric acid was also added to rings which were placed in the side chambers but did not enclose proteoid root segments. In all samples, the added citric acid was fully recovered (data not shown) after taking into account citrate efflux by the root, showing that microbial degradation had not occurred.

Extraction of root segments for assay of citrate

In some experiments, the proteoid root segments were removed from the Perspex™ rings, after being assayed for citrate efflux, and ground in a prechilled mortar and pestle with 0·4 cm3 ice-cold 0·6 kmol m–3 perchloric acid. The extract was centrifuged at 15 000g for 5 min, and 0·3 cm3 of the supernatant was collected and neutralized with 20 mm3 of 5 kmol m–3 K2CO3. The neutralized solution was centrifuged for 5 min at 15 000g, the supernatant diluted 10-fold and a 25 mm3 subsample assayed for citrate. Citrate in root exudates and root extracts was determined using an enzymatic method described by Dagley (1974) and modified by Delhaize, Ryan & Randall (1993).

Citrate efflux at different Pi solution concentrations

Citrate efflux from proteoid roots of white lupin was determined for plants grown at 0, 1, 2·5, 5·0, and 10·0 mmol m–3 Pi in 50 dm3 of nutrient solution. The experimental conditions and the maintenance of Pi concentrations were the same as for the solution culture experiment described above to examine plant growth at different Pi supplies. Citrate efflux from the 1–3 cm region of proteoid roots of 43-d-old plants was measured using the chamber technique described above. At each Pi concentration, six representative root segments were assayed from three proteoid roots from each of two plants.

PEPC, MDH and protein assays

The preparation of root segments for PEPC, MDH and protein assays was carried out using modifications of a method described by Macnicol & Jacobson (1992). Two representative proteoid roots from each of 15 plants grown without added Pi were assayed. The proteoid root segments were extracted by grinding them in a prechilled mortar and pestle with 0·4 cm3 of buffer [100 mol m–3 Hepes–KOH: pH 7·5, 1 mol m–3 MnCl2 and 10 mol m–3 dithiothreitol (DTT)]. The extract was centrifuged at 30 000g for 10 min, and aliquots of 50, 20 and 2 mm3 of supernatant were assayed for protein, PEPC and MDH, respectively. PEPC was assayed as described by Hatch & Oliver (1978), MDH by the method of Macnicol & Jacobsen (1992), and protein according to Bradford (1976).

Measurement of 32Pi uptake by root segments

The time course of 32Pi depletion from the nutrient solution was used to determine Pi uptake by segments of two proteoid and non-proteoid roots from each of five P-deficient plants grown in a – Pi nutrient solution, and two non-proteoid roots from five P-sufficient plants supplied with 50 mmol m–3 Pi. Perspex™ rings were placed along the proteoid and non-proteoid roots as described above, and 32Pi (660 Bq) was added to each ring in 2 cm3 of 50 mmol m–3 KH2PO4 solution. At 15 min intervals over a period of 150 min, 10 mm3 of solution was withdrawn from each chamber to measure the 32Pi that remained using a liquid scintillation counter (Beckman; model LS 6800). Preliminary experiments showed that Pi uptake by proteoid and non-proteoid roots supplied with 50 mmol m–3 Pi in the nutrient solution was linear, and that the relationship between Pi uptake and time was highly significant (P < 0·001). Therefore, the gradients of the 32Pi depletion curves were taken as the rate of Pi uptake by the root segments.

Electrophysiology

Pi-dependent changes in the electrical potential difference (Em) across membranes of root cells were measured in roots of six plants supplied with either 0 or 50 mmol m–3 Pi, and grown under the same conditions as for the 32P study. An entire lateral root was excised from each plant and held firmly with foam rubber in a narrow Perspex™ measuring chamber (6 cm3 volume). The root was left to stabilize for ≈ 40 min with a constant flow of Pi-free solution. The Pi-dependent depolarizations were measured in four different zones of lateral roots from the – Pi plants (see Results), and in the 0–1 and 1–3 cm zones from the tips of lateral roots from the 50 mmol m–3 plants.

Borosilicate glass electrodes were pulled on a Narishige puller (Narishige Scientific Instruments Ltd, Japan) and filled with 2 kmol m–3 KCl. Both the impaling electrode and a return electrode, which contained 2 kmol m–3 KCl in 2% agar, were connected to an A-M Systems Neuroprobe 1600 amplifier via Ag–AgCl half cells. The electrode was inserted into root tissue with a micromanipulator. The solution flow across the root was increased to ≈ 0·5 cm s–1 and held constant throughout the experiment. When Em had stabilized, a similar solution containing Pi was flowed into the chamber. Once the Pi-dependent change in Em had restabilized (within 2 min), the chamber was flushed with the Pi-free control solution. This procedure was repeated until each of six different Pi solutions (0·02, 0·1, 0·4, 1·0, 2·0, 5·0 mol m–3) had been replicated three times. Phosphate was added to the solutions as H3PO4 from a 1 kmol m–3 stock, and adjusted to pH 6·0 with 1 kmol m–3 NaOH. The sodium concentration was balanced with additions of 0·5 kmol m–3 Na2SO4 to all solutions except for the 5 mol m–3 Pi. Values of Vmax and Km were calculated by fitting the data to a Michaelis–Menten model using an iterative procedure to obtain the least squares solution.

Statistical analysis

An analysis of variance was performed on the data using GENSTAT 5 (Alvey, Galwey & Lane 1982).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Effect of Pi supply on proteoid root formation

Constant Pi supply

The large volume of the nutrient solution used and the daily or twice daily replenishment of the Pi concentrations provided the plants with relatively stable Pi supplies, even at the low Pi concentrations. Monitoring the Pi concentrations showed that they were depleted by only ≈ 8–18% daily for the treatments up to 25 mmol m–3 and by 1–8% for treatments above 25 mmol m–3. Proteoid root formation was greatest at the lower range of Pi supplies (1–10 mmol m–3), and was suppressed at Pi supplies of 25 mmol m–3 and higher (Table 1). Dry matter production was only reduced by the lowest Pi treatment (Fig. 3). Therefore, at Pi concentrations of 5–10 mmol m–3, the plants achieved maximum dry matter production, showed no symptoms of P deficiency, and yet produced the maximum proportion of proteoid roots. In addition, the shoot to root dry matter ratio was similar for all Pi concentrations supplied, regardless of the proportion of proteoid roots present, with the exception of the 1 mmol m–3 treatment (Fig. 3).

Table 1.  . Effect of Pi supply on the formation of proteoid roots (percentage of total root dry matter), P concentrations and P content of white lupin Thumbnail image of
image

Figure 3. . Effect of Pi concentration in the nutrient solution on shoot and root dry matter yields, and shoot to root dry matter ratios of white lupin. Each value is the mean of six replicates. The vertical bar represents the least significant difference for total plant dry matter at P = 0·05.

Download figure to PowerPoint

The concentration of total P in roots (both proteoid and non-proteoid), stems and leaves increased with increasing Pi supply, and showed the greatest increase at external Pi concentrations (10–25 mmol m–3) which coincided with the change from a root system dominated by proteoid roots to one largely composed of non-proteoid roots (Table 1). There were no significant (P = 0·05) differences in the P contents of plants grown on 50–100 mmol m–3 Pi, but the P contents of plants grown on 1–25 mmol m–3 Pi were lower (Table 1).

Altered Pi supply

Over the 37 d of the experiment, the proportional contribution that proteoid roots made to total root dry matter increased from 37 to 57% for the constant 1 mmol m–3 Pi plants, but remained at ≈ 8% for the constant 100 mmol m–3 Pi plants (Fig. 4a). The proportion of proteoid roots on the 1[RIGHTWARDS ARROW]100 mmol m–3 Pi plants decreased rapidly during the first 2 weeks after the change and then decreased more slowly, reaching the same proportion as the constant 100 mmol m–3 Pi plants by the end of the experiment. Most of the existing proteoid roots on the 1[RIGHTWARDS ARROW]100 mmol m–3 Pi plants died within the first 2 weeks after the change, and the new lateral roots that subsequently formed were usually devoid of proteoid roots. Plants from the 100[RIGHTWARDS ARROW]1 mmol m–3 Pi treatment produced prolific proteoid roots after an initial lag of ≈ 2 weeks and, by end of the experiment, the proportion of proteoid roots was similar to that of the constant 1 mmol m–3 Pi plants (Fig. 4a).

image

Figure 4. . Time course of the effect of constant and altered Pi supply on: (a) the proportions (%) of the root system as proteoid roots on a dry weight basis; (b) concentrations of inorganic P (Pi) in fresh tissue of leaflets from the youngest mature leaf (YML); (c) concentrations of total P in shoots; and (d) concentrations of total P in roots (proteoid and non-proteoid roots combined). The changes in Pi concentrations in the nutrient solution were made at 21 d after transplanting the seedlings into their initial Pi concentrations. Control plants grown on constant Pi concentrations of 1 and 100 mmol m–3 are shown as 1 and 100, respectively; plants changed from 1 to 100 and from 100 to 1 mmol m–3 Pi are shown as 1[RIGHTWARDS ARROW]100 and 100[RIGHTWARDS ARROW]1, respectively. The vertical bars represent least significant differences at P = 0·05.

Download figure to PowerPoint

Phosphate concentrations in the YML were inversely related to proteoid root formation, with Pi concentrations in the constant 1 mmol m–3 Pi plants remaining very low during the experiment (Fig. 4b). Conversely, Pi concentrations in the YML of the constant 100 mmol m–3 plants remained high throughout the experiment. Concentrations of Pi in the 100[RIGHTWARDS ARROW]1 mmol m–3 plants decreased rapidly in the first week after the change, and then gradually declined to the same concentration as in plants maintained at 1 mmol m–3 Pi. The 1[RIGHTWARDS ARROW]100 mmol m–3 Pi plants showed a large increase in their YML Pi concentration, which rose above that of the constant 100 mmol m–3 plants (Fig. 4b). These plants developed symptoms of P toxicity and shed some of their leaflets, whereas plants grown on the constant 100 mmol m–3 Pi supply showed no P toxicity symptoms. Total P concentrations in shoots followed similar patterns to Pi concentrations in the YML, although the increase in total P in shoots of the 1[RIGHTWARDS ARROW]100 mmol m–3 Pi plants was less rapid than the increase in the concentrations of Pi in the YML (Fig. 4c). Changes in concentrations of total P in roots reflected those of the shoot, except that total P concentrations in roots of the constant 100 mmol m–3 plants decreased rather than increased over time (Fig. 4d).

Proteoid root functions

Citrate efflux by different segments of the proteoid root axis

Citrate efflux was measured on proteoid roots ≈ 12 cm long from the – Pi plants and, except for the tip region (0–1 cm), these roots were densely covered with rootlets (40–70 rootlets cm–1; see Fig. 2). The morphology of the proteoid roots was similar to those grown in sand culture as described by Johnson et al. (1994). The greatest citrate efflux occurred from the proteoid root segment 1–3 cm from the root tip, whether expressed on a fresh weight, root length or protein basis (Table 2). There was no significant variation in citrate efflux along non-proteoid lateral roots of the – Pi plants or along the lateral roots of the 50 mmol m–3 Pi plants (data not shown), and the average effluxes were 0·68 ± 0·23 (standard error; n = 30) and 0·19 ± 0·07 (n = 30) nmol h–1 cm–1 root, respectively. Excluding the 1–3 cm segment, the average citrate efflux from the remainder of the proteoid root (0·58 nmol h–1 cm–1 root) was similar to the efflux from a non-proteoid root of a – Pi plant.

Table 2.  . Fresh weight (FW), protein concentration, and average citrate efflux over 6 h from different segments of proteoid roots of white lupin grown for 42 d in minus-Pi nutrient solution Thumbnail image of

The citrate concentration in proteoid roots ranged from 3·7 to 17·0 μmol g–1 fresh weight, with a mean value of 10·5 μmol g–1 fresh weight, comparable to the value of 7·2 μmol g–1 fresh weight found by Johnson et al. (1994). There was no significant correlation between the internal citrate concentration of proteoid root segments and the efflux of citrate from these segments (data not shown).

Citrate efflux from proteoid roots of plants grown at different Pi supplies

All measurements were made on the 1–3 cm segments of proteoid roots, where the highest citrate efflux occurred (see Table 2). Citrate efflux was ≈ 3 times greater in plants grown without Pi compared to plants supplied with some Pi, regardless of whether it was expressed on a fresh weight, root length or protein basis (Table 3). However, citrate efflux was the same (P < 0·05) in plants supplied with 1, 2·5, 5 and 10 mmol m–3 Pi.

Table 3.  . Effect of Pi supply on the citrate efflux from proteoid roots (1–3 cm root segment) of white lupin Thumbnail image of

The morphology of proteoid roots changed as the Pi supply increased due to a reduction in rootlet density. For example, the total length of rootlets in the 1–3 cm root segment of the – Pi plants was 18·1 cm compared with 14·8, 11·2 and 5·2 cm for plants grown on 1, 5 and 10 mmol m–3 Pi, respectively. Furthermore, the density of root hairs on the rootlets appeared to be reduced at higher Pi supplies.

PEPC and MDH activities in different segments of proteoid roots and in non-proteoid roots

As with citrate efflux, PEPC and MDH activities varied along the proteoid root axis of – Pi plants, with the greatest activities occurring in the 1–3 cm segment (Table 4). No significant differences were found in PEPC and MDH activities along the non-proteoid roots of these plants. The average PEPC and MDH activities of non-proteoid roots were 0·10 ± 0·04 (standard error) and 8·2 ± 2·3 (n = 30 for each segment) μmol NADH min–1 mg–1 protein, respectively, both of which were less than the values for the 1–3 cm segment of proteoid roots but comparable to the activities in the other proteoid root segments.

Table 4.  . Activity of phosphoenolpyruvate carboxylase (PEPC) and malate dehydrogenase (MDH) in different segments of proteoid roots of white lupin Thumbnail image of
Phosphate uptake by different root segments

In contrast to citrate efflux and enzyme activities, the highest Pi uptake capacity was not localized in the 1–3 cm segment of the proteoid root axis (Table 5). Uptake of 32Pi by proteoid roots of – Pi plants was greater at the root apex (0–1 cm) than in the other segments, whether expressed on either a fresh weight or a root length basis (Table 5). Phosphate uptake by the 0–1 cm region of non-proteoid roots of these plants was less than that of the same region of proteoid roots. Uptake of Pi per unit length by the 1–3 cm region of non-proteoid roots was not significantly different from that of the same region of proteoid roots, but it was greater when expressed on a fresh weight basis (Table 5).

Table 5.  . Rate of Pi uptake from different segments of proteoid and non-proteoid roots of white lupin Thumbnail image of

Because of differences in morphology between proteoid and non-proteoid roots, the effect of P deficiency on 32Pi uptake was assessed by comparing the rates of Pi uptake by non-proteoid roots (0–1 and 1–3 cm segments) from P-deficient (– Pi) and P-sufficient (50 mmol m–3 Pi) plants. Uptake of 32Pi by non-proteoid roots of the – Pi plants was ≈ 3-fold greater than by equivalent regions of roots from the 50 mmol m–3 Pi plants (Table 5).

Electrophysiology

Pi-dependent depolarizations were examined in four zones of the lateral root system of – Pi plants: (1) 0–1 and (2) 1–3 cm zones of non-proteoid roots; (3) the 0–1 cm zone of proteoid roots (no proteoid rootlets present); and (4) the 1–3 cm zone of proteoid roots (region of greatest citrate efflux from proteoid rootlets). Measurements were also made in the 0–1 and 1–3 cm zones of lateral roots of plants grown on 50 mmol m–3 Pi. The addition of Pi had no significant effect on the Em of roots from the 50 mmol m–3 Pi plants (< 2 mV), but in the – Pi plants, all four zones showed Pi-dependent changes in Em. The data from all root zones of the – Pi plants conformed well to a Michaelis–Menten model for transport kinetics where Vmax is expressed in millivolts (r = 0·81–0·89; Fig. 5). Pi-dependent depolarizations at the 0–1 cm and 1–3 cm zones of non-proteoid roots from the – Pi plants were not significantly different (Table 5). Similarly, measurements from the 0–1 cm and 1–3 cm zones of proteoid roots were not different. The Vmax estimates from the non-proteoid roots of the – Pi plants were ≈ 50% of the values estimated from the proteoid roots, but the Km values were not significantly different (Table 5).

image

Figure 5. . Effect of Pi concentration on the Pi-dependent changes in Em. Data show mean and standard error of six measurements, one from a representative root of each of six plants. The curves represent the least squares solution to the Michaelis–Menten model. (▵), lateral root from a P-sufficient plant (grown on 50 mmol m–3 Pi); (○), non-proteoid root from a P-deficient plant (no Pi supplied); (▵), main axis of a proteoid root from a P-deficient plant (0–1 cm); (○), young rootlet from a proteoid root of a P-deficient plant (1–3 cm).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Effect of constant and altered Pi supply on proteoid root formation and plant P status

Constant Pi supply

Our results are consistent with suggestions that internal P concentrations influence proteoid root formation in white lupin, and that proteoid root formation may represent a mechanism for avoiding P deficiency (Marschner et al. 1987). However, we have demonstrated for the first time that proteoid root formation is not necessarily a response to severe P deficiency, as previously proposed (Gardner, Parbery & Barber 1982b; Marschner et al. 1986; Louis, Racette & Torrey 1990; Gerke et al. 1994). Except for plants grown on 1 mmol m–3 Pi, the total P concentrations in leaves and stems of plants which produced a high proportion of proteoid roots were above the range (1·2–1·6 g P kg–1 dry weight) reported as inadequate or marginal for maximum growth of lupins (Gardner et al. 1982b; Moraghan 1993). The high proportion of proteoid roots (> 60%) formed in plants supplied with Pi concentrations of up to 10 mmol m–3 indicates that the formation of proteoid roots would be common in white lupin grown in most acidic and calcareous agricultural soils, where Pi concentrations in the soil solution are typically in the range of 5–30 mmol m–3 (Mengel & Kirkby 1982). Indeed, we observed that proteoid roots comprised 30–45% of the root system (despite losses in excavation) of plants sampled from 12 commercial white lupin crops in southern New South Wales, Australia, which had received recommended rates of P fertilizer (20–25 kg P ha–1). That the formation of proteoid roots by white lupin did not reduce either plant dry weight or shoot to root dry weight ratios, except at the lowest Pi concentration of 1 mmol m–3, indicates that the formation of proteoid roots and citrate exudation does not necessarily occur at the expense of dry matter production.

Altered Pi supply

It is clear from our results that the production of proteoid roots is a facultative response, because plants grown on a Pi supply sufficiently high to suppress proteoid root formation had the capacity to develop an extensive proteoid root system when they were transferred to a low Pi supply. Conversely, plants that had a prolific proteoid root system were capable of ceasing production of these roots when they were transferred to a Pi concentration that normally suppressed proteoid root formation. There was an inverse relationship between the concentration of P in the plant and the formation of proteoid roots. The most sensitive index was the concentration of Pi in the YML, as this responded to the changes in Pi supply more rapidly than total P in the shoot or roots. The start of prolific proteoid root production on plants changed from 100 to 1 mmol m–3 Pi occurred when the concentration of Pi in the YML had declined almost to the concentration in the YML of the constant 1 mmol m–3 plants. The very rapid rise in the YML Pi concentration of plants changed from 1 to 100 mmol m–3 Pi to well above that in the constant 100 mmol m–3 plants is consistent with other studies of P-deficient plants given access to an adequate P supply (e.g. Clarkson & Scattergood 1982; Cogliatti & Clarkson 1983). These results suggest that the internal Pi concentration in white lupin rather than the Pi concentration in the rooting medium determines the initiation of proteoid roots. This suggestion is supported by work showing that foliar application of Pi to white lupin (Marschner et al. 1987) and Myrica cerifera (Louis et al. 1990) increased the P concentration within the plant and decreased proteoid root formation.

Proteoid root functions

Citrate efflux and enzyme activities

We have shown that citrate efflux was greatest from the 1–3 cm segment of proteoid roots which had the youngest rootlets, whether expressed on a fresh weight, root length or protein content basis. However, citrate efflux from the remaining segments of the proteoid roots was similar to that from the non-proteoid lateral roots of the – Pi plants. Previous studies of citrate efflux from proteoid roots of white lupin have been based on the collection of leachates from plants grown in vermiculite (Gardner et al. 1983), soil (Dinkelaker et al. 1989) or sand (Johnson et al. 1996a). In these studies, either the possibility of microbial degradation was not considered or citrate efflux was determined for the entire root system or for whole proteoid roots rather than for particular regions of the proteoid root axis. The present study unequivocally demonstrates the spatial variation of citrate efflux along the proteoid root axis, and highlights the importance of assaying selected regions of proteoid roots to understand the mechanism of citrate efflux.

Growing plants without added Pi resulted in the greatest citrate efflux from proteoid roots (≈ 3-fold greater than in treatments where some Pi was supplied), suggesting an additional degree of citrate efflux regulation by plant P status beyond the initiation of proteoid root formation. Such regulation may occur at the level of citrate biosynthesis or transport of citrate out of the proteoid rootlets. Although citrate efflux from proteoid roots was significantly greater in plants grown without Pi compared to plants supplied with some Pi, there were no significant differences in citrate efflux from plants supplied with Pi concentrations of 1–10 mmol m–3. This shows that proteoid roots formed at P supplies considered adequate for white lupin are also capable of secreting citrate.

The PEPC and MDH assays demonstrated that the activities of these enzymes also vary considerably along the proteoid root axis. Consistent with the citrate efflux measurements, the highest activities were found in the younger segments of the proteoid root (Table 4). Johnson et al. (1994) showed an increase in the activities of PEPC (60%) and MDH (25%) on a protein basis in proteoid roots compared with non-proteoid roots of white lupin. However, our study showed even greater variation of PEPC and MDH between different segments of a proteoid root. For example, the 1–3 cm root segment had 118% greater PEPC and 134% greater MDH activity on a protein basis than the 7–9 cm segment. Such variation in PEPC and MDH activities along the proteoid root axis has not been previously reported. As noted above for citrate efflux, enzyme activities on a protein basis were considerably greater in the 1–3 cm segment of the proteoid root than the 0–1 cm segment which had the highest protein concentration. However, unlike citrate efflux, PEPC activity of the 3–5 cm segment was not significantly different from the 1–3 cm segment, indicating that greater citrate efflux is not necessarily a direct consequence of greater organic acid metabolism. Another indication that the biosynthesis of citrate is not the only factor regulating efflux is the observation that citrate concentration in a proteoid root segment did not correlate with its citrate efflux. Although the internal citrate concentration ranged from 3·7 to 17·0 μmol g–1 fresh weight, there was no correlation with citrate efflux in different segments of the proteoid root axis.

Phosphate uptake by proteoid root segments

Phosphate uptake into plant cells is thought to occur via secondary-active transport with protons (Ullrich-Eberius, Novacky & van Bel 1984). While there is debate concerning the stoichiometry between protons and Pi, the Pi-dependent depolarization reported in P-deficient plants indicates that uptake is associated with a transfer of net positive charge across the membrane (Dunlop & Hay 1985; Dunlop & Gardiner 1993). Recently, plant genes that encode phosphate transporters have been cloned (Muchhal, Pardo & Raghothama 1996; Leggewie, Willmitzer & Riesmeier 1997; Mitsukawa et al. 1997; Smith et al. 1997). Leggewie et al. (1997) showed that when a potato cDNA encoding a phosphate transporter was expressed in yeast, Pi uptake was dependent on the pH of the external medium and was inhibited by compounds that destroyed the pH gradient across the cell. Similar results were obtained by Mitsukawa et al. (1997) when they overexpressed a phosphate transporter gene in Arabidopsis. These observations are consistent with the hypothesis that Pi uptake in plants is mediated by secondary-active transport with protons.

Because the morphology of proteoid roots is so different from that of non-proteoid roots, the estimated rates of 32Pi uptake by the two types of roots could vary considerably depending on whether they were expressed on a fresh weight, length or surface area basis, thus making direct comparisons difficult. However, the morphology of non-proteoid roots from Pi-deficient and Pi-sufficient plants is the same, and 32P uptake by roots from the Pi-deficient plants was 3-fold greater than that by roots of the Pi-sufficient plants. This is consistent with previous studies of a number of plant species where higher maximum rates of Pi uptake by roots were associated with P deficiency (e.g. Lefebvre & Glass 1982; Lee, Ratcliff & Southon 1990; Bieleski & Läuchli 1992).

The use of electrophysiology to estimate Pi uptake allows comparisons to be made between proteoid and non-proteoid roots without the results being confounded by differences in root morphology, as with the 32P technique. The Pi-dependent changes in Em were at the limits of detection in P-sufficient plants, which is consistent with studies of other species (Ullrich-Eberius et al. 1984; Dunlop & Gardiner 1993). However, the Pi-dependent changes in Em we observed in both proteoid and non-proteoid roots of the – Pi plants suggest that electrogenic uptake of Pi was occurring. The estimated Vmax of the depolarizations measured in the non-proteoid roots was lower than the Vmax estimated at the apex of the proteoid roots (0–1 cm region), which is consistent with the 32P results (Table 5). In contrast, the affinities of the Pi uptake systems in proteoid and non-proteoid roots were not significantly different, with the average Km being ≈ 100 mmol m–3.

There were some differences in the results obtained with the two techniques for measuring Pi uptake. For instance, the 5-fold greater uptake of 32Pi by the 0–1 cm compared with the 1–3 cm region of the proteoid root (per unit FW), was not apparent with the electrophysiological measurements. The reasons for this discrepancy are unclear, but it may be that the isotopic technique is much more sensitive than electrophysiology to differences in the surface area of these two regions of the proteoid root. Despite their differences, both techniques show that: (1) proteoid and non-proteoid roots are capable of Pi uptake; and (2) the roots of P-deficient plants take up Pi faster than P-sufficient plants.

Concluding comments

We have shown for the first time that proteoid roots are formed by white lupin plants grown at Pi concentrations that are likely to occur in fertile agricultural soils. Perhaps more importantly, proteoid root formation and citrate exudation do not necessarily occur at the expense of dry matter production. In addition, white lupin is capable of changing its root system from one dominated by proteoid roots to one which has only a few proteoid roots, with no change in dry matter resource allocation between the shoot and the roots. The responses of white lupin to the changes in Pi concentrations in the nutrient solution show that well-established plants have the facultative capacity to either initiate or cease the formation of proteoid roots, and that this appears to be linked to the internal Pi concentration in the plant.

We have also shown that only the youngest segment (1–3 cm) of a proteoid root has significantly higher citrate efflux than the rest of the proteoid root or non-proteoid roots. Although citrate efflux varies 10-fold between the young and older regions of a proteoid root, rates of Pi uptake along the proteoid root axis are more uniform. The ability of the entire proteoid root to take up Pi can be viewed as a strategy to maximize the uptake of any P released by the action of the citrate exuded from the rootlets in the young region of the proteoid root. White lupin is non-mycorrhizal (Trinick 1977; Gardner et al. 1982a), and has a relatively coarse root system. The development of proteoid roots and the intensive mobilization of P from the relatively small volume of soil surrounding a proteoid root by citrate exudation appears to be an alternative strategy to mycorrhizal associations or the development of a fine fibrous root system to maximize Pi uptake from soils.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

We thank L. Mason and T. R. James for technical assistance.

Footnotes
  1. Present address: Joint FAO/IAEA Division, International Atomic Energy Agency, PO Box 100, A-1400 Vienna, Austria.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References
  • 1
    Alvey N., Galwey W., Lane P. (1982) An Introduction to Genstat. Academic Press, London.
  • 2
    Barber S.A., Walker J.M., Vasey E.H. (1963) Mechanisms for the movement of plant nutrients from soil and fertilizer to the plant root. Journal of Agricultural Food Chemistry 11, 204 207.
  • 3
    Bieleski R.L. & Läuchli A. (1992) Phosphate uptake, efflux and deficiency in the water fern, Azolla. Plant, Cell and Environment. 15, 665673.
  • 4
    Bradford M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Analytical Biochemistry 72, 248254.
  • 5
    Clarkson D.T. & Scattergood C.B. (1982) Growth and phosphate transport in barley and tomato plants during the development of, and recovery from, phosphate-stress. Journal of Experimental Botany 33, 865875.
  • 6
    Cogliatti D.H. & Clarkson D.T. (1983) Physiological changes in, and phosphate uptake by potato plants during development of, and recovery from phosphate deficiency. Physiologia Plantarum 58, 287294.
  • 7
    Dagley S. (1974) Citrate: UV spectrophotometric determination. In Methods of Enzymatic Analysis, Vol. 3 (ed. H.U. Bergmeyer), pp. 1562–1565. Academic Press, New York.
  • 8
    Delhaize E., Ryan P.R., Randall P.J. (1993) Aluminium tolerance in wheat (Triticum aestivum L) II. Aluminium-stimulated excretion of malic acid from root apices. Plant Physiology 103, 695702.
  • 9
    Dinkelaker B., Hengeler C., Marschner H. (1995) Distribution and function of proteoid roots and other root clusters. Botanica Acta 108, 183200.
  • 10
    Dinkelaker B., Römheld V., Marschner H. (1989) Citric acid excretion and precipitation of calcium citrate in the rhizosphere of white lupin (Lupinus albus L.). Plant, Cell & Environment 12, 285292.
  • 11
    Dowling E.J. & Bouma D. (1985) Field evaluation of a leaf test for assessment of the phosphorus status of subterranean clover and for prediction of its response to phosphorus. Australian Journal of Experimental Agriculture 25, 331336.
  • 12
    Dunlop J. & Gardiner S. (1993) Phosphate uptake, proton extrusion and membrane electropotentials of phosphorus-deficient Trifolium repens L. Journal of Experimental Botany 44, 18011808.
  • 13
    Dunlop J. & Hay M.J.M. (1985) Electrophysical evidence for the permeability of white clover stolons to nutrient ions. Plant and Soil 84, 153165.
  • 14
    Ernst M., Römheld V., Marschner H. (1989) Estimation of phosphorus uptake capacity zones of the primary root of soil-grown maize (Zea mays L.). Zeitschrift für Pflanzenernährung und Bodenkunde 152, 2125.
  • 15
    Fox T.R., Comerford N.B., McFee W. (1990) Phosphorus and aluminium release from a spodic horizon mediated by organic acids. Soil Science Society of America Journal 54, 17631767.
  • 16
    Gardner W.K., Barber D.A., Parberry D.G. (1983) The acquisition of phosphorus by Lupinus albus L. III. The probable mechanism by which phosphorus movement in the soil/root interface is enhanced. Plant and Soil 70, 107124.
  • 17
    Gardner W.K., Parbery D.G., Barber D.A. (1982a) The acquisition of phosphorus by Lupinus albus L. I. Some characteristics of the soil/root interface. Plant and Soil 68, 1932.
  • 18
    Gardner W.K., Parbery D.G., Barber D.A. (1982b) The acquisition of phosphorus by Lupinus albus L. II. The effect of varying phosphorus supply and soil type on some characteristics of the soil/root interface. Plant and Soil 68, 3341.
  • 19
    Gerke J., Römer W., Jungk A. (1994) The excretion of citric and malic acid by proteoid roots of Lupinus albus L., effects on soil solution concentrations of phosphate, iron, and aluminium in the proteoid rhizosphere in samples of an oxisol and a luvisol. Zeitschrift für Pflanzenernährung und Bodenkunde 157, 289294.
  • 20
    Hatch M.D. & Oliver I.R. (1978) Activation and inactivation of phosphoenolpyruvate carboxylase in leaf extracts of C4 species. Australian Journal of Plant Physiology 5, 571580.
  • 21
    Hoffland E., Findenegg G.R., Nelemans J.A. (1989) Solubilization of rock phosphate by rape. II. Local root exudation of organic acids as a response to P-starvation. Plant and Soil 113, 161165.
  • 22
    Johnson J.F., Allan D.L., Vance C.P. (1994) Phosphorus stress-induced proteoid roots show altered metabolism in Lupinus albus. Plant Physiology 104, 657665.
  • 23
    Johnson J.F., Allan D.L., Vance C.P., Weiblen G. (1996a) Root carbon dioxide fixation by phosphorus-deficient Lupinus albus. Contribution to organic acid exudation by proteoid roots. Plant Physiology 112, 1930.
  • 24
    Johnson J.F., Vance C.P., Allan D.L. (1996b) Phosphorus deficiency in Lupinus albus. Altered lateral root development and enhanced expression of phosphoenolpyruvate carboxylase. Plant Physiology 112, 3141.
  • 25
    Lee R.B., Ratcliff R.G., Southon T.E. (1990) 31P NMR measurements of the cytoplasmic and vacuolar Pi content of mature maize roots: relationship with phosphorus status and phosphate fluxes. Journal of Experimental Botany 41, 10631078.
  • 26
    Lefebvre D.D. & Glass A.D.M. (1982) Regulation of phosphate influx in barley roots: effects of phosphate deprivation and reduction of influx with provision of orthophosphate. Physiologia Plantarum 54, 199206.
  • 27
    Leggewie G., Willmitzer L., Riesmeier J.W. (1997) Two c-DNAs from potato are able to complement a phosphate uptake-deficient mutant: identification of phosphate transporters from higher plants. The Plant Cell 9, 381392.
  • 28
    Louis I., Racette S., Torrey J.G. (1990) Occurrence of cluster roots on Myrica cerifera L. (Myricaceae) in water culture in relation to phosphorus nutrition. New Phytologist 115, 311 317.
  • 29
    Macnicol P.K. & Jacobsen J.V. (1992) Endosperm acidification and related metabolic changes in the developing barley grain. Plant Physiology 98, 10981104.
  • 30
    Malajczuk N. & Bowen G.D. (1974) Proteoid roots are microbially induced. Nature 251, 316317.
  • 31
    Marschner H., Römheld V., Cakmak I. (1987) Root-induced changes of nutrient availability in the rhizosphere. Journal of Plant Nutrition 10, 11751184.
  • 32
    Marschner H., Römheld V., Horst W.J., Martin P. (1986) Root-induced changes in the rhizosphere: importance for the mineral nutrition of plants. Zeitschrift für Pflanzenernährung und Bodenkunde 149, 441456.
  • 33
    Mengel K. & Kirkby E.A. (1982) Principles of Plant Nutrition, 3rd edn, pp. 66–67. International Potash Institute, Worblaufen-Bern.
  • 34
    Mitsukawa N., Okumura S., Shirano Y., Sato S., Kato T., Harashima S., Shibata D. (1997) Overexpression of an Arabidopsis thaliana high-affinity phosphate transporter gene in tobacco cultured cells enhances cell growth under phosphate-limited conditions. Proceedings of the National Academy of Science, USA 94, 70987102.
  • 35
    Moraghan J.T. (1993) Phosphorus nutrition of white lupin grown on a calciquoll. Communications in Soil Science and Plant Analysis 24, 10771086.
  • 36
    Muchhal U.S., Pardo J.M., Raghothama K.G. (1996) Phosphate transporters from the higher plant Arabidopsis thaliana. Proceedings of the National Academy of Science, USA 93, 10519 10523.
  • 37
    Norrish K. & Hutton J.T. (1977) Plant analysis by X-ray spectroscopy. I. Low atomic number elements, sodium to calcium. X-ray Spectrometry 6, 611.
  • 38
    Purnell H.M. (1960) Studies of the family Proteaceae. I. Anatomy and morphology of the roots of some Victorian species. Australian Journal of Botany 8, 3850.
  • 39
    Smith F.W., Ealing P.M., Dong B., Delhaize E. (1997) The cloning of two Arabidopsis genes belonging to a phosphate transporter family. The Plant Journal 11, 8392.
  • 40
    Trinick M.J. (1977) Vesicular-arbuscular infection and soil phosphorus utilization in Lupinus spp. New Phytologist 78, 297304.
  • 41
    Ullrich-Eberius C.I., Novacky A., Van Bel A.J.E. (1984) Phosphate uptake in Lemna gibba G1: energetics and kinetics. Planta 161, 4652.
  • 42
    Vorster P.W. & Jooste J.H. (1986a) Potassium and phosphate absorption by excised ordinary and proteoid roots of the Proteaceae. South African Journal of Botany 52, 276281.
  • 43
    Vorster P.W. & Jooste J.H. (1986b) Translocation of potassium and phosphate from ordinary and proteoid roots to shoots in the Proteaceae. South African Journal of Botany 52, 281285.