Journal of Experimental Botany, Vol. 57, No. 2, pp. 413–423, 2006
Phenotypic Plasticity and the Changing Environment Special Issue
doi:10.1093/jxb/erj004 Advance Access publication 15 December, 2005
Systemic suppression of cluster-root formation and net
P-uptake rates in Grevillea crithmifolia at elevated
P supply: a proteacean with resistance for developing
symptoms of ‘P toxicity’
Michael W. Shane* and Hans Lambers
School of Plant Biology, Faculty of Natural and Agricultural Sciences, The University of Western Australia,
Crawley, WA 6009, Australia
Received 31 May 2005; Accepted 8 September 2005
Abstract
Grevillea crithmifolia R. Br. is a species of Proteaceae
that is resistant to developing P-toxicity symptoms at
phosphorus supplies in the root environment that induce P-toxicity symptoms in the closely related Hakea
prostrata (Proteaceae). It was discovered previously
that development of P-toxicity symptoms in H. prostrata
is related to its low capacity to down-regulate net
P-uptake rates (i.e. its low plasticity). The plasticity of
net P-uptake rates and whole-plant growth responses in
G. crithmifolia has now been assessed in two separate
experiments: (i) a range of P, from 0 to 200 lmol P d21,
was supplied to whole root systems; (ii) using a splitroot design, one root half was supplied with 0, 3, 75, or
225 lmol P d21, while the other root half invariably
received 3 lmol P d21. Fresh mass was significantly
greater in G. crithmifolia plants that had received
a greater daily P supply during the pretreatments, but
symptoms of P toxicity were never observed. Clusterroot growth decreased from about half the total root
fresh mass when the leaf [P] was lowest (c. 0.1 mg P g21
DM) to complete suppression of cluster-root growth
when leaf [P] was 1–2 mg P g21 DM. Split-root studies
revealed that cluster-root initiation and growth, and net
P-uptake rates by roots were regulated systemically,
possibly by shoot P concentration. It is concluded that,
in response to higher P supply, G. crithmifolia does not
develop symptoms of P toxicity because of (i) greater
plasticity of its net P-uptake capacity, and (ii) its greater
plasticity for allocating P to growth and P storage in
roots. This ecologically important difference in plasti-
city is most probably related to a slightly higher nutrient
availability in the natural habitat of G. crithmifolia when
compared with that of H. prostrata.
Key words: Hakea prostrata, net P uptake, phosphorus toxicity,
plant growth, proteoid roots, split-root design, systemic
regulation.
Introduction
The Proteaceae comprise some 1800 species in 80 genera.
The largest numbers of Proteaceous species occur in the
South West Botanical Province of Western Australia
(c. 650–682) and the Cape Floristic Region of South Africa
(c. 331 species) (Cowling and Lamont, 1998; Pate et al.,
2001); both areas have a Mediterranean climate, and are
recognized as ‘Global Biodiversity Hotspots’ (Myers et al.,
2000).
Common factors linking species from this diverse group
are a well-known ability for growth on oligotrophic soils,
low in available phosphorus, and a remarkable plasticity of
their roots: the acclimation and functioning of their root
systems has been a major focus of ecophysiological investigations (Grundon, 1972; Lamont, 1982; Dinkelaker et al.,
1995; Neumann and Martinoia, 2002; Shane and Lambers,
2005). Many essential nutrient resources (e.g. phosphorus
and manganese) in soils from south-western Western
Australia occur mostly in a chemically bound (fixed) form
(Pate and Dell, 1984; Foulds, 1993), and are thus only
sparingly accessible to plants not adapted to these conditions (Handreck, 1997a, b). Almost all of the Proteaceae
* Present address and to whom correspondence should be addressed. Department of Botany, University of Cape Town, Private Bag, Rondebosch, 7701,
South Africa. E-mail: [email protected]
ª The Author [2005]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved.
For Permissions, please e-mail: [email protected]
414 Shane and Lambers
form very dense clusters (termed ‘proteoid’ roots; Purnell,
1960) of short, determinate (branch) rootlets, at discrete
regions along the growing root axis (Lamont, 1982; Watt
and Evans, 1999). At low external P supplies, cluster roots
may comprise half or more of the total mass of the root
system, whereas higher P supplies suppress the proportion
of total root mass invested in cluster roots (Lamont, 2003;
Shane and Lambers, 2005). Furthermore, when P is limiting
growth, the carbon metabolism of the cluster roots is geared
toward synthesis and exudation of large quantities or
carboxylates (e.g. citrate) (Dinkelaker et al., 1989; Neumann
and Martinoia, 2002; Shane et al., 2004a). The released
carboxylates solubilize sparingly available nutrients, especially mineral-bound phosphorus (Ryan et al., 2001; Vance
et al., 2003). Cluster roots also exude phosphatases that
release P from organic P sources (Adams and Pate, 1992;
Grierson and Comerford, 2000; Vance et al., 2003).
Once P is in soluble inorganic form, roots acquire it by
the energy-dependent uptake of the inorganic phosphate
(Pi) from the soil (Abel et al., 2002; Epstein and Bloom,
2004). The active transport of Pi across the root epidermal
plasmalemma is mediated by both high- and low-affinity
transport systems (Smith et al., 2003). The high-affinity Pi
transporters are H+-dependent, and are encoded by Pht1
multigene families (Raghothama, 1999). It is essential that
the uptake of phosphate be regulated in order to prevent
P toxicity when soil P is not limiting and Pht1 gene transcript amounts are generally greatly suppressed when P in
the root zone is abundant, and enhanced when P is
limiting (Raghothama, 1999; Smith et al., 2003).
Many species of Proteaceae, either soil-grown or grown
in hydroponics, are extremely sensitive to the amount
of P supplied for growth, and develop P-toxicity symptoms
at [P] in the root environment that are harmless to other
species (e.g. Banksia ericifolia, Handreck, 1991; Parks
et al., 2000; B. grandis, Lambers et al., 2002; H. prostrata,
Shane et al., 2004b). However, many other species of
Proteaceae tolerate far greater amounts of P in their root
environment (e.g. Grevillea crithmifolia, Handreck, 1997a,
b). Previous findings have uncovered the mechanisms involved in this ‘P-sensitivity’ by showing that a ‘P-sensitive’
species, H. prostrata, had a very low capacity to reduce its
net P-uptake rates. The low capacity to down-regulate net
P-uptake rates in H. prostrata was interpreted as an adaptation to allow for continued P acquisition and storage during
the wet season (Shane et al., 2004b, c). In the Mediterranean
climate in the south-west of Western Australia, cluster-root
initiation and functioning occur exclusively in the wet
winter season (Jeschke and Pate, 1995), whereas shoot
growth predominantly occurs in spring and summer (into
mid-summer for H. prostrata; Lamont, 1976). In the dry
season, stored P is remobilized for growth.
In the present study, the susceptibility of G. crithmifolia
R.Br. (Proteaceae), a species that is in the sister genus to
that of H. prostrata (Briggs, 1998), to develop ‘P-toxicity’
symptoms at a range of P supplies for growth was tested.
G. crithmifolia occurs naturally along a narrow strip of
coastal land in the south-west of Western Australia near
Perth (Fig. 1A). This species ‘inhabits calcareous sand
and coastal limestone soil in coastal scrub, eucalypt and
Fig. 1. Distribution maps of (A) Grevillea crithmifolia and (B) Hakea prostrata in Western Australia, Australia. Permission to reproduce the species
distribution maps was kindly provided by Paul Gioia (Department of Conservation and Land Management, Western Australia).
Suppression of cluster roots in Grevillea
banksia woodland and in open coastal heath’ (Olde and
Marriott, 1995). Like most natural environments in Western
Australia, the habitat of G. crithmifolia is considered relatively nutrient poor, but not quite as nutrient-impoverished
as that of H. prostrata, which is ‘a common species of
south-western Australia and found on white or grey sand,
gravel or sandy clay in low dense heath, Eucalyptus
woodland, coastal heath or jarrah woodland’ (Barker
et al., 1999) (Fig. 1B). The ‘P-insensitive’ G. crithmifolia
was grown in nutrient solutions in a similar manner as done
previously for ‘P-sensitive’ H. prostrata; that is, supplied
with external [P] from severely limiting for growth to [P]
that is toxic to H. prostrata. Plasticity to P supply was
determined for its growth response, cluster-root formation,
internal [P], and net P-uptake rates.
Materials and methods
In all experiments 2-month-old soil-grown G. crithmifolia R.Br. were
used. Seedlings were obtained from Kings Park and Botanic Garden
(Perth, Western Australia). Soil was gently washed from root systems using tap water, and each plant was grown in aerated nutrient
2+
solution (pH 5.8) containing (in mmol mÿ3): 400 NOÿ
3 ; 200 Ca ,
2+
ÿ
2+
200 K+, 154 SO2ÿ
;
54
Mg
,
20
Cl
,
2.0
Fe-EDTA,
0.24
Mn
,
4
2+
4+
2+
0.10 Zn , 0.02 Cu , 2.4 H3BO3, and 0.3 Mo made up in deionized
water. Plants were grown in a glasshouse at min/max 20/32 8C, and
root temperature was maintained at 18–20 8C with the pots in a
temperature-controlled tank (Fig. 2A, B).
Experiment 1: growth at a range of phosphorus supplies
Twenty-three plants of uniform size were selected, and each was
grown in 2.0 l of nutrient solution as above, supplemented with P
(supplied as KH2PO4) to give six P supplies: 0, 0.4, 0.8, 2, 20, and
200 lmol P dÿ1 (n=4 plants in each P treatment, except the zero P
treatment, where n=3) (Fig. 2A). The P treatments were carried out
for 8 weeks (December, 2003 to January, 2004), and nutrient
solutions containing the daily P supply were replaced every day.
Experiment 2: growth in a split-root design
Seedlings were prepared for growth in a split-root design, as
described for H. prostrata (Shane et al., 2003a). Twenty-four plants
of uniform size and with equal numbers of lateral roots on each ‘root
half’ were transferred to split-root containers, and shoots were
supported in the centre of a rectangular, grey-foam lid. Each root
half was supplied with 3.0 l of continuously aerated nutrient solution
as above (Fig. 2B, D, E). All plants had one root half supplied with
3 lmol P dÿ1, while the other half was either: (i) deprived of P, or
supplied with (ii) 3 lmol P dÿ1, (iii) 75 lmol P dÿ1, or (iv) 225 lmol
P dÿ1 (n=6). The complete nutrient solution in each pot was changed
daily for the duration of the experiment (12 weeks).
Measurements of net P-uptake rates
Net P-uptake rates were determined for whole root systems as
described in Shane et al. (2004c). Briefly, P depletion was measured
from an external solution having an initial [P] of 5 mmol P mÿ3 over
3 h. It was determined, in preliminary experiments, that 6.0 l of
nutrient solution containing 5 mmol P mÿ3 was required to give
a linear P-depletion rate over 3 h. Plants remained in the glasshouse in
which they were grown during all P-uptake measurements. A small
volume of concentrated KH2PO4 solution (less than 100 ll) was
415
added to each pot containing fresh nutrient solution minus P (i.e. one
plant per 6.0 l pot) to give a final concentration of 5 mmol P mÿ3.
After 120 s of mixing by vigorous aeration in each pot, a 1 ml sample
(time zero) was taken from each pot, and subsequent samples taken at
30-min intervals, between approximately 10.30 h and 13.30 h.
Harvests and determination of phosphorus concentrations
Plants were harvested after an 8- (P-range experiment) or a 12-weeks
(split-root experiment) treatment, and immediately after P-depletion
measurements. Plants were separated into roots (cluster roots and
non-cluster roots separately) and shoots (young: still expanding
leaves, and mature leaves: fully expanded, and stems). Root [P] was
measured only in young growing roots, therefore non-cluster roots
were further subdivided into regions 30 mm proximal to the root tip,
and young (white) cluster roots only were used for analysis of root
[P]. Samples were weighed fresh, and again after drying for 7 d at
80 8C. Dried samples were digested in concentrated HNO3:HClO4
(3:1. v:v) at 175 8C. Total [P] in tissue digests and in solutions
collected from P-depletion studies were determined using the
Malachite green colorimetric method (Motomizu et al., 1983).
Statistics
Data were analysed with one-way analysis of variance (GenStat 7.1,
Lawes Agricultural Trust; Rothamsted Experimental Station).
Tukey’s pair-wise multiple comparison tests were used to determine
which levels differed significantly (a=0.05). To ensure normality and
homogeneity of variances, data were log transformed when necessary.
Results
Experiment 1: Plants grown at a range of
phosphorus supplies
Symptoms of P toxicity (i.e. premature leaf senescence,
chlorosis and necrosis, and stunted growth) were never
observed in G. crithmifolia, not even at a P supply of
200 lmol P dÿ1. There was no significant influence of the
lowest three P-supply rates (0, 0.4, and 0.8 lmol P dÿ1) on
root, stem, and leaf fresh mass (Fig. 3A) and the fresh mass
invested in cluster roots, as a percentage of total root fresh
mass, was approximately 50%. The ratio of total plant mass
to total root mass (i.e. root mass ratio) decreased with
increasing P supply, from 0.48 (no P supplied) to about
0.27 (20 or 200 lmol P dÿ1) (Fig. 3A). When the external
P-supply rate was 2 lmol P dÿ1, the fresh mass of noncluster roots, stems, and leaves was almost double, whereas
that of cluster roots was about c. 25% of the total root mass
(Fig. 3A). Higher P-supply rates, either 20 or 200 lmol
P dÿ1, were associated with c. 3-fold higher fresh mass of
non-cluster roots, stems, and leaves, whereas cluster-root
fresh mass was c. 6% of the total root mass at a P supply rate
of 20 lmol dÿ1, and cluster roots were completely suppressed on plants supplied with 200 lmol P dÿ1.
Leaf [P] increased from 0.15 mg P gÿ1 DM (no P
supplied) to 2.5 mg P gÿ1 DM at the highest external
P-supply rate of 200 lmol P dÿ1 (Fig. 3B). Leaf [P] was
always less than that of roots, but leaf and root [P] were
most similar (c. 1.6 and 2.1 mg P gÿ1 DM, respectively) in
plants grown at 20 lmol P dÿ1. When the P supply for
416 Shane and Lambers
Fig. 2. Grevillea crithmifolia plants grown in nutrient solutions in the greenhouse at a range of P concentrations. (A) Plants in the ‘P-range’ experiment
received: 0, 0.4, 0.8, 2, 20, or 200 lmol P plantÿ1 dÿ1 increasing from left to right in the photograph. (B) Plants grown in the ‘split-root’ experiment all
had one root-half supplied with 3 lmol P dÿ1, while the other root half was either, deprived of P, or supplied with 3 lmol P dÿ1, 75 lmol P dÿ1, or
225 lmol P dÿ1. (C) Young, developing cluster roots of G. crithmifolia are the ‘simple’ type (Lamont, 1982) of ‘proteoid’ root as produced in Hakea
spp. and was stimulated when plants were grown at <1 lmol P dÿ1. (D) Split-root G. crithmifolia plant grown with one root half receiving 3 lmol
P dÿ1 (left side of photograph) while the other root half received 225 lmol P dÿ1 (right side of the photograph). Cluster roots were not produced on
either root half. (E) Split-root G. crithmifolia plant grown where each root half received 3 lmol P dÿ1. Cluster roots were produced on both root halves
but their initiation and growth was not synchronized, i.e. the root half shown on the left side of the photograph had young (white) developing cluster
roots, while the most recently developed cluster roots on the other root half (supplied with the same amount of P) were already senesced.
growth was 200 lmol P dÿ1, root [P] was sharply higher at
7.7 mg P gÿ1 DM, whereas that of leaves was not
significantly higher (Fig. 3B). Water content of mature
leaves was approximately 65% in plants supplied with 0,
0.4, and 0.8 lmol P dÿ1, whereas it was c. 72% when P was
supplied at rates of 2, 20, and 200 mmol mÿ3 dÿ1 (data
not shown).
The fastest net P-uptake rate (i.e. 0.09 nmol P gÿ1 root
FM sÿ1) at a standard external concentration of 5 mmol
P mÿ3 was measured for G. crithmifolia plants that had been
deprived of P (Fig. 4), and these plants also had the lowest
leaf and root [P] (Fig. 3B). Net P-uptake rates decreased
linearly (r2=0.93) to 0.047 nmol P gÿ1 FM sÿ1 as external
P-supply rates supplied for growth during the 12 weeks in
Suppression of cluster roots in Grevillea
120
leaf
80
60
0.42
40
0.48
0.48
20
0.48
Tissue [P] (mg P g-1 dry mass)
0
B
10
root [P]
leaf [P]
Grevillea crithmifolia
Hakea prostrata
0.26
stem
non-cluster root
cluster root
100
Fresh mass (g)
0.12
0.28
Net P-uptake rate (nmol P g-1 root FM s-1)
A
0.1
417
a
ab
0.08
b
0.06
abc
ab
bc ab
bc
c
a
0.04
0.02
d
8
0
0
0.4
0.8
2
20
200
6
4
2
c
a
0
ab
ab
a
0
ab
0.4
d
bc
d
bc
0.8
c
2
20
200
P supplied for growth (µmol P plant-1 day-1)
Fig. 3. Growth and P concentrations of Grevillea crithmifolia plants
grown at a range of P supplies in nutrient solution. (A) Fresh mass of
leaves, stems, non-cluster and cluster roots. Value at the top of each bar is
the root mass ratio. (Error bars are SE, n=6 (P <0.05). (B) Phosphorus
concentrations of mature leaves and ‘young’ white roots. Different letters
indicate significant differences in [P] within leaf and root tissue. Error
bars are SE, n=6, (P <0.05).
the pre-treatment increased from 0 to 200 lmol P dÿ1
(Fig. 4). However, the net P-uptake rates, at a standard
external concentration of 5 mmol P mÿ3, were the same for
G. crithmifolia plants that had been grown at 2, 20, and
200 lmol P dÿ1 (Fig. 4), despite further significant increases
in root and leaf [P].
Experiment 2: plants grown in a split-root design
All plants grew well in the split-root design, and none
showed symptoms of P toxicity, even at the highest P
supply of 225 lmol P dÿ1 to one root half. Plants that had
one root half deprived of P, while the other root half was
supplied with 3 lmol P dÿ1 were small (c. 38 g FM plantÿ1,
Fig. 5A), whereas the fresh mass of G. crithmifolia plants
nearly doubled (c. 66 g plantÿ1) for plants that had both root
halves supplied with 3 lmol P dÿ1; the higher mass resulted
mostly from larger shoot biomass (Fig. 5A). When the
P-supply rate to one root half was either 75 or 225 lmol
Fig. 4. Net P-uptake rates for intact whole root systems, calculated from
P-depletion curves; the nutrient solution contained 5 mmol P mÿ3,
irrespective of the P supply during growth. Uptake rates were plotted
against the external [P] supplied during growth of Grevillea crithmifolia.
The values shown for Grevillea crithmifolia from the present study, while
the data points for Hakea prostrata were from Shane et al. (2004c).
Different letters indicate significant differences. Error bars are SE,
n=6, P <0.05.
P dÿ1, the fresh mass of G. crithmifolia was nearly 3-fold
greater (c. 200 g), accompanied by a significant increase in
root as well as shoot biomass. The root mass ratio (using the
combined mass of each root half on individual plants) was
0.54 for plants with one root half deprived of P, but lower
(0.22) for plants with one root half supplied with either 75
or 225 lmol P dÿ1 (Fig. 5A).
The fresh mass of root halves supplied with 3 lmol P dÿ1
varied from 14–25 g, depending upon the P supply to the
other root half (Table 1). Root halves receiving 75 lmol
P dÿ1 had the greatest mass (61 g), whereas the mass of root
halves supplied with 225 lmol P dÿ1 was similar (21 g) to
that of the other half (25 g) supplied with 3 lmol P dÿ1
(Table 1).
All plants had one root half that received the same
‘relatively low’ supply of P (i.e. 3 lmol P dÿ1). Cluster-root
formation in G. crithmifolia was completely suppressed on
both root halves of plants when either 75 or 225 lmol P dÿ1
was supplied to one root half (Table 1). When one root half
was either deprived of P or supplied with 3 lmol P dÿ1
(while the other root half was supplied with 3 lmol P dÿ1)
the percentage of root fresh mass allocated to cluster roots
was similar (range 32–44%) on each root half of individual
plants (Table 1). On plants that developed cluster roots, the
timing of cluster-root initiation and growth was not synchronized between root halves on individual plants.
418 Shane and Lambers
The [P] of old leaves was less than that of young leaves
in G. crithmifolia when one root half received either 0 or
3 lmol P dÿ1 (Fig. 5B). When 75 lmol P dÿ1 was supplied
to one root half, the [P] in old leaves was 6-fold higher than
in the absence of P, equalling that of young leaves, in which
[P] was 2-fold higher than in the absence of P; on plants
with one root half receiving 225 lmol P dÿ1 the [P] in old
leaves was 11-fold higher (c. twice that of young leaves)
(Fig. 5B) than in the absence of P. Root [P] was measured
only in young white roots (Fig. 5C). The [P] of these roots,
for root halves supplied with 3 lmol P dÿ1, was approximately 1.4 mg P gÿ1 DM, and increased to 2.1 mg P gÿ1
DM as the P supply to the other root-half was increased
from 0 to 225 lmol P dÿ1; however, the increase in root [P]
(c. 50%) was significant only for those plants whose other
root half received 225 lmol P dÿ1 (Fig. 5C). The [P] of root
halves with a variable P supply (i.e. 0, 3, 75, and 225 lmol
P dÿ1) increased significantly from 0.6 to 4.0 mg P gÿ1 DM
(c. 600%).
Net P-uptake rates, measured at a standard external P
concentration of 5 mmol P mÿ3, were the same on both root
halves of individual G. crithmifolia plants, regardless of the
large difference in P supply to the root halves during growth
(Fig. 6). For root-halves supplied with 3 lmol P dÿ1, the net
P-uptake rate was fastest (0.08 nmol P gÿ1 root FM sÿ1)
when the other half was deprived of P. The net P-uptake
rates for both root halves decreased incrementally to 0.02
nmol P gÿ1 root FM sÿ1 as the P-supply to one root-half
was increased from 0 to 3, 75, and 225 lmol P dÿ1,
respectively (Fig. 6).
Discussion
The results of the present experiments confirm earlier work
(Handreck, 1997a, b) that Grevillea crithmifolia (Proteaceae) is resistant to developing symptoms of P toxicity at
external P supplies that cause P toxicity in a ‘P-sensitive’
Proteaceae, for example, Hakea prostrata (Shane et al.,
2004b, c). These results are particularly exciting, in that
they show a distinct difference in ecophysiological functioning of two species of closely related (paraphyletic;
Briggs, 1998) genera. This difference in functioning, in
species endemic to the south-western corner of Western
Australia, a Global Biodiversity Hotspot (Myers et al.,
2000; Hopper and Gioia, 2004), is most likely linked to a
subtle difference in habitat characteristics, as discussed
below.
Fig. 5. Analysis of Grevillea crithmifolia plants grown in split-root
culture for 12 weeks under the indicated P treatments. Influence of P
supply (x-axis) on (A) root and shoot fresh mass. The value above each
bar is the root mass ratio based on the entire root system (sum of both root
halves). (B) Total P concentration in young and old leaves. The value
above young and old leaves for each treatment is the total plant P (mg)
based on average mass and [P] of the organs. (C) The phosphorus
concentration in root halves; the 3 lmol P dÿ1 root halves (open bars)
were compared to one another, as were the 0, 3, 75, and 225 lmol P dÿ1
root halves (filled bars). Bars indicate standard errors; n=6, different
letters denote significant differences (P <0.05).
Suppression of cluster roots in Grevillea
419
Table 1. Influence of [P] supplied to root halves of Grevillea crithmifolia R.Br. on the percentage of total root fresh mass (grams)
allocated to cluster roots after 12 weeks P treatment; standard error in parentheses
P supplied to root halves (lmol P dÿ1)
Total root
Cluster root
% Cluster root
(3
0)
(3
3)
(3
75)
(3
225)
14 (1.4)
6.2 (1.4)
45 (8)
6.6 (1.6)
2.1 (0.5)
35 (7)
11 (1.4)
4.0 (0.7)
35 (5)
16 (1.3)
5.7 (0.5)
38 (7)
25 (4)
0
0
61 (14)
0
0
21 (4.1)
0
0
25 (5)
0
0
Fig. 6. Net P-uptake rates for intact, whole root halves of Grevillea
crithmifolia plants grown in split-root culture after 12 weeks under the
indicated P treatments. Uptake rates were calculated from P-depletion
curves constructed using a nutrient solution containing 5 mmol P mÿ3.
Uptake rates were plotted against the external P supplied to each root half
during growth in split-root culture. The 3 lmol P dÿ1 root halves (open
bars) were compared to one another, as were the 0, 3, 75, and 225 lmol
P dÿ1 root halves (filled bars). Bars indicate standard errors (n=3); different letters denote significant differences (P <0.05).
What accounts for variable development of
P-toxicity symptoms amongst species of
Proteaceae grown at low P supply?
It is improbable that native plants in their own habitat are
ever exposed to levels of soil P that induce P toxicity,
because the levels in that habitat are too low for that. The
development of P-toxicity symptoms generally only happens on fertilized soils or when, because of regular watering, plants can continually take-up P (Shane et al., 2004b,
c). The difference in susceptibilities among species for
developing P-toxicity symptoms, at the same external P
supply, might result from differences in (i) ‘P sensitivity’ of
cells and tissues, (ii) growth responses, (iii) nutrient
allocation, and (iv) rates of net P uptake.
The leaf [P] required to cause P toxicity symptoms does
not differ greatly among species. P-toxicity symptoms (e.g.
reduced growth, leaf chlorosis, and necrosis) are similar
among species of Proteaceae (Grundon, 1972; Handreck,
1997b; Silber et al., 1998; Table 2 in Shane et al., 2004b)
and crops (Asher and Loneragan, 1967; Marschner, 1995),
and generally follow the accumulation of P to approxi-
mately 10 mg P gÿ1 in the leaf DM. In the present study,
leaf [P] in G. crithmifolia never exceeded 2.5 mg gÿ1 DM
(Figs 2B, 3B), which is well below that reported to cause P
toxicity. Although whole tissue nutrient analysis is useful, it
reveals little about the nutrient concentrations in individual
cells (Pate and Dell, 1984; Karley et al., 2000). [P] had
previously been measured in vacuoles of mesophyll cells of
intact leaves of H. prostrata, and it was found that it was
similar (millimolar range) to that in other species (Shane
et al., 2004b) when whole tissue analysis of leaf [P]
exhibited concentrations of less than 1 mg P gÿ1 dry mass.
Thus, enhanced P sensitivity of leaf tissues and cells does
not explain why species differ in their susceptibility to
developing P toxicity (i.e. does not explain why H. prostrata
is very sensitive to P toxicity compared with G. crithmifolia
at relatively low external P supply).
In terms of nutrient allocation, G. crithmifolia accumulated more P in young roots (Figs 3B, 4C) compared with
bulked root samples of H. prostrata grown at the same
range of P supplies (Table 2 in Shane et al., 2004b). The
significant accumulation of much higher [P] in roots of
G. crithmifolia was accompanied by a non-significant
increase in leaf [P] when plants were grown at the highest
P supplies. The allocation of ‘excess’ P to roots of
G. crithmifolia buffered the leaf [P] (Fig. 3A, B). Several
other species of Proteaceae are also known to accumulate
or store P in roots and stems (Jeschke and Pate, 1995;
Parks et al., 2000; Shane et al., 2004b). Stem [P] was
not measured in G. crithmifolia, but quantifying tissueP-allocation patterns between species with differences in
susceptibility to developing symptoms of P toxicity, and
assessing their net P-uptake rates, would help to examine
further the importance of nutrient-allocation patterns in
determining the susceptibility to developing P toxicity.
Fresh mass of G. crithmifolia more than doubled when
the P supply for growth increased from 2 to 20 lmol P dÿ1,
but increasing the P supply to 200 lmol P dÿ1 had no
further influence on growth. Growth of H. prostrata also
increased in response to a similar range of P supplies (Fig.
1A, D; Shane et al., 2004b, c, respectively), albeit not as
strongly as that of G. crithmifolia. Differences between the
two species in susceptibility to P toxicity cannot be
accounted for entirely by P dilution by growth, because
growth saturated before the highest P-supply rate was
reached in the P-insensitive G. crithmifolia (Fig. 3A). The
420 Shane and Lambers
same growth pattern was found in P-sensitive H. prostrata
(Fig. 1A, B in Shane et al., 2004b; Fig. 2 in Shane et al.,
2003a). Because H. prostrata developed P-toxicity symptoms at low external P supplies that had no harmful effects
on G. crithmifolia, the toxic effects of P accumulation might
be responsible for the decreased growth of H. prostrata,
rather than decreased growth leading to P accumulation.
In this study, P-toxicity symptoms were not observed
in G. crithmifolia, even at the highest P supplies (200 lmol
P dÿ1; Figs 1, 3), whereas in H. prostrata P-toxicity
symptoms developed when plants were grown at the same
range of P supplies (Shane et al., 2004b). The accumulation
of P to toxic levels in leaves of H. prostrata was related to
the species’ low capacity to down-regulate net P-uptake
rates (Shane et al., 2004c). Here, it was shown that net
P-uptake rates, measured at a standard external P concentration, decreased linearly in G. crithmifolia plants grown at
the same P-supply rates used for H. prostrata (i.e. from 0 to
2 lmol P dÿ1, Figs 2, 4; cf. Shane et al., 2003a, 2004b, c).
The fact that leaf P did not accumulate to toxic concentrations in leaves of G. crithmifolia, and that this species
had a strong capacity to down-regulate net P-uptake rates,
supports the hypothesis that development of P-toxicity
symptoms is related to a low capacity to down-regulate net
P-uptake rate (Shane et al., 2004c). It is concluded that
susceptibilities for developing P-toxicity symptoms, at the
same external P supply, in species of Proteaceae adapted
to nutrient-impoverished soils, is related to a species’ capacity to down-regulate net P-uptake rates by its roots.
Signalling of P status in ‘P-toxicity sensitive’ and
‘P-toxicity resistant’ Proteaceae
In the split-root experiment with G. crithmifolia, it was
shown that suppression of cluster-root growth (Table 1) and
net P-uptake rates (Fig. 6) occurred systemically with
respect to elevated shoot [P]: i.e. cluster-root growth and
net P-uptake rates were similar on both root halves of
individual plants despite significant differences in P supply
(and [P]) to the root halves. Systemic regulation of clusterroot growth has also been found for white lupin (Lupinus
albus) grown in a similar split-root design (Fig. 2; Shane
et al., 2003b; Tian et al., 2004; Li and Liang, 2005; Shen
et al., 2005). These results contrast strongly with previous
observations on H. prostrata, which continues to develop
cluster roots on ‘low-P’ root halves despite suppression of
cluster-root growth on ‘high-P’ root halves and increased
shoot [P] (Table 1; Shane et al., 2003a). This may mean
that species that are tolerant of a higher P supply in the root
environment (i.e. G. crithmifolia and L. albus), and which
do not readily develop P-toxicity symptoms (Keerthisinghe
et al., 1998; Shane et al., 2004c), react systemically with
a reduction in both cluster-root growth and net P-uptake
rates in response to enhanced P supply. This requires
further investigation.
It is known that systemic signalling in plants during
phosphate starvation is responsible for P-deficiencyinduced gene expression in Lycospersicon esculentum,
and that the expression of these genes is suppressed by
elevated shoot [P] (Burleigh and Harrison, 1999). Many
studies have tried to identify components in the signaltransduction pathway between ‘sensing’ shoot [P] status
and the observed systemic alterations in root growth and
metabolism. For example, auxins and cytokinins are most
likely involved in cluster-root growth, but their influence is
probably to be at the end of the signal-transduction pathway
(Gilbert et al., 2000; Skene and James, 2000; Neumann and
Martinoia, 2002). At the cellular level it is known that
cytoplasmic [P] often remains relatively constant, even
though the external P supply is increased (Schachtman
et al., 1998). Therefore, where and how the low [P] in
tissues (e.g. young or old leaves) generates intracellular
signals, the factors that modify gene expression in the nucleus, are unknown (Schachtman et al., 1998). Recently,
levels of sugars and sugar metabolites (i.e. glucose,
fructose, and sucrose) were found to stimulate LaPT1
(phosphate transporter) and LaSAP1 (secreted acid phosphatase) gene expression in the cluster-root forming
L. albus (Liu et al., 2005). Moreover, expression of these
genes in P-deficient plants was able to be manipulated
depending upon whether or not the plants were in the light
(photosynthesis) (Liu et al., 2005). Furthermore, in Arabidopsis, several phosphate-starvation-inducible genes are
also sugar-induced (Müller et al., 2005). Investigations of
species with distinct differences (e.g. H. prostrata and
G. crithmifolia) in their systemic regulation of (cluster-)
root growth and net P-uptake rates may help identify
components of the signal-transduction pathway between
‘sensing’ [P] and the regulation of gene expression leading
to changes in root growth and P uptake.
Searching for an ecophysiological explanation for
differences in a species’ predisposition to
developing symptoms of P toxicity
A more plastic growth response to added nutrients, the
enhanced allocation of P to roots, a ‘stronger’ systemic
P-signalling system, and a greater capacity to down-regulate
net P-uptake rates in G. crithmifolia compared with that in
H. prostrata are likely to be adaptive responses of a plant
functioning at slightly elevated P concentrations in the
rhizosphere. Because Grevillea is paraphyletic with respect
to Hakea (Briggs, 1998), an examination of the natural
environments in which H. prostrata and G. crithmifolia
characteristically occur may provide significant clues as to
the ecophysiological significance of these differences in the
response to P availability. Both species are endemic to the
south-west of Western Australia. H. prostrata is widely
distributed (Fig. 1A), invariably on severely phosphorusimpoverished acid soils, whereas G. crithmifolia has a much
Suppression of cluster roots in Grevillea
narrower habitat range (Fig. 1B), occurring along the coast
on soils over limestone. Limestone formation is the result of
leaching of calcium carbonate; soils over limestone are
presumably also enriched in other nutrients that leached
towards these sites. In south-western Western Australia,
somewhat nutrient-enriched ‘kwongan’ sites are associated
with soils over limestone, having a pH of 7.0 or more
(Foulds, 1993). In southern South Africa, ‘fynbos’ soils over
limestone (pH 7–8) have P, N, and micronutrient levels up to
10 times higher than those in nearby acidic soils (Esler et al.,
1989). The present differences between H. prostrata and
G. crithmifolia to regulate their P status might have resulted
from G. crithmifolia adapting to less P-impoverished soils.
It was found that G. crithmifolia grows better when P levels
are somewhat enhanced (Fig. 1A), and in its natural habitat
it is associated with slightly nutrient-enriched sites over
limestone (Olde and Marriott, 1995), which have slightly
elevated nutrient levels (Foulds, 1993).
Species-specific differences in ability to prevent excessive uptake of ions at the root surface are related to edaphic
factors (Rajakaruna et al., 2003), and to their root growth
allocation patterns (Poot and Lambers, 2003). The present
discovery that H. prostrata does not, whilst G. crithmifolia
does show plasticity in its P-uptake capacity offers an
explanation for the question why H. prostrata cannot enter
the habitat of G. crithmifolia. However, why G. crithmifolia
does not occur in the habitats occupied by H. prostrata
remains unknown. In South African ‘fynbos’, distribution
of Protea species that are restricted to either soils over
limestone (P. obtusifolia) or acid soils (P. susannae) is
thought to result, in part, from competitive exclusion
(Mustart and Cowling, 1993). It is speculated that this is
also related to species differences in the efficiency of nutrient
capture or nutrient use. For example, P retranslocation from
senescing leaves varies considerably among species adapted
to infertile soils (Güsewell, 2005). Although some species
of Proteaceae can remove up to 90% of the P from senescing leaves (Chapter 6, Table 24 in Lambers et al., 1998),
it is likely that this efficiency is less for other species of
Proteaceae, for example, G. crithmifolia. In addition, there
might be differences in P-allocation patterns and tissues used
for temporary P storage (Dixon et al., 1983; Pate and Dell,
1984; Shane et al., 2004b) amongst species of Proteaceae.
The present findings suggest that physiological aspects
of root functioning in Proteaceae, in relation to soil P
characteristics are involved in niche preferences. Further
examinations of Proteacean species that are restricted to
specific soils may yield alternative adaptive mechanisms
that allow species to avoid Al or Mn toxicity, and cope with
low availability of Ca, Mg, or Mo. Species restricted to
calcareous soils may also have adaptations to prevent
bicarbonate toxicity, and to tolerate excessive Ca and the
limited availability of Fe, Mn, and Zn. These aspects
deserve to be studied in greater detail and may provide
alternative explanations, in addition to what we have shown
421
for the role of soil P, in the selective colonization potential
of species of Proteaceae to edaphically different habitats.
Concluding remarks
The present results help to explain why species of closely
related genera of the Proteaceae fill marginally different
niches. They also provide the start of an ecophysiological
framework for understanding the enormous plant species
richness in the south-west of Australia, a global biodiversity
hotspot (Myers et al., 2000; Hopper and Gioia, 2004), as
related to edaphic factors. In G. crithmifolia, the net
P-uptake rates by roots were down-regulated strongly and
systemically, in contrast to H. prostrata. Moreover, better
growth and P allocation to roots of G. crithmifolia helped to
prevent toxic quantities of P accumulating in the leaves.
These results on two closely related taxa endemic on
different soil types indicate that a larger and more comprehensive survey of native species should be carried out.
Species that differ in their susceptibility to developing Ptoxicity symptoms in the species-rich flora of south-western
Australia will provide valuable information on the significance of root adaptation and root physiology in the process
of species diversification.
Acknowledgements
We are grateful to Professor Kingsley Dixon at Kings Park and
Botanic Garden for providing the Grevillea crithmifolia R.Br.
seedlings for our experiments. We are also grateful to Dr Patrick
Finnegan and Professor Stephen Hopper for providing valuable
comments during the preparation of this manuscript. We thank
Stuart Pearse for help with growing the plants and Paul Gioia for
permission to reproduce the species distribution maps. We also thank
the Journal of Experimental Botany, which supported the SEB
Symposium Plant Frontiers Meeting 2005, at which the data included here were presented first, and two anonymous reviewers for
helpful comments. This work was supported by the Australian
Research Council.
References
Abel S, Ticconi CA, Delatorre CA. 2002. Phosphate sensing in
higher plants. Physiologia Plantarum 115, 1–8.
Adams MA, Pate JS. 1992. Availability of organic and inorganic
forms of phosphorus to lupins (Lupinus spp.). Plant and Soil 145,
107–113.
Asher CJ, Loneragan JF. 1967. Response of plants to phosphate
concentration in solution culture. I. Growth and phosphorus
content. Journal of Soil Science 103, 225–233.
Barker RM, Haegi L, Barker WR. 1999. Hakea. In: Proteaceae 3:
Hakea to Dryandra. Flora of Australia 17B. Melbourne: CSIRO
Publishing, 31–170.
Briggs B. 1998. Proteaceae: diverse approaches toward a new
synthesis. Australian Systematic Botany 11, 631–633.
Burleigh SH, Harrison MJ. 1999. The down-regulation of Mt4-like
genes by phosphate fertilization occurs systemically and involves
422 Shane and Lambers
phosphate translocation to the shoots. Plant Physiology 119,
241–248.
Cowling RM, Lamont BB. 1998. On the nature of Gondwanan
species flocks: diversity of Proteaceae in Mediterranean southwestern Australia and South Africa. Australian Journal of Botany
46, 335–355.
Dinkelaker B, Hengeler C, Marschner H. 1995. Distribution and
function of proteoid root clusters and other root clusters. Botanica
Acta 108, 183–200.
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 and Environment
12, 285–292.
Dixon KW, Kuo J, Pate JS. 1983. Storage reserves of the seed-like,
aestivating organs of geophytes inhabiting granite outcrops in
south-western Australia. Australian Journal of Botany 31, 85–103.
Epstein E, Bloom AJ. 2004. Mineral nutrition of plants: principles
and perspectives, 2nd edn. Sunderland, Massachusetts: Sinauer
Associates, Inc. Publishers.
Esler KJ, Cowling RM, Witkowski ETF, Mustart PJ. 1989.
Reproductive traits and accumulation of nitrogen and phosphorus
during the development of fruits of Protea compacta R.Br.
(calcifuge) and Protea obtusifolia Buek. Ex Meisn. (calcicole).
New Phytologist 112, 109–115.
Foulds W. 1993. Nutrient concentrations of foliage and soil in
south-western Australia. New Phytologist 125, 529–546.
Gilbert GA, Knight JD, Vance CP, Allan DL. 2000. Proteoid root
development of phosphorus deficient white lupin is mimicked by
auxin and phosphonate. Annals of Botany London 85, 921–928.
Grierson PF, Comerford NB. 2000. Non-destructive measurement
of acid phosphatase activity in the rhizosphere using nitrocellulose
membranes and image analysis. Plant and Soil 218, 49–57.
Grundon NJ. 1972. Mineral nutrition of some Queensland heath
plants. The Journal of Ecology 60, 171–181.
Güsewell S. 2005. Nutrient resorption of wetland graminoids is
related to the type of nutrient limitation. Functional Ecology 19,
344–354.
Handreck KA. 1991. Interactions between iron and phosphorus in
the nutrition of Banksia ericifolia L. f. var. ericifolia (Proteaceae)
in soil-less potting media. Australian Journal of Botany 39,
373–384.
Handreck KA. 1997a. Phosphorus requirements of Australian native
plants. Australian Journal of Soil Research 35, 241–289.
Handreck KA. 1997b. Phosphorus needs of Australian plants:
removing the uncertainty regarding phosphorus toxicity. In:
Australian plants 8 ’SGAP Journal, South Australian Region of
the Society for Growing Australian Plants. Available online at
http://farrer.riv.csu.edu.au/ASGAP/APOL8/dec97-4.html.
Hopper SD, Gioia P. 2004. The southwest Australian floristic
region: evolution and conservation of a global hot spot of biodiversity. Annual Review of Ecology, Evolution, and Systematics
35, 623–650.
Jeschke DW, Pate JS. 1995. Mineral nutrition and transport in
xylem and phloem of Banksia prionotes (Proteaceae), a tree with
dimorphic root morphology. Journal of Experimental Botany 46,
895–905.
Karley AJ, Leigh RA, Sanders D. 2000. Differential ion accumulation and ion fluxes in the mesophyll and epidermis of barley.
Plant Physiology 122, 835–844.
Keerthisinghe G, Hocking PJ, Ryan PR, Delhaize E. 1998. Effect
of phosphorus supply on the formation of proteoid roots of
white lupin (Lupinus albus L.). Plant, Cell and Environment 21,
467–478.
Lambers H, Chapin III FS, Pons TL. 1998. Plant physiological
ecology. New York: Springer-Verlag.
Lambers H, Juniper D, Cawthray GR, Veneklaas EJ, MartinezFerri E. 2002. The pattern of carboxylate exudation in Banksia
grandis (Proteaceae) is affected by the form of phosphate added to
the soil. Plant and Soil 238, 111–122.
Lamont B. 1976. The effects of seasonality and waterlogging on the
root systems of a number of Hakea species. Australian Journal of
Botany 24, 691–702.
Lamont B. 1982. Mechanisms for enhancing nutrient uptake in plant
with particular reference to Mediterranean South Africa and
Western Australia. Botanical Review 48, 597–689.
Lamont B. 2003. Structure, ecology and physiology of root clusters:
a review. Plant and Soil 248, 1–19.
Li CJ, Liang RX. 2005. Root cluster formation and citrate
exudation of white lupin (Lupinus albus L.) as related to
phosphorus availability. Journal of Integrative Biology 47,
172–177.
Liu J, Samac DA, Bucciarelli B, Allan DL, Vance CP. 2005.
Signaling of phosphorus deficiency-induced gene expression in
white lupin requires sugar and phloem transport. The Plant Journal
41, 257–268.
Marschner H. 1995. Mineral nutrition of higher plants, 2nd edn.
London: Academic Press.
Motomizu S, Wakimoto T, Toei K. 1983. Spectrophotometric
determination of phosphate in river waters with molybdate blue
and malachite green. Analyst 108, 361–367.
Müller R, Nilsson L, Nielsen LK, Nielsen TH. 2005. Interaction between phosphate starvation signalling and hexokinaseindependent sugar sensing in Arabidopsis leaves. Physiologia
Plantarum 124, 81–90.
Mustart PJ, Cowling RM. 1993. The role of regeneration stages in
the distribution of edaphically restricted fynbos Proteaceae.
Ecology 74, 1490–1499.
Myers N, Mittermeier RA, Mittermeier CG, da Fonseca GAB,
Kent J. 2000. Biodiversity hotspots for conservation priorities.
Nature 403, 853–858.
Neumann G, Martinoia E. 2002. Cluster roots—an underground
adaptation for survival in extreme environments. Trends in Plant
Science 7, 162–167.
Olde P, Marriott N. 1995. The Grevillea book, Vol. 2. Species A–L.
Kenhurst: Kangaroo Press, 103.
Parks SE, Haigh AM, Creswell GC. 2000. Stem tissue phosphorus
as an index of the phosphorus status of Banksia ericifolia L. f.
Plant and Soil 227, 59–65.
Pate JS, Dell B. 1984. Economy of mineral nutrients in sandplain
species. In: Pate JS, Beard JS, eds. Kwongan: plant life of the
sandplain. Nedlands, Australia: University of Western Australia
Press, 227–252.
Pate JS, Verboom WH, Galloway PD. 2001. Co-occurrence of
Proteaceae, laterite and related oligotrophic soils: coincidental
associations or causative inter-relationships? Australian Journal of
Botany 49, 529–560.
Poot P, Lambers H. 2003. Are trade-offs in allocation pattern and
root morphology related to species abundance? A congeneric comparison between rare and common species in the south-western
Australia flora. Journal of Ecology 91, 58–67.
Purnell HM. 1960. Studies of the family Proteaceae. I. Anatomy and
morphology of the roots of some Victorian species. Australian
Journal of Botany 8, 38–50.
Raghothama KG. 1999. Phosphate acquisition. Annual Review
of Plant Physiology and Plant Molecular Biology 50, 665–693.
Rajakaruna N, Siddiqi MY, Whitton J, Bohm BA, Glass ADM.
2003. Differential responses to Na+/K+ and Ca2+/Mg2+ in two
edaphic races of the Lasthenia californica complex (Asteraceae):
a case for parallel evolution of physiological traits. New Phytologist 157, 93–103.
Suppression of cluster roots in Grevillea
Ryan PR, Delhaize E, Jones DL. 2001. Function and mechanism
of organic anion exudation from plant roots. Annual Review of
Plant Physiology and Plant Molecular Biology 52, 527–560.
Schachtman DP, Reid RJ, Ayling SM. 1998. Phosphorus uptake by
plants: from soil to cell. Plant Physiology 116, 447–453.
Shane MW, Cramer MD, Funayama-Noguchi S, Cawthray
GR, Millar AH, Day DA, Lambers H. 2004a. Developmental
physiology of cluster-root carboxylate synthesis and exudation
in Harsh Hakea. Expression of phosphoenolpyruvate carboxylase and the alternative oxidase. Plant Physiology 135,
549–560.
Shane MW, de Vos M, de Roock S, Cawthray GR, Lambers H.
2003a. Effects of external phosphorus supply on internal phosphorus concentration and the initiation, growth and exudation of
cluster roots in Hakea prostrata R.Br. Plant and Soil 248,
209–219.
Shane MW, de Vos M, de Roock S, Lambers H. 2003b. Shoot P
status regulates cluster-root growth and citrate exudation in
Lupinus albus grown with a divided root system. Plant, Cell and
Environment 26, 265–273.
Shane MW, Lambers H. 2005. Cluster roots: a curiosity in context.
Plant and Soil 274, 101–125.
Shane MW, McCully ME, Lambers H. 2004b. Tissue and cellular
phosphorus storage during development of phosphorus toxicity
in Hakea prostrata (Proteaceae). Journal of Experimental Botany
55, 1033–1044.
423
Shane MW, Szota C, Lambers H. 2004c. A root trait accounting
for the extreme phosphorus sensitivity of Hakea prostrata
(Proteaceae). Plant, Cell and Environment 27, 991–1004.
Shen J, Li H, Neumann G, Zhang F. 2005. Nutrient uptake, cluster
root formation and exudation of protons and citrate in Lupinus
albus as affected by localized supply of phosphorus in a split-root
system. Plant Science 168, 837–845.
Silber A, Ganmore-Neumann R, Ben-Jaacov J. 1998. Effects of
nutrient addition on growth and rhizosphere pH of Leucadendron
‘Safari Sunset’. Plant and Soil 199, 205–211.
Skene KR, James WM. 2000. A comparison of the effects of auxin
on cluster root initiation and development in Grevillea robusta
Cunn. ex R.Br. (Proteaceae) and in the genus Lupinus (Leguminosae). Plant and Soil 219, 221–229.
Smith FW, Mudge SR, Rae AL, Glassop D. 2003. Phosphate
transport in plants. Plant and Soil 248, 71–83.
Tian ZM, Wang B, Song CX, Li WP, Qin FL. 2004. The response
of Lupinus albus roots to the signal from phosphorus-deficient
substrate. Russian Journal of Plant Physiology 51, 396–401.
Vance CP, Uhde-Stone C, Allen DL. 2003. Phosphorus acquisition
and use: critical adaptations by plants for securing a nonrenewable
source. New Phytologist 157, 423–447.
Watt M, Evans JR. 1999. Linking development and determinacy
with organic acid efflux from proteoid roots of white lupin grown
with low phosphorus and ambient or elevated atmospheric CO2
concentration. Plant Physiology 120, 705–716.