Plant Physiol. (1990) 94, 1620-1625
Received for publication February 21, 1990
Accepted May 2, 1990
0032-0889/90/94/1 620/06/$01 .00/0
Assessing the Rhizotoxicity of the Aluminate Ion, Al(OH)4Thomas B. Kinraide
Appalachian Soil and Water Conservation Research Laboratory, Agricultural Research Service, U.S. Department of
Agriculture, Beckley, West Virginia 25802-0867
ABSTRACT
Dissolved aluminum (Ill) in acidic soils or culture media is often
rhizotoxic (inhibitory to root elongation). Alkaline solutions of Al
are also sometimes rhizotoxic, and for that reason toxicity has
been attributed to the aluminate ion, AI(OH)4-. In the present
study, seedlings of wheat (Triticum aestivum L. cv Tyler) and red
clover (Trifolium pratense L. cv Kenland) were cultured in aerated
aluminate solutions at pH 8.0 to 8.9. The bulk phases of these
solutions were free of reactive polynuclear hydroxy-AI (including
the extremely toxic species AlO4Al12[OH]24[H20'J2 [AI13J) according to the ferron (8-hydroxy-7-iodo-5-quinolinesulfonic acid) assay. At an aluminate concentration of 25 micromolar (23 micromolar activity) and a pH of 8, root elongation was less than 40%
of Al-free controls, but at pH 8.9 elongation was 100% of controls.
The hypothesis is offered that aluminate is nontoxic and that the
inhibition at lower pH values is attributable to Al13 postulated to
have formed in the acidic free space of the roots where the ratio
IA13 1/1H+13 may rise above 1010. At this value hydroxy-AI in oversaturated, alkaline solutions begins to undergo rapid conversion
to polynuclear species.
that H+ toxicity has been taken into account (because the pH
of the solutions must change) and one is willing to ignore the
uncertain species composition of the root free space. If possible, all polynuclear Al should be avoided because it may be
unpredictably accompanied by Al,3 at low, but toxic, levels
that are difficult to measure. These considerations were taken
into account in two recent investigations (13, 21), and it was
concluded that A13+ is rhizotoxic to wheat.
Similar investigations with dicotyledonous plants produced
contrary results (14). Al3+ seemed to be nontoxic, and it was
concluded that one or more mononuclear hydroxy-Al species
(e.g. AlOH2" or Al[OH]2+) may be toxic. That conclusion is
tenuous, however, because toxicity to these species cannot be
distinguished from Al3" toxicity that is ameliorated at lower
pH values (14). Perhaps protonation of cell surface sites
reduces Al3+ binding. In addition, Al,3 may have accumulated
in the root free space as mononuclear hydroxy-Al increased
in the bulk media.
Al in alkaline fly ash (9, 25), bauxite residue (6), and
hydroponic culture media (6) appears to be toxic, and that
toxicity has been attributed to aluminate (Al[OH]4) which,
according to species computations, would constitute >99% of
the mononuclear hydroxy-Al at pH > 7.9. However, these
media were complex, and, appropriately, the authors made
no attempt to compute the species composition. In addition,
no attempt was made to detect Al13, whose toxicity was not
widely known at the time of those studies.
The present experiments were performed as part of an
ongoing investigation of the relative rhizotoxicity of Al species. The major emphasis of the study was to employ simple
solutions whose species composition could be computed and
to ensure the absence of Al,3 in the bulk phase of the culture
media.
Aluminum (III) is a major constituent of soil minerals and
is present in the soil solution in a dozen or more chemical
species (15, 26). That one or more of these species is rhizotoxic
has been known for many decades (7), but only recently has
there been progress in determining the relative toxicities of
the various species (4, 8, 12-14, 19, 21, 22). Evidence for the
toxicity of a polynuclear Al species has been offered several
times (2, 21, 22, 27), and this species has now been identified
as Al13' (22). This species is easily prepared in artificial solutions, but upon aging, All3 undergoes a conversion to other
Al species that are unreactive with commonly used assay
reagents and may comprise microcrystalline gibbsite or other
solid-phase Al (3, 22). Al,3 is at least 10-fold more rhizotoxic
than A13+ (22), and the possibility of its occurrence is a
complicating factor in the study ofthe toxicity of other species
(13).
The toxicity of Al3+ is widely assumed, but few experiments
convincingly demonstrate its toxicity. Evidence for its toxicity
may be adduced only in experiments in which plants are
cultured in rooting media whose chemical composition may
be accurately computed. If the toxicity of the media increases
as Al3+ increases and if all other Al species remain constant
or decrease, then one may conclude that Al3+ is toxic provided
MATERIALS AND METHODS
Seedling Culture
Seedlings of wheat (Triticum aestivum L. cv Tyler) and red
clover (Trifolium pratense L. cv Kenland) were cultured at
25°C in aerated, 500-mL solutions as described previously
(10, 13). All alkaline solutions were aerated overnight prior
to the introduction of seedlings. After 2 d, root lengths were
measured and the culture media were assayed for Al by the
ferron method to be described. In a typical experiment intended to test the apparent toxicity of aluminate, the culture
media were prepared by adding 2.95 mL of 1 M NaHCO3, 0
to 50 mL of a 1 mm stock solution of LiAl02 (in 1 mM
NaOH), 50 to 0 mL of a 1 mm stock solution of LiCl (in 1
'Abbreviations: A,3, Al04Al,2(OH)24(H20)1; feffon, 8-hydroxy-7iodo-5-quinolinesulfonic acid; IXI, chemical activity of X in mol/L.
1620
Downloaded from on June 17, 2017 - Published by www.plantphysiol.org
Copyright © 1990 American Society of Plant Biologists. All rights reserved.
ASSESSING THE RHIZOTOXICITY OF ALUMINATE
Table I. Equilibrium Constants Used to Compute Species Activities
Constants for Al species from Nordstrom and May (20); all others
from Lindsay (15).
Equilibrium
Equilibrium Reactions
Constants
log K
Al3+ + H20 = AIOH2+ + H+
Al3+ + 2 H20 = AI(OH)2+ + 2 H+
Al3+ + 3 H20 = AI(OH)30 + 3 H+
Al3+ + 4 H20 = AI(OH)4- + 4 H+
CO2(g,a + H20 = H2CO30
H2CO30 = H+ + HC03H2CO30 = 2 H+ + CO2+
Ca2+ + H2CO30 = CaHCO3+ + H+
Ca2+ + H2CO30 = CaCO30 + 2 H+
-5.00
-10.1
-16.8
-22.7
-7.46
-6.36
-16.69
-5.24
-13.55
-5.27
CaCO3(.dte) = CaCO30
a C02(g) expressed as ,uL C02 per L atmosphere.
mM NaOH), 447 mL water, and, after mixing the preceding,
0.25 mL of 0.1 M CaCl2. Final concentrations were 50 ,gM Ca,
100 ,uM Li, 6 mM Na, 0 to 100 ,M Al, 200 to 100 ,M Cl, and
6 mm base equivalents from NaHCO3 and NaOH. The solutions were set to aerate overnight, and the seedlings were
transferred the next day when the pH was about 8.8.
Polynuclear Al
To assess the stability of mononuclear Al in the stock
solutions and culture media, several sets of solutions were
prepared in order to test the effects of pH and [Al] on the
assayed by the ferron method described previously
(23), but because of some errors in previous descriptions the
details are given here. Three stock solutions were prepared.
Stock solution A contained 1 g 8-hydroxy-7-iodo-5-quinolinesulfonic acid (ferron) and 0.11 g 1,10-phenanthrolineH20 per 1 L; stock solution B contained 580 g sodium acetate.
3H20 per 1 L; and stock solution C contained 100 g hydroxylamine HCl and 40 mL of concentrated HCI per 1 L. The
working solution (1.58 mM ferron) was prepared by mixing
40 mL of each of the stock solutions B and C then adding
100 mL of stock solution A. The pH of this solution was
adjusted to 5.2 by dropwise additions of 6 M HCI. This
working solution was prepared each week for use the following
week. This solution and stock A were stored in the dark.
Al
was
0.112
0.113
0.106
'
Al Assays
~ 2-
Assays were performed by adding a 2 mL combined volume
of water plus Al sample to 12 x 88 mm cuvettes. The ratio
ferron/Al should exceed 50 to accomplish detailed kinetic
analyses (23). In the present study, the ratio was as low as 40
for some assays. At time zero, 1 mL of the ferron working
solution was injected into the cuvette; then the cuvette was
swirled vigorously and inserted into a Bausch and Lomb
Spectronic 21 spectrophotometer set at wavelength 370 nm
and connected to a chart recorder. The absorbance was recorded for at least 5 min, and solutions were considered to be
free of A1l3 if the absorbance was constant after 3 min. Al
fractions have been designated Ala, Alb, and Al, according to
their rates of reaction with assay reagents. Ala (labile or
mononuclear Al) reacts completely with ferron within 3 min;
Alb (A1l3, at least in part) reacts completely within 30 min;
Al, (solid-phase hydroxy-Al or other unreactive Al) reacts
more slowly, if at all (3, 22).
Al Stock Solutions
Stable, acidic, mononuclear stock solutions were prepared
by adding 1 mmol AlCl3-6H20 or AlK(SO4)2. 12H20 to 1 L
of 1 mM HCI. The AlCl3 stock solution was used to prepare
certain growth media, and the AlK(S04)2 stock was used as
an Al standard, because the crystals of AlK(S04)2. 12H20
neither gained weight in humid summer air nor lost weight
at 70°C. Aluminate stock solutions were prepared by adding
1 mmol LiAl02 to 1 L of 1, 10, or 100 mM NaOH in which
A102- hydrates to AI(OH)4- (5). Dissolution of LiAl02 was
slow, and the stocks were prepared by stirring and heating the
mixtures in closed flasks overnight. Occasionally, traces of
solid material persisted. Furthermore, a significant precipitate
eventually formed in the 1 mM NaOH solutions, and in 10
mM NaOH a few new crystals eventually formed also.
1621
_
W
0.106
<c
m
0.097
0
0.098
m
005
--
3
35
6
7
0.05 Abs. units
0
2
4
6
8
TIME (min)
10
Figure 1. Chart records of the ferron assays for Al. At time zero, 1
mL of the assay reagent was injected into the 2 mL mixture of sample
plus water. The written values refer to absorbance after 5 min of
reaction. For records 1 through 4, the cuvettes contained 40 nmol Al
from (1) a 1 mm AIK(SO4)2 stock (in 1 mm HCI), (2) a 1 mM LiAI04
stock (in 100 mm NaOH), (3) a 1 mM LiAI04 stock (in 10 mm NaOH),
and (4) end-of-the-experiment culture medium corresponding to the
left-hand-most open symbol from the lower panel of Figure 6. This
medium contained 25 lM LiAI04 from the stock used for chart record
3. For records 5 through 7 the cuvettes contained 31.6 nmol Al from
1 mm solutions of AICI3 also containing (5) 1 mm HCI, (6) 5 mm NaOH,
and (7) 2 mM NaOH (i.e. no = -1, 5, and 2, respectively).
Downloaded from on June 17, 2017 - Published by www.plantphysiol.org
Copyright © 1990 American Society of Plant Biologists. All rights reserved.
0.120
0
0
O mM NaHCO3 also contained 9 to 0 mM NaCI so that [Nae]
and ionic strength were constant and [Cl-] decreased as pH
and bicarbonate increased.
_
80
0
LLJ
0
6
0.080
Computation of Species Activities
0
z
Species activities were computed by a computer program
written by the author that incorporated the hydrolysis constants compiled by Nordstrom and May (20), the Davies
equation for activity coefficients, and other thermodynamic
data presented by Lindsay (15) (Table I).
m
a:
00
0
U)
m
Plant Physiol. Vol. 94, 1990
KINRAIDE
1 622
0.040
00
0
0.000
0
1~~ ~
9
10
log[{AlI
RESULTS
0
Preliminary Growth Experiments
00
I
Alkaline solutions in equilibrium with air can support only
11
}i/{H
12
i}3
Figure 2. Effects of [Al] and pH on the stability of mononuclearr Al
solutions. Solutions of variable initial pH and [AIC3] were assaye d 3
d after mixing. Forty nmol of Al from the various solutions wo
assayed with ferron, and the absorbance values after a 5 min react
were plotted against log[$A13+l,1/H+Ji3]. Al3+ was computed on the
basis of total added Al.
formation of polynuclear Al. In one set, 10 mL of fres]hly
prepared 10 mM AlCl3 (in plain water) were mixed with
variable amounts of water, then 1 M HCl or 10 mM NaC)H
was quickly mixed in to yield 100-mL solutions that wer(e I
mM Al throughout. The solutions contained mM HCl o Ir O
to 6 mM NaOH so that the ratio of added base to Al (no) vvas
-1, 0, 1, 2, 3, 4, 5, or 6. These solutions were assayed II d
later.
Another set of 28 solutions was prepared by mixing nnL
of 1 M NaHCO3 and 83 mL water then adjusting the pH to
6.8, 7.1, 7.4, 7.7, 8.0, 8.3, or 8.6 with HCl in replicates of
four. Each of these seven groups of four was used to prep;are
four Al solutions in which 2, 4, 8, or 16 mL of 1 mm LiAl102
(in 1 mm NaOH) and 14, 12, 8, or 10 mL of 1 mM LiCl (irn 1
mM NaOH) were added so that the resulting 28 solutions w,ere
approximately factorial in pH (seven values prior to Ithe
addition of Li) and Al (four values equal to 20, 40, 80, or I160
/lM), but the Li and accompanying NaOH were const;ant
throughout. The pH was monitored as the Li solutions wiere
added; the initial postmixing pH values were recorded; Ithe
flasks were stoppered; and the solutions were assayed: d
later.
low [Ca] because of the low solubility of CaCO3 (Table I), but
in alkaline media the requirement for Ca is apparently low,
because 25 uM Ca was adequate for good growth of wheat,
and 50 gM produced only slight improvement. In contrast,
0.4
mm
Ca
was
minimally adequate
report
(1
1),
Li' is toxic
at
at
pH 4.5. According
to
concentrations above 10
mM, but the low levels used here had little effect upon wheat
root elongation. Elongation was sensitive to
[bi]carbonate or
the consequent elevation in pH. Mean root lengths of 49.7,
42.5, and 37.1 mm were obtained in culture media containing
6, 8, or 10 mM NaHCO3, respectively. Because of better
growth at 6 mm bicarbonate, that level of initial base equivalent (NaHCO3 or NaOH) was used in most experiments. The
80
I
I
I
I
A
E
60
bo8
H-
z
0
40
w
J
6
S
20
0
0
0
0
20
40
60
80
100
Effect of pH on Aluminate Toxicity
Two experimental designs were used to test the effect of pH
(or [bi]carbonate level) on apparent aluminate toxicity. In the
first design, the solutions contained 1 to 10 mM NaHCO3 (at
1 mM intervals) plus 50 ,M CaCl2 and 25 jM LiAl02 or LiCl
from 1 mm stocks in 10 mM NaOH. In these experiments,
[Na+] and ionic strength increased as pH and bicarbonate
increased. In the second design, the solutions containing 1 to
[Al] (pM)
Figure 3. Length of wheat roots as a function of [Al] at high and low
pH. Circles refer to seedlings cultured at pH 8.7 in media prepared
from a 1 mm LiAI04 stock in 1 mm NaOH (closed circles) or 10
mM NaOH (open circles). Triangles refer to seedlings cultured at
pH 4.5 in media prepared from AIC13. At the higher pH 1AI(OH)4-1
0.929[AI], and at the lower pH A13+1
Downloaded from on June 17, 2017 - Published by www.plantphysiol.org
Copyright © 1990 American Society of Plant Biologists. All rights reserved.
=
0.360[AI].
=
ASSESSING THE RHIZOTOXICITY OF ALUMINATE
25
E
20
I
I
F
I
0*
E
15
A
0
0
z
w
10
A
I
I
0
0
5
S
0
A
AAAA
0
0
20
40
60
80
100
[Al] (gM)
Figure 4. Length of red clover roots as a function of [Al] at high and
low pH. See legend for Figure 3.
culture conditions suitable for wheat also proved adequate for
red clover.
Formation of Polynuclear Al
According to the ferron assays none ofthe Al stock solutions
contained Alb, i.e. absorbance readings were constant after 3
min (Fig. 1, chart records 1, 2, and 3). However, there were
small losses of solid-phase Al in the 1 and 10 mm NaOH
stock solutions, but in the stock solutions used for growth
experiments solid-phase losses were <10%. (Cf the terminal
absorbance values of records 1, 2, and 3 in Fig. 1.)
The alkaline, Al-containing culture media were oversaturated with respect to gibbsite if one assumes that K5 = JA13+}/
JH+13 = 108.' for the dissolution of synthetic gibbsite (16)
according to the reaction Al(OH)3(gibbsite) + 3H+ = A13+ + 3
H20. Nevertheless, the ferron assays indicated no loss of
mononuclear Al added to culture media despite the fact that
the ratio IAl3+I/tH+3 was occasionally as high as 10101 (Cf
records 3 and 4 in Fig. 1.)
To test the minimal conditions required to support mononuclear Al in oversaturated solutions, several solution series
were prepared. The first set of solutions was intended, in part,
to test the stability of 1 mM Al stock solutions adjusted to no
values of -1, 0, 1, 2, 3, 4, 5, and 6. Loss of mononuclear Al
occurred only in the middle four solutions after 1 d of
incubation. (Cf record 7 with records 5 and 6 in Fig. 1.) The
assays for solutions no = 1 and 2 indicated copious production
of Alb, and solutions no = 3 and 4 contained large amounts
of Alc. Solution no = 4 contained an abundant, visible precipitate, but all other solutions were clear. According to equilibrium computations all of the solutions no 0 through 6 were
saturated with respect to gibbsite ( 16). Thus the mononuclear
solutions no = 0, 5, or 6 were actually metastable if the
1623
reported K, for gibbsite is correct, but the stock solution
prepared by adding 1 mm LiAI04 to 10 mm NaOH should
have been stable with respect to gibbsite.
The second solution series was prepared from an aluminate
stock and entailed much finer pH increments and four [Al].
The solutions were assayed 3 d after mixing. In Figure 2 the
absorbance reading after 5 min reaction with ferron was
plotted against log[JAl3`Ji/$H+i3] in which JAl3"i is the Al3"
activity computed as though all the added Al were present in
mononuclear form and IH+ji is the H+ activity computed
from the initial, postmixing pH. One can see from Figure 2
that Al remained mononuclear for at least 3 d in solutions in
which log[JAl3`Ji/JH+Ji3] < 10. As log[JAl3`i/JH i13] increased, the extent of polynuclear aggregation increased, but
ALC, rather than Alb, predominated.
Occasionally, measured mononuclear Al in the culture
media was a few per cent less than the nominal value when
assayed against the acidified AlK(SO4)2 standard, but this loss
was attributable to small amounts of AL in the aluminate
stocks (see above) rather than losses of mononuclear Al in the
culture media. (See records 1 through 4 in Fig. 1.)
Growth Responses to Aluminate Concentration
Aluminate-containing solutions were toxic to both wheat
and red clover, but in comparison to some other Al species
the apparent toxicity of aluminate was weak. Figures 3 and 4
present root lengths as a function of [Al] at both high and low
pH. Similar experiments at two [Al] ranges were performed
at the higher pH. In the first experiment, the solution inputs
were 0 to 100 AM LiAl02 (from a 1 mM NaOH stock), 100 to
O AM LiCl (from a 1 mM NaOH stock), 50 AM CaC12, and 5.9
mM NaHCO3. In the second experiment, the inputs were 0 to
20 AM LiAl04 (from a 10 mM NaOH stock), 20 to O AM LiCl
(from a 10 mM NaOH stock), 25 (wheat) or 50 (red clover)
AM CaCl2, and 5.8 mm NaHCO3. The initial pH of these
solutions after equilibrium with incubator air was 8.8, and
after 2 d of growth the pH was 8.7. The CO2 in the incubator
sometimes rose to 490 ML/L as measured with an Analytical
Development Co. model LCA-2 infrared CO2 analyzer calibrated at 0 and 585 ML/L CO2.
At the lower pH the solution inputs were 0 to 20 AM AlCl3
(from a 1 mM HCI stock), 100 AM LiCl, 0.4 mM CaCl2, 6.0
mM NaCl, and HCI to adjust the pH to 4.5. The computed
activities of the mononuclear Al species at both high and low
Table II. Computed Activities of Mononuclear Al Species in Culture
Media at High and Low pH
See text for details of the composition of the media. This tabulation
assumes a total [Al] of 10 gM and the nonformation of polynuclear
Al. Activities at other [Al] up to 100 AM would be proportional.
pH 8.7
pH 4.5
Species
M
Al3+
AIOH2+
10- 113
3.60
10o-7.43
Al(OH)2+
1 o-3.83
1.14
0.286
1 0-2.74
Al(OH)30
Al(OH)4-
0.0147
9.29
Downloaded from on June 17, 2017 - Published by www.plantphysiol.org
Copyright © 1990 American Society of Plant Biologists. All rights reserved.
10-4'14
.9A 0%0%
a.
1W
(root length in a similar, aluminate-free solution -c) where c
= 20 mm and is the length of wheat roots at transfer to the
test solutions plus an increment equal to the slight elongation
in saturating levels of Al (13). In red clover, the nontoxicity
of the more alkaline aluminate solutions was evident whether
root elongation was expressed on a relative or absolute basis.
Ferron assays of the culture media from these experiments
were performed with particular care, especially at the lower
pH values where polynuclear aggregation would be most
likely. Added mononuclear Al was recovered entirely and Alb
was absent according to the criterion of no rise in absorbance
after 3 min (cf records 3 and 4 in Fig. 1).
.
A
75 I
50 0
251-
_O
O
~~~~~0
0
I
z
UJ
0
-j
G
Plant Physiol. Vol. 94, 1990
KINRAIDE
1 624
14
B
DISCUSSION
75
50
50
Aluminate solutions (but not necessarily aluminate itself)
are sometimes toxic (6, 9, 25, and this study), but the present
study demonstrates that aluminate need not be toxic always
(Figs. 5 and 6). The relative root lengths of both wheat and
0
25
20
n
v
I
Ca
9.0
10.2
16
8.8
10.0
12F
8.6
9.8
9.6
8j
I
0
0
0
o
8.4
9.4
8.2
9.2
8.0
9.0
0
2
4
6
8
10
+
C.,
0)
0
[NaHCO3J (mM)
E
E
0
z
w
-J
0
0
S~~
0
0
0
41A
0
20
~
a.
.
16
a
0
0
0
0 12 -
Figure 5. Rhizotoxicity of aluminate solutions to wheat as a function
of pH or [bi]carbonate levels. Solutions contained 25 ,M LiCI (closed
circles) or 25 uM LiAI02 (open circles). Panel A refers to solutions in
which [NaHCO3] changed without compensation with NaCI. Panel B
refers to solutions in which NaCI compensated changes in NaHCO3
so that [Na+], [Cl-] + [HC03+], and ionic strength remained approximately constant. The curves in panel C refer to the terminal pH values
of the solutions in both upper panels or to the indicated log activity
ratios of the solutions in both panels. Each omitted data point in this
panel fits the drawn curve with an error of <5% of the corresponding
scale.
0
4-
B
0
9.0
I
I
10.2
I
10.0 I
9.8
9.6 'r
8.8
8.61
pH is presented in Table II. Apparently 3.6
more toxic than 93 AM $AI(OH)4-1.
gM tA131+
was
pH Dependency of Apparent Aluminate Toxicity
Aluminate solutions became nontoxic at higher pH (and
[bi]carbonate values (Figs. 5 and 6). In the Al-free control
solutions, wheat root elongation was inhibited by increases in
pH from 8.0 to 8.9, but elongation in the aluminate solutions
was unaffected so that relative root length increased. Relative
root length = (root length in the aluminate solution - c)/
-r
I
I
8.41.
9.4
8.2
9.2
R-f
9.0
0
2
4
6
8
10
C,,
0)
0
[NaHCO3J (mM)
Figure 6. Rhizotoxicity of aluminate solutions to red clover as a
function of pH or [bi]carbonate levels. See legend for Figure 5.
Downloaded from on June 17, 2017 - Published by www.plantphysiol.org
Copyright © 1990 American Society of Plant Biologists. All rights reserved.
ASSESSING THE RHIZOTOXICITY OF ALUMINATE
red clover rose from less than 40% to about
100% as the pH
aluminate solutions rose from about 8.0 to about
8.9. No changes in the chemistry of the culture media readily
explain this amelioration of toxicity with increasing pH. No
Alb could be detected, and no other known Al species, besides
aluminate, could achieve an activity exceeding 0.2 jLM at pH
8.0 if the hydrolysis constants in Table I are correct. Amelioration by Na+ or increased ionic strength was ruled out by
various experiments, and the possibility that C032- could
interfere with the action of the anionic aluminate (as cations
apparently interfere with the action of A13+ [1 1]) seems remote
for two reasons: [CO32-] < 0.5 mm, and S042- was not
ameliorative at concentrations up to 5 mm (data not presented); HC03- (8 mm at pH 9) may have been ameliorative,
but Cl- was not ameliorative at 9 mm.
of the 25
jM
My conclusion is that aluminate is not toxic but that A113,
another toxic Al species, was able to form in the acidic
root free space when the pH of the bulk solution was not too
or
high. The root free space was acidic even in the alkaline
culture media. When 2 mg of neutral red were added to
culture media at the end of a growth experiment, the medium
turned amber but the roots turned bright magenta indicating
a pH well below 7 (data not presented). As the pH of the
culture media declined, the pH of the root free space probably
declined also so that A1l3 may have formed as the ratio $Al3+j/
IH+ 3 rose above 10'0. The failure of Alb to appear in the bulk
media is not surprising. If A1l3 formation were confined to
the root free space, its high positive charge would keep the
A113 electrostatically bound to the fixed negative charges of
the root (17), and any A113 that escaped to the medium may
have undergone conversion to Al. Furthermore, very little
Al13 would need to form because of its extreme toxicity (22).
The present results may appear to be discrepant with previous reports in which toxicity was observed at pH values
higher than 9 (6). In those studies, however, the total [Al] was
very high (1 mM); the bulk solutions contained copious
amounts of polynuclear Al; and exposures were of longer
duration. Consequently, A1,3 may have been present in the
root free space even at the higher bulk-solution pH.
Future research should include an attempt to detect A113
under mixing conditions similar to those that may prevail in
the root free space: slow injection of alkaline Al into pHstated, weakly acidic solutions, perhaps. Direct tests of the
present hypothesis would be possible if sensitive in situ assays
for A1,3 were available. The effort to develop such assays may
be rewarded by the discovery that several previously observed
instances of Al toxicity that were not attributed to A113 may,
in fact, be attributable to this species (1, 14, 18, 24).
ACKNOWLEDGMENTS
I am grateful to Dr. David Parker for many conversations and
suggestions and for the assistance of Lea Ann Bosley, USDA-ARS
Summer Aide and student at Marshall University.
LITERATURE CITED
1. Baker JP, Schofield CL (1982) Aluminum toxicity to fish in
acidic waters. Water Air Soil Pollut 18: 289-309
1 625
2. Bartlett RJ, Riego DC (1972) Toxicity of hydroxy aluminum in
relation to pH and phosphorus. Soil Sci 114: 194-201
3. Bertsch PM (1989) Aqueous polynuclear aluminum species. In
G Sposito, ed, The Environmental Chemistry of Aluminum.
CRC Press, Boca Raton, FL, pp 87-115
4. Cameron RS, Ritchie GSP, Robson AD (1986) Relative toxicities
of inorganic aluminum complexes to barley. Soil Sci Soc Am
J 50: 1231-1236
5. Cotton FA, Wilkinson G (1966) Advanced Inorganic Chemistry.
Interscience Publishers, New York, p 437
6. Fuller RD, Richardson CJ (1986) Aluminate toxicity as a factor
controlling plant growth in bauxite residue. Environ Toxicol
Chem 5: 905-915
7. Hartwell BL, Pember FR (1918) The presence of aluminum as
a reason for the difference in the effect of so-called acid soils
on barley and rye. Soil Sci 6: 259-279
8. Hue NV, Craddock GR, Adams F (1986) Effect of organic acids
on aluminum toxicity in subsoils. Soil Sci Soc Am J 50: 28-34
9. Jones LH (1961) Aluminum uptake and toxicity in plants. Plant
Soil 13: 297-3 10
10. Kinraide TB, Arnold RC, Baligar VC (1985) A rapid assay for
aluminium phytotoxicity at submicromolar concentrations.
Physiol Plant 65: 245-250
11. Kinraide TB, Parker DR (1987) Cation amelioration of aluminum toxicity in wheat. Plant Physiol 83: 546-551
12. Kinraide TB, Parker DR (1987) Non-phytotoxicity of the aluminum sulfate ion, AlSO4'. Physiol Plant 71: 207-212
13. Kinraide TB, Parker DR (1989) Assessing the phytotoxicity of
mononuclear hydroxy-aluminum. Plant Cell Environ 12: 479487
14. Kinraide TB, Parker DR (1990) Apparent phytotoxicity of mononuclear hydroxy-aluminum to four dicotyledonous species.
Plant Physiol 79: 283-288
15. Lindsay WL (1979) Chemical Equilibria in Soils. John Wiley,
New York, pp 14, 79-98
16. May HM, Helmke PA, Jackson ML (1979) Gibbsite solubility
and thermodynamic properties of hydroxy-aluminum ions in
aqueous solution at 25'C. Geochim Cosmochim Acta 43: 861868
17. Marschner H (1986) Mineral Nutrition of Higher Plants. Academic Press, London, pp 9-12
18. Neville CM, Campbell PGC (1988) Possible mechanisms of
aluminum toxicity in a dilute, acidic environment to fingerlings and older life stages of salmonids. Water Air Soil Pollut
42: 311-327
19. Noble AD, Summer ME, Alva AK (1988) Comparison of aluminon and 8-hydroxyquinoline methods in the presence of
fluoride for assaying phytotoxic aluminum. Soil Sci Soc Am J
52: 1059-1063
20. Nordstrom DK, May HM (1989) Aqueous equilibrium data for
mononuclear aluminum species. In G. Sposito, ed, The Environmental Chemistry of Aluminum. CRC Press, Boca Raton,
FL, pp 29-53
21. Parker DR, Kinraide TB, Zelazny LW (1988) Aluminum speciation and phytotoxicity in dilute hydroxy-aluminum solutions.
Soil Sci Soc Am J 52: 438-444
22. Parker DR, Kinraide TB, Zelazny LW (1989) On the phytotoxicity of polynuclear hydroxy-aluminum complexes. Soil Sci
Soc Am J 53: 789-796
23. Parker DR, Zelazny LW, Kinraide TB (1988) Comparison of
three spectrophotometric methods for differentiating monoand polynuclear hydroxy-aluminum complexes. Soil Sci Soc
Am J 52: 67-75
24. Pettersson A, Hillbom L, Bergman B (1985) Physiological and
structural responses of the cyanobacterium Anabaena cylindrica to aluminum. Physiol Plant 63: 153-158
25. Rees WJ, Sidrak GH (1955) Plant growth on fly-ash. Nature
176:352
26. Stevenson FJ, Vance GF (1989) Naturally occurring aluminumorganic complexes. In G. Sposito, ed, The Environmental
Chemistry of Aluminum. CRC Press, Boca Raton, FL, pp
117-145
27. Wagatsuma T, Kaneko M (1987) High toxicity of hydroxyaluminum polymer ions to plant roots. Soil Sci Plant Nutr 33:
57-67
Downloaded from on June 17, 2017 - Published by www.plantphysiol.org
Copyright © 1990 American Society of Plant Biologists. All rights reserved.