a"

 

 

 

a an» a
5.41:5: . a
,

..ﬁ

.35,

.... ax!“ 5,
in...“ «3:.

 

 

it! 4 1.-
li: A .4.

3')...- .H.
. - unit. .a: (Jan. “-9...
Ex. 2 ﬁx: LT. .

‘Y a. . 5.. v
I: . .3“... 9; r.

 

"HESlS

MICHIG

l

NWEHSITY LIBRARIES

ll ill/IIll/lI/lI/II/l/I/II/I/lIll/Ill!!!” NIH/Ill”!

 

 

 

This is to certify that the

dissertation entitled

GROWTH-DEPENDENT ALUMINUM UPTAKE AND DISCRETE
LOCALIZATION IN CULTURED CELLS OF Nicotiana tabacum L.
cv.! BY-Z: A POSSIBLE ROLE FOR MEMBRANE TRAFFIC AT

THE CELL SURFACE

 

presented by

Victor Ao Vitorello

has been accepted towards fulﬁllment
of the requirements for

PhoDo degree in Botany and Plant Pathology

 

Palfggaw dot/«>14. t/Héag/

Major professor

Date OQJOSXqu

MS U i: an Afﬁrmative Action/Equal Opportunity Institution 0-12771

 

 

 

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE N RETURN BOX to remove this checkout from your record.
TO AVOID FINES rotum on or More data duo.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

GROWTH-DEPENDENT ALUMINUM UPTAKE AND DISCRETE LOCALIZATION
IN CULTURED CELLS OF Nicotiana tabacum L. cv. BY-2:
A POSSIBLE ROLE FOR MEMBRANE TRAFFIC AT THE CELL SURFACE
By

Victor A. Vitorello

A DISSERTATION
Submitted to
Michigan State University
in partial ﬁﬂﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Botany and Plant Pathology

1 996

ABSTRACT
GROWTH-DEPENDENT ALUMINUM UPTAKE AND DISCRETE LOCALIZATION
IN CULTURED CELLS OF Nicotiana tabacum L. cv. BY-2:
A POSSIBLE ROLE FOR MEMBRANE TRAFFIC AT THE CELL SURFACE
By

Victor A. Vitorello

Short-term aluminum (Al) uptake and cellular localization of Al were studied in
tobacco (Nicotiana tabacum L.) cv. BY-Z cultures, by using an in viva Al-morin ﬂuorescence
assay and graphite furnace atomic absorption spectrometry to detect Al. Results from these
two methods were highly correlated (r2=0.94). Al uptake was dependent on cell growth. In
growing cells, Al uptake was detectable within 5 min and had an initial rate of 16 nmol Al
(10" cells) " h'l , but was negligible in stationary-phase cells. With respect to cellular
localization, Al accumulated in discrete regions (2-30 pm long) at the cell periphery, usually
one region per cell. Frequently, Al-morin ﬂuorescence occurred along the interface (20-30
pm) of adjacent cells, as if along the plane of cellular division. The proportion of cells
displaying localized Al varied from 0 to about 80%. Al-morin ﬂuorescence was frequently
associated with a protuberance of the cell surface and with vesicle-like or globular structures
attached to the exterior surface of the cell and presumably formed in response to Al
treatment. The plasma membrane and the cell wall tended to remain combined at regions of
Al-morin ﬂuorescence, following plasmolysis, but Al was found solely associated with the
protoplast in rare cases of separation. The hypothesis that Al-accumulating regions
correspond to regions of cell growth is raised. Increased external CaClz, low temperature

(4°C), sorbitol (0.1 and 0.4 M), and pre-chelation of Al to citrate all largely reduced Al

uptake (by 75-90%). Low pH (4.5) caused an increase in the percentage of cells permeable
to trypan blue, in contrast to pH 5.6 where almost all cells were impermeable. This low pH-
induced increase in membrane permeability was dependent on cell growth, it could be
ameliorated by factors such as increased CaCl2 and it may play an important role in Al
uptake. The inhibitors cremart (15-30 uM), brefeldin A (10 uM), caffeine (1-5 mM) and 2,6-
dichlorobenzonitrile (20 uM) had no detectable effect on Al uptake. Growing cells lost their
capacity for Al uptake when incubated in a minimal medium (0.5 to 3 h) prior to exposure
to Al. However, Al uptake was sustained by treatment of cells with low concentrations (10
uM) of brefeldin A, an inhibitor of the Golgi secretory pathway. A physiological change in
these cells, associated with vesicular trafﬁc, is apparently necessary for cessation of Al

uptake, but is reversibly blocked by brefeldin A.

To Claudia, Tomas and Gabriela,

To my parents and sisters,

To those who love and truly dedicate themselves to mentoring.

iv

ACKNOWLEDGMENTS

I would like to thank Dr. Alﬁed Haug, my research advisor, without whom this work
would not have been possible. I am also grateful to Dr. Robert Bandurski, my academic
advisor, and Dr. Patrick Hart and Dr. John Ohlrogge of my guidance committee.

I wish to express my appreciation for the ﬁnancial support from the Brazilian federal
granting agency CNPq, grant # 200892/91-6 and by the University of 850 Paulo, Brazil. I
also thank additional ﬁnancial support from the Department of Botany and Plant Pathology
and the Graduate School at MSU. My thanks to those who helped me in several ways. Dr.
Robert Hausinger, Dr. Matthew Zabik, Dr. Alan Jones, Dr. Leah Bauer and Debbie Miller
for use of equipment. Dr. Joanne Whallon and Dr. Shirley Owens for help on the confocal
microscope. David Shintani and Mike Sullivan for providing me with the tobacco cell
cultures. Kurt Stepnitz for the preparation of slides and photographs.

Thanks to members of this department and to Dr. De Zoeten for their support and
encouragement. I am also grateful to Dr. Holly Schaeffer, for support and critical readings.
To my ﬁiends, I thank Ray Pacovsky, for ﬁ'uitful discussions and encouragement, and Jorge
Rodrigues and Maeli Melotto, for their years of friendship and for their always prompt help.
I also thank my Brazilian ﬁiends in East Lansing and at CENA, University of 850 Paulo.

Finally, special thanks to Dr. Ken Poff, who helped me change my perspective on
science and life, who gave me encouragement, and who. through his example, gave me

important lessons on mentoring.

TABLE OF CONTENTS

LIST OF TABLES ..................................................... ix

LIST OF FIGURES .................................................... x

ABBREVIATIONS ................................................... xiii
CHAPTER 1

INTRODUCTION AND REVIEW OF LITERATURE ....................... 1

Chemistry of Aluminum ...................... y ...................... 2

General aspects of Al toxicity in plants ................................. 5

Aluminum toxicity in animals ....................................... 7

Low pH effects on roots ............................................. 9

A1 uptake in plants ................................................ 10

Sites of Al accumulation in roots ............................... 10

Al uptake at the cellular level: apoplastic or symplastic? ............ 1 1

Factors impacting Al uptake .................................. 13

Mechanisms of Al toxicity .......................................... 14

Al tolerance: mechanisms, genetics and Al-induced genes and proteins ....... 15

Processes involved in cell growth/secretion and Al toxicity ................ 16

REFERENCES .................................................. 21
CHAPTER 2

DEVELOPMENT AND VALIDATION OF AN IN VIVO ALUMINUM-MORIN
FLUORESCENCE ASSAY FOR THE VISUALIZATION AND QUANTITATIVE
DETECTION OF ALUMINUM IN TOBACCO BY-2 CELL SUSPENSIONS . . . 32

ABSTRACT ..................................................... 32
INTRODUCTION ................................................ 34
MATERIALS AND METHODS ..................................... 37
Cell cultures and Al uptake ................................... 37
Morin solutions and cell loading ............................... 38
Fluorescence intensity measurements ........................... 39
Graphite Furnace Atomic Absorption Spectrometry ................ 40
RESULTS ...................................................... 41
Morin uptake by cells ........................................ 41
Spectra ................................................... 41
Optimization of Al-morin ﬂuorescence intensity .................. 45
Comparison of the Al-morin method to GFAAS ................... 49

vi

DISCUSSION ................................................... 53
REFERENCES .................................................. 56

CHAPTER 3
SHORT-TERM ALUMINIUM UPTAKE BY TOBACCO CELLS: GROWTH
DEPENDENCE AND EVIDENCE FOR INTERNALIZATION IN A DISCRETE

PERIPHERAL REGION ............................................... 58
ABSTRACT ..................................................... 58
INTRODUCTION ................................................ 58
MATERIAL AND METHODS ...................................... 59

Chemicals and labware ...................................... 59
Cell cultures ............................................... 59
Aluminium uptake .......................................... 59
Analysis of Al by graphite furnace atomic absorption spectrometry . . . . 59
Morin ﬂuorescence assay ..................................... 60
RESULTS ...................................................... 6O
Dependence of Al uptake on cell growth ......................... 60
Time-course of Al uptake .................................... 60
Aluminium uptake depends on Ca *, temperature and
pre-chelation to citrate ................................. 61
Distribution of morin in cells and cellular localization of Al ......... 63
Membrane permeability at low pH and Al uptake .................. 64
DISCUSSION ................................................... 64
REFERENCES .................................................. 66
CHAPTER 4

CESSATION OF ALUMINUM UPTAKE AND ALUMINUM-INDUCED
ABNORMALITIES OF THE CELL SURFACE DEPENDS ON BREFELDIN

A-SENSITIVE MEMBRANE TRAFFIC IN TOBACCO BY-2 CELLS ......... 67
ABSTRACT ..................................................... 67
INTRODUCTION ................................................ 68
MATERIALS AND METHODS ..................................... 70

Chemicals and labware ...................................... 70
Cell cultures and Al uptake procedure ........................... 70

Treatment of cells with inhibitors of growth-associated processes . . . . 71
Labelling of cells with morin and measurements of

ﬂuorescence intensity .................................. 72
Microscopy ................................................ 73
RESULTS ...................................................... 74
Confocal laser scanning microscopy of Al-accumulating
regions in cells ....................................... 74
Cells lose their capacity for Al uptake when incubated in a minimal
medium ............................................ 82
Effect of inhibitors of growth-associated processes on Al uptake ...... 82
Brefeldin A sustains Al uptake capacity by cells ................... 86

vii

DISCUSSION ................................................... 93

REFERENCES ................................................. 100
CHAPTER 5
GENERAL DISCUSSION AND PERSPECTIVES ......................... 104
Future perspectives ........................................ 110
REFERENCES ................................................. 113
APPENDIX A
GROWING TOBACCO CELLS RESPOND TO RAPID MEDIUM CHANGES
WITH A TRANSIENT INCREASE IN LOCALIZED Ca2+ ACTIVITY ....... 116
APPENDIX B
CELLULAR ASPECTS OF ALUMINUM TOXICITY IN PLANTS .......... 130
APPENDIX C
ALUMINUM INTERACTION WITH PHOSPHOINOSITIDE-ASSOCIATED
SIGNAL TRANSDUCTION ............................................ 137
APPENDIX D
ALUMINIUM COORDINATION TO CALMODULIN: THERMODYNAMIC
AND KINETIC ASPECTS ............................................. 144

viii

LIST OF TABLES

Chapter 3.

Table 1. Effects of low temperature, pre-chelation of Al to citrate, and
pre-wash pH on Al uptake in cultured tobacco cells, 3 days after transfer ....... 62

Table 2. Percentage of cells (3 days after transfer) permeable to trypan blue
when incubated in medium at pH 4.5 or pH 5.6, and as affected by Ca2+
concentrations and presence of 20 uM Al. ............................... 64

Table 3. Percentage of cells permeable to trypan blue after 1h Al incubation
as affected by growth stage of cells and pH of pre-wash medium .............. 64

Chapter 4.

Table 1. Growth of tobacco BY-2 cells following transfer to fresh culture medium,
incubation in minimal medium, or culture in the presence of brefeldin A (BF A). . 85

Table 2. The effect of cremart, dichlorobenzonitrile (DCB), caffeine
and brefeldin A (BF A) on Al uptake in tobacco BY-2 cells. .................. 87

ix

LIST OF FIGURES

Chapter 2.

Figure 1. Structure of the ﬂavonoid morin (3, 5, 7, 2', 4' - pentahydroxyﬂavone) and
presumed structure of its complex with Al ............................... 36

Figure 2. Uptake of morin into cells of tobacco BY-2 cultures and ﬂuorescence
intensity from Al-treated cells as a function of duration of incubation
with morin ........................................................ 42

Figure 3. Excitation and emission spectra of free morin and Al-morin in tobacco
BY-2 cells (A) and in an aqueous or ethanol/water/acetate solution (B). ........ 43

Figure 4. Dependence of cell Al-morin ﬂuorescence intensity on the concentration
of morin in the incubation medium and the cell titer during readings (insert). . . . . 46

Figure 5. Readings of Al-morin ﬂuorescence intensity from tobacco BY-2 cells
as affected by movement and sedimentation of cells and stability of
the Al-morin ﬂuorescence signal with time (insert). ........................ 48

Figure 6. Correlation between results of Al uptake by tobacco BY-2 cells
measured by the Al-morin method or graphite furnace atomic absorption
spectrometry (GFAAS). ............................................. 50

Figure 7. Aluminum uptake by tobacco BY—2 cells and by cellulose phophate,
at pH 4.5 and 5.6, as reported by Al-morin ﬂuorescence intensity or GFAAS. . . . 52

Chapter 3.
Figure 1. Growth stage-dependent Al uptake in cultured tobacco cells .......... 61
Figure 2. Time course of Al uptake by cultured tobacco cells ................. 61

Figure 3. Photomicrographs of ﬂuorescence from morin-labelled tobacco
cells in the absence of added Al ........................................ 62

Figure 4. Photomicrographs of ﬂuorescence from morin-labelled tobacco
cells in the presence of added Al ....................................... 63

Chapter 4.

Figure l. Photomicrographs of Al accumulation in deformities of
the cell surface which originated after Al treatment. ........................ 75

Figure 2. Photomicrographs of Al accumulation in structures at the
region of contact between two adjacent cells. ............................. 77

Figure 3. Photomicrographs of plasmolyzed, Al-treated cells. ................ 79

Figure 4. Aluminum uptake ceases in tobacco BY-2 cells after incubation
in a minimal medium, but not in cells treated with brefeldin A. ............... 83

Figure 5. The effect of the duration of pre-incubation with EPA, prior
to incubation in minimal medium, on blockage of the cessation of A1
uptake in tobacco BY—2 cells, by BFA. .................................. 88

Figure 6. Dose dependence curve for the suppression, by brefeldin A,
of the cessation of Al uptake by tobacco BY-2 cells, induced by incubation
in minimal medium. ................................................. 90

Figure 7. Reversibility of the effects of brefeldin A on loss of Al uptake
capacity (A) and its effect after some loss of Al uptake capacity has
already occurred (B) in tobacco BY-2 cells. .............................. 91

Chapter 5.
Figure l. A schematic representation of a model for uptake of Al into regions

of growing cells involved in the incorporation of new cell wall and
plasma membrane. .................................................. 108

Appendix A

Figure 1. A: Cellular Ca2+ activity, reported by chlorotetracycline ﬂuorescence,
as a function of the growth stage of tobacco cultures. ....................... 121

Figure 2. Changes in Ca2+ activity of growing tobacco cells, as reported by
chlorotetracycline ﬂuorescence, following incubation in a minimal medium, or

maintenance in culture medium. ....................................... 122

Figure 3. Photomicrographs of growing tobacco cells (2 days after subculture)
labelled with chlorotetracycline (50 MM, 15 min). ......................... 123

xi

Figure 4. Growth of tobacco cells following incubation in different media,
measured 24 h after beginning of treatment. .............................. 126

Appendix C

Fig. 1 Simplified view of signal transduction along the Ca2‘/phosphoinositide
pathway .......................................................... 138

xii

ABBREVIATIONS

ATP ......................... adenosine triphosphate

BF A ......................... brefeldin A

CCCP ........................ carbonyl cyanide m-chlorophenylhydrazone
DCB ......................... 2,6-dichlorobenzonitrile

DMSO ....................... dimethylsulfoxide

DNP ......................... 2, 4 dinitrophenol

EDTA ........................ ethylenediamine tetraacetic acid

GFAAS ...................... graphite furnace atomic absorption spectrometry
GTP ......................... guanosine triphosphate

Mes .......................... 2-(N-morpholino) ethanesulfonic acid
NTA ......................... nitrilotriacetate

PCV ......................... packed cell volume

PE ........................... phosphatidylethanolamine

PS . . . . . . ..................... phosphatidylserine

RGR ......................... relative growth rate (% per day)

SE ........................... standard error

xiii

CHAPTER 1

INTRODUCTION AND REVIEW OF LITERATURE

INTRODUCTION AND REVIEW OF LITERATURE

Aluminum toxicity is a major soil constraint to food and biomass production
throughout the world (F oy et al., 1978; Haug, 1984; Godbold et al., 1988). Its importance
stems from the area covered, effects on crop productivity and often difﬁcult correction.
Estimates indicate that 40% of the arable soils of the world are acidic and may therefore
present aluminum toxicity hazards (Osmond et al., 1980). Most of these areas are located in
third world countries in South America, central Africa and southeast Asia. In Brazil, 580
million hectares are affected by acidic soils, covering 70% of its total territory. This is the
largest area of acidic soils within a single country and corresponds almost to 75% of the
continental USA. The problem is complicated by the difﬁculty in liming subsoil layers or
forested areas (F oy et al., 1978), even when the farmer does have access to the necessary
capital and machinery.

Although Al toxicity has received considerable attention over the last 30 years, it is
still poorly understood. There are numerous, but unsubstantiated, theories on the causes of
Al toxicity in the literature (Kochian, 1995; Jones and Kochian, 1995). The cellular
components and processes which have been reported to be affected by A1 are wide ranging,
such as mitosis and cell division (Matsurnoto et al., 1976), physical properties and structure
of the plasma membrane (V ierstra and Haug, 1978; Zhao et al., 1987), uptake of Ca2+ and
other ions (Huang et al., 1992), phosphoinositide-mediated signal transduction (Haug et al.,

1994 - Appendix B; Jones and Kochian, 1995), and actin and tubulin (MacDonald et al.,

1

1987; Grabski and Schindler, 1995).

Given the potential for Al to interact with numerous crucial cellular components such
as ATP, GTP and phosphates, and thereby disrupt Mg-associated functions (Petterson and
Bergman, 1989), the search for the site(s) of toxicity may reveal numerous targets. To
illustrate, Crapper Mclachlan (1989) compiled a partial list of 46 processes and functions
reportedly damaged by Al. Many of these processes involved phosphate metabolism and
many were performed in vitro. That is, Al can be deterimental to many functions if it has
access to them. Just as important, or maybe even more so, is understanding how Al is taken
up by cells and reaches its targets. However, current understanding of most apects of Al
uptake is also controversial and ambiguous (Lazof et al., 1994). Limitations for advances in
this ﬁeld are largely methodological (Taylor, 1988; Delhaize and Ryan, 1995).

It was my objective in this study to characterize Al uptake, with respect to kinetics
and cellular localization, and to examine factors affecting uptake. To achieve this goal, an
in vivo ﬂuorescence technique, based on the complex of morin with Al, was developed and
optimized to detect Al in cells. This technique was then used to examine factors impacting

Al uptake by Nicotiana tabacum L. cv BY-2 cells.

Chemistry of Aluminum

Aluminum is the most abundant metal in the earth’s crust (Hem, 1986). Mineral soils
contain signiﬁcant amounts of Al, most of which is locked in aluminosilicates or aluminum
oxides (Sposito, 1989) of the clay fraction and does not pose a toxicity hazard. Upon

acidiﬁcation of a soil, a fraction of the total Al becomes soluble and potentially toxic to

3

plants. Thus, an acidic soil is almost synonymous with an Al-toxic soil. Acid soils cover up
to 40 % of the arable soils of the world (Osmond et al., 1980) and are located mainly in
tropical and subtropical areas, where soil acidity is a natural occurrence. The largest areas
are in northern and central South America, central Africa and southeastem Asia. In temperate
zones, soil acidity has become an increasing problem as a result of acid rain. In North
America, problem areas are in the northeast, in the industrial regions of the USA and Canada,
mainly because of acid rain, and also in the southeastern USA.

With its high ionic charge and small ionic radius, Al has a ratio of charge-over-radius
(z/r = 5.9) second only to beryllium (Orvig, 1993). Because of this, Al strongly polarizes the
water molecules in its hydration shell (Baes and Mesmer, 1976). With a coordination
number of six, Al is coordinated by six water molecules in an octahedral conﬁguration. The
high degree of polarization of the OH bond can result, depending on the pH of the medium,

in the dissociation of a proton:
Al(H20)63+ -‘= Al(H20)5(OH)2+ + H+

For simplification, these forms of A1 are represented without designating the associated
water molecules, as in Al3+ or Al(OH)x’. Aluminum undergoes a pH-dependent hydrolysis

series, as represented below:

A1” a, ARCH)” ‘—-r A1(0H); *7 AI(OH)3° r AKOH)!

I l
Aln (0H),,‘3“"”+ precipitates
polynuclear complexes

4

The neutral Al hydroxide species, Al(OH)3°, is insoluble and thus Al is largely insoluble in
the pH range between 5 and 7. At pH values above 7.5, Al(OH)4' is formed and Al is again
soluble. Under conditions with a high Al/OH' ratio, polynuclear species can form from
Al(OH)2*, of which the “AID“ tridecameric polycation is the most important (Parker and
Bertsch, 1992). In addition, hydrolysis and solubility of aluminum can be affected by
chelation. In the presence of such agents, prediction of the fate of Al is difficult.

Because of its high ratio of charge-to-radius, Al is a hard, highly reactive cation, and
binds preferentially to hard negative donor groups (Da Silva and Williams, 1991). Fluoride,
being the most electronegative of the anions, is the preferred inorganic monodentate ligand.
However, since it is not incorporated into multidentate ligands, it is oxygen containing
moieties of multidentate molecules which bind Al (Orvig, 1993). The most important ones
are carboxyl (-COOH), hydroxyl (-OH), carbonyl (-CO) and phosphate (-PO3) groups.
Sulfhydryl groups do not bind Al, even when part of a chelate ring (Toth et al., 1984).
Amines bind Al only when part of multidentate ligands such as NTA and EDTA (Martin,
1992)

Of the biological elements, the Al ion is closest in size (r=0.054 nm) to Mg2+ (r=0.072
nm) and the fen'ic ion F e3+ (r=0.065 nm), and it is size, not charge which is most important
in metal ion substitution (Martin, 1988; Williams, 1992). Magnesium is closely associated
with phosphate metabolism and use in cells (Williams, 1993). Al can bind to nucleoside
triphosphates with an association constant 107 times that of Mg”, a cofactor for hydrolysis
of nucleotides. Of special importance is the fact that Al has a slow rate of exchange in and
out of its coordination sphere (Orvig, 1993). Ligand exchange rates for A1 are of the order

of 1.3 s", 10’-fold slower than for Mi (Martin, 1992). These properties alone have the

capacity to explain toxicity of Al.

Fluoroaluminate complexes, most likely AlF4', can act as pseudophosphates, given
structural similarities. By binding to nucleoside diphosphates, AlF4' can act as an analogue
of the terminal y-phosphate of triphosphates. F luoroaluminates are not considered as the
toxic species of Al, but, like GTPyS, have been used as a valuable experimental tool in
studies of biochemical pathways involving GTP binding proteins. Since the ﬂuoroaluminate
acts as a non-hydrolyzable y-phosphate, these proteins are locked in an “activated” state
(Chabre, 1990; Haug et al., 1994 - Appendix B). Complexation of A1 with ﬂuoride can
reduce Al toxicity and actually stimulate growth, as exempliﬁed with pollen tubes (Konishi
and Miyamoto, 1983). Some plants (e.g. tea) accumulate ﬂuoride, and soils can be
contaminated with ﬂuoride as a result of industrial activity (Braen and Weinstein, 1985).
However, the role of ﬂuoride, in Al toxicity, from uncontaminated soils and from

endogenous levels in plants, still needs to be well characterized.

General aspects of Al toxicity in plants

Aluminum toxicity is primarily root associated. Although symptoms of toxicity are
manifested in the shoots, these are generally considered to be a consequence of the unhealthy
root system. With the exception of Al accumulating plants (a classic example is tea), little
Al is transported into the shoot (F oy et al., 1978). The initial and most dramatic symptom of
Al toxicity is inhibition of root growth. Al-inj ured roots become short, stubby and thickened,
with a brownish coloration. Fine branching and root hairs are reduced and the root system

takes on a “corraloid” appearance (Taylor, 1988). Cells which are affected are the root cap,

6

meristem, elongating cells, root hairs and branch initials (Bennet et al., 1987; Jones and
Kochian, 1995). Consequences can be very severe, with total loss of crop productivity. The
site of Al-induced inhibition of root growth is localized in the root apex, and has been
demonstrated by exposing only certain regions of the root to Al (Ryan et al., 1993). Root
growth inhibition is rapid, and can be detected within 1-2 hours of Al exposure (Van et al.,
1994; Ownby and Popham, 1989; Ryan et al., 1992). At least initially, this rapid inhibition
is primarily the result of inhibited cell elongation, rather than reduced cell division (Kochian,
1995; Jones and Kochian, 1995).

With the exception of some of the alterations in root morphology, overall toxicity
symptoms can resemble Ca2+ deﬁciency. Since this was ﬁrst recognized, a connection
between A1 toxicity and Ca2+ metabolism has been sought and continues to be a major area
of research in the ﬁeld (see Hang and Vitorello, 1996 - Appendix D).

Because of its pH—dependent solubility behavior, Al toxicity occurs only at soil pH
values roughly below 5.0 or 5.5. Toxicity is most severe in cation-poor soils with low base
saturation. This is frequently the case for highly weathered, acidic soils with low cation
exchange capacity. Al toxicity can be totally eliminated by amending soils with lime, but
because of the poor mobility of CaCO3 within the soil proﬁle, this will be true only in those
top layers that are tilled. The result will be a relatively shallow root system, particularly
vulnerable to drought (Foy et al., 1978). In some cases, liming is not a feasible alternative
for cash-strapped farmers in third world countries or remote locations. This practice is also
not feasible in forests, such as those affected by acid rain in industrialized regions, and
perennial crops. Although soil pH is not corrected, application of CaSO4 has recently gained

wider use to ameliorate Al toxicity by increasing soil Ca2+ levels. CaSO4 is more mobile than

7

CaCO3 in the soil, reaching greater depths. and is inexpensive.

Other major factors affecting severity of Al toxicity to roots are the concentrations
of Ca2+ and other cations in the external solution (Kinraide and Parker, 1987), the ionic
strength of solutions, temperature (Foy et al., 1978; Camargo, 1983), presence of chelators
(Suhayda and Haug, 1986) and genotype of the plant (Foy et al., 1978; Wheeler et al..
1993)

The chemical species of Al which are toxic are presumably Al3+ and the mononuclear
hydroxides, Al(OH)2* and Al(OH)2+ (Kinraide, 1991). High toxicity has also been attributed
to the “Al,3“ tridecameric polycation (Parker and Bertsch, 1992), but this and its relevance
to Al toxicity are questionable. These conclusions are based on correlations between root
growth inhibition and changes in relative concentrations of Al species in solution, obtained
predominantly by changing the pH of the solution. This co-linearity between pH and Al

species is problematic, given the important effects pH has on the plant root (Kinraide, 1991).

Aluminum toxicity in animals

Although Al has generally been regarded as being of low toxicity to humans, in the
last two decades, considerable evidence has implicated Al in various diseases, particularly
neurological disorders, such as Alzheimer’s disease, amyotrophic lateral sclerosis and
Parkinson-dementia (Ganrot, 1986; Kruck and McLachlan, 1988). That Al is neurotoxic has
been shown directly by intracerebral injection of Al, causing progressive encephalopathy
(Klatzo et al., 1965; Selkoe et al., 1979; Troncoso et al., 1985). It has also been clearly shown

to be a causative agent of dialysis encephalopathy (Kerr et al., 1992). However, its role in

8

Alzheimer’s and other neurological disorders of natural cause are not clear.

It is controversial whether brains afflicted by Alzheimer’s disease have an increased
bulk Al content (Crapper et al., 1973; McDermott et al., 1979), but Al is claimed to be
present in affected brain regions at concentrations sufﬁcient to be toxic (McLachlan et al.,
1992). It is associated with neuroﬁbrillary tangles (Perl and Brody, 1980) and amyloid
plaques (Candy et al., 1986) and Al was found to promote in vitro aggregation of amyloid
p-protein, which accumulates in senile plaques (Mantyh et al., 1993; Kawahara et al., 1994).

Because of the ubiquitous nature of Al, humans unintentionally ingest rather large
amounts of Al (30-50 mg/day) in their diet (Ganrot, 1986). For the most part (99%), this Al
is not absorbed by the intestines, and Al neurotoxicity is not expressed in normal, healthy
individuals. Why normal exclusion of A] may fail is not clear. The accumulation of Al in the
brain of patients afﬂicted with Alzheimer’s disease may be a consequence, rather than the
cause, of nerve cell damage and other primary disfunctions, resulting in access to binding
sites. A current hypothesis is that Al, although probably not a causative agent, is involved in
the pathogenesis of Alzheimer’s disease (McLachlan et al., 1992). Because of the neutral
internal pH values of animals, such as in the blood plasma, which would lead to
precipitation, Al is transported in the chelated form. Citrate and transferrin are the major
chelators of Al in the blood (Martin, 1988).

In other animals, the toxic effects of A1 are most pronounced in ﬁsh, particularly the
epithelium of gills (Driscoll et al., 1980; Youson and Neville, 1986). Until recently, Al
concentrations in surface waters were generally below toxic levels (MacDonald and Martin,
1988). However, acidiﬁcation due to acid rain has mobilized Al to toxic levels in many

bodies of water. In addition to low pH, low Ca” concentrations (and probably other cations

9

as well) are also associated with Al toxicity to ﬁsh (Yousson and Neville, 1986)

Low pH effects on roots

As mentioned above, Al toxicity occurs in plants only at low pH, whether in soils or
nutrient solutions. Low pH (<5.0) in itself is well known to be directly detrimental to root
growth (Zsoldos and Erdei, 1981; Marschner, 1991; Yan et al., 1992), particularly in
solutions of low ionic strength and low cation concentrations (Lund, 1970). As in A1 toxicity,
increasing the concentration of Ca2+ and other cations in the external solution reduces or even
abolishes the detrimental effects of acidity (F ageria et al., 1989). An evaluation of these low-
pH effects is necessary for greater understanding and correct interpretation in studies of Al
toxicity. Unfortunately, physiological effects related to low pH are usually not taken into
consideration in Al toxicity studies and in some cases may have led to incorrect interpretation
of results.

At the cellular level, low pH has detrimental and distinct effects on the plasma
membrane, particularly enhanced permeability (Zsoldos and Erdei, 1981; Yan et al., 1992).
In addition, microviscosity of the plasma membrane is altered with decreasing pH (Berczi
et al., 1981; Suhayda et al., 1988). These membrane effects are in part responsible for altered
patterns of nutrient accumulation at low pH (Marschner, 1991). Although K" permeates the
plasma membrane more readily at low pH, enhancement of efflux is greater, resulting in
reduced net uptake (Zsoldos and Erdei, 1981). Low pH-induced membrane permeability can
be alleviated by Ca2+ and other cations (Yan et al., 1992; Kinraide, 1993). Aluminum

reportedly decreased H’-induced membrane permeability to K+ (Sasaki et al., 1994). It has

10

been suggested that Al-induced growth stimulation results from amelioration of proton
toxicity (Llugany et al., 1995; Kinraide, 1993). At low pH, low concentrations of Al can be
beneﬁcial to root growth, particularly in Al-tolerant plants (Taylor, 1988).

Recently, it was found that regions of the wheat root apex with high Al
concentrations were correlated with regions in which a depletion in K* had occurred
(Delhaize et al., 1993a). This was interpreted as being Al-induced loss of K’ from cells.
However, since they did not present data clearly showing that this depletion of K” may have
resulted regardless of the presence of Al, this depletion may well have been the result of the
low pH of the medium, as has been shown to occur (Zsoldos and Erdei, 1981; Sasaki et al.,
1994). If this is the case, would this mean that Al accumulates preferentially in cells that have

low-pH-altered membranes?

Al uptake in plants

Sites of Al accumulation

In intact roots, Al accumulates in the epidermis and in the cortex, although less in the
inner layers of the cortex (Wagatsuma et al., 1987; Rengel, 1992; Jentschke et al., 1991,
Ownby 1993; Delhaize et al, 1993a). The endoderrnis acts as a barrier (Rengel, 1992a) and
transport to the shoot and leaves is generally small, although there are Al-accumulating
species. Examples are the tea plant, pine trees (F oy et al., 1978) and some species native to
the “cerrado” region of central Brazil (Haridasan, 1982; Haridasan and Araujo, 1988).
Unfortunately, there is little in the literature as to mechanisms, cellular localization and

chemical forms of the Al which accumulates in these plants. In tea, Al accumulates in older

11

leaves, particularly in the epidermis (Foy et al., 1978). With regard to the chemical form of
Al in these leaves, it was concluded that most Al was chelated to the catechin group of
polyphenols, and to a lesser extent to phenolic and organic acids and as Al-F complexes
(Nagata et al., 1992). Whereas in some cases there is evidence that Al is chelated within the
leaf cells (Haridasan et al., 1986), another report concluded that Al is deposited on the cell
wall of the epidermis and palisade parenchyma (Memon et al., 1981). The Al species present
after absorption from the soil is not yet known, but there is evidence that it may be

transported as Al-F species (Nagata et al., 1993).

Al uptake at the cellular level: apoplastic or symplastic?

At the cellular level, a major unresolved issue is whether Al actually enters the cell
or whether it accumulates solely in the cell wall or in the mucigel surrounding the root apex
(Delhaize and Ryan, 1995; Kochian, 1995). Attempts to make this distinction have received
considerable attention because of its implication to models of Al toxicity. While Al has been
found in the nucleus, presumably bound to DNA (Matsumoto et al., 1976), it has been
reported by others to be localized solely or predominantly in the cell wall (Clarkson, 1967;
Ownby, 1993; Marienfel and Stelzer, 1993). This ambiguity results partially from differing
experimental conditions, some of which may have been inadvertently ﬂawed (Kochian,
1995). Earlier reports have employed long periods of exposure to Al, which raises questions
as to the integrity of these stressed cells, and/or high concentrations of Al, which in turn
raises questions as to precipitation of Al. Precipitation of Al phosphates in the cell wall can
also occur.

Maj or limitations to progress in Al uptake studies have been largely methodological.

12

Reasons for this are the complex chemistry of Al (Martin, 1988), interactions with the cell
wall, the lack of a suitable Al radioisotope (Rengel, 1992; Schaedle et al., 1989; Taylor,
1988), and the limited sensitivity and resolution of microanalytical techniques (Delhaize et
al., 1993a; Lazof and Lauchli, 1991; Lazof et a1, 1994). Because of the uncertain
effectiveness of chelator washes in removing Al from the cell wall (Taylor, 1988),
quantiﬁcation of Al by graphite atomic absorption spectrometry (GFAAS), a sensitive and
standard analytical technique, does not provide certainty as to cellular sites of accumulation.
X-ray microanalysis, the major microanalytical technique, is limited by sensitivity and spatial
resolution, the latter being relatively poor because of the interactions of the radiation with
the specimen layer and distortions in the cell samples during preparation (Goldstein et al.,
1981; Lazof and Lauchli, 1991; Lazof et al., 1994).

Nevertheless, attempts have been made to surmount or circumvent these limitations
and to address the problem of Al uptake and its cellular site of accumulation. The kinetics
of short-term A1 uptake was found to be biphasic, and was ﬁrst used in an attempt to
distinguish cell wall A1 from uptake across the plasma membrane (Petterson and Strid, 1989;
Zhang and Taylor, 1989; Zhang and Taylor, 1990). The rapid, initial, non-linear phase of
uptake was interpreted as being accumulation of exchangeable A1 in the cell wall, and could
be desorbed with citrate (Zhang and Taylor, 1990). The slower, second phase of uptake was
linear and interpreted to be uptake across the plasma membrane (Pettersson and Strid, 1989)
or metabolism-dependent accumulation in the cell wall (Zhang and Taylor, 1990). A more
direct ﬁactionation procedure was also used (Tice et al., 1992). Following external washes
aimed at removing apoplastic Al, soluble symplastic Al was washed out after cell membranes

had been ruptured by freeze—thawing. The residual fraction was interpreted as being non-

l3

soluble symplasmic Al and contained 50 to 70% of the total cell-associated Al.

To improve upon microanalytical techniques for the detection of Al, secondary-ion
mass spectrometry was used to detect Al in soybean root tips exposed to A1 for a 30 min
period (Lazof et al., 1994). Although this method has a spatial resolution of about 2 pm,
approximately one order of magnitude better than X-ray microanalysis, the association
between A1 signal images and actual cell images was difﬁcult (Figs. 4 and 6 of Lazof et al.,
1994). Nevertheless, the authors concluded that Al was internalized by cells based on the fact
that the Al signal was concentrated in circular areas similar in size (15 um in diameter) to
individual cells. They argued that if present in the cell wall, the A1 signal image should show
“rings” corresponding to the cell wall. In summary, there now seems to be evidence, although
inconclusive, favoring the notion that Al is indeed rapidly internalized by cells (Kochian,

1995).

Factors impacting Al uptake

Whereas efforts have been made to determine the cellular sites of Al accumulation,
much less attention has been given to factors impacting Al uptake. With regard to
dependence of A1 uptake on metabolism and energy, results are again ambiguous. Depending
on the criteria used, accumulation of Al has the characteristics of non-metabolic uptake
(Clarkson, 1967; Wagatsuma, 1983; Foy et al., 1978). Metabolic inhibitors, such as DNP or
CCCP, increase Al uptake (Wagatsuma, 1983; Rincon and Gonzales, 1992; Zhang and
Taylor, 1991 ), rather than decrease it. Because this increase in uptake was more dramatic
in Al-tolerant cultivars, an energy-dependent exclusion mechanism has been proposed

(Zhang and Taylor, 1991; Rincon and Gonzales, 1992). In an apparent paradox, Al

14

accumulates in the root tip, which is physiologically very active (Bennet et al., 1985; Rincon
and Gonzales, 1992) and a metabolism dependent-accumulation of Al in the cell wall has
been proposed (Petterson and Strid, 1989; Zhang and Taylor, 1991). Low temperature has
been shown to substantially decrease Al uptake (Pettersson and Strid, 1989; Zhang and
Taylor, 1990). Calcium is known to ameliorate Al toxicity (Kinraide and Parker, 1987).
Information on the effects of Ca2+ on Al uptake is scarce but in at least one report, has been
shown to decrease Al uptake (Alva et al., 1986).

A1 uptake has also been studied in non-plant systems lacking a cell wall, such as
liposomes (Shi and Haug, 1988; Akeson and Munns, 1989) or mammalian cells (Shi and
Haug, 1990). In neuroblastoma cells, Al uptake was suggested to depend on some sort of
carrier, although it depended in part on the physical state of the membrane (Shi and Haug,

1990).

Mechanisms of Al toxicity

The search for the primary target of Al injury is probably the most active area in Al
toxicity studies, and too many potential targets for Al toxicity have been suggested to
adequately cover in this review. Several of them involve phosphate groups and their
utilization by cells. A1 is capable of binding tightly to DNA, presumably to its phosphate
backbone, or alternatively to associated histones (Dyrssen et al., 1987; Matsurnoto, 1991) and
this led to one of the earlier hypotheses of toxicity, that cell division was impaired because
of interactions of A1 with nuclear DNA. A1 also has a high binding afﬁnity to free nucleotide

triphosphates and a model for A1 toxicity based on its binding to ATP in the cytoplasm has

15

been proposed (Pettersson and Bergman, 1989).

A1 has also been shown to alter the structure of the plasma membrane (Vierstra and
Haug, 1978; Zhao et al., 1987) and has pronounced effects on ion ﬂuxes across the
membrane, particularly Ca2+ uptake (Huang et al., 1992). Al toxicity has frequently been
linked to Ca2+ (Rengel, 1992) either because of Al-induced perturbations in cellular
Caz’ metabolism or because of Ca2+ amelioration of A1 toxicity (Kinraide and Parker, 1987).
Several investigations found profound Al-induced alterations in the structure of calmodulin,
the chief mediator of intracellular Ca2+ signalling (Siegel and Haug, 1983; Siegel et al.,
1983), and initiated considerable research on the role of calmodulin in Al toxicity (Haug and
Vitorello, 1996 - Appendix C). More recently, phosphoinositide-mediated signal
transduction, a pathway which also involves Caz” as an intracellular messenger, has been
investigated as a primary site of Al toxicity (Fig. 1) in mammalian (Hang et al., 1994;
Appendix B) and plant cells (Jones and Kochian, 1995). In both cases A1 treatment
presumably inhibited the activity of phospholipase C or possibly the action of the trimeric

G-protein (Shi and Haug, 1992; Shi et al., 1993).

A1 tolerance: mechanisms, genetics and Al-induced genes and proteins

Plant genotypes present wide variability with respect to their sensitivity to A1, and
this has been known for some time. Traditional screening and breeding programs have
resulted in some success. The mechanisms for tolerance, like those for toxicity, are not
known (Delhaize and Ryan, 1995; Kochian, 1995). However, one mechanism, that of

exclusion of Al from the root, has recently been revisited and has gained favor in this ﬁeld

16

of research. In some cases, Al has been shown to accumulate more in the roots of A1-
sensitive cultivars of wheat than in Al-tolerant ones (Delhaize et al., 1993b; Rincon and
Gonzales, 1992). Recently, three independent groups have demonstrated that malate-
secretion is enhanced in Al-tolerant cultivars compared to Al-sensitive ones (Basu et al.,
1994; Delhaize et al., 1993b; Pellet et al., 1995). Tolerance would be the result of chelation
of A1 to malate.

From the few studies that have been performed, Al tolerance is a trait governed by
only one or a few dominant genes (Aniol and Gustafson, 1984; Aniol, 1991; Berzonsky,
1992). In tobacco cell cultures selected for Al tolerance, a single dominant gene mutation
was attributed to tolerance (Conner and Meredith, 1985). On the other hand, Al induces the
synthesis of several proteins (Basu et al., 1994) and also the expression of several genes
(Snowden et al., 1995). However, many of these proteins are induced in both Al-sensitive
and Al-tolerant genotypes (Delhaize and Ryan, 1995). Some of these proteins have been
identiﬁed and include phenylalanine ammonia-lyase, metallothionein-like proteins,
proteinase inhibitors and asparagine synthetases (Snowden et al., 1995). Thus, to date, the

proteins synthesized in response to A1 appear to be general stress-response proteins.

Processes involved in cell growth/secretion and Al toxicity

In plants, Al affects meristematic and elongating cells of the root apex (Ryan et al.,
1993), root cap cells (Bennet and Breen, 1991) root hairs (Jones et al., 1995) and pollen tubes
(Konishi and Miyamoto, 1983; Searcy and Mulcahy, 1990). In mammalian cells,

neurological cells appear to be the most sensitive (Ganrot, 1986) and in ﬁsh A1 affects the

17

epithelium of the gill (Youson and Neville, 1986). These cells are engaged in growth,
secretion, or synapsis, all of which involve similar processes at the cell surface, particularly
membrane trafﬁc and turnover. A1 may inhibit growth or disrupt secretory activity in these
cells by direct toxicity to these regions. With regard to growth in plant cells, A1 inhibits tip
growth in root hairs and pollen tubes, it reduces the incorporation of sucrose into cell walls
(Huck, 1972) and induces callose synthesis (Schaeffer and Walton, 1990). Al was found to
cause deformities in root hair tips, but only in those that were growing (Jones et al., 1995).
For these reasons and because the relationship between cell growth and Al toxicity is a major
theme of my dissertation, a brief review of processes associated with cell growth and their
potential relevance to A1 toxicity is warranted.

For a plant cell to expand and grow, new cell wall and plasma membrane must be
incorporated to the cell surface. This involves many cellular components, ranging from the
endomembrane system, the cytoskeleton and the plasma membrane. Cellulose microﬁbrils
are synthesized on the outer side of the plasma membrane by a membrane-associated
cellulose synthase complex (Delrner, 1987). Other cell wall components are synthesized in
the Golgi dictyosomes and transported and secreted to the cell surface (Staehelin and Moore,
1995). The plasma membrane is under continual turnover in these regions. The number of
exocytotic fusions in tip‘growing cells has been estimated to be SOOO/min (Heath, 1990). In
plants, cell growth has been better studied in tip-growing cells (Steer and Steer, 1989).
Regions of cellular growth may be involved in A1 toxicity for two reasons. First, access of
Al to sites of toxicity may be facilitated in these regions, because of the dynamic nature of
these membranes. Second, several components or processes involved with cell growth are

potential targets for Al.

18

Membrane fusion, turnover and trafﬁc are essential to growth and secretion. The
mechanisms of membrane fusion are not well understood, with most knowledge deriving
from information on liposomes rather than on actual fusion events in the cell. Nevertheless,
from these liposome models, it is thought that non-bilayer lipid conﬁgurations are required
as intermediate steps for fusion to occur. Lipid fUSIOl'I is promoted by lipids that do not form
bilayers and by cations (Ca2+, H‘) and heat, which disrupt bilayers (Zirnmerberg, et al., 1993).
Phosphatidylserine (PS) and phosphatidylethanolamine (PE) are considered important for
fusion. In PE-deﬁcient Escherichia coli cells, high concentrations of divalent cations were
necessary for growth, and these results were interpreted to involve regulation of lipid
polymorphism (Rietveld et al., 1994). Proteins are widely accepted as mediators of
membrane fusion, however, with few exceptions (viral proteins), little information is
available for such proteins (Zimmerberg et al., 1993).

The plasma membrane probably plays an important role in Al toxicity, either as a
permeability barrier or as a site of injury. At micromolar concentrations, Al reportedly
induced rigidiﬁcation and membrane ﬁrsion in unilamellar vesicles containing PS (Deleers
et al., 1986). Binding of A1 to phosphatidylcholine was found to occur with an affinity 560
times higher than with Ca2+ (Akeson et al., 1989). A role of the plasma membrane in A1
toxicity was suggested by the fact that the average surface charge of protoplasts from Al-
tolerant species was less negative than Al-sensitive species, and this property was used to
separate and collect Al-tolerant and Al-sensitive protoplasts (Wagatsuma et al., 1995).
Plasma membrane domains rich in negatively charged sites may constitute regions for
binding and uptake of A1. Membrane instabilities, enhanced by low-pH, low Caz’

concentrations and possibly even by Al, may give rise to special conditions for uptake of A1.

19

Phosphatidylserine, a negatively charged, fusogenic lipid, may be present in such regions.

Compared to the mechanisms of membrane fusion, other aspects of membrane
trafﬁcking are better understood, such as docking and the regulation of vesicular trafﬁc. Both
small and heterotrimeric GTP-binding proteins are involved in most phases, and GTP-
binding proteins can be found in all cellular membranes (V errna et al., 1994; Nuoffer and
Balch, 1994). There is experimental evidence that A] can inhibit the activity of GTP-binding
proteins (Shi et al., 1992; Haug et al., 1994 - Appendix C). Perhaps Al has greater access to
GTP-binding proteins of the plasma membrane in regions of cell growth.

Calcium is necessary for membrane fusion to occur and is involved in the regulation
of exocytosis and endocytosis (Steer, 1988). Modulation of fusion by Ca2+ can occur by
interaction with phospholipids or by means of annexins, which are Ca2’-dependent
phospholipid-binding proteins (Clark and Roux, 1995). In tip-growing cells such as pollen
tubes and root hairs, Ca2+ gradients are established between the tip and basal regions (Miller
. et al., 1992). These Ca2+ gradients are presumably generated and maintained by localized
Ca2+ inﬂux mediated through Ca2+ channels which accumulate in growing regions (Steer and
Steer, 1989). Levels of calmodulin are high in the cytoplasm proximal to sites of growth
(Hausser et al., 1984). Aluminum is thought to impact Ca2+ metabolism of cells and has been
shown to interact with calmodulin, in vitro (Siegel and Haug, 1983; Haug and Vitorello,
1996- Appendix D) and Ca2+ channels (Huang et al., 1996). It is not known whether Al
impacts annexins.

Microtubules and microﬁlarnents are involved in directing and transporting secretory
vesicles from the Golgi to the cell surface (Cole and Lippincott-Schwartz, 1995).

Microtubules of the cortical array are involved in orienting cellulose microﬁbril deposition

2O

(Delmer, 1987). Tubulin polymerization requires Mg-dependent hydrolysis of GTP. In vitro,
A1 promoted microtubule assembly (MacDonald etal., 1987). However, Al has not yet been
shown to interact with microtubules in vivo. The dynamics of actin are also dependent on
GTP hydrolysis and Al reportedly induced an increase in the rigidity of transvacuolar strands
of actin in cultured soybean cells (Grabski and Schindler, 1995). It has not yet been

demonstrated whether Al actually reaches these actin strands.

REFERENCES

Akeson, M.A., Munns, D.N., Burau, R.G. 1989. Adsorption of Al to phosphatidylcholine
vesicles. Biochim. Biophys. Acta 986:33-40.

Akeson, M.A., Munns, ON. 1989. Lipid bilayer permeation by neutral aluminum citrate and
by three alpha-hydroxy carboxylic acids. Biochim. Biophys. Acta 984:200-206.

Alva, A.K., Asher, C.J., Edwards, DC. 1986. The role of calcium in alleviating aluminium
toxicity. Aust. J. Agric. Res. 37:375-382.

Aniol, A., Gustafson, J.P. 1984. Chromosome location of genes controlling aluminum
tolerance in wheat, rye, and triticale. Plant Physiol. 104:701-705.

Aniol, A. 1991. Genetics of acid tolerant plant. In Wright, R.J. (eds). Plant-soil interactions
at low pH. Kluwer Academic, Dordrecht, The Netherlands, pp 1007-1017.

Baes, C.F., Mesmer, RE. 1976. The hydrolysis of cations. J .Wiley & Sons, New York. pp
112-113.

Basu, U., Godbold, D., Taylor, GJ. 1994. Aluminum resistance in T riticum aestivum
associated with enhanced exudation of malate. J. Plant Physiol. 144:747-753.

Bennet, R.J., Breen, C.M., Fey, M.V. 1985. The primary site of aluminium injury in the root
of Zea mays L. S.Aﬁ'. J. Plant Soil 228-17.

Bennet, R.J., Breen, CM. 1991. The aluminium signal: new dimensions to mechanisms of
aluminium tolerance. Plant Soil 134:153-166.

Bennet, R.J., Breen, C.M., Fey, M.V. 1987. The effects of aluminium on root cap function
and root development in Zea mays L. Environ. Exp. Bot. 27:91-104.

Berczi, A., Erdei, L., Zsoldos, F. 1981. Membrane transport and ion transport properties in
cereals under acidic and alkaline stress. Physiol. Plant. 53:475-478.

Berzonsky, WA. 1992. The genomic inheritance of aluminum tolerance in Atlas 66 wheat.
Genome 35:689-693.

21

22

Braen, S.N., Weinstein, L.H. 1985. Uptake of ﬂuoride and aluminum by plants grown in
contaminated soils. Water Air Soil Pollut. 24:215-223.

Camargo, C.E. de 0. 1983. Efeito da temperatura da solucao nutritiva na tolerancia ao
aluminio de cultivares de trigo. Bragantia 42: 51-63.

Candy, J .M., Oakley, A.E., Klinowski, J. et al. 1986. Aluminosilicates and senile plaque
formation in Alzheimer’s disease. Lancet 1:354-357.

Chabre, M. 1990. Aluminoﬂuoride and berylloﬂuoride complexes: new phosphate analogs
in enzymology. Trends Biochem. Sci. 1526-10.

Clark, G.B., Roux, SJ. 1995. Annexins of plant cells. Plant Physiol. 109:1133-1139.

Clarkson, D.T. 1967. Interactions between aluminium and phosphorus on root surfaces and
cell wall material. Plant Soil 27: 347-356.

Cole, N.B., Lippincott-Schwartz, J. 1995. Organization of organelles and membrane trafﬁc
by microtubules. Current Opinion in Cell Biology 7: 55-64.

Conner, A.J., Meredith, GP. 1985. Strategies for the selection and characterization of
aluminum-resistant variants from cell cultures of Nicotiana plumbaginifolia. Planta 166:
466-473.

Crapper, D.R., Krishnan, S.S., Dalton, A.J. 1973. Brain aluminum distribution in
Alzheimer’s disease and experimental neuroﬁbrillary degeneration. Science. 180:511-
513.

Crapper Mclachlan, DR. 1989. Aluminum neurotoxicity: criteria for assigning a role in
Alzheimer’s disease. In Lewis, T.E. (ed). Environmental Chemistry and Toxicology of
Aluminum. Lewis Pub1., Boca Raton, FL. pp 299-315.

Da Silva, J.J.R.F., Williams, R.J.P. 1991. The biological chemistry of the elements. The
inorganic chemistry of life. Oxford University Press, Oxford.

Deleers, M., Servais, J .P., Wiilfert, E. 1986. Biochim. Biophys. Acta 855: 271-276.

Delhaize, E., Craig, S., Beaton, C.D., Bennet, R.J., Jagadish, V.C., Randall, P.J. 1993a.
Aluminum tolerance in wheat (T riticum aestivum L.) I. Uptake and distribution of
almninum in root apices. Plant Physiol. 103:685-693.

Delhaize, B, Ryan, P.R., Randall, P.J. 1993b. Aluminum tolerance in wheat (Triticum
aestivum L.). II. Aluminum-stimulated excretion of malic acid from root apices. Plant
Physiol. 103:695-702.

23

Delhaize, E., Ryan, PR. 1995. Aluminum toxicity and tolerance in plants. Plant Physiol.
107: 315-321.

Delrner, DP. 1987. Cellulose biosynthesis. Ann. Rev. Plant Physiol. 38: 259-290.

Driscoll, C.T., Baker, J .P., Bisigni, J .J ., Schoﬁeld, CL. 1980. Effect of aluminium speciation
on ﬁsh in dilute acidiﬁed waters. Nature 284: 161-164.

Dyrssen, D., Haraldsson, C., Nyberg, E., Wedborg, M. 1987. Complexation of aluminum
with DNA. J. Inorg. Biochem. 29:67-75.

Foy, C.D., Chaney, R.L., White, MC. 1978. The physiology of metal toxicity in plants.
Annu. Rev. Plant Physiol. 29: 511-566.

Fageria, N.K., Baligar, V.C., Wright, R.J. 1989. Growth and nutrient concentrations of
alfalfa and common bean as inﬂuenced by soil acidity. Plant Soil 119: 331-333.

Ganrot, PD. 1986. Metabolism and possible health effects of aluminum. Environ. Health
Perspect. 65: 363-441.

Godbold, D.L., Fritz, E., Hiittermann, A. 1988. Aluminum toxicity and forest decline. Proc.
Natl. Acad. Sci. USA. 85:3888-3892.

Goldstein, J .I., Newbury, D.E., Echlin, P., Joy, D.C., Fiori, C., Lifshin, E. 1981. Practical
techniques of x-ray analysis. In Goldstein, J .I., Newbury, Echlin, P., Joy, D.C., Fiori, C.,
Lifshin, E. (eds). Scanning electron microscopy and x-ray microanalysis. pp 393-446.
Plenum Press, New York.

Grabski, S., Schindler, M. 1995. Aluminum induces rigor within the actin network of
soybean cells. Plant Physiol. 108:897-901.

Haridasan, M. 1982. Aluminium accumulation by some cerrado native species of central
Brazil. Plant Soil 65:265-273.

Haridasan, M., Paviani, T.I., Schiavini, I. 1986. Localization of aluminimn in the leaves of
some aluminium-accumulating species. Plant Soil 94:435-437.

Haridasan, M., De Araujo, GM. 1988. Aluminimn-accumulating species in two forest
communities in the Cerrado region of central Brazil. Forest Ecol. Manag. 24:15-26.

Haug, A. 1984. Molecular aspects of aluminum toxicity. Crit. Rev. Plant Sci. 1: 345-373.

Haug, A., Shi, B., Vitorello, V. 1994. Aluminum interaction with phosphoinositide-
associated signal transduction. Arch Toxicol 68: 1-7

24

Haug, A., Vitorello, V. 1996. Aluminum coordination to calmodulin: Thermodynamic and
kinetic aspects. Coordination Chemistry Reviews. (In press).

Hausser, 1., Herth, W., Reiss, H.D. 1984. Calmodulin in tip growing plant cells, visualized
by ﬂuorescing calmodulin-binding phenothiazines. Planta 162:33-39.

Heath, LB. 1990. Tip growth in plant and fungal cells. Academic Press, Orlando, Fla.

Hem, J .D. 1986. Geochemistry and aqueous chemistry of aluminum. Kidney International,
29(18):s3-s7.

Huang, J .W., Pellet, D.M., Papernik, L.A., Kochian, L.V. 1996. Aluminum interactions with
voltage-dependent calcium transport in plasma membrane vesicles isolated from roots
of aluminum-sensitive and -resistant wheat cultivars. Plant Physiol. 110: 561-569.

Huang, J. W., Shaff, J.E., Grunes, D.L., Kochian, L.V. 1992. Aluminum effects on calcium
ﬂuxes at the root apex of aluminum-tolerant and aluminum-sensitive wheat cultivars.
Plant Physiol. 98:230-237.

Huck, MG. 1972. Impairment of sucrose utilization for cell wall formation in the roots of
aluminum-damaged cotton seedlings. Plant Cell Physiol. 13:7-14.

Jentschke, G., Schlegel, H., Godbold, D.L. 1991. The effect of aluminium on uptake and
distribution of magnesium and calcium in roots of mycorrhizal Norway spruce seedlings.
Physiol. Plant. 82:266-270.

Jones, D.L., Kochian, L.V. 1995. Aluminum inhibition of the inositol 1,4,5-triphosphate
signal transduction pathway in wheat roots: a role in aluminum toxicity? Plant Cell
7:1913-1922;

Jones, D.L., Shaff, J .S., and Kochian, L.V. 1995. Role of calcium and other ions in directing
root hair tip growth in Limnobium stoloniferum. 1. Inhibition of tip growth by aluminium.
Planta 197:672-680.

Kawahara, M., Muramoto, K., Kobayashi, K, Mori, H., Kuroda, Y. 1994. Aluminum
promotes the aggregation of Alzheimer’s amyloid B-protein in vitro. Biochem. Biophys.
Res. Comm. 198:531-535.

Kerr, D.N.S. , Ward, M.K., Ellis, H.A., Simpson, W., Parkinson, LS. 1992. Aluminium
intoxication in renal disease. In Chadwick, D.J., Whelan, J. (eds). Almninium in biology
and medicine. pp 123-141. John Wiley & Sons, New York.

Kinraide, T.B., Parker, DR. 1987. Cation amelioration of aluminum toxicity in wheat. Plant
Physiol. 83: 546-551.

25

Kinraide, TE. 1991. Identity of the rhizotoxic aluminium species. Plant Soil 134: 167-178.

Kinraide, TE. 1993. Aluminum enhancement of plant growth in acid rooting media. A case
of reciprocal alleviation of toxicity by two toxic cations. Phys. Plant. 88:619-625.

Klatzo, I., Wisniewki, H., Streicher, E. 1965. Experimental production of neuroﬁbrillary
degeneration. 1. Light microscopic observations. J. Neuropathol. Exp. Neurol. 24:187-
199.

Kochian, L.V. 1995. Cellular mechanisms of aluminum toxicity and resistance in plants.
Annu. Rev. Plant Physiol. Plant Mol. Biol. 46: 237-60.

Konishi, S., Miyamoto, S. 1983. Alleviation of aluminum stress and stimulation of tea pollen
tube growth by ﬂuorine. Plant Cell Physiol. 24:857-862.

Kruck, T.P.A., McLachlan, DR. 1988. Mechanisms of aluminum neurotoxicity - relevance
to human disease. In: Sigel, H., Sigel, A. (eds). Metal ions in biological systems:
Aluminum and its role in biology, vol. 24, pp 285-314. New York, Marcel Dekker.

Lazof, D., Lauchli, A. 1991. Complementary analysis of freeze-dried and frozen-hydrated
plant tissue by electron-probe X-ray microanalysis: Spectral resolution and analysis of
calcium. Planta 184:327-333.

Lazof, D.B., Goldsmith, J .G., Rufty, T.W., Linton, R. 1994. Rapid uptake of aluminum into
cells of intact soybean root tips. A microanalytical study using secondary ion mass
spectrometry. Plant Physiol. 106: 1107-1114.

Llugany, M., Poschenrieder, C., Barcelo, J. 1995. Monitoring of aluminium-induced
inhibition of root elongation in four maize cultivars differing in tolerance to aluminum
and proton toxicity. Physiol. Plant. 93:265-271.

Lund, Z.F. 1970. The effect of calcium and its relation to several cations in soybean root
growth. Soil Sci. Soc. Am. Proc. 34:456-459.

MacDonald, T.L., Humphreys, W.G., Martin, RB. 1987. Promotion of tubulin assembly by
aluminum ions in vitro. Science 236: 1 83-1 86

Macdonald, T.L., Martin, RB. 1988. Aluminum ion in biological systems. Trends Biochem.
Sci. 13:15-19.

Mantyh, P.W., Ghilardi, J .R, Rogers, 8., DeMaster, E., Allen, C.J., Stimson, E.R., Maggio,
J .E. 1993. Aluminum, iron, and zinc ions promote aggregation of physiological
concentrations of B-amyloid peptide. J. Neurochem. 61:1171-1174.

26

Marienfeld, S., Stelzer, R. 1993. X-ray microanalyses in roots of Al-treated Avena sativa
plants. J. Plant Physiol. 141 :569-573.

Marschner, H. 1991. Mechanisms of adaptation of plants to acid soils. In Wright, R.J.,
Baligar, V.C., Murrmann, R.P. (eds). Plant-soil interactions at low pH. pp 683-702.
Kluwer Academic Publishers. Netherlands.

Martin, RB. 1988. Bioinorganic chemistry of aluminum. In: Sigel, H., Sigel, A. (eds). Metal
ions in biological systems: Aluminum and its role in biology, vol. 24, pp 2-57. New
York, Marcel Dekker.

Martin, RB. 1992. Aluminium speciation in biology. In Chadwick, D.J., Whelan, J. (eds).
Aluminium in biology and medicine. pp 5-25. John Wiley & Sons, New York.

Matsumoto, H., Hirasawa, E., Torikai, H., Takahashi, E. 1976. Localization of absorbed
aluminium in pea root and its binding to nucleic acids. Plant Cell Physiol. 17: 127-137.

Matsumoto, H. 1991. Biochemical mechanism of the toxicity of aluminium and the
_ sequestration of aluminium in plant cells. In Wright, R.J., Baligar, V.C. Murrrnann, R.P.
(eds). Plant-soil interactions at low pH. pp 825-838. Dordrecht, Kluwer Acad. Publ.

McDermott, J .R., Smith, A.I., Iqbal, K., Wisniewski, HM. 1979. Brain aluminum in aging
and Alzheimer disease. Neurology 29:809-814.

McLachlan, D.R., Fraser, P.E., Dalton, A]. 1992. Aluminium and the pathogenesis of
Alzheimer’s disease: a summary of evidence. In Chadwick, D.J., Whelan, J. (eds).
Aluminium in biology and medicine. pp 87-108. John Wiley & Sons, New York.

Memon, A.R., Chino, M, Yatazawa, M. 1981. Microdistribution of aluminum and
manganese in the tea leaf tissues as revealed by X-ray microanalyzer. Commun. Soil Sci.
Plant Anal. 12:441-452.

Miller, D.D., Callaham, D.A., Gross, D.J., Hepler, PK. 1992. Free Ca2+ gradient in growing
pollen tubes of Lilium. J. Cell Sci. 101 :7-12.

Nagata, T., Hayatsu, M, Kosuge, N. 1992. Identiﬁcation of aluminium forms in tea leaves
by 2"Al NMR. Phytochemistry 31:1215-1218.

Nagata, T., Hayatsu, M, Kosuge, N. 1993. Aluminium kinetics in the tea plant using 27Al and
"F NMR. Phytochemistry 32:771-775.

Nuoffer, C., Balch, WE. 1994. GTPases: multifunctional molecular switches regulating
vesicular trafﬁc. Ann. Rev. Biochem. 63: 949-990.

27

Orvig, C. 1993. The aqueous coordination chemistry of aluminum. In Coordination
chemistry of aluminum, (G.H. Robinson, ed), pp 85-121. VCH Publishers, New York,
NY. ISBN 1-56081-059-9.

Osmond, C.B., Bjorkman, 0., Anderson, DJ. 1980. Physiological processes in plant ecology,
Springer-Verlag, p. 61.

Ownby, J .D. 1993. Mechanisms of reaction of hematoxylin with aluminium-treated wheat
roots. Phsysiol. Plant. 87:371-380.

Ownby, J.D., Papham, HR. 1989. Citrate reverses the inhibition of wheat root growth
caused by aluminum. J. Plant Physiol. 135:588-591.

Parker, D.R., Bertsch, PM. 1992. Formation of the "All3" tridecameric polycation under
diverse synthesis conditions. Environ. Sci. Technol. 26:914-921.

Pellet, D.M., Grunes, D.L., Kochian, L.V. 1995. Organic acid exudation as an aluminum-
tolerance mechanism in maize (Zea mays L.). Planta 196:788-795.

Perl, D.P., Brody, AR. 1980. Alzheimer’s disease: x-ray spectrometric evidence of
aluminum accumulation in neuroﬁbrillary tangle-bearing neurons. Science 208:297-299.

Petterson, A., Bergman, B. 1989. Effects of aluminium on ATP pools and utilization in the
cyanobacterium Anabaena cylindrica: A model for the in viva toxicity. Physiol. Plant.
76:255-261.

Pettersson, S., Strid, H. 1989. Initial uptake of aluminum in relation to temperature and
phosphorus status of wheat (Triticum aestivum L.) roots. J. Plant Physiol. 134: 672-677.

Rengel, Z. 1992. Role of calcium in aluminum toxicity. New Phytol. 121:499-513.

Rietveld, A.G., Chupin, V.V., Koorengevel, M.C., Wienk, K.L.J., Dowhan, W., Kruijff, B.
1994. Regulation of lipid polymorphism is essential for the viability of
phosphatidylethanolamine-deﬁcient Escherichia cali cells. J. Biol. Chem. 269: 28670-
28675.

Rincon, M., Gonzales, RA. 1992. Aluminum partitioning in intact roots of aluminum-
tolerant and aluminum-sensitive wheat ( T riticum aestivum L.) cultivars. Plant Physiol.
99:1021-1028.

Ryan, P.R., DiTomasso, J .M., Kochian, L.V. 1993. Aluminium toxicity in roots: an
investigation of spatial sensitivity and the role of the root cap. J. Exp Bot 44:437-446.

28

Ryan, P.R., Shaif, J.E., Kochian, L.V. 1992. Aluminum toxicity in roots. Correlation among
ionic currents, ion ﬂuxes, and root elongation in aluminum-sensitive and aluminum-
tolerant wheat cultivars. Plant Physiol. 99:1 193-1200.

Sasaki, M., Kasai, M., Yamamoto, Y., Matsumoto, H. 1994. Comparison of the early

response to aluminum stress between tolerant and sensitive wheat cultivars: root growth,
aluminum content and efflux of K*. J. Plant Nutr. 17: 1275-1288.

Searcy, K.B., Mulcahy, D.L. 1990. Comparison of the response to aluminum toxicity in
gametophyte and sporophyte of four tomato (Lycapersican esculentum Mill.) cultivars.
Theor. Appl. Genet. 80:289-295.

Schaeffer, H.J., Walton, J .D. 1990. Aluminum ions induce oat protoplasts to produce an
extracellular (1-3)B-D-glucan. Plant Physiol. 94: 13-19.

Selkoe, D.J., Liem, R.K.H., Yen, S.H., Shelanski, ML. 1979. Biological and immunological
characterization of neuroﬁbrillary degeneration by aluminum. Brain Res. 163:235-252.

Schaedle, M., Thornton, F.C., Raynal, D.J., Tepper, HE. 1989. Response of tree seedlings
to aluminum. Tree Phsyiol. 52337-356.

Shi, B., Haug, A. 1988. Uptake of aluminum by lipid vesicles. Tox. Environ. Chem. 17: 337-
349.

Shi, B., Haug, A. 1990. Aluminum uptake by neuroblastoma cells. J. Neurochem. 55: 551-
558.

Shi, B., Haug, A. 1992. Aluminum interferes with signal transduction in neuroblastoma cells.
Pharmacol. Toxicol. 71 :308-313.

Shi, B., Chou, K., Haug, A. 1993. Almninium impacts elements of the phosphoinositide
signalling pathway in neuroblastoma cells. Mol. Cell. Biochem. 121:109-118.

Siegel, N., Coughlin, R.T., Haug, A. 1983. A thermodynamic and electron paramagnetic
resonance study of structural changes in calmodulin induced by aluminum binding.
Biochem. Biophys. Res. Commun. 1 15:512-517.

Siegel, N., Haug, A. 1983. Aluminum interaction with calmodulin: evidence for altered
structure and function from optical and enzymatic studies. Biochim. Biophys. Acta
744:36-45.

Snowden, K.C., Richards, K.D., Gardner, RC. 1995. Aluminum-induced genes. Induction
by toxic metals, low calcium, and wounding and pattern of expression in root tips. Plant
Physiol. 107: 341-348.

29

Staehelin, L.A., Moore, 1. 1995. The plant Golgi apparatus: structure, functional organization
and trafﬁcking mechanisms. Ann. Rev. Plant Physiol. Plant Mol. Biol. 46: 261-288.

Steer, M.W. 1988. The role of calcium in exocytosis and endocytosis in plant cells. Physiol.
Plant. 72: 213-220.

Steer, M.W., Steer, J .M. 1989. Pollen tube tip growth. New Phytol. 111:323-358.

Suhayda, C.G., Weis, C., Shi, B., Haug, A. 1988. pH-related changes in maize root plasma
membranes. Commun. Soil Sci. Plant Anal., 19:1165-1175.

Suhayda, C.G., Haug, A. 1986. Organic acids reduce aluminum toxicity in maize root
membranes. Physiol. Plant. 682189-95.

Sposito, G. 1989. The environmental chemistry of aluminum. CRC Press, Inc., Boca Raton,
FL. 336 p.

Taylor, G.J. 1988. The physiology of aluminium phytotoxicity. In: Sigel, H., Sigel, A. (eds).
Metal ions in biological systems: Aluminum and its role in biology, vol. 24, pp 123-163.
New York, Marcel Dekker.

Tice, K.R., Parker, D.R., DeMason, DA 1992. Operationally deﬁned apoplastic and
symplastic aluminum fractions in root tips of aluminum-intoxicated wheat. Plant Physiol.
100: 309-318.

Toth,1., Zekany, L., Brucher, E. 1984. Equilibrium study of the systems of Al(III), Ga(III),
and 1n(III) with mercaptacetate, 3-mercaptopropionate and 2-mercaptobenzoate.
Polyhedron 7:871-877.

Troncoso, J.C., Hoffman, P.N., Grifﬁn, J.W., Hess-Koslow, K.M., Price, D.L. 1985.
Aluminum neurotoxicity: a disorder of neuroﬁlament transport in motor neurons. Brain
Res. 342:172-175.

Van, H.L., Kuraishi, S., Sakurai, N. 1994. Aluminum-induced rapid root inhibition and
changes in cell-wall components of squash seedlings. Plant Physiol. 106:971-976.

Verrna, D.P.S., Cheon, C., Hong, Z. 1994. Small GTP-biding proteins and membrane
biogenesis in plants. Plant Physiol. 106: 1-6.

Viersna, R., Haug, A. 1978. The effect of Al” on the physical properties of membrane lipids
in Thermaplasma acidaphilum. Biochem. Biophys. Res. Comm. 84: 138-143.

Wagatsuma, T. 1983. Effect of non-metabolic conditions on the uptake of aluminum by
plant roots. Soil Sci. Plant Nutr. 29: 323-333.

3O

Wagatsuma, T., Minoru, K., Hayasaka, Y. 1987. Characteristics of upward translocation of
aluminum in plants. Soil Sci. Plant Nutr. 30:345-358.

Wagatsuma, T., Jujo, K., Ishikawa, S., Nakashima, T. 1995. Aluminum-tolerant protoplasts
from roots can be collected with positively charged silica microbeads: a method based
on differences in surface negativity. Plant Cell Physiol. 36: 1493-1 502.

Wheeler, D.M.,. Edmeades, D.C., Christie, R.A., Gardner, R. 1993. Comparison of
techniques for determining the effect of aluminium on the grth of, and the inheritance
of aluminium tolerance in wheat. In Randall, P.J., Delhaize, E., Richards, R.A., Munns,
R. (eds). Genetics aspects of plant mineral nutrition. pp 9-16. Kluwer Acad. Publ.,
Dordrecht.

Williams, R.J.P. 1992. Aluminium in biology: an introduction. In Chadwick, D.J., Whelan,
J. (eds). Aluminium in biology and medicine. pp 14 John Wiley & Sons, New York.

Williams, R.J.P. 1993. Magnesium: An introduction to its biochemistry. In Birch, N.J. (ed).
Magnesium and the cell. pp 15-30. Academic Press, London.

Yan, F., Schubert, S., Mengel, K. 1992. Effect of low root medium pH on net proton release,
root respiration, and root growth of corn (Zea mays L.) and broad bean (Viciafaba L.).
Plant Physiol. 99: 415-421 .

Youson, J ., Neville, C. 1986. Deposition of aluminum in the gill epithelium of rainbow trout
(Salma gairdneri Richardson) subjected to sublethal concentrations of the metal. Can.
J. Zool. 65:647-656.

Zhang, G., Taylor, G.J. 1989. Kinetics of aluminum uptake by excised roots of aluminum
tolerant and aluminum sensitive cultivars of T riticum aestivum L.. Plant Physiol. 91:
1094-1099.

Zhang, G., Taylor, G.J. 1990. Kinetics of aluminum uptake in T riticum aestivum L.. Identity
of the linear phase of aluminum uptake by excised roots of aluminum-tolerant and
aluminum-sensitive cultivars. Plant Physiol. 94: 577-584.

Zhang, G., Taylor, G.J. 1991. Effects of biological inhibitors on kinetics of aluminium
uptake by excised roots and puriﬁed cell wall material of aluminium-tolerant and
aluminium-sensitive cultivars of T riticum aestivum L. J. Plant Physiol. 138:533-539.

Zhao, X.J., Sucoff, E., Stadelmann, E.J. 1987. Al3+ and c5: alteration of membrane
permeability of Quercus rubra root cortex cells. 83:159-162

Zirnmerberg, J ., Vogel, S.S., Chemomordik, L.V. 1993. Annu. Rev. Biophys. Biomol. Str'uct.
22: 433-466.

31

Zsoldos, F ., Erdei, L. 1981. Membrane and ion transport properties in cereals under acidic
and alkaline stress. I. Effects of pH on potassium uptake and grth of rice and wheat.
Physiol. Plant. 53: 468-470.

CHAPTER 2

DEVELOPMENT AND VALIDATION OF AN IN VIVO ALUMINUM-MORIN
FLUORESCENCE ASSAY FOR THE VISUALIZATION AND QUANTITATIVE
DETECTION OF ALUMINUM IN TOBACCO BY-2 CELL SUSPENSIONS

ABSTRACT

The development, optimization and validation of an in viva aluminum-morin
ﬂuorescence method for the detection of A1 in cultured Nicatiana tabacum L. cv. BY-2 is
reported. Morin, a ﬂuorogenic reagent for Al, was added to cells in an aqueous medium (pH
5.6) compatible with cellular integrity, and was found to accumulate rapidly (<30 min) in
these tobacco cells. Accumulation occurred presumably via pH-dependent trapping in the
cytoplasm, due to virtually complete dissociation (99%) of the 3-hydroxyl group at the higher
cytoplasmic pH, as opposed to that (80%) in the loading medium (pH 5.6). The excitation
(AW=42O-450 nm) and emission ()11m =500-520 nm) spectra of the Al-morin complex in
cells and in aqueous solution were similar. This fact and the intensity and spectra of cell
autoﬂuorescence conﬁrmed that ﬂuorescence from cells treated with morin and/or Al
originated almost entirely from morin or its complex with Al. Based on the emission spectra
from cells, the Al-morin complex can be distinguished visually from free morin by intensity
and coloration. Maximum ﬂuorescence intensity was obtained by incubating cells with 100
MM morin. At higher concentrations, ﬂuorescence intensity is decreased by quenching. The
validity of this method for assessing Al in cells was tested by comparing results with graphite
furnace atomic absorption spectrometry (GFAAS). Under various experimental conditions
leading to wide-ranging levels of A1 uptake, results obtained ﬁom the Al-morin method were
highly correlated (r2=0.94) with those obtained with GFAAS. The advantages of this method

are several. It is quick, easy and inexpensive to perform, and permits quantitative

32

33

measurements as well as visualization under the microscope, with resolution comparable or

better than other methods. This method is also apparently insensitive to precipitated Al.

INTRODUCTION

The toxic effects of aluminum on plants and its impact on crop productivity and
forest decline in large areas throughout the world are well documented in the literature
(Haug, 1984; Delhaize and Ryan, 1995; Kochian, 1995). Increases in Al contents of surface
waters above previously innocuous levels, due to acid rain, have made it necessary to
monitor Al in water and aquatic organisms. There is, however, a dearth of knowledge
regarding mechanisms of Al uptake and toxicity in cells.

Difficulties in Al toxicity studies result from problems such as the complex chemistry
of Al (Martin, 1988; Delhaize and Ryan, 1995), particularly its pH-dependent hydrolysis.
Precipitation of Al hydroxides, particularly within the cell wall, can result in false
estimations of uptake and localization of A1 within the cell, when measured by graphite
furnace atomic absorption spectrometry (GFAAS) or by microanalytical techniques. Also,
the binding of Al to cell walls and its uncertain removal by chelating agents have made the
distinction between uptake into the protoplast and apoplast difficult (Zhang and Taylor,
1990; Tice et a1, 1992; Rincon and Gonzales, 1992; Delhaize and Ryan, 1995; Kochian,
1995). Because progress in this ﬁeld has in part been limited by methodology, further
progress may require the use and further development of new techniques (Delhaize and Ryan,
1995)

Ideally, a method for the detection of Al in cells should allow both the quantiﬁcation

and intracellular localization of A1 with high sensitivity and resolution. It should also

34

35

distinguish between precipitated and non-precipitated Al and should be easy, quick and
inexpensive to perform. Most studies have used GFAAS, X-ray microanalysis and more
recently, secondary-ion mass spectrometry for the detection and/or localization of cell-
associated Al (Zhang and Taylor, 1990; Delhaize et al., 1993; Lazof, et a1. 1994). However,
these methods do not meet several of the above mentioned requirements (e.g. see Lazof et
al., 1994). Although application of secondary-ion mass spectrometry has improved sensitivity
and resolution over X-ray microanalysis, it is still difﬁcult to relate images of Al content with
any cellular structure (Figs. 5 and 6 of Lazof et al., 1994) and requires extensive
instrumentation and sample preparation.

An alternative method involves the use of a ﬂavonoid, morin (3, 5, 7, 2', 4' -
pentahydroxyﬂavone). This highly sensitive ﬂuorogenic reagent is known in analytical
chemistry for the qualitative and quantitative determination of Al (Fig. l) in solutions
(Katyal, 1968; Katyal and Prakash, 1977). It has also been used in histochemical applications
as a dye for Al in microscope slide preparations (Eggert, 1970; Bani et al., 1974; Wen and
Wisniewski, 1985) and has recently been used by Tice et a1. (1992) in the examination of
apoplastic and symplastic Al fractions of wheat roots. Morin can bind Al even at low pH
values of 4.5, thus avoiding precipitation of Al, but because it is poorly soluble in water at
this pH, a water/alcohol/acetate mixture was generally employed. However, usage of morin
in aqueous medium, at other pH values, was demonstrated in studies on the aqueous
chemistry of morin (Browne et a1, 1990).

The development, optimization and validation of an in viva Al-morin ﬂuorescence
method for the detection of Al in intact tobacco BY-2 cells is reported. Validation of this

method was assessed by comparison of results with GFAAS.

36

 

0H + AI(OH),,<3'">+

 

o
\AI(0H),<2-n)+

 

 

Figure 1. Structure of the ﬂavonoid morin (3, 5, 7, 2', 4' - pentahydroxyﬂavone) and

presumed structure of its complex with A1.

 

MATERIALS AND METHODS

Cell cultures and Al uptake

Tobacco (Nicatiana tabacum L. cv BY-2) cell cultures were grown as described
(Nagata et al., 1992), at 28°C, in 50 ml of liquid medium in 250 ml ﬂasks and on a gyratory
shaker set at 165 rpm. The medium consisted of the Murashige and Skoog salts (Nagata et
al., 1992) supplemented with 30 g sucrose, 0.5 g MES, 100 mg myo-inositol, 1 mg thiamine-
HCl, 180 mg KHZPO4, and 0.22 mg 2,4-D (2,4-dichlorophenoxyacetic acid) per liter of
medium and the pH was adjusted to 5.7. Subcultures were made every 7 days with a 2.5 ml
inoculum (5% v/v dilution). Cells at two or three days after subculture were used in the
experiments.

To prevent precipitation of Al, or formation of Al complexes, Al uptake experiments
were performed in a medium consisting of 10 mM MES, 5 mM sucrose, 5 mM KCl and 2
mM CaCl2 (pH 4.5). Unless stated otherwise, the concentration of Al was 20 uM.
Experiments were performed at room temperature (23 °C). Stock solutions of AlCl3 (lmM)
were employed. Cells were ﬁrst pre-washed for 10 min in uptake medium at a ratio of 40
ml/cm3 of packed cells on a gyratory shaker at 60 rpm. Cells were then distributed into 15
ml plastic tubes and, following removal of the pre-wash supernatant, 4 m1 of Al-containing
(20 uM Al) uptake medium were added. The ﬁnal cell titer was 4 x 10’ cells/ml. Cells were
incubated with A1 during their allocated time on a shaker at 60 rpm. The pH of the cell
suspension was monitored in both the pre-wash and incubation steps and was found to be

stable during experiments.

37

38

Incubation of cells with Al was terminated by adding 0.1 M citrate (pH 4.5) to a ﬁnal
concentration of 5 mM, followed by removal of the Al-containing supernatant (operationally
deﬁned as the ﬁrst wash). For samples destined to GFAAS, cells were washed twice more
with 4 ml of uptake medium containing 5 mM citrate (pH 4.5) to further reduce the presence
of extracellular and cell wall A1.

For some of the optimization experiments, different levels of A1 uptake were desired
and generated by using cells at two different growth stages (2 or 3 days after subculture). In
these cases, results were normalized and expressed as percentage of response.

Ion exchange cellulose (pre-swollen high capacity carboxymethyl ion exchange
cellulose) from Pierce Chemical Co. (Rockford, IL, USA) was used as a control for A1
precipitation. Cellulose treatment was identical to that of cells, with 5 mg dry weight of

cellulose substituting for the 0.2 cm3 of packed cells.

Morin solutions and cell loading

Morin (3, 5, 7, 2', 4' - pentahydroxyﬂavone, C,5HIOO7, F.W. 302.2 g) was from Sigma
(St. Louis, MO). Stock solutions ﬁ'om two different lots of morin differed in intensity of
coloration, but did not produce different results in the ﬂuorescence assays. Morin was
prepared as a 0.1 M solution in DMSO. This concentrated stock solution was stable and
resulted in only a small carry-over of DMSO, which is commonly used in experimental cell
biology.

In aqueous solution, morin was poorly soluble at pH values between 4 and 5. Upon

addition of morin to water, the pH of the solution slowly dropped from neutral to about 4.5

39

with resultant precipitation. To prevent this, aqueous solutions were always prepared in
solutions buffered with 5 or 10 mM MES. Solubility and the development of a yellow
coloration increased with increasing pH. Consistent with this pH-dependent behavior, the
absorbance peak of morin in aqueous solution was found to shift from 360 to 385 nm, and
to increase in intensity by 64% in solutions of pH 4.8 to 6.5 (data not shown), as previously
reported in aqueous solutions (Browne et al., 1990). For comparisons, morin was also
prepared in an alcoholic solution of 0.2 % morin in 85% ethanol and 0.5% acetic acid, pH
4.5.

Following exposure to Al, cells were incubated with morin in fresh uptake medium
(pH 5.6) at a titer of 4 x 105 cells/m1. Morin concentrations ranging from 10 to 200 ,uM and
lengths of incubation ranging from 5 to 60 min were employed. The composition of this
medium was the same as that used for Al uptake except for a pH of 5.6. Although more
soluble at higher pH, morin was sufﬁciently soluble at pH 5.6 and presented a light yellow
coloration. After incubation with morin, the supernatant was removed and cells were
resuspended with 3.6 ml of uptake medium (pH 5.6) plus 0.2 ml of 0.1 M EDTA (pH 5.6).
The purpose of the presence of EDTA was to quench extracellular Al-morin ﬂuorescence.

Final cell titer was 4 x 10’ cells/ml.

Fluorescence intensity measurements

Measurements of ﬂuorescence intensity were performed on a Perkin Elmer (N orwalk,

CT, USA) LS-5 spectroﬂuorometer coupled to a microcomputer. Spectra were scanned at

120 nm/min, with a step interval of 0.5 nm. For ﬂuorescence intensity measurements, at

40

ﬁxed wavelengths, readings were monitored for 5 or 15 s at 0.5 5 intervals, for solutions and
cell suspensions, respectively. In all measurements slit widths of 10 and 5 nm for excitation

and emission, respectively, were employed.

Graphite Furnace Atomic Absorption Spectrometry

Following Al uptake and citrate washes to remove extracellular Al and Al bound to
the cell wall, cell pellets or cellulose material were digested with 250 [21 of 8 N HNO3 at
135 °C in Teﬂon screw-cap vials. After excess acid was dried off, samples were redissolved
in 1 or 2 ml of 25 mM HNO3 and analyzed for A1 with a Varian (Sunnyvale, CA, USA)
S pectrAA 400 graphite furnace atomic absorption spectrometer with Zeeman background
correction. The furnace program consisted of seven temperature steps set at 85, 95, 120,
1000, 1400, 2500 and 2600 °C for 5, 40, 10, 5, 6, 2.8, and 2 s, respectively. Pyrolytically
partitioned graphite tubes were ﬂushed with N2 gas. Sample volume was 20 a1 and readings
were performed during atomization at 2500°C, at 396.2 nm, with a slit width of 0.5 nm.
Absorbance values were based on peak height and standard solutions of AlCl3 from Sigma

(St. Louis, M0) were used for calibration.

RESULTS

Morin uptake by cells

Uptake of morin into cells (incubation at pH 5.6) occurred largely within the ﬁrst 30
min of incubation, as judged by the development of ﬂuorescence from the cells (Fig. 2).
Based on the decrease in absorbance at 380 nm of the external solution following a 20 min
period of uptake, it was estimated that approximately 14 % and 10% of the morin applied
was taken up by the cells at pH 5.6 and 6.5, respectively, at three different concentrations of
morin (20, 50 and 100 aM). Morin accumulation was found to be homogeneous throughout
the cytoplasm, with little or no accumulation of morin within the vacuole or cell wall
(V itorello and Haug, 1996). Incubation of cells with morin or with concentrations of DMSO
equivalent to those carried over with morin, was found to have no affect on cell viability, as

judged by trypan blue.

Spectra

The excitation and emission spectra of morin and Al-morin in cells were similar to
those recorded in aqueous solution and in a water/ethanol/acetate mixture (Fig. 3). The
excitation peak ranged from 420-450 nm and the emission peak ranged from 500-520nm,
similar to results previously reported in the literature. Excitation maxirna have been reported
to range from 410-440 nm and emission from 510-525 nm (Ueno et al., 1992; Saarl and

Seitz, 1983; Browne et al., 1990). The excitation spectrum of Al-morin in cells was

41

 

     

 

 

 

 

 

140
- 100
120 -
.L

A cells + Al
.3 + morin
c in—
E‘ v

o
g 80 A g
'2 - 60 g
3 m

o
3 ::
S :
° — 4o 8
8 cells - Al 3
3 + morin a
a
B I

- 20
= = = ———— + Al
autofluorescence - Al
O I I I I I I - O

o 10 ' 20 30 4o 50 60 70
time of incubation with morin (min)

Figure 2. Uptake of morin into cells of tobacco BY-2 cultures and ﬂuorescence intensity
from Al-treated cells as a function of duration of incubation with morin. Cells (2 and 3 days
after subculture) were pro-washed in uptake medium pH 4.5, then incubated for 1 h with 0
or 20 aM A1 (pH 4.5). Following this, cells were incubated with 100 11M morin in uptake
medium, pH 5.6, then resuspended in the same medium, free of morin and containing 5 mM
EDTA. To verify whether the concentration of Al in the cells determined the time needed for
morin incubation, differing levels of Al uptake were obtained by using cells of different ages,
and the resulting Al-morin ﬂuorescence intensities were normalized (ordinate on the right).
Uptake medium consisted of 10 mM MES, 5 mM sucrose, 5 mM KCl and 2 mM CaCl,, pH

4.5 or 5 .6. Bars represent SE (n=3).

43

Figure 3. Excitation and emission spectra of free morin and Al-morin in tobacco BY-2 cells
(A) and in an aqueous or ethanol/water/acetate solution (B). Scanning was performed at 120
nm/min. The slit widths were 10 and 5 nm for excitation and emission, respectively. A.
Aluminum uptake by cells (2 days aﬁer subculture) and morin loading were performed as
described in Fig. 2. B. The morin concentration was 10 and 50 “M in the aqueous solution
(uptake medium, pH 5.6) and 6.6 mM in the 85% ethanol/0.5 % acetic acid solution. The Al

concentration was 10 MM.

ﬂuorescence (arbitrary units)

ﬂuorescence (arbitrary units)

 

 

  
   
    
 

200
excitation emission
150 -
+ Al, + morin + A' I monn

100 —

5O -

" Al, "’ ”0'1“ - Al, + morin
f“? A
‘~ g = = _ -- "-=. g
0 autofluorescence autoﬂuoresce-r: “-
B excitation emission
200 — 10.3.32?" CQUQOUS
'0' AI
150 -'
ethanol!
acetate
100 '7
aqueous.
aqueous 50 Pr£°"“
5O - 50 uM morin
4' Al
O " — = = "" = = = = __ __ ._

 

 

380 400 420 440 460 480 500 520 540 560 580

wavelength (nm)
Figure 3.

45

closer to that found in aqueous medium than to that obtained from a water/ethanol/acetate
mixture whereas the opposite occurred for the emission spectrum (Fig. 3). These small
differences in the excitation and emission spectra of the Al-morin complex in cells compared
to those in solution are expected given the more diverse physico-chemical environments
within cells as opposed to that in simple solution. Shifts in the spectra of Al-morin depend

on the chemical environment (Escriche et al., 1983; Saarl and Seitz, 1983).

Optimization of Al-morin ﬂuorescence intensity

As a method for detection of Al in these tobacco cells, the use of Al-morin
ﬂuorescence was optimized, aiming for maximum ﬂuorescence intensity and sensitivity. At
a ﬁxed duration of incubation with morin (10 min), the ﬂuorescence intensity from cells
treated with Al depended on the concentration of morin in the labelling medium (Fig. 4),
increasing with concentration up to about 100 “M, after which it decreased as a result of
quenching of ﬂuorescence intensity, presumably by free morin. This quenching pattern was
also observed in solution albeit at concentrations of morin one order of magnitude lower.
Quenching probably explains why the Al-morin ﬂuorescence signal was fully developed
within the ﬁrst 5-10 min of incubation with morin (Fig. 2), although morin uptake continued
to increase and was completed only after about 30 to 60 min of incubation (Fig. 2). Thus, the
greater accumulation of morin within the cell was counter-balanced by quenching. Since
quenching did occur, optimization of cell titer for ﬂuorescence intensity readings was also
performed (insert Fig. 4). At the cell densities used, no decrease in ﬂuorescence at higher cell

densities was observed, but an inﬂection in the ﬂuorescence

46

 

 

 

 

 

 

 

 

 

120
100 -
$
2' 80 -
0
g 250
a A 2 days
In 60 — 0 3 200 - +N
2 g 'E
E 3 3 150 —
8 3 E 3 days
a 40 "‘ 8 g 100 _ '0' AI
e a e
h
5 50 - cells-Al 3‘2?
.— a— -- -" ‘—
20 _ 0 ”re 2 days
I l l I
o 200 400 600 800
cell titer (1o3 cells/ml)
o l l l l
O 50 100 150 200

Figure 4. Dependence of cell Al-morin ﬂuorescence intensity on the concentration of morin
in the incubation medium and the cell titer during readings (insert). Al uptake by cells (1 h,
20 uM A1, pH 4.5) and morin loading were performed as described in Fig. 2. The effect of
morin concentration was tested with different levels of A1 uptake, obtained by using cells of
different ages (2 and 3 days after subculture). The resulting Al-morin ﬂuorescence intensities

were normalized and expressed as percentage of maximum response. Bars represent SE

(n=3).

morin concentration (uM)

250

47

intensity curve was observed between 3 and 5 x 105 cells/ml. Although less pronounced, this
presumably resulted from quenching by free morin and/or absorption of light by the cells
themselves.

Because fluorescence readings ﬁom these cells are performed on a suspension rather
than in solution, readings ﬂuctuated because of the random movement of the cells within the
cuvette and as a result of sedimentation of cells in the cuvette. From time course readings of
ﬂuorescence intensity (Fig. 5), reductions in ﬂuorescence intensity due to settling of cells
were detectable after about one minute. Despite the problem of sedimentation of cells, the
signal was sufﬁciently stable for enough time so that measurements could be made. To
compensate for ﬂuctuations in readings during this time, measurements were made at 0.5 5
intervals and averaged over 10 to 30 5. When cells were allowed to settle to the bottom of
the cuvette aﬁer a few minutes (Fig. 5), the ﬂuorescence signal decreased and approached
zero, reaching even lower values if the cells were centrifuged. Thus, ﬂuorescence originating
from the solution bathing the cells was minimal, but was higher in cells treated with morin
in an alcohol/acetate solution, which compromises cell integrity. Additionally, EDTA was
added at a concentration of 5 mM to the cell suspensions following incubation with Al and
morin and was always present during ﬂuorescence intensity readings. In solution, 5 mM
EDTA (A1 binding affinity constant of log K=16.5; Orvig, 1993) was capable of quenching
all ﬂuorescence (99% reduction) from the Al-morin complex (binding afﬁnity constant
estimated to be log K= 4.2-6.5; Katyal and Prakash, 1977; Saarl and Seitz, 1983) and can
thus by utilized to further reduce the contribution of ﬂuorescence emanating from the bathing
solution. Citrate, with an Al binding afﬁnity constant (log K=7.8) closer to that of morin, was

less effective at quenching Al-morin ﬂuorescence (37% reduction).

b
00

 

 

  
 
  

 

  

 

 

 

 

100
I a
V100 _ m
8
r: 80 -
8
e"? 80 - a, 50 _
a: 2
C a 40 -
3 5
E‘ 2 20 -
(U ‘ or
E 60 _ D. 0 I 1 ﬂ 1 1
ﬁ- 0 20 40 so 80100
3' time following morin
u incubation (min)
5 40 -
o
0
in
o
I.
3 -Al
: 20 - + morin
o I . I ' .
O 50 100 150 200 250

time (seconds)

Figure 5. Readings of Al-morin ﬂuorescence intensity from tobacco BY-2 cells as affected
by movement and sedimentation of cells and stability of the Al-morin ﬂuorescence signal
with time (insert). For readings during sedimentation, measurements were made at l 5
intervals and collected with a microcomputer. Cells (3 days after subculture) had almost
completely sedimented within the time frame involved. To test the stability and
reproducibility of the Al-morin ﬂuorescence signal, readings were made on the same cell
samples at different times following morin incubation. This was tested in cells with different
concentrations of A1, by using cells at 2 and 3 days after subculture and the resulting Al-
morin ﬂuorescence intensities were normalized and expressed as percentage of maximum
response. Cells were incubated with Al (1 h, 20 uM A1, pH 4.5) and morin (100 uM, 10 min)

as described in Fig. 2. Bars represent SE (n=3).

49

The Al-morin ﬂuorescence signal from cells was found to be repeatable overtime
(insert Fig. 5). Essentially the same readings could be obtained from the same sample over
a period of about an hour, with small decreases thereafter. In a spectrophotometric
application for the determination of Al by morin (Ahmed and Hossan, 1995), absorbance

values were found to be stable for 48 hours.

Comparison of the Al-morin method to GFAAS

From the optimization experiments described above, a desirable compromise of
conditions was established, taking into consideration maximum ﬂuorescence intensity for
greater sensitivity, quenching, time spent, stability and reproducibility of results. This was
achieved by loading cells with morin at pH 5.6, 100 uM morin, for 10 min.

To ultimately validate this Al-morin technique, a comparison was made between
results obtained from this technique and graphite furnace atomic absorption spectrometry,
a standard analytical technique for quantiﬁcation of Al, and the predominant analytical
method employed in Al uptake studies. Good agreement (r2=0.94) was found between results
from these two techniques (Fig. 6) when both were used to measure cell-associated A1 in
parallel samples. This comparison covered wide-ranging levels of Al content, which had
been obtained under various experimental conditions (V itorello and Haug, 1996).

The above comparisons were evaluated at a pH of 4.5, at which Al precipitation does
not occur at [A1] 3 20 uM. Control experiments with cellulose provided supporting evidence
that precipitation did not occur. This control experiment was undertaken because GFAAS

will also quantify precipitated Al present in the cell pellets. In contrast, the Al-morin

50

 

    
  
  

 

 

 

25
20 _ y=x*0.13+0.71 O
O
a; r2=0.94
2
'5 15 _
2o”
< E
(“5 a
.0 10-
E
r:
5..
0 l i I l l I i l

0 20 40 60 80 100 120 140 160

AI-morin ﬂuorescence
(arbitrary units)

Figure 6. Correlation between results of A1 uptake by tobacco BY-2 cells measured by the
Al-morin method or graphite furnace atomic absorption spectrometry (GFAAS). Al uptake
was performed under different experimental conditions and resulted in wide-ranging levels
of Al uptake and concentrations of cellular A1. Factors which were varied were: age of cells,
time of incubation with Al, concentrations of CaCl2 in the uptake medium, concentrations
of sorbitol in uptake medium, temperature, pre-chelation of A1 to citrate and pH of the pre-

wash solution (V itorello and Haug, 1996 - chapter 3). In all cases, cells were incubated with

100 uM morin for 10 min, following Al uptake.

51

technique appeared to be insensitive to precipitated Al (Fig. 7). When Al (20 uM) was
incubated with cells in a solution at pH 5.6, a pH value which will knowingly lead to at least
some precipitation of unchelated Al (Kragten, 1978), uptake results differed between the A1-
morin technique and GFAAS. Compared to pH 4.5, uptake at pH 5.6 measured with GFAAS
indicated increased levels of Al associated with the cells, whereas the Al-morin method
reported less Al associated with the cells (Fig. 7). This discrepancy in results can be
attributed to precipitation of Al, which can cause overestirnations of Al uptake. That
precipitation of Al did indeed occur was evidenced by control experiments with ion exchange
cellulose (Fig. 7). At pH 4.5, but not at pH 5.6, Al was quantitatively removed from the
cellulose after citrate washes, as determined by GFAAS. At pH 5.6, even after the citrate
washes, about 60 % of the applied Al was found to be associated with the cellulose, with
nearly identical results in cells, indicating precipitation of Al. Intemalization of Al by the cell
should be expected to decrease at pH 5.6 since Al is not toxic at this pH and a correlation
between A1 toxicity and Al uptake has been demonstrated in tobacco cell cultures
(Yamamoto et al., 1994). Thus, GFAAS can overestimate Al uptake when precipitation
occurs whereas the Al-morin method probably reports results closer to genuine uptake. It
should be noted that precipitation can also occur at lower pH values if rather high Al
concentrations are used. In addition, Al solutions are most often unbuffered. Thus, even
though Al may be added in a solution of sufficiently low pH, high A1 levels, close to the limit
of solubility, associated with ﬂuctuations in pH might cause precipitation of Al and

overestimation of Al uptake.

52

 

 

    
   

   

 

 

 

 

 

150 (55
125 - — 50
‘6
E
_ — 45 a
E. 100 J <
I- x <
E in"
I
5, 75 - [15 :
at <
O In
a a, a %
o _ o
g 50 g g ~10 E
g o 2
a: g g.
25 ' o 3 — 5
3 O
E
O
o — o
a b c a b c
pH 4.5 pH 5.6

Figure 7. Aluminum uptake by tobacco BY-2 cells and by cellulose phophate, at pH 4.5 and
5.6, as reported by Al-morin ﬂuorescence intensity or GFAAS. Cells were incubated with
Al (1 h, 20 “M Al) and morin (100 uM, 10 min) as described in Fig. 2, with the exception
of the pH of the medium.Cellulose phosphate was treated identically to that used for the

cells, with 5 mg substituting for each 0.2 cm3 of packed cells. Bars represent SE (n=3).

DISCUSSION

The major ﬁndings of this report are that morin can be successfully loaded into intact
cells of tobacco BY-2 cells in an aqueous medium, and that morin can be used to reliably
detect Al in these cells following uptake. The conditions for use of this technique were
optimized for maximum ﬂuorescence intensity. As discussed below, this technique has
several advantages over previously used methods.

Loading of intact cells with morin, in an aqueous medium, was possible presumably
because of its pH-dependent solubility behavior (Fig. 2). Compared to the extracellular
solution employed (pH 5.6), the higher pH value of the cytoplasm allows for more extensive
dissociation of the 3-hydroxyl group, higher solubility of morin and thus presumably results
in a pH-dependent trapping of the dye. The pK., for the ﬁrst hydroxyl group (3-hydroxyl
group) to undergo dissociation has been estimated to be 5.0 (Browne et al, 1990). At pH 5.6,
about 80% of morin will be dissociated in contrast to 99% at pH 7.0. Thus, at the lower pH,
there is still a reasonable quantity of undissociated, neutral morin to traverse the plasma
membrane whereas there is almost none once the compound reaches the interior of the cell.
That the dye is contained within the protoplast has been demonstrated by plasmolysis
experiments (V itorello and Haug, 1996). The same principle of pH-dependent ion trapping,
which can explain entry of morin into the cell can also explain why the dye was not detected
to accumulate within the vacuole. Loading of cells with naringin (4', 5, 7 -
trihydroxyﬂavanone 7- rhamnoglucoside) a glycosylated ﬂavonoid, with higher solubility in
water than morin, was also attempted but was unsuccessful.

From the excitation and emission spectrum of Al and morin loaded cells and from

53

54

relative intensity values (Fig. 3.), it is safe to conclude that the ﬂuorescence measured from
these cells originated largely from morin and its complex with Al, and that the ﬂuorescence
behavior of this reagent is not signiﬁcantly altered by the cellular medium. Additionally, as
conﬁrmed by the emission spectrum of morin and Al-morin in cells, a distinction between
complexed and non-complexed morin can be achieved based on coloration and intensity,
which is of importance for the observation of cells under the microscope. Complexed morin
is bluish-green, in agreement with its emission maximum, while free morin, because of its
low and broad emission spectrum, appears olive-green or yellowish green. Actual color tones
may, however, vary depending on the ﬁlter set used in the microscope.

In addition to determining conditions for maximizing ﬂuorescence intensities, the fact
that quenching of the morin signal occurs, may be advantageous for reproducibility of
results. Employing morin concentrations in the plateau area probably helps stabilize the
ﬂuorescence signal against small ﬂuctuations in morin uptake resulting from variances in
duration of incubation or possible heterogeneities in cell populations (Figs. 2 and 4).

The advantages of this Al-morin method for the detection of Al in cells are manifold.
It is easy, quick, and inexpensive to perform. It allows both for quantitative measurements
and for visualization and localization under an epiﬂuorescence microscope, using ﬁlter sets
similar to those used for the common ﬂuorescent dye ﬂuorescein. In comparison to other
methods for the cellular localization of Al, it has higher resolution, which theoretically
should approach the limit of optical microscopy. Additionally, superior imaging of cellular
Al could probably be obtained if confocal microscopy were utilized. Another advantage of
this method is that it is apparently not sensitive to precipitated Al.

Another possible advantage of this technique is that it probably distinguishes between

55

apoplastic and symplastic Al. Evidence for this and a discussion are presented by Vitorello
and Haug, 1996. Whether or not Al can be totally removed from the cell wall, from sites with
unknown binding affinity, by use of chelating agents has been disputed. If contained within
the cell wall, ﬂuorescence of the Al-morin complex is achieved either because morin strips
Al off these sites or (more likely) because of the possibility of multiple ligand binding. In
either case it should still be possible, in principle, to quench Al-morin ﬂuorescence with
EDTA, because the relative binding afﬁnities of morin and EDTA are known, regardless of
whether Al is actually removed from its site on the cell wall.

Whereas the application of this technique to cell suspensions is straightforward, its
application to other plant model systems, such as roots, may require some adaptations, but
should nevertheless prove to be advantageous. As for observations with the microscope,
tissues made up of several layers of cells could be observed by confocal microscopy.

Despite its numerous advantages, this method reports only relative amounts of Al,
not absolute ones. Thus it is imperative that when using this method in a new cell or tissue
system, that results obtained with Al-morin be initially checked or calibrated with some other
standard method of Al quantiﬁcation. This may also be important since the chemical
environment, particularly whether hydrophobic or hydrophilic, of the Al-morin complex can
alter ﬂuorescence levels.

Al-morin is an advantageous technique to be employed when numerous samples are
being studied, such as in screening programs. If not used as the primary method for detection
of Al, it would be advantageous as a preliminary method, such as in exploratory experiments,

which could be later conﬁrmed with other methods of analysis such as GFAAS.

REFERENCES

Ahmed, M.J., Hossan, J. 1995. Spectrophotometric determination of aluminium by morin.
Talanta 42:1135-1142.

Boni, U. De, Scott, J.W., Crapper, DR. 1974. Intracellular aluminum binding: a
histochemical study. Histochemistry 40:31-37.

Browne, B.A., McColl, J.G., Driscoll, CT. 1990. Aluminium speciation using morin: I.
Morin and its complexes with aluminium. J. Environ. Qual. 19: 65-72.

Delhaize, B, Ryan, RR. 1995. Aluminum toxicity and tolerance in plants. Plant Physiol.
107: 315-321.

Delhaize, E., Craig, S., Beaton, C.D., Bennet, R.J., Jagadish, V.C., Randall, P.J. 1993.
Aluminum tolerance in wheat (T riticum aestivum L.) I. Uptake and distribution of

aluminum in root apices. Plant Physiol. 103: 685-693.

Eggert, DA. 1970. The use of morin for ﬂuorescent localization of aluminum in plant
tissues. Stain Technol. 45: 301-303.

Escriche, J .M., De la Guardia Cirugeda, M., Hernandez, F .H. 1983. Increase in the
sensitivity of the ﬂuorescent reaction of the complexing of aluminium with morin using
surfactant agents. Analyst. 108: 1386-1391.

Haug, A. 1984. Molecular aspects of aluminum toxicity. Crit. Rev. Plant Sci. 1: 345-3 73.

Katyal, M. 1968. Flavones as analytical reagents - a review. Talanta 15: 95-106.

Katyal, M., Prakash, S. 1977. Analytical reactions of hydroxiﬂavones. Talanta 24: 367-375.

Kochian, L.V. 1995. Cellular mechanisms of aluminum toxicity and resistance in plants. -
Annu. Rev. Plant Physiol. Plant Mol. Biol. 46: 237-60.

Kragten, J. 1978. Atlas of metal-ligand equilibria in aqueous solution. John Wiley & Sons,
New York, pp 16-39, 52-68.

56

57

Lazof. D.B., Goldsmith, J .G., Rufty, T.W., Linton, R. 1994. Rapid uptake of aluminum into
cells of intact soybean root tips. A microanalytical study using secondary ion mass
spectrometry. Plant Physiol. 106: 1107-1114.

Martin, RB. 1988. Bioinorganic chemistry of aluminum. In: Sigel, H., Sigel, A. (eds). Metal
ions in biological systems: Aluminum and its role in biology, vol. 24, pp 2-57. New
York, Marcel Dekker.

Nagata, T., Nemoto, Y., Hasezawa, S. 1992. Tobacco BY-2 cell line as the “HeLa” cell in
the cell biology of higher plants. Int. Rev. Cytol. 132:1-30.

Orvig, C. 1993. The aqueous coordination chemistry of aluminum. In Robinson GH. (ed).
Coordination chemistry of aluminum . pp 85-121. VCH Publishers. New York.

Rincon, M., Gonzales, RA. 1992. Aluminum partitioning in intact roots of aluminum-
tolerant and aluminum-sensitive wheat (T riticum aestivum L.) cultivars. Plant Physiol.
99:1021-1028.

Saarl, L.A., Seitz, W.R. 1983. Immobilized morin as ﬂuorescence sensor for determination
of aluminum (1H). Anal. Chem. 55: 667-670.

Tice, K.R., Parker, D.R., DeMason, D.A. 1992. Operationally deﬁned apoplastic and
symplastic aluminum fractions in root tips of aluminum-intoxicated wheat. Plant Physiol.
100: 309-318.

Uemura, E., Minachi, M. and Lartius, R. 1992. Enhanced neurite growth in cultured
neuroblastoma cells exposed to aluminum. Neuroscience Letters 142: 171-174.

Ueno, K., Irnamura, T., Cheng, KL. 1992. Handbook of Organic Analytical Reagents. CRC
Press, Boca Raton, pp 25-29, 158.

Vitorello, VA. and Haug, A. 1996. Short-term aluminium uptake by tobacco cells: Growth
dependence and evidence for internalization in a discrete peripheral region. Physiol.
Plant. (in press).

Wen, G.Y. and Wisniewski, HM. 1985. Histochernical localization of aluminum in the
rabbit CNS. Acta Neuropathologica 68: 175-184.

Zhang, G., Taylor, G.J. 1990. Kinetics of aluminum uptake in Triticum aestivum L.. Identity
of the linear phase of aluminum uptake by excised roots of aluminum-tolerant and
aluminum-sensitive cultivars. Plant Physiol. 94: 577-584.

Yamamoto, Y., Rikiishi, 8., Chang, Y., Ono, K., Kasai, M., Matsumoto, H. 1994.
Quantitative estimation of aluminium toxicity in cultured tobacco cells: correlation
between aluminium uptake and growth inhibition. Plant Cell Physiol. 35: 575-583.

CHAPTER 3

SHORT-TERM ALUMINIUM UPTAKE BY TOBACCO CELLS: GROWTH
DEPENDENCE AND EVIDENCE FOR INTERNALIZATION IN A DISCRETE
PERIPHERAL REGION

Vitorello, V.A., Haug, A. 1996. Physiologia Plantarum (in press)

58

PHYSIOLOGA MW 96: xxx-an. 1996
mum-aw“

womb-Int!"
must-an

Short-term aluminium uptake by tobacco cells: Growth
dependence and evidence for internalization in a discrete

peripheral region

VidorLVltorelloandAlh'edﬂang

V.A.aud A.l996.Shut-unmaluminimunakebytohaeeoeens:
“mile. m . . il I

wuemtammham

96:

ii
2
5

E

m-

222

222
22

22

=1

2

5%

2"

232
2,21% .
f
2

.222-

,3
:2
2
E
3
2
E

F
i
2
22
E
E
E
2

mm-Wmmmwmmnw
www.mmmmammm

uh. W.

SA:CaInde£ns iaNadeerna

Bo “Plant? M 'mSmUuEmEaser-
W m W mascara».
..Brusil (W “1.1. Hang (eon!-

MW; Dept alum um. Sum um. East mm. MI
43324. .

lntruduetion

Aluminium toxicity is a major factor limiting plant
pounhmacidicsoilseoveﬁngvastregionsoftheworld
(Foyetal. 1978.ng l984).andhasalsoreeeived rec-
ognition asapotential probleminhumaus and animals
(Ganrot 1986). Despite considerable research endeav-
ans.meehanismsofAltoxieityarestillpoorlymder-
modpuucularlywithrespeetwcellularsitesofwxie-
ity(l(oehian l995).ltisstilldisprnedwhetherinitial
miceﬂ’eetsofAlnemanifestedintheeellwalla
withiutheplanteellLStudiesouAluptakeanditseellu-

Received 180mb.l995:uvisad6m1996

“~91!”

h localiztim are central to the elucidation of such
Ryan 1995.Kochim 1995).Diacrepaneies.inpart.result
from methodological diﬁcuhies. such as the complex
aolutionchenusuyofAHOrvig l993).thelaekofasuit-
able Al radioisotope. the potential fa ”face binding 6
AlandformatimofAlruecipitatesintheeellwallmin-
laidel991.'liceetal.l992).aswellastheratherlinuted
resolution of microanalytical techniques. such as elec-
uoupmbex-raynucmaualyais (Lazofetal. 1994).
Concerning Al uptake studies. Zhang md Tayla'
(1989.1990)analyaedthekinetiesofAluptakewiththe

59

purposeofdistinguishingbetweenAlresidingintheap-
oplastarrdthatinthesymplast'l'tceetal.(1992)deviscd
afractionationprocedureinroots.andmorerecently
Lard et al. (1994) utilized secondary-ion mass spec-
nometrytoexaminethedistributionot'Alinrootapices.
Onbalance.evidenceappearstofavourtheideathatAl
israpidlytakenupintotherootapicalsymplamal-
thoughconsensusisapparentlylackingandpoblems
esistineachoftheseapproaches(lcocbianl995).
Apmﬁomhausﬁkewmpuuuremwnmd
Strid 1989. Zhang and Taylor 1990). genotype (Zhang
and Taylor 1989). and metabolic inhibitors (Wagatsuma
1983). thereeaistsadearthofinformationregardingpa-
rameters impacting A1 uptake. To illustrate. althouﬂr
Ca"iswellknowntoameliorateAltoxicity(Kinraide
andParker1987.DelhaiuandRyan1995).thereis
scantinformation(Alvaetal.1986)astotheactualup-
takeofAlbyplantcellsanditsdepardencementerual
Ca” concentrations.
ForthisAluptakesmdy.weae1ectedamluuedto-
baccocelllinecharactetisedbyhighgrowthratesand
high homogeneity (NagataetaL1992). Guidedinpart
byornprevious experimentsonmammaliancellsﬁhi
andl-laug1990).Aluptakewasassessedbypaphitefur-
nace atomic absorption spectromeu'y (GFAAS).Inaddi-
tion. a novel ﬂuorescent dye assay was developed. em-
ploying morin in aqueous solution. Our results indicate
thatAlisrapidlytakenupbyactivelygrowingtobacco
cellswhereAlisapparentlyloulizﬁathastinitially.
indisceteaorresneuthecellularpaiphery.

Abbreviations-HM

graphitel'tanaceauuulcabsuptiou
mam.mmm

Materialsand methods
Chm-Id hbwue

MunshigeandSkoogbasalsaltmixuueandmmin
(2'.3'.4',S.7-pentahydroxyﬂavone) were purchased from
Sigma (St. Louis. MO. USA). All other chemical re-
agemsusedwereofanalyticalgnde.61ssswarewasem-
ployed only as the culture vessel and for some pipetting.
andwuwashedm'th 10%lm03htordertodinu'nish
backgrormd levels of Al. Otherwise polypropylene or
polyethylene labware was utilized. KOl-l uHCl (0.1 toS
M)wasusedforadjustingpl-lofallsolutions.1-lighqual-
itywater. obtaineduaptoductafterreverseosmosis.
deioninticnanddistillation.wasuscd'mallertperimems.

Cellular.

CensuspensionatlnuesofNiearimmbemLev.
Bright Yellow2(BY-2)weregrownatpl-15.6underpe-
viously described conditions (Nagataet al. 1992). Cul-
ttu'esweregrown.at_28'C.in50mlofliquidmediumin
250ml£rlenmeyerﬂaskashakingatl65rpmmagyrr
tory shaker. Subculttues were prepared every 7 days
witha25m1-inoculum.

2

Cellgrowthwasassessedbymeastuingpackedcell
vohmeG—SmhtcennifugationatZOOgingaduatcd
tubes). Eachcm’ ofpackedcellscontained7.9x 10‘
cellsandcorrespondedtoo.6goffreshweigbt.Cells
werecotmtedafterdisaw'egationwithsﬁo-Oﬁaque-
ous).ar70C.fcr30min.Proteincontentwasl.3tol.S
mgptoteinperlo‘cells.meastuedbythe3radfud
(1976)microassay .

Cellmembrane "tywasassmdwithtrypan
b1ue.Cellswereincubatedfor$mininal:1(v/v)mix-
uueofcellstrspensionarrdo.4%(wlv)u'ypanblue.Cells
inculturewerevirtuallyimpermeabletouypmblueo
98%ofcells).

Ankh-uptake
Uptake medium consisted of 10 mM 2-(N-mor-
pholino)etlnnesulfonicacid(MES).5mMsucrose.and
SmMKQpH45.Calciumconcentmimswereeither
0.2.2.0,a’mmMCa03.'l'heAlconcenuationwas20
uMinallenperiments.AlthoughMEShasapK.of6.1.
itwasusedasbuﬁerbecauscorganicacids ' em-
ployedintlu‘spHrangecanchelateAl(Orvigl993).ln
1ieuofamtuesuitab1ebuffer.MBSwillat1eastresistin-
creasesinpl-Lamaja'probleminAltmticitystudies.
Allertperimentswereperformedatroomwmperature
(23C)un1essstatedotherwise.A1uminiumstoeksolu—
tims(1mM)wereprepared.gena-allythesamedayof
anexperimartbyadditioncfAlClﬁI-Iﬁtointenaely
stirredwaterpeviouslyadjustedtoapl-lofSSto3A.
likewiseAlChstockwassimilarlyaddedtouptakeme-
diumimmedmelypriortoincubatiatofcells.
Cellswetelne-washedinuptakemeditmrataratioof
40mlcm‘ofpackedcellsandincubatedfa10minma
shakerat60rpn.8tandardpe-washsohuionshadapll
d45andcootained20mMCaCh. Subscqueotly.cells
waedisuibutedirno15mlplasticurbes,eachutbereceiv-
ingOlcm’ dpaekedcellal-‘ollowmgremovalotthepe-
wuhsupanatmtcellswaeincubatedwith4mlofthe
Al-containinguptakemediquH45).onashakerat60
maﬁadhrgaﬁnaleelltiued’4xto’cellsmr'. Total
elapsedtimewasappoahnatelywminfrunthebcgin-
uingofpe-washsteptothebegimingofAlinatbatiut.
StanthrduptakemedimncontainedOlmMCaCla. The
pl-lofthecellsuspensionwasfoundtobestableduring
experimentsAluptahewashahedbyadditionofOJmI
ofO.1Mciune.pH45. followedbyremovaloftheAl-
calmingsuperana’sunplesdcsdnedeAAS.
cellswerewashedtwicemorewith4mloftmtakeme-
dimnphrsSmMcitrate(pl-l4.5.2mMCaCl,J.Elapaed
tinnwasZSminfromtheeodofAlwonﬁuitialad-
ditionofciuate)bcompleticnofwashes.

A-lylselAlbygrapﬂschrr-esasc-Ieabam-pﬂco
m

Sampleswereoven-driedat115Cinlooselycappcd.5-
mlehnsaew-capﬁahthendigestedwichSOulof

“‘97.!“

60

8Mlm0,intighdycappedvialsat135C.Exces'sacid
was dried off and digested samples were redissolved in
lor2mlof25li-lNO,. Sampleswereanalyacdfa’
AlwithaSpectrAAItOOgraphiteﬁnnaceatomicab—
sorption spectrometer (Varian. Stmnyvale. CA. USA)
wicheemanbackgrormdcorrection. Readingswerere-
corded in duplicate. at 396.2 nm. Pyrolytically coated
pardtionedgraphitembeswereflushedwithNggasA
seven-step furnace program was applied as recom-
mended by the manufacturer. Amintion temperature
wassettoZSOOCwithapmdingchartemper-anueof
IOOO’C.
RecoveryofappliedAlfromthecellpelletsandwash
supernatantsofseveralsampleswasdetermincdtobe
102:3.4%(se).simi1artoﬁndingsonAlrecoverycm-
ductedonnetu'oblastomacells(8hiandl-laug1990).
However. cell pellets yielded more remoducible results
and werethereforeused the ' 'tm.
BackgrotmdlevelsofaboutZumolAl(10‘cells)“ftom
controlcells(noaddedAl)weresubtractedfromthere-
sultsofAl-treatedsamples.

Mailman-y

ln=paralle1toGPAAS.theAlconsentincellswasalso
evaluatcdinvivowithmain.aﬂuorescentdyewith
highsensitivityforAHinviuodetectabilitylimitof
about 0.04 uM). commonly used in analytical applica-
donsmltyllmdPrakashlm).Morinisuaditionally
employed in a methanol and/or acetic acid based solu-
nonmltydandPrakashlm.‘l’tceetal.1992).1-low-
ever.wehavesucceededinloadingthisdyeintointl:t
cells (see Results. Fig. 3).inabuffered aqueous solution
(pl-1 5.6).freecfmethanol and acetate'l'beaqueoussol-
ubihtyofmainaweakpmnprndcaddislowbutin-
creaseswithincreasingpl-lmenoetal. 1992).inaccord
with the pit-dependent dissociation of its hydroxyl
poupaTheB'hydmylpwphuapnofSDandis
theonlyoneoftheﬁvetomtdergomajorprotondissoci-
ationatapl-llessthan7.0(Browneetal. 1990). Thus.
the dye was insoluble and precipitated in aqueous solu-
tionsun'thapllbetween4t05.butwhenbuffetedatpll
5.6 yielded a yellow. translucent solution.

Stock solutions of morin (0.1 M in dimethyl sulfox-
ide)werestoredinthedarkandwerepepsredfreshev-
erytSdastmmediatelyafterAluptakeandciMe
wash.cellswereresuspendedin4rnlof100uMmorin
inuptake meditnn (pl-l 5.6.2mMCaC13) and incubated
for 10 min.at60rpm. Following removal of the super-
natanLcellswereresuspeodedwith3.6mlofuptakeme-
diurn(pl-1'.a.n.2mMCaC1;)p1usO.2mlofO.1Methyl-
enediaminetetraacetate (EDTA) (pH 5 6.) affording a ﬁ-
naltitreof4x10’ce11sml". Tominimizedyeself-
quenching. these conditions were selected following ex-
perimentswheremorinconcenuation (10—500 uM).in-
cubau'on time (5-60 min). and cell titre (1.5-8 x 10’
cellsml“)weretestedandoptimiaed.inthepresence
andabsenceofAl(20|rM.1hincubationofce11s).

“M9119”

CellswereobsetvedwithaWLstandardresearchmi-
eroscope equipped withanepifluorescence condenser
NFL (Carl Zeiss. Oberkochen, Germany). The ﬁne set
usedconsistedofabandpassﬁlterml’wO-MOnm)
andalong-wavepassﬁlterMHO nm).ﬁlmutilized
fuphotomiaographywaskoyalGolleOOCEssunan
KodakCo..Rochester.NY.USA).

Fluorescence intensities were derivcdfromsuspen-
sionswithals-SlauninesceoceSpcctromcteerin-
Elmer. Norwalk. CT. USA) coupled to a microcomputer
and chart recorder. The excitation and emission wave-
lengthsweresettoMO:5nmand515:2.5nm.re-
spectively.Cellsweresuspendedinacuvetteandthe
ﬂuorescence intensity was subsequently monitored at
0.5sintervalsEachsamplereadingwastakenastheav-
erageofthcftrst30sofmonitoting.llesultingfromset-
dmgofcenstedttcdonsinﬂuorescenceintensitycould
bedetecsedonlyafteratleastlmin.Whencellswereal-
lowcdtosettlecompletelleuorewencereadingsofthe
super-unantapproachedsero.dernonsu-atingtheabsence
dmhuitscompleawithALinthemedimFluo-
rescenceintensityofcellsampleswasstableforatleast
1h.

Ranks
minty-keno!”
AltuninitunupnkeaotrMJhincubatiomwasassesaed
atdifferentstagesofthecellculunecyclem. 1A).and
wasfoundtobecriticallydepeodentontheageofthe
culuue. Under standard conditions (pH 4.5. 0.2 mM
CaCl.).Aluptakepeahcdat2daysaftersubcultmebut
wasnegligibleatSand6days(Frg.1A).whichcorre-
spondcdrespeetivelynoperiodsofhighandlowgrowth
activity(l’tg.18).Relativegrowthrate(RGR)isanindi-
rectindicatorofthepercentageofcellsengagedin
growthanddivisionatagiventimemditsmaximum
valuewasformdtobe1129day"at2daysaftertransfer
(ﬁg.18).lndeed.tbemitoticindexwasfoundtopeakin
dtistobaccocelllineatZ-Bdaysaftersubculun'eanda
similarRGRwasreportchilagataetal. 1992).Asignif-
icantcorrelationbetweenAluptakemdrelativegrowth
ratewasfouud(insert.ﬁg. 1A). Therefore. activecell
growth is required fa A1 uptake under conditions of
short-tam expostu'etoAl. Regardinglong-termexpo-
stne. Aluptake(100uMAl.pH4.0)occurredaftera10
hperiodinlogarithmic-phasecellsofawbaccocellcul-
mbmwunotobsu'vableataninstationmy-phaae
cells('Yamamotoeta1. 1994).
Allsubscquentuptakeesperimentswereperformed
withceusathayspost-uansferratherthmwithcellsat
2dayspost-transfer.becauseofthegreaterquantityof
cellsavailableandthebetterreproducibilityofdata.

The-coursedAluptake

Aluminiumuptakecouldbeatectedjustminutesafter
initialincubation(pl-l45,0.2mMCaCl,)byboththe

3

 

 

 

 

 

 

€200 ' g
.3
€150 r- i
.- '5
100 - =—
2
E so . E
g
0

1‘ nus-u stunt were) “i: 8‘

:9 125
g i
g 1m 5
e ’5 l
2' mi
5 25 i

'0

 

 

and
inmeditunthatcontainedSliTA Acelltitreof xlo’
cellsmr'wasusedintbpre-washstep. nd4x10’cellsmr'
gcmu'clforfluo-

10mMMBS.5mMsucrcse.5mMKCl.
and2.0(me-wash)cr0.2ﬂ(Alincubatiat)mHCaC1..Valnsle
themeanofatleastd forAl-morinllutaesaence-d
anhcanonsquFAAS. anrerxesentsalnsutCorrelation
betweenAluptake.asaessedbybotbGFAASandAl-morinﬁu-
aeacence.andrelativegrowthrle(RGR. Fig. lB)of¢lls.L'm-
earmpession onsandthecerrespondiugcoelftcierasof
determimtiar. .werey-0.13RGR+1..9 t’nOJJfor
OFAAS.andy-1.25RGR+8..8.r’-O79forAl-mminﬂuoru-
cence. B. Growthoftobaccocellculunesandrelativegrowth
rate(RGR).AWeibull-typemotblwasusedsoﬁtagrowth
eurvetothecelluuedultelativegrowtbrasewucalculaed
bydseWRGR-(lltineﬂduen

Al-mtxinﬂuuescencemetltodandGFAAsmg.2).Up-
takeratewashighestduringtheﬁrstwminfmitialrue
ca 16 umolAl [10‘ cells)" h").afterwhichitdeaaased
(ca4 nmol A1 [10‘ cellsl’| h"). but saturation was not
observable within the experimental period.
OurvaluesonAluptakeintobaccocellsngMe-
semble those observed in bean cell suspensions (Me-

4

61

 

nrnolAl (to'cotrar‘
is a 8

 

 

0&0

 

 

Montreal-rum unit»
5

o883

 

 

 

Fig.2.‘l'lrnc

daysafter

timand

mean .

2110.0 “pli‘dgx'ath of 032 (O). 2.0 (I). and 20 mM

tn rescue

CaC1.(A).respecti y.Dnerminedideraieally.butintheab-
senceofaddchLconu'olvalues forﬂuaescence'mtenslty

2
E:
fr
.3
E
ii

(2hangand'l'hylor1989.1990).‘l'heinitialrateofup-
takeinorntobaccocellﬁafterconvertingornrmitsmd
assruningSﬁdryweiut)wasl.8ugAl(gdryweidrt)"
min".1..anofetal.(1994)estimatedatissueconcenua-
tiarof70nmolAl(gfreshweight)"insoybeanroottips
aher30nrinofincubatimavaluecomparablet064
nmolA1(gfreshweight)" forutdinotn'cellsafter30
minofincubation.

”WWQCR’J-pmn-dpu-
madam

lnpnttoexaminefactorscthnthancellMandin
puttoftlthersu'engthenthegoodcorrelaimtbetween

““911”

Tab.l.Effectsoflow 'onofAltocin'ate.
andpe-washpl-lonAluptakeinculnrredtobwcocellstays sf-
tertransfer C.ellswerepre-washed(pl-l4.5ta'5620mMC
followcdbyAlC13inarbation(20pM,l;npna;15.0.2mMCaClz) fal

h.at23C.unlessind_icatedodrawise.

5 mM citric sc-
id.adjustedtopl-l4.5 wi KOl-l. Allotherexpaimentalcondi'
and solutions was as described (Fig. 1A). Conuol values for ﬂuo-

intens (mean equalto ﬂuorescence units).

rescence ity
identicallybutinthesbsenceofALwaembtrmdfromdretotal
“ ‘1,“ "Bheplications.

 

 

Aluptake-lhincuban'on

 

Al-mtrin GFAAS
(nmol Al
Uptake conditions (arbiuury units) [10‘ cells]")

Pre-wash at pH 4.5

Standardincubation 825:3.1 9.11:1.3

4Cpre-washandincubanon 20.7:55 2.3103

Prechelati ntocitrate ll..7:4l 1.6304
Pro-wash at pH 5.6

Standard incubation 39.0 t 4.4 5.4 t 0.6

 

GFAAS and Al-morin fluorescence. we examined the in-
ﬂuence of Ca”. low temperature. and pre-chelation to
citrate on Al uptake. Aluminium uptake was only re-

duced by about 10% (Fig. 2) when the level of Ca“ in

A

62

the uptake solution was increased from 0.2 to 2.0 mM
Ca". However. the presence of 20 mM Ca 2' in both the
pre-wash and uptake solutions drastically reduced Al up-
take by 90% (Fig. 2). even after 4 h of incubation. Post-
application of 20 mM Ca”. however. failed to aﬁect cel-
lular Al levels (data not shown). It is noteworthy that
Ca" reportedly can ameliorate Al toxicity (Kinraide and
Parker 1987). This rs one of the ﬁrst reports to examine
eﬁects of Ca“ on Al uptake in intact plant cells. Prior to
this, Alva et al. (1986) found that Ca” concenu'ations
varying from 0.5 to 15 mM reduced water-extractable
and acid-enn'actable (0.1 M HNO,) Al in soybean and
subterranean clover roots. While the water-exnactablc
Al can be expected to be associated with the cell wall, it
is. however. not clear how much of the acid-extractable
Al was previously associated with the cell wall or with
the symplasm.

Pro-chelation cfAl to 5 mM citrate also greatly re-
duced Al uptake by 85% (Tab. 1). Upon lowering the
temperattne of the standard pre-wash and uptake solu-
tionsfrom23Cto4C,Aluptakewasdiminishedby
about 75% (Tab. 1). in accord with observations on Triti-
cum aestivum L. roots (Pettersson and Strid 1989. Zrang
and Taylor 1990). The expression of Al toxicity has also
been shown to be lass pronounced at lower temperatures
(Camargo 1983).

 

Fig. 3. Photomicrographs of ﬂutxesoence from morin-labelled tobacco cells (3 days aha transfer)i III the absence of added Al. and

autoﬂuoresce ence fran control cells (not treated with morin).

ment (atretoﬂuoresce

nansmt tted images of the same cells lsbe morin

Experimental manipulation of cells was identical to that tnA
ms(Fig. 1A).)brg with Al absent. A. ﬂuorescence and B

luptake

geseofth samecellswithout morintreat-

lit-tranrmrgh
.Flutrescence' Images of cellsl labesllcgsrthggnn (0.1 mM 10 min); E. Fluorescence and F. light-
ssm wt 5 M sorbitol. showin the ofmorin ﬂo-
rescence in the cell protoplast but undeteéltsble in the cell wall: The scyale 3 presence I:

ebumpmeuuﬁmmdaumiaographsareonthesame

scale. A camera shutter release speed of 0.5 s was utilized in gene-sting all ﬂuorescence' images

nmmnlm

Distribution of morin in cells and cellular localization of Al

As judged by comparison with autoﬂuorescence of con-
trol cells (Fig. 3A. B). the distribution of morin ﬂuores-
cence in cells not exposed to Al was homogeneous
throughout the cytoplasm and nucleus. and morin did
not detectably accumulate in the vacuole (Fig. 3C. D) or
in the cell wall of plasmolyzed cells (Figs. 3E. F). Typi-
cal ﬂuorescence intensities from morin-loaded cells in
the absence of Al were 18 to 26 arbitrary units (e.g. see
Fig. 1A) compared to 4—8 units from autoﬂuorescence
of control cells.

Following exposure to Al and subsequent chelate
wash. brightly ﬂuorescent Al-mon'n zones could be vi-
sualized at the periphery of cells (Figs. 4A, B). in addi-
tion to apparently non complexed morin distributed
throughout the cytoplasm. In cells not treated with Al.
morin ﬂuorescence appeared pale green. as opposed to a
darker (and much more intense) green ﬂuorescence from
these regions where the dye was seemingly complexed
to Al. The proportion of cells showing these discrete
zones (ranging from almost none to about 70—80%) de-
pended on the extent of Al uptake. When present. one or
two ﬂuorescent zones per cell were visible. Often. these
ﬂuorescent zones appeared at the region of contact be-

 

63

tween two adjacent cells (Fig. 4C). as it along the plane
of cellular division; these ﬂuorescent zones were gener—
ally larger (about 20—30 pm) than those located in other
regions (about 2—18 pm) of the cell periphery (Fig. 4A.
B vs C). Visualization of the ﬂuorescent zones became
possible shortly ( 5 min) after A1 incubation. In cells 5
and 6 days after subculture. in which Al uptake was neg-
lible (Fig. 1A). none or very few discrete ﬂuorescent re-
gions were perceptible and the ﬂuorescence pattern of
these cells was the same as that of cells not treated with
A] (Fig. 3C, D).

This discrete localization of Al does not appear to re-
sult from artefacts such as unequal distribution of morin
in cells (above), or from morin being unable to compete
with cellular ligands. At longer AI exposure times (>1
h). some cells (less than 5% of the population) displayed
Al-morin ﬂuorescence throughout the cytoplasm. with
an even brighter region presumably originating from the
nucleus (Fig. 4D). demonstrating that morin can com-
pete with intracellular ligands. Virtually the same pattern
of Al-morin distribution was rapidly generated in almost
all cells when incubated with Al in the presence of 5 mM
acetate. which drastically affects membrane penneabil-
ity at pH 4.5 (Jackson and St. John 1980). Interestingly.
post-application of acetate, following Al incubation. did

  

N. ‘. . ‘2. - n
. ' ,./’ -

Fig. 4. Photomicrpgraphs of ﬂuorescence from morin-labelled tobacco cells (3 days afte- transfer. except D) in the presence of
added AI. Aluminium uptake conditions were as described in Fig. 1A. A and B. Cells incubated with AI (20 M. 30 min) under stan-
dard conditions and labelled With morin (0.1 mM. 10 min). AI-morin ﬂuorescence is localized in adiscrete region of the cell periph-
cry (arrow). Fluorescence from remainderof cell is from non complexed morin. similar to control cells labelled with morin in ab-
scence of applied Al (Figs. 3C. D). C. Similar to A and B. but demonsu’ating Al-mon'n ﬂuorescence located at the region of contact
between two adjacent cells (arrow 1. D. Cells (2 days after u-ansfcr) incubated with Al for 90 min. showing localization of Al
throughout cell. but parti.t:ularly in nucleus (arrow). E. Fluorescence and F. light-transmitted images of the same morin and trypan
blue-labelled cells. showmg the presence of an Al-morin region in a cell (arrow in E) im cant to trypan blue (F). as opposed to
one that is permeable (arrow in F). Scale bars in A and C represent 20 and 25 um. respectively. B and D are on same scale as A. E
and F the same as C. A camera shutter release speed of 0.2.5 5 was utilized in generating all ﬂuorescence images.

6 Physiol. Plant. 97, I996

nor ageueialiaeddisuibudonofAanrdidit
detectablyalterthediseietepatternofAldisuibutiuiat
thecell periphery (data not shown).

Specialmentionismadeoftheobservatimthatthe
smespatialpattemofAldisuibutionocciuredirrespee-
tiveoftheorderofAlandrnorinloading(i.e.ineells
ﬁrstexposedtomorinmdthentoAlorviceversa);evai
whencellswerestainedwithniorininau-aditionalrneth-
anal/acetic acid solution. the same Al distribution in dis-
ereteperiphaalregionswasobtained(dannotshowu).
However. in cells preloaded with morin. the ﬂuores-
ceneeintensitywaslessintense.pesinnahlyuiginling
ﬁcmmorinleakagebyeellsduingpe—washandineu-
bationwithAl.

TodistinguishwbetherthedisueteAl-inorinﬂues-
centiegioriwasassociatedwiththesymplamcellwall.
orboth.plasinolysisofcellsneuedwithAlwasat-
tempted. similarlytothatperforinedinnon-Al-ueated
cells (Figs 3E.P').However.theaecellsalmostalways
plasmolyzed poorly. shrivelling instell. with little disso-
ciadonofthesymplaunn'omthecenwanlntheiare
occasions in which plasmolysis was more favorable.
thesyrnplaststillrernainedasaociatedwiththeeellwall
alongaseetionhubotuingmeeeAl-mainﬂuaeseent
regions.

ltisnotewonhythatthefluaeseeueeofAl-m’uiwas
notquenchedbythepreaenceofMAﬂmMﬁnatleast
aSO—foldexeessoverrna'in(appliedato.lrnMandex-
eesswashedout).ED‘l‘Ahasabindingaﬂinityemstant
for Al (log K = 16.5; Orvig 1993) about 10" times
higherthanthatofniorinaogK1-65;Xatyaland
Prakashl977)andeanbeutiliaedtoqueuehAl-inuin
{lg-gm in analytical applications (Ueno et a1.
1 ).

H.DI'IIC pain-hilly I IowpﬂudAluptaha

Althoughcellsinculuueandthoseineubatedlnuptake
mediumatpHS.6werevirtuallyinipei-ineableu>u'ypui
blue. lowering media pl! to4.5 enhanced dye permeabil-
ity.partieularlyafter4hofineubation('lhh. 2). 'l'histype
of H‘-induwd membrane permeability could be amelio-
medbyseveralfactors. Indeed. cellage(i..e yowth-
stage). coneenn'ationofCa". orpreseueeofAlinthe
medium.andternpentureallimpaetedmernbnneper-
nieabiliryatpll450‘abs2and3).asjudgedbyuypan
bhieuptakeintobaeeocellsAlthoughlessinform
is availdile for cell ailtures. increased membrane per-
meabilityofrootsreportedlyresultedfiurndeaeasedex-
terrnIleYanetal. 1992).pat‘ticularlyasillustrandby
enhanced efflux of K’ from cells (Zsoldos and Erdei
1981. Sasakietal. 1994).
Thus.pH-uiggeredinernbraneperinediiliryniaybea
factorinﬂueneingAluptakeHowevermellperuieahility
touypanblueitselfisapparentlynotarequirenientfu
Aluptake. Fineduallabellingofcellswithinorinand
u'ypanbluerevealedﬂuoreseentAl-rnorinsouesincells
thatwerenOtperuiediletouypanblueﬂsmBJee-

“M.W.!”

1a». 2Wof¢lhadaysmmpmbleo ny-
pnbluewluiinaihatedinrnediumatpl-HSa'pHSLanduﬂ-

head? Ca’enncuiuauonsandthepnaeneeoi‘zoiiMALCells
(2x1 cellsinr‘)waehiaibuadinrnediaidentialineunposi-
umbthatutilizadinAluptakeespawoeelegendl-‘rgJM
mmmaummmm anirdrhl
lOnunhnibaucupammeuednuuwasreplacedbyanequal
vohneotheahnndhundaunemhiean3sehun3
septic-5on3

 

 

 

Mun ﬁCellspaindleuiuypnbhie
lbw 4h“
prise
01ml)” 1830.4 2230.8
pins
Minna" 19.4328 73333.0
2.0inMCa” 5.5313 29.1315
MEMO“ 1.7306 3331.4
Madly-039”“ 1230.2 8.1313

 

md.eellspe-washedatp85.6.ratherthanatpll4.5.
renainediinperineablemuypanbluewitbin l hofAJ
uptakeahb.3).butAluptakewasstillhalfofthatob-
tainedunderstandarduptakeconditionst'l‘ab. l).'l'hird.
thepercentageofcellsthatshoweddisereteregionswith
Al-niorinﬂuoreaceucewas higherthanthepercentage
permeableuiu'ypanhlue.'l'hatis.wefoundthatabout
mwMofMay-oldeelhdemonsuatedAl-minau-
mtregimwhileaboutMoftheeellswerealso
permeabletouypanblueaab.3)afteral-hAlineuba-
tionperiod. Lastly.thediseretecellularlocalinriouof
Atasmonimiedbymoiin.peisistedregaidhssofeellu-

-hrperniediilitytouypanblue.

M

Ou'resultssupportthenotiouthatrapidAluptakeiseiit-
icaﬂydepadentmceﬂMInmwingtobaccoeells
aeonsiderableunarntofAlisapparentlylocaliaed.at
leastiiiitially.iriadiscieteregionofthecell.podhly
comprisingpeiipheraleytosolandloraplasinarneinbrme
aeeum.’ oralteinauvely' .inthedyouung‘” cellwallregion.

1&3. MdalhpumlenuypnhheamrlhAlh-
mam» deellsand d
nedthellswerepe—mwashed pH M
bwdbymwhhawhhanmmuumﬂ4ﬁ. 02ml!
cmfclhltllother conditions andsohiuoi-
mnmwmlALMen3sam3replicanoas.

 

 

 

§Cellsperrnnhletotmlnblue
Dania-tam +Al -Al
mill-l4:
2m 3532.1 46.0345
Bay 14.9322 28333.8
Shy 23305 1.7305
mum
3m 2.830.? 15.3333

 

Some or possibly most of the Al within this discrete
region was apparently sequestered into at least one com-
partment which was not directly accessible to the exter-
nal solution. i.e. suggesting that Al was indeed internal-
iaed by the cell. In view of uncertainties as to whether
chelate washes are capable of removing all Al from the
cell wall (Zhang and Taylor. 1990). supposedly because
of high affinity binding sites within the wall. evidence
for Al internalization was derived from several aspects
of our studies. Whereas a considerable portion of the q)-
plied Al was still associated with the cell after several ci~
trate washes following Al uptake. prechelatiorr of Al to
citrate reduced Al uptake by about 85% (Tab. 1). This
large difference is unexpected ifAl is binding to a site in
the cell wall. Similarly. CaCl, (20 mM) was ineffectual
in reversing a preceding Al accumulation by the cell. but
application of these levels of CaCl, during incubation
with Al greatly reduced Al uptake (Fig. 2). Furthermore.
even if it is assumed that the observed ﬂuorescence orig-
inaes from Al tightly bound in the cell wall. application
of EDTA. while possibly not removing the Al. should
still be capable of displacing morin and quenching fluo-
rescence. because of its much higher binding affinity.
However. ﬂuorescence of Al-morin was not quenched
by externally applied high concentrations of EDTA. ln-
ternalization of Al is further suggested by the observa-
tion that morin accumulated in the cytoplasm (Figs 3C-
F) and that membrane permeability appears to play a
role in Al uptake.

Apart from the possibility that Al may be localized in
a discrete region of the cell wall. which in turn com-
prises a mosaic of wall domains with different composi-
tion and functions (Roberts 1990). our studies indicated
the absence of detectable amounts of Al in the remaining
larger portion of the cell wall. These ﬁndings were in
part further documented by the virtually negligible up-
take of AI by older cells. In this context it is worth men-
tioning that Al was not associated with pectin uronic ac-
idsashasbeenshowninthreedifferentpearootsections
treated with 1 mM AlCl,. pH 4.5 (Matsumoto et al.
1977).

Taken together. our evidence for Al internalization
supports the emerging notion that Al mainly accumu-
lates in the symplasm rather than the apoplasm (Kochian
1995). In animal cells. which lack a cell wall. it was
demonstrated that Al internalization is required to inhibit
formation of inositol phosphates. key products of phos-
phoinositide-related signal transduction (Shi et al. 1993).
It is a reasonable assumption that Al-triggered charges
in these types of second messengers are early indicators
of Al toxicity in cells.

The plasma membrane of growing cells. possibly a
discrete region. therefore plays a role in Al uptake.
Given the pronormced dependence of Al uptake on cell
growth and its absence in stationary-phase cells. we can
only speculate a to the underlying mechanisms respon-
sible for the results assented here. A certain region of
theplasmamembranethatpmticipatesingrowthpro—

65

cesses. such as membrane turnover and cell wall synthe-
sis. might be less effective as a permeability barrier and
could serve as an entry port for Al uptake. This nright
explain the discrete localization of a major amount of Al
at the cell periphery. In addition. the physical state of
this membrane region could be modulated by the exter-
nal low pH (Lindblom and Rilfors l992). which in turn
might favour Al-tn'ggered membrane phase changes (Vi-
erstra and Haug 1978) and facilitate Al entry. as was
demonsu'ated on phosphatidylserine-containing lipid
vesicles (Shi and Haug 1988). pl-l-related changes in the
plasma membrane of some tobacco cells were indicated
by enhanced permeability to trypan blue (Tabs 2 and 3).
especially upon longer exposure to low pH. These u'y-
pan blue-leaky cells were probably alive because cell
death occurred at an external pH < 2 according to "P nu-
clear magnetic resonance studies on suspension cultures
(Fox and Ratcliﬁe 1990). and cellular ATP levels were
most likely maintained under our atlture conditions. Un-
der high external Ca” concentrations (20 mM. Fig. 2). or
at low temper-mire (4'C. Tab. 1). greatly diminished Al
uptakemightresultfromalterationsofthediscretere-
gioninacellsubjecttothistypeoftreatment.

OurﬁndingsthatAluptakeiscriticallydependeuton
cellulm'growthareinaccm'dwith wholeplantobserva-
nonsManifestationsofAltoxicityinrootsoccurin
meristematiccells.elongatingcellsandroothairs(Mat-
sumoto et al. 1976. Ryan et al. 1993). which are indeed
theactivelygiowingcellsofaroot.

Tothebestofourknowledge. this isthefirstreport
where morin in a solely aqueous medium has been ap-
plied to cells. i.e. free of commonly used methanol and
acetic acid. Since cellular integrity is better maintained
upon application of aqueous solutions. a methanol/ace-
tare-free procedure was preferable because Al redistribu-
tion is less probable. Additionally. extracellular morin
was reduced. allowing for cell-derived ﬂuorescence
measurements. Given the simplicity and resolution of
light microscopy (about 0.2 pm). and the ease of speci-
men preparation. the morin ﬂuorescence method is
thoughttobesuperiortothesecondaryion mass spec-
u'omeu'y technique recently applied to ascertain Al dep-
osition in plant cells (Lazof et al. 1994). Thus. a morin-
based technique should prove advantageous in Al uptake
studies. particularly when many data points are desired.
or when Al incorporation by genetically different culti-
vars is evalutated.

AW-VA.V.wmsupputedinpartbyafellow-
shiny-rt! 2013921916. from theConselho Nacional deDe~
senvolvimento Cientffrco e Tecuoldgico (CNPq). Brasil. This
work was also partially funded by the nt of Botany
and PM! Pamology at MSU. We are grateful to Drs R.
Hausinger. M. Zahik. A. lures. and 1.. Bauer for use of the
GFAAS. thior'raneter. cpl-fluorescent miaoscope. and water
shaker. respectively. We also thank R. Pacovsky for useful dis-
cussions and D. Shiatsu and M. Sullivan fr: sumlyiug us with
use cell culture.

met-391.1006

66

References

Alva. A. K.. Asher. C. J. 8: Edwards. D. G. 1986. The role of
calcium in alleviating aluminium mxicity. - Aust. J. Apic.
Res. 37:375-382.

Bradford. M. 1976. A rapid and sensitive method in the quarti-
tation of microgram quantities of protein utilizing the princi-
ples of protein-dye binding. - Anal. Biochem. 72: 248-254.

Browne. B. A., McColl. J. G. 8: Driscoll. C. T. 1990. Alumino
ium speciation using morin: l. Morin and its complexes with
aluminium. - J. Environ Qual. 19: 65-72.

Carnargo. C. E. de O.1983.E.feito da unmet-sun da soluclo
nutritiva na tolerlncia so aluminio dc cultivates dc trigo. —
Bragantia 42: 51-63.

Delhaize. E. 8:. Ryan. P. R. 1995. Alumirnim toxicity lid tolero
ancc in plains. - Plant Physiol. 107 : 315—321.

Fox. G. G. at Rstcliffe. R. G. 1990. ”P-NMR obsu'vation on the
effects of the external pH on use inuacellular pH values in
plant cell suspension culnires. - Plain Physiol. 93: 512-321.

Foy. C. D.. Chaney. R. l... a White. M. C. 1978. The physiology
of metal toxicity in plants. - Annu. Rev. Plant Physiol. 29:
511-566.

Ganrot. P. O. 1986. Metabolism and possible health effects of
alumiiaim. - Envirar. Health Perspect. 65: 363-441. .

Haug. A. 1984. Molecular aspects of alum'utum toxicity. - Crit.
Rev. Plant Sci. 1: 345-373.

Jackson. P. C. at St. John. J. B. 1980. Charges in memu'ane lip-
idsofrootsassociatcd withchanesirrpermesbil'uy.-Plant
Physiol. 66: 801-804.

Katyal. M. at Prakash. S. 1977. Analytical raciicnsofhymy-

.. ﬂavones. - Talarsa 24: 367-375.

Kinraide, T. B. 1991.1dcntity ofthe thiamnxic alt-aim spe-
cies. - Plant Soil 134: 167-178.

- &Parker.D.R.1987.Cationainelicatirsrdahmiimimeox-
icity in wheat - Pill! Physiol. 83: 546-551.

Kochian. L. V. 1995. Cellular meclnnisms of aluminum msicity
and resistance in plants. - Annu. Rev. Plait Physiol. Plant
M01. Biol. 46: 237-60.

Lazof. D. B.. Goldsmith. J. G.. Rufty. T. W. & Liam R. 1994.
Rapid upmke of shimiraim into cells of imact soybem root
tips. A micromalytical snidy using in mass
spectrometry. - Plant Physiol. 106: 1107-1114.

Lindblom. G. at Rilfors. 1... 1992 Cubic lid hexagonal lipid
phases -lnStructureandDynamic PropertieeofLipidsand
Membranes (P. J. Quinn and R. J. Cherry. eds). FP- 51-76.
Portland Press. Lonbn. ISBN 1-855-78u014-3.

Matsurnoto. l-l.. Hirasawa. E.. Torikai. H. dt Takahashi. E. 1976.
Localization ofabscrbed aluminiuminpearootanditsln'nd-
ing to nucleic mids. - Plant Cell Physiol. 17: 127-137.

- . Mortmura S. at Takahashi. E. 1977. Less involvernern of
pectin in the precipitation of aluminium in pea root. - Plant
Cell Physiol. 18: 325-335.

McDonald-Stephens. J. 1... Jr Taylor. G. J. 1995. Kinetics of alu-
minum uptake by cell suspensions of Phased» vulgaris l...
— J. Plant Physiol. 145: 327—334.

Nagatlal. gheNemoto. II. at Hasmws. S. 1992 ‘lbbaeco BY-2
cc ' asthe“e1.a"cellintleccllbrology' ofhi
plats-1m. Rev. Cytol. 132: 1-30. 8""

Why?!

“The”. 1996

Orvig, C. 1993. The aqueous coadination chemistry of alumi-
num. - In Coordination Chemistry of Alumiiaim (G. 11.
Robinson. ed). pp. 85-121. VCl-l Publishers. New York
NY. lsBN 1-56081-059-9.

Peuersson. S. a Su-id. 1-1. 1989. Initial uptake of aluminum in re-
lxion to temperature and phosphorus status of wheat (Triu-
eiim antiwar 1..) roots. - J. Plant Physiol. 134: 672-677.

Robats. K. 1990. Structures at the plant surface. - Curr. Opirt
Cell Biol. 2: 920-928.

Ryan. P. R.. Ditomaso. J. M. a Kochian. L. V. 1993. Aluminium
toxicity in roots: an investigation of spatial sensitivity and
the role of the root cap. - J. Exp. Bot 44:437-446.

Sasaki. M., Kasai. M.. Yamamoto, Y. 6r Matsumoto, H. 1994.

'stsr of the ecly response to aluminum sums be-
tween tolerant and sensitive wheat cultivars: root growth.
alumirmm cement and eﬂ'lux of K‘. - J. Pill! Nun. 17:
1275-1288.

Shi.B.&Heug. A. 1988.Uptakeofaluminumby lipidvesicles.
-Tox. Environ. Chan. 17: 337-349.

-&1-lnig.A. 1990.A1uminumuptahbyneuroblastoun

.-J. Neurochem. 55: 551-558.

- Chou. K & Haug. A. 1993. Aluminium impacts elements
of the phosphoinositide sipslling pathway in neuroblas-
tans cells - Mol. Cell. Biochem. 121:109-118.

Tice. K. R.. Parka. D. R. a DeMason. D. A. 1992. Opermion-
ally defined apoplastic and symplastic aluminum fractions
in root tips of aluminum-intoxicmed wheat. - Plant Physiol.
1(1): 309-318.

Ueuo. L1mamura.T.&ChengK.L. 1992.1-llrdbookof0r-
ganicAnalyticalReagents.-CRCPiess.BocaRatou.FL.pp.
2.5-29. 158. ISBN 0-8493-4287-2

Vita-sue R. a Bang. A. 1978. The effect ofAl"on the physical
matte} ofmanbranelip'dsinmmplom acrdoplrr' '

.-Biochem. Bligpbys. Res. Comm. 84: 138-143.

Wagmuma. T. 1983. act of non-metabolic conditions on do
uptakcofaluminumbyplantroots.-Soil Sci. Plant Nutr.
29: 153-333.

Yamamom. Y.. Rikiishi. 8.. Grang. Y.. Ono. K.. Kasai. M. a
Matsuinom. 1-1. 1994. Qumtitative estimation of aluminium
toxicity in culuired mbacco cells: correlaion between alu-
minium upmke mid growth inhibition. - Plant Cell Physiol.
35: $75—$83.

VII. F.. Schubert. S. & Mengel. K. 1992. Effect of low root me-
d'annplionnctprotonreleasc.roarespiration.androot

ofcorniZeasinysLMndbroadbethciafabo

). - Plain Physiol. 9% 415-421.

Zhang. G. & Taylor. G. J. 1989. Kinetics of aluminum uptake by
excised roots of aluminum tolaant and aluminrnn sensitive
cgigvms of Tritieisn oern‘vum 1.... - Plant Physiol. 91: 1094-

- R Taylor. G. J. 19%. Kinetics of aluminum uptake in Thri-
cisn oestr'vum 1.... ldemity of the linear phase of aluminum
spake by excised roots of aluminum-tolerant and alumi-
niirnoaensitive cultivars. - Plant Physiol. 94: $77-$84.

Zsoldos, F. & Erdei. L. 1981. Membrane and ion transport prop-
ertiesincereals underacidicmrdalkalinestrees.1.£ffectsd

onpotassitmtuptakeandgrowthofriceandwhea.-
$291.35:sz

E

70.

CHAPTER 4

CESSATION OF ALUMINUM UPTAKE AND ALUMINUM-INDUCED
ABNORMALITIES OF THE CELL SURFACE DEPENDS ON BREFELDIN A-
SENSITIVE MEMBRANE TRAFFIC IN TOBACCO BY-Z CELLS

ABSTRACT

Continuing previous work [Vitorello and Haug, 1996], the relationship between cell
growth and Al uptake in tobacco BY-Z cells was further examined. As observed by confocal
laser scanning microscopy of morin-labelled cells, new aspects of Al-accumulating regions
emerged. Al-morin ﬂuorescence was frequently associated with a protuberance of the cell
surface and with vesicle-like or globular structures attached to the exterior surface of the cell
and presumably produced in response to Al treatment. Similar structures were also frequently
observed along the plane of contact between adjacent cells. These events occurred only in
growing cells, consistent with the hypothesis that Al accumulates in regions of the cell where
new cell wall and plasma membrane are being incorporated. Treatment of cells with cremart
(IO-30 uM), brefeldin A (10 uM), caffeine (1-5 mM) and 2,6-dichlorobenzonitrile (20 uM),
inhibitors of microtubule polymerization, Golgi-mediated secretion, cell plate formation and
cellulose synthesis, respectively, had no discernible effect on Al uptake. Brefeldin A also had
no effect on Al-induced release of material by the growing cells. However, cells almost
completely lost there capacity for Al uptake when incubated in a minimal medium, at pH 5.6,
for 1 to 3 h prior to exposure to Al. Incubation of cells in minimal medium reduced
subsequent growth in culture medium. Brefeldin A (10 11M) suppressed loss of Al uptake
capacity, upon incubation in minimal medium, in a dose-dependent and reversible manner.

Once cells lost their capacity for Al uptake, brefeldin A did not restore it.

67

INTRODUCTION

An understanding of Al uptake processes is central to the elucidation of mechanisms
of Al toxicity. Al toxicity was shown to be correlated with long-term uptake in tobacco cell
cultures (Yamamoto et al., 1994). In mammalian cells, Al internalization was required for
inhibition of phosphoinositide-mediated signal transduction (Shi et al., 1993).

Short-term aluminum uptake has recently been shown to be dependent on cell growth
in Nicotiana tabacum L. cv BY-2 cells (V itorello and Haug, 1996), consistent with the fact
that Al is detrimental to and predominantly accumulates in growing cells of the root apex
(Ryan et al., 1993; Rincon and Gonzales, 1992). Al accumulated predominantly in a discrete
region at the cell periphery, oﬁen between adjacent cells, as if along the plane of cell
division. It is hypothesized that these regions might coincide with regions of growth in cell
surface and volume (V itorello and Haug, 1996). Access of Al to target sites may be
facilitated in such regions.

Recently, it was shown that Al is detrimental (within 30 min) to tip growth in root
hairs, demonstrating that Al can rapidly inhibit cell elongation in non-dividing cells (Jones
et al., 1995). This provided additional evidence that inhibition of cell elongation, rather than
cell division, is the primary cause of rapid, Al-induced root growth inhibition (Kochian,
1995). The tips of these root hairs were swollen and deformed and aﬁer prolonged exposure
to Al, sometimes burst (Jones et al., 1995). Moreover, A1 (0.3 mM) has been found to cause

rapid changes in the ultrastructure of the Golgi apparatus and to reduce secretory activity of

68

69

peripheral root cap cells (Bennet and Breen, 1991). In addition, Al reportedly inhibited
utilization of sucrose for cell wall synthesis (Huck, 1972) and induced callose production
(Schaeffer and Walton, 1990; Schreiner et al., 1994). Taken together, circumstantial evidence
supports the notion that Al may accumulate in and disturb regions of the cell that are
involved in processes of cell growth and/or secretion, although it is not clear how this occurs.

In this chapter, the relationship between cell growth and Al uptake and accumulation
is further examined. By using confocal laser scanning microscopy and low morin
concentrations for the detection of Al, improved images are generated showing Al-induced
abnormalities of the cell surface of growing cells. Subsequent studies, using inhibitors of
processes involved in cell growth, led to the discovery that a functional, brefeldin A-sensitive
secretory pathway is required for cells to lose their capacity for Al uptake, upon prior

incubation in a minimal medium and concomitant reduction in growth.

MATERIALS AND METHODS

Chemicals and labware

Brefeldin A, morin and Murashige and Skoog basal salts were from Sigma (St. Louis,
MO) and caffeine was from Aldrich (Milwaukee, WI). Cremait (butamiphos) was a giﬁ from
Dr. K. Saito of Sumitomo Chemical Co. (Hyogo, Japan), and 2,6-dichlorobenzonitn'le (DCB)
was a reference standard from the Environmental Protection Agency (Research Triangle
Park, NC). All other chemical reagents were of analytical grade. Acid-washed glassware
or, preferably, plastic labware was employed. The water used in all experiments was

sequentially puriﬁed by reverse osmosis, deionization and distillation.

Cell cultures and Al uptake procedure

Cells of tobacco (Nicotiana tabacum L.) cv BY-2 were grown as described previously
(V itorello and Haug, 1996). Growing cells, at 2 or 3 days after subculture, were used in all
experiments. Procedures used for Al uptake were also as described previously (V itorello and
Haug, 1996) with the exception of Al uptake prior to plasmolysis experiments. Al uptake was
performed in a minimal medium composed of 10 mM Mes, 5 mM sucrose, 5 mM KCl, 2
mM CaClz, pH 4.5 (or pH 5.6).

In some experiments, cells were plasmolyzed with 0.5 M sorbitol, following Al
treatment, to allow separation of the plasma membrane from the cell wall and thus

distinguish whether Al-morin ﬂuorescence was in the cell wall or in the protoplast. However,

70

71

following incubation with A1 at low pH, cells plasmolyzed poorly, tending to shrivel, with
incomplete separation of the plasma membrane from the cell wall. Attempting to improve
this, cells were also brieﬂy pre-washed at pH 5.6 and incubated with Al for shorter periods

of time (<30 min), also in minimal medium at pH 5.6.

Treatment of cells with inhibitors of growth-associated processes

Cremart, brefeldin A, caffeine and 2,6-dichlorobenzonitn'le are inhibitors of
microtubule polymerization (Doonan et al., 1988), the Golgi secretory pathway (Driouich et
al., 1993), cell plate formation (Hepler and Bonsignore, 1990), and cellulose synthesis
(Shedletzky et al., 1992), respectively. Stock solutions of these inhibitors of growth-
associated processes were prepared as follows. A 10 mM stock solution of brefeldin A (BF A;
MW= 280.4) was prepared in methanol (Yasuhara et al., 1995) and stored refrigerated with
desiccant. A stock solution of cremart [O-ethyl-O-(3-methyl-6-nitrophenyl) N-sec-
butylphosphorothioarnidate; MW = 332] was prepared by adding 10 til of cremart per ml of
DMSO (ca. 30 mM). Caffeine (MW=194.2) was prepared as an aqueous 0.1 M stock
solution, and 2,6-dichlorobenzonitrile (DCB; MW=172) was prepared as a 20 mM stock
solution in DMSO.

To test the effects of these inhibitors on Al uptake, these compounds were
individually added to 10 ml of cell culture which had been transferred to 50 ml Erlenmeyer
ﬂasks, and were incubated for 30-60 min on a gyratory shaker 160-180 rpm, at 28°C.
Following this treatment, Al uptake (1 h, pH 4.5) was performed as described above. The

ﬁnal concentrations of these compounds used were 15 and 30 1.1M cremart, 1 and 5 mM

caffeine, 10 iiM BFA and 20 iiM DCB.

In the experiments regarding the effects of BFA on the cessation of Al uptake, cells
were pre-incubated with BF A prior to other experimental treatments by transferring a volume
of suspension culture (generally 10 to 20 ml) to 50 ml Brlemneyer ﬂasks, followed by
incubation with BFA on a gyratory shaker at 160-180 rpm, 28 °C. Concentrations of BFA
employed ranged from 1 to 50 uM, but was generally 10 uM. Unless otherwise stated, pre-
incubation with BFA was for 30 min. Control cells were treated identically, with an

equivalent amount of methanol, but no BF A.

Labelling of cells with morin and measurements of ﬂuorescence intensity

For the detection of Al, cells were labelled with the Al-binding ﬂuorescent dye morin
(100 uM, 10 min) and measurements of ﬂuorescence intensity were made with a LS-S
luminescence spectrometer (Perkin-Elmer, Norwalk, CT) coupled to a microcomputer, as
described previously (V itorello and Haug, 1996). However, for observations under the
microscope, cells were treated with lower concentrations of morin (5 or 10 uM) to reduce

glare and saturation of Al-ﬂuorescent regions, thus improving the quality of the images.

73

Microscopy

Aluminum and morin-treated cells were observed with a Zeiss 210 laser scanning
confocal microscope (Carl Zeiss, Inc., Thomwood, NY). Images were collected in laser
scanning transmitted brightﬁeld and confocal ﬂuorescence modes using the blue 488 nm line
from a 10 mW dual-line argon ion laser. For confocal ﬂuorescence, a long pass 520 nm
barrier ﬁlter was used. A 40 x Plan-Neoﬂuor oil immersion (numerical aperture of 1.3)
objective was used. For black and white photography, Kodak Tmax 100 ﬁlm (Eastman
Kodak Co., Rochester, NY, USA) was used on a Matrix Multicolor computerized camera
unit (Agfa-Matrix, Orangeburg, NY), directly coupled to the Zeiss microscope.

Conventional, epiﬂuorescence microscopy was also employed to examine cells. A
band pass 400-440 nm and a long pass 470 nm ﬁlter set were used with a Zeiss WL standard
research microscope equipped with an epiﬂuorescence condenser IV FL (Carl Zeiss,

Oberkochen, Germany).

RESULTS

Confocal laser scanning microscopy of Al-accumulating regions in cells

Improved images of Al accumulating regions were obtained (Figs. 1-3), compared
with our previous report (Vitorello and Haug, 1996), by using confocal microscopy and
lower concentrations of morin for detection of Al. This resulted in reduced glare and
saturation of ﬂuorescence. As reported previously, Al accumulated in a discrete region at the
cell periphery, frequently along the boundary between two cells, as if along the plane of cell
division. A bulge or protuberance of the cell surface was often observed at sites of Al-morin
ﬂuorescence (Fig. 1A). Adjacent to sites of Al accumulation, but exterior to the cell, Al-
morin ﬂuorescence was also frequently observed in what appeared to be large vesicle-like
or globular structures (Fig. 1B, C). These structures were attached to the cell, as evidenced
when cells rotated and moved freely in a newly placed drop of cell suspension on a
microscope slide. Al-morin ﬂuorescence was not uniform along the plane of contact between
adjacent cells, where Al frequently accumulated, but rather was associated with distinct
structures (Fig. 2), somewhat similar to the vesicle and globular-like structures described
above (Fig. 1). Another new observation was that Al-morin ﬂuorescence occurred at the
margins of the plane of contact between cells, as judged by focussing the confocal images
up and down the specimen.

These events could be observed after short-term exposure of the cells to Al. However,
these structures were more pronounced following longer periods of exposure to Al (60 min)

than shorter periods (15 min). These structures also became more pronounced with time,

74

75

Figure l. Photomicrographs of Al accumulation in deformities of the cell surface which
originated aﬁer A] treatment. Cells (2 days after subculture) were pre-washed (pH 4.5) then
incubated with 10 or 20 uM AlCl3 (pH 4.5) for 15 or 30 min. Uptake was halted by addition
of 5 mM citrate. For detection of Al, cells were incubated with 5 pM morin for 10 min (pH
5.6), then resuspended in fresh medium (pH 5.6) containing 5 mM EDTA. Images were
collected with a confocal laser scanning microscope. Images of each specimen (A-C) were
collected in both transmitted laser scanning brightﬁeld and confocal ﬂuorescence modes. The
left half side of each image is a transmitted laser scanning brightﬁeld image showing a
general view of the cell. The right half side is a confocal ﬂuorescence image of the same cell
showing only Al-morin ﬂuorescence. A) Protuberance of the cell surface (arrow in left),
containing A1 (image on right), which formed in response to Al treatment. The ﬂuorescence
from the remainder of the cell is weak and does not show up under the conditions used for
observation. B) A globular structure attached to the cell surface (arrow in B), also containing
A1 (image on right). C) F orrnation of a vesicle-like structure (arrow in C) at the cell surface

and associated Al-morin ﬂuorescence (image on right).

76

 

Figure 1.

77

Figure 2. Photomicrographs of Al accumulation in structures at the region of contact between
two adjacent cells. Cells (2 days after subculture) were treated with Al and morin as
described in Fig. 1. From top to bottom, images are confocal ﬂuorescence, transmitted laser
scanning brightﬁeld and an overlay of both modes. Images were collected with a confocal
laser scanning microscope. A, B and C) A view of Al accumulation between two cells. D,
E, and F) A higher magniﬁcation View of Al accumulation between two cells, showing more

detailed structure.

78

 

Figure 2.

79

Figure 3. Photomicrographs of plasmolyzed, Al-treated cells. Cells were treated with Al (10
uM, 15 min) as described in Fig. 1. Aﬁer labelling with morin, cells were plasmolyzed with
0.25 M sorbitol, followed 5-10 min later with a ﬁnal concentration of 0.5 M sorbitol. Images
are overlays of confocal ﬂuorescence images over transmitted laser scanning brightﬁeld
images. A and C) Images illustrating tendency for the cell wall and plasma membrane to
remain attached in the region containing A]. B) Plasmolyzed cell with Al-morin ﬂuorescent
region associated with the protoplast rather than with the cell wall. However, separation of

the protoplast and cell wall was rarely achieved in the region containing Al.

80

 

Figure 3.

81

after treatment with Al had terminated. None of these events were observed to occur in
growing control cells treated identically but without Al, or in non-growing cells (stationary-
phase cells) treated with Al. These observations are similar to Al-induced deformities in root
hair tips observed by Jones et al. (1995), although longer periods of exposure to Al were
required in root hairs. After 2 to 3 hours of exposure to Al (20 uM), the root hair tips of a
pondweed, Limnobium stoloniferum, were deformed and swollen and, after prolonged
exposure, were sometimes observed to burst.

Al-treated cells were plasmolyzed to visually distinguish Al possibly in the cell wall
and in the protoplast. This distinction has been difﬁcult, because of shrivelling and
incomplete separation of the plasma membrane from the cell wall. Modest improvements
were obtained when Al exposure was performed in a medium of pH 5.6 and was decreased
to 15 or 30 min. From repeated plasmolysis experiments, at pH 4.5 or 5.6, it became apparent
that there was a tendency for the protoplast to remain associated with the cell wall at these
regions of Al-morin ﬂuorescence (Fig. 3A, B). Occasionally, however, it was possible to
observe apparent dissociation of the plasma membrane from the cell wall, with Al-morin
ﬂuorescence associated with the protoplast (Fig . BB). Plasmolysis of cells prior to Al
treatment could be accomplished successfully. However, Al uptake was reduced by 76 to
84% when cells were treated with 0.1 or 0.3 M sorbitol prior to and during exposure to Al,
as determined by both Al-morin ﬂuorescence and graphite furnace atomic absorption
spectrometry. Al-accumulating regions could not be observed in cells plasmolyzed prior to

Al treatment.

82

Cells lose their capacity for A1 uptake when incubated in a minimal medium

Al uptake in the tobacco cells is dependent on growth and is highest in cells 1 to 3
days after subculture (V itorello and Haug, 1996). When 2-or 3-day cells were removed from
their culture medium, and ﬁrst incubated in a minimal medium at pH 5.6 prior to exposure
to Al, subsequent Al uptake (1 h, pH 4.5) decreased signiﬁcantly (Fig. 4). After 30 min of
incubation, Al uptake was reduced by 30-50%, and most of the Al uptake capacity of these
cells was abolished after 1 or 2 h. Al-morin ﬂuorescent regions could not be observed, under
the microscope, in cells which had undergone this loss of Al uptake capacity and, in this
respect, resembled stationary-phase cells.

Incubation of cells in minimal medium for a deﬁned period of time (1 or 3 h),
followed by fresh culture medium, signiﬁcantly reduced cell growth when compared to
uninterrupted culture or to direct transfer into fresh culture medium (Tab. 1). Thus, the
cessation of Al uptake, induced by incubation in minimal medium, was presumably caused

by reduced or halted cell growth.

Effect of inhibitors of growth-associated processes on Al uptake

Cells were incubated for one hour, prior to treatment with A1, with cremart (15 and
30 pM), BFA (10 uM), 2,6—dichlorobenzonitrile (ZOuM) and caffeine (1 and 5 mM), which
are inhibitors of microtubule polymerization, Golgi-mediated secretion, cell wall synthesis
and cell plate formation, respectively (Doonan et al., 1988; Driouich et al., 1993; Hepler and

Bonsignore, 1990; Shedletzky et al., 1992). These processes are intimately involved in cell

83

Figure 4. Aluminum uptake ceases in tobacco BY-2 cells after incubation in a minimal
medium, but not in cells treated with brefeldin A. Cells at 2 and 3 days after subculture (A
and B, respectively) were used. Prior to any other treatment, cells were ﬁrst pre-incubated
with O or 10 uM brefeldin A for 30 min, then removed from their culture medium and
incubated (4 x 105 cells/ml) in a minimal medium (pH 5.6) also containing 0 or 10 uM
brefeldin A, for 0.5 to 3 h. Cells (2 or 3 days after subculture) were pre-washed (pH 4.5) then
incubated with 20 uM AlCl3 (pH 4.5) for l h. Uptake was halted by addition of 5 mM citrate.
For detection of Al, cells were incubated with 100 iiM morin for 10 min (pH 5.6), then
resuspended in ﬁ'esh medium (pH 5.6) containing 5 mM EDTA. Control cells were
manipulated identically but were maintained in culture medium prior to Al uptake. Bars

represent SE (n=3).

AI uptake (1 hr)
fluorescence (arbitrary units)

Al uptake (1 hr)
fluorescence (arbitrary units)

Figure 4.

.3 .3
N uh
O

_a.

O

O
l

O

84

 

' J
r

A 2 day cells

.. ________ —o
culture medium (control)

 

j

minimal medium

 

 

6° ‘ (+ 10 “M BFA)
40 -*
20 a minimal medium
(- BFA)
0
B 3 day cells

minimal medium

30 ‘ (+ 10 "M BFA:E

O)

.5
O
l

N
O
I

O

  
  
  
  

culture medium —.

(control)

 

 

 

minimal medium
(- BFA)
ﬁ.
1 I I
0 1 2 3

duration of incubation (hours)

 

85

Table 1. Growth of tobacco BY-2 cells following transfer to fresh culture medium,
incubation in minimal medium, or culture in the presence of brefeldin A (BF A). Growth was
measured 24 h after the beginning of each treatment, as the increment in cell titer (106
cells/ml) of the suspension. Cells were utilized 2 days after subculture. a) Cells were left
undisturbed in the culture medium; b) Cells were transferred to a fresh culture medium; c)
and (1) Cells were incubated for 1 and 3 h, respectively, in a minimal medium, followed by
incubation in fresh culture medium; e) and f) Cells in culture medium were treated with 10
uM brefeldin A (BF A) continuously, or for 1 h, followed by incubation in fresh culture

medium, respectively.

 

 

Growth of cells -
increment in cell titer
Treatment (10° cells/ml)
a) continuous culture 1.05 i 0.05
b) fresh medium 0.69 i 0.04
c) l h minimal medium 0.40 i 0.06
d) 3 h minimal medium 0.33 i 0.04

e) 10 uM BF A - continuous culture 0.03 d: 0.03
f) 1 h 10 uM BFA ' 0.56 :t 0.04

 

86

growth. However, treatment of growing cells in culture medium with these inhibitors prior
to exposure to Al had no signiﬁcant effect on Al uptake (Tab. 2).

In the speciﬁc case of BFA, an inhibitor of the Golgi apparatus and associated
secretory pathway (Klausner et al., 1992; Driouich et al., 1993), the effects of this compound
on Al-induced production of extracellular material, as described above (Fig. 1 and 2), was
also tested. As judged by observations with the microscope, treatment of cells with 20 uM
BF A bad no effect on Al-induced formation of extracellular material. Cells were treated with
BF A one hour before application of Al, but BFA was also maintained during Al and morin
incubations, as a precaution, because the effects of BF A are known to be highly reversible

(Driouich et al., 1993).

Brefeldin A sustains Al uptake capacity by cells

Although application of BF A to continually growing cells did not cause any
signiﬁcant changes in Al uptake, it suppressed the loss of cell Al uptake capacity induced by
incubation in minimal medium (Fig. 4). This suppression was complete in 3 day cells (Fig.
4B), but not in 2 day cells (Fig. 4A). Simply co-incubating cells with BFA and minimal
medium was sufﬁcient to inhibit most of the reduction in Al uptake capacity, but the effects
of BFA were more pronounced if growing cells, in culture medium, were pre-incubated with
BF A for 15-30 min prior to exposure to minimal medium (Fig. 5). In preliminary
experiments, cremart (30 11M) or caffeine (5 mM) did not suppress this dr0p in Al uptake
capacity (data not shown).

Tests were performed to verify whether these were pleiotropic effects of BFA. A

87

Table 2. The effect of cremart, dichlorobenzonitrile GJCB), caffeine and brefeldin A (BF A)
on Al uptake in tobacco BY-2 cells. Growing cells (2 and 3 days) in culture medium were
incubated with these inhibitors for 1 h prior to Al uptake, with the exception of BF A which
was applied 0.5 hr before. Al uptake (1 h) was performed as described in Fig. 4. Al uptake
was determined by Al-morin ﬂuorescence. Results were pooled (n=3), and are expressed as

percentage of Al uptake in control cells not treated with these inhibitors.

 

 

Al uptake
treatment percentage of control (%)
(mean :1: SE)
Cremart (15 and 30 uM) 100.2 i 11.3
DCB (20 uM) 98.2 :t 2.9
Caffeine (1 and 5 mM) 118.4 :t 13.8

BFA(10 11M) 104.8 s 5.41

 

88

 

 

 

 

 

 

120
Al uptake
_ Zcontrol
100 - A
E ZOpMBFA
S a
1:: 3‘ 8° ‘
i g SpMBFA
27%
83 60
g o
< 3 4° ‘ \ AI uptake
'5 minimal medium
a (no BFA)
20 —
0 l l I l l I
O 5 10 15 20 25 30 35

time (min), pre-incubation BFA

Figure 5. The effect of the duration of pre-incubation with BF A, prior to incubation in
minimal medium, on blockage of the cessation of Al uptake in tobacco BY-2 cells, by BFA.
Cells 2 days after subculture were used. Prior to any other treatment, cells were pre-incubated
with 5 and 20 uM BFA for 0 to 30 min, then incubated in minimal medium + BFA for 1 h.
Al uptake (1 h) and morin treatment were performed as described in Fig. 4. The lower and
upper dashed lines represent control cells not treated with BF A and control cells not treated

with both BFA and minimal medium, respectively.

89

dose-dependence curve, for BFA inhibition of Al uptake reduction, was performed (Fig. 6).
Effects were noticeable at concentrations as low as 1 uM. Although complete saturation of
the response was not observed, 5 to 10 uM BFA produced a response close to maximum.
Further increments, with use of up to 50 uM of BFA, were small (Fig. 6). These
concentrations and duration of treatment with BFA are low, consistent with a primary and
speciﬁc effect upon the Golgi apparatus. In plant cells, concentrations of BFA which have
been employed range from about 0.5 to 700 11M (Satiat-Jeunemaitre et al., 1996). In
mammalian systems, concentrations vary from 0.2 to 20 11M depending on whether intact
cells or a cell-free system is employed (Robinson, 1993). A hallmark of BFA effects upon
the Golgi apparatus is its rapid reversibility (Driouich et al., 1993). Once BF A was removed
from the minimal medium, the capability of cells for Al uptake dropped similarly to those
cells treated only with minimal medium (Fig. 7 A). Finally, addition of BFA to cells that had
already experienced some loss in their A] uptake capacity did not result in recovery of Al
uptake, rather it had the effect of maintaining the same levels of Al uptake and avoiding

further drops in Al uptake capacity (Fig. 7 B).

90

 

120 _ AI uptake

   
 
 

 

 

 

 

g 100—
C
3
‘i‘t‘
.1: _
3g 80
CE
$3 60
‘58
38
<01
2 40
O ..
3
C
20—
0 l l l l T
o 10 20 30 40 so 60

BFA concentration (11M)

Figure 6. Dose dependence curve for the suppression, by brefeldin A, of the cessation of Al
uptake by tobacco BY-2 cells, induced by incubation in minimal medium. Cells after 2 days
of subculture were used. Cells were incubated in minimal medium for 1 h, followed by
incubation with Al (1 h), and treatment with morin, as described in Fig. 4. The dashed line
represents Al uptake in control cells neither incubated in minimal medium nor with BFA

prior to uptake. Bars represent SE (n=3).

91

Figure 7. Reversibility of the effects of brefeldin A on loss of Al uptake capacity (A) and its
effect after some loss of A1 uptake capacity has already occurred (B) in tobacco BY-2 cells.
Prior to any other treatment, cells were pre-incubated with O or 10 uM BFA for 30 min, then
incubated in minimal medium (i BFA) for various times. Cells were treated with Al for 1
h and with morin as described in Fig. 4. Cells 3 days after subculture were used. A) After 1
h incubation in minimal medium and BF A, some cells were reincubated in minimal medium
free of BFA (dashed line). B) BFA (10 uM) was added, at two different times, to cells
incubated in minimal medium, after some loss of Al uptake capacity had occurred.Bars

represent SE (n=3).

AI uptake (1 hr)

Al uptake (1 hr)
ﬂuorescence (arbitrary units)
3
l

100

92

 

 

     
  
  
 
  
  

 

 

 

  

A /
.75 _
'E 80 —
3 . . .
Z‘ ' . minimal medium
E 41 (+10 11M BFA)
s: o _
'9 6 release g -
3 from BFA -.
o A .
0 ~, minimal
5 40 A medium
9
in
E
g 20 A
a:
minimal medium
(-BFA)
O
B
50 1.
addition of
10 11M BFA

 

   
 
 
 

addition of
10 11M BFA

minimal medium
(- BFA)

 

0

Figure 7.

l H I l

1 2 3 4
duration of incubation (hours)

 

DISCUSSION

Three major new ﬁndings emerge from this work. First, treatment of growing cells
with Al promotes deformities in the cell surface, presumably in regions where new cell wall
and plasma membrane are being added. Second, upon reduction of cell growth following
incubation in minimal medium, Al uptake is also reduced. Third, Al uptake is sustained when
cells are treated with BF A, despite incubation in minimal medium and reductions in growth.
It is suggested that alterations in the plasma membrane or cell wall occur concomitant with
cessation of growth, and result in reduced Al uptake. Cells treated with BF A are incapable
of these alterations and therefore continue to take up Al.

Based on observations with the microscope and using Al-morin ﬂuorescence, it is
clear that Al is associated with deformities (protuberances, vesicle-like and globular
structures) of the cell surface (Fig. 1). It is concluded that Al induced these abnormalities of
the cell surface because: 1) These structures were not observable in control cells treated
identically but without Al; 2) These structures were more pronounced after longer periods
of incubation with Al; 3) Presumably because of these abnormal structures, there was a
tendency for the cell wall and protoplast to remain attached at these sites following
plasmolysis. This was observable only in Al-treated cells.

With respect to the nature of these regions, it is important to try to distinguish
whether these events are occurring because of interactions of A1 with the plasma membrane,
with the cell wall, or both. In at least some cells plasmolyzed after Al treatment, Al was
apparently completely associated with the protoplast (Fig. 3B). Evidence favoring the notion

that Al uptake involves passage across the plasma membrane was provided previously

93

94

(Vitorello and Haug, 1996 - chapter 3). Involvement of the plasma membrane is again
suggested by the fact that Al uptake was almost completely eliminated in growing cells
treated with sorbitol prior to exposure to Al.

These regions of Al accumulation possibly coincide with regions of cell growth,
where new cell wall and plasma membrane are being added (V itorello and Haug, 1996). This
is suggested because of the dependence of Al uptake on cell growth and because of the
patterns of Al accumulation in the cell. The ﬂuorescence observed along the region of
contact between adjacent cells (Fig. 2) was suggestive of recent observations made on
cytokinesis in this same BY-2 cell line, which were in the process of fusing their cell plate
with the parent plasma membrane and cell wall (Samuels et al., 1995). Fusion occurs at the
edges of the plane separating the daughter cells, which seems to colocalize with Al-morin
ﬂuorescence. Furthermore, fusion occurs through ﬁnger-like projections at the margins of
the cell plate. If Al accumulates at these sites of ﬁrsion, this would explain the presence of
several discrete ﬂuorescing regions observed along the plane of separation (Fig. 2).
Alternatively, this pattern of Al accumulation may coincide with maturation of the cell plate,
which involves closing of plate fenestrae (also towards the margins) and cellulose synthesis
(Samuels et al., 1995).

The hypothesis that aluminum accumulates in regions of cell wall and plasma
membrane synthesis is consistent with the fact that Al toxicity is manifested in growing cells
of the root apex (Ryan et al., 1993; Kochian, 1995), that it inhibits tip growth in root hairs
and pollen tubes (Konishi and Miyamoto, 1983; Searcy and Mulcahy, 1990; Jones et al.,
1995), that it disrupts secretory activity and causes derangements of the Golgi in root cap

cells (Bennet et al., 1991), that incorporation of sucrose into cell walls is impeded (Huck,

95

1972) and that callose synthesis is induced (Schaeffer and Walton, 1990). Most compelling
are recent observations that Al induces deformities only in the tips of growing root hairs,
following a few hours of exposure to Al (Jones et al., 1995).

It seems reasonable that the observed abnormalities occurred because of
derangements caused by Al in the incorporation of cell wall and plasma membrane. Such
derangements can explain the tendency for the cell wall and plasma membrane to remain
attached in Al-treated, plasmolyzed cells. It may also explain some previously reported
observations in the literature. In wheat roots, Al uptake was interpreted to depend on an
intact protoplast and to involve metabolism dependent uptake into the cell wall (Zhang and
Taylor, 1990). In one of the few previous applications of morin to studies of Al in plant cells,
accumulation of Al in the cell walls of elongating cells may have been wrongly interpreted
as being a consequence of Al precipitation (Fig. 4 of Tice et al., 1992).

Although it is not yet clear how and why Al accumulates in these discrete regions,
intriguing information on the cessation of Al uptake in cells, and its suppression by BFA, has
been obtained. This may contribute to a further understanding of Al uptake. On the other
hand, inhibitors of cell growth processes had almost no effect on Al uptake (Tab. 2), whereas
cells largely lost their capacity for Al uptake following incubation of cells in a minimal
medium. This occurred presumably because of reduced growth upon incubation in minimal
medium. The manner in which growth is reduced may determine whether Al uptake is
affected or not, since BF A halted cell grth (Tab. 1) but did not reduce Al uptake.

An intriguing and important ﬁnding was that BF A sustained the capacity of cells to
incorporate Al, following incubation in minimal medium and concomitant reduction in

growth (Fig. 4). Two points should be stressed. Brefeldin A did not reverse the loss in Al

96

uptake capacity of cells (Fig. 7b) but rather prevented its occurrence (Fig. 4). Brefeldin A
itself had no signiﬁcant effect on Al uptake, except possibly aﬁer prolonged exposure of
cells to BFA (Tab. 2, Figs. 4 and 7).

The antibiotic brefeldin A, a fungus-derived macrocyclic lactone, is an inhibitor of
vesicular trafﬁc between the endoplasmic reticulum (ER) and the cistemae of the Golgi
apparatus. Its application to cells results in disassembly of the Golgi apparatus and inhibition
of the secretory pathway, particularly as assessed by protein secretion (Misumi et al, 1986;
Driouich et al., 1993). Brefeldin A is widely used as an experimental tool, particularly in
mammalian cells, for the study of Golgi-associated events (Klausner et al., 1992; Staehelin
and Moore, 1995). However, in plant cells the effects of BF A on the Golgi are not as well
deﬁned (Satiat-Jeunemaitre et al., 1996). It does not always result in disassembly of the
Golgi, and its effects have been observed to be more pronounced on the trans-Golgi network
(Driouich et al., 1993). Brefeldin A is thought to inhibit binding of GTP to the ADP
ribosylation factor (ARF), a small GTP-binding protein. This prevents the association of
ARF and coatomer proteins with the membranes of the ER and Golgi, and consequently
budding of vesicles is impeded (Torii et al., 1995). Prior application of GTPyS and AlF,‘ can
inhibit the effects of BFA (Klausner et al., 1992).

It is concluded that the effects of BFA on maintaining Al uptake in these tobacco
cells resulted from the known interference of BF A on vesicular trafﬁc, rather than secondary
effects because: 1) The concentration of BFA employed was low (10uM), similar to those
used in mammalian cells and is among the lowest reportedly used for plant cells (Robinson,
1993; Satiat—Jeunemaitre et al., 1996). In these tobacco BY-2 cells, 20 1.1M BFA was

sufficient to promote disassembly of the Golgi (Yasuhara et al., 1995); 2) The effect of BF A

97

was highly reversible (Fig. 7A), a hallmark of the effect of this compound on the Golgi
apparatus and its associated secretory pathway (Driouich et al., 1993); and 3) Addition of
BFA to the cells only during Al incubation had no effect on Al uptake, i.e. there is no
evidence that it somehow facilitated uptake of Al.

Thus, by apparently inhibiting the Golgi secretory pathway, BFA maintains the cells
capable of Al uptake upon incubation in minimal medium. In growing cells, the bulk of
vesicular trafﬁc is expected to occur between the trans Golgi apparatus and regions of cell
growth and expansion. Indeed, in tobacco BY-2 cells, BF A inhibited the formation of cell
plates (Yasuhara et al., 1995) and it inhibited auxin-mediated growth and secretion of cell-
wall proteins in maize coleoptiles (Schindler et al., 1994). Aluminum accumulates in regions
at or close to the cell surface (F igs.1, 2 and 3; Vitorello and Haug, 1996). This putative role
of the Golgi secretory pathway in Al uptake lends support to the hypothesis that Al
accumulates in regions of cell growth.

The mechanism by which BFA inhibits the loss of A1 uptake capacity in cells is not
clear. The effects of BFA on vesicular traffic do not seem to be limited to the Golgi
apparatus, per se. There is evidence that it affects vesicular trafﬁc along the exocytotic and
endocytotic pathway also (Pelham, 1991; Driouich et al., 1993; Torii et al., 1995; Satiat-
J eunemaitre et al., 1996). Thus, it is likely that in addition to reduced secretion, plasma
membrane turnover is inhibited by BFA, caused either by inhibition of the Golgi secretory
pathway or by inhibition of exocytosis and endocytosis. It is also likely that there are
important changes in the composition of the cell surface when a region ceases growth (or
vice versa). These changes require, in part, plasma membrane turnover and the transport of

proteins, precursors etc. to the cell surface. Indeed, BFA reportedly inhibited the secretion

98

of auxin-binding protein in maize suspension cultures (Jones and Herman, 1993) and of a
“wall-loosening factor”in maize coleoptiles (Schindler et al., 1994). Taken together, it is
hypothesized that alterations in the composition of the cell surface are responsible for
cessation of Al uptake. Maintenance of Al uptake by BFA results from the inhibition of these
changes at the cell surface.

What are these changes? Proteins or lipids could be involved. The transit of
glycolipids and sphingomyelin through the Golgi to the plasma membrane was inhibited by
BF A in BHK21 cells (Kallen et al., 1993). However, the transport of
phosphatidylethanolamine, in hepatocytes, and cholesterol, in chinese hamster ovary cells,
to the plasma membrane were not, even though protein secretion was greatly reduced (Urbani
and Sirnoni, 1990; Vance et al., 1991). With respect to proteins, BF A differentially inhibited
secretion of proteins in tobacco cell cultures cv. Havanna (Kunze et al., 1995).

A clue to this puzzle may be the fact that low pH induces enhanced membrane
permeability (Yan et al., 1992). This was found to occur in growing cells, but not in non-
growing cells of these tobacco BY-2 cells, and this low-pH induced increase in membrane
permeability may play a role in Al uptake (V itorello and Haug, 1996). Protonation of
proteins and lipids involved in membrane fusion could be responsible for the destabilization
of the plasma membrane.

For the purpose of speculation and to illustrate a possible scenario, one possibility
may be annexins, which are Ca21-dependent proteins which bind to acidic phospholipids and
can be regulated by phosphorylation. These proteins can be transported to the cell surface by
the Golgi secretory pathway and play a role in a wide variety of essential processes, among

which are exocytosis, membrane fusion and vesicle trafficking. GTP-binding proteins are

99

sometimes associated in these processes (Clark and Roux, 1995). Annexins have been
immunolocalized to tip-growing regions of pollen tubes and fern rhizoids (Blackboum et al.,
1992; Clark et al., 1995) and also to secretory cells such as outer cells of the root cap (Clark
et al., 1992;1995). In addition to involvement in exocytosis and secretory activity, annexins
have been implicated with cellulose and callose synthases (Andrawis et al., 1993; Clark and
Roux, 1995), and also with interactions of microﬁlaments, particularly actin, with the plasma
membrane (Clark and Roux, 1995). These processes have all been reported to be affected
by Al (Grabski et al., 1995; Schaeffer and Walton, 1990), and it is conceivable that a member
of the annexin family is involved with Al accumulation in cells, either by generating
membrane instability or perhaps by direct binding of Al. BF A treatment may cause this
protein to remain at the cell surface upon incubation of cells in minimal medium and

cessation of growth.

REFERENCES

Andrawis, A., Solomon, M., Delrner, DP. 1993. Cotton ﬁber annexins: a potential role in
the regulation of callose synthase. Plant J. 3: 763-772.

Bennet, R.J., Breen, CM. 1991. The aluminium signal: new dimensions to mechanisms of
aluminium tolerance. Plant Soil 134:153-166.

Blackboum, H.D., Barker, P.J., Huskisson, N.S., Battey, NH. 1992. Properties and partial
protein sequence of plant annexins. Plant Physiol. 99:864-871.

Clark, G.B., Dauwalder, M., Roux, SJ. 1992. Puriﬁcation and immunolocalization of
annexin-like protein in pea seedlings. Planta 187: 1-9.

Clark, G.B., Tumwald, S. Tirlapur, U.K., von der Mark, K., Roux, S.J., Scheuerlein, R.
1995. Induction and polar distribution of annexin-like proteins during phytochrome-

mediated rhizoid initiation and growth in spores of the ferns Diyopteris and Anemia.
Planta 197:376-3 84.

Clark, G.B., Roux, SJ. 1995. Annexins of plant cells. Plant Physiol. 109:1133-1139.

Doonan, J .H., Cove, D.J., Lloyd, CW. 1988. Microtubules and microﬁlaments in tip growth:
evidence that microtubules impose polarity on protonemal growth in Physcomitrella
patens. J. Cell Sci. 89:533-540.

Driouich, A., Zhang, G.F., Staehelin, LA. 1993. Effect of brefeldin A on the structure of the
Golgi apparatus and on the synthesis and secretion of proteins and polysaccharides in

sycamore maple (Acer pseudoplatanus) suspension-cultured cells. Plant Physiol.
101:1363-1373.

Grabski, S., Schindler, M. 1995. Aluminum induces rigor within the actin network of
soybean cells. Plant Physiol. 108:897-901.

Hepler, P.K., Bonsignore, CL. 1990. Caffeine inhibition of cytokinesis: ultrastructure of cell
plate/degradation. Protoplasma 157: 182-192.

Huck MG (1972) Impairment of sucrose utilization for cell wall formation in the roots of
aluminum-darnaged cotton seedlings. Plant Cell Physiol 13:7-14

100

101

Jones, AM. and Herman, EM. 1993. KDEL-containing auxin-binding protein is secreted
to the plasma membrane and cell wall. Plant Physiol. 101:595-606.

Jones, D.L., Shaff, J .S., and Kochian, L.V. 1995. Role of calcium and other ions in directing
root hair tip growth in Limnobium stoloniferum. 1. Inhibition of tip growth by aluminium.
Planta 197:672-680.

Kallen, K.J., Quinn, P., Allan, D. 1993. Effects of brefeldin A on sphingomyelin transport
and lipid biosynthesis in BHK21 cells. Biochem. J. 289:307-312.

Klausner, R.D., Donaldson, J .G., Lippincott-Schwartz, J. 1992. Brefeldin A: insights into the
control of membrane trafﬁc and organelle structure. J. Cell. Biol. 116: 1071-1080.

Kochian, L.V. 1995. Cellular mechanisms of aluminum toxicity and resistance in plants.
Annu. Rev. Plant Physiol. Plant Mol. Biol. 46: 237-60.

Konishi, S., Miyamoto, S. 1983. Alleviation of aluminum stress and stimulation of tea pollen
tube growth by ﬂuorine. Plant Cell Physiol. 24:857-862.

Kunze, I., Hillmer, S., Kunze, G., Muntz, K. 1995. Brefeldin A differentially affects protein
secretion from suspension-cultured tobacco cells (Nicotiana tabacum L.).

Misumi, Y., Miki, A., Takatsuki, A., Tamura, G., Ikehara, Y. 1986. Novel blockade by
brefeldin A of intracellular transport of secretory proteins in cultured rat hepatocytes. J.
Biol. Chem. 261 :1 1398-1 1403.

Pelham, HR. 1991. Multiple targets for Brefeldin A. Cell 67:449-451.

Rincon, M., Gonzales, RA. 1992. Aluminum partitioning in intact roots of aluminum-
tolerant and aluminum-sensitive wheat (Triticum aestivum L.) cultivars. Plant Physiol.
99:1021-1028.

Robison, D.G. 1993. Brefeldin A: a tool for plant cell biologists? Bot. Acta 106: 107-109.

Ryan, P.R., DiTomasso, J.M., Kochian, L.V. 1993. Aluminium toxicity in roots: an
investigation of spatial sensitivity and the role of the root cap. J. Exp Bot 44:437-446.

Samuels, A.L., Giddings, T.H., Staehelin, LA. 1995. Cytokinesis in tobacco BY-2 and root
tip cells: a new model of cell plate formation in higher plants. J. Cell Biol. 130: 1345-
1357.

Satiat-Jeunemaitre, B., Hawes, C., 1992. Reversible dissociation of the plant Golgi apparatus
by brefeldin A. Biol. Cell 74:325-328.

102

Satiat-Jeunemaitre, B., Cole, L., Bourett, T., Howard, R., Hawes, C. 1996. Brefeldin A
effects in plant and fungal cells: something new about vesicle trafficking. J. Microscopy
181:162-177.

Schaeffer, H.J., Walton, J .D. 1990. Aluminum ions induce oat protoplasts to produce an
extracellular (1-3)B-D—glucan. Plant Physiol. 94:13-19.

Shi, B., Chou, K., Haug, A. 1993. Aluminium impacts elements of the phosphoinositide
signalling pathway in neuroblastoma cells. Mol. Cell. Biochem. 121:109-118.

Schindler, T., Bergfeld, M., Hohl, M., Schopfer, P. 1994. Inhibition of Golgi-apparatus
function by brefeldin A in maize coleoptiles and its consequences on auxin-mediated
growth, cell-wall extensibility and secretion of cell-wall proteins. Planta 192: 404-413.

Schreiner, K.A., Hoddinott, J ., Taylor, G.J. 1994. Aluminum-induced deposition of (1,3)-
beta-glucans (callose) in Triticum aestivum L. Plant Soil 162:273-280.

Searcy, K.B., Mulcahy, D.L. 1990. Comparison of the response to aluminum toxicity in
gametophyte and sporophyte of four tomato (Lycapersican esculentum Mill.) cultivars.
Theor. Appl. Genet. 80:289-295. '

Shedletzky, B., Shrnuel, M., Trainin, T., Kalman, S., Delmer, D. 1992. Cell wall structure
in cells adapted to growth on the cellulose-synthesis inhibitor 2,6-dichlorobenzonitrile.
Plant Physiol. 100: 120-130.

Staehelin, L.A., Moore, 1. 1995. The plant Golgi apparatus: structure, functional organization
and trafficking mechanisms. Ann. Rev. Plant Physiol. Plant Mol. Biol. 46: 261-288.

Tice, K.R., Parker, D.R., DeMason, D.A. 1992. Operationally deﬁned apoplastic and
symplastic aluminum fractions in root tips of aluminum-intoxicated wheat. Plant Physiol.
100: 309-318.

Torii, S., Banno, T., Watanabe, T., Ikehara, Y., Murakami, K., Nakayama, K. 1995.
Cytotoxicity of Brefeldin A correlates with its inhibitory effect on membrane binding of
COP coat proteins. J. Biol. Chem. 270: 11575-1580.

Urbani, L., Sirnoni, RD. 1990. Cholesterol and vesicular stomatitis virus G protein take
separate routes from the endoplasmic reticulum to the plasma membrane. J. Biol. Chem.
265:1919-1923.

Vance, J .E., Aasman, E.J., Szarka, R. 1991. Brefeldin A does not inhibit the movement of
phosphatidylethanolamine from its sites of synthesis to the cell surface. J. Biol. Chem.
266:8241-8247.

103

Vitorello, VA. and Haug, A. 1996. Short-term aluminium uptake by tobacco cells: Growth
dependence and evidence for internalization in a discrete peripheral region. Physiol.
Plant. (in press).

Yasuhara, H., Sonobe, S., Shibaoka, H. 1995. Effects of brefeldin A on the formation of the
cell plate in tobacco BY-2 cells. Eur. J. Cell Biol. 66:274-281.

Yamamoto, Y., Rikiishi, S., Chang, Y., Ono, K., Kasai, M., Matsumoto, H. 1994.
Quantitative estimation of aluminium toxicity in cultured tobacco cells: correlation
between aluminium uptake and growth inhibition. Plant Cell Physiol. 35: 575-583.

Yan, F., Schubert, S., Mengel, K. 1992. Effect of low root medium pH on net proton release,
root respiration, and root growth of corn (Zea mays L.) and broad bean (Viciafaba L.).
Plant Physiol. 99: 415—421.

Zhang, G., Taylor, G.J. 1990. Kinetics of aluminum uptake in T riticum aestivum L.. Identity
of the linear phase of aluminum uptake by excised roots of aluminum-tolerant and
aluminum-sensitive cultivars. Plant Physiol. 94: 577-584.

CHAPTER 5

GENERAL DISCUSSION AND PERSPECTIVES

GENERAL DISCUSSION AND PERSPECTIVES

In this study, some novel ﬁndings on Al uptake by cultured tobacco cells are reported.
A major theme to emerge from this research is that cell growth is somehow implicated with
Al uptake. Non-growing cells do not take up Al, neither do they appear to be sensitive to it,
as evidenced from the absence of observable symptoms of toxicity in non-elongating root
cells (Ryan et al., 1993). Another major novelty is the discovery of discrete Al-accumulating
regions at the periphery of the cell and Al-induced abnormalities at the cell surface. An
intriguing ﬁnding is that brefeldin A-sensitive membrane/vesicular traffic plays a role in
whether a cell becomes incapable of Al uptake. What is the relationship between cell growth
and Al uptake? What role do these Al-accumulating regions play? How does
membrane/vesicular trafﬁc affect a cell’s capability to take up Al? These are some of the
questions that are posed. The objective of this chapter is to summarize and discuss the
ﬁndings on Al uptake, to present a model for uptake of A1 and to discuss future research
perspectives.

Using the ﬂuorescence assay developed for the detection of Al in these tobacco BY-2
cells (chapters 2 and 3) and also GFAAS, Al uptake was found to be correlated with and
dependent on cell growth (Chapter 3). When growing cells were incubated in minimal
medium, leading to reduced or arrested growth, Al uptake almost ceased to occur (Chapter
4). The underlying causes for the association between A1 uptake and cell growth are not yet

known. However, greater availability of metabolic energy for Al uptake, during cell growth,

104

105

does not appear to be the cause. In intact and excised roots, metabolic inhibitors were
demonstrated to increase Al uptake (Wagatsuma, 1983; Zhang and Taylor, 1989; Zhang and
Taylor, 1991) rather than decrease it. However, metabolic inhibitors (DNP) had no effect on
Al uptake in Neuroblastoma cells (Shi and Haug, 1990).

Some additional ﬁndings are summarized. The kinetics of Al uptake in these cultured
cells was found to be similar to those found in roots, which is important if use of these cells
are desired as a plant cell model for Al uptake. Other than age, factors which were found to
affect A1 uptake were Ca2+ concentration, low temperature, increased osmoticum, and pre-
chelation of Al to citrate (Chapters 3 and 4). The low pH of 4.5, required in Al toxicity
studies, was itself detrimental to cells, causing increased membrane permeability. Some
evidence for a correlation between A1 uptake and this low-pH induced membrane
permeability was found (Chapter 3). The ameliorative effects of factors such as cations (ionic
strength) and low temperature, upon Al uptake, may result from their modulating effect on
low-pH induced alterations in membrane structure. The physiological effects of the low pH
of solutions which are necessary for Al toxicity to occur, must obviously be taken into
consideration when examining the effects of Al.

As judged with this Al-morin ﬂuorescence technique, Al accumulated in discrete
peripheral regions of the cell, in particular in protuberances and in material presumably
originating therefrom (Chapters 3 and 4). These structures were induced by treatment with
Al. Generalized binding of Al to the cell wall was not detected. The cell wall and the plasma
membrane appeared to become closely associated in Al-accumulating regions, as evidenced
by plasmolysis of Al- and morin-labelled cells, in which the plasma membrane tended to

remain in contact with the cell wall in regions where Al accumulated. Plasmolysis of cells

106

prior to exposure to Al resulted in very low levels of Al uptake and virtually no localized
regions of Al-morin ﬂuorescence. It appears that any accumulation of Al in the cell wall, in
these discrete Al-accumulating regions, is the result of the release of cellular material,
induced by Al itself, into the cell wall. First, however, Al is taken up by the protoplast. This
is supported by the statements made above, by the observation of Al associated with the
symplast of plasmolyzed cells, and by arguments in Chapter 3, which support Al
internalization.

It is suggested that the regions of Al accumulation are regions where the cells are
growing, i.e., where new membrane and cell wall are being added. This is suggested because
1.) A1 uptake occurs only in growing cells; 2) Al accumulation is localized to a discrete region
close to the cell surface, and the pattern of Al localization is suggestive of events occurring
during cytokinesis in these tobacco BY-2 cells (Samuels et al., 1995); 3) Al induces
deformities of the cell surface, suggestive of deformities induced by Al in growing root hair
tips (Jones et al., 1995); 4) BFA-sensitive membrane/vesicular traffic affects the cell’s
capability of Al uptake.

Two major experimental tools were instrumental for generating the new information.
A ﬂuorescence assay for the detection of A1 and use of a cell culture as a plant cell model.
The ﬂuorescence method was adapted, optimized and a comparison of results obtained with
this technique and with GFAAS validated this method as a technique for the detection of Al
(Chapter 2). This method has several advantages over other methods of detection previously
employed, such as GFAAS itself, X-ray microanalysis and the recently employed method of
secondary-ion mass spectrometry (Lazof et al., 1994). It is sensitive and allows both for

quantitative measurements and intracellular localization, with higher resolution than

107

previously possible. It is performed on intact cells, without sample preparation and is easy,
rapid and inexpensive to perform. It is also apparently insensitive to Al precipitates. Because
of these advantages, this method should be useful in future studies on Al toxicity, a ﬁeld
which has been limited by appropriate methodology (Taylor, 1988; Delhaize and Ryan, 1995)
and is potentially applicable to other systems such as root tips.

A tentative model for uptake of Al by plant cells is proposed (Fig. 1). Al has access
to potential targets because of events occurring in regions of cell growth, during membrane
turnover and cell wall deposition. Membrane instabilities which occur during membrane
fusion events serve as a port of entry for Al, or expose binding sites of membrane-bound
proteins. The observed increases in membrane permeability at low pH (Chapter 3)
presumably results from enhanced membrane instability of these regions. Low pH has been
shown to promote non-bilayer conﬁgurations in some lipids (Lindblom and Rilfors, 1992).
Alternatively, membrane-bound proteins become protonated at low pH and low cation
concentration, and could be responsible for membrane instabilities. Thus, any treatment
which has an effect on membrane stability, could affect Al uptake. This would be consistent
with the effects of low temperature, high Ca2+ and cation concentration, reduced cell turgor
and metabolic inhibitors on Al uptake. Such a model based on membrane properties could
explain why non-metabolic conditions provided by cold treatment and metabolic inhibitors
produce Opposing effects on Al uptake (Wagatsuma, 1983; Zhang and Taylor, 1989, 1990).

In this model, the role of BF A-sensitive membrane/vesicular trafﬁc in determining
a cell’s capability for Al uptake rests on its ability to transport material (proteins, lipids,
polysaccharides etc.) to and from the cell surface. Thus, the arrival or departure of key

components at the cell surface may greatly determine whether AI will accumulate in this

108

 

 

 

 

 

growing
PM CW
ER cell
BFA
i vesicles
% A|3+
membrane
instability at
low pH and low
> ionic strength

 

 

 

 

Figure l. A schematic representation of a model for uptake of Al into regions of growing
cells involved in the incorporation of new cell wall and plasma membrane. Membrane
instabilities in these regions resulting from fusion events, and enhanced by low-pH and low
ionic strength, facilitate access of Al to the interior of the cell or to Al-binding proteins of
the plasma membrane. Accumulation of Al in these regions causes derangements in
processes involved in growth, and eventually in deformations of the cell surface. Upon
cessation of growth, changes in the cell surface impede Al uptake. These changes depend on
vesicular traffic, which is inhibited by action of BFA. ER-endoplasmic reticultun, PM-

plasma membrane, CW-cell wall, BFA-brefeldin A.

109

region or not. The question is what are these components?

Following Al uptake, Al-induced derangements in this cell region could cause the
observed release of cellular material (Chapter 4). Since this material was found to be exterior
to the rest of the cell, this might help explain why there has been much controversy and
ambiguity over whether Al accumulates in the apoplast or the symplast. Tice et a1. (1992),
in one of few previous applications of morin on plant tissue, observed localized Al in the cell
wall, which was interpreted to be Al precipitation, but is perhaps better explained by the
release of Al-containing material. Furthermore, this can also explain Zhang and Taylor’s
(1990, 1991) “metabolism dependent” accumulation of Al within the cell wall.

Yarnamoto et al. (1994) have recently demonstrated a correlation between A1 uptake
and Al toxicity in tobacco cell cultures. Exclusion of A] may be a mechanism for avoiding
Al toxicity. Increased cation concentration and chelation of Al to citrate in the solution both
reduced Al uptake (Chapter 3) and both are well known to ameliorate Al toxicity in roots
(Kinraide and Parker, 1987; Suhayda and Haug, 1986). Recently, it has been reported that
tolerance to Al may be associated with enhanced malate exudation (Delhaize et al., 1993,
Basu et al., 1994; Pellet et al, 1995) which would ﬁmction to keep Al outside of the cell. That
A1 is excluded from non-growing cells is clearly demonstrated in Chapter 3.

The root apex, speciﬁcally the meristem and elongating cells, has recently been
shown to be the primary site of inhibition of root grth by Al (Ryan et al., 1993). In
addition to these cells, Al has also been shown to cause injuries to root cap and root hair cells
(Bennet and Breen, 1991). Our results and suggested model point to a common denominator
between these cells, namely the fact that they are growing or actively involved in secretory

processes.

110

Future perspectives

The Al-morin ﬂuorescence assay is potentially very useful in Al toxicity studies and
also in plant screening programs. In a very recent application (Larsen et al., 1996), morin was
utilized to examine Al in Al-sensitive mutants of Arabidopsis. Imaging of Al in intact root
tissues could be accomplished with confocal microscopy and the Al-morin technique.

Hopefully, the work described here will foster continued investigations into the
matter of how and where Al accumulates in cells. Further supporting evidence for the
hypothesis that Al accumulates in cell regions involved in new plasma membrane and cell
wall synthesis might be obtained by examining Al-morin ﬂuorescence in tip-growing cells
such as pollen tubes and root hairs. The growing region of these cells is well deﬁned. Similar
examinations could be performed in budding yeast.

Use of markers of regions of cell growth would be ideal to determine if Al
accumulation coincides with these regions. Annexins, Cay-dependent phospholipid-binding
proteins involved in membrane fusion, are possible candidates for use as markers.
Immunolocalization studies have determined that annexins are concentrated in the tip region
of polarly growing cells such as pollen tubes and fern rhizoids (Blackbourn et al., 1992;
Clark et al., 1995) and also in secretory cells such as the outer cells of the root cap (Clark et
al., 1992). In a recent study, Al induced rigidiﬁcation in transvacuolar strands of actin in
soybean (Glycine max) cell suspensions (Grabski and Schindler, 1995). It has not been
shown whether Al reaches this actin. IfAl accumulates in regions at or near the cell surface,
how does it interact with actin? Membrane-associated annexins have been shown to interact

with and modulate actin bundling (Clark and Roux, 1995; Glenney et al., 1987). Thus,

III

studies relating the cellular distribution of Al to immunolocalization of annexins and actin
would be informative. In such studies, it would also be of interest to examine if actin and
annexins are present in Al-induced extracellular material (chapter 4.). In preliminary studies
(data not shown) these external structures were found to be detergent-insoluble, as are
cytoskeleton fractions. The feasibility of such studies would depend, in part, on the
accessibility to needed antibodies.

Another approach to examine the nature of these Al-accumulating regions would be
the use of yeast as a model system. Processes involved in cell growth and in the secretory
pathway are better understood in yeast than in plants, and a larger selection of experimental
tools is available. In yeast there are many mutants of growth-associated processes, such as
the Golgi-mediated secretory process. The genetic dissection of these processes is an active
area of research (Chant, 1994). Such an approach could prove very useful. However, little
is known about Al toxicity in yeast and their susceptibility to A1.

The possible relationship between low pH, plasma membrane permeability to ions,
Al uptake and cellular regions of growth should be examined. One possibility would be to
perform patch-clamp studies on growing cells. Tip-growing cells, such as pollen tubes or root
hairs, could be used to examine whether changes in membrane permeability are localized to
regions of cell growth, and whether there is a correlation with accumulation of Al. An
obvious obstacle to this approach is the presence of the cell wall. In at least one study,
however, laser microsurgery was used to expose the plasma membrane of the growing apex
of F ucus spiralisn rhizoid cell so that conventional patch clamp studies could be performed
(Taylor and Brownlee, 1992). A better alternative to patch clamping is the use of

extracellular vibrating ion-selective microelectrodes, which permits non-invasive

112

measurements of ion ﬂuxes (Smith et al., 1994).This technique has recently been used to
determine the magnitude and spatial localization of Ca”, K’ and H" ﬂuxes in root hairs of
Limnobium stolom'ferum (Jones et al., 1995). This technique could be used in conjunction
with the localization of Al by morin ﬂuorescence to answer important questions. Is Al
accumulation co-localized with low-pH increases in plasma membrane permeability to ions?
ls Al uptake modulated by factors which similarly modulate low-pH induced plasma
membrane ion permeability? What is the role of Ca2+ in stabilizing the membrane and in
modulating Al uptake? A question which arises from these considerations is whether a
correlation can be found between tolerance to low pH and tolerance to Al in plants.
Finally, to understand what processes are involved with Al uptake in growing cells,
studies comparing the physiology of cells capable of Al uptake and those which were
induced to lose their capacity for Al uptake could be performed. For example, are there
changes in plasma membrane composition (both in terms of lipids and proteins) between
these cells? Does BFA block these changes? An obvious problem with this approach is that
changes in a relatively small region of the plasma membrane may not be detectable if whole
cells or even plasma membrane preparations are used for analysis. Furthermore, ﬁnding such

differences would not necessarily establish a causal link.

REFERENCES

Basu, U., Godbold, D., Taylor, G.J. 1994. Aluminum resistance in Triticum aestivum
associated with enhanced exudation of malate. J. Plant Physiol. 1442747-753.

Bennet, R.J., Breen, CM. 1991. The aluminium signal: new dimensions to mechanisms of
aluminium tolerance. Plant Soil 134:153-166.

Bennet, R.J., Breen, C.M., Fey, M.V. 1987. The effects of aluminium on root cap function
and root development in Zea mays L. Environ. Exp. Bot. 27:91-104.

Blackboum, H.D., Barker, P.J., Huskisson, N.S., Battey, NH. 1992. Properties and partial
protein sequence of plant annexins. Plant Physiol. 99:864-871.

Chant, J. 1994. Cell polarity in yeast. Trends in Genetics 10: 328-333.

Clark, G.B., Dauwalder, M., Roux, SJ. 1992. Puriﬁcation and immunolocalization of
annexin-like protein in pea seedlings. Planta 187: 1-9.

Clark, G.B., Roux, SJ. 1995. Annexins of plant cells. Plant Physiol. 109:1133-1139.

Clark, G.B., Tumwald, S. Tirlapur, U.K., von der Mark, K., Roux, S.J., Scheuerlein, R.
1995. Induction and polar distribution of annexin-like proteins during phytochrome-
mediated rhizoid initiation and growth in spores of the ferns Dryopteris and Anemia.
Planta 197:376—384.

Delhaize, B., Ryan, P.R., Randall, P.J. 1993. Aluminum tolerance in wheat (Triticum
aestivum L.). 11. Aluminum-stimulated excretion of malic acid from root apices. Plant
Physiol. 103:695-702.

Delhaize, B., Ryan, PR. 1995. Aluminum toxicity and tolerance in plants. Plant Physiol.
107: 315-321.

Glenney, J.R., Tack, B., Powell, MA. 1987. Calpactins: two distinct Ca“ -regulated
phospholipid- and actin-binding proteins isolated from lung and placenta. J. Cell. Biol.
104: 503-511.

Grabski, S., Schindler, M. 1995. Aluminum induces rigor within the actin network of
soybean cells. Plant Physiol. 108:897-901.

113

114

Jones, D.L., Shaff, J .S., and Kochian, L.V. 1995. Role of calcium and other ions in directing
root hair tip growth in Limnobium stoloniferum. 1. Inhibition of tip growth by aluminium.
Planta 197:672-680.

Kinraide, T.B., Parker, DR. 1987. Cation amelioration of aluminum toxicity in wheat. Plant
Physiol. 83: 546-551.

Larsen, P.B., Tai, C.Y., Kochian, L.V., Howell, SH. 1996. Arabidopsis mutants with
increased sensitivity to aluminum. Plant Physiol. 110:743-751.

Lazof, D.B., Goldsmith, J .G., Rufty, T.W., Linton, R. 1994. Rapid uptake of aluminum into
cells of intact soybean root tips. A microanalytical study using secondary ion mass
spectrometry. Plant Physiol. 106: 1107-1114.

Lindblom, G., Rilfors, L. 1992. Cubic and hexagonal lipid phases. In Structure and dynamic
properties of lipids and membranes, (P.J. Quinn & R.J. Cherry, eds), pp 51-76. Portland
Press, London. ISBN 1-855-78-014-3.

Pellet, D.M., Grunes, D.L., Kochian, L.V. 1995. Organic acid exudation as an aluminum-
tolerance mechanism in maize (Zea mays L.). Planta 196:788-795.

Ryan, P.R., DiTomasso, J .M., Kochian, L.V. 1993. Aluminium toxicity in roots: an
investigation of spatial sensitivity and the role of the root cap. J. Exp Bot 44:437-446.

Samuels, A.L., Giddings, T.H., Staehelin, LA. 1995. Cytokinesis in tobacco BY-2 and root
tip cells: a new model of cell plate formation in higher plants. J. Cell Biol. 130: 1345-
1357.

Shi, B., Haug, A. 1990. Aluminum uptake by neuroblastoma cells. J. Neurochem. 55: 551-
558.

Smith, J .S., Sanger, R.H., Jaffe, LR 1994. The vibrating Caz’ electrode: a new technique for
detecting plasma membrane regions of Ca2+ inﬂux and efflux. Meth. Cell Biol. 40: 115-
134.

Suhayda, C.G., Haug, A. 1986. Organic acids reduce aluminum toxicity in maize root
membranes. Physiol. Plant. 68:189-195

Taylor, G.J. 1988. The physiology of aluminium phytotoxicity. In: Sigel, H., Sigel, A. (eds).
Metal ions in biological systems: Aluminum and its role in biology, vol. 24, pp 123-163.
New York, Marcel Dekker.

Taylor, A.R., Brownlee, C. 1992. Localized patch clamping of plasma membrane of a
polarized plant cell. Plant Physiol. 99:1686-1688.

115

Tice, K.R., Parker, D.R., DeMason, D.A. 1992. Operationally deﬁned apoplastic and
symplastic aluminum ﬁactions in root tips of aluminum-intoxicated wheat. Plant Physiol.
100: 309-318.

Wagatsuma, T. 1983. Effect of non-metabolic conditions on the uptake of aluminum by
plant roots. Soil Sci. Plant Nutr. 29: 323-333.

Yamamoto, Y., Rikiishi, S., Chang, Y., Ono, K., Kasai, M., Matsumoto, H. 1994.
Quantitative estimation of aluminium toxicity in cultured tobacco cells: correlation
between aluminium uptake and growth inhibition. Plant Cell Physiol. 35: 575-583.

Zhang, G., Taylor, G.J. 1989. Kinetics of aliuninum uptake by excised roots of aluminum
tolerant and aluminum sensitive cultivars of Triticum aestivum L.. Plant Physiol. 91:
1094-1099.

Zhang, G., Taylor, G.J. 1990. Kinetics of aluminum uptake in Triticum aestivum L.. Identity
of the linear phase of aluminum uptake by excised roots of aluminum-tolerant and
aluminum-sensitive cultivars. Plant Physiol. 94: 577-584.

Zhang, G., Taylor, G.J. 1991. Effects of biological inhibitors on kinetics of aluminium
uptake by excised roots and puriﬁed cell wall material of aluminium-tolerant and
aluminium-sensitive cultivars of Triticum aestivum L. J. Plant Physiol. 138:533-539.

APPENDIX A

GROWING TOBACCO CELLS RESPOND TO RAPID MEDIUM CHANGES
WITH A TRANSIENT INCREASE IN LOCALIZED Cit2+ ACTIVITY

Vitorello, V.A., Haug, A. 1996. Plant Cell Reports (in press).

ABSTRACT

A suspension of tobacco cells, Nicotiana tabacum L. BY-2, was subjected to a rapid medium
change, resulting in a disturbance of growth. A subpopulation of growing cells responded to
such a nutritional signal by establishing a transient, localized Caz’ accumulation, as judged
by chlorotetracycline ﬂuorescence. Residing near or at the plasma membrane, this initial Ca2+
signal began to relax after I h to a value presumably corresponding to an equilibrium Ca2+
level. This response was susceptible to treatment with brefeldin A, an agent impacting
vesicular trafﬁc, as indicated by a ﬁirther ﬂuorescence increase. By contrast, undisturbed
growing and non-growing cells did not display a Ca2+ response, regardless of the presence

of brefeldin A.

116

INTRODUCTION

Plant cells are confronted with the task of continually monitoring their
microenvironment and of adjusting intracellular parameters in response to varying external
conditions. Given the signiﬁcance of nutrient availability in the plant's environment,
signalling pathways must exist coupling growth promotion or arrest to nutrition-linked
signals. How plants sense their environment, and how signals are recognized and transduced
are poorly understood mechanisms. A potential participant in this type of signal
transmission is the Ca2+ signal, which in turn is triggered or modulated via upstream
pathways involving components like phosphoinositides, GTPases, and protein kinases
(Benidge 1995). In plant cells, Ca2+ is a key player in diverse cellular processes such as tip
growth during pollen germination and elongation (Herth et al. 1990; Zonia and Tupy 1995).

Evidence also indicates that Ca2+ exerts a profound inﬂuence upon the transport and
distribution of Golgi-synthesized, noncellulosic material, destined for the cellular surface,
via specialized carrier vesicles (Schmid and Damke 1995). For example, in animal cells
(Robinson et al. 1996), in maize, and growing pollen tubes (Clark and Roux 1995), annexin
proteins have been identiﬁed which interact in a Cay-dependent manner with secretory
vesicles mediating exocytotic processes. The Golgi apparatus of growing plant cells, as
opposed to that of mammalian cells (Rothman and Orci 1990; Cole and Lippincott-Schwartz
1995) is intensively engaged in synthesizing complex cell wall polysaccharides (Staehelin

and Moore 1995). To decipher mechanisms of vesicular traffic, in part Ca2’-dependent,

117

118

usage of yeast mutants, defective in certain steps of vesicular pathways (Chant 1994), and
treatment of mammalian cells (Klausner et al. 1992; Jackson and Kepes 1994), and plant
cells (Staehelin and Moore 1995), with a macrocyclic lactone, brefeldin A (Merck 1989),
have provided considerable insight into vesicular trafﬁc.

Given the paucity of information on pathways mediating signals in plant cells, we
examined the intracellular Ca2+ activity in growing tobacco cells, stressed by a rapid change
in medium. In response to such a signal, a transient localized Ca2+ accumulation is
observed, based on chlorotetracycline ﬂuorescence. Furthermore, the transient increase in

Ca2+ activity responds to brefeldin A-treatrnent.

MATERIALS AND METHODS

Cell culture and treatments

Cell suspension cultures of Nicotiana tabacum L. BY-2 were grown in a culture medium
(Nagata et al. 1992), at pH 5.6, 28°C, and an extracellular CaCl2 concentration of 3 mM.
Cell growth was determined by measuring packed cell volume (3-5 min centrifugation at 200
g in graduated tubes); each cm3 of packed cell volume corresponded to 7.9x10" cells.
Cellular grth was disturbed by rapidly (about 5 min) transferring cells from the culture
medium into a minimal medium composed of 10 mM MES, 5 mM sucrose, 5 mM KCl, 2
mM CaClz, pH 5.6, 23°C. Brefeldin A (Sigma, St. Louis, MO) was disssolved in methanol

(10 mM), applied to cells at a ﬁnal concentration of 10uM, and incubated for 30 min prior

119

to and during the duration of the treatment.

Fluorescence assay

Employing a titer of 4x1 05 cells/ml, intracellular Caz’ was evaluated with chlorotetracycline
(Sigma, St. Louis, MO). After removal of the supernatant, and resuspension of cells in
minimal medium devoid of the dye, but with 5 mM EDTA, ﬂuorescence measurements were
performed on a Perkin-Elmer LS-S luminescence spectrometer coupled to a microcomputer.
Wavelength settings were 395:5 nm for excitation and 520:2.5 nm for emission. Each
sample reading was taken as the average of the ﬁrst 15 s of monitoring. Reductions in
ﬂuorescence intensity, attributable to cell settling, were detectable only after at least 1 min.
Loading of cells with chlorotetracycline presented no difﬁculties in contrast to problems
encountered in loading plant cells with Ca2+ indicators like Fluo-3 (Timmers et al. 1991),
because of cell wall-associated esterases (Callaham and Hepler 1991). After entering the cell,
this ﬂuorescent antibiotic reports Ca” in the solution adjacent to hydrophobic sites like

membranes, with which the Ca2+ is in equilibrium (Tsien 1989).

Microscopy

Cells were observed with a Zeiss (Carl Zeiss, Oberkochen, Germany) WL research
microscope equipped with an epiﬂuorescence condenser IV FL. The ﬁlter set used consisted
of a band pass ﬁlter (BP 400-440 nm) and a long-wave pass ﬁlter (LP 470 nm). For

photomicroscopy a Kodak Tmax p-3200 (Eastman Kodak Co., Rochester, NY) ﬁlm was

120

employed.

RESULTS AND DISCUSSION

When tobacco cells were rapidly transferred from the culture medium into a minimal
medium, then labelled with chlorotetracycline (50 uM, 30 min), a pronounced ﬂuorescence
intensity was observed. This ﬂuorescence intensity was dependent on the growth stage of the
culture, and their relative growth rate (ﬁgs. 1A and B). Actively growing cells, at about 1 to
3 days after subculture, displayed the highest ﬂuorescence levels, while cells close to or at
stationary phase (5 and 6 days after subculture), manifested ﬂuorescence levels close to that
of autoﬂuorescence. If maintained in culture medium, rather than incubated in minimal ‘
medium, actively growing cells (2 days after subculture), labelled with chlorotetracycline,
also displayed a weak ﬂuorescence signal close to that of autoﬂuorescence.

As to cells transferred to minimal medium, the ﬂuorescence intensity increase was
detected after a rather short period (15 min) of incubation in this type of medium, peaked
after about 1 h, and then gradually declined towards the initial level (ﬁg. 2). To follow this
type of response, chlorotetracycline labelling of cells was reduced to 15 min, and was
performed at the end of each incubation period.

As observed by epiﬂuorescence microscopy on growing cells, exposed to minimal
medium, the enhanced Ca2+ activity was found to be initially localized in a discrete region
near or at the plasma membrane (ﬁg. 3A). However, with longer exposures to minimal

medium (2 1 h), ﬂuorescence was often observed to occur over the entire cellular periphery.

121

 

 

 

 

 

 

 

1e
0
i- A
or». _
€1512
”i
5 _
85’
(I
534..
c
5
i: A
g ~125§
E ~100§
= ﬂ
8 45%
o
3 ~50 e
3 or
c. c
"25 >

a: \ 3

0 Fa r T l '1‘ ‘0 2

0 1 2 3 4 5 6
days after subculture

Fig. 1 A: Cellular Ca2+ activity, reported by chlorotetracycline ﬂuorescence, as a function
of the growth stage of tobacco cultures. Cells in minimal medium were labelled with 50 12M
dye for 30 min. Values are reported after subtraction of autoﬂuorescence. Bars represent SE
(n=3). B: Growth curve and relative growth rates (RGR) of the cultures. The growth curve
can be described by PCV=67.6 - 64.4exp(- 0.00599), 1'2 = 0.98, where PCV is the packed cell
volume (%), and t is the time (days). The relative growth rate was calculated from the

expression RGR =d(PCV)/dt(1/(PCV).

122

 

0 minimal medium
I continuous culture

_ _ 1g m)brefeldin A

...\
x§\\\§

-. removal of
hnhMMA

..a ...a
O N
I I

co
1

.5
l
\

 

\

chlorotetracycline ﬂuorescence
(arbitrary units)
or
1

 

O

 

I l l l I

0 30 60 90 120150180
time (min)

Fig. 2. Changes in Ca2+ activity of growing tobacco cells, as reported by chlorotetracycline
ﬂuorescence, following incubation in a minimal medium, or maintenance in culture medium.
Cells (2 days after subculture) were labelled with the dye (50 HM) during the last 15 min of
the incubation period. Some cells were also treated with 10 11M brefeldin A (30 min before
and throughout medium treatment). To test the reversibility of brefeldin A effects, the
treatment solution was replaced after 1 h with a fresh minimal medium devoid of the
antibiotic. The mean chlorotetracycline ﬂuorescence intensity of cells close to the stationary
phase (5 days after subculture) was about two ﬂuorescence units, regardless of incubation

in minimal medium. Values are reported aﬁer subtraction of autoﬂuorescence. Bars represent

SE (n=3).

123

Fig. 3 Photomicrographs of growing tobacco cells (2 days after subculture) labelled with
chlorotetracycline (50 MM, 15 min). A: Growing cells maintained in culture medium,
displaying lack of localized ﬂuorescence; B: growing cells transferred to and incubated in
minimal medium for 1 h, displaying localized ﬂuorescence. The camera's shutter release

speed (0.5 s) was‘the same for both images.

124

 

 

20 um

Figure 3.

125

By contrast, continuously growing and stationary phase cells lacked any localized
ﬂuorescence (such as in ﬁg. 3B). In minimal medium, only a subpopulation of the cells,
ranging from 7 to 16 % of the total cell population, displayed a transient, localized Ca2+
response. The observed changes in ﬂuorescence intensity (ﬁg. 2) can in part be attributed to
variations in the percentage of cells displaying localized ﬂuorescence. At 15, 60, and 180
min of exposure to minimal medium, 7, 13, and 9 % of the cells displayed a response,
respectively. In this subpopulation, the Ca2+ signal relaxed within a few hours to a value
presumably corresponding to a global equilibrium Ca2+ level, representative of the stationary
phase. Since the cell population is not synchronized, and only a subpopulation displayed a
transient Ca”, the response is expected to be sharper and more intense in a synchronized
culture. In this respect, the tobacco cultivar employed may serve as a good plant model
because these cells can be highly synchronized (Nagata et al 1992).

When cells were growing (2 days after subculture) in culture medium, cell growth
decreased in response to medium changes (ﬁg. 4). This reduction in growth was presumably
temporary, indicative of a new lag phase. Alterations in cellular growth can be expected to
be accompanied by changes in the Golgi-mediated secretory pathway, wherein Caz’ is known
to participate in the regulation of vesicular trafﬁc (Schmid and Damke 1995). Application
of 10 “M brefeldin A to growing cells resulted in further enhancement of cellular Ca2+
activity, following transfer of cells to a minimal medium, but not when these cells were kept
in culture medium (ﬁg. 2). The effects of brefeldin A were found to be highly reversible
(ﬁg. 2). With 10 1.1M brefeldin present, 12, 14, and 16% of the cells responded, respectively.
The lack of detectable local Ca2+ accumulations in tobacco cells growing in culture medium,

both in the presence or absence of brefeldin A (ﬁgs. 2 and 3A), seems to be in accord with

126

 

  

 
  

     
       

        
    

        

      
 

   

 
   
 

   
 

 
   
 

     

  

    

 
    

 
         
 

  
             

      
        

   
  
  
  

  
   

    
 
  

16

14 — s ,,

° 12 — \ ° :5 a
E \ E ,, u y
i R 8 E 3 2
31° ‘ R a 8 8 4:2
3 \ ._ __ .5 Z/sz
3 a Rs " s “f a:
o - \‘:$ n w?
32 4 ‘ RR Z632
“3 E a”:
R.. 1 W
Rm W}
2 — R33? 0 {/3
J R /
0 ;\\\\ :::::':: 7%

 

 

   

abcdef

(D

Fig. 4. Growth of tobacco cells following incubation in different media, measured 24 h after
beginning of treatment. Cells were utilized 2 days after subculture. Treatment were (a) cells
leﬁ undisturbed in the culture medium; (b) the culture medium was replaced by a fresh one;
(c) and (d) cells were incubated for 1, and 3 h, respectively, in a minimal medium, followed
by incubation in fresh culture medium; (e) and (f) cells in culture medium were treated with
10 11M brefeldin A (BFA) continuously, or for 1 h, followed by incubation in fresh culture
medium, respectively; (g) cells in culture medium were grown continuously in the presence
of 0.1 % methanol, as a control for methanol introduced as a solvent for brefeldin A. Values
are reported after subtraction of the initial percentage of packed cell volume (8.1%). Bars

represent SE (n=3)_

127

recent observations on growing green algae where cytosolic [Cay] determinations and

2* gradients

injection of Ca2+ buffers also failed to provide evidence for detectable Ca
(Holzinger et al. 1995).

In growing cells, the Ca” response to medium change was enhanced by brefeldin A
treatment (ﬁg. 2). This did not appear to be an artifact since chlorotetracycline ﬂuorescence
was not affected by brefeldin A in stationary phase or undisturbed growing cells (ﬁg. 2).
Furthermore, the high degree of reversibility of this effect is typical of brefeldin A action on
the Golgi apparatus (Klausner et al. 1992), where considerable vesicular trafﬁc, in part under
Ca2+ control, originates from.

V Although poorly understood, Cay-mediated signalling pathways are assumed to be
operative in plant cells (Bush 1995). Progress has been largely hampered by methodology

and lack of good model systems. As reported here, the tobacco cell line, BY-2, seems to be

a good model system for studies on signalling pathways in plant biology (Nagata et al. 1992).

REFERENCES

Berridge, M.J. 1995. Calcium signalling and cell proliferation. Bioessays 17: 491-500

Bush, D.S. 1995. Calcium regulation in plant cells and its role in signaling. Annu. Rev.
Plant Physiol. Plant Mol. Biol. 46: 95-122

Callaham, D.A., Hepler, PK. 1991. Measurement of free calcium in plant cells. In:
McCormack, J .G., Cobbold, P.H. (eds) Cellular Calcium, A Practical Approach. IRL
Press, Oxford, pp 383-410

Chant, J. 1994. Cell polarity in yeast.Trends Genetics 10:328-333
Clark, G.B., Roux, SJ. 1995. Annexins of plant cells. Plant Physiol. 109 :1133-1139

Cole, N.B., Lippincott-Schwartz, J. 1995. Organization of organelles and membrane trafﬁc
by microtubules. Curr. Opin. Cell Biol. 7:55-64

Herth, W., Reiss, H.D., Hartrnann, E. 1990. Role of calcium ions in tip growth of pollen
tubes and moss protonema cells. In: Heath IB (ed) Tip Growth in Plant and Fungal Cells.
Academic Press, San Diego CA, pp 91-118

Holzinger, A., Callaham, D.A., Hepler, P.K., Meindl, U. 1995. Free calcium in
Micrasterias: local gradients are not detected in growing lobes. Eur. J. Cell Biol. 67:363-
371

Jackson, C.L., Kepes, F. 1994. BFRI, a multicopy suppressor of brefeldin A-induced
lethality, is implicated in secretion and nuclear segregation in Saaccharomyces cerevisae.
Genetics 137: 423-437

Klausner, R.D., Donaldson, J .G., Lippincott-Schwartz, J. 1992. Brefeldin A: Insights into the
control of membrane traffic and organelle structure. J. Cell Biol. 11621071-1080

Merck Index , 11-th edition 1989. Merck Inc, Rahway NJ, Nr 1363

Nagata, T., Nemoto, Y., Hasezawa, S. 1992. Tobacco BY-2 cell line as the “HeLa” cell in
the cell biology of higher plants. Int. Rev. Cytol. 132:1-30

128

129

Robinson, M.S., Watts, C., Zerial, M. 1996. Membrane dynamics in endocytosis. Cell
84:13-21

Rothman, J .E., Orci, L. 1990. Movement of proteins through the Golgi stack: a molecular
dissection of vesicular transport. F ASEB J. 4: 1460-1468

Schmid, S.L., Damke, H. 1995. Coated vesicles: a diversity of form and function. FASEB

J. 921445-1453 '3

Staehelin, L.A., Moore, 1. 1995. The plant golgi apparatus: structure, functional organization
and trafﬁcking mechanisms. Annu. Rev. Plant Physiol. Plant Mol. Biol. 46:261-288

Timmers, A.C.J., Reiss, H.D., Schel, J .H.N. 1991. Digitonin-aided loading of ﬂuo-3 into
embryogenic plant cells. Cell Calcium 12:515-521

Tsien, KY. 1989. Fluorescent indicators of ion concentrations. Meth. Cell Biol. 30:127-156
Zonia, L., Tupy, J. 1995. Lithium-sensitive calcium activity in the germination of apple

(Malus x domestica Borkh.), tobacco (Nicotiana tabacum L.), and potato (Solanum
tuberosum L.). J. Exptl. Bot. 46:973-979

APPENDIX B

CELLULAR ASPECTS OF ALUMINUM TOXICITY IN PLANTS

Haug, A., Vitorello, V. 1996. In: Yasui, M., Ota, K., Strong, M.J., Verity, M.A. (eds).

Mineral and metal neurotoxicology. CRC Press, Boca Raton, F L (in press).

130
Chapter 4

Cellular Aspects of Aluminum
Toxicity in Plants

 

 

 

 

 

CONTENTS

4.1 Introduction........... .... - ...................................................................

4.2 Al(III) Chernisuy .............................................................................................

4.3 Al(III) and Plant Growth ..................................................................... -

4.4 Genetics and Aluminum Tolerance ............................................................................................

4.5 Molecular Features - .. ........................................................................
4.5.1 Al(III) Uptake .............................................................................................
4.5.2 Interiorized Al(III) .........................................................................................................

4.6 Concluding Remarks ..................................................................................................................

Acknowledgment ..................................................................................................................................

References .......................... ........................................................................................

 

4.1 INTRODUCTION

Soil acidity is a major factor limiting plant growth and biomass production. In the heterogeneous
soil system the pH of the aqueous phase plays a fundamental role in properties like the solubility
and hydrolysis of ions. the electric charges of soil colloids, and the uptake of nutrients by plant
roots. As the pH decreases, the solubility of ions like Al(III), Mn(II). and Cu(II) generally increases,
eventually reaching levels toxic for many crap plants.

Regarding metal toxicity. that caused by aluminum is by far the major factor for plant growth
limitation on naturally acidic soils (pH <5.0). For example, 70% of the soils in tropical America
are affected by aluminum toxicity. Industrial activities and fertilizer usage have also contributed to
the acidiﬁcation and thus to the levels of free aluminum in terrestrial and aquatic environments
previously not afﬂicted with aluminum toxicity."3

Despite the long history of studies on aluminum-related limitation of plant growth. there still
exists a dearth of knowledge concerning the molecular basis for aluminum toxicity and the impact
of low pH on plant root biochemistry."3 In this chapter we are therefore focusing our attention
primarily on molecular aspects of aluminum toxicity in plant cells rather than on phenomenological
descriptions of the anatomy/physiology of plants under aluminum stress. Regarding mechanisms
of aluminum toxicity in plant cells. information will also be provided, and conclusions drawn, from
appropriate experiments conducted on mammalian cells. We hope that such a strategy will stimulate
interest in studies directed towards elucidating primary mechanisms of aluminum toxicity mani-
fested in both plant and mammalian cells.

 

08493-7664-5/96601XkSJO
c 1996 by crtc Press Inc. 35

Al(III) IN PLANTS 13 l 36

42 Al(III) CHEMISTRY

In soils. aluminum is predominantly found as aluminosilicate and clay minerals. In the environment,
chemical reactions of aluminum are mostly occurring in water. For these reasons, and also because
the water content of a typical biological cell is approximately 80% of the cellular weight. it is
necessary to understand reactions of aluminum in aqueous solution. Concerning details about
aluminum properties (thermodynamic data, kinetics. chelation) in minerals and in solution, infor-
mative articles are available.”

In short, in aqueous solution the charged Al(III) exists as a hexahydrated species. Al(H20)63‘,
at low pH. As the pH increases, in particular in the range from 5 to 6.2, the hydrated metal ion
has a strong tendency to form various hydrolytic species. A series of polymeric aluminum species
is also generated, particularly at higher pH values. Besides the pH value and Al(III) concentration.
the extent of hydrolysis is strongly dependent on the presence of chelating ligands. The hydrated
Al(III) ion has a considerable aﬂinity to oxygen-containing negatively charged ligands such as
carboxyl, hydroxyl, carbonyl, and phosphate groups associated with organic molecules. Apart from
binding to hydroxyl, strong inorganic complexes are formed with fluoride, sulfate. and phosphate
ligands. The mode of metal uptake across the plant plasma membrane may depend upon a chelate
formed by Al(III). Following interiorization, the movement of Al(III) to biochemically critical
targets in the cytosol and/or membranes is presumably also dependent upon the types of ligands
available. Participation of Al(III) in chemical reactions is in part governed by the slow exchange
rate of hydration water. a consequence of the small ionic radius (0.053 nm) of Al(III).

4.3 Al(III) AND PLANT GROWTH

A pH value of 5.0 is the approximate threshold below which aluminum toxicity becomes manifest
in whole plants. This threshold is rather ill-deﬁned since it also depends on factors such as the
presence of ions Ca(II) and phosphate, chelators, and the plant genotype.L3 In a “typical" acidic
soil, the pH is around 4.5 and the concentration of mobile aluminum may vary between 10 and
40 M (1 ppm Al(III) equals 37 1.1M). However, in acidic forest ecosystems displaying forest decline.
pH values around 3.8 and mobile aluminum levels of about 1 mM have been reported.. Toxicity
has been tentatively ascribed to the hexahydrated, trivalent Al(III) species, rather than to its
hydrolysis products with lower electric charge.“3

Al(III)-stressed plants show inhibition of growth. mainly that of the roots; the root apex is
apparently a critical tissue.9 Al(III)-related derangements of root metabolism reportedly impact
transport of essential nutrients between roots and plant tissues above ground. thus reducing plant
biomass accumulation."3 Symptoms on roots include root thickening and discoloration. At the
cellular level, perturbations of plasma membrane structure. vacuolation of root cap cells, and
changes in nuclear structure are observable.3~9 Since primary target(s) of Al(III) action on cells have
not been clearly identiﬁed, some or all of these symptoms may reﬂect secondary injuries to the cell.

Regarding growth of Al(III)-stressed plants. considerable attention has been paid to interactions
between Al(III) and Ca(II). Not attributable to ionic strength effects, elevated Ca(II) or Mg(II)
levels in the root medium have been found to ameliorate the effects of toxic Al(III) on plant growth.1
An atn'active feature of these observations is that an increase in soil acidity is generally accompanied
by a loss of exchangeable cations like Ca(II) and Mg(II).‘ Findings derived from these types of
amelioration studies led to the hypothesis that toxic Al(III) ions are displacing Ca(Il) from the
membrane surface, thus impairing the functioning of surface ligands and transport channels critical
for proper root growth."3 Recent experiments on wheat discredited this type of Ca(II)-displacement
hypothesis for explaining Al(III)-related rhizotoxicity.lo Rather than unspecifically residing at the
cell surface, the initial Al(III)-induced lesion seems to be associated with regulatory processes
which are in part dependent upon signal-transducing elements of the plasma membrane, at least in
mammalian cells.“-12

Al(III) IN PLANTS 132 37

4.4 GENETICS AND ALUMINUM TOLERANCE

Among plant species and genotypes of a given species there exist wide variations regarding tolerance
to Al(IIl).‘ In the acidic cerrado region of central Brazil, for example, Al(III)-accumulating tree
species have been found which apparently even require Al(III) in the growth medium for normal
growth.13 However, the biochemical and genetic control mechanisms underlying tolerance to Al(III)
are poorly understood. At this time, hard evidence is lacking whether tolerance to Al(III) is
associated with processes outside the cell (exclusion) and/or inside the plant cell (internal detoxi-
ﬁcation). Given this dearth of knowledge, and ﬁndings that rhizotoxicity seems to be a pronounced
feature of toxic Al(III) on plants,1 parameters like root length or root elongation of Al(III)-stressed
plants are generally selected as indicators when studying the inheritance of Al(III) tolerance in crop
plants. Based on these selection techniques, there is evidence that a single incomplete dominant
gene is responsible for the inheritance of aluminum tolerance in wheat." Al(III) resistance of a
tobacco variant was also attributed to a single dominant gene mutation when tobacco cells were
selected for Al(III) resistance in cell cultures.” Furthermore. by selecting biochemical parameters
as indicators for Al(III) sensitivity, experiments on two wheat varieties suggested that the effects
of Al(III) on K(I) uptake are not primary factors associated with sensitivity towards toxic Al(III)."
Despite the paucity of information on the molecular basis for Al(III) resistance, signiﬁcant progress
has been achieved. solely relying on conventional plant breeding programs. in selecting and
releasing Al(III)-tolerant plants for crop production on acidic soils.‘

4.5 MOLECULAR FEATURES
4.5.1 Al(III) Uptake Across The plasma membrane

Numerous studies have been conducted concerning Al(III) uptake and distribution in intact plants
or in plant tissue!“ However. considerable ambiguity remains concerning the relationships between
the measured uptake kinetics and putative Al(III)-binding compartments. Indeed, few data are
available on the effects of Al(III) at the cellular and molecular levels, such as mode(s) of Al(III)
uptake across the plasma membrane, role of metal speciation in membrane traversal, and initial
targets critical for cellular physiology. Outside the plant cell, in the Donnan free space of the root
cell, major aluminum-binding sites are negatively charged carboxyl groups of pectic substances."3
However, we believe that these surface-exposed. unspeciﬁc binding sites can be dismissed as
primary aluminum targets. Rather, likely primary targets for aluminum-triggered rhizotoxicity are
expected to reside in the cytosol and/or in the plasma membrane of the plant cell,“ i.e., Al(III)
must be internalized by the cell. This notion is also supported by ﬁndings on mammalian cells,
where key elements of phosphoinositide-associated signal u'ansduction pathways. coupled with
Ca(II) regulation, were implicated as potential primary targets for toxic Al(III) ions.“-‘2

In oak root cortex cells, application of Al(III) (pH 4; 37 11M to 3.7 mM) enhanced membrane
permeability to nonelectrolytes (methyl urea. urea), and lowered permeability to water. '9 The plasma
membrane was also implicated as a putative Al(III) target in studies on Amaranthus protoplasts,
where “Ca“ uptake was inhibited within minutes by Al(III) application (pH 4.5; 10 and 50 1.11%)”
These kinds of permeability changes are presumably related to pronounced Al(III)-triggered mem-
brane lipid phase changes, in vivo and in membrane vesicles isolated from Thermoplasma acido-
philum (pH 2).2 Moreover. compared with nonstressed mycelia, the proportion of less-ordered
membrane domains was decreased when spin-labeled fungal mycelia were exposed to Al(III) stress
(1 to 10 mM).21 Al(III)-induced changes in lipid packing were also observed in response to Al(III)-
binding (20 M) to the erythrocyte membrane.22 The phosphodiester group was found to be the
primary binding site of Al(III) with lipids.23 Furthermore, based on ﬁndings that Al(III) neutralizes
the surface charge of phosphatidyl choline vesicles. this type of interaction was implicated in the
nonspeciﬁc. Al(III)-related inhibition of cation uptake by root cells.“ In the presence of Al(III) (20
to 50 1.1M), changes in the membrane surface charge are generated by displacing virtually all bivalent

as 133

cations, such as Ca(II), present at bulk concentrations as high as 1 mM.15 To reiterate, Al(III)-
related Ca(II) displacement at the membrane surface does not seem to be the initial lesion for
rhizotoxicity of plants.” Nevertheless, it is reasonable to assume that nonspeciﬁc Al(III) binding
to the membrane surface. followed by traversal of the membrane, are sequential events required
for Al(III) interiorization by the cell. Membrane traversal, perhaps involving some sort of carrier?‘5
may in part depend on the physical state of the membrane.

4.5.2 Interiorized Al(III)

Once interioriwd, Al(III) may interact with membrane-bound, cytosolic, and nuclear constituents
of the cell. Given the slow solvent exchange rate of Al(III)] it must be kept in mind, however, that
initially established interactions with certain sites may gradually change until a ﬁnal equilibrium
has been reached where Al(III) is irreversibly coordinated to a ligand, perhaps different from the
initial one.

Forming the interface between the cell and its microenvironment. the plasma membrane detects
external stimuli at the surface, processes these signals across the membrane, then converts them
into a cascade of reactions involving second messengers like Ca(II). In animal cells, chief among
cellular communication channels is the phosphoinositide signaling pathway. A major product of
this pathway is a second messenger, inositol-1.4,S-triphosphate (1P3) which, in turn, plays a crucial
role in mobilizing intracellular Ca(II).” In plants, the elements of this pathway are present but
evidence is lacking as to the Operation of this system in response to a speciﬁc stimulus.” Regarding
aluminum toxicity, short-terrn studies on neuroblastoma cells demonstrated that application of
Al(III) (pH 6.8, 2 W to 1 mM) inhibited intracellular Ca(II) mobilization which, in turn. coincided
with a decrease in inositol phosphate production. Being crucial for 1?, formation. phosphatidyli-
nositol-4,5-diphosphate-speciﬁc phospholipase C probably harbors an Al(III) target. A second
Al(III) target seems to reside on molecular switching elements of the phosphoinositide signaling
pathway, viz.. on Mg(II)/guanine nucleotide binding (Gp) proteins. It was suggested that Al(III) is
occupying a site at or near the GTP binding center on the Gp protein."-12 Besides direct interactions
with proteins, Al(III) may perturb the protein's lipid milieu, especially at higher.Al(III) For example.
the hydrolyzing efﬁcacy of phospholipases is in part dependent on the physical state of their lipid
environment.‘2 Evidently further studies are necessary to identify Al(III) targets on elements of the
signal transduction pathway in eukaryotic cells.

Concerning Al(III) interactions with membrane-bound enzymes. studies on barley root plasma
membranes indicated that Al(III) ions are capable of inhibiting a membrane-associated Ca(II)-
ATPase activity. The Al(III)-induced lower activity could in part be restored by applying massive
amounts (50 mM) of monovalent cations. As a result of increased ionic strength. Al(III) was
presumably displaced from lipids in the boundary lipid region of the enzyme, thus restoring the
motional requirements of the intact protein.29 There is also the possibility that ATPase activity is
diminished resulting from the formation of an Al(III)-ATP chelate when millimolar AlCl, levels
are employed.30 Furthermore. applying Al(III) to vesicles enriched in right-side-out plasma mem-
branes isolated from maize roots, Al(III) (pH 6.0; 40 11M) inhibited NADH-dependent electron
transfer processes," i.e., reactions generally associated with the uptake of iron by the plant root.32

Following internalization by the wheat root cell, Al(III) was distributed in the cytosolic fraction
(generally 55% of total Al[III]), nuclear fraction (around 30%), and the mitochondrial fraction
(around 13%).33 Interestingly, a roughly similar distribution of Al(III) was found in neuroblastoma
cells."5 In the cytosol there exists a multitude of ligands capable of binding Al(III), e.g., ATP,30
citric acid,“-7 various organic acids.’ and acidic polypeptides.2 However, evidence is lacking whether
intracellular chelators are instrumental in intrinsic Al(III) tolerance mechanisms of plant cells. On
the other hand, like in mammalian cells."5 this large chelator pool may contribute to the high and
rapid accumulation of Al(III) by plant cells. By maintaining a low intracellular-free [Al(III)] the
equilibrium is shifted in favor of Al(III) inﬂux.

Al(III) 1N PLANTS 134 39

Al(III) ions reportedly also interact with mitochondrial membranes. Experiments on isolated
mitochondria from untreated wheat roots indicated that Al(III) ions (pH 7; 12 to 75 11M) interacted
with the mitochondrial respiratory pathway, presumably by interfering with the oxidation of sub-
strates that donate electrons to Complexes I and 11.34 Also involved in energy-generating processes
are NADP-dependent dehydrogenases. which were strongly inhibited by Al(III) (pH 6.85; 2 11M
A1C13) with respect to the substrate. isocitrate.”

In the cytosol, interiorized Al(III) may bind to calmodulin.2 a highly conserved protein (about
17 kDa) crucial for Ca(II)-mediated regulation in eukaryotic cells. Responding to altered intracel-
lular Ca(II) levels. e.g., triggered via signaling pathways, this protein is activated by undergoing
conformational changes generated by speciﬁcally binding four Ca(II) per protein. The activated
protein can then associate with and stimulate certain calmodulin-dependent partner enzymes."
Concerning Al(III)-related alterations of calmodulin, only in vitro experiments are available. Pre-
sumably binding only weakly to calmodulin,” the protein undergoes helix-coil transitions in the
presence of Al(III), thus changing the protein's internal dynamics and its efﬁcacy in correctly
docking with partner proteins.3“‘° Al(III)-induced changes on calmodulin are instrumental in
malfunctions of calmodulin-dependent reactions, e.g., that of 3’.5’-cyclic nucleotide phosphodi-
esterase, certain plant AT'Pases,2 and NAD kinase activity in wheat root tips.“ The latter region
harbors high levels of calmodulin.‘2 and has been implicated as being a site for Al(III) injury? At
this time evidence is lacking whether Al(III) inactivates calmodulin in a cell. Abundantly present,
cytosolic chelators are possibly protecting this important regulatory protein from Al(III) injury.‘3

The nucleus may be an important sink for interiorized Al(III), especially in cells under long-
term Al(III) stress. For example, when onion root tips were treated with AlCl, (pH 5.5, 1 mM,
several hours), cell division was inhibited.“ Various ﬁndings indicate that Al(III) accumulates on
DNA,“ where the phosphodiester group is apparently a binding Site.45 Under cytosolic conditions,
Al(III) binding to the DNA is presumably weak.“5 Given the condensed state of DNA in the nucleus
of a cell, interstrand DNA cross-linking via Al(III) bridges is highly probable."I Such cross-linking
would be consistent with observations that Al(III)-treated pea chromatin is more heat stable.“
Moreover, Al(III)-induced changes in DNA structure might be expected to disrupt the physiolog-
ical processes of gene expression. It is noteworthy that the mRNA levels of calmodulin and tubulin
were considerably diminished in Al(III)-treated rabbit brains.“ Compared with Al(III)-induced
derangements of membrane structure and signaling processes (detectable within minutes),”-‘3
Al(III) interactions at the DNA, and numerous toxicity responses (growth, cell secretion, callose
formation, etc.) are long-term effects.‘-3~3-‘9 Whether some of these long-term responses are directly
caused by Al(III) is questionable.

4.6 CONCLUDING REMARKS

Only rudimentary information is currently available on primary targets and the initial phases in the
temporal progression of aluminum toxicity, both in plant and animal cells. What. at present. appears
to be a multitude of Al(III) effects in cells, may eventually originate from a single (or a few) primary
target(s) critical for the temporal development of the aluminum toxicity syndrome. Given the
signiﬁcance of an intact plasma membrane and of functional communication pathways for cellular
survival, any Al(III)-uiggered change therein is expected to have severe repercussions on the cell's
ability to respond to extemal/internal signals. By focusing on identifying initial key interaction
sites for toxic Al(III), we may also contribute to. and in fact simplify, our endeavors of deve10ping
Al(III)-resistant varieties through biotechnological means.

ACKNOWLEDGMENT

V.V. was supported in part by a fellowship, grant Nr. 200892/91-6, from the Conselho Nacional
de Desenvolvimento Cientiﬁco e Tecnologico (Brasil).

40

BS

REFERENCES

I.

la)

sev-

IO.

11.

12.

13.

14.

16.

I7.

18.

19.

20.

21.

22.

23.

24.
25.
26.
27.
28.
29.
30.

31.

Foy CD. Chaney RL. and White MC: The physiology of metal toxicity in plants. Annu. Rev. Plant Physiol.. 1978;
29: 511-566.

. Haug A: Molecular aspects of aluminum toxicity. CRC Crit. Rev. Plant Sci.. 1984; 1: 345-373.
. Taylor GJ: The physiology of aluminum phytotoxicity. In Sigel I-I (ed): Metal Ions in Biological Systems. Vol. 24.

New York, Marcel Dekker. 1988. pp 123-163.

. Sposito G (ed): The Environmental Chemistry of Aluminum. Boca Raton, CRC Press. 1989.

Akitt JW: Multinuclcar studies of aluminum compounds. Prog. Nucl. Magn. Reson. Spectrosc. 1989: 21: 1-149.
Martin RB: The chemistry of aluminum as related to biology and medicine. Clin. Chem. 1986: 32: 1797-1806.
Orvig C: The aqueous coordination chemistry of aluminum. In Robinson GH (ed): Coordination Chemistry of
Aluminum. New York. VCH Publishers. 1993. pp 85-121.

. Godbold Dl.. Fritz E. and Hilttermann A: Aluminum toxicity and forest decline. Proc. Natl. Acad Sci. U.S.A.. I988;

85: 3888-3892.

. Bennet RJ and Breen CM: Aluminium toxicity: Towards an understanding of how plant roots react to the physical

environment. In Randall PJ. Delhaize E. Richards RA. and Munns R (eds): Genetic Aspects of Plant Mineral Nutrition.
Dordrecht. Kluwer Academic. 1993. pp 103-116.

Kinraide TB. Ryan PR. and Kochian LV: AIP-Ca" interactions in aluminum rhizotoxicity. 11. Evaluating the Ca"-
displacement hypothesis. Planta. I994; 192: 104-109. '

Shi B, Chou K. and Haug A: Aluminium impacts elements of the phosphoinositide signalling pathway in neuroblas-
toma cells. Mol. Cell. Biochem.. 1993: 121: 109-118.

Haug A. Shi B. and Vitorello V: Aluminum interaction with phosphoinositide-associated signal transduction. Arch.
Toxicol. 1994; 68: 1-7.

Haridasan M. and de Araujo G M: Alurninium-accurnulating species in two forest communities in the cerrado region
of central Brazil. Forest Ecol. Manage. 1988; 24: 15-26.

Wheeler DM. Edrneades DC, Christie RA. and Gardner R: Comparison of techniques for determining the effect of
aluminium on the growth of. and the inheritance of aluminium tolerance in wheat. In Randall PJ. Delhaize E. Richards
RA. and Munns R (eds): Generic Aspects of Plant Mineral Nutrition. Dordrecht. Kluwer Academic. 1993. pp 9-16.

. Conner AJ. and Meredith CP: Strategies for the selection and characterization of aluminum-resistant variants from

cell cultures of Nicotiana plumbaginifolia. Planta, 1985; 166: 466-473.

Pettersson S, and Strid H: Effects of aluminium on growth and kinetics of K’('°Rb’) uptake in two cultivars of wheat
with different sensitivity to aluminium. Physiol. Plant. 1989; 76: 255-261.

Zhang G. and Taylor GJ: Effects of biological inhibitors on kinetics of aluminium uptake by excised roots and
puriﬁed cell wall material of aluminium-tolerant and aluminium-sensitive cultivars of Triticum aesri'vum L. J. Plant
Physiol.. 1991: 138: 533-539.

Haug A. and Shi B: Biochemical basis of aluminium tolerance in plant cells. In Wright RJ. Baligar VC. and Murrmann
RP (eds): Plant-soil interactions at low pH. Dordrecht. Kluwer Academic. 1991. pp 839-850.

Zhao XJ. Sucoff E. and Stadelmann El: Al" and Ca” alteration of membrane permeability of Quercus rubra root
cortical cells. Plant Physiol., 1987; 83: 159-162.

Haug and Vitorellols

Rengel Z. and Elliott DC: Aluminium inhibits "Ca” uptake by Amaranthus protOpIasts. Biochem. Physiol. Pﬂanz.
1992: 188: 177-186

Zel J. Svetek J. Cine H. and Schara M: Effects of aluminum on membrane fluidity of the mycorrhizal fungus Amanita
mascaria. Physiol. Plant. 1992: 89: 172-176.

Weis C. and Haug A: Aluminum-altered membrane dynamics in human red blood cell white ghosts. Thromb. Res.
1989: 54: 141-149.

Hauser H. and Phillips MC: Interactions of the polar groups of phospholipid bilayer membranes. In Cadenhead DA.
and Danielli JF (eds): Progress in Surface and Membrane Science. Vol. 13. New York. Academic Press. 1979. pp
297-413.

Akeson MA. Munns DN. and Burau RG: Adsorption of A13’to phosphatidylcholine vesicles. Biochim. Biophys. Acta.
1989: 986: 33—40.

Deleers M: Cationic atmosphere and cation competition binding at negatively charged membranes: Pathological
implications of aluminum. Res. Commun. Chem. Pathol. Pharmacol. 1985; 49: 277-294.

Shi B. and Haug A: Aluminum uptake by neuroblastoma cells. J. Neurochem.. 1990: 55: 551-558.

Berridge MJ: Inositol triphosphate and calcium signalling. Nature, 1993: 361: 315-325.

Cote' GG. and Crain RC: Biochemistry of phosphoinositidcs. Annu. Rev. Plant Physiol.. 1993: 44: 333-356.
Caldwell CR. and Haug A: Divalent cation inhibition of barley root plasma membrane-bound Gait-ATPase activity
and its reversal by monovalent cations. Physiol. Plant. 1982: 54: 112-118.

Panchalingam K. Sachedina S. Pettegrew JW. and Glonek T: Al-ATP as an intracellular carrier of Al(III) ion. Int. J.
Biochem.. 1991; 23: 1453-1469.

Loper M. Brauer D. Patterson D. and Tu 81: Aluminum inhibition of NADI-l-linkcd electron transfer by com root
plasma membranes. J. Plant Nutr.. 1993: 16: 507-514.

Al(III) IN PLANTS 136 41

32

33.
34.
35.

36.
37.

38.
39.
40.
41.
42.
43.

44.

45.
46.

47.
48.

49.

Holden Ml, Luster DO. and Chaney RL: Enzymatic iron reduction at the root plasma membrane: Partial puriﬁcation
of the NADI-l-Fe chelate reductase. 1n Manthey JA. Crowley DE. and Luster DG (eds): Biochemistry of Metal
Micronutrients in the Rhizosphene. Boca Raton, Lewis Publishers. 1994. pp 285-294.

Nieda'ela G. and Aniol A: Subcellular distribution of aluminium in wheat roots. Acta Biochim. Pol.. 1983: 30: 99-105.
de Lima ML. and Copeland L: The eﬁect of aluminium on respiration of wheat roots. Physiol. Plant. 1994; 90: 51-58.
Yoshino M. Yamada Y. and Murakami K: Inhibition by aluminum ion of NAD- and NADP-dependent isocitrate
dehydrogenases from yeast. Int. J. Biochem. 1992: 24: 1615—1618.

Cohen P. and Klee CB (eds): Calmodulin. Amsterdam. Elsevier. 1988.

Martin RB: Bioinorganic chemistry of aluminum. In Sigel H. and Sigel A (eds): Metal Ions in Biological Systems.
Vol 24. New York. Marcel Dekker. 1988. pp 1-57.

Yuan S. and Haug A: Frictional resistance to motions of bimane-labelled spinach calmodulin in response to ligand
binding. FEBS Lett.. 1988: 234: 218-223.

Weis C. and Haug A: Aluminum-altered conformational changes in calmodulin alter the dynamics of interaction with
rnelittin. Arch. Biochem. Biophys.. 1987: 254: 304-312.

l-Iaug A. and Vitorello V: Aluminium coordination to calmodulin. thermodynamic and kinetic aspects. Coord Chem.
Rev.. in press.

Slaski II: Effect of aluminium on calmodulin-dependent and calmodulin-independent NAD kinase activity in wheat
root tips. J. Plant Physiol.. 1989: 133: 696-701.

Allan E. and Trewavas A: Quantitative changes in calmodulin and NAD kinase during early cell develoth in the
root apex of Pisum sativum L Planta. 1985: 165: 493-501.

Suhayda CG. and I-Iaug A: Citrate chelation as a potential mechanism against aluminum toxicity in cells: the role
of calmodulin. Can J. Biochem. Cell Biol. 1985: 63: 1167-1175.

Matsurnoto 1-1: Biochemical mechanism of the toxicity of aluminium and the sequestration of aluminium in plant
cells. In Wright RJ. Baligar VC. and Murrrnann RP (eds): PlantoSoil Interactions at Low pH. Dordrecht. Kluwer
Academic. 1991. pp 825-838. '

Dyrssen D. Haraldsson C. Nyberg E. and Wedborg M: Complexation of aluminum with DNA. J. Inorg. Biochem..

" 1987; 29: 67-75.

Martin RB: Aluminium speciation in biology. 1n Aluminium in Biology and Medicine (Ciba Foundation Symposium
no. 169). Chichester. John Wiley 8: Sons. 1992. pp 5-25. .

Karlik SJ. and Eichhom GL: Polynucleotide crosslinking by aluminum. J. Inorg. Biochem.. 1989: 37: 259—269.
Crapper Mclachlan DR: Aluminum neurotoxicity: Criteria for ”signing a role in Alzheimer's disease. In Lewis TE
(ed): Environmental Chemistry and Toxicology of Aluminum. Boca Raton, Lewis Publishers. 1989. pp 299-315.
Basu A. Basu U. and Taylor GI: Induction of microsornal membrane proteins in roors of an alurninumoresistant
cultivar of Triticum aestivum 1.. under conditions of aluminum stress. Plant Physiol.. 1994: 104: 1007-1013.

APPENDIX C

ALUMINUM INTERACTION WITH PHOSPHOINOSITIDE-ASSOCIATED
SIGNAL TRANSDUCTION

Haug, A., Shi, B., Vitorello, V. 1994. Arch. Toxicol. 68:1-7.

Arch Toxicol (1994) 68: l-7

Review article

137

 

ﬁhcology

© Springer-Verlag 1994

 

Aluminum interaction with phosphoinositide-associated signal

transduction

Alfred Haug, Biao Shi, and Victor Vitorello

Department of Microbiology and Pesticide Research Center. Michigan State University. East Lansing. MI 48824. USA

Received 14 June l993/Accepted 15 September 1993

Abstract. Concerning molecular and cellular mechanisms
of aluminum toxicity. recent studies support the hypothesis
that interactions of aluminum ions with elements of signal
transduction pathways are apparently primary events in
cells. In the case of the phosphoinositide-associated sig-
nalling pathway of neuroblastoma cells. guanine nucleo-
tide-binding proteins (G proteins) and a phosphatidylino-
sitol-4.5-diphosphate (PIP2)-speciﬁc phospholipase C are
probable interaction sites for inhibitory actions of alumi-
num ions. Following interiorization of aluminum by the
cell. metal interactions decrease the accumulation of ino-
sitol phosphates. especially that of inositol-1.4.5-triphos-
phate (1P3). concomitant with derangements of intracellular
Ca1+ homeostasis. In the presence of high concentrations of
Calt. formation of 1?: is also diminished in aluminum-
pretreated cells. presumably involving a process not re-
quiring Mgzt-dependent G proteins. At higher aluminum
doses. metal-induced changes in the lipid milieu of the
membrane-bound phospholipase may play a role. These
types of primary interactions of aluminum ions with ele-
ments of cellular communication channels are probably
crucial in the manifestation of the multifacetted aluminum
toxicity syndrome. If present as a phosphate-like fluoro-
aluminate. a stimulatory role of aluminum ions is displayed
in G protein-coupled transmembrane signalling.

Key words: Aluminum toxicity - Calcium regulation -
G proteins - Neuroblastoma - Phosphoinositides - Phos-
pholipase C - Signal transduction

 

Introduction

Over the past decade. concern has intensified regarding the
toxic effects of aluminum on man (Ganrot 1986; Marquis
1989; Schmidt et al. 1991). In the environment. aluminum
is recognized as a serious global problem since vast areas of

Correspondence to: A. Haug

the world suffer from natural soil acidity which in turn
mobilizes aluminum for uptake by crop plants. Moreover.
industrial development and usage of fertilizers have con- .
tributed to the acidiﬁcation of agricultural areas and aquatic
habitats. Aluminum can therefore enter the food chain
(Haug 1984; Lewis 1989; Taylor 1991).

Although considerable evidence has emerged regarding
clinical aspects (Wills and Savory 1985: Nebelter and Co-
burn 1986). only scant information is available on specific
cellular and molecular mechanisms associated with alu-
minum toxicity. Nevertheless. there are indications that
aluminum interferes with signalling pathways in cells
(Woods et al. 1990: Miller et al. 1989; Schofl et al. 1990:
Shi and l-Iaug 1992).

Indeed. the hypothesis can be formulated that elements
of signalling pathways. like that associated with phos-
phoinositides, harbor a primary target for aluminum inter-
action(s) with a cell. Depending on conditions. these in-
teractions may either stimulate or inhibit the signalling
pathway. It is therefore our objective to provide a critical
review of cellular and molecular aspects pertinent to the
above hypothesis. To set the stage for such a focussed
discussion. we first summarize some background infon'na-
tion on the biochemistry of the interrelated Ca2+lphos-
phoinositide signalling pathway. the best-known commu—
nication channel.

Signal transduction along the Gaze/phosphoinositide
pathway

Cells require communication with their environment. By
carrying messages along signalling pathways. input signals
are transformed into specific biological responses. such as
changes in intracellular Ca2+ concentrations (Rasmussen
1990). In eukaryoes. chief among cellular communication
channels is the phosphoinositide signalling pathway
(Fig. 1). Following binding of a primary messenger (hor-
mone. neurotransmitter) to a receptor at the cell surface.
second messengers. viz. inositol-1.4.5otriphosphate (1P3)
and diacylglycerol. are produced through hydrolysis of

138

L)

 

 

   

 

r. 0 W extracellular
"W matrix
lrocetitt'ir
"-r ~ ' ". ~r wk w t“ ._. a "2 DAG
- " ’ . ' w L. f‘ . 1 — Q‘
a- "1'9 I a
i ... ~ ‘ Osman ac: ‘
g . (’3’ ... _ -- an . .. :25:
i ' .W -..Jxa. '. J .....I-a «L... ---...1. ' I I :23:
. 3%?1
; El ...=~ @ al,—.e- £25:
. v' a
i ’3 'P P 751’.-
l ;~::":
r 2. f‘ 2:-
\ “m, I“ .25.“.
I retictm 9‘3: :
cellular .-- .

 

 

Fig. 1. Simpliﬁed view of signal transduction along the Cab/phos-
phoinositide pathway. The shaded area indicates signal transduction
elements which are presumably impacted by aluminum ions. Signal
input occurs by binding of a primary messenger to a receptor at the
plasma membrane surface. thus triggering transmission of a signal to a
heterotrimeric G, protein. The Ca subunit dissociates from the Gay
subunits in exchange of GDP for GTP. The Go subunit activates a
phosphoinositide (PIP2)-speciﬁc phospholipase C (PLC). thus produ-

membrane-associated phosphatidylinositol-4.5-diphosphate
(PIPz) by a speciﬁc phospholipase C (Fig. l). 11’: plays a
key role in mobilizing intracellular Ca2+ levels (Berridge
1993). while diacylglycerol activates protein kinase C
which in turn is involved in protein phosphorylation pro-
cesses. Following an array of reactions. 11’: and dia-
cylglycerol are metabolized and PIP: is resynthesized. thus
closing the phosphatidylinositol cycle (Berridge 1987;
Rana and Hokin I990: Majerus l992).

Heterotrimeric Gp proteins (M around 100 kDa) serve as
intermediaries (Fig. 1) that link cell surface receptors to
phospholipase C (Bimbaumer et al. 1990: Hall 1992). In
response to the activation of an appropriate receptor. the Gp
protein undergoes a conformational change which facil-
itates the Mgzt-dependent exchange of bound GDP for
GTP. GTP binding to the alpha subunit reduces the afﬁnity
to the receptor. Following dissociation from the beta/
gamma subunits. the Mgzhactivated alpha subunit (Fig. I)
interacts with the speciﬁc effector. phospholipase C. Signal
transduction is terminated by the hydrolysis of bound GTP
through the GTPase activity of the alpha subunit. Sub-
sequent reassociation of this subunit with the beta/gamma
subunits leads to reconstitution of the G, protein (Freiss-
muth et a1. 1989).

Fluoroaluminate-Induced phosphoinositide metabolism
and Cu2t mobilization

Concerning signal transduction. fluoroaluminates have be-
come valuable tools in elucidating functional aspects of
signalling pathways. For example. in vitro experiments

cing two second messengers as signal output. viz. 11’: and diacylglyo
cerol (DAG). The off-phase of signal transmission is initiated by GTP
hydrolysis allowing reassembly of G, subunits. Through a series of
reactions. e.g. dephosphorylations leading to IP: and 1P. [Pi and
diacylglycerol are metabolized and PIP: is resynmesized via PI and
PIP (Berridge I987). PKC. protein kinase C: protein-P. phosphorylated
protein

demonstrated that application of millimolar fluoride. in the
form of NaF. activated regulatory uimeric G proteins.
which in turn are known to participate in stimulating G
protein-coupled transmembrane signalling. This activation
process has been shown to be dependent on the presence of
trace amounts of aluminum. leading to the formation of
ﬂuoroaluminates (Sternweis and Gilman 1982). Indeed.
under ordinary experimental conditions. aluminum con-
tamination from chemicals and glassware was sufficient for
this ﬂuoride-related activation. Even when precautions
against contamination were taken. aluminum levels close to
1 M were found in neuroblastoma tissue culture cells (Shi
and Haug 1990).

The fluoroaluminate species which activates G proteins
is not known unambiguously. With regard to thermo-
dynamic equilibria. aluminum apparently formed soluble
complexes A1133“. where x = 1-6. in aqueous solutions
(Savchenko and Tananaev 1959). G proteins were maxi-
mally activated in the presence of millimolar NaF con-
centrations. Under these conditions. the resulting major
species. AIFI. was proposed to act as a pseudophosphate by
binding to the G protein adjacent to GDP. in a tetrahedral
conﬁguration. However. unlike the gamma-phosphate of
GTP. this fluoroaluminate bound to GDP does not undergo
hydrolysis and therefore the catalytic center of the protein
is locked into the activated conﬁguration (Chabre 1990).

In support of the notion that certain ﬂuoroaluminate
species behave as a pseudophosphate are studies showing
that ﬂuoroaluminates mimicked phosphate binding to the
Cazr-free conformation of sarcoplasmic reticulum ATPase.
In the presence of Mg“. phosphorylation by phosphate (Pi)

was inhibited as a result of AlFI binding. Fluoroaluminate
also acted as an analogue of the gamma-phosphate of ATP
in an ADP/AIFI complex of the Ca2t-bound form of the
enzyme. Bound on the external high afﬁnity sites. the Ca3+
ions became occluded upon binding of an ADP/ﬂuoro-
aluminate complex (T roullier et al. 1992).

Observed cellular responses to fluoroaluminates are in
accord with the action that these compounds are interacting
with G proteins. Treatment of hepatocytes with NaF mi-
micked the effects of Cab-mobilizing hormones. Cytosolic
free Ca2+ levels rose as a result of G protein-mediated PIP:
hydrolysis and release of 1P3 (Blackmore et al. 1985).
Following treatment of rat parotid acinarcells with milli-
molar concentrations of NaF. the observed amylase release
was mediated by phosphoinositide degradation and in-
tracellular Ca2+ mobilization. even in the absence of ex-
tracellular Cazt Sustained increases in intracellular Ca1+
levels probably resulted from enhanced entry of extra-
cellular Cazt. Trace amounts of aluminum probably suf-
ﬁced to form the active chelate AIFI (Tojyo et al. 1991).
Moreover. by stimulating human umbilical vein endothelial
cells with ﬂuoroaluminates. inositol phosphates were gen-
erated in a dose- and time-dependent manner. arachidonic
acid was released. and prostacyclin was produced. The
arachidonic acid release was dependent on intracellular
Ca?t stores. while prostacyclin formation was in part re-
lated to protein kinase C activity (Magnusson et al. 1989).
In rat cortical slices. NaF (2-20 mM) application stimu-
lated phosphoinositide hydrolysis. rapidly producing IP as
the major product within a few min (lope I988). Em-
ploying murine keratinocytes. fluoroaluminates (rnicro-
molar AlClalmillimolar NaF) stimulated total inositol
phosphate formation. in the presence of 0.05 mM Ca” in
the medium. The enhanced phosphoinositide metabolism
was. however. not sufficient to induce cellular differentia-
tion (Lee and Yuspa 1991). On the other hand. in osteo-
blasts low concentrations of aluminum (10-25 M
Alz(SOa)3. or AlCls. or Al(N03)3) stimulated DNA syn-
thesis and alkaline phosphatase activity through a me-
chanism apparently not involving ﬂuoride (SO—200 itM
NaF) (Lau et al. 1991). At these fluoride concentrations one
would not expect formau'on of phosphate-like fluoro-
alurninates.

Various cells have been reported to exhibit oscillations
in cytosolic Ca2+ concentrations periodically released from
intracellular Ca“ reservoirs (Berridge 1993). For example.
in single rat hepatocytes such Cazt oscillations could be
induced by cell surface receptor agonists. However. when
generated by direct activation of G proteins with ﬂuoro-
aluminates (1 uM AlClsll mM NaF). these oscillations
became less regular in both frequency and temporal de-
pendence compared with agonist-induced ones. The am-
plitudes of the oscillations ranged from a resting value of
about 200 nM to maximally 600 nM of intracellular Ca2+
levels. which were monitored with rnicroinjected aequorin.
The irregularity in the fluoroaluminate-induced. receptor-
independent oscillations was attributed to a modulatory role
in the transient generation by the agonist-receptor complex
(Woods et al. 1990).

The notion that all effects of ﬂuoroaluminates can be
ascribed to the formation of a pseudo-phosphate in the

139

3

GDP-G protein complex was thrown into doubt by ex-
periments on the bacterial peptide chain elongation factor
EF-Tu (Kraal et al. 1990). a member of the family of G
proteins (Bosch et al. 1989). However. taking into account
the three-dimensional location of the three consensus ele-
ments involved in guanine nucleotide binding (La Cour et
al. 1985: Dever et al. 1987) and data from NMR experi-
ments (Hazlett et al. 1990). it appears that the guanine
binding sites of ET-Tu and other G proteins are structurally
non-equivalent.

At higher concentrations. the presence of aluminum in
cells generally leads to inhibition rather than stimulation
of biochemical processes. Using ﬂuoride-stimulated
(5—10 mM NaF) murine neuroblastoma cells. A103 (50 uM
to 1 mM) reduced inositol phosphate formation in a dose-
dependent manner. Aluminum apparently did not interfere
with subsequent kinase or phosphatase reactions. down-
stream of 1?; production. The primary aluminum action
rather occurred at an upstream site(s) responsible for 1?;
production. At low aluminum doses (about 0.2 uM to
50 uM). the fluoride-stimulated inositol phosphate forma-
tion was virtually unaffected (Shi et al. 1993).

As opposed to heterotrimeric G proteins discussed
above. there also exists a second family of smaller
(20—25 kDa). monomeric G proteins (Downward 1990).
These smaller G proteins cannot be activated by aluminum.
fluoride. and Mg2+ (Kahn 1991).

In the preceding paragraphs fluoroaluminates were in-
troduced as valuable tools in studying signal transduction.
To generate biochemically active. i.e. phosphate-like. spe-
cies of fluoroaluminates. typically millimolar NaF is ap-
plied-to cellular suspensions or enzyme solutions. Given
these evidently high doses of NaF. the question has to be
raised regarding the toxicological signiﬁcance of fluoride
in aluminum toxicology.

The normal levels of ﬂuoride in soft tissues varied be-
tween25and53uM(l ppm=53uMF).whereasinteeth
and bones concentrations typically ranged from 5 to about
16 mM (Clarkson 1991). In the blood plasma of healthy
patients. baseline aluminum levels varied from about 0.2 to
0.4 uM; levels above 3 M were considered indicative of
increased aluminum burden (Van der Voet et al. 1991).
Employing laser microprobe mass analysis. in-
tracytOplasmic aluminum concentrations of about 1 mM
were found in tangle-bearing neurons of patients afﬂicted
with amyotrophic sclerosis (Perl and Good 1987). Given
the above micromolar fluoride levels found in soft tissues.
it is thus fairly improbable that phosphate-like ﬂuoro-
aluminates are being formed. This conclusion is based on
known dissociation constants of complex fluoroaluminates
in aqueous solution (Savchenko and Tananaev 1959) as-
suming the existence of thermodynamic equilibria. To the
best of our knowledge. there exists no information re-
garding fluoroaluminate speciation taking into account ki-
netic equilibria. i.e. conditions prevalent in a biological
microenvironment.

When millimolar levels of NaF are applied to cells.
there is the possibility that NaF elicits responses in-
dependent of those triggered by the simultaneously formed.
phosphate-like AIFI. This was demonstrated on rabbit fe-
moral arteries where aluminum ﬁuoride stimulated phos-

140

4

pholipase C whereas NaF (10 mM NaF plus deferoxamine
to chelate contaminating aluminum) apparently stimulated
calcium channels (Ruth and Blackmore 1990). A dual effect
of ﬂuoride was also observed on phosphoinositide meta-
bolism in rat brain cortex (Claro et al. 1990).

Interactions of aluminum with elements of the
phosphoinositide-associated signalling pathway

Concerning direct effects of aluminum ions on elements of
the phosphoinositide-linked signalling pathway. doses of
AlCl3 (l—lOO uM) inhibited PIP: hydrolysis by bovine
heart phospholipase C activity (McDonald and Mamrack
1988). and also the release of total inositol phosphates in rat
cortical slices stimulated by carbachol (Johnson and lope
1986). or ﬂuoride (lope 1988). Applying patch-clamp
techniques to single internally perfused pancreatic acinar
cells. pulsatile Cab-dependent Cl- currents. normally
evoked by acetylcholine. disappeared when 1 mM AlCh
was present in the perfusion solution (Wakui et al. 1990).
Furthermore. in hormone-stimulated hepatocytes. A103
(10 M) reportedly perturbed oscillatory phosphoinositide-
mediated Ca2+ signalling (Schofl et al. 1990).

In intact neuroblastoma cells. the bradykinin-triggered
1P3 release and the concomitant change in intracellular Ca2+
levels were appreciably less in alurninum-pretreated cells.
compared with corresponding values in the control group
(Shi and Haug l992). Employing digitonin-permeabilized
murine neuroblastoma cells. stimulated with nonhydrolyz-
able guanosine-5'-(gammathio) triphosphate (GTPISl). ap
plication of various aluminum doses (2 11M to 1 mM)
considerably interfered with the formation of IE. IPz. and
IP. Competition experiments (5 mM GTP. 5 mM Mg“.
50 uM AlCls) showed that the aluminum binding site on
the Gp protein was probably different from that for GTP
(Shi and Haug 1992). Furthermore. at enhanced in-
tracellular free CaZt concentrations a Gilt-mediated path-
way seemed to become Operative. bypassing the Mgzthp-
mediated pathway. In the Cal's-mediated pathway. appli-
cation of 50 itM aluminum mainly diminished the release
of 1P3. Following interiorization (Shi and Haug 1990).
aluminum therefore seems to target two elements crucial
for PIP: hydrolysis. viz. the Gp protein and a phosphoino-
sitide-speciﬁc phospholipase C (Fig. l) (Shi et al. 1993).

Rather than binding directly to proteins. e.g. phospho-
lipase C. the possibility cannot be excluded that aluminum
perturbed the enzyme‘s lipid microenvironment (Pan-
chalingam et al. 1991) because the hydrolyzing efficacy of
phospholipases was reportedly dependent on the physico-
chemical pr0perties of their lipid milieu (Dennis 1983).
Furthermore. these types of lipid-mediated effects may be
also transmitted to G proteins since certain speciﬁc lipids.
e.g. is0prenoids. were found to be covalently attached to G
proteins (Casey 1992). To illustrate. on measuring GTP-
induced inositol phosphate formation in digitonin-permea-
bilized neuroblaStoma cells exposed to increasing AlC13
doses (about 0 1.1M to 1000 uM). a biphasic inositol pro-
duction was observed. The data may in part be interpreted
in terms of aluminum-related changes in the phospholipase
lipid environment at aluminum doses higher than 50 M

(Shi et al. 1993). This notion is consistent with observations
that ﬂuorescent hydrophobic membrane probes exhibited
constraints in motional parameters when human eryth-
rocyte white ghosts were exposed to 20 uM AlC13 (Weis
and Haug 1989). Since in vitro experiments had indicated
that aluminum induces structural changes in calmodulin
(Yuan and Haug 1988). aluminum may also impair Calt-
calmodulin-dependent phospholipase C activity (Levine et
al. 1990), and other calmodulin-dependent enzymes. Sub-
units of certain G proteins are known to inhibit calmodulin-
sensitive adenylate cyclase (Mangels et al. 1992). In this
context it is noteworthy that aluminum ions. administered
as AlCls. reportedly bound to calpains. i.e. non-lysosomal
intracellular cysteine proteinases which harbor calmodulin-
like CaZt-binding domains and are activated by Celt
(Zhang and Johnson 1992). Moreover. calpain-mediated
proteolysis of low molecular weight subunits of human
neuroﬁlaments was inhibited by AlCl3 (ICso = 200 uM).
mainly by affecting the substrate directly (Nixon et al.
1990). In the absence of experimental data. theoretical ar-
guments suggested an avid binding of aluminum to P1P:
and/or 1?: (Birchall and Chappell 1988). which in turn may
adversely affect the phosphoinositidemediated Ca2+ sig-
nalling pathway. Studies on receptor-generated Ca2+ tran-
sients in hepatocytes indicated. however. that aluminum
does not interact with 1P3 by perturbing its metabolism or
its Cazt-releasing properties. It was concluded that alumi-
num might rather interact with either phospholipase C or G
proteins (Shears et al. 1990). consistent with recent ﬁndings
on neuroblastoma cells (Shi and Hung 1992: Shi et al.
1993). Clearly more studies are necessary to further de-
lineate aluminum target(s) with elements of the phospho—
inositide signalling pathway.

The aluminum concentrations adminisrered in the above
experiments on cells generally varied from 0 to about
1000 M. thus ranging from intracellular background level.
around 1 RM. to pathophysiological. and ﬁnally to very
high concentrations. Given this range of aluminum doses
applied. data derived from isolated cells appear to be re-
levant regarding the impact of aluminum on the signalling
pathway in cells associated with tissue. in particular that of
the brain. Regarding pathophysiological levels in experi-
mental toxicology. total brain aluminum concentrations
average about 100 uM, and in some neuronal populations
aluminum levels may reach 400 M as a result of non-
uniform accumulation (Nixon et al. 1990). even as high as
1 mM in tangle-bearing neurons of patients afflicted with
amyotrophic sclerosis (Perl and Good 1987). As opposed to
the extracellular aluminum dose. the intracellular free Al3+
concentration is difﬁcult to estimate. because of un-
certainties surrounding the kinetics of aluminum. binding to
various intracellular ligands. and the lack of information
regarding metal speciation in a cellular environment.
Nevertheless. model calculations indicate that the intra-
cellular free Al3+ level is in the subnanomolar range. at
physiological pH values (Martin 1986; Shi et al. 1993).

As indicated above. the majority of experimental studies
were conducted by exposing cells to aluminum salts.
mostly AlCla. However. when dealing with the bioavail-
ability of toxic aluminum in tissues like the brain. the
nature of the metal’s coordination sphere (speciation)

seems to play an important role. This is illustrated in ex-
periments on aluminum-related changes in the permeability
of the blood-brain barrier to sucrose of rats. Employing
compounds like aluminum lactate and aluminum maltolate.
the permeability characteristics were dependent on metal
speciation (Favarato et al. 1992). Similar observations were
made in studies on the aluminum distribution into brain and
liver of rats and rabbits in response to the intravenous ad-
ministration of aluminum lactate or citrate (Yokel et al.
1991).

Aluminum binding to G proteins other than G, proteins

At subnanomolar free Al3+ concentrations. the trivalent
cation could compete with Mgzr. at 1 mM. for the ex-
changeable nucleOtide (GTP or GDP) binding site. the so-
called E site of tubulin (Macdonald et al. 1987). a member
of the G protein family (Bosch et al. 1989). All“ inhibited
GTPase activity once GTP was bound to the protein. Given
the slow ligand exchange rate of A13‘ (Martin 1986). dis-
sociation of GTP was inhibited on Apr-bound GTP-tubu-
lin, leading to aberrant rrricrotubule assembly and dis-
assembly in a pure tubulin system (Macdonald et al. 1987).
On the other hand. employing a more realistic system. viz.
in vitro assays with tubulin/microtubule associated pro-
teins. no structural peculiarities of microtubules have been
found in response to various AlC13 doses (0.2-l mM)
(Schmidt et al. 1991). Additionally. in studies of mice fed
high dietary aluminum (up to l mg/g diet. several weeks).
impaired brain microtubule function did not appear to be an
early biochemical lesion in aluminum toxicosis (Oteiza et
al. 1989).

Fluoroaluminates reportedly did not bind to soluble tu-
bulin even in the presence of GDP. However. ﬂuoro-
aluminates bound with rrricromolar afﬁnity to tubulin in-
corporated in microtubules. and inhibited their depoly-
merization (Carlier et al. 1988). The ﬂuoride complex ap-
parently mimicked the terminal phosphate only after its
dissociation from the nucleotide but before its release from
the protein (Chabre 1990).

Aluminum ions also inhibited the activation of the G
protein-related transducin (Miller et al. 1989), which cou-
ples rhodopsin to the stimulation of a guanosine-3’-5'-
cyclic monophosphate (cGMP)-speciﬁc phosphodiesterase
in retinal rod outer segments (Stryer 1986). By competing
with Mg2+ for a high afﬁnity binding site on transducin.
subnanomolar concentrations of intracellular free Ali“.
following addition of A103. markedly inhibited GTP
binding to the protein. rather than inhibiting GTPase ac-
tivity once GTP was bound. as in the case of Al3+ inter-
actions with tubulin. At these ion aetivities. the rod cGMP
phosphodiesterase was only slightly inhibited (Miller et al.
1989). In the presence of A13». cells are therefore impaired
in processing light-induced signals in the vision mecha-
msm.

Fluoroaluminates also impacted transducin by binding
to the Mg2+-GDP-G protein complex. Behaving like a
pseudOphosphate. binding of ﬂuoroaluminate to u'ansducin
led to the formation of a GTP-alpha subunit complex of the
G protein. This latter complex could activate cGMP

141

5

phosphodiesterase in response to rhodopsin photoexcitation
(Kanaho et al. 1985). As opposed to the effect of free Al?"
(see above). ﬂuoride-chelated aluminum thus enhanced
signal transduction from a photoexcited receptor.

Another cellular communication channel that operates

via G proteins is the receptor-regulated adenylate cyclase

system which produces a second messenger. cyclic AMP
(Freissmuth et al. 1989). As to aluminum interactions with
this communication channel. experiments on rat brains
demonstrated that intracerebroventricularly administered
aluminum citrate (1 M. pH 7.2) caused pronounced. long-
lasting changes in cyclic AMP synthesis in the cortex
(Johnson 1988). Moreover. dual effects of aluminum as
activator and inhibitor were observed on adenylate cyclase
in a parasite. the liver ﬂuke. Micromolar amounts of MCI;
augmented the enzyme activation by ﬂuoride. Ale in-
hibited the GTP-stimulated cyclase at Mgzt concentrations
ranging from 1 to 4 mM (Mansour et al. 1983).

Since aluminum has been implicated in contributing to
neuronal dysfunction in Alzheimer‘s disease (Crapper
McLachlan I989). experiments were performed to de-
termine whether the level of alpha subunits of ﬁve G pro-
teins was altered in Alzheimer patients. Employing im-
munoblotting techniques. no signiﬁcant differences in the
level of G protein subunits between Alzheimer‘s diseased
and age-matched postmortem brains were detectable
(McLaughlin et al. 1991).

Conclusions

In the past few years. our understanding of primary inter-
actions of aluminum with cells has appreciably increased.
Current ﬁndings provide evidence that aluminum interacts
with key elements of cellular communication channels
(Fig. 1). At the very least. some of the features of the
multifacetted aluminum toxicity syndrome presumably
originate from these types of interactions. As a result.
aluminum-triggered derangements occur on the levels of
second messengers and the temporal progression of signal
events which in turn are fairly rapidly reﬂected in meta-
bolic dysfunctions. Regarding long-term changes. alumi-
num-related modiﬁcations of signalling pathways may lead
to altered gene expression and thus to impaired cellular
malfunctions since signal transduction pathways are in-
volved in regulating gene expression by modulating the
activity of nuclear transcription factors (Karin and Smeal
l992).

Evidently more work is needed to clearly identify direct
target(s) of aluminum with wellopuriﬁed G proteins. and
subunits thereof. What aluminum-triggered structural and
functional changes do take place following binding to the
protein? Application of recombinant DNA technology tools
is expected to enhance our understanding of aluminum-
sensitive regions of these proteins. A decision must be
reached whether or not. and at which doses. aluminum
indirectly interacts with these proteins via metal-related
changes in the lipid matrix of the membrane. Additionally.
the impact of aluminum on phospholipase C and its lipid
microenvironment deserves particular attention. Given ap-
parent differences in tissue sensitivity towards aluminum.

142

6

do these variations in part result from differential sensi-
tivities of phospholipase isozymes found in the respective
tissue? In aluminum-tolerant tissues. have cells adapted to
continual aluminum stress by employing communication
channels less susceptible to aluminum interference? A hint
of such a possibility is given by observations that in the
presence of higher Cab concentrations second messengers
can be generated by bypassing G proteins. Elucidation of
primary aluminum interactions with key elements of cel-
lular communication pathways will ultimately yield insight
into initial reactions associated with the pathophysiology of
aluminum toxicity.

References

Berridge Ml (I987) lnositol triphosphate and diacylglycerol: two in-
teracting second messengers. Annu Rev Biochem 56: 159—193
Berridge Ml (19931 lnositol u'iphosphate and calcium signalling.
Nature 361: 315-325

Birchall JD. Chappell JS (1988) Aluminium. chemical physiology. and
Alzheimer's disease. Lancet 11: 1008-1010

Bimbaumer L. Abramowitz l. Yatani A. Okabe K. Mattera R. Graf R.
Sanford J. Codrna J. Brown AM (1990) Roles of G proteins in
coupling of receptors to ionic channels and other effector systems.
CRC Crit Rev Biochem Mol Biol 25: 225-244

Blackmore PF. Bocckino SB. Wayniek LE. Exton JH (1985) Role of
guanine nucleotide-binding regulatory protein in the hydrolysis of
hepatocyte phosphatidylinositol-4.5-biphosphate by calcium-mo-
bilizing hormones and the control of cell calcium. J Biol Chem 27:
14477-14483

Bosch L. Kraal B. Parmeggiani A (eds) ( I989) The guanine-nucleotide
binding proteins. Common structural and functional properties.
Plenum Press. New York

Carlier MF. Didry D. Melki R. Chabre M. Pantaloni D (1988) Stabi-
lization of microtubules by inorganic phosphate and its skeletal
analogues. the ﬂuoride complexes of aluminum and beryllium.
Biochemisu'y 27: 3555-3559

Casey P (1992) Visual differences: G proteins. Nature 359: 671-672

Chabre M (1990) Aluminoﬂuoride and berylloﬂuoride complexes:
new phosphate analogs in enzymology. Trends Biochem Sci 15:
6—10

Clarkson TW (1991) Inorganic and organometal pesticides. ln: Hayes
WJ. Laws ER (eds) Handbook of Pesticide Toxicology. vol 2.
Academic Press. New York. pp 559—565

Claro E. Wallace MA. Fain IN (1990) Dual effect of ﬂuoride on
phosphoinositide metabolism in rat brain cortex. Biochem J 268:
733-737

Crapper Mclachlan DR (1989) Aluminum toxicity: criteria for as-
signing a role in Alzheimer's disease. In: Lewis TE (ed) En-
vironmental chemistry and toxicology of aluminum. Lewis. Boca
Raton. Ha. pp 299-315

Dennis EA (1983) Phospholipases. In: Boyer PD (ed) The Enzymes.
vol 16. Academic Press. New York. pp 307-353

Dever TE. Glynias Ml. Merrick WC (1987) GTP-binding domain:
three consensus sequence elements with distinct spacing. Proc Natl
Acad Sci USA 84: 1814-1818

Downward J (1990) The ras superfarnily of small GTP-binding pro-
teins. Trends Biochem Sci 15: 469-472

Favarato M. Zatta P. Perazzolo M. Fontana I... Nicolini M (1992)
Aluminum (Ill) inﬂuences the permeability of the blood-an
barrier to ["CJsucrose in rats. Brain Res 569: 330-335

Freissmutlr M. Casey Pl. Gilman AG (1989) G proteins control diverse
pathways of transmembrane signaling. FASEB J 3: 2125-2131

Ganrot P0 (1986) Metabolism and possible health effects of alumi-
num. Environ Health Perspect 65: 363-441

Hall A (1992) Signal transduction through small GTPases - a tale of
two GAPs. Cell 69: 389-391

Haug A (1984) Molecular aspects of aluminum toxicity. CRC Crit Rev
Plant Sci 1: 345-373

Hazlett TL. Higashijima T. Jameson DM (1990) An aluminum ﬂuoride
binding site was not detected on elongation factor Tu-GDP. Bio-
phys] 57: 289a

Johnson GVW. lope RS (1986) Aluminum impairs glucose utilization
and cholinergic activity in rat brain in vitro. Toxicology 40: 93-l02

Johnson GVW (1988) The effects of aluminum on agonist-induced
alterations in cyclic AMP and cyclic GMP concentrations in rat
brain regions in vivo. Toxicology 51: 299-308

lope RS (1988) Modulation of phosphoinositide hydrolysis by NaF and
aluminum in rat conical slices. J Neurochem 5|: 1731-1736

Kahn RA ( I991) Fluoride is not an activator of the smaller (20—25 kDa)
GTP-binding proteins. J Biol Chem 266: 15595-15597

Kanaho Y. Moss 1. Vaughn M (1985) Mechanism of inhibition of
transducin GTPase activity by ﬂuoride and aluminum. J Biol Chem
260: “493-11497

Karin M. Smeal T (1992) Control of transcription factors by signal
transduction pathways: the beginning of the end. Trends Biochem
Sci 172418-422 _

Kraal B. de Graaf JM. Mesters JR. van Hoof PJM. laquet E. Par-
meggiani A (1990) Fluoroaluminates do n0t affect the guanine
nucleotide binding centre of the peptide chain elongation factor
EF-Tu. Eurl Biochem 192: 305-309

La Cour TFM. Nyborg J. Thin 5. Clark BFC (1985) Structural details
of the binding of guanosine diphosphate to elongation factor
Tu from E. coli as studied by X-ray crystallography. EMBO J 4:
2385-2388 .

Lau KHW. Yoo A. Wang SP (1991) Aluminum stimulates the pro-
liferation and differentiation of osteoblasts in vitro by a mechanism
that is different from ﬂuoride. Mol Cell Biochem 105: 93-105

Lee E. Yuspa SH (1991) Aluminum ﬂuoride stimulates inositol
phosphate metabolism and inhibits expression of differentiation
markers in mouse keratinocytes. J Cell Physiol 148: 106-115

Levine SN. Sonnier GB. Abroe K (1990) Effects of diabetes mellitus
and aluminum toxicity on myocardial calcium transport. Toxicol-
ogy 65: 137-147

Lewis TE (ed) (1989) Environmental chemistry and toxicology of
aluminum. Lewis. Boca Raton. Fla.

Macdonald TL. Humphreys WG. Martin RB (1987) Promotion of tu-
bulin assembly by aluminum ions in vitro. Science 236: 183-186

Magnusson MK. Halldorsson H. Kjeld M. Thorgeirsson G (1989)
Endothelial inositol phosphate generation and prostacyclin pro-
duction in response to G protein activation by AlFI. Biochem J 264:
703-711

Majerus PW (1992) lnositol phosphate biochemistry. Annu Rev Bio-
chem 61: 225-250

Mangels LA. Neubig RR. Hamm HE. Gregy ME (1992) Calmodulin
binding distinguishes between beta gamma subunits of activated G
proteins and transducin. Biochem J 283: 683—690

Mansour JM. Ehrlich A. Mansour TE (1983) The dual effects of alu-
minum as activator and inhibitor of adenylate cyclase in the liver
ﬂuke Fasciola hepatica. Biochem Biophys Res Commun 112:
911-918

Marquis lK (1989) Neurotoxicity of aluminum. In: Lewis TE (ed)
Environmental chemistry and toxicology of aluminum. Lewis.
Boca Raton. Fla. pp 289-298

Martin RB (1986) The chemistry of aluminum as related to biology
and medicine. Clin Chem 32: 1797-1806

McDonald Ll. Manuack MD (1988) Aluminum affects phosphoino-
sitide hydrolysis by phosphoinositidase C. Biochem Biophys Res
Comrnun 155: 203-208

McLaughlin M. Ross BM. Milligan G. McCulloch J. Knowler IT
(1991) Robustness of G proteins in Alzheimer‘s disease: an im-
munoblot study. J Neurochem 57: 9-14

Miller lL. Hubbard CM. Litrnan Bl. Macdonald TL (1989) Inhibition
of transducin activation and guanosine uiphosphatase activity by
aluminum ion. J Biol Chem 264: 243—250

Nebelter HG. Coburn JW (1986) Aluminum and renal osteodystrophy.
Annu Rev Med 37: 79-95

Nixon RA. Clark JF. Logvinenko KB. Tan MKH. Hoult M. Grynspan F
(1990) Aluminium inhibits calpain-mediated proteolysis and in-
duces human neurofilament proteins to form proteaseoresistant
high molecular weight complexes. J Neurochem 55: 1950—1959

Oteiza Pl. Golub MS. Gershwin ME. Donald JM. Keen CL (1989) The
inﬂuence of high dietary aluminum on brain microtubule poly
merization in mice. Toxicol Lett 47: 279-285

Panchalingam K. Sachedina S. Pettegrew lW. Glonek T (1991) Al-
ATP as an intracellular carrier of Al(lll) ion. Int J Biochem 23:
1453-1469 .

Perl DP. Good PF ( I987) The association of aluminium. Alzheimer's
disease and neuroﬁbrillary tangles. J Neural Transm 24: 205-211

Rana RS. Hokin LE (1990) Role of phosphoinositides in transmem-
brane signaling. Physiol Rev 70: 115-164

Rasmussen H (1990) The complexities of intracellular Ca“ signalling.
Biol Chem Hoppe-Seyler 371: 191-206

Rath HP. Blackmore PF (1990) Differential activation of rabbit fe-
moral arteries by aluminum ﬂuonde and sodium ﬂuoride. l Phar-
macol Exp Ther 254: 514—520

Savchenko GS. Tananaev IV (1959) Types of complex ﬂuoro-
aluminates in aqueous solution. Soviet research in ﬂuorine cheni-
istry. 1949-1956. part 11. English edition. Consultants Bureau. New
York 203-212

Schmidt R. BOhm K. Vater W. Unger E (1991) Aluminium induced
osteomalacia and encephalopathy - an aberration of the tubulin
assembly into microtubules by Al"? Progr Histo- Cytochem 23:
355-364

Schdﬂ C. Sanchez~Bueno A. Dixon Cl. Woods NM. Lee JAC. Cuth-
bertson KSR. Cobbold PH. Birchall JD (1990) Aluminium perturbs
oscillatory phosphoinositide-mediated calcium signalling in hor-
mone-stimulated hepatocytes. Biochem J 269: 547-550

Shears SB. Dawson AP. Loomis-Husselbee lW. Cullen Pl ( 1990) The
perturbation. by aluminium. of receptor-generated calcium tran-
sients in hepatocytes is not due to effects of 1ns(1.4.5)P3-stimulated
Ca" release or lns(1.4.5)P3 metabolism by the 5—phosphatase and
3okinase. Biochem J 270: 837

Shi B. Haug A (1990) Aluminum uptake by neuroblastoma cells.
J Neurochem 55: 551-558

Shi B. Haug A (1992) Aluminium interferes with signal transduction in
neuroblastoma cells. Pharmacol Toxicol 71: 308-313

143

7

Shi B. Chou K. Haug A (1993) Aluminium impacts elements of the
phosphoinositide signalling pathway in neuroblastoma cells. Mol
Cell Biochem 121: 109-118

Stemweis PC. Gilman AG (1982) Aluminum: a requirement for acti-
vation of the regulatory component of adenylate cyclase by
ﬂuoride. Proc Natl Acad Sci USA 79: 4888-4891

Stryer L (1986) Cyclic GMP cascade of vision. Annu Rev Neurosci 9:
87-119

Taylor GJ (1991) Current views of the aluminum stress response; the
physiological basis of tolerance. Curr Top Plant Biochem Physiol
10: 57-93

Tojyo Y. Tanimura A. Matsui S. Matsumoto Y. Sugiya H. Furuyama S
(1991) NaF-induced amylase release from rat parotid cells is
mediated by P1 breakdown leading to Ca3° mobilization. Am
J Physiol 260: C194-C200

Troullier A. Girardet lL. Dupont Y (1992) Fluoroaluminate complexes
are bifunctional analogues of phosphate in sarcoplasmic reticulum
Cale-ATPase. J Biol Chem 267: 22821—22829

Van der Voet GB. Marani E. de Wolff FA (1991) Aluminium neuro-
toxicity. Progr Histo- Cytochem 23: 235-242

Wakui M. ltaya K. Nirchall D. Peterson H (1990) Intracellular alu-
minium inhibits acetylcholine- and caffeine-evoked Calt mobili-
zation. FEBS Len-267: 301—304

Weis C. Haug A (1989) Aluminum-altered membrane dynamics in
human red blood cell white ghosts. Thromb Res 54: 141-149

Wills MR. Savory l (1985) Water content of aluminum. dialysis de-
mentia. and osteomalacia. Environ Health Perspect 63: 141-147

Woods NM. Dixon Cl. Cuthbertson KSR. Cobbold PH (1990)
Fluoroaluminate mimics agonist application in single rat hepato-
cytes. Biochem J 265: 613-615

Yokel RA. Lidums V. McNamara Pl. Ungerstedt U (1991) Aluminum
distribution into brain and liver of rats and rabbits following in-
travenous aluminum lactate or citrate: a microdialysis study.
Toxicol Appl Pharrnacol 107: 153-163

Yuan S. Haug A (1988) Frictional resistance to morions of bimane-
labelled spinach calmodulin in response to ligand binding. FEBS
Lett 234: 218-223

Zhang H. Johnson P (1992) Differential effects of aluminum ion on
smooth muscle calpain land calpain ll activities. Int J Biochem 24:
1773-1778

APPENDIX D

ALUMINUM COORDINATION TO CALMODULIN: THERMODYNAMIC AND
KINETIC ASPECTS

Haug, A., Vitorello, V. 1996. Coord. Chem. Rev. (in press)

144

 

 

lilllllllllNllllllN
CHEMISTRY

' . ' Coordination Chemistry Reviews 0 (1995) REVIEWS
ELS EVI R 000-000

       

 

 

Aluminium coordination to calmodulin:
Thermodynamic and kinetic aspects

Alfred Haug ', Victor Vitorello ”'1

‘ Department ofMiaobiology. Michigan State University, East Busing, MI 48824. USA
" Department of Boraw. Michigan State University. East Lansing. MI 48824. USA

Received 23 January 1995; revised 3 April 1995

 

Contents
Abstract ...................................................... .000
1. Introduction ................................................ 000
2. Calmodulin ................................................. 000
3. Al( III) in biology ............................................. 000
3.1. A1( III) in aqueous solution ..................................... 000
3.2. Al(III) in cells ............................................ 000
4. Al(III) impacts calmodulin ........................................ 000
4.1. Thermodynamic aspects of Al(III) binding to calmodulin ................... 000
4.2. Kinetics of Al(III) removal by dtric acid from calmodan .................. 000
4.3. Al(III) impairs recomition of calmodan by target proteins ................. 000
4.4. Al(III)-related changes in calmodulin's internal dynamics ................... 000
Summary .................................................... 000
Acknowledgement ............................................... 000
References .................................................... 000

Abstract

Calmodulin is an acidic, Ca(II)-binding protein (about 17000 d) which plays a key role in
regulating a broad spectrum of target enzymes in eukaryotic cells. Within each lobe of the
dumbbell-shaped molecule there are two adjacent Ca(II)-binding domains. which form a
single cooperative unit. Hydrophobic forces are crucial for interfacing the Ca(II)-activated
regulatory protein with its target enzyme. Presumably not involving the {our speciﬁc Ca(II)-
binding domains on calmodulin, binding of Al(III) to calmodan generates considerable
dehydration entropies. Al(III )-triggered changes in the a-helix content of calmodulin are
involved in the mismatch between calmodulin and its target protein. By altering calmodulin's

‘ Permanent address: Centro de Energia Nuclear na Agrintltura/USP. Caixa Postal 96. Piracicaba. 13400
Sao Paulo. Brasil.

0010-8545/95/309.50 c 1995 ElseVier Science S.A. All rights reserved
5501 0010~8545(95)01168-4

145
2

internal dynamies. Al(III) apparently interferes with the proteins capacity to search out
conformational substates suitable for proper docking with a speciﬁc target protein.

Keywords: Al(III) binding; Al(III)-related dehydration; Al(III) removal; Calmodulin;
Calmodulin partners; Protein dynamics

 

1. Introduction

In eukaryotic cells a small protein, calmodulin, plays a pivotal role in regulating
a broad spectrum of target enzymes which in turn are involved in vital cellular
processes, such as protein phosphorylation, cyclic nucleotide metabolism, motility,
cell division, and sensory perception [1-3]. The regulatory action is accomplished
by binding Ca(II) to calmodulin, thereby modulating its conformation in such a
manner as to promote speciﬁc interactions between the small protein and the
respective target. As a cytosolic protein, calmodulin senses changes in intracellular
free Ca(II) concentrations, [Ca(II)], ranging from about 10 nM to 1 itM. Changes
in [Ca(II)] are brought about by release of Ca(II) from intracellular stores, or from
Ca(II) signals generated in response to the transduction of external input signals
across the plasma membrane. Chief among signal transduction pathways is that
associated with membrane-bound phosphoinositides [4-6]. Calmodulin is therefore
an essential communication link necessary for transforming extracellular information
into speciﬁc biological responses.

2. Calmodan

The key to calmodulin’s ability to function as a Ca(II)-dependent transducer
resides in its primary protein structure. This acidic, globular protein (about 17000
d, 148 amino acid residues, isoelectric point about 4.1) has been highly conserved
during the evolution of eukaryotic cells. suggesting the existence of an ancestral gene
E 7,8]. X-ray diffraction studies demonstrated that calmodulin resembles a dumbbell
(length about 6.5 nm) with the two lobes interconnected by an eight-turn a-helix.
This central helix (about 2 nm long, amino acid residues 65-92) carries an appreciable
net dipole which may play a role in the sequential binding of Ca(II) to the apoprotein
[9.10]. Each lobe (diameter about 2 nm) harbours two Ca(II)-binding domains
whose helix-loop—helix arrangement (11:12:11 residues reSpectively) is similar to
the structure of Ca(II)—binding units in other Ca(II)-binding proteins [11]. In
- calmodulin, only the loop regions contain antiparallel B-sheet strands. Coupling
antiparallel strands of adjacent Ca(II)-binding loops, hydrogen bonds assist in
stabilizing the macromolecule [9,10]. Counting from the amino terminus of calmo-
dulin, the Ca(II)-binding domains are designated as I, II, III and IV respectively.
Thermodynamic studies on domain organization in calmodulin indicated that within
a lobe the two adjacent Ca(II)-binding domains form a single cooperative block,
especially between domains I and II. The two lobes seem to repel each other [12].

146
3

The four domains in calmodulin permit speciﬁc binding of 4 mol Ca(II) per mol of
protein. At approximate vertices of an octahedron, the liganding residues harbour
oxygen-containing side chains, such as those of aspartic acid, and the peptidyl
carbonyl oxygen. Hydration water is probably part of the ligand sphere [13]. Despite
considerable similarities in the primary sequence of the domains’ loops, the dissoci-
ation constants for Ca(II) vary between 1 uM and 10 MM, at physiological conditions.
The Ca(II)-binding properties and the sequence of Ca(II)-binding to the protein are
critically dependent on the presence of monovalent ions, pH, and divalent cations
like Mg(II). Current models indicate that domains III and IV (residing on the
C-terminus lobe) are c00peratively binding Ca(II) with high amnity, as compared
with the low aﬁnity cooperative binding in domains I and II [2,14]. NMR experi-
ments on calmodulin in solution further corroborate the existence of high and low
afﬁnity Ca(II)-binding domains [14]. For the high afﬁnity domains, the rates of
Ca(II) dissociation range from 3 to about 10 3”, while those from the low aﬁnity
domains are in the 300 to 550 5" range, at room temperature [15,16]. Ca(II)
association constants are expected to range from 107 to 10‘I s“, since these values
are related to the rates of water exchange of hydrated Ca(II) ions [17].

Upon binding of 1 to 4 mol Ca(II) per mol of apoprotein, pronounced structural
transformations take place in the protein such as increase of the a-helix content. and
enhanwd formation of hydrophobic regions on the protein surface. Ca(II)-binding
generates a readjustment/formation of the ﬁ-sheet within the binding 100ps of the
domains [18]. These helical rearrangements may in part result from coupled move-
ments of a pair of helices [l9]. Elicited by coordination reactions [20], especially
by those involved in the binding of two Ca(II) in the high afﬁnity domains of the
C-terminus lobe, local perturbations in structure are also transmitted to the
N-terminus lobe of the ﬂexible macromolecule [21]. Ca(II) binding also eﬂ'ects the
protein to elongate by about 7% [22]. Ca(II)-triggered short and long range
conformational changes cause an enhanced exposure of hydrophobic regions on the
protein surface [23], thus preparing calmoduhn for docking with and the activation
of the target protein.

The molecular basis for calmodulin’s ability to activate numerous target proteins
has not been elucidated. Nevertheless, experiments on calmodulin-binding oligopep-
tides. like the amphiphilic melittin (2845 d, about 1.5 nm long, 26 amino acid
residues), demonstrated that hydrophobic forces are crucial for interfacing the
Ca(II)-calmodulin with its target [24]. Regarding calmodulin interactions with its
partner (molar ratio 1:1), binding constants vary from 10 itM to 100 pM [25]. The
Ca(II)-calmodulin structure apparently collapses around its partner [26], involving
hydrophobic and salt bridges [27].

3. Al(III) in biology

3.1. Al(III ) in aqueous solution

Since the water content of a typical cell comprises about 80% of the cellular
weight, it is essential to understand the behaviour of Al(III) in aqueous solution.

I47

4

Regarding details of the aqueous chemistry of Al(III) we are referring to informative
review articles [28—30]. Brieﬂy, the coordination of the trivalent cation, Al(III), in
aqueous solution is generally either tetrahedral, pH > 6.2, like in Al(OH): , or octahe-
dral, pH<5, like in Al(H,O)2‘. The ionic radius is 0.054 nm; the hydrated radius
has a value of 0.475 nm [31]. As the pH increases, the hexahydrated species hydro-
lyzes, especially in the range from 5<pH<6.2. The deprotonation of Al(III) in
aqueous solution can be partially or completely inhibited by ligation of Al(III) to
ligands, in particular those containing oxygen and ﬂuorine as donor atoms [32].
Given the multitude of ligands in biological ﬂuids, the slow solvent exchange rate
of Al(III) is of great import. To illustrate, the formation rate constants of solvated
Al(III) with multidentate ligands vary from about 0.1 to 10 M" s" [17,32], while
those of Ca(II) range from about 107 to 10‘0 M" s" [17]. In the absence of
ligands (pH<7), formation of polynuclear Al(III) species in aqueous solution is
unclear for [Al(III)] < 100 uM [33,34].

3.2. Al(III ) in cells

Formation of polynuclear species is also highly improbable in cells, where numer-
ous Al( III) ligands are available. According to model calculations, the intracellular
free Al(III) concentration, in the neutral pH range, is in the subnanomolar range
[35]. ATP seems to be a key ligand for Al(III) [36,37], affording subnanomolar
concentrations of free [Al(III)] for typical experimental conditions (neutral pH,
millimolar ATP, Al(III) doses less than 1 mM) [38]. Concerning the total [Al(III)]
in cells, background levels were found to be around 1 uM in neuroblastoma cells,
but may reach close to 1mM in tangle-bearing neurons of patients afﬂicted with
amyotrophic sclerosis [39].

4. “(III ) impacts calmodulin
4.1. Thermodynamic aspects of Al ( III ) binding to calmodulin

Calmodulin harbours numerous electron pair donor atoms (e.g. carboxyl and
carbonyl oxygens, amino and guanidino nitrogens) which are potential ligands
capable of participating in metal coordination. In pretransition metals (e.g. Ca(II),
Mg(II), Al(III)) the complex formation is critically dependent on the size of the
ligand and the nature of the metal ion; electrostatic interactions play a key role [40].

Concerning Al(III) binding to bovine brain calmodulin, equilibrium dialysis
(pH 6.5, 10 mM Tris buﬂ'er, 5 mol Ca(II) per mol of protein, 5 uM protein) studies
demonstrated [41] that the macroscopic dissociation constant is 1.2 x 10" M for
the ﬁrst Al(III) bound to the protein. This reaction is associated with an entrOpy
change of AS°=22 cal deg" mol". Binding of a second and third mole Al(III) to
calmodan is characterized by an average entropy change of about AS°=41 cal
deg" mol " per each mol Al(III), corresponding to an average dissociation constant
of about 0.6 x 10'6 M [42]. The positive entropy changes observed may be attributed

148 5

to the considerable dehydration entrepies gained as the salvation structures of the
ligand and the metal cation are breaking up upon formation of the complex. In view
of spectroscopic evidence [41-43] (see below), that the calmodulin conformation is
appreciably perturbed and becomes more randomized upon binding of the second
and third Al(III), it is reasonable to expect that these Al(III)-triggered effects are in
part associated with conformational entrOpy changes. These latter changes may thus
contribute to the overall entropy changes (AS° =41 cal deg" mol ") observed upon'
binding of the second and third Al(III) to calmodulin. For comparison, the average
entropy change associated with the binding of a mole of Ca(II) per calmodulin is
16.1 cal deg" mol", mostly attributable to dehydration processes [44]; the dissoci-
ation constants vary in the range from 10” to 10" M [45].

The magnitude of the dissociation constants and the extent of dehydration
observed suggest that the ﬁrst three Al(III) ions are binding to localized regions on
the protein. Via some form of multidentate chelation, Al( III) ions seem to coordinate
directly to a speciﬁc set of ligands, because the binding is tighter by several orders
of magnitude than expected for Debye-Hilckel screening ions [46]. At physiological
pH, potential metal coordination sites are carboxyl groups and the imidazoyl moiety
of histidine as indicated by the deprotonation pH ranges of amino acids [47].
Important coordination sites are also donor atoms in the lateral groups of amino
acid residues (aspartyl, glutamyl), rather prevalent in the acidic calmodulin.

Despite the absence of detailed structural information, it is nevertheless safe to
assume that the high-aﬁnity coordination sites of Al(III) do not involve those of
the four speciﬁc Ca(II)-binding domains on calmodulin. Equilibrium dialysis and
- atomic absorption studies indicated that Ca(II) remained bound to calmodulin,
following addition of Al(III). Sequential addition of Al(III) simultaneously caused
an almost complete inhibition of ' the Ca(II)—calmodulin-dependent cyclic 3':5'-
nucleotide of phosphodiesterase activity when the molar ratio of [AU-[calmodulin]
reached a value 4:1 [48]. The observed lack of Al(III)-Ca(II) exchange is not
surprising since metal coordination with proteins is generally governed by ionic size
and charge [49]. Having similar ionic radii as that of Ca(II) (radius 0.099 nm) [31],
lanthanides are able to replace speciﬁcally bound Ca( II) in calmodan which remains
enzymatically active However, with an ionic radius of 0.132 nm, Pb(II) can also
replace Ca(II) in speciﬁc Ca(II) binding regions [14,50]. Besides the nature of the
cation, other factors like stereoselectivity thus also contribute to metal coordination
to ligands in calmodulin.

Since Al(III) apparently does not displace Ca(II) from its speciﬁc coordination
domains, one can only speculate about the location of putative Al(III) binding sites
on calmodulin. As a ﬁrst approximation, Al(III) may bind to sites harbouring Mg(II)
which has an ionic radius (0.065 nm) close to that of Al(III). Measuring the infrared
absorption maxima in the amide I' region (1644 cm ‘ 1) of calmodulin, conformational
rearrangements in the a-helical regions were observed following Mg(II) binding
[51]. These ﬁndings are in accord with those tentatively localizing Mg(II)-binding
sites in the N-terminal a-helix of each of the four Ca(II)-binding domains [24], and
possibly in the central a-helix [l4], connecting the two lobes. The Ca(II) and Mg(II)
sites are, however, not identical, and the Mg(II) dissociation constants are in the

149
6

millimolar range [24]. As to putative Al(III) binding sites, there is spectroscopic
evidence [41] that Al(III) application to calmodulin produces a considerable
decrease in ar-helix content of the protein. This decrease in helix content may be
attributed to Al(III) residing in Mg(II) binding sites mentioned above. Further,
these sites harbour aspartyl and glutamyl residues [24] which are favourable for
metal coordination. Since the binding afﬁnity of Al( III) is appreciably higher than
that of Mg(II), the divalent cation is presumably displaced.

Given the current information. further experiments are necessary to quantify the
effect of Al(III) binding to calmodulin, in particular the role of the ionic strength,
pH, and the presence of Mg(II). Identifying peptide regions which bind Al(III) with
high afﬁnity, Al(III)-chelate aﬂinity chromatography [52] may be applied to purify
peptides produced by trypsin digestion of calmodulin. Site-directed mutagenesis [53]
can be used to introduce speciﬁc amino acids probably involved in interactions
between the metal and the protein ligand. Additionally, in studies on calmodulin,
efforts should be made to accurately assess the impact of low concentrations of
Al(III) by using appropriate metal ion buﬂ'ers [54]. To better evaluate
Al(III)—peptide-protein interactions, data should be generated regarding complex
formation of Al(III) and amino acids, in particular aspartic and glutamic acids.

4.2. Kinetics of Al ( III ) removal by citric acid from calmodulin

The kinetics of Al(III) dissociation from apocalmodulin and Ca(II)-saturated
calmoduhn was studied by atomic absorption spectroscopy [55]. Following preload-
ing of apocalmodulin with Al(III) (molar ratio 1:3; pH 6.5, 10 mM Pipes buﬂ‘er,
1011M calmodulin, 22°C), Al(III) was quantitatively removed in the presence of
citrate ([citrate]-[Al(III)]=9:l), a strong Al(III) chelator [29]. The kinetics of
Al(III) removal could be best described as the sum of two exponential expressions
characterized by rate constants: k,-=4.1i0.7 min" and k,-=0.6i0.08 min". The
velocity of removal of the ﬁrst Al(III) exceeded the instrumental deadtime (15 s). In
the presence of Ca(II) ([Ca(II)]-[calrnodulin</RM>]=6:1), the removal of
Al(III) was accelerated, as indicated by k3- 2551-05 min" and k“ n0.5i0.04
min". As judged by the rates of Al(III) removal of the second and third Al(III),
the binding sites are heterogeneous and probably noninteracting. The diﬂerences in
the rates, i.e. k, vs. k“, presumably arise from the accessibility of the chelator to
the respective Al(III) binding sites in the ﬂexible apocalmodulin [56], or the more
compact Ca(II)-saturated calmodulin [10,57]. Given these observations, one can
speculate that the ﬁrst Al(III) is coordinated to ligands in the central helix region
which is readily accessible in the absence or presence of Ca(II). The second and
third Al(III) are coordinated to ligands in the vicinity of, or at, the Ca(II)-binding
domains, as hypothesized above, where chelator access might be restricted in the
more compact Ca(II)-saturated protein compared with the more open apocalmo—
dulin. These types of restriction and the profound denaturation initially produced
by Al(III) in the protein [41] may account for the lack of ready reversibility of the
Al(III)—calmodulin complex upon applican'on of citrate. It is noteworthy that the
rate constants, k, and k,, correspond to half-lives of about 10 and 70 s respectively.

150
7

Similar time constants, ranging from 10 to about 100 s, have been observed in
thermal unfolding reactions of ribonuclease A [58], whose molecular weight (M :13
700) is close to that of calmodulin. Citrate-related removal of Al(III) is a considerably
slower process than dissociation of Ca(II) from calmodulin. Employing Ca(II)-
sensitive ﬂuorescence chelators, the latter rate constants, kg“, 550 s" and about 5
s" (19‘C), correspond to the fast and slow release of two Ca(II) each from the
protein [44].

Al(III) application inhibited the Ca(II)- and calmodulin-dependent 3',5'-cyclic
nucleotide phosphodiesterase activity by impairing calmoduﬁn rather than the basal
activity of the enzyme, examined in the absence of calmodulin [41]. Following
exhaustive citrate removal by dialysis, the Al(III)-depleted calmodulin stimulated
the Ca(II)—calmodulin-dependent phosphodiesterase activity in a manner identical
to a calmodulin which had never seen Al(III) and citrate. Despite such a complete
restoration of biological activity, restoration of the a-helix content of calmodulin
remained incomplete, irrespective of the presence or absence of Ca(II) [55]. This
lack of complete restoration may result from local rearrangements of ligands in the
region previously occupied by Al(III).

Clearly, more experiments are necessary to understand mechanisms of chelator-
triggered dissociation of Al(III) from calmodulin. Al(III)-sensitive, ﬂuorescent chela-
tors may be used to observe processes related to the removal of the ﬁrst Al(III).
The Al(III)-related pathway of calmodulin refolding should be analysed relative to
that operative upon Ca(II) dissociation.

4.3. Al(III ) impairs recognition of calmodulin by target proteins

Al(III)-triggered structural changes in calmodulin reportedly led to the inhibition
of Ca(II)—calmodulin-dependent enzymes like 3',5'-cyclic nucleotide phosphodiester-
ase [41], a plant membrane-bound ATPase activity [59], and calpains [60.61]. To
get insight into the tOpography of the enzyme’s binding region with Ca(II )-activated
calmodulin, a high-amnity, amphiphilic, 26-residue peptide [24,25] from bee venom,
melittin, was used [43]. The steady state and time-dependent ﬂuorescence character-
istics of melittin’s single tryptOphanyl residue are sensitive parameters to derive
information on the peptide’s interface with calmodulin, with Al(III) absent or present.
With Al( III) present. the tryptophanyl residue experienced a more polar microenvi-
ronment at the interface of melittin with Ca(II)-calmodulin. These ﬁndings are
probably a consequence of the Al(III)-triggered breakage of a-helices in
Ca(II)—calmodulin [41] which in turn would lead to a more open Ca(II)—calmodulin
structure compared with the compact Ca(H)-calmodulin.structure [62], observed
in the absence of Al(III). Al(III)-induced structural changes are coupled with a
rearrangement of salvation water. which plays a key role in protein dynamics [63,64]
and as a hydrogen bond sink [65]. An open protein structure is also indicated by:
(a) a decreased immobilization of a spin probe covalently attached to Al(III )-treated
Ca(II)—calmodulin [42]; (b) an improved access of diffusion-controlled, collisional
quenching molecules to the ﬂuorophore of melittin, associated ( 1 : l) with calmodulin;
(c) an enhanwd rotational correlation time of melittin’s ﬂuorophore. This faster

151
3

correlation time (3.5 as) is indicative of segmental motions in the vicinity of the
tryptophanyl residue compared with restricted motions (8.1 ns) observed in the
hydrophobic-hydrogen bonding [65,66] interplay of the compact melittin-calmo-
dulin complex, existing in the absence of Al(III) [43]. Since the basic moiety of
melittin apparently associates with acidic amino acid residues [26,67] of the central
helix in biochemically active Ca(II)-calmodulin [9.10], it is noteworthy that forma-
tion of time-dependent hydrogen bonding patterns by the residues Arg 74, Arg 86,‘
and Arg 90 are crucial in the compaction process during complexation [68]. The
central helix has also been suggested to harbour a Mg(II)-binding site [14] which
in turn is a putative site for Al(III), as discussed previously. Arg 86 is ﬂanked by
three glutamate residues [24] (Glu 82, Glu 83, Glu 84) potentially capable of forming
chelate rings through side chain donor atoms [47]. Given the proximity of key
residues for the formation of hydrogen bonding patterns and the putative Al(III)—
binding site on the central helix, Al(III)-triggered breakage of hydrogen bonds would
have profound repercussions on the compaction process during calmodulin’s inter-
action with melittin. Taken together, the presence of Al(III) apparently induces a
mismatch between calmodulin and its partner. Such a mismatch may be instrumental
in Al(III)-related malfunctions of Ca(II)—calmodulin dependent enzymatic reactions.

Concerning the Al(III)-related lack of recognition between calmodulin and its
speciﬁc partner, experiments assessing solely the binding constants for the two
proteins are obviously insuﬁcient. Detailed structural and dynamic studies are
required to ascertain Al(III)-related changes in the complementary surface, crucial
for recognizing calmodulin’s speciﬁc partner protein. Particular attention should be
paid to the role of hydration water presumably involved in the Al(III)-disturbed
compaction process upon complexation.

4.4. Al(III )-related changes in calmodulin's internal dynamics

Like other globular proteins, calmoduﬁn has a distinct equilibrium structure [10].
Aﬂ'ording calmodulin the capacity to associate with speciﬁc partners, other conforma-
tions. separated by low energy barriers, and different from the average equilibrium
structure, must exist [63,64,69].

Although a clear picture has not yet emerged, the space-time dynamics in calmo-
dulin was illustrated by nanosecond motions of a speciﬁc site [70.71], by thermally
induced ﬂuctuations modulating dipole—dipole interactions [72], by nanosecond
responses elicited via ligand-triggered conformational perturbations [21], and by
internal and side chain mobilities in NMR experiments [15,44]. Whether low fre-
quency motions (wave numbers less than 50 cm") are important for calmodulin’s
biological activity has not been elucidated [73].

For the purpose of diagnosing Al(III)-related anomalies in the dynamics of calmo-
dulin, relative to that in the presence of Ca(II), the frictional resistance to local
motions of a protein-bound ﬂuorophore was evaluated [74]. The ﬂuorescence anisot-
ropy and lifetime of protein-associated ﬂuorophore are sensitive to global motions
of the protein and to local rotations in the ﬂuorophore's microenvironment From
the temperature dependence of these ﬂuorescence parameters the thermal coeﬁicient

152
9

of the frictional resistance to the ﬂuorophore motions can be extracted [75,76]. As
to calmodulin, Al(III)-generated changes in ﬂuorescence parameters were measured
(0.1 M KCl, 10 mM Mops buﬂ‘er, pH 6.5, 5 mol Ca(II) and 3 mol Al(III) per mol
calmodulin, range 10 to 40 °C) through a ﬂuorophore covalently anchored at site 26
of spinach calmodulin where a cysteinyl residue is residing. Both in the presence or
absence of Al(III), calmodulin underwent a substate change at a critical temperature
of 25 °C [74]. This temperature is appreciably lower than that of the major unfolding
of the protein: at 55 °C for apocalmodulin and above 90°C in the presence of Ca(II)
[77]. In the presence of Al(III), the enthalpy, AH=8.2 kcal mol", was close to
that of apocalmoduﬁn (8.0 kcal mol“), as opposed to biochemically active
Ca(II)-calmodulin (5.3 kcal mol"), measured in the absence of Al(III). The lower
AH value in biochemically active calmodan apparently reﬂects the sensitive dynamic
aspects essential for the transfer of information within the protein in response to
regulatory stimuli. As indicated by a AH =8.2 kcal mol"1 and the thermal resistance
coeﬂicients, the ﬂuorOphore motions appeared to be less disposed to temperature
changes when Al(III) was associated (3: 1) with calmodulin. Since A1( III) application
to calmoduhn caused hydrogen bond breakage [41], some of the temperature-
dependent changes in AH and motional characteristics described above may be
attributed to alterations in protein hydration (solvent ﬂuctuations) relative to that
seen by biochemically active Ca(II)—calmodulin. The observed restricted ﬂuorophore
mobility at site 26, with Al(III) present, may in part result from the increased
viscosity in the ﬂuorophore‘s microenvironment [74], because solvent atoms exert
a damping eﬂ'ect on the atoms at the surface of the protein [64]. Taken together,
by altering calmodulin’s conformational substates, Al(III) apparently interferes with
the protein’s capacity to search out substates suitable for proper docking with a
speciﬁc partner protein harbouring melittin-like amino acid sequences.

5.Summary

To fully understand the regulatory properties of multifunctional
Ca(II)-calmodulin, the systematic features of the protein's space-time dynamics
must be clariﬁed. Links must be established between the critical conformational
substate at the onset of ligand binding (e.g. Ca(II) coordination) and the ﬁnal
conformational substate(s) responsible for correct docking between calmodulin and
its respective target. By the same token, the temporal and structural gap must be
bridged between the initial substate of the Al(III)-coordinated protein and the
dysfunctional substate, nonOperative for docking, to understand molecular details of
the Al(III)-induwd lesion on calmodulin. By employing bioengineered calmodulin
with a single intrinsic ﬂuorescence emitter (tryptOphan), located at selectively deﬁned
sites along the protein backbone, ﬂuctuations of the local state can be monitored at
various regions of calmodulin subjected to Al(III) coordination.

Recent studies had indicated that Al(III) is primarily bound to membrane-
associated elements of the Ca(II)-phosphoinositide signalling pathway [38,39],
rather than to cytosolic calmodulin. However, the possibility cannot be excluded

10 153

that Al(III) is also binding to calmodulin-dependent enzymes associated with the
membrane [78], or to calmodulin involved in membrane channel functions [79].

Given the considerable amount of information available on structural details
(X-ray structure, amino acid sequence), and the availability of genetic tools [53] to
design calmodulins differing in amino acid composition, we believe that calmodulin
is a unique instrument to derive a microscopic picture on the impact of Al(III)
coordination on an important macromolecule.

Acknowledgement

One of us (V.V.) was supported in part by a fellowship, grant no. 200892/91-6,
from the Conselho Nacional de Desenvolvimento Cientiﬁco e Tecnologico (Brasil).

References

[1] CB. Klee and TC Vanaman. Adv. Protein Chem. 35 (1982) 213.
[2] C.L.A. Wang and PC Leavis, in M.P. Thompson (Ed), Calcium-Binding Proteins. VoL 1., CRC
Press, Boca Raton. FL. 1988, p. 97.
[3] DC. Wylie and T.C. Vanaman. in P. Cohen and CB. Klee (eds), Calmodulin. Elsevier. Amsterdam,
1988, p. 1.
[4] MJ. Berridge, Nature. 361 (1993) 315.
[5] P.W. Majerus, Annu. Rev. Biochem., 61 (1992) 225.
[6] H. Rasmussen, Biol. Chem. Happe-Seyler, 371 (1990) 191.
[7] M. Goodman, Progr. Biophys. Melee. Biol, 37 (1981) 105.
[8] A.R. Means. LA. Putkey and P. Epstein. in P. Cohen and CB. Klee (Eds), Calmodulin. Elsevier,
Amsterdam, 1988, p. 17.
[9] Y5. Babu. 1.5. Sack. TJ. Greenhough. CE Bugg, A.R. Means and WJ. Cook, Nature. 315 (1985) 37.
[10] Y.S. Babu. CE. Bugg and WJ. Cook. in P. Cohen and CB. Klee (Eds), Calmodulin, Elsevier.
Amsterdam. 1988, p. 83.
[11] H. Kawasaki and R.1-l. Kretsinger, in P. Shelterline (Ed), Protein Proﬁle, Vol. 1, Academic Press,
. London, 1994, p. 343.
[12] TN. Tsalkova and PL Privalov, .1. MOL Biol, 181 (1985) 533.
[13] N.CJ. Strynadka and M.N.G. James. Annu. Rev. Biochem., 58 (1989) 951.
[14] GB. Klee, in P. Cohen and CB. Klee (Eds), Calmodulin, Elsevier, Amsterdam. 1988. p. 35.
[15] IS. Evans, BA. Levine. R.J.P. Williams and MR. Womald, in P. Cohen and CB. Klee (Eds),
Calmodulin. Elsevier. Amsterdam, 1988. p. 57.
[16] P. Bayley. P. Ahlstrom. S.R. Martin and S. Forsen. Biochem. Biophys. Res. Commun., 120 ( 1984) 185.
[17] D.W. Margerum. G.R. Cayley, D.G. Weatherburn and GK Pagenkopf, in AB. Martel! (Ed),
Coordination Chemistry. VoL 2, American Chemical Society, Washington. 1978. p. l.
[18] C.L.A. Wang, Q. Zhan. T. Tao and J. Gergeley, Biophys. J., 22 (1986) 48a.
[19] O. Herzberg. 1. Moult and M.N.G. James, 1. BioL Chem. 261 (1986) 2638.
[20] 6.1... Eichhorn. Coord Chem. Rev. 128 (1993) 167.
[21] S. Yuan and A. Haug. Eur. J. Biochem.. 175 (1988) 119.
[22] B.A. Seatom J.P. Had. D.M. Engelman and EM. Richards, Biochemistry, 24 (1985) 6740.
[23] BE. Finn. T. Drakenberg and S. Forsen. FEBS Lett. 336 (1993) 368.
[24] LA. Cox. M. Comte, A. Mamar-Bachi, M. Miles and JJ. Schaer. in C. Gerday, L. Bolis and R
Gilles (Eds), Caldum and Calcium Binding Proteins. Springer-Veda; Berlin. 1988. p. 141.

154 u

[25] SR. Anderson and DA. Malencik. in W.Y. Cheung (Ed), Calcium and Cell Function, Vol. V1,
Academic Press. Orlando. FL. 1986. p. 1.

[26] P. James. T. Vorherr and E Carafoli. Trends Biochem. Sci, 20 (1995) 38.

[27] WE Meador, A.R. Means and EA. Quiocho. Science, 262 (1993) 1718.

[28] CF. Baes and 11.5. Mesmer, The Hydrolysis of Cations. Wiley, New York, 1976, p. 112.

[29] A.P. Martel] and RJ. Motekaitis. in T.E. Lewis (Ed), Environmental Chemistry and Toxicology
of Aluminum. Lewis. Boa Raton, FL. 1989, p. 3.

[30] R.B. Martin. in H. Sigel (Ed), Metal Ions in Biological Systems. Vol. 24, Marcel Dekker. New.
York. 1988. p. 1.

[31] ER. Nightingale. J. Phys. Chem. 63 (1959) 1381.

[32] C. Orvig. in OH. Robinson (Ed), Coordination Chemistry of Aluminum. VCH. New York. 1993.
p. 85.

[33] J.W. Akin. Progr. NMR Spectr. 21 (1989) 1.

[34] RM. Bertsch. in G. Sposito (Ed), The Environmental Chemistry of Aluminum. CRC Press, Boca
Raton. FL. 1989, p. 81.

[35] R.B. Martin. Clin Chem. 32 (1986) 1797.

[36] K. Panchalingam. S Sachedina. J.W. Pettegrew and T. Glonek. Int. J. Biochem.. 23 (1991) 1453.

[37] SJ. Karlik. GA. Egavish and GI. Eichhorn, J. Am. Chem. Soc.. 105 (1983) 602.

[38] B. Shi. K. Chou and A. Haug. Molec. Cell. Biochem.. 121 (1993) 109.

[39] A. Haug, B. Shi and V. Vitorello, Arch. ToxicoL. 68 (1994) 1.

[40] D.A. Buckingham. in GI... Eichhorn (Ed), Inorganic Biochemistry, Elsevier. Amsterdam. 1975. p. 3.

[41] N. Siegel and A. Haug. Biochim. Biophys. Acta. 744 (1983) 36.

[42] N. Siegel R. Coughlin and A. Haug. Biochem. Biophys. Res. Commun.. 115 (1983) 512.

[43] C. Weis and A. Haug. Arch. Biochem. Biophys., 254 (1987) 304.

[44] S. Forsen. HJ. Vogel and T. Drakenberg. in W.Y. Cheung (Ed), Calcium and Cell Function.
Vol. VI, Academic Plus. FL. 1986, p. 113.

[45] GA. MePhalen. N.C.J. Strynadka and M.N.G. James. Adv. Protein Chem. 42 (1991) 77.

[46] MT Record. CF. Anderson and T.M. Lohman. Quart. Rev. Biophys.. 11 (1978) 103.

[47] '1'. Kiss. in K. Burger (Ed), Biocoordination Chemistry. Ellis Horwood. New York. 1990, p. 56.

[48] N. Siegel and A. Haug. Inorgan. Chim. Acta. 79 (1983) 230.

[49] J.P. Glusker. Adv. Protein Chem, 42 (1991) 1.

[50] J.P. McManus. CW. Hague, BJ. Manden. M. Sikorska and AG. Szabo, J. BioL Chem. 265
(1990) 10358.

[51] J. Trewhella. WK. Liddle, D.B. Heidorn and N. Strynadka. Biochemistry, 28 (1989) 1294.

[52] G. Lindeberg. H. Bennich and A. Engstroem. Int. J. Peptide Protein Res. 38 (1991) 253.

[53] G.K. Ackers and ER. Smith. Annu. Rev. Biochem.. 54 (1985) 597.

[54] 1.1.. Miller. CM. Hubbard. BJ. Litman and T.L Macdonald, J. Biol. Chem. 264 (1989) 243.

[55] CG. Suhaya and A. Haug. Can. J. Biochem. Cell BioL. 63 (1985) 1167.

[56] P.K. Lambooy. Rf. Steiner and H. Sternberg. Arch. Biochem. Biophys.. 217 (1982) 517.

[57] DC. Laporte, B.M. Wierman and D.R. Strom, Biochemistry. 19 (1980) 3814.

[58] CR. Cantor and PR Schimmel. Biophysical Chemistry. Freeman. San Francisco. 1980. p. 1095.

[59] N. Siegel and A. Haug. Physiol. Plant. 59 (1983) 285.

[60] H. Zhang and P. Johnson. Int. J. Biochem.. 24 (1992) 1773.

[61] R.A. Nixon. 1.1". Clarke. K.B. Logvinenko, M.K.H. Tan. M. Hoult and F. Grynspan. J. Neurochem..
55 (1990) 1950.

[62] H. Weinstein and EL Mehler. Annu. Rev. Physiol., 56 (1994) 213.

[63] G. Careri. Order and Disorder in Matter. Benjamin—Cummings. Menlo Parks. CA. 1984.

[64] W17. van Gunsteren and M. Karplus. Biochemistry. 21 (1982) 2259.

[65] IL Finney. J. Phys (Paris) Colloq. 45 (C7) (1984) 197.

[66] A.R. Persht Trends Biochem. Sci, 12 (1987) 301.

[67] RC. Weber. TJ. Lukas, 12A. Craig, E. Wilson. M.M. King, A.P. Wiatkowski and D.M. Watterson.
Proteins”. Struct. Funct. Genet. 6 (1989) 70.

[68] LL Pascual-Ahuir. EL Mehler and H. Weinstein. Mol. Eng. 1 (1991) 231.

,2 155

[69] J.A. McCammon and SC. Harvey, Dynamics of Proteins and Nucleic Acids, Cambridge
Univiversity Press London, 1987.

[70] S. Pundak and RS. Roche. in B. de Bernard, G.l... Sottocasa G. Sandri, E Carafoli, A.N. Taylor.
T.C. Vanaman and R.J.P. Williams (Eds). Calcium Binding Proteins, Elsevier, Amsterdam, 1983.
p. 171.

[71] M. Chabberl. M.C. Kilhoﬂer. D.M. Watterson. J. Haiech and H. Lami, Biochemistry. 28 (1989) 6093.

[72] PB. O'Hara. KM. Gorski and M.A. Rosen. Biophys. J., 53 (1988) 1007.

[73] KC. Chou. BiOphys. Chem. 30 (1988) 3.

[74] S. Yuan and A. Haug. FEBS Lett.. 234 (1988) 218.

[75] G. Weber. S. Smrlata and M. Rholam. Biochemistry. 23 (1984) 6785.

[76] S. Scarlau. M. Rholam and G. Weber. Biochemistry, 23 ( 1984) 6789.

[77] S.R. Martin and RM. Bayley, Biochem. J. 238 (1986) 485.

[78] S.N. Levine. G.B. Sonnier and K. Abroe, Toxicol.. 65 (1990) 137.

[79] H.C. Yang. M.M. Reedy, C.L. Burke and GM. Strasburg. Biochemistry. 33 (1994) 518.

 

- "Illllllllllllllll