AN ABSTRACT OF THE THESIS OF
Steven D. Verhey for the degree of Doctor of Philosophy
in Botany and Plant Pathology
Title:
presented on August 17. 1993
Analysis of Putative Elements of Plant Signal
Transduction Chains.
Abstract approved:
Redacted for Privacy
Terri L. Lomax
The thesis begins with an introduction to signal transduction
and an analysis of current understanding of plant signal
transduction.
There are similarities between plants and animals,
but also key differences, including lack of protein kinase C and of a
cAMP signaling pathway in plants, and presence in plants of
calcium dependent protein kinase (CDPK), which has a kinase
catalytic domain contiguous with a C-terminal calmodulin-like
domain.
The next section examines protein kinase activity in the
plasma membrane (PM) of zucchini hypocotyls.
Zucchini PM
contains four or more polypeptides with calcium-requiring protein
kinase activity.
The enzymes appear to be tightly associated with
the PM, and at least three are recognized by monoclonal antibody
to soybean soluble CDPK.
Total proteins from several different
organs of zucchini seedlings contain kinases with molecular
weights similar to the hypocotyl PM enzymes.
In the third section
details of partial purification of the solubilized PM kinases are
presented.
Kinases which do not crossreact with anti-CDPK
monoclonal antibody were resolved by anion exchange from ones
which do crossreact.
Peptide mapping was used to test the
relationship between the kinases.
Results of peptide mapping
suggest that at least three types of protein kinase are present in
zucchini PM, two of which are immunologically similar to CDPK and
one of which is not. The last section concerns the potential for
testing interactions between PM protein kinases and plasma
membrane auxin binding proteins (ABP's) by use of photoaffinity
labeling of ABP's.
Causes of variable photoaffinity labeling by an
azido-IAA are considered.
Labeling of both the tomato mutant
diageotropica and the parent VFN membranes was inexplicably
inconsistent.
Analysis of Putative Elements of Plant
Signal Transduction Chains
by
Steven D. Verhey
A THESIS
submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Doctor of Philosophy
Completed August 17, 1993
Commencement June 1994
APPROVED:
Redacted for Privacy
Associate Professor of Botany and Plant Pathology in charge of
major
Redacted for Privacy
Chair, Department of Botany and Plant Pathology (r,
Redacted for Privacy
Dean of Grasp
School
Date thesis presented
Typed by Steve Verhey for
August 17, 1993
Steven D. Verhey
ACKNOWLEDGEMENTS
This thesis might not have been completed without the
support and encouragement of a number of people and at least one
cat.
The financial and less tangible support of Terri Lomax helped
make this work possible.
constant lessons.
Her enthusiasm and optimism teach me
All members of the Lomax lab have helped in
ways great and small, and all deserve mention:
Catherina Coen,
Chris Gaiser, Rosie Hopkins, Kyoung-Hee Kim, Andreas Nebenfiihr,
Dave Ray le, Peggy Rice. Dave Ray le deserves mention twice, for
his excellent editing and counsel, as well as his technical help.
Discussions with Constance White were always useful, even during
her Rush Limbaugh phase, and I always appreciated the advice of
Tom Wolpert and Chris Mundt. My parents and family were
always supportive, even when they must have doubted my
judgement.
Finally, much of the strength necessary to complete
this project derived from my partner Denise.
TABLE OF CONTENTS
Signal Transduction In Plants
Introduction
Second Messengers
G-Proteins
Protein Phosphorylation
Conclusion
Acknowledgements
Zucchini Plasma Membrane Contains Multiple CalciumRequiring Protein Kinases
Introduction
Materials And Methods
Results
Discussion
Acknowledgements
1
1
6
19
28
44
47
48
48
51
56
69
76
Zucchini Plasma Membrane Contains at Least Three Different
Species of Protein Kinase
77
Introduction
Materials And Methods
Results
Discussion
Acknowledgements
Analysis Of Factors Affecting Labeling Of Tomato And
Zucchini Auxin Binding Proteins By Azido Auxin
Introduction
Materials And Methods
Results
77
81
87
99
106
107
107
112
116
Discussion
Acknowledgements
124
133
Conclusion
134
Bibliography
137
Author Index
173
LIST OF FIGURES
Eau
Figure
1.
Elements of signal transduction:
plants and animals.
distribution in
46
Size and diversity of protein kinases in zucchini,
soybean, and oat.
60
Calcium requirement of the plasma membrane
kinases.
62
4.
Phosphoamino acid analysis.
63
5.
Western analysis: antigenic similarity of the
zucchini PM kinases to soybean CDPK.
65
6.
Plasma membrane kinase solubilization.
67
7.
Typical results of Q-Sepharose chromatography of
2.
3
CHAPS-solubilized PM.
91
8.
Typical G-100 elution profile.
92
9.
Silver-stained 2-D gels prepared with
phosphorylated and unphosphorylated postHAP material.
10.
In situ phosphorylation and western blot analysis
of Q-Sepharose peaks A, B, and C.
11.
12.
93
Cleveland maps of radiolabeled bands from QSepharose peaks A-C.
HPLC analysis of unphotolyzed and photolyzed
azido-IAA.
94
95
12 0
13.
HPLC analysis of azido-IAA following preparative
121
HPLC purification.
14.
Effect of azido-IAA purification on photoaffinity
labeling.
15.
Results of labeling experiments comparing dgt and
VFN.
122
123
LIST OF TABLES
1z
Table
1.
Differences and similarities between plants and
animals of relevance to signal transduction.
5
2.
Purification table.
96
3.
Protein purification techniques tested with
solubilized protein kinases.
97
Calculation of P(x) from results of Cleveland maps.
98
4.
ANALYSIS OF PUTATIVE ELEMENTS OF PLANT SIGNAL
TRANSDUCTION CHAINS
Signal Transduction in Plants
INTRODUCTION
Like all organisms, vascular plants face a wide variety of
changeable environmental conditions, including light, gravity, wind,
The genius of
plants is that, unlike mesozoa, they are able to cope with variable
conditions without resorting to locomotion. Partial solutions to
many problems are inherent in the anatomy and physiology of all
plants but active responses to changing conditions are necessary as
well. Plants must coordinate developmental processes to produce
structural preparations for and responses to environmental
eventualities, as well as adapt to changing conditions on a more
temperature, predators, pathogens, and moisture.
rapid timescale.
Active responses to both normal and plastic
developmental control take place at the cellular and subcellular
levels via recognition of intercellular and intracellular signaling
molecules.
Mechanisms by which intracellular signaling molecules exert
their effects on cells are generically known as "signal transduction"
which implies, by analogy to electronic transducers, the linkage of
a number of components which convert information from one form
to another.
In the animal paradigm, this involves perception of the
signal (for example, a peptide hormone) at the plasma membrane,
2
which leads directly or indirectly to the release of "second
messengers" inside the cell.
Changes in the intracellular
concentration of second messengers affects the biochemical status
of the cell, either directly, by affecting enzyme activity, or
indirectly, by affecting protein kinase or phosphatase activity.
Phosphorylation or dephosphorylation then modulates the activity
of cellular enzymes. In this chapter, we will focus on the events
which take place within cells following perception of stimuli in the
form of extracellular chemical messengers.
Plant hormone
mutants, control of gene expression, and the effects of
phytochrome are other important topics in the field of signal
transduction, but are beyond the scope of this chapter.
Because many aspects of signal transduction in animals are
understood in considerable detail, we will begin with a description
of signal transduction paradigms derived primarily from research
on animals.
Research on animal signal transduction is driven by
questions about neurotransmission, muscle action, inflammation,
immune responses, vision, carcinogenesis, and hormone action,
among many others.
While there is little obvious relationship
between these phenomena and the activities of plants, it has been
reasonably and widely assumed that there is room for similarity in
terms of the molecular and biochemical mechanisms that underlie
responses to extracellular signals. Arguments for such assumptions
include evolutionary relationships as well as the fundamental
biochemical similarities between all organisms (Janssens, 1988;
Palme, 1992; Westheimer, 1987).
Some important issues must be
3
borne in mind in this context, however. First, plants are indeed
different from animals in a great many ways. Some differences of
relevance to signal transduction are listed in Table 1.
In addition,
animal responses depend upon systems of biochemical events in
which each component depends upon the presence of every other
component.
Many elements of animal signal transduction chains
(or "webs," as it is now fashionable to call them) make no sense out
the context of the system of which they are a part. For example,
cyclic AMP is an extremely important signaling molecule in a
variety of evolutionarily divergent groups, and is part of a system
which includes receptors, linkages to adenylate cyclase, and cAMPdependent protein kinase.
Because cAMP is generally accepted to
have no widespread signaling role in vascular plants (see below),
one would not expect to find key elements of that system in plants,
except where involved with routine metabolism.
It appears, based
on analysis of eukaryotic microbes, that animals and plants
contained similar signal transduction systems when they diverged
in evolution (Janssens, 1988). However, there is little other a priori
reason, beyond biophysical realities, to expect animal signal
transduction systems to have simple parallels with signal
transduction systems in plants or other distantly related
organisms.
With these issues in mind, we will intitally describe in
intentionally general terms important components of wellunderstood (primarily animal) signal transduction systems.
The
details of the systems themselves will be left to the numerous
4
excellent reviews which will also give a greater sense of the degree
of complexity possible.
After introducing each nonplant paradigm
we will consider biochemical and molecular evidence for the
activity of each element in plants, and then discuss examples of
possible roles, if any, for each element in plant responses to
extracellular stimuli.
5
Table 1:
Selected differences between animals and plants
of relevance to signal transduction
Animals
Most rapid macroscopic
responses:
msec
Many highly specialized cell
Plants
Most rapid macroscopic
responses: sec or min
Fewer specialized cell types
types
Peptide hormones mostly
Peptide hormones rarely
Active circulatory system
Passive circulatory system
Centralized hormone production
Except meristems,
decentralized hormone
production
Motile
Sessile
Deterministic development--
Most cells remain totipotent-adult architecture plastic
adult architecture fixed
No cell wall
Cell wall present
Principal energy storage in
glycogen/fat
Principal energy storage in
sucrose/starch
Nervous system
No nervous system
6
SECOND MESSENGERS
Because many intercellular messengers in animal systems are
large and/or hydrophilic, a considerable amount of attention has
focused on the plasma membrane (PM), which forms a barrier
between such messengers and the cellular processes which they
modulate. Imbedded in the PM are specific receptors which
initially interact with signaling molecules and begin the process of
signal transduction across the membrane by stimulating the
production or release of "second messengers" inside the cell. We
will describe the primary second messengers of animal systems
and their possible presence in plants, leaving aside the question of
the extent to which plant growth regulators may exert their effects
at the PM.
Calcium.
Two important second messengers in animal
systems are calcium and cyclic AMP. Changes in calcium levels are
involved in muscle contraction, fertilization, glycogen synthesis and
Ion channels in the PM and
the endoplasmic reticulum (ER), which respond to various stimuli
breakdown, and many other processes.
including inositol 1,4,5-trisphosphate (IP3), mediate the flux of
calcium in animal cells, while Ca2+-ATPases maintain a low
cytosolic calcium concentration.
A primary mechanism by which
Ca2+ affects enzyme activity involves the binding of Ca2+ to
calmodulin (CAM), a protein which itself has no enzymatic activity,
but which acts by binding to other proteins in response to Ca2+.
7
Important targets of Ca2+-CAM regulation include protein kinases.
Calcium signaling pathways also interact with cyclic AMP signaling
pathways.
This interaction, or "crosstalk," between the systems
allows very complex modulation of responses.
Cyclic AMP in plants
will be discussed below.
Interest formed in the 1980s regarding the possible
involvement of Ca2+ in signal transduction in plants (for reviews,
see Hepler and Wayne, 1985; Gilroy et al., 1987).
An important
role for Ca2+ in plant signaling initially was supported by
circumstantial evidence such as the effects of inhibitors and
chelators (Hepler and Wayne, 1985).
With the demonstration of
the presence of calmodulin, Ca2+-dependent protein kinases
(Roberts and Harmon, 1992; see section on kinases), Ca2+ channels
(Johannes et al., 1991) and elements of an inositol phosphate
system (see next section), the case became much stronger.
Additional evidence linking changes in Ca2+ levels and signal
transduction has come from analysis of Ca2+ levels and flux
between the cytosol and extra- and intracellular Ca2+ stores. This
has been made much more quantitative by the use of fluorescent
Ca2+ indicator dyes and the tools of electrophysiology (Trewavas
and Gilroy, 1991).
In animals, membrane bound intracellular stores are one
very important source of Ca2+, which is released in response to the
binding of inositol 1,4,5-trisphosphate (IP3) to receptor channels.
Ca2+ is normally present in the plant cytosol at a concentration of
around 100-300 nM (Drobak, 1992).
Ca2+ concentrations in the cell
8
wall ranges around 1 mM; because the surface of the PM is
negatively charged it has been speculated that Ca2+ concentrations
there will be considerably higher than the average Ca2 +
concentration in the wall (Drobak, 1992), however this has not
been directly measured.
In plants, the ER, which is the primary
source of intracellular signaling Ca2+ in animals, contains
approximately 10 µM Ca2+. The highest concentration of Ca2+
inside plant cells is found in the vacuole, which contains roughly 10
mM Ca2+ (Drobak, 1992). Low plant cytosolic Ca2+ concentrations
are maintained by Ca2+-pumping ATPases and Ca2+/nH+
antiporters in the PM, ER, and tonoplast (Evans et al., 1991).
One promising area being examined by several laboratories is
the role of Ca2+ in stomatal control.
Stomata close in response to
abscisic acid (ABA) when guard cells lose osmoticum (in the form
of K+) and become less turgid (see Tallman, 1992, for review).
Schroeder and Hagiwara (1989) used patch-clamp techniques to
show that cytosolic Ca2+ levels in Vicia faba guard cells influence
the status of PM K+ channels. Effects of Ca2+ included blockage of
inward-rectifying K+ channels and activation of voltage-dependent
depolarizing anion channels.
Using the fluorescent Ca2+ indicator
dye Fura-2, McAinsh et al. (1990) found that application of
extracellular ABA to Commelina guard cells triggered an increase in
cytosolic free Ca2+, which was followed by stomatal closure.
There
is also evidence that stomatal movements are preceeded by
changes in cytosolic pH as well as cytosolic Ca2+ levels (Irving et al.,
1992).
9
The notion that Ca2+ was involved in stomatal closure was
supported by the experiments of Gilroy et al. (1990) who used a
photolabile caged form of Ca2+ to increase intracellular Ca2+ levels
in guard cells. Release of Ca2+ upon photolysis was followed by
narrowing of the stomatal aperture.
Release of a similarly caged
form of 1P3 caused a rise in intracellular Ca2+ and subsequent
stomatal narrowing (Gilroy et al., 1990).
Blatt et al. (1990) used a
similar approach with Vicia guard cells and found that release of
caged 1P3 inactivates K+ channels thought to be involved with K+
uptake and stimulates a second class of channels to promote K+
efflux.
Presumably these effects are mediated by Ca2+ (Blatt et al.,
1990), supporting the findings of Schroeder and Hagiwara (1989).
More recent results of Schroeder and Hagiwara (1990) suggest that
Ca2+-permeable ion channels present in the PM are indirectly
activated by ABA, rather than being ligand-gated ion channels that
respond directly to ABA.
Hormonal control of ion channel gating
has recently been reviewed (Blatt and Thiel, 1993).
While the above experiments seem to indicate Ca2+ is
involved in stomatal control, the extent of the role of Ca2+ in
stomatal control is called into question by other experiments.
When guard cells of Commelina were treated with ABA in one set
of experiments, stomatal closure occurred in 54 of 54 cells tested,
but strongly elevated Ca2+ levels were observed in only 4 of 54
cells.
Twenty-four cells showed no changes in Ca2+ levels (Gilroy et
al., 1991).
Schroeder and Hagiwara (1990) found effects of ABA on
10
cytoplasmic calcium in only 37% of the cells studied. More work is
necessary to sort out these apparent contradictions.
Adding to the confusion are results from barley aleurone.
Aleurone responds to gibberellic acid (GA) by producing several
hydrolytic enzymes, including a-amylase, which mobilize food
reserves stored in the starchy endosperm of the seed. Calcium is
required for this effect of GA, which is reversed by ABA (for
review, see Jones and Jacobsen, 1991). Use of Ca2+-sensitive dyes
showed that GA treatment was followed within 3-4 hours by a 3fold increase in Ca2+ in the peripheral cytoplasm of aleurone cells.
Subsequent treatment with ABA was followed by reduction of
cytoplasmic calcium levels (Gilroy and Jones, 1992).
This effect of
ABA is the opposite of ABA's effect (when an effect is observed) on
calcium levels in guard cells (cf. Gilroy et al., 1991).
Inositol 1.4.5-trisphosphate.
In animals, two important
second messengers are derived from the membrane itself when
receptors activate phospholipase C (PLC), an enzyme which is
specific for inositol phospholipids.
PLC cleaves the phosphodiester
bond of phosphatidylinositol 4,5-bisphosphate (PIP2) in the inner
leaflet of the PM, releasing the products inositol 1,4,5trisphosphate (IP3) and diacylglycerol (DAG). In animal cells, IP3 is
a primary effector of Ca2+ release, interacting with receptors on
intracellular Ca2+ storage sites to release Ca2+ into the cytoplasm
(Berridge, 1993).
Increased intracellular Ca2+ leads to changes in
the activity of a number of enzymes, some of which were discussed
above.
The role of DAG is to activate several subspecies of protein
11
kinase C (PKC). PKC is discussed at more length in the section on
protein kinases. DAG is also formed in mammalian cells by the
action of phospholipase D on phosphatidylcholine and
phosphatidylethanolamine, while the action of phospholipase A2
(PL A2) on phosphatidylcholine produces free fatty acids and
lysophosphatidylcholine (Asaoka et al., 1992).
A primary
mechanism by which PLC is activated in animals involves G-
proteins, which are themselves activated when receptors in the PM
bind their ligands.
See the section on G-proteins for more detail.
Research on the role of inositol phospholipid-derived
molecules in plants has focused on several areas. These include
demonstration of the presence of PIP2 and IP3 in plant cells,
characterization of the enzymes of inositol phospholipid
metabolism, and investigations on the effects of IP3 and DAG in
plant cells. Issues related to the role of IP3 in plants are, of course,
closely related to the effects of Ca2+.
Phosphatidylinositol (PI) is about twice as abundant in plant
membranes as it is in animal membranes, comprising 8-15 mol% in
plant cell membranes.
In contrast, the phosphorylated forms of PI,
phosphatidylinositol 4-phosphate (PIP) and phosphatidylinositol
4,5-bisphosphate (PIP2), which are so important in animal signal
transduction, are present in much lower amounts in plants than in
animals. This, along with technical difficulties and the presence of
numerous interfering compounds and metabolic pathways, has
hampered research in this area (for reviews see Boss, 1989;
Einspahr and Thompson, 1990; Dr Obak, 1992; Cote and Crain,
12
In spite of these difficulties, it is now clear that vascular
1993).
plants do contain PIP and PIP2 (Cote et al., 1989; Irvine et al.,
1989), as do algae (Einspahr et al., 1988; Irvine et al., 1989).
The enzymes of PI metabolism and IP3 release, which
catalyze the sequential phosphorylation of PI to form PIP and PIP2,
as well as the cleavage of IP3 from PIP2, are also present in plant
PI-4-hydroxy kinase and PIP-5-hydroxy kinase have been
cells.
demonstrated to be present mainly in plant PM (Sandelius and
Sommarin, 1986; Sommarin and Sandelius, 1988).
In animals these
enzymes are associated with the cytoskeleton as well as with the
PM.
An association with the cytoskeleton has recently been shown
in plants as well (Xu et al., 1992; Tan and Boss, 1992). IP3-specific
phosphatases, necessary to help remove second messenger after
signaling events, are present in the cytosol of barley (Martinoia et
al., 1993).
PLC activity has been reported to be associated with
plant PM (Einsphar and Thompson, 1989; Yotsushima et al., 1993).
Thus, most or all of the components needed for generation of an
IP3 signal are present in plants.
Presumably these or similar
enzymes are also involved in metabolism of inositol phospholipids
and phosphorylated inositols.
If products of PIP2 hydrolysis are involved in signaling, it is
expected that changes in concentration of components of the
phosphatidylinositol system should be detectable in response to
stimuli.
Changes in cytoplasmic Ca2+ in response to IP3 should
likewise be observable, as should changes in cell function. There
have been several reports correlating changes in metabolism of PI
13
cycle components with the application of different stimuli including
light, auxin, fusicoccin, cytokinin, gibberellic acid, osmotic shock,
and cell wall-degrading enzymes (reviewed in Drobak, 1992).
Some of these reports have been subject to criticism on technical
grounds (e.g. Boss, 1989), and few appear to have been reproduced
in other laboratories. This lack of replication is due in part to the
complexity of the system and in part to the use of a number of
different research organisms.
For example, at least four
laboratories have looked at the effects of auxin using four different
organisms, and have found effects on five different parameters,
none of which were assayed by any other research group (see
Drobak, 1992, Table 2).
There have been reports which attempt to
directly link release of caged IP3 inside cells with changes in cell
function, but these results are not as straightforward as they might
appear; see the section on Ca2+. In addition, Moreau and Preisig
(1993) suggest that treatments with materials such as fungal
elicitor and cellulase may have less specific effects on membrane
lipids than may be evident from the plant IP3 literature.
It has been suggested that an IP3 system is only functional in
a limited number of often-stimulated cells, including guard cells,
pulvini, and stem epidermal cells (Trewavas and Gilroy, 1991).
An
additional possibility is that heterogeneity of plant tissues and
asynchrony of responses act to compound quantitation problems
which stem from low concentrations of IP3 during a response and
the small cytoplasmic volume of vacuolated cells (Drobak, 1992).
Drobak (1992) has calculated that typical plant cells might be
14
expected to contain less than 0.2 amol [attomoles, 10-18 mol] of 1P3
during a response.
The most sensitive assays for 1P3 operate in the
sub-pmol range, necessitating the near-synchronous response of
more than 106 cells for the production of a detectable signal even
under optimal conditions.
This calculation underscores an
unfortunate aspect of the use of guard cells as models for signal
While it is possible to obtain relatively
large quantities of guard cell protoplasts (Kruse et al., 1989),
transduction research.
producing enough material for extensive classical biochemical
analysis (i.e. hectogram or kilogram quantities) would be laborious,
or more likely, impossible.
This difficulty will hamper important
avenues of research in this area.
Cyclic AMP. As noted above, cyclic AMP (3',5'-cyclic
adenosine monophosphate; cAMP) and Ca2+ are two vital second
messengers in animals.
Levels of cAMP in animal cells are
controlled by adenylate cyclase, which interacts with a variety of
PM-localized receptors and produces cAMP from ATP in response
to extracellular stimuli, and by phosphodiesterase, which degrades
cAMP. As with Ca2+, signaling by cAMP in animals culminates in
activation of protein kinases, in this case the cyclic AMP-dependent
protein kinases (PKAs).
Cyclic AMP has been demonstrated to function in signal
transduction in diverse organisms including bacteria, eukaryotic
microbes, and animals. After the discovery of a widespread role for
cAMP in these organisms, there developed considerable interest in
possible functions for cAMP in plants (Amrhein, 1977).
However,
15
in spite of concerted efforts spanning many years, no such role for
cAMP has yet been found.
The very presence of cAMP in plant tissues remains
controversial.
Spiteri et al. (1989) examined seeds and other
tissues of five species of plants, at least three of which had
previously been reported to contain cAMP (cf. Brown and Newton,
1981, Table 1).
Spitieri and coworkers used a highly sensitive
radioimmunoassay technique, and carefully eliminated possible
artifactual origins of cAMP.
Levels of cAMP above the detection
limit (<0.5 pmol/gFW) were found in only two of 14 tissues tested;
cAMP in the two positive samples was reported to be of nonplant
origin.
Cyclic AMP levels in animals are on the order of 100-500
pmol/gFW (Robison et al., 1971; cited in Amrhein, 1977).
Reports
identifying cAMP in plants appear from time to time (e.g. Lemna,
Gangwani et al., 1991; maize, Janistyn and Ebermann, 1992). Such
recent reports evidently do not suffer from the technical problems
which have plagued the field in the past (Amrhein, 1977), but, as is
the case in the Lemna and maize reports, quantitative data are
often not shown.
Most reliable accounts make it clear that plants,
when they contain cAMP at all, contain much lower amounts than
do animal tissues (i.e. <<100 pmol/gFW). As with 1P3, it has been
suggested that the presence of cAMP in a minority of signaling cells
in a plant tissue might be undetectable in a homogenate of the
entire tissue (Trewavas and Gilroy, 1991).
Although direct, unequivocal demonstration of cAMP in
vascular plant tissues is problematic, several papers have
16
suggested that DNA and protein sequence comparison data may
indicate a role for cAMP in plants (Trewavas and Gilroy, 1991).
Working with tobacco, Katagiri et al. (1989) found two DNAbinding proteins, TGA 1 a and TGA lb, with homology to basic-
leucine zipper regions of mammalian nuclear factors including CREB
(cyclic AMP response element binding protein) and two others.
DNA binding by CREB is thought to be modulated via
phosphorylation by cAMP-dependent protein kinase (PKA), protein
kinase C, and/or casein kinase II (Gonzalez et al., 1989; Hunter and
Karin, 1992).
Katagiri et al. (1989) suggest TGA 1 a and TGA 1 b may
be involved in PKA-mediated signaling, although only TGA 1 b
contains a consensus PKA phosphorylation site, KRXXS/T; in
animals RRXS/T is most commonly used (Scott, 1992). A third
CREB-like sequence, VBP1, from V. faba contains neither PKA
phosphorylation site (Ehrlich et al., 1992).
Like TGA1 a and TGA lb,
VBP1 contains homology to the basic-leucine zipper region and can
bind DNA.
All three plant CREB-like proteins contain
phosphorylation sites for casein kinase II (S/TXXE; Kemp and
Pearson, 1990), as well as numerous other serine residues, leaving
open the possibility that DNA binding is modulated by kinases
other than PKA. Examination of plant CREB-like sequences thus
actually appears to provide evidence against the involvement of
cAMP in plant signal transduction.
Caution should also be used in
drawing conclusions from the presence of CRE binding sites
adjacent to plant genes: of the many CRE-binding proteins, CREB is
the only one which has been identified as a mediator of cAMP
17
action (for reviews see Karin and Smeal, 1992; Hunter and Karin,
1992).
Another line of evidence concerning a role for cAMP comes
from experiments on protein kinase activity.
PKA activity is often
assayed using an artificial substrate known as Kemptide, a 7residue peptide containing the PKA phosphorylation site RRAS
(Kemp, 1979).
Petunia extracts contain kinase activity which is
capable of phosphorylating this substrate (Polya et al., 1991). The
petunia Keptide kinase activity is responsive to bovine PKA
regulatory subunit in a cAMP-dependent manner, but endogenous
The lack of
endogenous cAMP regulation could be the result of differential
cAMP-responsive regulation was not observed.
degradation of the regulatory subunit, or it could be indicative of
the presence of a cAMP-independent kinase with the same
phosphorylation target sequence as PKA (Polya et al., 1991).
It
should be noted that PKA is not the only kinase capable of
phosphorylating Kemptide (Kemp and Pearson, 1990).
Plants have
been demonstrated to contain proteins that are responsive to
phosphorylation by mammalian PKA, such as maize leaf sucrose-
phosphate synthase (Huber and Huber, 1991) and
phosphoenolpyruvate carboxylase (Terada et al., 1990).
Endogenous kinases capable of phosphorylating these plant
proteins are apparently cAMP-independent; phosphorylation site
specificity for these kinases is unknown.
There are recent reports which suggest the involvement of
cAMP in some plant responses, indicating continued interest in this
18
topic.
One example is the downregulation of elicitor-induced
phenylalanine ammonia lyase (PAL) in suspension cultures of
Phaseolus vulgaris (Bolwell, 1992). In this work, downregulation
of PAL was observed in response to added forskolin, an activator of
adenylate cyclase activity in animal cells.
Addition of both elicitor
and forskolin led to increased levels of cAMP of two to seven-fold
over the basal level of around 4 pmol/gFW (Bolwell, 1992). The
suggestion by Morsucci et al., (1991) that cAMP may be involved in
stomatal movement in V. faba is based on transmission electron
microscope cytochemistry and incubation of epidermal peels in
cAMP solutions.
Concentrations of 0.1-0.5 mM cAMP led to
maximum stomatal apertures.
However, given the low cAMP levels
present in plants, it could be argued that the levels of cAMP used
by Morsucci et al. (1991) were not physiological.
Pending further investigations of the possible second
messenger regulation of the Kemptide kinase and other aspects of
cAMP in vascular plants it must be stated that evidence for the
involvement of cAMP in plant signal transduction is extremely
limited, or negative.
There is a need for a comprehensive,
authoritative, impartial, and critical review of this subject.
19
G-PROTEINS
Guanine nucleotide-binding regulatory proteins (G-proteins)
carry out a wide variety of regulatory roles, and their presence has
been documented in organisms ranging from E. coli to mammals
(for reviews see Simon et al., 1991; Hepler and Gilman, 1992).
G-
proteins themselves are a diverse group, but all share the ability to
undergo conformational changes in response to the binding of GTP
or GDP which then affects the ability of G-proteins to interact with
other proteins.
One class of G-proteins, consisting of a single polypeptide
chain, comprises the so-called "small G-proteins."
Small G-proteins
are involved in functions including intracellular vesicle trafficking
and secretion, cell growth and differentiation, protein synthesis,
and cytoskeleton organization.
One example is the well-known
oncogene ras, which encodes a small G-protein (for review, see Hall,
1990; Hall, 1992).
Genes with homology to ras have been found in
Arabidopsis (Anai et al., 1991), and Zea mays contains genes with
homology to the ras-related ypt gene family (Palme et al., 1992).
The function(s) of these genes in plants has not yet been
established.
Heterotrimeric G-proteins make up the second class of G-
proteins, and the group most associated with transmembrane
signal transduction (Simon et al., 1991; Hepler and Gilman, 1992).
Heterotrimeric G-proteins (more commonly simply called G-
20
proteins) consist of three subunits, a (39-46 kD), 13 (37 kD), and y (8
kD). The f3 and y subunits are present as a tightly associated
complex that interacts with a as a unit. G-proteins are attached to
the inner leaflet of the PM, via prenylation of y and myristoylation
of some species of a. The unstimulated G-protein complex includes
a molecule of GDP, tightly bound to a. The presence of GDP
maintains the affinity of a for 13T, when the complex makes contact
with a hormone receptor which has bound ligand, the rate of
dissociation of GDP is increased. GDP is exchanged for GTP, and a
loses its affinity for 13y. The a subunit is then free to interact with
an effector such as PLC, ion channels, or adenylate cyclase.
There
is growing evidence, at least in some cases, that 13y interacts with
the same or a different effector as well. The a subunit possesses a
constitutive GTPase activity. Once GTP has been cleaved to GDP the
subunits reassociate, returning the system to the unstimulated
state (Hepler and Gilman, 1992). The rate of GTP hydrolysis thus
determines the lifetime of the interaction between a subunits and
their effectors.
Biochemical examination of multiple G-proteins, along with
molecular cloning, has led to the identification of many different Gprotein a subunits (Ga). All share the biochemical characteristics
of GDP binding and GTP binding and hydrolysis. More than 30
genes for Ga have been cloned (Simon et al., 1991). Analysis of the
cDNA clones suggests that 20% of the amino acids are invariantly
conserved, and that Ga fall into four main classes (Gas, Gai, Gaq,
and Goc12). Each class consists of two or more subclasses, all of
21
which differ in terms of various biochemical properties, including
rate of GTP hydrolysis, nucleotide binding affinities, and
susceptibility to ADP-ribosylation by pertussis toxin (PTX) and
cholera toxin (CTX). PTX dissociates susceptible G-proteins from
their receptors, blocking signal transduction, while CTX activates Gproteins by inhibiting GTPase activity.
By treating intact cells with
PTX or CTX and examining the effect, evaluation of G-protein
function and correlation with G-protein type and subtype is
possible (Hepler and Gilman, 1992). Different Go also display
different tissue distribution and receptor associations (Simon et al.,
1991).
Over one hundred G-protein-linked receptors have been
described in mammals alone.
With a single possible exception
(Nishimoto et al., 1993), all G-protein-linked receptors described to
date contain seven transmembrane domains.
Distinct subtypes of
receptors that respond differently to the same ligand have been
identified.
Because receptor subclasses can be coupled to different
second messenger pathways and to ion channels, or, alternatively,
multiple receptor subtypes can activate a single effector, very
complicated signaling networks are possible (Simon et al., 1991).
G-protein-linked receptors have not yet been identified in plants.
Initial clues to the presence of G-proteins in vascular plants
came in the form of evidence of binding of [35S] GTP-yS, a
nonhydrolyzable analogue of GTP, to solublized microsomal
proteins from pea epicotyls (Hasunuma et al., 1987).
A somewhat
more thorough report of [35S] GTP-yS binding to microsomal
22
proteins from etiolated zucchini hypocotyls also provided the first
immunological evidence of plant polypeptides resembling Gproteins (Jacobs et al., 1988). Two bands, at 50 and 33 kD, were
seen on western blots probed with antibody to Gas. Kinetic
analysis of binding showed an overall Kd of 300 nM; Scatchard
analysis suggested the presence of more than one type of binding
site (Jacobs et al., 1988).
Proteins with characteristics of animal Ga, such as [35S] GTP-
yS binding and antigenic similarity, were reported in a number of
plants including V. faba, A. thaliana, and C. communis by Blum et
al. (1988).
Antigenic similarity was based on crossreactivity using
antibody raised to the highly conserved nucleotide binding region
of mammalian Ga subunits (Ga_common).
Immunostaining plant
proteins were described at 31-38 kD, slightly smaller than the
range of sizes of mammalian Ga (39-52 kD; Stryer and Bourne,
1986).
Using a different approach to identifying GTP-binding
proteins by molecular weight, Zbell et al. (1990) employed [35S]
GTP-TS overlays on nitrocellulose blots, and reported binding
proteins with molecular weights of 24 and 28 kD in soybean PM.
Similar experiments with zucchini PM yielded a binding protein at
30 kD (Perdue and Lomax, 1992). The zucchini and soybean PM
GTP-binding proteins are considerably smaller than the size range
for known Ga subunits. Parallel experiments with membrane
proteins from rat revealed a single [35S] GTP-yS binding protein, at
28 kD, indicating that Ga does not bind [35S1 GTP-TS under these
conditions (Zbell et al., 1990). The soybean and zucchini [35S] GTP-
23
yS binding proteins are similar in size to small G-proteins (Hall,
1990).
Conservation of Ga sequences allows the application of
molecular approaches to questions about G-proteins in plants.
Using PCR, a gene for a Ga homologue was cloned from Arabidopsis
(Ma et al., 1990). This sequence was subsequently used to clone
cDNA sequences for a related Ga from tomato (TGA1; Ma et al.,
1991).
The predicted sequences of the tomato and Arabidopsis
proteins are very similar to one another, with 85% identical and
93% similar (identical plus conservative replacements; Ma et al.,
1991). TGA1 is 27% identical and 51% similar to rat Gas and 34%
identical and 59% similar to the Ga of bovine transducin. The
tomato sequence encodes a 44.9 kD protein which includes all of
the conserved regions characteristic of known Ga, as well as a
consensus myristoylation and cholera toxin ADP-ribosylation sites.
Southern analysis of tomato and Arabidopsis genomic DNA indicate
that the plant Ga sequences are each single copy genes.
Stringency
of the tomato Southern experiment was such that sequences of
>60% similarity were detectable. Known Ga sequences that are less
than 60% homologous are considered to be in different classes,
leaving open the possibility that tomato contains genes encoding
members of other Ga classes (Ma et al., 1991). Indeed, lower
stringency probing of Arabidopsis Southerns indicated other Ga
sequences were present (Ma et al., 1990)
As it became clear that plant cells contain GTP-binding
proteins, workers began to try to link GTP-binding with application
24
of extracellular stimuli or activation of effectors.
Thus, Zaina et al.
(1990, 1991) reported preliminary evidence of an effect by auxin
on GTP-binding and on GTPase activity.
Other laboratories have
detected changes in phospholipase C and phospholipase A2 activity
in response to auxin (Ettlinger et al., 1988; Zbell et al., 1989; Andre
and Scherer, 1991). In animals, PLC and PLA2 activity are often
controlled via G-proteins (Simon et al., 1991).
Ion channels are one class of effectors controlled by Gproteins in animals. Regulation of plant ion channels has recently
been reviewed (Blatt and Thiel, 1993).
Using patch-clamp
experiments, Fairley-Grenot and Assmann (1991) provided
evidence for G-protein regulation of potassium channels in guard
cells of V. faba. Perfusion of the cytosol with 100-500 gM GDPI3S,
which binds tightly to Ga and locks G-proteins in the inactive GDP-
bound state, resulted in an increase in inward-directed K+ channel
current. Treatment with GTP-7S, which locks G-proteins in the
active GTP-bound state, had the opposite effect in that current was
reduced. Thus, the results indicate that activation of a G-protein
results in a decrease of inward K+ current.
Interestingly, both CTX
and PTX decrease K+ current (Fairley-Grenot and Assmann, 1991).
This is contrary to the expectation that CTX would decrease current
(by activating a G-protein) and PTX would either have no effect or
increase current, depending upon the nature of the upstream
activation of the G-protein, i.e. the type of receptor binding
involved, and whether the Ga involved had a target site for PTX
(Simon et al., 1991). The authors suggest the possibility of a PTX-
25
sensitive G-protein with a function opposing that of the CTXsensitive G-protein (Fairley-Grenot and Assmann, 1991). Lee et al.,
(1993) have recently reported that pretreatment of epidermal
peels of C. communis with PTX, or microinjection with GTP-yS into
guard cells promotes stomatal opening.
Further work will be
necessary to reconcile this result with those of Fairley-Grenot and
Assmann (1991).
There have been at least two reports of changes in G-protein
behavior in response to extracellular stimuli.
Working with PM
from etiolated pea epicotyls, Warpeha et al. (1991) found that lowfluence blue light induces a GTPase activity 1-2 minutes after
A 40 kD polypeptide present in the PM was recognized
by antiserum directed against the Go subunit of transducin. CTX
irradiation.
and PTX both ADP-ribosylated a protein with the same molecular
weight as the immunostained protein.
Interestingly, PTX had no
effect on the 40 kD protein in the presence of blue light and GTP,
while CTX was effective only in the presence of blue light and GTP.
Finally, using a nonhydrolyzable UV-crosslinkable analogue of GTP,
a 40 kD protein was found to bind GTP only in the presence of blue
light (Warpeha et al., 1991).
Given the characteristic structure of
animal G-protein-linked receptors which recognize chemical
signals, it will be interesting to elucidate the nature of the receptor
which links blue light with a G-protein.
A second approach to the role of GTP-binding proteins in
responding to extracellular stimuli involved the use of soybean
suspension culture cells (Legendre et al., 1992). The soybean cells
26
contained a 45 kD protein which cross-reacted with antibody
raised against the Gacommon sequence. The antigen binding
fragment (Fab) of the anti-Gacommon antibody was delivered into
the cell via receptor-mediated endocytosis, a process facilitated by
biotinylation of the protein.
When this was followed by treatment
with the fungal elicitor polygalacturonic acid (PGA), cells responded
by producing up to 10-fold more H202 than controls involving
heat-treated Fab or biotinylated BSA.
The elicitor response could
also be triggered by treatment of the cells (in the absence of PGA)
with mastoparan, a 14-residue oligopeptide from wasp venom. In
mammalian cells, mastoparan activates G-proteins by mimicking
the G-protein-recognition region of the activated receptor proteins.
These results suggest that at least two G-proteins may be involved
in the oxidative response to PGA: because introduction of Fab into
the cell, with the presumed neutralization of the 45 kD G-protein,
prolongs and intensifies the response to PGA, the authors suggest
that this G-protein is involved in turning off 11202 production.
On
the other hand, mastoparan, which directly activates G-proteins,
may affect a G-protein involved in the initiation of the response to
elicitor, since it activates the oxidative burst in the absence of
elicitor (Legendre et al., 1992).
Compared to the situation with plant protein kinases,
relatively little is known about G-proteins in plants. Still, even at
this early stage in the development of this field, several cellular
events have been linked with G-proteins.
These include perception
of blue light (Warpeha et al., 1991), response to auxin (Zaina et al.,
27
1990), response to elicitor (Legendre et al., 1992), regulation of ion
channels (Fairley-Grenot and Assmann, 1991), and stomatal
opening (Lee et al., 1993). There is every reason to believe that
the latter half of 1993 and 1994 will bring a rapid increase in our
understanding the roles of G-proteins in plants.
28
PROTEIN PHOSPHORYLATION
Protein kinases. Protein kinases are a large and diverse
group of phosphoryl-transferases which transfer the terminal
phosphate from ATP (or, less commonly, GTP) to substrate proteins.
Covalent phosphorylation induces conformational changes in the
structure of enzymes, modulating their activity (Sprang et al.,
1988; Browner and Fletterick,1992).
Protein kinase C (PKC) and PKA are the most well understood
of the many protein kinases which have been described in animals.
There are ten subspecies of PKC, comprising three main groups.
The PKC subspecies differ in their response to a number of
activators, including DAG, Ca2+, phosphatidylserine,
lysophosphatidylcholine, and cis-unsaturated fatty acids.
Not all
PKC subspecies are activated by DAG or Ca2+, but all require
phosphatidylserine.
PKC is also activated by limited proteolysis
which renders the enzyme Ca2+-independent (Nishizuka, 1988).
PKA is activated by the presence of cAMP. Inactive PKA consists of
two catalytic subunits complexed with two cAMP-binding
regulatory subunits.
The catalytic subunits are activated when the
regulatory subunits bind cAMP and dissociate.
As with the PKC
family, there are multiple forms of PKA, and further complexity is
possible because there also exist multiple forms of the regulatory
subunit.
For example, in humans there are three genes for
29
catalytic subunits and four genes for regulatory subunits (for
review, see Scott, 1991).
It would be difficult to overemphasize the degree to which
protein kinases are now known to be involved with the functioning
of animal cells (Fischer and Krebs, 1989). If work on
phosphorylation in animal systems is still in its fast-growing
adolescence, research on the role of protein kinases and
phosphatases in plant biology may only now be leaving its infancy.
Indeed, our understanding of the presence and roles of protein
phosphorylation in animal cells has grown so vast that the topic is
rarely if ever reviewed as a whole. While this review cannot be
comprehensive, published reports on phosphorylation in plants will
still fit in a few, if bulging, file folders. Progress toward an
understanding of phosphorylation in plants began as it did in
animals, with biochemical approaches.
Recently molecular
techniques have greatly aided the search.
Most initial attempts to identify protein phosphorylation
activity in plants focused on cAMP-regulated phosphorylation,
which has still not been demonstrated in vascular plants.
The first
biochemical evidence for cAMP-independent protein
phosphorylation in carrot root and tobacco and chinese cabbage
leaf appeared in 1972 (Ralph et al., 1972). Two more reports
appeared in 1973 (Keates, 1973;
By 1976 there
was sufficient evidence for and interest in phosphorylation in
Trewavas, 1973).
plants to warrant a review of protein modification by
phosphorylation (Trewavas, 1976).
This review provides a
30
snapshot of an early phase of both animal and plant kinase
research and is a useful source of perspective. In spite of general
agreement at that time that cAMP was not involved in signaling in
plants (Amrhein, 1977; Trewavas, 1976), demonstration of cyclic
AMP independence was a key feature of plant kinase reports
through at least 1982 (Gowda and Pillay, 1982; Erdmann et al.,
1982; Yan and Tao, 1982). Regulation of phosphorylation in these
reports was at the substrate level (i.e. histone-specific, caseinspecific, etc.).
The lack of a kinase-activating molecule such as
cAMP hampered work in plants as much as its presence helped
progress in animals.
The first reports of protein kinase activity modulation at
other than the substrate level involved Ca2+-stimulated protein
kinase activity in pea shoot membranes (Hetherington and
Trewavas, 1982) and zucchini hypocotyl membranes (Salimath and
Marine, 1983).
Soluble Ca2+-calmodulin activated kinase activty
was reported in wheat germ extracts (Polya and Davies, 1982), and
phosphorylation of specific proteins in corn was shown to be Ca2+sensitive (Veluthambi and Poovaiah, 1984).
Reports of Ca2+
stimulated phosphorylation were particularly interesting because
changes in cytoplasmic Ca2+ had been demonstrated to be involved
with developmental and other changes in plant cells (Hepler and
Wayne, 1985).
Work on phosphorylation and dephosphorylation in plants
began to accelerate to the point that a 1987 review (Ranjeva and
Boudet, 1987) was able to report considerable phenomenology
31
which confirmed that phosphorylation occurs in all parts of plant
cells including nuclei, mitochondria, plastids, and microsomal
membranes.
In addition, phosphorylation-dephosphorylation had
been shown to affect the activity of four enzymes including, of
most relevance to signal transduction, the PM H4--ATPase (Ranjeva
and Boudet, 1987).
Interest in Ca2+-stimulated phosphorylation in soybean
(Putnam-Evans et al., 1986) led to the discovery of a Ca2+ activated
protein kinase which was calmodulin independent (Harmon et al.,
1988).
The soybean protein kinase was purified (Putman-Evans et
al., 1990), and subsequently microsequenced and cloned (Harper et
al., 1991).
The deduced amino acid sequence for the clone
demonstrated that it was a member of a unique class of protein
kinases, the calcium-dependent protein kinases (CDPK).
The CDPK
are characterized by a structure which includes an N-terminal
kinase catalytic domain contiguous with a C-terminal calmodulinlike Ca2+-binding domain (Harper et al., 1991).
Sequence analysis
is currently the sine qua non for identification of a kinase as a
CDPK (Harper et al., 1993). Monoclonal antibodies to soybean CDPK
are available (Putman-Evans et al., 1989), but immunological
results must be interpreted with caution because the antibodies
recognize an epitope near the nucleotide binding site in a region
which is fairly well conserved in all kinases (Alice Harmon,
University of Florida, personal communication).
Anti-CDPK
antibodies have been found to crossreact with proteins from many
organisms and subcellular locations including oat and zucchini PM,
32
soybean root nodules, corn, onion, Mougeotia, Tradescantia, and
Chara (Roberts and Harmon, 1992).
CDPKs have recently been
comprehensively reviewed (Roberts and Harmon, 1992;
Roberts,
1993).
Because of the important functions present there, as well as
by analogy to animal signal transduction systems, it has long been
assumed that the PM would be the site of some degree of protein
kinase activity.
This supposition was supported by reports of
protein kinase activity associated with PM of pea bud
(Hetherington and Trewavas, 1984), suspension cultured ryegrass
(Polya et al., 1984), oat root (Schaller and Sussman, 1988; Schaller
et al., 1992), silver beet leaf (Kulcis and Polya, 1988), corn (Ladror
and Zielinski, 1989), and zucchini (Verhey et al., 1993). Several of
these reports involved Ca2+-stimulated kinase activity
(Hetherington and Trewavas, 1984; Schaller and Sussman, 1988;
Schaller, et al., 1992; Verhey et al., 1993).
An enzyme, termed pp18, which was presumed to be
responsible for at least part of the kinase activity in pea was
purified (Blowers et al., 1985) and found to have several curious
characteristics, including its size (18 kD), which was considerably
smaller than known protein kinases (Blowers and Trewavas, 1989).
However, an enzyme with similar molecular weight and
biochemical characteristics in sugar cane has recently been
demonstrated to be a nucleotide dinucleotide kinase (Michael
Harrington, University of Hawaii, personal communication),
33
suggesting the identification of ppl8 as a protein kinase may have
been premature.
Other protein kinases have been identified in the PM fraction
of several species. Oat root PM contains a 79 kD lipid-activated,
calcium-stimulated protein kinase which crossreacts with antibody
to soybean CDPK (Schaller et al., 1992). Upon limited digestion
with trypsin, the 79 kD kinase is converted into a smaller, soluble,
lipid-independent polypeptide which is unlike PKC.
There have
been several reports of soluble and PM-associated protein kinases
which share some biochemical similarities with one another (Kulcis
and Polya, 1988; Klimczak and Hind, 1990). PM from various
organs of zucchini contain a number of calcium-requiring protein
kinase subspecies, some of which crossreact with anti-CDPK
antibody (Verhey et al., 1993; Verhey and Lomax, unpublished
data).
This biochemical demonstration of multiple protein kinases
with similar characteristics is predicted by molecular data
indicating the presence of a gene family (Harper et al., 1991),
although their localization to the PM is not.
Biochemical approaches to identifying protein kinases have
the advantage that they identify enzymatically active kinases, but
technical difficulties can make the biochemical route arduous.
Because of these difficulties, most current work on plant
kinases involves molecular approaches.
protein
Analysis of the nucleic
acid and amino acid sequences of a great many non-plant protein
kinase and putative protein kinases has made molecular
approaches possible in plants, because protein kinase sequences
34
contain invariant amino acid residues in each of eleven conserved
subdomains (Hanks et al., 1988).
The first example of the use of conserved sequence
information in the molecular cloning of plant transcripts encoding
putative protein kinase homologues involved bean and rice
(Lawton et al., 1989) This approach used sequential probing of
cDNA clones with partially degenerate oligonucleotides to two
subdomains, based on data presented by Hanks et al. (1988).
Sequences corresponding to a 609 amino acid polypeptide from
bean (PVPK-1) and a 536 amino acid polypeptide from rice (G11A)
were cloned. The plant sequences bore considerable similarity to
known kinase sequences, with only one significant single amino
acid difference, the nonconservative substitution of an aspartate
(single letter code D) for an invariant glycine (G) in the virtually
invariant subdomain VII sequence which for nonplant sequences
reads DFG. Subdomain VII is associated with nucleotide binding
(Hanks et al., 1988; Knighton et al., 1991). No further reports have
linked PVPK-1 or G1 1A with catalytic activity or a functional role
in plants.
One advantage of molecular approaches using partial or
complete kinase sequences to screen cDNA libraries is the ability to
work with tissue that would be difficult to address with
biochemical tactics.
Biermann et al. (1990) probed a maize root cap
cDNA library sequentially with three partially degenerate
oligonucleotides and isolated a putative kinase clone called 90.7.
Interestingly, in addition to containing considerable homology to
35
the bean and rice transcripts cloned by Lawton et al. (1989), the
maize sequence also contained the same DFD triplet that
characterized PVPK-1 and Gl1A.
Northern analysis and in situ
hybridization indicated similar levels of expression in coleoptiles,
root meristems, and the root elongation zone, with lower levels
expressed in leaf (Biermann et al., 1990).
Kinase-specific primers were used with the polymerase chain
reaction (kinase-specific PCR) by Lin et al. (1991) to clone five
members (PsPK1-5) of a family of putative protein kinases from
greening pea seedlings.
The members of the PsPK family showed
different rates and levels of expression during the deetiolation
process.
A similar PCR strategy was used to identify and clone a
family of six putative protein kinases from soybean (GmPK1-6;
Feng et al., 1992) and one sequence from Arabidopsis (Atpk7;
Hayashida et al., 1992). The pea and Arabidopsis sequences, and
four of the six soybean sequences all contained the DFD feature in
subdomain VII.
The other two soybean sequences contained the
canonical DFG motif; because the authors listed DFD as the expected
sequence for plant kinases, they found themselves trying to
explain this apparently anomalous result (Feng et al., 1992).
To
our knowledge, none of these sequences (PVPK-1, G11A, PsPK1-5,
GmPK1-6, or Atpk7) has been associated with an expressed protein
kinase activity, or even expressed protein, in spite of the fact that
they were all cloned from cDNA libraries and all could be shown to
hybridize to sequences in mRNA on northern blots. It seems
unlikely that, for example, sequences for (previously active)
36
kinases containing DFG in subdomain VII were converted to
inactive DFD-containing sequences through a cloning artifact,
because both PCR and library screening were used to obtain the
clones.
However, it is hard to imagine a role for mRNA which is
apparently either untranslated or codes for inactive protein
kinases. In any event, all other plant sequences with protein
kinase homology cloned to date, including several with catalytic
activity, possess subdomains VII with the DFG triplet.
Knowledge of conserved kinase sequences has been used to
clone putative protein kinase cDNAs which do encode protein i n
vivo.
For example, an abscisic acid-inducible transcript encoding a
putative protein kinase (PKABA1) has been isolated from wheat
embryos by screening a library derived from ABA-treated tissue
with a probe produced by kinase-specific PCR (Anderberg and
Walker-Simmons, 1992).
PKABA1 transcripts were dramatically
induced when plants were subjected to dehydration stress or to
low levels of ABA, as were levels of a protein recognized by an
antibody raised to a synthetic peptide based on the cDNA sequence
(M.K. Walker-Simmons, Washington State University, personal
communication).
The phosphorylation state of several proteins in
carrot somatic embryos is apparently affected by treatment with
abscisic acid (Koontz and Choi, 1993).
The first plant protein kinase to be cloned using peptide
sequence information from a purified protein kinase was soybean
CDPK (Harper et al., 1991). Sequence analysis showed this kinase
to be the prototype for a previously undescribed kinase which
37
contained an N-terminal catalytic domain contiguous with a C-
terminal calmodulin-like Ca2+ binding domain (see Roberts and
Harmon, 1992; Roberts 1993 for reviews).
This type of kinase is
apparently unique to plants and protists, and has yet to be
demonstrated in an animal species (Harper et al., 1993). A second,
partial clone of a CDPK from carrot has been obtained by PCR (Suen
and Choi, 1991).
In addition to CDPK, which can respond to Ca2+ without the
agency of calmodulin, a putative Ca2+-calmodulin dependent
protein kinase has recently been cloned from apple (Watillon et al.,
1993).
The sequence, CB1, was identified by screening an
expression library for the ability to bind 125I-labeled calmodulin
(Watillon et al., 1992).
CB1 shares significant sequence homology
to rat Ca2+/calmodulin protein kinase II in both the kinase catalytic
and calmodulin binding domains (Watillon et al., 1993).
Thus
plants may have at least two different routes by which changes in
Ca2+ levels may effect changes in phosphorylation.
Several plant kinases have been described which have
characteristics of putative receptor kinases.
In animals, receptor
kinases are a class of transmembrane proteins in which the
extracellular domain is associated with recognition of and binding
to extracellular signaling molecules, while the intracellular domain
is generally a tyrosine kinase. The plant kinase ZmPK1, identified
by kinase-specific PCR of corn cDNAs, was the first putative protein
kinase sequence with a transmembrane sequence to be found in
plants (Walker and Zhang, 1990). The catalytic domain resembles
38
raf-1 , a soluble serine/threonine kinase from human, and also
human epidermal growth factor receptor, which is a membranespanning tyrosine kinase.
The extracellular domain of ZmPK1 is
strikingly similar to the extracellular domain of S-locus
glycoproteins (SLG) found in Brassica species. In the Brassicaceae,
SLGs are thought to be involved in the expression of selfincompatibility in pistils (Ebert et al., 1989).
However, in maize
ZmPK1 mRNA is expressed primarily in shoots and roots of
seedlings, and to a reduced degree in silks.
Southern analysis
indicated the presence of a ZmPK gene family (Walker and Zhang,
1990).
The suggestion that a protein kinase might be involved with
the regulation of self incompatibility was extended when sequence
analysis of SLG clones from B. oleracea identified regions with
kinase catalytic domain homology in polypeptides with membranespanning regions (Stein et al., 1991).
The S-receptor kinase (SRK)
catalytic-like and SLG-like domains were each found to hybridize
to similar-sized transcripts from pistil and, to a lesser extent,
anther.
Unlike ZmPK1, the Brassica sequences were not detected in
leaf or shoot tissue (Stein et al., 1991).
Similar sequences have
been cloned from Arabidopsis and B. napus (Tobias et al., 1992;
Goring and Rothstein, 1992). The Al 0 allele of an SRK from a selfcompatible B. napus has recently been shown to encode a message
with a 1-bp deletion, causing a frameshift which is predicted to
lead to premature termination of translation (Goring et al., 1993).
39
This supports the concept that a functional SRK gene is required for
expression of self-incompatibility.
A fourth report of a plant protein kinase sequence with a
transmembrane region involves a protein with an extracellular
domain resembling the leucine-rich repeat motif found as part of
the extracytoplasmic domains of a number of mammalian
membrane proteins, rather than the S-protein homologous domains
of ZmPK1 or the SRK sequences (Chang et al., 1992). Only a few
transmembrane serine/threonine kinases have been reported from
animal species; the majority phosphorylate only tyrosine (Ullrich
and Schlessinger, 1990).
Complementation of yeast protein kinase mutants, which has
been a very fruitful approach to the identification of animal
protein kinase sequences, has been used to indicate the function of
a protein kinase sequence cloned from rye endosperm (Alderson et
al., 1991).
The cDNA sequence (cRKIN1), which had been isolated
as part of an unrelated experiment, was found to have sequence
similarity to SNF1, a serine/threonine kinase from yeast which is
involved in derepression of several glucose repressible genes.
Subsequent complementation experiments showed the sequence
was able to restore function to SNF1 deletion mutants of yeast. The
RKIN1 message was found in mRNA from developing endosperm
but not in shoots, and genomic Southern analysis suggested it was a
member of a small multigene family. Hybridizing sequences were
also found in Arabidopsis, barley, and wheat.
40
Much interest has centerd on the question of whether plants
contain protein kinase C. There have been reports of protein
kinases which crossreact with antibody to bovine brain PKC (Elliott
and Kokke, 1987), as well as reports of kinases which are activated
by Ca2+ and fatty acids (Lucantoni and Polya, 1987). Protein
kinase activity in zucchini microsomes is stimulated by platelet-
activating factor, a phospholipid (Scherer et al., 1988). However,
no plant kinases have yet been identified which meet the
biochemical definition for PKC. PKC activity is phosphatidylserineand (usually) calcium-dependent.
The presence of diacylglycerol
dramatically increases the affinity of PKC for Ca2+; limited digestion
with trypsin or calcium-dependent protease activates the enzyme
and renders it totally independent of Ca2+ (Kikkawa and Nishizuka,
1986). CDPK from Arabidopsis (Harper et al., 1993) and oat
(Schaller et al., 1992) has been analyzed for lipid dependence and
response to proteolysis. The Arabidopsis kinase is stimulated by
lysophosphatidylcholine and phosphatidylinositol;
phosphatidylcholine and phosphatidylserine had no effect.
Calcium
and lipid stimulated the enzyme synergistically (Harper et al.,
1993).
CDPK from oat is similarly activated by calcium and
The presence of lipid has no effect on the affinity of
the oat enzyme for calcium, and limited proteolysis converts the
phospholipid.
enzyme into a form that is lipid-independent but still calciumdependent (Schaller et al., 1992). Thus, CDPK from both
Arabidopsis and oat have biochemical properties which are
dissimilar from those of PKC.
41
Protein phosphatases.
Much less research has been devoted
to protein phosphatases than to protein kinases.
This belies their
importance, as the phosphorylation state of proteins reflects a
balance between kinase and phosphatase activity, and
dephosphorylation as well as phosphorylation affects enzyme
activity.
For example, sucrose phosphate synthase is inactivated
by dephosphorylation (Huber and Huber, 1991).
Four main classes
(or types) of protein phosphatases have been identified in
mammalian cells (for review see Cohen, 1989).
Phosphatase
classes are differentiated on the basis of the effects of inhibitors
and ions, and at least three of the classes have been found in
plants.
Type 1 phosphatases have been identified in pea and
carrot (MacKintosh et al., 1991), B. oleracea (Rundle and Nasrallah,
1992), and maize (Smith and Walker, 1991). The Brassica and
maize enzymes have been cloned (Rundle and Nasrallah, 1992;
Smith and Walker, 1991).
Phosphatase types 2A and 2C are
present in carrot, pea, and wheat; type 2A is also present in B.
napus (MacKintosh et al., 1991).
Phosphatase types 1 and 2A are very sensitive to the
inhibitor okadiac acid, and the effects of this and other inhibitors
have been exploited in efforts to link the activity of phosphatases
to various cellular processes.
For example, okakaic acid causes
mitotic arrest in tobacco (Zhang et al., 1992).
Application of
okadiac acid and the phosphatase inhibitors microcystin-LR and
microcystin-RR at specific points during the course of mitosis
affects metaphase transit times as well as the pattern of sister
42
chromatid separation in Tradescantia (Wolniak and Larsen, 1992).
A cDNA clone of an Arabidopsis type 1 phosphatase has been
reported to be able to rescue the Schizosaccharomyces pombe
mutant dis2-11, which fails to complete chromosome disjunction
under nonpermissive conditions (Nitschke et al., 1992).
Roles of phosphorylation/dephosphorylation. While
considerable molecular and biochemical phenomenology is being
collected, relatively few processes or enzymes are known to be
controlled by phosphorylation in plants.
Changes in protein kinase
activity or phosphorylation have recently been correlated with a
number of extracellular signals, including syringomycin (Bidwai
and Takemoto, 1987), blue light (Short and Briggs, 1992),
oligouronides (Farmer et al., 1989, 1991), cutin (Bajar et al., 1991),
fungal elicitor (Schwacke and Hager, 1992), the elicitor aaminobutyric acid (Raz and Fluhr, 1993). Examples of processes or
enzymes which may be modulated by phosphorylation or
dephosphorylation include self incompatibility (Goring et al., 1993),
initiation of mitosis (Li and Roux, 1992, Duerr et al., 1993),
isoprenoid biosynthesis (MacKintosh et al., 1992), cytoplasmic
streaming (Tominga et al., 1987; McCurdy and Harmon, 1992),
sucrose phosphate synthase activity (Huber and Huber, 1991),
MsERK1 activity (Duerr et al., 1993), and phosphoenolpyruvate
carboxylase activity (Bakrim et al., 1992).
In addition, plant growth regulators have been demonstrated
to affect protein kinases, and phosphorylation in general.
Transcription of one putative kinase sequence is stimulated by
43
ABA (Anderberg and Walker-Simmons, 1992), and
phosphorylation of several proteins is stimulated by ABA (Koontz
and Choi, 1993).
Treatment with ethylene stimulates
phosphorylation in tobacco leaves (Raz and Fluhr, 1993).
Phosphorylation in oat coleoptiles is affected by auxin (Veluthambi
and Poovaiah, 1986).
It appears that kinases in plants, as in animals, are
fundamentally involved with the regulation of cellular function.
However, it is also becoming clear that there are important
differences: there is no conclusive evidence to indicate that plants
contain either PKA or PKC, which act together as well as
independently to modulate many vital cellular processes in
animals.
In addition, while transmembrane kinases are present,
they are thus far exclusively serine/threonine type kinases in
plants.
On the other hand, plants contain CDPK, a class of kinase
which may not be present in animals.
In the case of phosphatases,
it appears that plants are similar to animals in that plants contain
most or all of the major types of protein phosphatases.
44
CONCLUSION
The rate at which progress is being made toward
understanding signal transduction in plants is increasing greatly.
However, it is not likely to match even early rates of progress on
animal signal transduction.
This is only partly because of the
relatively small number of researchers arrayed against the
problem or because of the difficulties of working with vacuolated
cells surrounded by cell walls. Progress in animal systems has
proceeded at a rate directly proportional to the availability of the
experimental tissue. Thus, some of the earliest work on animal
signal transduction utilized rabbit skeletal muscle (e.g. Krebs and
Fischer, 1956) and liver (e.g. Sutherland and Cori, 1956). Insights
gained from abundant tissues were then helpful in beginning to
sort out signaling in other, more precious tissues such as rod outer
segments from bovine retina and hair cells from bullfrog inner ear.
It would have been difficult to proceed in the opposite direction,
yet that is what appears to be necessary in plant signal
most experimental tissues are relatively difficult to
obtain in pure form. Comparisons with animal systems, which
transduction:
were initially expected to substitute for abundant tissues, have
certainly been helpful, but it is now clear that there are important
differences between plant and animal signal transduction (see
Figure 1), just as there are differences between plant and animal
physiology (see Table 1).
In spite of the current lack of a rabbit
45
skeletal muscle equivalent in the plant world, these are exciting
times to be involved with research on plant signal transduction.
If considered at the lowest level of resolution, it is clear that
plants contain, in some form, the major components of signal
transduction systems found in many organisms ranging from
bacteria to mammals.
For example, phosphorylation-
dephosphorylation is known to control the activities of a number of
plant enzymes, small- as well as heterotrimeric G-proteins are
present, and plant cells appear to possess the machinery for
generation of several different second messengers. Looked at more
closely, however, the differences become more striking than the
similarities:
presently, plants do not appear to contain a signaling
pathway based on cyclic AMP; nor do they appear to contain an
enzyme regulated precisely like PKC; plants do contain CDPK, an
enzyme that may not be utilized by animals.
It is much too early to
draw conclusions based on these observations, but the observations
do have important consequences. At this stage of the development
of our understanding of plant signal transduction, we have not yet
identified the variety in transduction systems that would lead to a
robust network of signaling pathways, finely tuned by crosstalk
between the pathways.
Such crosstalk is a crucial part of well-
understood animal models.
If crosstalk is part of the mechanism by
which a handful of plant growth regulators are able to have diverse
effects in different tissues, it is to be expected that one or more
fundamental signaling pathways are yet to be discovered.
outside
Transi-ribraile
.Seahr: :
Tyr
Present in animals; apparently absent or rare in plants
Present in animals; apparently present in plants
CDPK
1
Apparently absent in animals; present in plants
çnss
Status in plants unknown
Figure 1
47
ACKNOWLEDGEMENTS
This chapter appears under the same title and in somewhat
edited form by Steven D. Verhey and Terri L. Lomax in the Journal
of Plant Growth Regulation, 1993 (In Press).
I thank members of
our laboratory, in particular David L. Ray le and Andreas
Nebenfiihr, and Sally Assmann (Harvard University), Wendy Boss
(North Carolina State University), Alice Harmon (University of
Florida), and Jeff Harper (Scripps Research Institute), for critical
reading of the manuscript.
I also thank Alice Harmon (University
of Florida) and Michael Harrington (University of Hawaii) for
sharing unpublished data.
48
Zucchini Plasma Membrane Contains Multiple Calcium-
Requiring Protein Kinases
INTRODUCTION
Protein phosphorylation, mediated by protein kinases (E.C.
2.7.1.37), is one of a small number of readily reversible covalent
protein modifications available for intracellular control of
structure, activity, and targeting of enzymes and other proteins.
Protein kinases have been found in virtually every cellular
compartment and serve myriad functions, from post-translational
modification of proteins to involvement in signal transduction and
amplification (for review, see Hunter, 1991).
In plants,
polypeptides capable of acting as protein kinases have been
identified in the cytosol and in nuclei, mitochondria, and plastids,
as well as the plasma membrane (PM) (for review, see Ranjeva and
Boudet, 1987).
Activities of several plant enzymes are affected by
phosphorylation-dephosphorylation, including pyruvate
dehydrogenase (Rao and Randall, 1980); quinate: NAD
oxidoreductase (Refeno et al., 1982); sucrose phosphate synthetase
(Huber and Huber, 1990); and pyruvate, Pi dikinase (Ranjeva and
Boudet, 1987). The PM proton ATPase is a target of
49
phosphorylation, presumably via a PM protein kinase (Zocchi,
1985; Schaller and Sussman, 1988).
Although involvement of protein kinases in signal
transduction across the PM has not been established in plants,
changes in protein kinase activity or protein phosphorylation have
been reported to correlate with the application of extracellular
stimuli.
Examples of this include tomato tissue culture cell
responses to elicitor (Felix et al., 1991), Phytophthora infection of
soybean (Feller, 1989),
tomato proteinase inhibiting factor (Farmer
et al., 1989), citrus exocortis viriod infection of tomato (Vera and
Conejero, 1990), and syringomycin treatment (Bidwai and
Takemoto, 1987).
A predominant protein kinase activity of plant plasma
membranes is stimulated by calcium (Schaller and Sussman, 1988;
Ladror and Zielinski, 1989; Klimczak and Hind, 1990). Several
Ca2+-stimulated, PM-bound protein kinases have been identified in
plants.
Among these are two kinases from silver beet leaf (Kulcis
and Polya, 1988), one from pea epicotyl (Blowers et al., 1985), one
from barley leaf (Klimczak and Hind, 1990), and one from oat root
(Schaller et al., 1992). The oat root PM enzyme bears
immunological homology to soybean calcium-dependent protein
kinase (CDPK, Harmon et al., 1987). Soybean CDPK is a soluble
enzyme which contains a C-terminal calmodulin-like Ca2+-binding
domain contiguous with an N-terminal kinase catalytic domain
(Harper et al., 1991; for reviews see Roberts and Harmon, 1992;
Roberts, 1993).
50
In the course of our studies on Ca2+-dependent
phosphorylation in zucchini (Cucurbita pepo L., cv Dark Green) we
have identified several protein kinases which are present to
different degrees in total protein preparations from roots,
hypocotyls, hooks, cotyledons, and plumules of dark- and lightgrown seedlings.
In addition, we have identified at least four
species of plasma membrane-bound calcium-requiring protein
kinases (CRPK) with apparent molecular weights between 58-68
kD. As is the case for the soluble 55 kD soybean CDPK, these
zucchini hypocotyl PM protein kinases all require micromolar
concentrations of Ca2+ for activity and they phosphorylate serine
and, to a lesser extent, threonine. Like the oat PM enzyme, three of
the zucchini PM protein kinases are recognized by monoclonal
antibodies to the soluble soybean CDPK. The nature of the
relationship of the zucchini kinases to one another and to the
soybean and oat enzymes remains to be elucidated.
51
MATERIALS AND METHODS
Chemicals and radiochemicals. Radio labeled ATP ([y-32P]ATP;
specific activity, 3000 Ci/mmol) was obtained from New England
Nuclear, Boston, MA. The divalent cation-chelating resin Chelex,
and reagents used for SDS-PAGE (acrylamide; bisacrylamide; SDS; N,
N, N', N'-tetra-methylethylenediamine; ammonium persulfate;
molecular weight standards) were obtained from Bio-Rad
(Richmond, CA).
Biotrace NT nitrocellulose was acquired from
Gelman, Inc. (Ann Arbor, MI).
Anti-CDPK monoclonal antibody
(Putnam-Evans et al., 1990) was the generous gift of Dr. Alice
Harmon at the University of Florida, Gainesville, FL.
Goat anti-
mouse alkaline phosphatase-conjugated secondary antibody was
from Promega (Madison, WI). Unless otherwise noted, all other
chemicals were obtained from Sigma (St. Louis, MO).
Plant material. Seeds of zucchini (Cucurbita pepo L., cv Dark
Green), and oat (Avena sativa L. cv Aberdeen) were obtained from
Lilly-Miller.
Seeds of soybean (Glycine max L.) were obtained
from First Alternative (Corvallis, OR).
After surface sterilization for
10 min with 20% (v/v) bleach containing a few drops/L of Tween-
20, seeds were rinsed with tap water and sown directly on ca. 3 cm
of moist vermiculite (Avena) or on moist vermiculite covered with
several layers of unbleached absorbant paper (Kimtowels;
Kimberly-Clark, Rosewell, GA)(Cucurbita and Glycine). After 5 days
52
at 27°C in either dark ("dark-grown") or fluorescent-lighted ("lightgrown," 35 gm m-2 s-1) growth chambers, seedlings were harvested
and processed as described below.
Preparation of membrane vesicles.
Microsomal membrane
vesicles were prepared by homogenizing dark-grown zucchini
hypocotyls, using a Brinkman (Westbury, NY) Polytron at setting 8
for 20-30 s, with an equal volume of ice-cold Buffer 1 (250 mM
sucrose, 10 mM Tris-Cl pH 7.5, 1 mM EDTA, 1 mM DTT, 0.1 mM
Mg SO4) plus protease inhibitors (0.2 mM PMSF, 1 gg/mL each
leupeptin and pepstatin).
The brei was filtered through 4 layers of
cheese cloth, and the retentate rehomogenized with an additional
volume of Buffer 1. The rehomogenized brei was filtered, and the
filtrates pooled.
After a 15 min centrifugation at 1000 x g
(Beckman GPR, GH3.7 rotor) to remove cell walls and undisrupted
cells, microsomes were pelleted by a 30 minute centrifugation at
150,000 x g (Beckman L5-65, Ti50.2 rotor). Microsomal pellets
were resuspended in Buffer 2 (5 mM KPi, pH 7.8, 250 mM sucrose,
4 mM KC1) flash-frozen in liquid N2, and stored at -70°C until use.
Plasma membrane vesicles were prepared by aqueous twophase (dextran/PEG) separation of microsomal membrane
preparations from zucchini seedlings as described by Hicks et al.
(1989).
Protease inhibitors were replenished after each
centrifugation.
After the final centrifugation, PM vesicles were
solubilized as described below or resuspended in a small volume of
Buffer 1 plus protease inhibitors, flash-frozen in liquid N2, and
stored at -70°C. Marker enzyme assays indicate that zucchini
53
hypocotyl PM prepared in this manner are >95% pure (Hicks et al.,
Protein concentrations were estimated using either the
method of Schaffner and Weissmann (1973), which eliminates
1989).
interference by detergents, or a kit from Bio-Rad based on the
Bradford assay.
BSA was used as the standard for protein assays.
Preparation of liquid nitrogen-ground tissues. For some
experiments, in order to minimize exposure to protease activity,
organs were harvested directly into liquid N2 and stored at -70°C
until they could be used. Each sample was subsequently ground
with liquid N2 in a mortar and pestle. The frozen powder was
dropped, slowly and with stirring, into a small beaker containing
SDS-PAGE loading buffer and 100 mM DTT. The small beaker sat
inside a boiling water bath, which maintained the contents of the
small beaker at a temperature of at least 65°C. Several aliquots of
each organ extract were frozen at -70°C to await analysis on SDS-
PAGE. A fresh aliquot was prepared just prior to each analysis by
thawing and centrifugation in a microfuge for 5 min. The
supernatant was reheated to 95°C for two min before loading the
The amount of material needed for similar protein loads was
determined empirically on minigels.
gel.
Electrophoresis.
Protein electrophoresis was carried out
according to instructions furnished with the Bio-Rad Protean II
electrophoresis unit.
Prior to SDS-PAGE, samples were heated at
95°C for 2 min in Laemmli sample buffer containing 100 mM DTT,
and electrophoresed on 10% (w/v) polyacrylamide gels (0.75 mm x
15 cm) unless otherwise noted.
54
In situ phosphorylation on nitrocellulose. In situ
phosphorylation on nitrocellulose blots was carried out as
described by Celenza and Carlson (1986), with the following
modifications.
Subsequent to SDS-PAGE, proteins were
electroblotted onto nitrocellulose in an LKB semidry blotting
apparatus.
The blotting buffer contained 12.5 mM Tris/35 mM
glycine, pH 8.3. Nonfat dry milk (5% w/v) was used as the blocking
agent.
Renaturation buffer contained 50 mM bis-Tris-
propane/Hepes pH 6.8, 100 mM NaC1, 2 mM DTT, 1 mM EGTA, 0.1%
(v/v) NP-40, 5 mM MgC12, 1.055 mM CaC12, and 0.25% (w/v) nonfat
dry milk; renaturation was carried out at 4°C for at least 9 h.
Phosphorylation buffer contained 50 mM bis-Tris-propane/Hepes
pH 6.8, 1 mM EGTA, 5 mM MgC12, 1.055 mM CaC12, and 0.25% (w/v)
nonfat dry milk.
The labeling reaction was started by the addition
of 20 tCi of [y- 32P]ATP (specific activity 3000 Ci/mmol)/10 mL
phosphorylation buffer (0.7 nM ATP) and continued for 60 min. No
unlabeled (carrier) ATP was used.
After incubation the filter was
washed with several changes of 10% (w/v) TCA-1M H3PO4, then
with several changes of double-distilled water.
visualized by autoradiography;
Bands were
exposures were usually for 16 h
without an intensifier screen.
Immunostaining.
Following autoradiography, blots were
incubated with anti-CDPK monoclonal antibody, then with goat
anti-mouse-alkaline phosphatase secondary antibody as described
in Sambrook et al. (1989).
55
Protein kinase solubilization.
Pelleted plasma membranes
were resuspended at 2 mg protein/mL in Buffer 3 (10 mM Tris pH
8.5, 50 mM NaC1, 1 mM DTT, 10% (v/v) glycerol) containing 500
mM NaC1/5 mM EDTA or 0-3% (w/v) detergent and held on ice for
30 min.
Detergents were 3-[(3-cholamidopropyl)dimethyl-
ammonio]-1-propanesulfonate (CHAPS, Research Organics,
Cleveland, OH) or Triton X-100. After centrifugation at 140 kxg
(Beckman TL-100, rotor TLA 100.3) for 5 min, the supernatants
The pellets were taken up in the appropriate
solubilization buffer, and the volumes adjusted to match those of
were recovered.
the corresponding supernatants.
Aliquots were then analyzed by
SDS-PAGE followed by in situ labeling on nitrocellulose.
Phosphoamino acid analysis. Following SDS-PAGE, PM
proteins were blotted to PVDF and labeled in situ as described for
nitrocellulose, except nonfat dry milk was omitted.
After
localization by autoradiography, bands were excised and
hydrolyzed by treatment for two h with 6 N HC1 at 105°C in
microfuge tubes.
HC1 was evaporated under a stream of air, and
the hydrolyzate was taken up in a small volume of
chromatography buffer (15 mM KPi, pH 3.8) filtered, and analyzed
by isocratic HPLC (Swarup et al. 1981). HPLC utilized a 4.6 x 250
mm Beckman SAX anion exchange column and internal
phosphoamino acid standards.
Following addition of scintillation
fluid (Dupont Formula 989), fractions were analyzed for
radioactivity by scintillation counting.
56
RESULTS
We surveyed the activity of protein kinases from several
different light-grown and etiolated organs of zucchini, as well as
from dark-grown roots of oat and hypocotyls of soybean.
Whole
plant material was ground in liquid N2 and extracted with hot SDSPAGE loading buffer to minimize protease action.
Total proteins so
prepared were separated on one dimensional SDS-PAGE, blotted to
nitrocellulose, and incubated with [i- 32P]ATP following blocking of
nonspecific binding sites with nonfat dry milk (in situ
As can be seen in Figure 2, lanes loaded with
protein from etiolated zucchini hooks contained the greatest
phosphorylation).
number of kinases active under these conditions, a total of eight
bands between approximately 55 and 68 kD (Fig. IA, lane 3). Lanes
containing protein from light- and dark-grown zucchini cotyledons
showed the least activity, with only 2-3 bands visible (Fig. 1A,
lanes 4 & 7, respectively). Plumules contained a kinase profile
most similar to etiolated hooks (Fig. 1, c.f. lanes 3 & 8). In general,
more activity was present in lanes loaded with protein from
etiolated organs than from their light-grown counterparts.
Total protein from etiolated zucchini hypocotyls yielded
radiolabeled bands at six molecular weights, between
approximately 58 and 68 kD (Fig. 2A, lane 2). Lanes loaded with
protein from soybean hypocotyls contained five labeled bands,
with apparent molecular weights ranging from ca. 54 to 67 kD (Fig.
57
2B, lane 1).
Roots from etiolated plants displayed a kinase profile
similar to that of hypocotyls (Fig. 2A, lane 1). Lanes loaded with
protein from oat roots contained only two labeled bands, at ca. 57
and 61 kD (Fig. 2B, lane 2). None of the soybean or oat bands had
molecular weights identical to any zucchini band (Fig. 2, cf. Panels
A & B).
To assess which kinases were PM-associated, we examined
kinase activity in zucchini hypocotyls in greater depth.
When
plasma membrane proteins isolated by aqueous two phase
(dextran/PEG) partitioning were labeled as described above, four
bands, at 58.6, 60.5, 65.7, and 67.9 kD, were most heavily labeled
(Fig. 2C).
These four bands matched the most heavily labeled
bands seen in experiments involving total protein from etiolated
hypocotyls (c.f. Fig. 2A, lane 2). No labeled bands were seen on
blots incubated with [a-32P] ATP (data not shown).
Calcium-dependence of the PM kinases was examined by in
situ phosphorylation on nitrocellulose, using Chelex-treated bovine
serum albumin as a calcium-free blocking agent and Chelex-treated
solutions during blot preparation and labeling.
Renaturation was in
the presence of 87 I.LM free Ca2+. As shown in Figure 3, all four
kinases were active in 11 gM free Ca2+, while virtually no activity
was present in the same region following incubation of blots with
[7-32P]ATP in the presence of 1 mM EGTA (Fig. 3A & B). Increasing
free calcium to 87 1.LM had no further effect on labeling (Fig. 3C).
Similar results were obtained when blots were renatured in the
absence of Ca2+ (data not shown).
58
The amino acid specificity of phosphorylation reactions is
useful in classification of protein kinases (Krebs, 1986). To
determine amino acid specificity, the four zucchini hypocotyl PM
kinases were labeled with [y-3211ATP after blotting to PVDF. The
individual bands were excised, hydrolyzed with HC1, and labeled
amino acids were identified by HPLC and scintillation counting. As
can be seen in Figure 4, all four kinases autophosphorylated on
both serine and, to a lesser extent, threonine residues.
In order to assess the antigenic relatedness of the zucchini
PM kinases to soybean CDPK, Western blot analysis was carried out
following in situ phosphorylation. As shown in Figure 5, anti-CDPK
monoclonal antibodies (Putnam-Evans et al., 1990) crossreacted
with three major bands, having molecular weights identical to
those labeled by [y- 32P]ATP. An additional, somewhat larger band,
not labeled by [y- 32P]ATP in this experiment, was also recognized
by the monoclonal antibodies.
Finally, as shown in Figure 6, we examined the extent to
which the PM-associated zucchini enzymes were associated with
the membrane.
No detectable kinase activity was removed from
PM by washing the membranes with 0.5 M NaC1-5 mM EDTA (Fig.
6A & B, lanes 1). Relatively high detergent concentrations were
required to remove the proteins from the PM:
only at CHAPS (Fig.
6A) or Triton X-100 (Fig. 6B) concentrations greater than the
critical micelle concentration (0.49% w/v and 0.02% w/v,
respectively) of each detergent was significant kinase activity
released into the supernatant. While the largest kinases were
59
solubilized at lower detergent concentrations, all four kinases were
soluble in concentrations of >0.3% (w/v) Triton X-100 under these
conditions.
Further experiments demonstrated that all four kinases
could also be solubilized by 1% (w/v) CHAPS; similar results were
obtained with 1% (w/v) octyl glucoside (data not shown).
Following solubilization, removal of CHAPS during gel filtration led
to aggregation of the protein kinases (data not shown).
60
Figure 2.
Size and diversity of protein kinases in zucchini, soybean,
and oat. Proteins were separated by SDS-PAGE and
electroblotted to nitrocellulose.
After blocking and
renaturation, blots were incubated with 2 I.LCi/mL [y-3213]A TP
for 1 hour, then washed with 10% TCA-1M H3PO4 and
subjected to autoradiography.
Results from three
autoradiograms are shown. Panel A:
zucchini organs which
were frozen and ground in liquid N2 before loading. Lane 1,
dark-grown roots; lane 2, etiolated hypocotyls; lane 3,
etiolated hooks; lane 4, etiolated cotyledons; lane 5, lightgrown hypocotyls; lane 6, apical 5 mm of light-grown
hypocotyls; lane 7, light-grown cotyledons; lane 8, lightgrown plumule.
Panel B:
Lanes contain similar amounts of protein.
similar amounts of protein from etiolated soybean
hypocotyls (lane 1) and dark-grown oat roots (lane 2)
prepared as described for Panel A. Panel C: 40 gg of zucchini
hypocotyl PM protein prepared by aqueous two-phase
(dextran/PEG) partitioning.
Autoradiography was for 16
hours without an intensifying screen.
61
Figure 2
A
1
2345678
1
241-1
76
48
-h
28
19-
2
62
A
Figure 3.
B
C
Calcium requirement of the plasma membrane kinases.
Blots were prepared as described, blocked with Chelextreated BSA, and incubated with Chelex-treated buffers.
Both
blots were incubated in the same renaturation buffer
containing 87 1.1.M Ca2+, and rinsed with the appropriate
phosphorylation buffer before labeling.
Autoradiograms (16
hours, no intensifying screen) are shown of blots labeled in
the presence of (A) 11 I.LM Ca2+, (B) 1 mM EGTA, or (C) 87 I.LM
Ca2+.
Calcium concentration was controlled using an EGTA-
Ca2+-Mg2+ buffer, and free Ca2+ was calculated using a
computer program (Perrin and Sayce, 1967) with constants
from Martell and Smith (1974).
Bars indicate major
polypeptides which require calcium for activity.
63
Figure 4. Phosphoamino acid analysis.
Labeled bands were excised
from PVDF blots and hydrolyzed in microfuge tubes with 6 N
HC1.
After lyophilization, the residue was taken up in
chromatography buffer, spiked with phosphoamino acid
standards, and chromatographed on HPLC anion exchange.
Arrows indicate elution fractions of internal standards (P-S,
phosphoserine; P-T, phosphothreonine; P-Y,
phosphotyrosine).
Panels A-D represent results from largest
through smallest PM kinases, respectively.
64
Figure 4
1500
A
1000
500
P-T
0
=IA,
1
1000
B
500
C
500 -
\
I
riFiVta
0
1:13133-wrm
D
500
0
15
25
Fraction
35
65
Figure 5. Western blot analysis:
antigenic similarity of the
zucchini PM kinases to soybean CDPK.
After labeling and
autoradiography as described for Figure 1, the blot was
incubated with a mixture of three anti-CDPK monoclonal
antibodies.
Immunodecorated proteins were located with
alkaline phosphatase-conjugated-goat anti-mouse antibody.
Panel A, Coomassie-stained gel. Panel B, autoradiogram (16
hour exposure, no intensifying screen).
Panel C, Western
analysis of blot used to create Panel B. Bars indicate the
major phosphorylated and immunoreactive bands.
66
ABC
20011697-
66-
mg
42-
IM
67
Figure 6.
Plasma membrane kinase solubilization.
Autoradiograms
from blots of gels loaded as follows. Lanes 1-7 are
supernatants, and lanes 8-14 are the corresponding pellets.
Panel A, CHAPS; panel B, Triton X-100. Treatments: lanes 1
and 8, 500 mM NaC1/5 mM EDTA; lanes 2 and 9, 0.01%
detergent;
lanes 3 and 10, 0.03% detergent; lanes 4 and 11,
0.1% detergent;
lanes 5 and 12, 0.3% detergent; lanes 6 and
14, 1% detergent; lanes 7 and 13, 3% detergent. Exposures
were for 16 hours without an intensifying screen.
68
Figure 6
A
123456789
1011121314
WO 111111/41P
B
12 34 5 6 7 8 9 1011121314
69
DISCUSSION
In situ SDS-PAGE phosphorylation, either in gels or on
nitrocellulose blots, has been used in the analysis of purified
protein kinases from animals (Geahlen et al., 1986), yeast (Celenza
and Carlson 1986), and plants (Klimczak and Hind, 1990; Harmon et
Blowers et al., 1985). We employed this approach to
al., 1987;
identify and characterize protein kinase activity in zucchini with
respect to organ distribution, subcellular localization, molecular
weight, calcium requirement, amino acid specificity, and
immunological similarity to other kinases.
We examined total protein extracted from dark-grown roots,
hypocotyls, hooks, and cotyledons, as well as protein from lightgrown hypocotyls, opened hooks, cotyledons, and plumules (Fig.
2A).
We also analyzed dark-grown soybean hypocotyls and Avena
roots (Fig. 2B).
kinases.
In zucchini, hooks contained the largest diversity of
Eight polypeptides exhibiting protein kinase activity, with
molecular weights ranging from approximately 52 to 67 kD, were
present in hook extracts. Interestingly, although the hooks had
opened in light grown seedlings, the kinase pattern from the apical
5 mm of light-grown hypocotyls more closely resembled that of
etiolated hooks than that of the lower part of light-grown
hypocotyls.
Cotyledons, both light- and dark-grown, contained the
fewest protein kinases active under these conditions.
Intermediate
numbers and activities of kinases were present in the other
70
zucchini organs.
Although SDS-PAGE should separate the kinases
from potential inhibitors such as phenolics which may be present
in higher concentrations in light-grown material or in cotyledons,
this explanation for the difference in activity between light-grown
and etiolated tissues cannot be completely ruled out.
Total protein preparations from etiolated zucchini hypocotyls
yielded six radiolabeled bands with molecular weights between 58
and 78 kD. In comparison to zucchini, etiolated soybean hypocotyls
were found to contain approximately five labeled bands, with
molecular weights between ca. 54 and 67 kD.
The 54 kD band from
soybean hypocotyls is similar in size to a 55 kD CDPK which has
been purified from soybean suspension culture cells (PutnamEvans et al., 1990). Oat roots contained only two kinases active in
our in situ phosphorylation assay, at approximately 57 and 61 kD.
Schaller and co-workers (1992) described three oat root
polypeptides, with molecular weights of 57, 61, and 79 kD, which
crossreact with a mixture of monoclonal antibodies raised against
soybean CDPK. We did not detect phosphorylation products in the
79 kD size range identified by immunostaining of oat root proteins
(Schaller et al., 1992), Schaller et al. also observe only very limited
in vitro phosphorylation of the 79 kD immunoreactive polypeptide
of oat (E. Schaller, University of Wisconsin, personal
communication).
A potential trivial explanation for the presence of multiple
kinase bands is post-homogenization proteolytic activity, which
might convert one or more higher molecular weight kinases into a
71
series of smaller ones.
Evidence suggesting this can occur in plants
has been presented (Schaller et al., 1992).
However, the same
zucchini kinase bands were seen in total protein prepared by
freezing and grinding harvested hypocotyls directly in liquid N2,
then dropping the frozen powder into hot SDS loading buffer before
electrophoresis, blotting, and labeling (Fig. 2, Panel A). Heating at
high temperatures with SDS inactivates plant PM proteases (North,
1989).
Thus, the presence of kinase activity at different apparent
molecular weights may be indicative of separate kinases with
similar characteristics.
We cannot, however, rule out the
possibility that any of the kinase bands are derived in vivo from a
larger kinase or kinases via endogenous protease activity.
Examination of zucchini PM proteins via in situ
phosphorylation on nitrocellulose blots led to the identification of
at least four major PM-bound protein kinases (Fig. 2C). The
molecular weights of these polypeptides are identical to those of
the four most heavily labeled bands found among total protein
from hypocotyl (c.f. Fig. 2A, lane 2). Thus, the majority of protein
kinases active under these conditions are evidently PM-localized.
Using in situ phosphorylation on nitrocellulose, we
determined that Ca2+ is not required for effective renaturation of
the protein kinases.
The zucchini PM protein kinases require Cat 1-
for autophosphorylation, however, at about the same concentration
as does soybean CDPK. CDPK displayed 100-fold activation in the
presence of 10 gm free Ca2+, (Harmon et al., 1987; Putnam-Evans
et al., 1990). The zucchini PM kinases were virtually inactive in
72
the presence of 1 mM EGTA, but were active in the presence of 11
ilM free Ca2+ (Fig. 3).
Phosphoamino acid analysis suggests another similarity
between CDPK and the zucchini kinases.
The zucchini kinases were
phosphorylated on serine, and, to a lesser extent, threonine
residues (Fig. 4). Soybean CDPK autophosphorylates primarily
serine, but also threonine (Putnam-Evans et al., 1990).
Recent cloning of a gene encoding a kinase closely related to
CDPK has demonstrated that CDPK is the prototype for a new family
of calcium-regulated protein kinases in which kinase activity is
modulated by direct binding of calcium. Sequence analysis has
shown that CDPK contains a C-terminal calmodulin-like Ca2+-
binding domain contiguous with an N-terminal kinase catalytic
domain.
Southern blot analysis of genomic DNA from soybean and
Arabidopsis thaliana reveals multiple restriction fragments
hybridizing with a PCR-generated CDPK probe (Harper et al., 1991),
which is consistent with the possibility that multiple subspecies
exist. Because of similarities in terms of Ca2+-dependence, amino
acid specificity, and molecular weight between the zucchini PM
kinases and soybean soluble CDPK, we assessed the immunological
relatedness of kinases from the two organisms.
Western blot analysis of zucchini PM proteins following in
situ labeling showed that three of the major labeled polypeptides
were recognized by a mixture of three monoclonal antibodies
raised against CDPK (Fig. 5, bars; Putnam-Evans et al., 1990).
Presumably the smallest labeled band is not immunostained in this
73
experiment because it is more widely diverged than the other
three, because it is present in smaller quantity, or for some other
reason. A ca. 76 kD band, which was not radiolabeled by [y32P]ATP under these conditions, was also recognized by the
monoclonal antibodies (Fig. 5).
Interestingly, in oat root, three
bands are recognized by anti-CDPK monoclonal antibody (Schaller
et al., 1992), but little protein kinase activity is associated with the
largest band, at 79 kD (E. Schaller, personal communication).
It
should be noted that the anti-CDPK monoclonal antibodies
recognize epitopes in the kinase catalytic domain, remote from the
calmodulin-like Ca2+-binding domain (A. Harmon, personal
communication).
Thus, binding of the antibodies to a protein
kinase does not necessarily demonstrate that the kinase is a
member of the CDPK family of kinases; such a demonstration can
most reliably come from sequence analysis (Harper et al., 1993).
To avoid confusion, we will refer to the zucchini kinases as
calcium-requiring protein kinases (CRPK).
In contrast to the soluble soybean CDPK, the zucchini PM
CRPK are evidently integral membrane proteins or are otherwise
tightly associated with the PM.
They were not removed from PM
by 0.5 M NaC1/5 mM EDTA washes, and were only solubilized by
?_1% (w/v) CHAPS or _0.3% (w/v) Triton X-100. The larger of the
major kinase bands appeared to be somewhat more soluble or
more active after solubilization than the smaller kinases under
these conditions (Fig. 6). When detergent was removed by gel
filtration, the kinases aggregated (data not shown).
These results
74
suggest all of the 58-68 kD polypeptides with kinase activity are
In this sense,
they resemble the 79 kD PM-associated calcium- activated protein
membrane-bound and contain hydrophobic regions.
kinase from Avena
(Schaller et al., 1992).
Peptide mapping or individual sequence analysis will be
necessary to determine whether any of the PM bands represent
distinct polypeptides, or one or more sets of subspecies of one or
more types of protein kinase. Some types of protein kinase,
including protein kinase C (PKC, Nishizuka, 1988), cAMP-dependent
protein kinase (PKA, Dostmann et al., 1990), Ca24-/calmodulin-
dependent protein kinase II (Colbran and Soder ling, 1990), cGMP-
dependent protein kinase (Ruth et al., 1991), and raf (Sithanandam
et al., 1990), exist as related subspecies. The PKC family consists of
at least seven subspecies which display variable tissue distribution
and subtly different biochemical regulation (Nishizuka, 1988).
Different isozymes of some protein kinases, including those of PKC
(Nishizuka, 1988; Otte and Moon, 1992) and PKA (Ginty et al.,
1992), have been suggested to have particular functions.
In other
cases, the roles of different forms have not been elucidated.
Heterologous molecular weights of subspecies or isozymes arise by
mechanisms which include expression of separate genes on
different chromosomes (Nishizuka, 1988), and differential mRNA
splicing (Francis et al., 1988-89).
As noted above, although several lines of evidence suggest
that the zucchini PM kinases might be membrane-bound members
of the CDPK family described by Harper et al. (1991), such a
75
conclusion is premature.
Purification and sequence analysis of the
zucchini hypocotyl PM CRPK, as well as further characterization of
the kinases from other organs, should reveal more information
concerning their relationship both to one another and to other
classes of kinases.
It should also facilitate the generation of
reagents useful in establishing the regulation and function of these
enzymes.
76
ACKNOWLEDGEMENTS
This chapter appears, in slightly edited form, under the title
"Protein Kinases In Zucchini: Characterization of Calcium-requiring
Plasma Membrane Kinases" by Steven D. Verhey, J. Christopher
Gaiser, and Terri Lomax (Plant Physiology, 1993, In Press), as
Technical Paper Number 10,238 from the Oregon State University
Agricultural Experiment Station.
Dr. Gaiser initially adapted the
nitrocellulose labeling protocol for use in our laboratory.
I thank
Alice Harmon for generously sharing anti-CDPK monoclonal
antibodies and the computer program used to calculate free Ca2+
concentrations, and for helpful discussions.
77
Zucchini Plasma Membrane Contains at Least Three Different
Species of Protein Kinase
INTRODUCTION
As described in the previous chapter, zucchini plasma
membrane (PM) vesicles contain multiple calcium requiring protein
kinases.
When PM proteins are incubated with [y-32P ] ATP
following renaturation on nitrocellulose blots of SDS-PAGE gels, at
least four major bands become labeled.
The radiolabeled bands
apparently represent individual proteins, not proteolytic products
of some larger kinase, and have molecular weights in the 58-68 kD
range. All require calcium for phosphorylation activity, and at
least three are recognized by a monoclonal antibody raised against
a soluble soybean kinase known as calcium dependent protein
kinase (CDPK) (Verhey et al., 1993). Soybean CDPK is the
prototype for a unique class of kinases which have calcium-binding
calmodulin-like domains connected to kinase catalytic domains
(Harper et al., 1991).
In contrast to the sutuation in zucchini PM, a relatively
limited number of protein kinases have been reported to be
associated with the PM of other plants. Oat root PM contains a
single protein, with a molecular weight of 79 kD, that crossreacts
78
with the anti-CDPK monoclonal antibody but which phosphorylates
poorly; a smaller band is also present but is assumed to be a
proteolytic product of the larger protein (Schaller et al., 1992; Eric
Schaller, University of Wisconsin, personal communication).
At
least two kinases are associated with silver beet PM (Kulcis and
Polya, 1988) and one with barley leaf PM (Klimczak and Hind,
1990).
One reason more kinases have been identified in zucchini
than in other organisms may be that other laboratories have used
less specific approaches to identifying protein kinases. The silver
beet kinases, for example, are inferred from peaks containing
kinase activity which eluted from a gel filtration column (Kulcis
and Polya, 1988). Klimczak and Hind (1990) used in gel
phosphorylation, a technique very similar to in situ
phosphorylation on blots, on partially purified kinase preparations,
but did not scrutinize PM or microsomal membranes directly.
As discussed in the previous chapter, there are similarities
between several protein kinases found in the zucchini plasma
membrane and CDPK.
Clarification of the number and type of
protein kinases associated with the zucchini plasma membrane is
the focus of this chapter. There is molecular evidence from work
on other plants that there are multiple genes for CDPK, and several
important animal protein kinases, including PKC, PKA,
Ca2+/calmodulin-dependent protein kinase II, cGMP-dependent
protein kinase, and raf, are expressed as multiple isozymes or
subspecies.
In animals this multiplicity is considered to be crucial
for modulation of the diverse actions of signaling molecules.
79
Signaling molecules of plants also have pleiotropic actions, and the
biochemical basis for this pleiotropy is unknown.
It thus seems
possible that in plants, as in animals, multiple variants of
components of signal transduction systems are involved in the
complex responses to seemingly simple combinations of
extracellular signals.
With this possibility in mind, we examined
the degree to which the CRPK might be related as subspecies of a
kinase family.
The determination of the nature of the relationship between
the zucchini CRPK and the CDPK would also be interesting, and
would yield insights into the possible roles of both types of kinase.
The comparison of CDPK with other protein kinases is not eased by
the availability of monoclonal antibodies, because all four
monoclonal antibodies that have been raised against soybean CDPK
recognize a single epitope near or within the nucleotide binding
site (Alice Harmon, personal communication). Thus, immunological
crossreactivity reflects only the degree of relatedness of the
respective nucleotide binding regions.
Molecular analysis of cDNA
and/or genomic clones is presently the only definitive approach to
this question (Harper et al., 1993). Molecular comparison requires
that the enzymes of interest be isolated, sequenced, and the
corresponding cDNA cloned.
This has been accomplished for only
one plant kinase, soybean CDPK (Putnam-Evans et al., 1990),
although a number of plant kinase sequences have now been
cloned as the result of the application of alternate strategies such
as kinase-specific PCR and cloning by heterologous screening.
80
Nonmolecular approaches, such as peptide mapping, are
theoretically possible but have not been applied to comparisons of
plant protein kinases.
The use of molecular strategies such as PCR or heterologous
cloning was considered for this study.
However, because our long-
term interest was in the potential for kinase involvement in signal
transduction at the PM, the decision was made to begin with the
PM-localized protein kinases described in the previous chapter,
and to use the biochemical methods described here instead. These
methods are more specific in terms of permitting the study of
kinases associated with a particular subcellular fraction; such
specificity would be very difficult or impossible with molecular
techniques.
This chapter describes the partial purification of the
PM CRPK and results of a peptide mapping comparison of the
CRPK's to one another.
81
MATERIALS AND METHODS
Chemicals and radiochemicals.
Acrylamide was purchased as
a 40% solution from BDH (Poole, UK). Bio-Rad (Richmond, CA) was
the source for other reagents used in SDS-PAGE.
Hydroxylapatite
was manufactured by Bio-Rad and was the gift of Dr. Reg
McParland, OSU Center for Gene Research and Biotechnology.
Nitrocellulose (Biotrace NT) was obtained from Gelman Inc. (Ann
Arbor, MI), and [y- 32P]ATP was purchased from New England
Nuclear (Boston, MA).
Sephadex G-100, Q-Sepharose, proteases,
and chemicals not listed above were obtained from Sigma (St.
Louis, MO).
Preparation of membrane vesicles. and kinase solubilization.
Plasma membrane vesicles were prepared from 5-6 day old darkgrown zucchini hypocotyls as described by Hicks et al. (1989a) and
in the previous chapter. Seeds of zucchini (Cucurbita pepo L., cv
Dark Green) were obtained from Lilly-Miller.
After surface
sterilization for 10 min in 20% (v/v) bleach containing a few
drops/L of Tween-20 and rinsing with tap water, seeds were sown
on 3 cm of moist vermiculite covered with several layers of
unbleached paper towels (Kimberly-Clark, Rosewell, GA). PM
protein was solubilized by resuspension of PM vesicles in 0.3
mL/mg protein of cold solubilization buffer (10 mM Tris pH 8.5, 50
mM NaC1, 10% glycerol, 1.5% CHAPS, 100 I.LM DTT, and 100 gg/mL
each of leupeptin and pepstatin).
After incubation on ice for 30
82
min, the mixture was centrifuged for 30 min at 105,000 x g and 3°
C (Beckman L5-65, Ti50.2 rotor), and the pellet resuspended in 0.1
mL/mg original protein of solubilization buffer.
After another 30
min incubation on ice, the mixture was centrifuged and the
supernatants pooled.
Other techniques. SDS-PAGE was performed according to
instructions furnished with the Bio-Rad Protean II electrophoresis
unit. Before being loaded on gels, samples were heated to 95° C for
2 min in Laemmli sample buffer containing 100 mM DTT.
Electrophoresis was on 150 x 0.75 mm 10% (w/v) acrylamide gels
unless otherwise noted. In situ phosphorylation on nitrocellulose
was performed as described (previous chapter; Verhey et al. 1993).
Protein kinase assay. Protein kinase activity was assayed
using casein (1 gg/mL) as substrate in a total reaction volume of
100 gL containing 50 mM BTP-HEPES pH 6.8, 10 mM MgC12, 1 mM
CaC12, 2 mM MnC12. The phosphorylation reaction was started by
the randomized addition of radiolabeled ATP (specific activity 5001000 cpm/pmol) to a working concentration of 100 iLM. After the
appropriate period of incubation at 22° C, 50 pi. aliquots of each
reaction mixture was pipeted onto a 2 x 3 cm piece of Whatman
3MM paper, and the paper placed in 10% TCA-1 M H3PO4. Once all
reactions had been stopped in this way, all filter papers were
washed as follows: 10 min, 10% TCA-1 M H3PO4; 15 m in 5% TCA at
70° C; 5 min in each of the following, in sequence: 100% ethanol,
ethanol: diethyl ether (1:1), chloroform: methanol: 12M HC1
(200:100:1), diethyl ether (Sommarin et al., 1981).
After the final
83
wash, filters were dried and placed in scintillation vials for analysis
either by Cherenkov counting or, after the addition of 7 mL
scintillation fluid, by scintillation counting.
For assays used in
specific activity calculations, the initial rate was generally obtained
at the 1 min reaction timepoint.
0-Sepharose chromatography.
During optimization, anion
exchange chromatography was carried out in various modes
including step and gradient elution, with various combinations of
equipment including FPLC, gravity-flow open columns, and
pumped open columns.
A typical fractionation run involved a 7.9
cm x 1.5 cm (-14 mL) Bio Rad Econo column with flow adapter,
attached to a Bio Rad Econo System gradient former/elution
monitor unit in a coldroom. Protein solubilized from 93 mg PM
protein, in a total volume of 70 mL, was pumped onto the column
at a rate of 0.3 mL/min, followed by an additional column volume
of starting buffer (10 mM Tris pH 8.5, 50 mM NaC1, 10% glycerol,
1% CHAPS, 100 gM DTT, and 100 µg/mL each of leupeptin and
pepstatin).
The kinases were eluted by developing the column at
the same flow rate with a 5-column volume NaC1 gradient, to 200
mM. Fraction size was 2mL. Fractions were frozen at -20° C to
await analysis. Actual NaC1 concentration in fractions was
determined by use of a conductivity meter.
Sephadex G-100 chromatography. Gel filtration
chromatography utilized two 1.5 cm columns connected in series so
that one column could be loaded at the top and eluted downward
while the other eluted upward.
The latter column (85 cm long)
84
was manufactured by Pharmacia, and the former (116 cm) was
constructed from a length of glass tubing, two rubber stoppers, and
two hypodermic needles, with a plastic grid and nylon mesh
covering the inside of the bottom stopper. Elution was by gravity,
at 9 mL/hr (pressure head 43 cm). Fraction size was 4.5 mL (30
min fractions). The column was calibrated using (3- amylase, BSA,
carbonic anhydrase, cytochrome c, and DNPG.
If necessary,
samples were concentrated before gel filtration using a hollowfiber centrifugal concentrator (Spectrum, Los Angeles, CA), so that
sample size was less than 2% of the bed volume. To minimize
detergent use, the column was charged with detergent-containing
elution buffer (10 mM Tris pH 8.5, 20 mM NaCl, 10% glycerol, 1%
CHAPS) before the sample was loaded.
The sample, mixed with an
additional 5% glycerol, was loaded under the elution buffer; the
column run used elution buffer without detergent.
Fractions were
frozen to await analysis.
Hydroxylapatite chromatography.
Initial hydroxylapatite
(HAP) experiments were carried out using HAP equilibrated in
phosphate buffer (1 mM phosphate buffer pH 6.8, 1% CHAPS, 10%
glycerol; phosphate mode), or in G-100 buffer (10 mM Tris pH 8.5,
50 mM NaCl, 1% CHAPS, 10% glycerol; NaCl mode). Post-G-100
material was either diluted and reconcentrated with phosphate
buffer (for binding in phosphate mode), or used directly (for
binding in NaC1 mode).
Appropriately prepared aliquots of post-G-
100 material (300 4) were mixed with 250
(packed volume)
of the appropriately prepared form of HAP and incubated on ice
85
for 5 min in 1.5 mL microfuge tubes. After incubation, HAP was
pelleted via a brief centrifugation in a microfuge and the
supernatants aspirated.
The pellets were washed once with the
appropriate starting buffer, and then each type of preequilibrated
HAP was sequentially eluted with the following solutes (in the
appropriate starting buffer):
phosphate.
5 mM MgCl2, 1 M NaCl, 400 mM
Aliquots from each elution step of each type of HAP
were assayed for protein kinase activity, with appropriate controls
to allow identification of effects on kinase activity by buffer salts.
Once HAP chromatography conditions had been optimized,
chromatography involved batchwise separations operated in
phosphate mode, with stepwise elutions at phosphate
concentrations as indicated.
Protein determination.
Protein concentrations were
estimated by the method of Schaffner and Weissmann (1973) or by
use of the Bio-Rad DC protein assay kit. Both methods are resistant
to interference by detergents.
BSA was used as the protein
standard with both methods.
Preparation of labeled material for peptide mapping. Post-QSepharose or post-G-100 material was labeled as described for
assay conditions above, except that the total volume was 75
0../lane, no exogenous substrate was added, and the reaction was
terminated after 20 min by the addition of 30 pi, of 2X SDS-PAGE
loading buffer and heating to 95° C for 2 min.
This material was
run on SDS-PAGE as described above, except gels with reduced
crosslinker concentrations, in which the final bisacrylamide
86
concentration was 0.067% instead of the usual 0.27%, were used.
Best results were obtained when these gels were run until the 60
kD region was approximately 2/3 of the way down the gel. Gels
were dried onto cellophane without fixation or staining, and
labeled bands were located by autoradiography.
Peptide mapping. Cleveland (1983) maps of phosphorylated
kinases were prepared as follows. Labeled bands were excised
with a razor blade and inserted into wells in the stacking gel of a
15% SDS-PAGE gel. Gel slices were overlaid with 60-80 1.1.L of gel
slice equilibration buffer (12.5 mM Tris-glycine pH 6.8, 0.1% SDS,
10% glycerol, 1 mM EDTA, 0.3% 2-ME, trace bromphenol blue)
containing the appropriate protease.
TLCK-treated trypsin (300 ng-3 gg),
Initial experiments involved
TLCK-treated chymotrypsin
(300 ng-3 p.g), subtilisin (10 ng-1 gg), papain (130-260 pl), and V8
protease (30-300 ng).
Electrophoresis was carried out at 16 mA
until the dyefront was in the middle of the stacking gel, at which
time power to the gel was turned off.
After 30 min electrophoresis
was resumed and continued until the dyefront reached the bottom
of the resolving gel. The gel was dried and subjected to
autoradiography with an intensifier screen.
Immu nos tain in g.
Following autoradiography, blots were
incubated with anti-CDPK monoclonal antibody, then with goat
anti-mouse-alkaline phosphatase secondary antibody as described
in Sambrook et al. (1989).
87
RESULTS
Partial purification. Work described in the previous chapter
demonstrated that most of the protein kinase activity present in
zucchini microsomes remains associated with the PM fraction
following aqueous two-phase extraction of PM, and the majority of
this kinase activity is solubilized by Triton X-100 or by the
zwitterionic detergent CHAPS.
Because of its superior properties of
UV-transparency, low critical micelle concentration, and small
micelle size, CHAPS was chosen for use in further purification.
In a
typical purification run, PM proteins from 1.8 kg of tissue were
solubilized and loaded onto a Q-Sepharose column. Kinase activity
bound to the column at 50 mM NaC1, and was eluted with a salt
gradient to 200 mM. Activity eluted in two peaks, one major and
one minor, centered around 140 mM (peak A) and 200 mM (peak
B) NaCl, respectively. Typical results are shown in Figure 7. An
additional peak of activity eluted during a 1 M NaC1 wash of the
column (peak C; data not shown).
Fractions comprising peak A, which contained the 58-68 kD
kinases, were pooled and concentrated to <2% of the bed volume
(357 mL) of the 2 m G-100 column. After concentration, the
sample (<7 mL) was loaded onto the column. To minimize detergent
use the column was charged with CHAPS-containing elution buffer
before loading and was eluted using detergent-free buffer.
As
shown in Figure 8, kinase activity eluted in a symmetrical peak
88
centered around Ve=165 mL, which corresponded to a molecular
weight of 60 kD.
Fractions containing activity were pooled.
Protein kinase
assays of material from different steps in the purification are
shown in Table 2.
Pooled post-G-100 material was aliquotted for
future use, including use in pilot experiments with numerous other
Most of these proved to be
unsatisfactory, and are listed in Table 3.
One purification technique that did show promise was
potential purification methods.
hydroxylapatite (HAP) chromatography.
Chromatography in the
phosphate mode with elution with phosphate buffer was the most
adequate in terms of recovery of activity. Kinase activity eluted
only with elution solutions of pH 8.5, not pH 6.5, although proteins
did elute at pH 6.5. When protein was bound to HAP at pH 8.5 and
eluted with a step-gradient of increasing phosphate concentrations,
most activity eluted at 60-75 mM, while most protein eluted at
phosphate concentrations of <45 or >400 mM (data not shown).
2-D electrophoresis of post-HAP material. To determine the
level of contamination of the post-HAP kinase preparation by other
proteins, an aliquot of this material was labeled by incubation with
[y- 32P]ATP and analyzed by 2-D (IEF/SDS-PAGE) electrophesis.
Several phosphorylated spots were present, including several sets
of spots that appeared to comprise charge trains at four molecular
weights in the 58-68 kD range. To the extent that it was possible
to correlate individual stained proteins with radiolabeled spots,
few pairs of such spots were apparent (data not shown).
89
Further 2-D analysis was used to ascertain whether stained
spots could be seen to change pI on phosphorylation, in an effort to
gauge the amount of kinase protein present in the preparation.
Two samples were run on paired IEF and SDS gels. One sample was
incubated with ATP before electrophoresis, while the other sample
was not phosphorylated before electrophoresis. The gels were
analyzed to test the hypothesis that the sets of radiolabeled spots
arrayed in this molecular weight range were members of charge
trains. Although proteins did become radiolabeled in the
phosphorylated sample, the two staining patterns were nearly
identical in the molecular weight region of interest, as shown in
Figure 9. No changes in the location of silver stained spots were
evident, suggesting that at least some of the kinases were present
in amounts below the level of detection, or that phosphorylation
did not cause an appreciable change in the pI of the proteins.
Further analysis of Q-sepharose peaks. One hypothesis for
the genesis of peak B (see Figure 7) was that it represented a more
highly phosphorylated (and therefore more anionic) population of
the kinases in peak A. However, incubation of solubilized proteins
with ATP before Q-sepharose chromatography resulted in the same
peak A-peak B elution pattern as is shown in Figure 7 (data not
shown), which indicated that the proteins present in the two peaks
were distinct.
Material which eluted from the Q-sepharose column was
analyzed in more detail by western blotting and peptide mapping.
For western blotting, aliquots of fractions from within peak A and
90
peak B, as well as a more anionic peak, peak C, were subjected to in
situ labeling on nitrocellulose blots before being incubated with
primary and secondary antibody.
As shown in Figure 10, while
both peaks contained protein kinase activity, only peak B contained
material which was immunodecorated by a-CDPK monoclonal
antibody.
The third peak (peak C), which eluted at much higher
salt concentrations, also contained a higher molecular weight
kinase which was also recognized by the a-CDPK monoclonal
Immunostained and radiolabeled bands from peak B and
peak C comigrated. The immunostained and radiolabeled patterns
antibody.
were unchanged by preincubation of peak A or B material with
ATP or with calf intestinal phosphatase (data not shown).
Peptide mapping of kinases from peaks A. B & C.
Preliminary
mapping experiments led to the conclusion that V8 protease (from
Staphylococcus aureus) was the most appropriate protease for use
in further mapping experiments (data not shown). This protease
was used to prepare Cleveland maps from the radiolabeled bands
derived from peaks A, B, and C, as shown in Figure 11. Table 3
shows the results of analysis of the autoradiogram shown in Figure
11.
Maps of bands from within each peak are very similar, while
comparison of bands from peaks A and B reveals more differences.
91
.
0
10
20
30
Fraction
40
Figure 7. Typical results of Q-Sepharose chromatography of
CHAPS-solubilized PM. A 7.9 x 1.5 cm column of Q-Sepharose
was loaded with CHAPS-solubilized PM protein in starting
buffer (10 mM Tris pH 8.5, 50 mM NaCl, 10% glycerol, 1%
CHAPS, 100 i.i.M DTT, 100 µg /mL leupeptin and pepstatin).
After washing with one column volume of starting buffer, a 5
column-volume linear gradient to 200 mM NaC1 was initiated.
Flow rate was 0.3 mL/min, fraction size was 2 mL. NaCl
concentrations shown on the figure were determined with a
conductivity meter.
92
60
50
M
Ic,
,(
x
40
30
20
10
0
100
150
200
250
300
Ve
Figure 8. Typical G-100 elution profile. Fractions from peak A of
Q-Sepharose chromatography were concentrated and loaded
onto a 1.5 x 2 m G-100 column which had been
preequilibrated with Q-Sepharose starting buffer.
After
loading, column was eluted with CHAPS-free buffer at a flow
rate of 9 mL/hr; 4.5 mL fractions were collected. Molecular
weight standards were 0-amylase (Vo; 200 kD), BSA (66.2
kD), carbonic anhydrase (29 kD), and cytochrome c (12 1(13).
93
A.
kD
B.
...MIN*
.
-220
-117
--.
97
ti 4
1
441
-
'1
411'
66
43
.
Figure 9.
J.
31
Silver-stained 2-D gels loaded with unphosphorylated
and phosphorylated post-HAP material.
Panel A,
unphosphorylated proteins; Panel B, phosphorylated proteins.
The constellation of spots at 31 kD is a mixture of pI
standard proteins.
Acidic proteins migrate to the right.
94
P
32
AB
Western
C
AB
1111'.846
C
1(1)
241
117
75
48
Figure 10. In situ phosphorylation and western blot analysis of QSepharose peaks A, B, and C. The left panel shows results of
an overnight autoradiogram.
The same blot was
immunostained with anti-CDPK monoclonal antibody, as
shown in the right panel.
95
A
1
C
B
2
3
I
1
1
2
3
4 5
;
97.4
4 44
I
4
kD
4
1
66.2
42.7
31.0
e
21.5
14.4
Figure 11.
Cleveland maps of radiolabeled proteins from Q-
Sepharose peaks A-C.
Bands 1-3 from peak A, bands 1-5 from
peak B, and the high molecular weight band from peak C were
radiolabeled, resolved on SDS-PAGE, and subjected to cleavage with
Staphylococcus aureus V8 protease in the stacking gel of a 15%
acrylamide SDS-PAGE gel before electrophoresis to resolve labeled
fragments.
96
Table 2.
Sample
Purification Table
mg
Total
%
protein Activity
Microsomal membrane
Plasma membranes
68
CHAPS supernatant
42
CHAPS pellet
20
26.5
32.8
4.2
Q-Sepharose peak A
G-100
Hydroxylapatite
3.9
10.1
1.6
6.9
Specific Activity
(pmol mg-1 min-1)
0.18 (typical)
100 0.39
124 0.78
16 0.21
38
2.6
26 4.3
9.5 (estimated)
97
Table 3.
Protein Purification Techniques Tested
with Solubilized Protein Kinases.
Technique
Reason Unsatisfactory
Ammonium sulfate precipitation Activity precipitated over wide
range of [salt]
Immunoprecipitation (a-CDPK
No immunoprecipitation
antibodies)
Hydroxylapatite:
FPLC
Packing material collapsed ->
very high backpressure
Triton X-114 phase separation
Acivity equally divided
between both phases
ATP-agarose affinity
Activity not bound
chromatography
Q-sepharose + EDTA
Results same as -EDTA
Phenyl-sepharose
Very poor resolution, activity
chromatography
not bound
Cibacron Blue affinity
Lost enzymatic activity/
chromatography
activity not eluted
Mono Q: FPLC
Lost enzymatic activity
Affinity chromatography of a-
Activity not bound
CDPK-kinase mixtures on
Protein A agarose
98
Table 4.
Calculation of P(x) from results of Cleveland maps
1
# bands 40A
x
40A
40B
40B
40C
P(x)
x
P(x)
x
40A
40B
17
21
17 9.0 E-16
40C
25
13
15
17
16
2.3 E-11
5
68A
68B
68C
68D
68E
-
21
6
4
29
9
0.03
0.02
0.03
0.24
0.02
1
0.41
72
6
6
# bands 68A
x
40A
40B
21
40C
25
68A
68B
13
68C
68D
68E
72
40C
P(x)
20 2.0 E-15
6
0.02
6
0.04
4
0.24
6
0.21
0.21
0
0.23
5
68A
68B
P(x)
x
68B
P(x)
0.05
0.15
0.06
0.13
6
2
7
7
2
0
4.4 E-03
0.17
68C
68C
x
P(X)
17
-
15
11
17}
21
10
10
29
6j
8
1
6.9
2.5
3.9
7.9
E-10
E-07
E-06
E-03
0.40
-
14 2.2 E-13
13 2.5 E-09
11 1.3 E-04
1
0.41
15
11
1
2.5 E-11
7.5 E-04
0.41
99
DISCUSSION
When microsomal membranes from zucchini hypocotyls were
subjected to aqueous two-phase partitioning to isolate PM vesicles,
the protein kinase specific activity increased, indicating most of the
kinase activity was associated with the PM. This is consistent with
results from other laboratories (Kulcis and Polya, 1988, Ladror and
Zielinski, 1989). Most of the PM kinase activity was soluble in 1%
CHAPS, which was used in elution buffers of all chromatographic
systems.
Plasma membrane preparation and solubilization
resulted in a greater than 4-fold enrichment of kinase activity
relative to microsomes.
According to the casein phosphorylation assay, most of the
kinase activity eluted from the Q-sepharose column in peak A, at
around 140 mM NaCl.
Analysis by in situ labeling on nitrocellulose
showed this peak contained kinases in the expected molecular
weight range, and it was used in further purification efforts.
A
second, much smaller and somewhat more anionic peak of activity
was eluted at around 200 mM NaCl; this peak is discussed at
greater length below.
When material from peak A was fractionated on a 2 m G-100
column, all kinase activity eluted in a single peak around 60 kD,
approximately the same the molecular weight range seen on
labeled blots of SDS-PAGE gels.
This suggests the kinases are not
associated with other proteins under the conditions used for
elution.
100
Specific activity calculations for purification steps up to and
including G-100 chromatography are shown in Table 2.
The
zucchini enzymes are enriched approximately 25-fold over
It is impossible to compare these results to
those of Putnam-Evans et al. (1990), who purified CDPK to
microsomal levels.
homogeneity, because the CDPK purification used much different
techniques, including phenyl Sepharose, cellulose-DEAE, and
Cibacron blue-Sepharose.
Schaller et al. (1992) partially purified
an apparent CDPK from oat root plasma membrane, using
techniques similar to those detailed here, reporting a 114-fold
increase in specific activity. However, their calculations used data
from an assay time point that is well past the initial rate in our
hands.
Calculations using the approach of Schaller and coworkers
for our enzyme(s) yield a specific activity increase of at least 258
fold for the zucchini kinases.
A number of other purification techniques were tested on an
analytical scale. Most proved unsatisfactory for reasons ranging
from poor resolution to enzyme inactivation (see Table 3).
One purification technique that did show promise was
hydroxylapatite (HAP) chromatography. HAP chromatography can
be carried out in any of three different modes, depending on the
nature of the proteins of interest and the protein-matrix
interaction (Gorbunoff, 1985).
HAP chromatography with
phosphate buffer elution, the most general approach, was the most
adequate in terms of recovery of activity. Kinase activity eluted
only with elution solutions of pH 8.5, not pH 6.5, although many
101
proteins did elute at pH 6.5. Evidently chromatography on HAP at
this pH either inactivated the enzymes, or the enzymes bound
tightly to the matrix and failed to elute.
When protein was bound to HAP at pH 8.5 and eluted with a
step gradient of increasing phosphate concentrations, most activity
eluted at 60-75 mM, while most protein eluted at phosphate
concentrations of
45 or 400 mM.
Judging from the change in
protein concentration of the kinase fraction, this yielded an
estimated increase in specific activity of 5-fold. This material was
labeled and analyzed by 2-D electrophoresis to test the possibility
that stained spots corresponding to radiolabeled spots could be
identified.
The pattern of radiolabeled spots suggested a series of
charge trains in the molecular weight range of interest, and it was
difficult to correlate stained spots with radiolabeled spots (data not
This experiment did show that there here were relatively
few proteins associated with the 60-75 mM HAP fraction. Thus, it
shown).
seemed that it should be possible to make a comparison of
phosphorylated and unphosphorylated material on matching 2-D
gels, visually identifying changes in the stained spot pattern which
corresponded to kinase charge trains.
A comparison of 2-D gels
loaded with phosphorylated and unmodified material showed no
difference in the staining pattern in this region. This observation,
combined with the limited correlation between radiolabeled and
stained spots, suggested that even in at this level of enrichment,
estimated at over 100-fold, insufficient kinase protein was present
for localization of at least some spots by silver staining. Silver
102
staining can detect protein amounts of 2-10 ng, and each 2-D gel
was loaded with protein from the equivalent of 100 gFW of
zucchini hypocotyls, suggesting some kinases are rather low
abundance proteins from which it would likely be difficult to
obtain sequence data.
Peptide mapping of labeled fragments,
which does not depend on purified protein for success, was chosen
as a more efficient route to answering the question of relatedness
of the kinase bands to one another. It should also be useful in
determining the degree of similarity between the zucchini kinases
and CDPK.
Q-Sepharose chromatography removes enough of the
endogenous kinase substrates that phosphorylation of this material
before electrophoresis yields labeled bands at the same molecular
weights as bands detected by labeling following immobilization on
nitrocellulose or by immunostaining.
Few proteins of other
molecular weights are labeled, suggesting other substrate proteins
were removed by Q-Sepharose chromatography.
Thus, post-Q-
Sepharose material seems suitable as a source of labeled kinase for
mapping studies.
Further analysis of material eluting from Q-Sepharose led to
a reconsideration of the importance of peak B, which had
previously been discarded because it appeared to contain only
a
fraction of the activity present in peak A. Because peak B
contained material that was somewhat more anionic than peak A,
one initial hypothesis was that peak B contained more highly
phosphorylated kinases that were otherwise the same as those in
103
peak A.
This was not supported by data from experiments in
which proteins were phosphorylated (by incubation with ATP) or
dephosphorylated (by incubation with calf intestinal phosphatase)
before chromatography.
Analysis of material from peak A, peak B,
and a third much more anionic peak of activity (peak C) by in situ
phosphorylation and western blotting led to an interesting
observation: while all peaks contain kinase activity, only peaks B
and C contain material that is also immunodecorated by anti-CDPK
monoclonal antibody (see Figure 10).
The epitope recognized by
the antibody is in or near the nucleotide binding site, which is
composed of regions known to be well in from the N-terminal end
of PKA (the protein kinase for which the most structural
information is available; Knighton et al., 1991). Thus it is unlikely
that simple proteolysis could have removed the epitope to generate
peak A kinase from the peak B or C enzymes. Removal of the
epitope would also presumably affect kinase activity.
When the kinases present in Q sepharose peaks A-C were
phosphorylated and analyzed by SDS-PAGE, several radiolabeled
bands were resolved.
Each band was excised and analyzed by
Cleveland mapping, the results of which are shown in Figure 11.
Map patterns can be compared by calculation of P(x), the
probability that x fragments are identical by coincidence, through
application of the following equation from Calvert and Gratzer
(1978):
P(x) = m!n!(N-m)!(N-n)! / N!x!(m-x)!(n-x)!(N-m-n+x)!
104
where N is the total number of bands, m and n represent the
number of bands in the compared patterns, and x is the number of
bands with identical mobility.
P(x) < 0.01 is a common standard for
proteins which are highly related (Calvert and Gratzer, 1978).
P(x)
for various comparisons are shown in Table 4.
The three bands present in peak A yield very similar peptide
maps (P << 0.01). The five bands of peak B seem to fall into an
additional class:
each band of peak B is very similar to every other
band of peak B but is generally unrelated to bands of peak A. The
data are also consistent with the conclusion that the large
immunostaining band of peak C is a member of a third class of
kinase, unrelated to the other two.
The labeled polypeptide from
peak C is either not very highly phosphorylated, or has relatively
few cleavage sites, as it yields only 6 labeled fragments on
digestion.
Other polypeptides in this experiment formed an
average of about 17 labeled fragments on digestion.
It is
interesting to note that soybean CDPK autophosphorylates only
very poorly (Alice Harmon, University of Florida, personal
communication), as does the 79 kD oat root kinase (Schaller et al.,
1992; Eric Schaller, University of Wisconsin, personal
communication).
It is possible that one or more of the peak A or peak B bands
were contaminated by labeled protein from adjacent radiolabeled
bands, since all bands were of similar molecular weight. This does
not affect the observation that each peak contains material which
105
results in very different peptide maps, however.
Kinases which
are recognized by anti-CDPK monoclonal antibody thus produce two
very different patterns of labeled fragments on peptide maps,
suggesting that two classes of kinase with quite different
sequences share similar nucleotide binding sites, or that
phosphorylation patterns of the two members of a single class of
kinase are quite different.
106
ACICNOWLEDGEMEN'IS
I thank Reg McParland for advice concerning protein
purification.
Tom Wolpert suggested use of peptide mapping to
compare the different kinases.
107
Analysis of factors affecting labeling of tomato and zucchini
auxin binding proteins by azido auxin
INTRODUCTION
Protein kinases and/or other signal transduction elements
likely link intracellular proteins and processes to changes in
The search for elements of an
auxin signal transduction chain has also focused on G-proteins
(Zaina et al., 1990, 1991), phospholipases (Zbell et al., 1989; Andre
and Scherer, 1991), and ion channels (Blatt and Theil, 1993). While
the details of the linkage between auxin, signal transduction
extracellular auxin concentration.
elements, and changes in cellular processes are still unknown, the
first step in the auxin signal transduction chain is presumed to
involve some sort of receptor protein (Jones and Venis, 1989). As
is the case with protein kinases, by analogy to animal systems it
seems reasonable to expect that at least part of this hypothetical
receptor pool would be associated with the plasma membrane.
In
animal systems there is considerable interplay between protein
kinases and receptors.
Perhaps the most active single avenue of auxin signal
transduction research has involved auxin binding proteins (ABPs),
which are of interest because they may represent auxin receptors.
108
Other possible identities of ABPs include enzymes associated with
auxin metabolism, or mediators of auxin transport (Jones and
Venis, 1989).
An important change in our ability to work with
auxin binding proteins came in the mid-1980's, with the
development of auxin analogues which could be used as
photoaffinity probes for auxin binding proteins.
Photoaffinity probes consisting of a ligand coupled to an azido
group (-N3) have been used for over 20 years to identify and
characterize ligand binding proteins, beginning with the work of
Fleet et al. (1969). When photolyzed by light of the appropriate
wavelength, photoaffinity probes covalently crosslink with target
proteins, providing a convenient method for "tagging" proteins of
interest.
A plant hormone-based photoaffinity probe, azido auxin
(azido-IAA; 5-azido-[7-3H]-indole-3-acetic acid) apparently binds
specifically and reversibly in the dark to auxin binding sites (Jones
et al., 1984a) and is thus a potentially useful probe for auxin
binding proteins (ABPs) in plants.
Activation of azido auxin, or any molecule bearing an azido (-
N3) group, occurs when the compound is exposed to ultraviolet
light.
On photolysis, the azido group liberates nitrogen (N2) and
forms a highly reactive nitrene (-N:) group which is capable of
participating in reactions which include insertion into -C-H or -0-H
bonds.
Insertion reactions can lead to covalent attachment to
amino acids (Herspool, 1976).
Azido auxin binding proteins were intially reported in maize
(Jones et al., 1984b).
Several maize azido auxin binding proteins
109
have now been identified, including microsomal proteins with
molecular weights of 22 and 24 kD (Jones and Venis, 1989); and
PM proteins of 23, 58 and 60 kD (Feldwisch et al., 1992). The 23
kD maize PM ABP may be involved with auxin transport, as
suggested by the observation that it is also labeled by 5'- azido[3,6- 3H2] -1- napthylphthalamic acid, a photoactivatable analogue of
napthylphthalamic acid (Zettl et al., 1992).
Plasma membranes of zucchini and microsomes from tomato
contain polypeptides with molecular weights of 40 and 42 kD
which are photoaffinity labeled by azido auxin (Hicks et al., 1989a,
1989b).
Hypocotyls of the tomato mutant diageotropica (dgt) have
been reported to contain greatly reduced levels of the 40 and 42
kD proteins, while roots appeared to contain wild type levels (Hicks
et al., 1989b). Proteins of similar size were found to be labeled by
azido-IAA in a variety of plants, including conifers, monocots, and
other dicots (Lomax and Hicks, 1991).
In well-studied signal transduction systems there are
numerous modes of interaction between protein kinases and
receptors for signaling molecules.
These include both activation of
kinases by receptors, and modulation of receptor sensitivity by
phosphorylation-dephosphorylation (Collins et al., 1992).
It would
be of interest to examine the interaction between the protein
kinases present in plant plasma membranes and the potential
auxin receptors which are also present there. One important
question that might be asked is whether phosphorylation of an
ABP affects the protein's affinity for auxin.
Zucchini ABP's do form
110
multiple charge isoforms on 2-D gels, consistent with
phosphorylation or other posttranslational modifications (Hicks,
Some questions about interactions between protein kinases
and ABP's could, in principle, be answered with the help of
1992).
photoaffinity auxin analogues such as azido auxin.
Because of this
potential to link the protein kinase work described elsewhere in
this thesis with work on ABPs that is ongoing in this laboratory, a
subproject was initiated to determine the feasibility of using
photoaffinity labeling to help forge such a linkage.
Although considerable phenomenology has been collected on
azido-IAA labeling of maize, zucchini, and tomato proteins, most
work has focused on the putative auxin-binding proteins labeled
by azido auxins, and little time has been spent exploring factors
affecting labeling (Campos et al., 1991).
Such an analysis is
important to assess the potential for azido-IAA labeling, if it can be
applied reliably, to permit detection of the effects of protein
kinases on auxin binding and vice versa.
Azido labeling has
already been used to analyze genotype-specific anomalies in auxin
responses.
However, unless reliable azido-IAA labeling can be
validated, such uses may not be appropriate; inconsistency in
azido-IAA labeling has been experienced in both our laboratory
and at least one other laboratory with extensive experience in
azido-IAA labeling (Alan Jones, University of North Carolina,
personal communication).
The study described here was
undertaken to investigate labeling inconsistencies and to identify
and test parameters potentially critical to the use of photoaffinity
111
labeling in the analysis of ABPs. Factors including azido auxin
quality, duration and spectral characteristics of irradiation,
detection of radiolabel, source and preparation of biological
material for labeling, and reaction chemistry were considered.
112
MATERIALS AND METHODS
Radiochemicals and Chemicals. Tritiated azido auxin (5-
azido47-311]-indole-3-acetic acid) was obtained from Alan Jones
(University of North Carolina, Chapel Hill, NC). The ethanolic
solution (14 p,M, 16 Ci /mmol) was stored at -80° C. At the time of
these experiments, the lot of azido auxin in use was 6 years old.
Samples of azido auxin were available that had been exposed to
room temperature and to the possibility of illumination for various
numbers of cycles, ranging from once to numerous exposures.
Acrylamide was obtained as a 40% solution from BDH (Poole, UK).
Other reagents used in electrophoresis (including
dihydroxyethylenebisacrylamide, DHEB A) were obtained from BioRad (Richmond, CA). Except where noted all other chemicals were
from Sigma (St. Louis, MO).
Plant Material and Preparation of Membrane Vesicles. Seeds
of zucchini (Cucurbita pepo L. cv Dark Green) were obtained from
Lilly-Miller.
Seeds of the tomato genotypes dgt and VFN-8 (VFN)
were obtained from field-grown plants (OSU Botany Farm).
Seeds
of both species were grown for microsome production as follows.
Seeds were treated with 10% bleach containing a few drops/L of
Tween 20 for 10 minutes. After being rinsed with running tap
water, seeds were sown on several layers of moist unbleached
absorbent paper (Kimtowels; Kimberly-Clark, Rosewell, GA) in
polycarbonate boxes ("sweater boxes") which had previously been
113
treated with bleach and rinsed with tap water.
Seeds were grown
for 5-7 days at 27° C in the dark before harvest of hypocotyls.
Microsomal membrane vesicles were prepared from freshlyharvested hypocotyls by homogenization with a Brinkman Polytron
in an equal volume of ice-cold Buffer 1 (250 mM sucrose, 10 mM
Tris-Cl pH 7.5, 1mM EDTA, 1 mM DTT, 0.1 mM MgSO4) plus, except
where indicated, protease inhibitors (0.2 mM PMSF, 1
leupeptin and pepstatin).
1.1.g each
The brei was filtered through 4 layers
of cheesecloth, and the retentate was rehomogenized with an
additional volume of Buffer 1.
Filtrates were pooled and
centrifuged for 15 min at 1000 x g and 3° C (Beckman GPR, GH3.7
Microsomes were pelleted from the supernatant by a 30
rotor).
minute centrifugation at 105,000 x g and 3° C (Beckman L5-65,
Ti50.2 rotor).
Microsomal pellets were resuspended in Buffer 1
without DTT and either used immediately or flash-frozen in liquid
N2 and stored at -80° C until use.
Photoaffinity Labeling.
Except where noted, photoaffinity
labeling was carried out essentially as described by Hicks et al.
(1989a).
All operations were performed under dim red light.
Briefly, membrane samples were mixed with labeling buffer (10
mM BTP-MES pH 6.5, 250 mM sucrose, 0.1% Triton X-100) and
azido auxin (final concentration 0.28 I.LM) in UV-transparent plastic
cuvettes.
The cuvettes were transferred to an aluminum block
situated in a liquid nitrogen bath.
After several minutes (to allow
the samples to reach -196° C) they were irradiated from above for
30 sec with UV light (300 nm) from a Fotodyne Model 3-3000
114
The surface of the transilluminator filter was 2.5
cm from the samples. It was often convenient to transfer the
transilluminator.
samples at this point to -20° C and store them until they could be
analyzed by SDS-PAGE, usually overnight.
Before electrophoresis,
the samples were removed from the freezer, and SDS loading
buffer added as they warmed to room temperature.
Samples were
loaded on gels after heating for 2 min at 95° C.
Electrophoresis.
SDS-PAGE was carried out according to
instructions furnished with the Bio-Rad Protean II electrophoresis
unit.
Except where otherwise noted, all resolving gels contained
12% acrylamide and 0.32% bisacrylamide, and were 15 cm long and
0.75 cm thick. For dissolvable gels, bisacrylamide was replaced
with DHEBA. Slices from gels prepared with DHEBA dissolved in
less than 45 minutes in 0.5 mL aliquots of 200 mM periodic acid at
80° C.
Fluorography and scintillation counting. Following
electrophoresis and Coomassie staining, gels were rinsed with
water for 30 min and then soaked in Fluoro-Hance (Research
Products International, Mt. Prospect IL) for 30 min at room
temperature. Gels were dried and exposed to Kodak XAR film at
-80° C for the indicated times. For experiments involving film
hypersensitized by preflashing, a Vivitar Model 2200 flash unit
with a Kodak Wratten 22 filter and one layer of Whatman No. 1
filter paper covering the lamp was triggered at a distance of 1 m
from the film (Lackey and Mills, 1975).
115
For scintillation counting, samples were mixed with 7 mL of
DuPont Formula 989 (DuPont, Wilmington DE) and analyzed in a
Beckman Model LS 6800 scintillation counter on the tritium
channel.
High Pressure Liquid Chromatography. HPLC utilized a 4.6 x
250 mm Beckman ODS column attached to a Beckman System Gold
All elutions were carried out isocratically, with
50% methanol-water, at a flow rate of 1 mL/min. Fractions were
collected at 0.9 min intervals. Ultraviolet detection, when used,
chromatograph.
was at 244 nm.
To ensure minimum exposure of samples to room
light, initial analytical and preparative runs were performed after
sunset, using only a small flashlight equipped with a red filter for
light.
For later runs, the chromatography system was moved to a
darkroom, where red safelights were used for illumination.
After
preparative scale chromatography, fractions containing azido auxin
were pooled and concentrated in amber microfuge tubes using a
Speedvac (Savant, Farmingdale, NY), which was covered to prevent
light entry. Concentrated samples were reconstituted to the correct
volume with absolute ethanol.
Microbiology and axenic growth of tomatoes.
Microorganisms
associated with the roots of tomato seedlings were analyzed by
Gram staining and culture on NZY plates. For axenic growth, seeds
were surface sterilized as usual, rinsed with sterile distilled water,
and sown on sterile 1% water agar or sterile paper towels in a
laminar flow sterile cabinet. Axenically grown hypocotyls were
harvested and labeled as usual.
116
RESULTS
HPLC analysis. Analytical HPLC was used to determine the
radiochemical purity of the azido-IAA.
As shown in Figure 12,
chromatography of unphotolyzed samples yielded a symmetrical
peak which eluted from the C18 column at 7.9 min. This peak
contained the majority of 3H.
Numerous other UV-absorbing peaks
were present, including at least one, at 2.6 min, which also
contained 3H. Photolysis at 300 nm for 15 sec before
chromatography affected the size of many of the peaks; the 7.9 min
peak was strongly affected but not completely eliminated, while
the 2.6 minute peak was somewhat increased in size, and an
additional 3H-containing peak appeared at 15 min (Figure 12).
Intensity of peaks other than the one at 7.9 min was positively
correlated with the extent of exposure of the sample to light and
elevated temperature (data not shown). This indicated that the 7.9
min peak contained azido auxin, and that other peaks, including the
one at 2.6 min, represented photo- and/or thermal degradation
products.
Rechromatography of samples eluted with UV detection
showed that UV detection degraded the azido auxin as it passed
through the flow cell.
HPLC purification of azido auxin. Preparative-scale HPLC was
used to remove photoproducts from azido auxin, and was carried
out without UV detection. By concentrating the sample before
chromatography, it was possible to process up to 7 ml (-11 mCi) of
117
azido auxin stock solution per chromatography run.
Figure 13
shows that purification removed the 2.6 minute 3H-containing peak
as well as most of the untritiated UV-absorbing peaks.
Effect of azido auxin purification on labeling. Purified azido
auxin was compared with untreated azido auxin in three
experiments in order to ascertain whether purification resulted in
improved labeling efficiency.
Freshly harvested and frozen dgt
and VFN microsomal membrane preparations, and frozen zucchini
microsomal membranes were tested. In two of the experiments
purified azido auxin appeared to label the 40-42 kD bands slightly
more efficiently.
In the third experiment use of purified azido
auxin clearly yielded a greater signal than untreated azido auxin.
Figure 14 shows the results of all three experiments. Increasing
300 nm irradiation time from 30 sec to 5 min had no effect (data
not shown).
Effect of genotype.
Hicks et al. (1989b) reported that the
40-42 kD ABPs of the tomato mutant dgt are either present in
greatly diminished quantity, or have much reduced ability to bind
auxin relative to the wild type VFN.
Under the conditions used
here in attempts to repeat those experiments, genotype appeared
to have an unpredictable effect on labeling.
In two experiments,
equivalent labeling was seen in dgt and VFN, while in one
experiment dgt labeled better than VFN. In two other experiments
40-42 kD proteins from VFN were more heavily labeled. Figure 15
shows the results of four of these experiments; two more are
shown in Figure 14. In one of the experiments shown in Figure 14
118
both dgt and VFN labeled very poorly, while zucchini PM labeled
more efficiently.
In addition to experiments performed as part of
this study, at least 15 other experiments to compare azido-IAA
labeling of dgt and VFN have been completed, with qualitatively
similar results (Hicks, Hopkins, Lomax, Ray le, unpublished results).
Presence and effect of root contamination. Tomato plants
grown under nonaxenic conditions often showed signs of pathology,
including root necrosis; abnormally short (<1 cm), pinkish taproot;
and lack of root hairs or lateral roots. Boxes containing such
seedlings often had a foul smell.
Culture of roots indicated the
presence of at least four different microorganisms, based on colony
morphology and Gram's staining.
When surface-sterilized seeds
were grown on sterile water agar or sterile paper towels, root
morphology was normal and healthy-looking, with long white
taproots bearing profuse root hairs.
The possible effect of root
infections on photoaffinity labeling was examined.
There was no
difference in labeling between contaminated and axenically-grown
seedlings (data not shown).
Effects of other factors.
Several other factors with the
potential to affect labeling were identified and subsequently
tested.
These included presence or absence of elevated ethylene
levels during growth of tomatoes before harvest, presence or
absence of protease inhibitors during isolation of membranes and
preparation for labeling, presence or absence of added or residual
DTT during labeling, freezing of microsomes before labeling, and
preincubation of membrane vesicles with azido-IAA label before
119
photolysis.
Of these treatments, addition of DTT (final
concentration 100 ilM) had a clearly negative effect on labeling.
Added DTT inhibited labeling, while the presence of DTT during
only membrane isolation had no effect (data not shown).
In
general, none of the other individual treatments had any effect on
labeling.
Isotope detection.
Several parameters involved in detection
of the 3H signal were tested.
Use of preflashed (hypersensitized)
film clearly increased the sensitivity of fluorography.
This effect
included amplification of signals not detectable on unflashed film
(data not shown).
It was also observed that use of fresh
fluorographic enhancer was essential (data not shown).
Neither of
these observations could explain labeling variability.
Because fluorography, although sensitive, requires exposures
of many days, the possible utility of polyacrylamide gels
formulated with a crosslinker which could be dissolved by periodic
acid was tested.
Dissolution of gel slices permits analysis by
scintillation counting.
However, when dissolved gels were mixed
with scintillant and analyzed by scintillation counting, backgrounds
of 2000-11,000 cpm were detected (data not shown).
Overnight
incubation of samples in the dark, as recommended by the
manufacturer (to allow fluorescence to decay), did not reduce
background levels to acceptable levels.
120
A
B
0.02
5
10
0
5
10
15
Minutes
Figure 12.
IAA.
HPLC analysis of unphotolyzed and photolyzed azido-
Panel A shows HPLC analysis of unphotolyzed, 6-year-
old azido-IAA stock solution. Panel B shows HPLC analysis of
an aliquot from the same stock, after 15 sec of 300 nm
irradiation.
121
0.02
10
T
5
Minutes
Figure 13. HPLC analysis of azido-IAA following preparative HPLC
purification.
122
i
z
Figure 14.
P
d
U
1
v
z
d
P
v
d
U
v
d
P
v
U
Effect of azido-IAA purification on photoaffinity
labeling.
Results of three experiments are shown.
P, purified
azido-IAA; U, unpurified azido-IAA; z, zucchini plasma
membranes; d, dgt tomato microsomal membranes; v, VFN
tomato microsomal membranes.
123
d
D
C
B
A
v
d
v
d
v
ABP {
Figure 15.
Results of labeling experiments comparing dgt and VFN.
Results of four separate experiments are shown; see Figure
15 for results of two other comparisons. ABP, auxin binding
protein; d, dgt tomato microsomal membranes; v, VFN tomato
microsomal membranes.
124
DISCUSSION
Importance of azido auxin quality. This chapter describes the
synthesis and testing of a number of hypotheses in an attempt to
explain apparently inconsistent and irreproduc able photoaffinity
labeling of zucchini and tomato proteins by azido auxin (Figures 14
and 15, and data not shown).
Initial attention focused on
potential degradation of the azido auxin reagent itself, as it
appeared that the problems with labeling efficiency had developed
gradually.
Although it had been stored properly (-80° C, in the
dark), the azido auxin in use at the time of these experiments was
approximately 6 years old.
Radiochemical purity of the azido auxin was found to be
approximately 80%, while radiochemical purity of freshly synthesized azido auxin is >99% (Jones et al., 1984b). Because
photoproducts of azido auxin can compete with intact azido auxin
for binding to proteins (Melhado et al., 1984), it was deemed
important to purify a quantity of azido auxin for use in
optimization experiments.
Both dgt and VFN responsed similarly in
experiments comparing purified and unpurified azido auxin- -that
is, impurities in the azido auxin, if they had any effect, did not
have a differential effect on dgt and VFN (Figure 14). In one such
experiment, dgt and VNF proteins labeled very poorly, while
zucchini proteins labeled relatively well.
This experiment will not
125
be considered as part of the comparison between dgt and VFN,
although it is indicitive of the variability which was seen.
Purification of azido auxin did appear to improve labeling,
but the effect was not dramatic, suggesting other factors needed to
be considered.
Irradiation.
Azido auxin displays an absorption maximum at
244 nm that is absent in IAA (Campos et al., 1991). The azido
group is apparently responsible for this absorption.
Thus, it would
theoretically be best to irradiate labeling mixtures at this
For historical reasons, 300 nm is the wavelength used
in this laboratory. Azido auxin absorbs very little at 300 nm,
wavelength.
suggesting use of shorter wavelength light could improve labeling.
Diverse light sources are used by other laboratories working with
azido auxins:
Jones's laboratory uses light from segments of two
relatively low powered (6 W and 15 W), long wave (365 nm) tubetype UV lamps, filtering the emission through Pyrex (Melhado et
al., 1984). A water cooled 125 W high pressure UV lamp has also
been used to photolyze azido-IAA (Campos et al., 1991).
High
pressure UV lamps emit an intense, relatively smooth continuum of
light throughout the UV spectrum (Phillips, 1983), and appear to
be most commonly used for azido group photolysis (Bayley and
Knowles, 1977).
Low pressure lamps, such as those used by
Melhado et al. (1984) and in this study, emit line spectra primarily
at 254 and 365 nm (Phillips, 1983). The output of a low pressure
source can be modified by the use of filters, such as the 300 nm
filter used in our laboratory.
Because of the specialized nature of
126
the lamp used by Campos et al. (1991), it was not possible to test
the effect of such a light source on labeling in our system.
However, it was possible to test the effect of removing the 300 nm
filter before irradiation, as well as the effect of increasing
irradiation time.
Neither removal of the filter nor increasing the
irradiation time from 30 sec to 5 min had an apparent effect on
labeling intensity, although increased photolysis time can lead to
increased background (data not shown).
These results, together
with the observation that the irradiation of azido auxin for 15 sec
at 300 nm nearly completely removed the azido auxin peak
identified by HPLC, suggested that light was not a critical factor
under the conditions used here.
Detection of radiolabel. Compared with 32P or even 14C ,
tritium poses special detection problems.
Impregnation of gels
with scintillation agents is necessary in order to obtain any
fluorographic signal at all, even when gels contain large amounts of
tritium.
X-ray film suffers from failure of photographic reciprocity
at low light intensities, necessitating that correct procedures be
followed for detection of small amounts of tritium.
Otherwise,
incorrect or seriously misleading results may be obtained (Laskey
and Mills, 1975). When X-ray film is exposed to a brief flash of
light before fluorography, it becomes much more sensitive, and its
response is linear even with very low light levels.
Without
preflash hypersensitization, low levels of isotope may remain
undetectable (Laskey and Mills, 1975).
Because many azido-IAA
labeling experiments yielded labeling that was at the lower level of
127
detectability, and because use of preflashed film was not the norm,
experiments were conducted to verify, under conditions in our
laboratory, the utility of preflashed film. Preflashing did permit
detection of otherwise undetectable levels of labeling. In the
course of these experiments it was also observed that it was
particularly important to use fresh scintillation agent, but neither
this nor failure to preflash film can account for labeling
inconsistencies seen in adjacent lanes of a single experiment.
Fluorography is very time consuming, so an attempt was
made to find a more rapid approach to quantitating the amount of
tritium in a given region of a gel. The dissolvable crosslinker
DHEBA was incorporated into a polyacrylamide gel. After drying,
gel slices were dissolved in periodic acid, and the resulting
solutions mixed with scintillant and analyzed in a scintillation
counter.
This led to extremely high background counts of up to
11,000 cpm.
An alternate approach, involving counting strips of
nitrocellulose after the blotting of proteins from polyacrylamide
gels (Sunahara et al., 1990), may have some merit in this regard,
but was not tested.
Preparation of material for labeling.
Several factors
associated with the source of tissue or the preparation of
membranes for labeling were identified as having potentially
important effects on labeling efficiency and were tested.
Dithiothreitol (DTT), which is used in the membrane extraction
buffer, but not the membrane resuspension buffer, is capable of
reducing azides to amines at physiological pH and temperature
128
(Staros et al., 1978).
An experiment was performed to examine the
possible effect on labeling. Labeling was inhibited by DTT (100
gM) added to the labeling buffer before freezing and photolysis.
However, no effect was seen when membranes isolated in the
presence and absence of DTT were compared.
Another factor that was examined involved the observation
that gross microbial contamination of seedling roots occurred from
time to time, in spite of surface sterilization of seeds and growth
boxes.
Because contamination, like labeling, was variable, an
attempt was made to characterize the contamination and to
correlate it with labeling efficiency. No correlation was found
between labeling efficiency and root contamination, although
considerable pathology was associated with contaminated roots,
and at least four microorganisms were associated with such roots.
Tomato genotype was another factor found to have little
predictable effect on labeling.
One genotype, dgt, has been
reported to be deficient in either the amount of auxin binding
protein or the binding affinity of protein from hypocotyls (Hicks et
al., 1989b). In three of five comparable experiments performed as
part of this study, proteins from dgt were found to label at least as
well as VFN (Figures 14 and 15). In two experiments VFN clearly
labeled better than dgt. Taken together with the results of a large
number (>15) of similar results from experiments carried out by
others (data not shown), these results are consistent with the
possibility that there may be no difference, under the conditions
used here, between dgt and VFN in terms of the 40-42 kD auxin
129
binding proteins.
The results are also indicative of variability in
azido-IAA labeling.
Although every effort has been made to reproduce the
conditions used in earlier experiments, there are other possible
explanations for this inability to repeat the results of Hicks et al.
(1989b), some of which can be ruled out by the results presented
here, and some of which remain untested. All experiments, from
1988 through the experiments discussed here, were performed
using azido-IAA from the same lot.
The experiments reported here
mostly utilized azido-IAA which had been purified by preparative
HPLC.
Recently a fresh-synthesized lot of azido-IAA was received
and it has been tested (Lomax and Hopkins, unpublished results).
Qualitatively similar labeling has been seen using azido-IAA from
all three sources. This would appear to rule out variability as a
result of differences in reagent quality.
Another factor which could possibly influence labeling, either
by affecting reaction condition (see below) or by affecting amounts
or binding characteristics of auxin binding proteins, is greening of
tissue during harvest and before homogenization.
Because of the
smaller size of dgt seedlings, it often took longer to harvest them
than to harvest VFN. During greening the ABP's could be subject to
turnover, or secondary products which interfere with labeling
could be formed.
Although labeling results did not seem to
correlate with, for example, the number of plants harvested, this
variable has not been systematically tested.
130
Other possible variables have already been tested; others will
be more difficult to test. A variety of different growth conditions
have been utilized, but no consistent effect has been found.
A
different lot of seeds was used to grow the material for the original
experiments, but that lot has been exhausted. Various lots of seeds
have been tested over the course of two years, and again, no
consistent effect has been found.
Nevertheless, the possibility
remains that some subtle variable in growth conditions, membrane
isolation, or labeling protocol is as yet uncontrolled.
Reaction chemistry.
Because an extensive search failed to
identify variables responsible for unpredictable labeling, aspects of
the chemistry of the azido-IAA labeling reaction were considered.
One reaction of the nitrene that is formed on photolysis of an
azido group involves insertion into -C-H or -0-H bonds of amino
acids;
this is the primary reaction responsible for radioactive
tagging of target proteins, and it is a relatively high-energy
reaction.
A variety of side reactions can also occur, competing with
intermolecular insertion into amino acids.
Possible side reactions
include group migration, hydrogen abstraction, intramolecular
insertion into single or double bonds, coupling between nitrenes,
and polymerization (Horspool, 1976).
Which reactions prevail is
determined in part by the balance between nitrenes in each of two
possible states, the singlet state (electron spins opposite) or the
lower-energy triplet state (electron spins parallel).
Direct
irradiation leads to the formation of singlet nitrenes, while
photosensitized decomposition results in triplet states.
131
Photosensitized decomposition occurs following irradiation in the
presence of aromatic compounds, such as benzene and napthalene.
Triplet nitrenes can also be formed from singlet nitrenes by
intersystem crossing (Coxon and Halton, 1974).
Singlet nitrenes
generally participate in insertion (leading to crosslinking), while
triplet nitrenes more often participate in abstraction (Gilbert and
Baggott, 1991).
Thus, the spin state of nitrenes is very important
in determining the outcome of a labeling experiment.
This does not
appear to have been discussed in the azido auxin literature
previously (cf. Jones et al., 1984a), although it has been mentioned
in the general crosslinking literature (Bayley and Knowles, 1977).
If spin state is important, it follows that the presence during
labeling of compounds which may cause photosensitized
decomposition is of concern. Deciding which compounds are
potential sensitizers is complicated by the fact that each such
compound has a unique triplet energy; only compounds with a
triplet donor energy greater than the triplet energy of a given
nitrene will cause significant decomposition (Coyle, 1986).
In any
event, it is worth noting that inclusion of various hormones or
analogues in the labeling mixture at concentrations several orders
of magnitude higher than the azido-hormone, as potential
competitors with plant azido-hormone derivatives, is a common
method for determining whether azido-hormone binding to a
protein is specific (e.g. Jones et al., 1984b; Hicks et al., 1989a;
Macdonald et al., 1991; Feldwisch et al., 1992; Zettl et al., 1992).
Since most auxin analogues are potential triplet sensitizers
132
(because of their aromaticity) it seems possible that in some cases
excessive side reactions by triplet-state nitrenes, rather than
competition, may contribute to decreased labeling. Plant secondary
products or pigments, such as would be present in leaves, could
also conceivably have such an effect. Azido-IAA labeling has been
unsuccessfully attempted with material from leaves (Hicks, 1992).
In spite of the obvious importance of the balance between
singlet and triplet states to azido labeling in general, it is difficult
to imagine how it could affect a series of experiments such as those
shown in Figures 14 and 15. Variable labeling was seen under
labeling conditions which had been made as similar as possible.
The only difference between these experiments was the source of
plant material, which was either from different frozen lots or was
grown fresh for each experiment.
In conclusion, an extensive analysis of reaction components
and conditions has failed to identify a satisfactory explanation for
variability in azido auxin labeling.
Thus it appears that, at least
under conditions currently in use in this laboratory, azido-IAA
labeling is an inherently unpredictable and uncontrollable
technique.
Azido-IAA labeling has been used to identify auxin
binding proteins, probe differences between genotypes, and tag
material for purification.
With the caveats discussed above, and
with appropriate controls, two of these three uses are probably
valid. However, use of azido-IAA labeling to identify putative
genotypic differences does not seem appropriate without further
optimization of the labeling protocol.
133
ACICNOWLEDGFIVIENI'S
David Ray le, Rosie Hopkins, and Peggy Rice assisted with
azido-IAA labeling.
All members of the Lomax lab provided
helpful suggestions and discussions.
Alan Jones generously
provided azido-IAA and helpful suggestions.
134
CONCLUSION
Since the first publication concerning protein phosphorylation
in plants, analogies with well-understood animal signal
transduction systems have been used to guide plant signal
transduction research.
Such analogies are useful in suggesting
outlines of potential signal transductions systems, informing
researchers of further possibilities, and shaping experiments.
One
striking aspect of animal signal transduction systems is the
diversity in the system components. There are literally hundreds
of different protein kinases, at least 16 Ga subunits, more than a
hundred G-protein-linked receptors. The most widespread protein
kinases, PKA and PKC, comprise a total of at least 13 different
subspecies.
All this diversity is part of a system of interlocking
signal transduction chains, more recently called "webs," which link
phosphorylation-dephosphorylation-based control of metabolism
and gene function with signals from outside the cell.
The diversity
presumably allows interactions including potentiation, cooperation,
synergism, and antagonism between many different pathways.
The overall impression one obtains on surveying even a fraction of
the animal signal transduction literature is that of a robust,
complex, self-regulating, interlocking entirety.
There is every reason to expect that a picture with similar
attributes will emerge once plant signal transduction is more well
understood.
However, the current picture in plants is far simpler.
135
The major components of signal transduction have been identified
in plants, particularly with the aid of molecular techniques, but
their roles and often even their subcellular location are almost
completely unknown.
The plasma membrane, which in animals is
the most important signal transduction site in the cell, in plants is
known to be the location of only a handful of signal transduction
elements.
This belies the importance of the plant plasma
membrane in responding to diverse phenomena which include
pathogens, environmental conditions, nutrition status, disposition
of photosynthate, and signals of growth regulators.
Results from experiments using molecular approaches
suggest that there are genes for multiple subspecies of protein
kinases. There is also evidence for the expression of multiple
mRNAs with homologies to one another but which encode messages
of different sizes, although a number of these sequences appear to
be untranslated or to encode inactive enzymes.
In the work presented here I have tried to begin to
understand, using biochemical techniques, the degree to which
protein kinases may be involved in the function of the plasma
There have been scattered reports of single, or
occasionally two, protein kinases in the plasma membrane of a few
membrane.
different organisms.
This work has identified numerous zucchini
The kinases appear to fall into
three general groups A, B, and C, based on their peptide maps.
plasma membrane protein kinases.
Groups A and B appear to contain several members.
An
implication of this observation, by analogy with animals, is that
136
even in simple, unchallenged tissue, several different functions
may be affected by protein kinases. Details of the possible
functions remain to be elucidated, but this should be facilitated by
the availability of partially purified kinase.
It should be possible to examine the interactions that may
exist between the PM protein kinases, including ones yet to be
described, and other potential members of signal transduction
networks.
ABP's are one type of promising candidate for such
interactions.
The nature of the interplay between kinases and
ABP's, and strategies for elucidating the interactions, will depend
on their respective functions.
If the ABP's have a transport
function, experiments to test the effect of phosphorylation on
transport characteristics such as hormone binding would be in
If the ABP's represent receptors, it would be interesting to
examine the possibility that phosphorylation could be involved in
order.
modulating the responsiveness of the receptor(s) to auxin, as well
as the possible activation of kinases by receptors which have
bound auxin.
For all of these experiments it would be convenient
to use azido-IAA, as a probe to identify ABP's, as an irreversible
activator of receptor function, or in binding studies.
The work on
ABP's described in this thesis suggests such experiments would be
premature.
137
Bibliography
Alderson A, Sabel li PA, Dickinson JR, Cole D, Richardson M, Kreis
M, Shewry PR, Halford N (1991) Complementation of snf 1, a
mutation affecting global regulation of carbon metabolism in
yeast, by a plant protein kinase cDNA. Proc. Natl. Acad. Sci.
USA 88:8602-8605
Amrhein N (1977) The current status of cyclic AMP in higher
plants.
Ann. Rev. Plant Physiol.
28:123-132
Anai T, Hasegawa K, Watanabe Y, Uchimiya H, Ishizaki R, Matsui M
(1991)
Isolation and analysis of cDNAs encoding small GTP-
binding proteins of Arabidopsis thaliana.
Anderberg RJ, Walker-Simmons MK (1992)
Gene 108:259-264
Isolation of a wheat
cDNA clone for an abscisic acid-inducible transcript with
homology to protein kinases. Proc. Nat. Acad. Sci. USA
89:10183 -10187
Andre B, Scherer GF (1991)
Stimulation by suxin of phospholipase
A in membrane vesicles from an auxin-sensitive tissue is
mediated by an suxin receptor.
Planta
185:209-214
138
Asaoka Y, Nakamura S, Yoshida K, Nishizuka Y (1992) Protein
kinase C, calcium and phospholipid degradation.
Trends
Biochem. Sci. 17:414-417
Bajar A, Podila GK, Kolattukudy PE (1991) Identification of a
fungal cutinase promoter that is inducible by a plant signal
via a phosphorylation trans-acting factor.
Sci. USA
Proc. Natl. Acad.
88:8208-8212
Bakrim N, Echevarria C, Cretin C, Arrio-Dupont, M, Pierre JN, Vidal
J, Chollet R, Gadal P (1992) Regulatory phosphorylation of
Sorghum leaf phosphoenolpyruvate carboxylase
Biochem.
Eur. J.
204:821-830
Bayley H, Knowles JR (1977) Photoaffinity labeling.
Meth.
Enzymo1.46:69-114
Berridge MJ (1993)
Nature
Inositol trisphosphate and calcium signalling.
361:315-325
Bidwai AP, Takemoto JY (1987)
Bacterial phytotoxin,
syringomycin, induces a protein kinase-mediated
phosphorylation of red beet plasma membrane polypeptides.
Proc Natl Acad Sci USA 84:6755-6759
139
Biermann B, Johnson EM, Feldman LJ (1990)
Characterization and
distribution of a maize cDNA encoding a peptide similar to
the catalytic region of second messenger dependent protein
kinases.
Plant Physiol.
94:1609-1615
Blatt MR, Thiel G (1993) Hormonal control of ion channel gating.
Annu. Rev. Plant Physiol. Plant Mol. Biol.
Blatt MR, Thiel G, Trentham DR (1990)
44:543-567
Reversible inactivation of
K+ channels of Vicia stomata! guard cells following the
photolysis of caged inositol 1,4,5-trisphosphate.
Nature
346:766-769
Blowers DP, Hetherington A, Trewavas A (1985) Isolation of
plasma-membrane-bound
calcium/calmodulin-regulated
protein kinase from pea using Western blotting.
Planta 166:
208215
Blowers DP, Trewavas AJ (1989a)
Second messengers:
existence and relationship to protein kinases.
Morre DJ (eds)
Development.
In:
their
Boss WF,
Second Messengers in Plant Growth and
Alan R. Liss, Inc. New York, pp 1-28
Blowers DP, Trewavas AJ (1989b) Rapid cycling of
autophosphorylation of a Ca2+-calmodulin regulated plasma
140
membrane located protein kinase from pea.
Plant Physiol.
90:1 27 9-1 285
Blum W, Hinsch K-D, Schultz G, Weiler EW (1988) Identification of
GTP-binding proteins in the plasma membrane of higher
plants.
Biochem. Biophys. Res. Comm.
156:954-959
Bolwell GP (1992) A role for phosphorylation in the down-
regulation of phenylalanine ammonia-lyase in suspensioncultured cells of french bean. Phytochemistry
31:4081-
4086
Boss WF (1989)
Phosphoinositide metabolism:
its relation to
signal transduction in plants. In: Boss WF, Morre DJ (eds)
Second Messengers in Plant Growth and Development. Alan
R. Liss, Inc. New York, pp 29-56
Brown EG, Newton RP (1981) Cyclic AMP and higher plants.
Phytochemistry
20:2453-2463
Browner MF, Fletterick RJ (1992) Phosphorylase:
transducer.
Trends Biochem. Sci.
a biological
17:66-71
Calvert R, Gratzer WB (1978) Proteolytic digestion patterns of
spectrin subunits.
FEBS Lett. 86:247-249
141
Campos N, Feldwisch J, Zettl R, Boland W, Schell J, Palme K (1991)
Identification of auxin-binding proteins using an improved
assay for photoaffinity labeling with 5-N3-[7-31-1]-ind ole-3acetic acid.
Technique
3:69-75
Celenza JL, Carlson M (1986) A yeast gene that is essential for
release from glucose repression encodes a protein kinase.
Science
233: 1175-1180
Chang C, Schaller GE, Patterson SE, Kwok SF, Myerowitz EM, Bleeker
AB (1992)
The TMK1 gene from Arabidopsis codes for a
protein with structural and biochemical characteristics of a
receptor protein kinase.
Cleveland DW (1983)
Plant Cell
4:1263-1271
Peptide mapping in one dimension by
limited proteolysis of sodium dodecyl sulfate-solubilized
proteins.
Cohen P (1989)
Meth. Enzymol.
96:222-229
The structure and regulation of protein
phosphatases.
Ann. Rev. Biochem. 58:453-508
Colbran RJ, Soder ling TR (1990)
Calcium/calmodulin-dependent
protein kinase II. Curr Top Cell Regul 31:181-221
142
Collins S, Caron MG, Lefkowitz RJ (1992) From ligand binding to
gene expression:
new insights into the regulation of G-
protein-coupled receptors.
Trends Biochem. Sci.
17:37-39.
Cote GG, Crain RC (1993) Biochemistry of phosphoinositides. Ann.
Rev. Plant Physiol. Plant Mol. Biol. 44:333-356
Cote GG, Depass AL, Quarmby LM, Tate BF, Morse MJ, Satter RL,
Crain RC (1989)
Separation and characterization of inositol
phospholipids from the pulvini of Samanea saman. Plant
Physiol.
90:1422-1428
Coxon JM, Halton B (1974) "Organic Photochemistry." Cambridge
University Press, London.
Coyle JD (1986)
"Introduction to Organic Photochemistry." Wiley,
NY. pp 17-20
Dostmand WR, Taylor SS, Genieser HG, Jastorff B, Doskeland SO,
Ogreid D (1990)
Probing the cyclic nucleotide binding sites
of cAMP-dependent protein kinases I and II with analogs of
adenosine 3', 5'-cyclic phosphorothioates. J Biol Chem 265:
1 0 4 8 4-1 0 4 9 1
DrObak BK (1992) The plant phosphoinositide system.
288:697-712
Biochem. J.
143
Duerr B, Gawienowski M, Ropp T, Jacobs T (1993) MsERK1: a
mitogen-activated protein kinase from a flowering plant.
Plant Cell 5:87-96
Ebert PR, Anderson MA, Bernatzky R, Altschuler M, Clarke AE
(1989)
Genetic polymorphism of self-incompatibility in
flowering plants.
Cell 56:256-262
Ehrlich KC, Cary JW, Ehrlich M (1992) A broad bean cDNA clone
encoding a DNA-binding protein resembling mammalian
CREB in its sequence specificity and DNA methylation
sensitivity.
Gene 117:169-178
Einspahr KJ, Peeler TC, Thompson GA Jr (1988) Rapid changes in
polyphosphoinositide metabolism associated with the
response of Dunaliella salina to hypoosmotic shock. J. Biol.
Chem.
263:5775-5779
Einspahr KJ, Thompson Jr. GA (1990)
Transmembrane signaling
via phosphatidylinositol 4,5-bisphosphate hydrolysis in
plants. Plant Physiol. 93:361-366
Elliott DC, Kokke YS (1987) Cross-reaction of a plant protein kinase
with antiserum raised against a sequence from bovine brain
144
protein kinase C regulatory subunit.
Biochem. Biophys. Res.
Commun. 145:1043-1047
Erdmann H, Bucher M, Wagner KG (1982) Two protein kinases
from nuclei of cultured tobacco cells with properties similar
to the cyclic nucleotide-independent enymes (NI and NII)
from animal tissue. FEBS Letters 137:245-248
Ett linger C, Leh le L (1988) Auxin induces rapid changes in
phosphatidylinositol metabolites.
Nature 331:176-178
Evans DE, Briars SA, Williams LE (1991) Active calcium transport
by plant cell membranes.
J. Exp. Botany 42:285-303
Fairley-Grenot K, Assmann SM (1991)
Evidence for G-protein
regulation of inward K+ channel current in guard cells of
Fava bean.
Plant Cell 3:1037-1044
Farmer EE, Moloshok TD, Saxton MJ, Ryan CA (1991)
Oligosaccharide signalling in plants:
specificity of
oligouronide-enhanced plasma membrane protein
phosphorylation.
J. Biol. Chem.
266:3140-3145
Farmer EE, Pearce G, Ryan CA (1989) In vitro phosphorylation of
plant plasma membrane proteins in response to the
145
proteinase inhibitor inducing factor. Proc Natl Acad Sci USA
8 6:153 9-1 5 4 2
Feiler HS, Jacobs TW (1991) Cloning of the pea cdc2 homologue by
efficient immunological screening of PCR products.
Plant
Mol. Biol. 17:321-333
Feldwisch J, Zettl R, Hesse F, Schell J, Palme K (1992) An auxin-
bining protein is localized to the plasma membrane of maize
coleoptile cells:
identification by photoaffinity labeling and
purification of a 23-kDa polypeptide. Proc. Natl. Acad. Sci.
USA
89:475-479
Felix G, Grosskopf DG, Regenass M, Boller T (1991) Rapid changes
in protein phosphorylation are involved in transduction of
the elicitor signals in plant cells. Proc Natl Acad Sci USA
88:8831-8834
Feller K (1989) Characterization of a protein kinase from soybean
seedlings. Study of the variation of calcium-regulated kinase
activity during infection with the incompatible race 1 and
the compatible race 3 of Phytophtora megasperma F. Sp.
Glycinea by in vitro phosphorylation of calf thymus histone
Hl. Plant Sci. 60:67-75
146
Feng X-H, Bottino PJ, Kung S-d (1992) Molecular identification of a
soybean protein kinase gene family by using PCR. Plant Mol.
Biol. 18:581-584
Fischer EH, Krebs EG (1989) Commentary on 'The phosphorylase b
to a converting enzyme of rabbit skeletal muscle' by E.G.
Krebs and E.H. Fischer Biochem. Biophys. Acta 20 (1956)
150-157.
Biochem. Biophys. Acta 1000:297-301
Fleet GWJ, Porter RR, Knowles JR (1969) Affinity labelling of
antibodies with aryl nitrene as reactive group.
Nature
224:511-512
Francis SH, Woodford TA, Wolfe L, Corbin JD (1988-89) Types 1-
alpha and 1 beta isozymes of cGMP-dependent protein
kinase:
alternative mRNA splicing may produce different
inhibitory domains.
Second Mess Phosphoprot 12(5-6):301-
310
Gangwani L, Tamot BK, Khurana JP, Maheshwari SC (1991)
Identification of 3',5'-cyclic AMP in axenic cultures of Lemna
paucicostata th high-performance liquid chromatography.
Biochem. Biophys. Res. Commun.
178:1113-1119
147
Geahlen RL, Anostario Jr M, Low PS, Harrison ML (1986) Detection
of protein kinase activity in sodium dodecyl sulfatepolyacrylamide gels.
Anal Bioch
153: 151-158
Gilbert A, Baggott J (1991) "Essentials of Molecular
Photochemistry." CRC Press, Boca Raton. pp 450-456
Gilroy S, Blowers DP, Trewavas AJ (1987) Calcium: a regulation
system emerges in plant cells.
Development 100:181-184
Gilroy S, Fricker MD, Read ND, Trewavas AJ (1991) Role of calcium
in signal transduction of Commelina guard cells. Plant Cell
3:333-344
Gilroy S, Jones RL (1992) Gibberellic acid and abscisic acid
coordinately regulate cytoplasmic calcium and secretory
activity in barley aleurone protoplasts.
USA
Proc. Natl. Acad. Sci.
89:3591-3595
Gilroy S, Read ND, Trewavas AJ (1990) Elevation of cytoplasmic
calcium by caged calcium or caged inositol trisphosphate
initiates stomatal closure.
Nature 346: 769-771
Ginty DD, Fanger GR, Wagner JA, Maue RA (1992) The activity of
cAMP-dependent protein kinase is required at a
posttranslational level for induction of voltage-dependent
148
sodium channels by peptide growth factors in PC12 cells. J
Cell Biol 116: 1465-73
Gonzalez GA, Yamamoto KK, Fischer WH, Karr D, Menzel P, Biggs III
W, Vale WW, Montminy MR (1989) A cluster of
phosphorylation sites on the cyclic AMP-regulated nuclear
factor CREB predicted by its sequence.
Gorbunoff MJ (1985)
columns.
Nature 337:749-752
Protein chromatography on hydroxyapatite
Meth. Enzymol. 117:370-380
Goring DR, Glavin TL, Schafer U, Rothstein SJ (1993) An S receptor
kinase gene in self-compatible Brassica napus has a 1-bp
deletion.
Plant Cell
5:531-539
Goring DR, Rothstein SJ (1992) The S-locus receptor kinase gene in
a self-incompatible Brassica napus line encodes a functional
serine/threonine kinase.
Plant Cell 4:1273-1281
Gowda S, Pillay DTN (1982) Cyclic AMP independent protein
kinases from soybean cotyledons (Glycine max L.). Pl. Sci.
Lett. 25:49-59
Hall A (1990) The cellular functions of small GTP-binding
proteins.
Science 249:635-640
149
Hall A (1992)
two GAPs.
Signal transduction through small GTPases--a tale of
Cell 69:389-391
Hanks SK, Quinn AM, Hunter T (1988) The protein kinase family:
conserved features and deduced phylogeny of the catalytic
domains.
Science 241:42-52
Harmon AC, Putnam-Evans C, Cormier MJ (1987) A calcium-
dependent but calmodulin-independent protein kinase from
soybean.
Plant Physiol. 83:830-837
Harper JF, Binder BM, Sussman MR (1993) Calcium and lipid
regulation of an Arabidopsis protein kinase expressed in
Escherichia coli.
Biochemistry
32:3282-3290
Harper JF, Sussman MR, Schaller GE, Putnam-Evans C, Charbonneau
H, Harmon AC (1991)
A calcium-dependent protein kinase
with a regulatory domain similar to calcium.
Science
252:951-954
Hasunuma K, Furukawa K, Tomita K, Mukai C, Nakamura T (1987)
GTP-binding proteins in etiolated epicotyls of Pisum sativum
(Alaska) seedlings.
1 48:13 3-13 9
Biochem. Biophys. Res. Commun.
150
Hayashida N, Mizoguchi T, Yamaguchi-Shinozaki K, Shinozaki K
(1992)
Characterization of a gene that encodes a homologue
of protein kinase in Arabidopsis thaliana.
Gene 121:325-330
Hepler JR, Gilman AG (1992) G proteins. Trends Biochem. Sci.
17:383-387
Hepler PK, Wayne RO (1985)
Calcium and plant development.
Annu. Rev. Pl. Physiol. 36:397-439
Hetherington A, Trewavas A (1982)
kinase in pea shoot membranes.
Hetherington AM, Trewavas A (1984)
Calcium-dependent protein
FEBS Lett. 145:67-71
Activation of pea
membrane protein kinase by calcium ions.
Planta 161:409-
417
Hicks GR, Ray le DL, Jones AM, Lomax TL (1989a) Specific
photoaffinity labeling of two plasma membrane polypeptides
with an azido auxin. Proc. Natl. Acad. Sci. USA 86:4948-4952
Hicks GR, Ray le DL, Lomax TL (1989b) The diageotropica mutant
of tomato lacks high specific activity auxin binding site.
Science 245:52-54
151
Hicks GR (1992)
"Molecular Identification and Characterization of
Two Plasma Membrane Associated Auxin-Binding Proteins."
Ph.D Thesis.
Oregon State University.
Horspool WM (1976)
"Aspects of Organic Photochemistry."
Academic Press, NY. pp 267-270
Huber SC, Huber JL (1990) In vitro phosphorylation and
inactivation of spinach leaf sucrose-phosphate synthase by
an endogenous protein kinase. Biochem Biophys Acta 1091:
3 93 -4 0 0
Huber SC, Huber JL (1991)
Regulation of maize leaf sucrose-
phosphate synthase by protein phosphorylation.
Physiol.
Plant Cell
32:319-326
Hunter T (1991) Protein kinase classification.
Meth Enzymol
200:3-37
Hunter T, Karin M (1992) The regulation of transcription by
phosphorylation.
Cell
70:375-387
Irvine RF, Letcher AJ, Lander DJ, DrObak BK, Dawson AP, Musgrave
A (1989)
Phosphatidylinositol(4,5)bisphosphate and
phosphatidylinositol(4)-phosphate in plant tissues.
Physiol.
89:888-892
Plant
152
Irving HR, Gehring CA, Parish RW (1992) Changes in cytosolic pH
and calcium of guard cells precede stomatal mevements.
Proc. Natl. Acad. Sci. 89:1790-1794
Jacobs M, The len MP, Farndale RW, Astle MC, Rubery PH (1988)
Specific guanine nucleotide binding by membranes from
Cucurbita pepo seedlings. Biochem. Biophys. Res. Comm.
15 5:1 47 8-1485
Janistyn B, Ebermann R (1992) Biological activity of Zea mays
isolated adenosine and guanosine 3',5'-cyclic monophosphate
by protein phosphorylation of membrane fractions from
Drosophila brain. Phytochemistry 31:1879-1881
Janssens PMW (1988)
The evolutionary origin of eukaryotic
transmembrane signal transduction.
Comp. Biochem. Physiol.
90A:209-223
Johannes E, Brosnan JM, Sanders D (1991)
signal transduction in plant cells.
Calcium channels and
BioEssays 13:331-336
Jones AM, Melhado LL, Ho T-HD, Leonard NJ (1984a) Azido auxins:
quantitative binding data in maize.
301
Plant Physiol. 74:295-
153
Jones AM, Melhado LL, Ho T-HD, Pearce CJ, Leonard NJ (1984b)
Azido auxins:
photoaffinity labeling of auxin-binding
proteins in maize coleoptile with tritiated 5-azidoindole-3acetic acid.
Plant Physiol.
75:1111-1116
Jones AM, Venis MA (1989) Photoaffinity labeling of indole-3acetic acid-binding proteins in maize.
Proc. Natl. Acad. Sci.
USA 86:6153-6156
Jones RL, Jacobsen JV (1991)
Regulation of the sysnthesis and
transport of secreted proteins in cereal aleurone.
Cytol.
Intl. Rev.
126:49-88
Karin M, Smeal T (1992) Control of transcription factors by signal
transduction pathways:
Biochem. Sci.
the beginning of the end.
Trends
17:418-422
Katagiri F, Lam E, Chua N-H (1989) Two tobacco DNA-binding
proteins with homology to the nuclear factor CREB. Nature
340:727-730
Keates RAB (1973) Cyclic nucleotide-independent protein kinase
from pea shoots. Biochem. Biophys. Res. Comm. 54:655-661
Kemp BE (1979)
Phosphorylation of synthetic peptide analogues
of rabbit cardiac troponin inhibitory subunit by the cyclic
154
AMP-dependent protein kinase.
J. Biol. Chem.
254:2638-
2642
Kemp BE, Pearson RB (1990) Protein kinase recognition sequence
motifs.
Trends Biochem. Sci.
15:342-346
Kikkawa U, Nishizuka Y (1986) Protein kinase C. In: The
Enzymes, Vol. XVII. 3rd. Ed. Boyer PD and Krebs EG (eds).
Academic Press. New York, pp 167-189
Klimczak LJ, Hind G (1990) Biochemical similarities between
soluble and membrane-bound calcium-dependent protein
kinases of barley. Plant Physiol. 92:919-923
Klucis E, Polya GM (1988) Localization, solubilization and
characterization of plant membrane-associated calciumdependent protein kinases.
Plant Physiol 88: 164-171
Knighton DR, Zheng J, Ten Eyck LF, Ashford VA, Xuong N-H, Taylor
SS, Sowadski JM (1991)
Crystal structure of the catalytic
subunit of cAMP-dependent protein kinase.
Science
253:407-414
Koontz DA, Choi JH (1993) Protein phosphorylation in carrot
somatic embryos in response to abscisic acid.
Biochem.
31:95-102
Plant Physiol.
155
Krebs EG (1986)
The enzymology of control by phosphorylation.
In: P. Boyer, EG Krebs, eds, The Enzymes, Volume XVII.
Academic Press, NY. pp 3-20
Krebs EG, Fischer EH (1956) The phosphorylase b to a converting
enzyme of rabbit skeletal muscle.
Biochem. Biophys. Acta
2 0:1 5 0-15 7
Kruse T, Tallman G, Zeiger E (1989) Isolation of guard cell
protoplasts from mechanically prepared epidermis of Vicia
faba leaves. Plant Physiol. 40:1382-1386
Kulcis E, Polya GM (1988) Localization, solubilization and
characterization of plant membrane-associated calciumdependent protein kinases.
Plant Physiol. 88:164-171
Ladror US, Zielinski RE (1989) Protein kinase activities in
tonoplast and plasmalemma membranes from corn roots.
Plant Physiol. 89:151-158
Laskey RA and Mills AD (1975) Quantitative film detection of 3H
and 14C in polyacrylamide gels by fluorography.
Biochem.
56:335-341
Eur. J.
156
Lawton MA, Yamamoto RT, Hanks SK, Lamb CJ (1989) Molecular
cloning of plant transcripts encoding protein kinase
homologs.
Proc. Natl. Acad. Sci. USA 86:3140-3144
Lee JL, Tucker EB, Crain RC, Lee Y (1993) Stomatal opening is
induced in epidermal peels of Commelina communis L. by
GTP analogs or pertussis toxin.
Plant Physiol.
102:95-100
Legendre L, Peinstein PF, Low PS (1992) Evidence for
participation of GTP-binding proteins in elicitation of the
rapid oxidative burst in cultured soybean cells.
J. Biol. Chem.
267:20140-20147
Lin X, Feng X-H, Watson JC (1991) Differential accumualtion of
transcripts encoding protein kinase homologue in greening
pea seedlings.
Proc. Natl. Acad. Sci. USA 88:6951-6955
Lomax TL, Hicks GR (1991) Specific auxin binding proteins in the
plasma membrane:
receptors or transporters?
Biochem. Soc.
Trans. 20:64-69
Lucantoni A, Polya GM (1987) Activation of wheat embryo
calcium-regulated protein kinase by unsaturated fatty acids
in the presence and absence of calcium. FEBS Lett. 221:3336
157
Ma H, Yanofsky MF, Hauang H (1991) Isolation and sequence
analysis of TGA1 cDNAs encoding a tomato G protein a
subunit.
Gene 107:189-195
Ma H, Yanofsky MF, Myerowitz EM (1990) Molecular cloning and
characterization of GPA1, a G protein a subunit gene from
Arabidopsis thaliana. Proc. Natl. Acad. Sci. 87:3821-3825
MacDonald H, Jones AM, King PJ (1991) Photoaffinity labeling of
soluble auxin-binding proteins.
J. Biol. Chem. 266:7393-7399
MacKintosh C, Coggins J, Cohen P (1991) Plant protein
phosphatases.
Biochem J. 272:733-738
MacKintosh RW, Davies SP, Clarke PR, Weekes J, Gillespie JG, Gibb
BJ, Hardie DG (1992) Evidence for a protein kinase cascade
in higher plants:
reductase kinase.
3-hydroxy-3-methylglutaryl-CoA
Eur. J. Biochem.
209:923-931
Martell AE, Smith RM (1974) Critical stability constants. Vol. 1.
Plenum Press, New York
Martinoia E, Locher R, Vogt E (1993) Inositol trisphosphate
metabolism in subcellular fractions of barley (Hordeum
vulgare L.) mesophyll cells.
Plant Physiol. 102:101-105
158
McAinsh MR, Brownlee C, Hetherington AM (1990)
Abscisic acid-
induced elevation of guard cell cytosolic Ca2+ precedes
stomatal closure.
Nature 343:186-188
McCurdy DW, Harmon AC (1992)
Calcium-dependent protein
kinase in the green alga Chara. Planta 188:54-61
Melhado LL, Jones AM, Ho T-HD, Leonard NJ (1984) Azido auxins:
photolysis in solution and covalent binding to soybean.
Physiol.
Plant
74:289-294
Moreau RA, Preisig CL (1993) Lipid changes in tobacco cell
cuspensions following treatment with cellulase elicitor.
Phys.
Plant. 87:7-13
Morsucci R, Curvetto N, Delmastro S (1991)
Involvement of
cytokinins and adenosine 3',5'-cyclic monophosphate in
stomatal mevement in Vicia faba.
Plant Physiol. Biochem.
29:537-547
Nishimoto I, Okamoto T, Matsuura Y, Takahashi S, Okamoto T,
Murayama Y, Ogata E (1993) Alzheimer amyloid protein
precursor complexes with brain GTP-binding protein Go.
Nature 362:75-79
159
Nishizuka Y (1988)
The molecular heterogeneity of protein kinase
C and its implications for cellular regulation.
Nature
334:661-665
Nitschke K, Fleig U, Schell J, Plame K (1992) Complementation of
the cs cis2-11 cell cycle mutant of Schizosaccharomyces
pombe by a protein phosphatase from Arabidopsis thaliana.
EMBO J.
11:1327-1333
North MJ (1989) Prevention of unwanted proteolysis.
RJ
In:
Beynon, JS Bond, eds, Proteolytic Enzymes, a Practical
Approach.
IRL Press, NY, pp 105-124
Otte AP, Moon RT (1992) Protein kinase c isozymes have distinct
roles in neural induction and competence in Xenopus.
68:
Cell
1021-1029
Palme K (1992)
Molecular analysis of plant signaling elements:
relevance of eukaryotic signal transduction models.
Rev. Cytol.
Intl.
132:223-283
Palme K, Diefenthal T, Vingron M, Sander C, Schell J (1992)
Molecular cloning and structural analysis of genes from Zea
mays (L.) coding for members of the ras-related ypt gene
family.
Proc. Natl. Acad. Sci. USA 89:787-791
160
Perdue DO, Lomax TL (1992)
Characterization of GTP binding and
hydrolysis in plasma membranes of zucchini.
Biochem.
Plant Physiol.
30:163-172
Perrin DD, Sayce IG (1967)
Computer calculation of equilibrium
concentrations in mixtures of metal ions and complexing
species.
Talanta 14:833-842
Philips R (1983)
Sources and Applications of Ultraviolet Radiation.
Academic Press, New York.
Polya GM, Chung R, Menting J (1991) Resolution of a higher plant
protein kinase similar to the catalytic subunit of cyclic AMPdependent protein kinase. Plant Sci. 79:37-45
Polya GM, Davies JR (1982)
Resolution of calcium--calmodulin-
activated protein kinase from wheat germ.
FEBS Lett.
15 0:1 67-17 1
Putnam-Evans C, Harmon AC, Palevitz BA, Fechheimer M, Cormier
MJ (1989)
Calcium-dependent protein kinase is localized
with F-actin in plant cells. Cell Motil. Cytoskel. 12:12-22
Putnam-Evans CL, Harmon AC, Cormier MJ (1986)
Calcium-
dependent protein phosphorylation in suspension-cultured
161
soybean cells.
In:
Molecular and Cellular Aspects of calcium
in Plant Development.
AJ Trewavas, ed. Plenum, 1986.
Putnam-Evans CL, Harmon AC, Cormier MJ (1990) Purification
and characterization of a novel calcium-dependent protein
kinase from soybean. Biochemistry 29:2488-2495
Ralph RK, McCombs PJA, Tener G, Wojcik SJ (1972) Evidence for
modification of protein phosphorylation by cytokinins.
Biochem J. 130:901-911
Ranjeva R, Boudet AM (1987) Phosphorylation of proteins in
plants:
regulatory effects and potential involvement in
stimulus/response coupling.
Ann. Rev. Plant Physiol. 38:73-
93
Rao KP, Randall DD (1980) Plant pyruvate dehydrogenase
complex:
inactivation and reactivation by phosphorylation
and dephosphorylation.
Arch Biochem Biophys
200:461-
466
Raz V, Fluhr R (1993) Ethylene signal is transduced via protein
phosphorylation.
Plant Cell
5:523-530
Refeno G, Ranjeva R, Boudet AM (1982) Modulation of quinate:
NAD oxidoreductase activity through reversible
162
phosphorylation in carrot cell suspensions.
Planta
154:193-
198
Roberts DM (1993)
Protein kinase with calmodulin-like domains:
novel targets of calcium in plants. Curr. Opin. Cell Biol.
5:242-246
0
Roberts DM, Harmon AC (1992)
Calcium-modulated proteins:
targets of intracellular calcium signals in higher plants.
Annual Rev. Plant Physiol. Plant Mol. Biol. 43:375-414
Robison GA, Butcher RW, Sutherland EW (1971) Cyclic AMP.
Academic, New York.
Rundle SJ, Nasrallah JB (1992)
Molecular characterization of type
serine/threonine phosphatase from Brassica oleracea.
Plant Mol. Biol. 20:367-375
1
Ruth P, Landgraf W, Keilbach A, May B, Egleme C, Hoffmann F
(1991)
The activation of expressed cGMP dependent protein
kinase isozymes I alpha and I beta is determined by the
different amino-termini.
Salimath BP, Marine D (1983)
Eur J Biochem 202:1339-1344
Protein phosphorylation and its
regulation by calcium and calmodulin in membrane fractions
from zucchini hypocotyls. Planta 158:560-568
163
Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning,
Second Ed.
Cold Spring Harbor Laboratory Press.
pp. 18.69-
1 8 .7 5
Sandelius AS, Sommarin M (1986)
Phosphorylation of
phosphatidylinositols in isolated plant membranes.
FEBS
Lett. 201:282-286
Schaffner W, Weissmann C
(1973) A rapid, sensitive, and specific
method for the determination of protein in dilute solution.
Anal Biochem
56:502-514
Schaller GE, Harmon AC, Sussman MR (1992) Characterization of a
calcium- and lipid-dependent protein kinase associated with
the plasma membrane of oat.
Biochemistry
31:1721-1727
Schaller GE, Sussman MR (1988) Phosphorylation of the plasma-
membrane H-F-ATPase of oat roots by a calcium-stimulated
protein kinase.
Planta
173:509-518
Scherer GE, Andre B (1989) A rapid response to a plant hormone:
auxin stimulates phospholipase A2 in vivo and in vitro.
Biochem. Biophys. Res. Comm.
163:111-117
164
Scherer GFE, Martiny-Baron G, Stoffel B (1988) A new set of
regulatory molecules in plants:
a plant phospholipid similar
to platelet-activating factor stimulates protein kinase and
proton-translocating ATPase in membrane vesicles.
Planta
175 :241 -25 3
Schroeder JI, Hagiwara S (1989)
Cytosolic calcium regulates ion
channels in the plasma membrane of Vicia faba guard cells.
Nature 338:427-430
Schroeder JI, Hagiwara S (1990)
Repetitive increases in cytosolic
Ca2+ of guard cells by abscisic acid activation of nonselective
Ca2÷ permeable channels. Proc. Natl. Acad. Sci. USA 87:93059309
Schwacke R, Hager A (1992)
Fungal elicitors induce a transient
release of active oxygen species from cultured spruce cells
that is dependent on Ca2+
Planta
and protein kinase activity.
187:136-141
Scott JD (1991)
Cyclic nucleotide-dependent protein kinases.
Pharmac. Ther.
50:123-145
Short TW, Briggs WR (1990)
Characterization of a rapid, blue
light-mediated change in detectable phosphorylation of a
165
plasma membrane protein from etiolated pea (Pisum
sativum L.) seedlings. Proc. Natl. Acad. Sci. USA 92:179-185
Simon MI, Strathmann MP, Gautam N (1991) Diversity of G
proteins in signal transduction.
Science 252:802-808
Sithanandam G, Kolch W, Duh FM, Rapp UR (1990) Complete
coding sequence of a human B-raf cDNA and detection of B-
raf protein kinase with isozyme specific antibodies.
Oncogene 5: 1775-1800
Smith RD, Walker JC (1991) Isolation and expression of a maize
type
1
protein phosphatase.
Plant Physiol.
97:677-683
Sommarin M, Henriksson T, Jergil B (1981) Cyclic AMP-dependent
protein phosphorylation on the surface of rat hepatocytes.
FEBS Lett. 127:285-289
Sommarin M, Sandelius AS (1988)
Phosphatidylinositol and
phosphatidylinositol phosphate kinase in plant plasma
membranes.
Biochem. Biophys. Acta
958:268-278
Spiteri A, Viratelle OM, Raymond P, Rancillac M, Labouesse J,
Pradet A (1989)
plant tissues.
Artefactual origins of cyclic AMP in higher
Plant Physiol. 91:624-628
166
Sprang SR, Acharya KR, Goldsmith EJ, Stuart DI, Varvill K, Fletterick
RJ, Madsen NB, Johnson LN (1988)
Structural changes in
glycogen phosphorylase induced by phosphorylation.
Nature
336:215-221
Staros JV, Bayley H, Standring DN, Knowles JR (1978) Reduction of
aryl azides by thiols:
photoaffinity reagents.
implications for the use of
Biochem. Biophys. Res. Commun.
80:568-572
Stein JC, Howlett B, Boyes DC, Nasrallah ME, Nasrallah JB (1991)
Molecular cloning of putative receptor protein kinase gene
encoded at the self-incompatibility locus of Brassica oleracea.
Proc. Natl. Acad. Sci. USA 88:8816-8820
Stryer L and Bourne HR (1986) G-proteins: a family of signal
transducers.
Annu. Rev. Cell. Biol. 2:391-419
Suen K-L, Choi JH (1991) Isolation and sequence analysis of a
cDNA clone for a carrot calcium-dependent protein kinase:
homology to calcium/calmodulin-dependent protein kinases
and to calmodulin.
Plant Mol. Biol. 17:581-590
Sunahara RK, Murphy BJ, Tuana BS (1990) A sensitive and rapid
method for identification of low abundance receptors.
Biochem. 185:143-146
Anal.
167
Sutherland EW, Cori CF (1956) Effect of hyperglycemic-
glycogenolytic factor and epinephrine on liver
phosphorylase.
J. Biol. Chem.
188:531-543
Swarup G, Cohen S, Garbers DL (1981)
Selective
dephosphorylation of proteins containing phosphotyrosine
by alkaline phosphatases. J Biol Chem 256:8197-8201
Tallman G (1992)
revisited.
The chemiosmotic model of stomatal opening
Crit. Rev. Plant Sci.
11:35-57
Tan Z, Boss WF (1992) Association of phosphatidylinositol kinase,
phosphatidylinositol monophosphate kinase, and
diacylglycerol kinase with the cytoskeleton and F-actin
fractions of carrot (Daucus carota L.) cells grown in
suspension culture.
Plant Physiol. 100:2116-2120
Terada K, Kai T, Okuno S, Fujisawa H, Izui K (1990) Maize leaf
phosphoenolpyruv ate carboxylase:
phosphorylation of Ser15
with a mammalian cyclic AMP-dependent protein kinase
diminishes sensitivity to inhibition by malate.
259:241-244
FEBS Lett.
168
Tobias CM, Howlett B, Nasrallah JB (1992) An Arabidopsis thaliana
gene with sequence similarity to the S-locus receptor kinase
of Brassica oleracea. Plant Physiol. 99:284-290
Tominga Y, Wayne R, Tung HYL, Tazawa M (1987)
Phosphorylation-dephosphorylation is involved in Ca2+controlled cytoplasmic streaming of characean cells.
Protoplasma 136:161-169
Trewavas A (1973)
Lemna minor.
Trewavas A (1976)
phosphorylation.
The phosphorylation of ribosomal protein in
Plant Physiol. 51:760-765
Post-translational modification of proteins by
Annu. Rev. Pl. Physiol.
27:349-374
Trewavas A, Gilroy S (1991) Signal transduction in plant cells.
Trends Genet. 7:356-361
Ullrich A, Schlessinger J (1990)
Signal transduction by receptors
with tyrosine kinase activity.
Cell 61:203-212
Veluthambi K, Poovaiah BW (1984)
phosphorylation in plants.
Calcium-promoted protein
Science 223:167-169
169
Veluthambi K, Poovaiah BW (1986)
In vitro and in vivo protein
phosphorylation in Avena sativa L. coleoptiles.
Plant Physiol.
81:836-841
Vera P, Conejero V (1990) Citrus exocortis viroid alters the in
vitro pattern of protein phosphorylation of tomato leaf
proteins.
Mol. Plant-Microbe Interact
3:28-32
Verhey SD, Gaiser JC, Lomax TL (1993) Protein Kinases in zucchini:
characterization of calcium-requiring plasma membrane
kinases.
Plant Physiol. In Press.
Walker JC, Zhang R (1990) Relationship of a putative receptor
protein kinase from maize to the S-locus glycoproteins of
Brassica.
Nature 345:743-746
Warpeha KMF, Hamm HE, Rasnick MM, Kaufman LS (1991) A
blue-light-activated GTP-binding protein in the plasma
membranes of etiolated peas. Proc. Natl. Acad. Sci. USA
88:8925-8929
Watillon B, Kettmann R, Boxus P, Burny A (1992) Cloning and
characterization of an apple (Malus domestica L. Borkh)
cDNA encoding a calmodulin-binding protein domain similar
to the calmodulin-binding region of type II mammalian
170
Ca24-/calmodulin-dependent protein kinase.
Plant Sci.
81:227-235
Watillon B, Kettmann R, Boxus P, Burny A (1993) A
calcium/calmodulin-binding serine/threonine protein kinase
homologous to the mammalian type II calcium/calmodulindependent protein kinase is expressed in plant cells.
Plant
Physiol. 101:1381-1384
Westheimer FH (1987) Why nature chose phosphates.
Science
23 5:1173-117 8
Wolniak SM, Larsen PM (1992)
Changes in the metaphase transit
times and the pattern of sister chromatid separation in
stamen hair cells of Tradescantia after treatment with
protein phosphatase inhibitors. J. Cell Sci. 102:691-715
Xu P, Lloyd CW, Staiger CJ, DrObak BK (1992) Association of
phosphatidylinositol 4-kinase with the plant cytoskeleton.
Plant Cell 4:941-951
Yan T-FJ, Tao M (1982)
Purification and characterization of a
wheat germ protein kinase.
J. Biol. Chem. 257:7037-7043
Yotsushima K, Mitsui T, Takaoka T, Hayakawa T, Igaue I (1993)
Purification and characterization of membrane-bound
171
inositol phospholipid-specific phospholipase c from
suspension-cultured rice (Oryza sativa L.) cells. Plant
Physiol. 102:165-172
Zaina S, Mapelli S, Reggiani R, Bertani A (1991) Auxin and GTPase
activity in membranes from aeorobic and anaerobic rice
coleoptile.
J. Plant Physiol. 138:760-762
Zaina S, Reggiani R, Bertani A (1990) Preliminary evidence for
involvement of GTP-binding protein(s) in auxin signal
transduction in rice (Oryza sativa L.) coleoptile. J. Plant
Physiol. 136:653-658
Zbell B, Paulik M, Morre DJ (1990) Comparison of [35S]GTPyS
binding to plasma membranes and endomembranes
prepared from soybean and rat. Protoplasma 154:74-79
Zbell B, Walter-Back C, Hohenadel H, Schwendemann I (1989)
Polyphosphoinositide turnover and signal transduction of
auxin on isolated membranes of Daucus carota L. In: Pharis
RP and Rood SB (eds) Plant Growth Substances 1988.
Springer-Verlag. Berlin, pp 122-129
Zettl R, Feldwisch J, Boland W, Schell J, Palme K (1992) 5'-azido-
[3,6-3H2]-1-napthylphthalamic acid, a photoactivatable probe
for naphthylphthalamic acid receptor proteins from higher
172
plants:
identification of a 23-kDa protein from maize
coleoptile plasma membranes.
Proc. Natl. Acad. Sci. USA
89:480-484
Zhang K, Tsukitani Y, John PCL (1992) Mitotic arrest in tobacco
caused by the phosphoprotein phosphatase inhibitor okadaic
acid.
Pl. Cell Physiol. 33:677-688
Zocchi G (1985)
Phosphorylation/dephosphorylation of membrane
proteins controls the microsomal proton-translocating
ATPase activity of corn Zea mays roots. Plant Sci. 40:153159
173
Author Index
Alderson et al., 1991 39
Amrhein, 1977 14, 15, 30
Anai et al., 1991 19
Anderberg and WalkerSimmons, 1992 36, 42
Andre and Scherer, 1991 24,
107
Asaoka et al., 1992 11
Bajar et al., 1991 42
Bakrim et al., 1992 42
Bayley and Knowles, 1977
127, 133
Berridge, 1993 10
Bidwai and Takemoto, 1987
42, 49
Biermann et al., 1990 34, 35
Blatt and Thiel, 1993 9,
24,107
Blatt et al., 1990 9
Blowers and Trewavas, 1989
32
Blowers et al., 1985 32, 49, 69
Blum et al., 1988 22
Bolwell, 1992 18
Boss, 1989 11, 13
Brown and Newton, 1981 15
Browner and Fletterick,1992
28
Calvert and Gratzer, 1978 103,
104
Campos et al., 1991 110, 127,
128
Celenza and Carlson 1986 54,
69
Chang et al., 1992 39
Cleveland, 1983 86
Cohen, 1989 41
Colbran and Soderling, 1990
74
Collins et al., 1992 109
Cote and Crain, 1993 11
Cote et al., 1989 12
Coxon and Halton, 1974 133
Coyle, 1986 133
Dostmann et al., 1990 74
Drobak, 1992 7, 8, 11, 13
Duerr et al., 1993 42
Ebert et al., 1989 38
Ehrlich et al., 1992 16
Einspahr and Thompson, 1990
11
Einspahr et al., 1988 12
Einsphar and Thompson, 1989
12
Elliott and Kokke, 1987 40
Erdmann et al., 1982 30
Ettlinger et al., 1988 24
Evans et al., 1991 8
Fairley-Grenot and Assmann,
1991 24, 25, 27
Farmer et al., 1989 42, 49
Farmer et al., 1991 42
Feldwisch et al., 1992 109,
133
Felix et al., 1991 49
Feller, 1989 49
Feng et al., 1992 35
Fischer and Krebs, 1989 29
Fleet et al., 1969 108
Francis et al., 1988-89 74
Gangwani et al., 1991 15
Geahlen et al., 1986 69
Gilbert and Baggott, 1991 133
Gilroy and Jones, 1992 10
Gilroy et al., 1987 7
Gilroy et al., 1990 9
Gilroy et al., 1991 9, 10
Ginty et al., 1992 74
Gonzalez et al., 1989 16
Gorbunoff, 1985 100
174
Goring and Rothstein, 1992 38
Goring et al., 1993 38, 42
Gowda and Pillay, 1982 30
Hall, 1990 19, 23
Hall, 1992 19
Hanks et al., 1988 34
Harmon et al., 1987 49, 69, 71
Harmon et al., 1988 31
Harper et al., 1991 31, 33, 36,
49, 72, 74, 77
Harper et al., 1993 31, 37, 40,
73, 79
Hasunuma et al., 1987 21
Hayashida et al., 1992 35
Hepler and Gilman, 1992 19,
20, 21
Hepler and Wayne, 1985 7, 30
Herspool, 1976 108
Hetherington and Trewavas,
1982 30
Hetherington and Trewavas,
1984 32
Hicks et al., 1989 52
Hicks et al., 1989a 81, 109,
113, 133
Hicks et al., 1989b 109, 130,
131
Hicks et al., 1989 53
Hicks, 1992 110, 134
Horspool, 1976 132
Huber and Huber, 1990 48
Huber and Huber, 1991 17, 42
Huber et al., 1991 41
Hunter and Karin, 1992 16, 17
Hunter, 1991 48
Irvine et al., 1989 12
Irving et al., 1992 8
Jacobs et al., 1988 22
Janistyn and Ebermann, 1992
15
Janssens, 1988 2, 3
Johannes et al., 1991 7
Jones and Jacobsen, 1991 10
Jones and Venis, 1989 107,
108, 109
Jones et al., 1984a 108, 133
Jones et al., 1984b 108, 126,
133
Karin and Smeal, 1992 17
Katagiri et al. (1989) 16
Keates, 1973 29
Kemp and Pearson, 1990 16,
17
Kemp, 1979 17
Kikkawa and Nishizuka, 1986
40
Klimczak and Hind, 1990 33,
49, 69, 78
Knighton et al., 1991 34, 103
Koontz and Choi, 1993 36, 43
Krebs and Fischer, 1956 44
Krebs, 1986 58
Kruse et al., 1989 14
Kulcis and Polya, 1988 32, 33,
49, 78, 99
Ladror and Zielinski, 1989 32,
49, 99
Laskey and Mills, 1975 114,
128
Lawton et al., 1989 34, 35
Lee et al., 1993 25, 27
Legendre et al., 1992 25, 26,
27
Li and Roux, 1992 42
Lin et al. (1991) 35
Lomax and Hicks, 1991 109
Lucantoni and Polya, 1987 40
Ma et al., 1990 23
Ma et al., 1991 23
Macdonald et al., 1991 133
MacKintosh et al., 1991 41
MacKintosh et al., 1992 42
Martell and Smith (1974) 62
Martinoia et al., 1993 12
175
McAinsh et al. (1990) 8
McCurdy and Harmon, 1992
42
Melhado et al., 1984 126, 127
Moreau and Preisig, 1993 13
Morsucci et al., 1991 18
Nishimoto et al., 1993 21
Nishizuka, 1988 28, 74
Nitschke et al., 1992 42
North, 1989 72
Otte and Moon, 1992 74
Palme et al., 1992 19
Palme, 1992 2
Perdue and Lomax, 1992 22
Perrin and Sayce, 1967 62
Philips, 1983 127
Polya and Davies, 1982 30
Polya et al., 1984 32
Polya et al., 1991 17
Putman-Evans et al., 1989 31
Putnam-Evans et al., 1986 31
Putnam-Evans et al., 1990 31,
51, 58, 70, 71, 72, 79, 100
Ralph et al., 1972 29
Ranjeva and Boudet, 1987 30,
31, 48
Rao and Randall, 1980 48
Raz and Fluhr, 1993 42, 43
Refeno et al., 1982 48
Roberts 1993 37
Roberts and Harmon, 1992 7,
32, 37, 49
Roberts, 1993 32, 49
Robison et al., 1971 15
Rundle and Nasrallah, 1992
41
Ruth et al., 1991 74
Salimath and Marti* 1983 30
Sambrook et al. (1989) 54, 86
Sandelius and Sommarin,
1986 12
Schaffner and Weissmann
(1973) 53, 85
Schaller and Sussman, 1988
32, 49
Schaller et al., 1992 32, 33,
40, 49, 70, 71, 74, 78, 100,
104
Scherer et al., 1988 40
Schroeder and Hagiwara, 1989
8, 9
Schroeder and Hagiwara, 1990
9
Schwacke and Hager, 1992 42
Scott, 1992 16, 29
Short and Briggs, 1992 42
Simon et al., 1991 19, 20, 21,
24
Sithanandam et al., 1990 74
Smith and Walker, 1992 41
Sommarin and Sandelius,
1988 12
Spiteri et al. (1989) 15
Sprang et al., 1988 28
Staros et al., 1978 130
Stein et al., 1991 38
Stryer and Bourne, 1986 22
Suen and Choi, 1991 37
Sunahara et al., 1990 129
Sutherland and Cori, 1956 44
Swarup et al., 1981 56
Tallman, 1992 8
Tan and Boss, 1992 12
Terada et al., 1990 17
Tobias et al., 1992 38
Tominga et al., 1987 42
Trewavas and Gilroy, 1991 7,
13, 15, 16
Trewavas, 1973 29
Trewavas, 1976 29, 30
Ullrich and Schlessinger, 1990
39
176
Veluthambi and Poovaiah,
1984 30
Veluthambi and Poovaiah,
1986 43
Vera and Conejero, 1990 49
Verhey et al., 1993 32, 33, 77,
82
Walker and Zhang, 1990 37,
38
Warpeha et al., 1991 25, 26
Watillon et al., 1992 37
Watillon et al., 1993 37
Westheimer, 1987 2
Wolniak and Larsen, 1992 41
Xu et al., 1992 12
Yan and Tao, 1982 30
Yotsushima et al., 1993 12
Zaina et al., 1991 24
Zaina et al., 1990 24 ,26
Zbell et al., 1990 22
Zbell et al., 1989 24, 107
Zettl et al., 1992 109, 133
Zhang et al., 1992 41
Zocchi, 1985 49