Nuclear Proteins Bind Conserved Elements in the Abscisic Acid-Responsive Promoter of a Rice Rab Gene

Abstract

We have previously shown that the expression of a rice gene, rab-16A, is responsive to abscisic acid (ABA) and osmotic stress in plant tissues and cultured suspension cells. We demonstrate here that transcriptional elements between -294 and -52 of this gene are sufficient to confer ABA-dependent expression on the chloramphenicol acetyltransferase reporter gene in rice protoplasts. Sequence motifs within this 242-base-pair region of the rab-16A gene are conserved among the 5' upstream regions of other ABA-responsive genes. Gel retardation and DNAse I experiments show nuclear factor(s) binding to these sequences. This correlative data indicate that these motifs are involved in the transcription of the rab genes and suggest that they may be ABA-responsive-elements (ABREs).

Full text

Proc.
NatI.
Acad.
Sci.
USA
Vol.
87,
pp.
1406-1410,
February
1990
Botany
Nuclear
proteins
bind
conserved
elements
in
the
abscisic
acid-responsive
promoter
of
a
rice
rab
gene
(cis-acting
elements/DNAse
1
cleavage
inhibitor
pattern/osmotic
stress/PEG-mediated
transfection/trans-acting
factors)
JOHN
MUNDY*,
KAZUKO
YAMAGUCHI-SHINOZAKIt,
AND
NAM-HAI
CHUA
Department
of
Plant
Molecular
Biology,
Rockefeller
University,
1230
York
Avenue,
New
York,
NY
10021
Communicated
by
John
D.
Axtell,
December
4,
1989
(received
for
review
July
24,
1989)
ABSTRACT
We
have
previously
shown
that
the
expres-
sion
of
a
rice
gene,
rab-16A,
is
responsive
to
abscisic
acid
(ABA)
and
osmotic
stress
in
plant
tissues
and
cultured
suspension
cells.
We
demonstrate
here
that
transcriptional
elements
be-
tween
-294
and
-52
of
this
gene
are
sufficient
to
confer
ABA-dependent
expression
on
the
chloramphenicol
acetyl-
transferase
reporter
gene
in
rice
protoplasts.
Sequence
motifs
within
this
242-base-pair
region
of
the
rab-16A
gene
are
conserved
among
the
5'
upstream
regions
of
other
ABA-
responsive
genes.
Gel
retardation
and
DNAse
I
experiments
show
nuclear
factor(s)
binding
to
these
sequences.
This
cor-
relative
data
indicate
that
these
motifs
are
involved
in
the
transcription
of
the
rab
genes
and
suggest
that
they
may
be
ABA-responsive-elements
(ABREs).
The
hormone
abscisic
acid
(ABA)
appears
to
mediate
phys-
iological
processes
in
response
to
osmotic
stress.
Levels
of
endogenous
ABA
increase
in
tissues
subjected
to
osmotic
stress
because
of
high
osmoticum,
salt,
desiccation,
or
cold
(1-3).
Under
these
conditions,
specific
genes
are
expressed
that
can
also
be
induced
in
unstressed
tissues
by
the
appli-
cation
of
exogenous
ABA
(4-6).
A
number
of
these
ABA-
responsive
genes
are
normally
expressed
during
late
embryo-
genesis,
when
seed
tissues
desiccate
and
the
embryos
of
some
species
become
dormant
(7-9).
Therefore,
it
is
thought
that
some
of
these
ABA-responsive
genes
encode
proteins
with
osmoregulatory
or
other
protective
functions
(10-14).
We
are
interested
in
elucidating
how
ABA
regulates
spe-
cific
gene
expression
because
of
its
role
in
seed
development
and
in
the
response
of
plants
to
osmotic
stress,
two
agro-
nomically
important
traits.
To
this
end,
we
have
character-
ized
an
ABA-responsive
rice
gene,
initially
called
rab-21,
that
is
expressed
in
seeds
late
during
embryogenesis
and
that
is
induced
by
ABA
and
osmotic
stress
in
vegetative
tissues
(6).
We
now
have
completely
characterized
a
rice
locus
encoding
this
gene
and
its
three
tightly
linked
homologues
(15).
In
the
present
and
subsequent
publications
we
call
these
genes
rab-16A-D,
in
keeping
with
the
average
molecular
weights
(16,000)
of
the
encoded
RAB
proteins.
Comparison
of
the
5'
upstream
sequences
of
rab-16A-D
and
other
ABA-responsive
genes
reveals
two
conserved
motifs
that
could
be
involved
in
ABA
responsiveness.
To
identify
such
cis-acting
ABA
responsive
elements
(ABREs),
we
prepared
a
series
of
5'
deletion
mutants
and
chimeric
promoter
constructs
to
assay
their
activities
in
protoplasts
prepared
from
rice
suspension
cultures.
Using
these
con-
structs,
we
present
evidence
here
to
show
that
the
rab-16A
gene
is
transcriptionally
regulated
by
ABA.
Furthermore,
we
provide
in
vitro
data
indicating
that
the
conserved
sequence
motifs
in
the
rab-16A
promoter
specifically
bind
nuclear
protein
factors.
Therefore,
these
motifs
may
be
candidate
ABREs.
METHODS
DNA
Manipulations.
All
DNA
manipulations
were
per-
formed
by
standard
procedures
(16).
Deletions
of
the
5'
upstream
region
(Xba
I/Nhe
I;
positions
-2500
to
+27)
of
the
rab-16A
gene
(6),
cloned
in
pEMBL
12+
(17),
were
generated
by
using
BAL-31
exonuclease
digestion.
Plasmid
DNAs
used
in
all
experiments
were
purified
by
CsCl
gradient
centrifu-
gation,
chromatographed
on
Sephadex
G-SOF
in
TE
buffer
(10
mM
Tris/1
mM
EDTA,
pH
8.0)
(0.4
x
40
cm),
and
collected
by
ethanol
precipitation.
Reporter
Gene
Constructs.
Three
types
of
promoter
con-
structs
were
fused
to
the
bacterial
chloramphenicol
acetyl-
transferase
(CAT)
coding
region
with
a
pea
rbcS-E9
poly-
adenylylation
site
(18).
The
three
types of
plasmid
construct
are:
(i)
a
long
35S
promoter
(produces
a
35S
transcript)
from
position
-941
to
+8;
(ii)
rab-16A/35S
chimeric
promoters
containing
rab-16A
5'
deletion
fragments
(starting
at
posi-
tions
-1505,
-770,
and
-442)
that
are
truncated
at
-52
(Sac
I
site)
and
fused
to
the
35S
promoter
TATA
box
(from
-90
to
+8);
and
(iii)
rab-16A
5'
deletion
fragments
(starting
at
-442,
-294,
and
-52)
fused
to
the
CAT-encoding
region
at
+27
of
the
rab-16A
promoter.
Transient
Expression
Assay
in
Rice
Protoplasts.
Rice
sus-
pension
cells
were
cultured
from
embryo-derived
callus
of
Tapei
309
in
standard
media
(19).
PEG-mediated
transfection
was
used
to
introduce
the
constructs
into
rice
protoplasts
as
described
by
Krens
et
al.
(20).
Twenty-four
hours
after
transfection,
half
of
the
protoplasts
were
incubated
in
10
,uM
ABA,
harvested
18
hr
later,
and
assayed
for
CAT
activity
as
described
by
Nagy
et
al.
(21).
Preparation
of
Binding
Protein
Extracts.
Whole-cell
protein
extracts
were
prepared
from
roots
and
shoots
of
20-day-old
rice
seedlings
grown
hydroponically
as
described
by
Green
et
al.
(22).
Nuclear
extracts
were
regularly
prepared
from
shoots
of
9-day-old
dark-grown
plants
(22).
For
ABA
treat-
ments,
leaves
were
sprayed
with
100
,uM
ABA/0.02%
Tween
at
24
hr,
12
hr,
and
3
hr
before
harvest.
The
average
yield
of
protein
per
nuclear
extract
was
100
mg
per
kg
of
leaves
(3.5
liters
of
seeds
as
starting
material).
Gel
Retardation
and
DNAse
I
Cleavage
Inhibition
Patterns
(Footprinting).
Probes
were
labeled
at
the
polylinker
HindIII
site
(5'
end
of
rab-16A;
fragment
from
-290
to
+27)
with
Klenow
enzyme,
digested
with
Sac
I
(cleavage
at
-52,
for
gel
retardation
assays),
isolated
on
4%
polyacrylamide
gels,
and
characterized
by
isotachophoresis.
Competitor
fragments
were
Abbreviations:
ABA,
abscisic
acid;
ABRE,
abscisic
acid-responsive
DNA
element;
rab,
gene
responsive
to
abscisic
acid.
*Present
address:
Department
of
Biotechnology,
Carlsberg
Research
Laboratory,
Gl.
Carlsberg
Vej
6,
Dk-2500,
Copenhagen,
Denmark.
tPresent
address:
Department
of
Cell
Biology,
National
Institute
of
Agrobiological
Resources,
Yatabe,
Ibaraki
305,
Japan.
1406
The
publication
costs
of
this
article
were
defrayed
in
part
by
page
charge
payment.
This
article
must
therefore
be
hereby
marked
"advertisement"
in
accordance
with
18
U.S.C.
§1734
solely
to
indicate
this
fact.

Proc.
Natl.
Acad.
Sci.
USA
87
(1990)
1407
isolated
on
agarose
gels
by
using
DE-81
nylon
membranes
(Schleicher
&
Schuell
NA-45).
A
typical
binding
reaction
con-
tained
20
fmol
of
probe
(5,000-20,000
cpm)
and
a
100-fold
molar
excess
of
specific
inhibitor
DNA.
Gel
retardation
assays
were
performed
as
described
by
Green
et
al.
(22)
with
nuclear
extract
protein
at
0.6
ug/gl,
poly(dI)-poly(dC)
at
0.5
gg/pl,
and
poly(dA)-poly(dT)
at
2
gg/pl
as
competitors.
DNAse
I
footprint
assays
were
performed
with
extract
protein
at
4
gg/pl,
poly-
(dI)-poly(dC)
at
1
,ug/tl,
and
poly(dA)*poly(dT)
at
5
,ug/gl
as
competitors
(22).
RESULTS
5'
Upstream
Sequences
of
the
rab-16A
Gene
Confer
Respon-
siveness
to
ABA
upon
the
CAT
Reporter
Gene.
We
showed
previously
that
accumulation
of
rab-16
transcripts
is
induc-
ible
in
various
rice
tissues
by
ABA
and
osmotic
stress
(6).
To
examine
whether
this
induction
is
due
at
least
in
part
to
transcriptional
control
by
rab-16
5'
upstream
sequences,
we
constructed
gene
fusions
using
5'
upstream
fragments
of
the
rab-16A
gene
and
the
bacterial
CAT
reporter
gene.
These
gene
constructs
were
introduced
into
rice
protoplasts
by
using
PEG-mediated
transfection
to
assay
for
transient,
ABA-responsive
CAT
enzyme
expression.
Fig.
1A
outlines
the
gene
constructs
used,
and
Fig.
1B
shows
the
results
of
a
typical
CAT
enzyme
assay.
The
control
transfections
using
constructs
1
and
2
clearly
show
that
CAT
expression
is
promoter
dependent
and
that
transcription
from
the
strong
35S
viral
promoter
is
not
responsive
to
ABA.
Transfections
with
the
chimeric
rab-16A/35S
constructs
3-5
(rab
deletion
fragments
starting
at
-1505,
-770,
and
-442,
respectively)
show
low
levels
of
CAT
activity
in
protoplasts
incubated
without
ABA.
This
low-level,
consti-
tutive
expression
may
be
due
to
cumulative
effects
of
pro-
moter
elements
within
the
rab-16A
gene
sequences
upstream
of
-52
and
elements
within
the
35S
promoter
fragment
from
-90
to
+8.
This
region
of
the
35S
promoter,
which
contains
the
putative
TATA
box,
has
recently
been
shown
to
contain
elements
capable
of
enhancing
transcription
in
tobacco
cells
(18).
Therefore,
it
is
not
surprising
that
it
affects
basal
level
expression
in
our
system.
Of
greater
significance
is
the
fact
that
such
chimeric
promoters
are
responsive
to
ABA.
This
can
be
clearly
seen
in
the
transfection
experiments
with
the
shorter
promoter
containing
the
rab-16A
5'
upstream
fragment
from
position
-442
to
-52
(construct
5),
as
well
as
that
with
the
fragment
from
-770
to
-52
(construct
4).
A
comparison
of
the
three
chimeric
promoters
assayed
(constructs
3-5)
suggests
that
silencer-like
sequences
that
appear
to
diminish
ABA
respon-
siveness
reside
upstream
of
-440
in
the
rab-16A
DNA.
Similar
reductions
in
ABA-inducible
promoter
strength
with
increasing
promoter
length
have
been
seen
in
comparable
assays
of
5'
upstream
sequences
of
the
wheat
EM
gene
with
its
own
TATAA
box
(23).
Therefore,
it
is
likely
that
this
"silencing"
effect
is
independent
of
the
presence
of
35S
sequences
in
our
chimeric
constructs.
Transfections
with
constructs
6
and
7
show
that
sequences
between
-290
and
+27
of
the
rab-16A
gene
confer
strong,
ABA-dependent
expression
on
the
CAT
reporter
gene.
CAT
enzyme
activity
in
control
(-ABA)
incubated
protoplasts
was
undetectable,
while
incubation
with
10AM
ABA
resulted
in
CAT
levels
comparable
to
that
seen
in
protoplasts
trans-
fected
with
constructs
containing
the
full
35S
viral
promoter.
Protoplasts
transfected
with
construct
8
do
not
express
detectable
levels
of
CAT,
indicating
that
sequences
between
-52
and
+27
do
not
mediate
the
ABA
responsiveness
of
the
rab-16A
gene.
This
is
further
supported
by
the
fact
that
constructs
4
and
5,
which
lack
this
region,
are
clearly
responsive
to
ABA.
Therefore,
we
conclude
that
sequences
between
-290
and
-52
of
the
rab-16A
gene
contain
ABA-
A
1
promoter-less
2
35S
-1500
3
rab/35S
=
-941
-90
-770
4
rab/35S
I
-442
zzz
Eaz
5
rab/35S
-442
l'-
6
rab
7
rab
-290
E
._._
l
-52
EC3
8
rab
B
0
4
v@
1
2
2
3
3
-.
.
.
o
o
CONSTRUCT
4
4
5
5
6
ABA.
6
7
7
8
8
67788*
_
FIG.
1.
Transient
expression
of
CAT
gene
fusions
in
rice
proto-
plasts.
(A)
CAT
fusion
constructs
with
the
5'
upstream
sequences
of
the
cauliflower
mosaic
virus
35S
transcription
unit
and
rab-16A
gene
used
for
transfection.
The
constructs
are:
1,
negative
control,
"promoterless"
CAT
gene;
2,
positive
control,
constitutive
35S
promoter
(-941
to
+8);
3,
chimeric
rab-16A/35S
promoter
[rab-16A
fragment
from
-1505
to
-52
fused
to
35S
promoter
TATA
box
(-90
to
+8)];
4,
rab-16A
fragment
from
-770
to
-52
fused
to
35S
promoter
TATA
box;
5,
rab-16A
fragment
from
-442
to
-52
fused
to
35S
promoter
TATA
box;
6,
rab-16A
fragment
from
-442
to
+27
of
rab-16A
promoter,
fused
to
CAT
gene
(-442/+27
rab-16A/CAT);
7,
-294/+27
rab-16A/CAT;
8,
-52/+27
rub-16A/CAT.
(B)
CAT
as-
says
of
rice
protoplasts
transfected
with
the
above
constructs
and
incubated
without
or
with
10
,uM
ABA
as
marked.
responsive
DNA
elements
that
modulate
the
transcription
of
this
gene.
Sequence
Motifs
Between
-294
and
-52
of
the
rab-16A
Gene
Are
Conserved
Among
5'
Upstream
Regions
of
Other
ABA-
Responsive
Genes.
Fig.
2A
shows
the
sequence
of
the
prox-
imal
300
base
pairs
(bp)
of
the
rab-16A
promoter
(6).
Our
transfection
experiments
indicate
that
sequence
elements
within
this
region
are
capable
of
controlling
ABA-dependent
gene
expression.
Such
ABREs
might
be
expected
to
be
conserved
within
the
5'
upstream
sequences
of
the
rab-16A
and
other
ABA-responsive
genes.
We
have
previously
noted
that
several
G+C-rich
sequence
motifs
are
duplicated
within
the
rab-16A
promoter.
More
recently,
we
have
sequenced
three
other
rab-16
genes
(rab-16B-D;
ref
15).
These
genes
contain
single
copies
of
one
of
the
G+C-rich
motifs
noted
in
the
rab-16A
gene
(Fig.
2B,
motifs
Ila
and
hIb).
Furthermore,
the
upstream
regions
of
all
of
the
rice
rab-16
genes
contain
a
motif
that
is
also
found
in
the
upstream
regions
of
ABA-
responsive
genes
from
cotton
(Fig.
2B,
motif
I).
Botany:
Mundy
et
al.

Proc.
Natl.
Acad.
Sci.
USA
87
(1990)
A
ATCCACGGCG
AGCACTCATC
CAAACCGTCC
ATCCACGCGC
ACAGTACACA
-251
CACATAGTTA
TCGTCTCTCC
CCCCGATGAG
TCACCACCCG
TGTCTTCGAG
-201
AAACGCCTCG
CCCGACACCG
ITAcSTpCC
CACCOCCGCG
CCTGCCGCCT
-151
GGACACGTCC
GGCTCCTCTC
CCGCCGCGT
GGCCACCGTC
CACCGGCTCC
-101
CGCACACGTC
TCCCTGTCTC
CCTCCACCCA
TGCCGTGGCA
ATCGAGCTCA
-51
TCTCCTCGCC
TCCTCCGGCT
TATAAATGGC
GGCCACCACC
TTCACCTGCT
-1
Motif
I
Motif
IIa
Motif
IIb
A
PROBE
-52
0
MOTIF
I
lia
lib
-52
COMPETITOR
1
-294
2
RICE
-189**
*
*
-171**:*****
T:
-130:**:******
rab
16A
ACACCGTACGTGGCGCCA
ACCGCCGCGCCTGC
TCCCGCCGCGCTGG
-270**:
********
::
-242
**.*
rab
16B
ACACAGTACGTGGCA-GC
ACACGGCGCGCTAC
-242**
********
-196:**.******
rab
16C
ACACA-TACGTGGCGTGC
CCCCGGCGCGCTAC
-190**
**
-174
***
rab
16D
GTAC-GTACGTGGCGCGC
GGGCGCCGCGCTGA
CONCINSUS
ACAC--TACGTGGCG-GC
COTTO
D-7
D-19
D-29
D-34
D-113
CCGCCGCGCTG
422*
******
ACAAGATACGTGTTTCAT
745
**
TGTGCTTACGTGGATCAC
1162
**
TAGGGGATCGTGGCTATA
1078
******
TTGGGTTACGTGTTAAGG
1483
CTTGTATACGTGGCAGCT
FIG.
2.
5'
Upstream
sequences
of
the
rice
rab-16A
and
other
ABA-responsive
genes.
(A)
Sequence
of
the
region
from
-300
to
-1
of
the
rab-16A
gene
showing
deletion
points
used
in
this
study
(arrows),
conserved
motifs
I
(box)
and
Ila
and
Ilb
(underlined),
and
putative
CAAT
and
TATA
(bold
letters).
(B)
Sequence
comparison
of
motifs
I
and
II
found
in
the
rab-16A
(6)
and
other
ABA-responsive
genes.
rab-16B-D
are
the
three
other
tightly
linked
members
of
the
rab-16
locus
of
rice
(15).
Lea-D7-113
are
cotton
genes
described
by
Baker
et
al.
(8).
*,
Base
conserved
within
the
motif
among
all
four
rab-16
genes;
:,
base
conserved
within
the
motif
among
three
rab-16
genes.
Examination
of
the
motif
I
sequence
shows
that
it
has
some
homology
to
the
cyclic
AMP-responsive
element
described
by
Deutsch
et
al.
(24).
We
have
noted
(25)
that
motif
II
is
similar
to
the
binding
site
of
SP1,
a
mammalian
transcription
factor.
Visual
and
computer
analysis
did
not
identify
any
regions
of
significant
homology
between
rab-16A
upstream
sequences
and
steriod
hormone-responsive
elements
identi-
fied
in
mammalian
genes
(26).
The
Conserved
Sequence
Motifs
I,
IHa,
and
Ub
Are
Sites
for
Nuclear
Protein
Billding.
To
attempt
a
finer
delination
of
the
regulatory
regions
of
the
rab-16A
promoter,
we
examined
whether
sequences
within
the
region
from
-290
to
+27
specifically
bind
cellular
or
nuclear
protein
factors.
In
initial
experiments,
whole-cell
protein
extracts
were
prepared
from
control
and
ABA-treated
roots
or
shoots.
We
have
shown
previously
that
both
of
these
tissues
accumulate
rab-16A
mRNA
in
response
to
ABA.
No
discrete
binding
was
seen
between
proteins
in
these
extracts
and
the
rab-16A
probe.
However,
discrete
binding
was
achieved
in
experiments
with
proteins
extracted
from
shoot
nuclei
(Fig.
3).
Competition
experiments
with
unlabeled
upstream
fragments
(Fig.
3A)
suggest
that
two
different
binding
sites
occur
between
-194
and
-102,
one
of
them
lying
upstream
of
-154
(Fig.
3B,
see
arrows).
The
simplest
explanation
for
these
gel
retardation
results
is
that
the
slower
complex
is
due
to
factor(s)
binding
to
the
type
II
motif,
while
the
faster
one
is
a
complex
with
motif
I.
This
pattern
of
binding
was
consistently
seen
for
proteins
extracted
from
shoot
nuclei
of
both
control
and
ABA-treated
plants
(data
not
shown).
Although
the
level
of
this
binding
(cpm/,ug
of
protein)
in
extracts
from
ABA-treated
shoots
was
4
-194
-102
*
o
*
3
-294
-154
-168
-52
0
0
B
COMPLEX
E:|
PROBE
EXTRACT
-
+
+
+
+
+
COMPETITOR
-
-
1
2
3
4
FIG.
3.
Gel
retardation
assay
delineating
a
region
of
the
rab-16A
promoter
that
binds
nuclear
factor(s).
(A)
rab-16A
5'
upstream
fragments,
generated
by
BAL-31
exonuclease
digestion,
used
as
probes
and
as
the
following
competitors:
1,
cold
probe
fragment
from
-294
to
-52;
2,
region
containing
motifs
I,
Ila,
and
Ilb,
from
-194
to
-101;
3,
region
containing
motif
I
and
motif
Ila,
from
-294
to
-154;
4,
region
containing
motif
Ilb,
from
-168
to
-52.
Motifs
I
(*),
IIA
(o),
and
IIB
(e)
occur
as
marked.
See
Methods
for
probe/
competitor
concentrations.
(B)
Gel
retardation
experiment
assessing
binding
of
nuclear
proteins
from
ABA-treated
leaves
to
the
rab-16A
promoter
region
from
-294
to
-52.
sometimes
2-
to
3-fold
higher
than
in
extracts
from
control
shoots,
the
data
suggested
that
binding
of
these
factors
in
vitro
is
not
significantly
increased
by
ABA.
Addition
of
ABA
to
the
binding
reactions
also
had
no
effect
on
the
levels
of
specific
DNA-protein
complexes
formed.
To
further
define
the
sequence(s)
that
interact
with
nuclear
proteins,
we
carried
out
DNAse
I
footprinting
experiments.
Fig.
4
shows
what
appears
to
be
two
major
areas
of
protection
(positions
-105
to
-131
and
-155
to
-180)
that
coincide
with
the
conserved
sequence
motifs
I
and
Ila/Ilb.
The
presence
of
two
bands
in
the
gel
retardation
assays
(Fig.
3)
and
the
bipartite
nature
of
the
footprint
suggest
that
binding
of
a
factor
to
motif
I
may
be
independent
of
binding
to
motifs
Ila
or
lIb.
Taken
together
with
the
in
vivo
activities
of
the
5'
B
%f
-
1408
Botany:
Mundy
et
al.
-294
dr-
%
-410-

Proc.
Natl.
Acad.
Sci.
USA
87
(1990)
1409
G/A
0
2
5
,W
1-105
MOTIF
flb
40
~-131
q
N.
-155
MOTIF
Ila
_ff
.2
_*
MOTIF
_-.
4
-
S
a,
A
-180
_
__
WV"M
FIG.
4.
DNAse
I
footprint
experiment
showing
the
binding
sites
of
nuclear
protein(s)
in
the
region
from
-290
to
+27
of
the
rab-16A
promoter.
The
probe
is
the
3'
end-labeled
bottom
strand
of
the
deleted
promoter
region
shown
in
Fig.
2A.
Lane
G/A
shows
the
G+A
product
ladder
derived
by
chemically
sequencing
the
probe.
The
amount
(,Ag/IAd)
of
nuclear
extract
used
to
generate
the
footprint
is
given
above
the
remaining
lanes.
deletion
mutants,
these
in
vitro
results
indicate
that
motifs
I
and
II
may
be
important
in
some
aspect
of
the
transcription
of
rab
genes,
most
probably
their
response
to
ABA.
DISCUSSION
We
began
our
study
of
the
molecular
mechanism
of
plant
hormone
action
by
isolating
the
members
of
a
rice
gene
family,
the
rab-16A-D
genes,
whose
expression
is
strongly
induced
by
ABA
(6,
15).
We
are
analyzing
the
ABA-
responsive
expression
of
the
rab-16
genes
as
a
model
system
to
elucidate
the
mechanism
of
action
of
this
hormone.
In
the
present
work,
we
show
that
the
region
between
positions
-294
and
+27
in
the
5'
upstream
region
of
the
rab-16A
gene
is
sufficient
to
confer
ABA-responsive,
transient
expression
upon
the
CAT
reporter
gene
in
transfected
rice
protoplasts.
These
results
are
similar
to
those
reported
by
Marcotte
et
al.
(23),
who
showed
that
sequences
between
-550
and
+95
of
the
wheat
Em
gene
confer
ABA
induction
upon
the
GUS
reporter
gene
in
rice
protoplasts.
Our
experiments
also
indicate
that
the
rab-16A
region
between
positions
-290
and
-52
alone
acts
to
enhance
the
expression
of
the
CAT
gene
when
placed
5'
to
a
heterologous
TATA
box
from
the
constitutively
expressed
35S
cauliflower
mosaic
virus
gene.
These
results
indicate
that
this
242-bp
region
of
the
rab-16A
gene
contains
one
or
more
ABREs.
Comparison
of
the
5'
upstream
sequences
of
the
rab-
16A-D
genes
and
those
from
several
ABA
responsive,
late
embryogenesis
abundant
(lea)
genes
from
cotton
(8)
revealed
conserved
sequence
motifs
that
are
good
candidates
for
ABREs.
Two
motifs
were
found
to
be
conserved
in
all
four
rab-16
genes.
Motif
I
has
the
consensus
RTACGTGGR
(R
is
an
unspecified
purine
nucleoside),
which
is
similar
to
the
cAMP-responsive
element
(TGACGTCA)
that
binds
the
transcription
factor
CREB
(24).
More
importantly,
motif
I
is
found
in
the
5'
upstream
regions
of
five
of
six
lea
genes
and
in
that
of
the
ABA-responsive
wheat
Em
gene
(R.
Quatrano,
personal
communication).
Motif
II,
which
is
found
in
two
copies
(Ila
and
Ilb)
in
rab-16A
and
once
in
rab-16B-D
,
has
the
consensus
CGSCGCGCT,
in
which
S
is
G
or
C.
It
occurs
in
the
rab-16
genes
as
part
of
sequences
that
are
similar
to
the
degenerate
decanucleotide
binding
site
of
SP],
an
auxilliary
mammalian
transcription
factor
(25).
The
foregoing
results
prompted
us
to
test
whether
the
region
of
the
rab-16A
gene
from
-290
to
-52,
and
in
particular
the
conserved
motifs
I
and
II,
specifically
bind
cytosolic
or
nuclear
proteins.
Gel
retardation
studies
showed
that
this
upstream
region
indeed
binds
nuclear
protein(s),
forming
what
appears
to
be
two
discrete
DNA-protein
bind-
ing
complexes
limited
to
the
region
containing
motifs
I,
Ila,
and
lIb
(from
-192
to
-102).
At
this
level
of
resolution,
it
is
not
possible
to
determine
whether
a
single
protein
or
two
protein
species
bind
to
this
region.
These
complexes
are
formed
at
only
slightly
higher
levels
by
nuclear
proteins
extracted
from
ABA-treated
tissues
than
by
those
from
control
tissues.
This
suggests
that
binding
of
the
factor(s)
is
not
promoted
by
the
hormone
in
vitro.
This
"constitutive"
binding
has
been
noted
for
various
activator
and
regulatory
factors
(26,
27).
Such
factors
apparently
activate
transcrip-
tion
only
when
modified
by
interaction
with
other
proteins
or
by
changes
in
their
phosphorylation
state.
DNAse
I
footprinting
enabled
us
to
show
that
the
detect-
able
sites
of
nuclear
protein
binding
to
the
rab-16A
promoter
are
limited
to
the
sequences
containing
the
conserved
motifs
I,
Iha,
and
lIb.
These
correlative
data
indicate
that
these
sequences
are
involved
in
the
transcription
of
the
rab-16
genes,
and
possibly
of
the
lea
genes,
and
suggest
that
motifs
I
and
II
are
candidate
ABREs.
Work
on
animal
hormones
has
defined
two
major
pathways
of
hormone
action:
(i)
activation
of
regulatory
factors
by
direct
steroid
hormone
binding
(26)
and
(ii)
activation
via
"second
messenger"
pathways
(24).
Recent
studies
show
that
the
response
of
cells
to
auxin
involves
plasmalemma
receptors
and
phosphatidylinositol
metabolites,
suggesting
that
second
messenger
pathways
mediate
the
action
of
this
hormone
(28,
29).
The
results
presented
here
do
not
permit
us
to
discern
whether
these
mechanisms
mediate
ABA-
responsive
expression
of
the
rab-16
genes.
Evidence
from
other
tissue
systems
suggests
that
plasmalemma
ion-
transport
proteins
(30),
receptors
(31),
and
protein
phosphor-
ylation
(32)
are
involved
in
cellular
responses
to
ABA.
However,
we
were
unable
to
detect
changes
in
rab-16
gene
expression
in
cultured
rice
suspension
cells
incubated
with
bromo-cAMP,
phorbol
12-myristate
13-acetate
forskolin,
Ca",
or
a
Ca2l
ionophore-molecules
that
affect
second-
messenger
signaling
in
animal
cells
(unpublished
data).
We
hope
to
use
defined
ABREs
as
probes
for
identifying
clones
encoding
factors
binding
to
the
rab-16A
promoter.
The
char-
acterization
of
these
factors
will
provide
useful
tools
with
Botany:
Mundy
et
al.

1410
Botany:
Mundy
et
al.
which
to
dissect
the
pathway(s)
of
ABA-induced
gene
expres-
sion.
We
thank
Irene
Roberson
for
technical
help
and
Steve
Kay
for
expert
advice.
This
work
was
supported
by
a
grant
from
the
Rockefeller
Foundation.
1.
Henson,
I.
E.
(1984)
Ann.
Botany
54,
569-582.
2.
Davies,
W.
J.,
da
Costa,
A.
R.
&
Lodge,
T.
A.
(1987)
in
Advances
in
Agricultural
Biochemistry.
Plant
Growth
and
Development,
ed.
Purohit,
J.
(Nijhoff,
The
Hague),
Vol.
2,
pp.
151-170.
3.
Mohapatra,
S.
S.,
Poole,
R.
J.
&
Dhindsa,
R.
S.
(1988)
Plant
Physiol.
87,
468-473.
4.
Singh,
N.
K.,
LaRosa,
P.
C.,
Handa,
A.
D.,
Hasegawa,
P.
M.
&
Bressan,
R.
A.
(1987)
Proc.
Natl.
Acad.
Sci.
USA
84,
739-743.
5.
Gomez,
J.,
Sanchez-Martinez,
D.,
Stiefel,
V.,
Rigau,
J.,
Puig-
domenech,
P.
&
Pages,
M.
(1988)
Nature
(London)
334,
262-
264.
6.
Mundy,
J.
&
Chua,
N.-H.
(1988)
EMBO
J.
7,
2279-2286.
7.
Finkelstein,
R.
R.,
Tenbarge,
K.
M.
&
Crouch,
M.
L.
(1986)
Plant
Physiol.
78,
630-636.
8.
Baker,
J.
C.,
Steele,
C.
&
Dure,
L.,
III
(1988)
Plant
Mol.
Biol.
11,
277-291.
9.
Koorneef,
M.,
Hanhart,
C.
J.,
Hihorst,
H.
W.
M.
&
Karssen,
C.
M.
(1989)
Plant
Physiol.
90,
462-469.
10.
Mundy,
J.,
Hejgaard,
J.,
Hansen,
A.,
Hallgren,
L.,
Jorgensen,
G.
&
Munck,
L.
(1986)
Plant
Physiol.
81,
626-630.
11.
Richardson,
M.,
Valdez-Rodriguez,
S.
&
Blanco-Labra,
A.
(1987)
Nature
(London)
327,
432-434.
12.
Bartels,
D.,
Singh,
M.
&
Salamini,
F.
(1988)
Planta
175,
485-492.
13.
Dure,
L.,
lII,
Crouch,
M.,
Harada,
J.,
Ho,
T.-H.,
Mundy,
J.,
Quatrano,
R.,
Thomas,
T.
&
Sung,
Z.
R.
(1989)
Plant
Mol.
Biol.
12,
475-486.
Proc.
Natl.
Acad.
Sci.
USA
87
(1990)
14.
Mortenson,
E.
&
Dreyfuss,
G.
(1989)
Nature
(London)
337,
312.
15.
Yamaguchi-Shinozaki,
K.,
Mundy,
J.
&
Chua,
N.-H.
(1990)
Plant
Mol.
Biol.,
in
press.
16.
Maniatis,
T.,
Fritsch,
E.
F.
&
Sambrook,
J.
(1982)
in
Molecular
Cloning:
A
Laboratory
Manual
(Cold
Spring
Harbor
Labora-
tory,
Cold
Spring
Harbor,
New
York).
17.
Dente,
L.,
Cesarini,
G.
&
Cortese,
R.
(1983)
Nucleic
Acids
Res.
11,
1645-1655.
18.
Lam,
E.,
Benfey,
P.
N.,
Gilmartin,
P.
M.,
Fang,
R.-X.
&
Chua,
N.-H.
(1989)
Proc.
Natl.
Acad.
Sci.
USA
86,
7890-7894.
19.
Kao,
K.
N.
(1977)
Mol.
Gen.
Genet.
150,
225-230.
20.
Krens,
F.
A.,
Molendjik,
F.,
Wullems,
G.
&
Schilperoort,
R.
A.
(1982)
Nature
(London)
296,
72-74.
21.
Nagy,
F.,
Kay,
S.
A.
&
Chua,
N.-H.
(1988)
in
Plant
Molecular
Biology
Manual,
eds.
Gelvin
S.
V.
&
Schilperoort,
R.
A.
(Kluwer,
Dordrecht),
Vol.
B4,
pp.
1-29.
22.
Green,
P.,
Kay,
S.
A.,
Lam,
E.
&
Chua,
N.-H.
(1988)
in
Plant
Molecular
Biology
Manual,
eds.
Gelvin,
S.
V.
&
Schilperoort,
R.
A.
(Kluwer,
Dordrecht),
Vol.
B11,
pp.
1-21.
23.
Marcotte,
W.
R.,
Jr.,
Bayley,
C.
C.
&
Quatrano,
R.
S.
(1988)
Nature
(London)
335,
454-457.
24.
Deutsch,
P.
J.,
Hoeffler,
J.
P.,
Jameson,
J.
L.
&
Jabener,
J.
F.
(1988)
Proc.
Natl.
Acad.
Sci.
USA
85,
7922-7926.
25.
Briggs,
M.,
Kadonaga,
J.
T.,
Bell,
S.
P.
&
Tjian,
R.
(1986)
Science
234,
47-52.
26.
Beato,
M.
(1989)
Cell
56,
335-344.
27.
Johnson,
P.
F.
&
McKnight,
S.
L.
(1989)
Annu.
Rev.
Biochem.
58,
799-839.
28.
Ettlinger,
C.
&
Lehle,
L.
(1988)
Nature
(London)
331,
176-178.
29.
Hicks,
G.
R.,
Rayle,
D.
L.,
Jones,
A.
M.
&
Lomax,
T.
L.
(1989)
Proc.
Natl.
Acad.
Sci.
USA
86,
4948-4952.
30.
Schroeder,
J.
I.
&
Hedrich,
R.
(1989)
Trends
Biochem.
Sci.
14,
187-192.
31.
Hornberg,
C.
&
Weiler,
E.
W.
(1984)
Nature
(London)
310,
321-324.
32.
Goday,
A.,
Sanchez-Martinez,
D.,
Gomez,
J.,
Puigdomenech,
P.
&
Pages,
M.
(1988)
Plant
Physiol.
88,
564-569.