Abscisic acid and water-stress induce the expression of a novel rice gene

Abstract

We have identified a novel rice gene, called RAB 21, which is induced when plants are subject to water-stress. This gene encodes a basic, glycine-rich protein (mol. wt 16,529) which has a duplicated domain structure. Immunoblots probed with antibodies raised against beta-galactosidase/RAB 21 fusion protein detect RAB 21 protein only in cytosolic cell fractions. RAB 21 mRNA and protein accumulate in rice embryos, leaves, roots and callus-derived suspension cells upon treatment with NaCl (200 mM) and/or the plant hormone abscisic acid (10 microM ABA). The effects of NaCl and ABA are not cumulative, suggesting that these two inducers share a common response pathway. Induction of RAB 21 mRNA accumulation by ABA is rapid (less than 15 min in suspension cells) and does not require protein synthesis, indicating that preformed nuclear and/or cytosolic factors mediate the response to this hormone. We have characterized the RAB 21 gene by determining the complete nucleotide sequence of a nearly full-length cDNA and corresponding genomic copy, and by mapping the start site of its major transcript. The proximal promoter region contains various GC-rich repeats.

Full text

The
EMBO
Journal
vol.7
no.8
pp.2279-2286,
1988
Abscisic
acid
and
water-stress
induce
the
expression
of
a
novel
rice
gene
John
Mundy
and
Nam-Hal
Chua
Laboratory
of
Plant
Molecular
Biology,
Rockefeller
University,
1230
York
Avenue,
New
York,
NY
10021,
USA
Communicated
by
D.von
Wettstein
We
have
identified
a
novel
rice
gene,
called
RAB
21,
which
is
induced
when
plants
are
subject
to
water-stress.
This
gene
encodes
a
basic,
glycine-rich
protein
(mol.
wt
16
529)
which
has
a
duplicated
domain
structure.
Immunoblots
probed
with
antibodies
raised
against
B-
galactosidase/RAB
21
fusion
protein
detect
RAB
21
pro-
tein
only
in
cytosolic
cell
fractions.
RAB
21
mRNA
and
protein
accumulate
in
rice
embryos,
leaves,
roots
and
callus-derived
suspension
cells
upon
treatment
with
NaCl
(200
mM)
and/or
the
plant
hormone
abscisic
acid
(10
/tM
ABA).
The
effects
of
NaCI
and
ABA
are
not
cumulative,
suggesting
that
these
two
inducers
share
a
common
response
pathway.
Induction
of
RAB
21
mRNA
accumu-
lation
by
ABA
is
rapid
(<
15
min
in
suspension
cells)
and
does
not
require
protein
synthesis,
indicating
that
pre-
formed
nuclear
and/or
cytosolic
factors
mediate
the
re-
sponse
to
this
hormone.
We
have
characterized
the
RAB
21
gene
by
deternmining
the
complete
nucleotide
sequence
of
a
nearly
full-length
cDNA
and
corresponding
genomic
copy,
and
by
mapping
the
start
site
of
its
major
tran-
script.
The
proximal
promoter
region
contains
various
GC-rich
repeats.
Key
words:
genomic
sequence/cDNA
sequence/cyclohex-
imide/osmotic
regulation/water
deficit
Introduction
The
hormone
abscisic
acid
(ABA)
mediates
a
number
of
important
physiological
processes
in
plants
(King,
1976;
Jones
et
al.,
1987).
Developmental
studies
have
shown
that
ABA
induces
the
accumulation
of
specific
mRNAs
and
proteins
late
during
embryogenesis
in
seeds
of
diverse
species
(Galau
et
al.,
1986;
Finklestein
et
al.,
1985;
Litts
et
al.,
1987).
At
this
time,
the
level
of
endogenous
ABA
increases,
the
seeds
desiccate
and
the
embryos
of
some
species
become
dormant
(King,
1976;
Suzuki
et
al.,
1981).
In
cereal
seeds,
some
of
the
'late'
developmental
mRNAs
are
long-lived
in
mature,
dry
grains,
but
are
rapidly
degraded
during
seed
germination.
However,
exogenously
applied
ABA
causes
the
precocious
accumulation
of
these
mRNAs
in
immature
em-
bryos
and
their
reappearance
in
germinating
seeds
(Finkle-
stein
et
al.,
1985;
Mundy
et
al.,
1986).
Little
is
known
about
the
functions
of
these
ABA-inducible
proteins
or
about
their
intracellular
localization.
Some
of
them
are
storage
poly-
peptides
(Finklestein
et
al.,
1985;
Bray
and
Beachy,
1985),
while
others,
such
as
lectins
(Raikel
and
Wilkins,
1987)
or
an
enzyme
inhibitor
(Mundy
et
al.,
1986)
may
be
involved
in
seed
protection
and/or
the
maintenance
of
dormancy.
Physiological
studies
have
shown
that
endogenous
ABA
levels
increase
in
plant
tissues
subjected
to
water-stress
by
high
osmoticum,
NaCl,
or
drying
(Henson,
1984;
Jones
et
al.,
1987).
Under
these
conditions,
specific
mRNAs
and
proteins
accumulate
which
could
affect
intra-cellular
osmo-
larity
or
have
other
protective
functions
(Finklestein
and
Crouch,
1986;
Ramagopal,
1987).
One
such
salt-inducible
protein
whose
accumufation
is
increased
by
ABA
has
been
isolated
from
tobacco
and
shown
to
be
homologous
to
members
of
a
group
of
proteinase
inhibitors
(Singh
et
al.,
1987;
Richardson
et
al.,
1987).
These
results
suggest
that
some
of
the
ABA-inducible
mRNAs
and
proteins
which
accumulate
during
seed
desiccation
are
part
of
a
general
response
by
the
plant
to
water
deficit.
If
so,
then
part
of
this
response
involves
the
synthesis
of
enzyme
inhibitors
and
lectins
which
may
protect
plant
tissues
from
degradation
by
pathogens
during
periods
of
arrested
growth
and
development.
Both
the
developmental
studies
on
seeds
and
the
physio-
logical
studies
on
water-stress
indicate
that
ABA
controls
the
accumulation
of
specific
mRNAs
and
proteins.
However,
it
is
unclear
whether
ABA
acts
at
the
transcriptional
or
post-
transcriptional
level,
or
both
(Jacobsen
and
Beach,
1986;
Mozer,
1980).
The
mode
of
action
of
the
hormone
via
receptors
and/or
transducing
pathways
also
remains
obscure
(Homberg
and
Weiler,
1984)
and
it
is
not
known
whether
transduction
of
the
ABA
response
signals
requires
de
novo
protein
synthesis.
To
date,
no
genomic
sequences
have
been
reported
for
plant
genes
strongly
induced
by
ABA.
Charac-
terization
of
promoter
sequences
of
such
genes
will
provide
a
tool
with
which
to
dissect
the
mechanism
by
which
ABA
regulates
specific
gene
expression.
We
are
interested
in
determining
how
gene
expression
is
regulated
by
plant
hormones.
We
have
chosen
to
study
the
effect
of
ABA
on
gene
expression
in
rice
because,
as
we
demonstrate
here,
this
hormone
plays
a
central
role
in
seed
development
and
in
the
response
of
rice
plants
to
water-
stress,
two
important
agronomic
traits
(Chang
et
al.,
1986;
Seshu
and
Sorrells,
1986).
Knowledge
of
the
structure
and
function
of
ABA-responsive
proteins
will
aid
our
under-
standing
of
the
physiology
of
seed
maturation
and
of
drought
tolerance
in
cereals.
As
a
first
step,
we
have
isolated
several
cDNA
clones
whose
expression
is
induced
by
ABA
and
water-stress.
One
of
these
clones,
called
RAB
21
(for
Responsive
to
ABA),
was
fully
characterized.
We
present
here
the
sequence
of
this
novel
rice
gene,
and
a
characteriz-
ation
of
the
RAB
21
protein
product.
We
show
that
the
induction
of
the
RAB
21
by
ABA
and
water-stress
is
rapid
and
independent
of
de
novo
protein
synthesis.
©IRL
Press
Limited,
Oxford,
England
2279

J.Mundy
and
N.-H.Chua
Er
LUI
0
a
>-
0l
gn
0
M
z
LU
Lu
10
DAF
:
>-
0
m
0
2
z
LU
LU
20
DAF
_t
!
0
4
=
46-.
30--
Ce-
m
LU,
w~
cE
I
m
m
m
+
+
0t
W-
.e
*^6.
7
E
i4
26aX
14-
-A
..
NNW
1
2
3
4
5
6
B
0
m
U
7
8
NEpHGE
-
+
ABA
I
cn
0
*
~~~cn
0
-
0f
m
1
-D
'I
--
NEpHGE
-
B.A.
C/)
0
0
0)
M.
II.
Dl
-D
I
-
NEpHGE
HYBRID
SELECTION
0
,-
0W
co,~
Fig.
1.
Translation
products
of
developing
and
germinating
seed
mRNAs.
(A)
One-dimensional
SDS-PAGE
of
in
vitro
products
of
mRNAs
from
embryo
and
endosperm
half-seeds
harvested
10
days
after
flowering
(DAF)
(lanes
1
and
2),
embryo
and
endosperm
at
20
DAF
(lanes
3
and
4),
mature
embryo
at
30
DAF
(lane
5),
mature
embryo
germinated
3
days
and
then
incubated
12
h
without
(lane
6)
and
with
10
A1M
ABA
(lanes
7
and
8).
ABA-responsive
products
are
marked
on
the
far
left
with
dots;
23-,
21-
and
20-kd
polypeptides,
referred
to
in
panels
C
and
D,
are
marked
with
arrows
(lane
5).
Products
in
lanes
1-7
were
labelled
with
[35S]methionine,
those
in
lane
8
with
[35S]cysteine.
(B)
Two-dimensional
NepHGE/SDS-PAGE
of
[35S]methionine-labelled
translation
products
of
mRNAS
from
germinated
embryo
half-seeds
incubated
without
ABA
(as
in
A,
lane
6).
One-dimensional
gel
also
shown
at
left.
(C)
NEpHGE/SDS-PAGE
of
translation
products
of
mRNAs
from
germinated
embryo
half-
seeds
incubated
with
10
zM
ABA
(as
in
A,
lane
7).
(D)
NEpHGE/SDS-PAGE
of
translation
products
of
poly(A)
RNA
selected
by
hybridization
to
RAB
21
cDNA.
Mol.
wt
markers
are
indicated
at
the
left
in
kd.
Results
Isolation
of
an
ABA-inducible
cDNA
encoding
a
21-kd
polypeptide
As
a
first
step
toward
isolating
genes
from
rice
whose
expression
is
affected
by
ABA,
mRNA
populations
from
developing
rice
seeds
were
analyzed
by
in
vitro
translation
and
SDS
-PAGE
of
the
protein
products.
This
experiment
identified
mRNAs
encoding
prominent
polypeptides
of
mol.
wts
45,
39, 30,
25,
23,
21
and
20
kd
which
accumulate
late
during
rice
embryogenesis
(Figure
IA,
lanes
1-5,
see
dots).
These
mRNAs
are
long-lived
in
mature
grain
harvested
30
days
after
flowering
(DAF)
(Figure
lA,
lane
5)
and
disappear
completely
during
normal
germination
(Figure
lA,
lane
6).
However,their
accumulation
can
be
recapitulated
during
germination
by
a
12-h
incubation
with
ABA
(Figure
IA,
lane
7).
These
results
indicate
that
the
levels
of
mRNA
encoding
these
polypeptides
are
modulated
by
ABA.
Two-dimensional
separation
of
these
translation
products
revealed
that
the
23-,
21-
and
20-kd
polypeptides
are
com-
2280
A
kD
IL
in
1-D
C
kD
68-
A
C+_
t;)
-p
301-
14-I

ABA
and
water-stress
responsive
rice
gene
A
10
rO1-gal
fusion~
--
30
50
N-MEHQGQHGHV
TSRVDEYGNP
VGTGAGHGQM
GTAGMGTHGT
AGTGRQFQPM
REEHKTGGVL
70
90
110
QRSGSSSSSS
SEDDGMGGRR__KKGIKEKIKE
EKLPGGNKGEQ
QHAMGGTGTG
TGTGTGTGGA
130
150
163
YGQQGHGTGM
TTGTTGAHGT
TTTDTGEKKG
IMDKIKEKLP
GQH-C
r-
-IN
VITRO
----I
TOTAL
IMMUr
-PPT
<
~ct
<
CI:
m
m
m
m
B
<
<
<
<
I
+
I
+
kD
C
kD
r-
IN
;VfO
(+ABA)
_
TOT.;
D
LiU
0}
L
o
0
o
0
0
co
:D
z
IMMUNOBLOT
CL
i
tr
D
W
<
o
X
_J
Ir-
D
t)
CD
Z
68-
46-
30-
._.
14-
=
2
3
4
I
3
4
5
6
Fig.
2.
Structure
and
expression
of
the
RAB
21
protein.
(A)
Amino
acid
sequence
of
the
RAB
21
protein
deduced
from
the
ORF
of
the
cDNA
(see
Figure
5).
The
A-
and
B-type
sequence
repeats
are
underlined
with
solid
or
dashed
lines
respectively.
The
splice
site
used
in
constructing
the
,B-
galactosidase
fusion
proteins
is
marked
by
an
arrow.
(B)
Immunoprecipitation
of
in
vitro
synthesized
RAB
21
polypeptide.
Total
translation
products
of
mRNAs
from
embryo
half-seeds
germinated
3
days
and
then
incubated
without
(lane
1)
or
with
10
uM
ABA
(lane
2);
immunoprecipitates
of
these
products
with
antibodies
raised
against
'in
frame'
,B-galactosidase/RAB
21
fusion
protein
(lanes
3
and
4).
(C)
Immunodetection
of
RAB
21
protein
synthesized
in
vivo
in
leaves
sprayed
with
aqueous
100
uM
ABA
solutions.
Total
proteins
of
cytosolic,
organellar
and
nuclear
leaf
cell
fractions
were
silver-stained
(lanes
1-3),
or
electroblotted
onto
nitrocellulose
and
the
RAB
21
protein
was
then
detected
with
antibodies
raised
against
the
(3-galactosidase/RAB
21
fusion
protein
(lanes
4-6).
Mol.
wt
markers
are
indicated
at
the
left
in
kd.
prised
of
multiple
isoelectric
forms
(Figure
IC,
see
arrows).
The
even
spacing
of
spots
suggests
that
some
of
this
hetero-
geneity
may
be
due
to
serial
charge
differences,
a
common
electrophoretic
artefact.
Alternatively,
some
of
the
different
isoforms
may
be
products
of
different
genes
(see
Discussion).
Differential
labelling
experiments
suggest
that
the
23-,
21-
and
20-kd
polypeptides
are
rich
in
methionine
but
lacking
in
cysteine
(Figure
lA,
lanes
7
and
8).
A
cDNA
library
was
constructed
using
the
ABA-treated
seed
mRNA
as
template.
ABA-responsive
cDNA
clones
were
isolated
by
differential
screening
and
subsequent
North-
ern
blot
analysis.
One
clone,
called
pRAB
21,
was
chosen
for
further
characterization.
Figure
ID
shows
that
pRAB
21
hybridizes
to
mRNA(s)
that
encode
the
prominent
poly-
peptide(s)
of
apparent
mol.
wt
21 000.
At
the
low
hybridiz-
ation
stringency
shown
here,
mRNAs
encoding
the
23-
and
20-kd
polypeptides
are
also
hybrid-selected
(the
23-kd
group
is
very
faint
in
this
exposure).
At
higher
stringencies
of
hybrid-selection,
only
the
21-kd
polypeptides
are
seen
(not
shown).
This
cross-hybridization,
together
with
several
lines
of
evidence
discussed
later,
indicate
that
the
mRNAs
for
the
23-,
21-
and
20-kd
polypeptides
are
homologous.
RAB
21
is
a
basic,
glycine-rch
protein
that
accumulates
in
the
cytosol
of
ABA-treated
cells
The
amino
acid
sequence
of
RAB
21
was
determined
from
2281
68-
46-
A-
A-l

J.Mundy
and
N.-H.Chua
cr
LU
D
Q
A
L
LU
O
L-
's
flE
0
co
LLI
m
0
m
°
a:
<
<
E
z
E
z
mm
U
LU
LU
<
<
<
lOd
20d
>
I
+
_
*
1
2
43
4
5
6
7
ROOTS
SHOOTS
I
B
ABA ABA
NaCI
DRYING
0
6
12
24
0 6
12
24
6
12
24
6
12
hr
1
2
3
4
5
6
7
8
9
1011
12
13
Fig.
3.
(A)
Steady-state
RAB
21
mRNA
levels
in
rice
seeds,
leaves
and
roots.
15
,ug
of
total
RNA
was
used
per
lane.
Northern
blot
analysis
of
RAB
21
mRNA
levels
in
embryo
and
endosperm
half-seeds
harvested
10
DAF
(lanes
1
and
2),
embryo
and
endosperm
at
20
DAF
(lanes
3
and
4),
mature
embryo
at
30
DAF
(lane
5),
mature
embryo
germinated
3
days
and
the
incubated
12
h
without
(lane
6)
and
with
10
AM
ABA
(lane
7).
(B)
Accumulation
of
RAB
21
mRNA
in
roots
of
hydroponically
grown
rice
following
addition
of
10
AM
ABA
to
media
(lanes
1-4),
and
in
shoots
of
the
same
plants
following
addition
of
10
zM
ABA
to
media
(lanes
5-8),
addition
of
NaCl
(200
mM)
to
media
(lanes
9-11),
and
air-drying
the
whole
plants
(lanes
12,13).
Drying
was
accomplished
by
removing
plants
from
growth
media
and
air-drying
on
a
laboratory
bench
at
27°C
for
the
indicated
times.
Arrows
mark
the
single
major
transcript
of
850
bases.
the
nucleotide
sequence
of
the
RAB
21
cDNA
and
genomic
clones
(see
Figure
5).
The
RAB
21
open
reading
frame
(ORF)
(ATG
93
-TGA
665)
encodes
a
polypeptide
of
mol.
wt
16
529
(Figure
2A),
roughly
4500
smaller
than
the
mol.
wt
predicted
from
the
mobility
of
the
RAB
21
protein
on
SDS
-PAGE.
However,
several
lines
of
evidence
suggest
that
this
is
the
correct
sequence
of
the
RAB
21
protein.
First,
the
ORF
encodes
eight
methionines
and
no
cysteines,
con-
sistent
with
the
results
of
in
vitro
translation
experiments
with
labelled
cysteine
or
methionine
(Figure
lA).
Second,
the
ORF-encoded
polypeptide
is
basic
(approximate
pI
=
9.4),
in
keeping
with
mobility
of
the
RAB
21
polypeptide(s)
in
NEpHGE
(Figure
ID).
Third,
the
GC
content
of
the
ORF
is
high
(70%)
while
that
of
the
5'-
and
3'-untranslated
regions
is
low
(45%).
This
GC-codon
bias
has
been
noted
in
the
coding
regions
of
other
seed
protein
genes
(Rogers
et
al.,
1985).
To
obtain
direct
confirmatory
evidence,
anti-
bodies
were
raised
against
portions
of
the
RAB
21
protein.
Figure
2B
and
C
show
that
antibodies
against
a
j3-galacto-
sidase/RAB
21
fusion
protein
(RAB
21
residues
15-163)
specifically
recognize
in
vitro
and
in
vivo
synthesized
poly-
peptides
which
correspond
to
RAB
21.
Furthermore,
fusion
proteins
with
the
RAB
21
cDNA
in
all
three
reading
frames
were
produced
in
Escherichia
coli
and
analyzed
on
Western
blots
with
polyclonal
antibodies
raised
against
an
extract
of
total
soluble
proteins
from
mature
rice
seed.
Only
the
'in
[
+ABA--l
1
2
3
4
5
6
B-
-
+
+
INHIBS.
-
+
-
+
ABA
1
2
3
4
Fig.
4.
Steady-state
RAB
21
mRNA
levels
in
suspension
cells.
(A)
Time
course
of
accumulation
of
RAB
21
mRNA
in
suspension
cells
incubated
for
0
min,
15
min,
30
min,
1
h,
3
h and
6
h
with
10
/M
ABA
(lanes
1-6).
(B)
The
effect
of
pre-incubation
with
inhibitors
of
protein
synthesis.
RAB
21
mRNA
levels
after
6
h
incubations
without
(lane
1)
and
with
10
izM
ABA
(lane
2).
Lanes
3
and
4
show
the
same
incubations
following
a
1.5-h
pre-incubation
with
inhibitors
of
protein
synthesis
[cycloheximide
(10
MM),
anisomycin
(60
1M)
and
chloramphenicol
(100
/%M)].
The
arrows
mark
the
two
different
transcripts
which
accumulate.
The
size
of
the
smaller
transcript
is
850
bases.
frame'
fusion
was
immunoreactive
(results
not
shown),
demonstrating
conclusively
that
the
other
two
possible
reading
frames
are not
translated
in
vivo.
In
vitro
transcription/translation
of
RAB
21
cDNA
con-
structs
was
used
to
show
that
the
predicted
stop
codon
(TGA,
665)
terminates
the
RAB
21
ORF.
In
these
experiments,
ptzl9U
vectors
(USBC)
containing
either
the
NheI
-SacI
(57
-670)
fragment
which
terminates
at
the
expected
TGA,
or
the
NheI-NheI
(57-741)
fragment,
were
used
as
tem-
plates
to
produce
polypeptides
in
vitro.
These
two
polypep-
tides
had
identical
mobilities
in
SDS
-PAGE
(not
shown),
confirming
that
TGA
at
residue
665
is
used
as
the
stop
codon.
Analysis
of
the
RAB
21
protein
sequence
using
Harr
plots
revealed
that
the
protein
contains
a
duplicated
domain
struc-
ture.
Each
domain
contains
an
A
and
B
repeat
(Figure
2A).
The
A
and
B
repeats
are
adjacent
in
the
C-terminal
domain,
but
are
separated
in
the
N-terminal
domain
by
a
portion
of
the
ORF
containing
the
gene's
single
intron.
An
algorithm
which
predicts
the
secondary
structure
of
RAB
21
suggests
that
most
of
the
polypeptide,
including
the
A
repeats,
is
composed
of
random
turns
due
to
the
irregular
spacing
of
glycine
residues.
The
B-repeats
are
very
basic
and
contain
the
conserved
sequence
KIKEKLPG
which
may
be
part
of
a
helix
bounded
by
turns.
Further
computer
analysis
showed
weak
homology
between
the
entire
protein
(or
the
A
repeats
alone)
to
regions
of
several
viral
nuclear
proteins.
The
highest
homology
occurred
with
the
glycine
-alanine
co-
polymer
domain
of
the
major
nuclear
antigen
encoded
by
the
Epstein
-Barr
virus
(Hennessy
and
Kieff,
1983).
Not
surprisingly,
the
very
basic,
lysine-rich
B-repeats
are
weakly
homologous
to
various
DNA-binding
proteins
such
as
his-
tones.
These
results
suggest
that
the
RAB
21
polypeptide
may
be
a
nuclear
protein.
To
examine
this
possibility,
RAB
21
antibodies
were
used
to
probe
soluble,
organellar,
and
nuclear
fractions
prepared
from
leaves
of
ABA-treated
rice
plants.
These
experiments
show
that
the
RAB
21
protein
is
found
in
the
soluble
fraction
(Figure
2C).
Since
the
protein
has
not
been
detected
in
crude
organellar,
chloroplastic,
mitochondrial
or
nuclear
extracts,
we
suggest
that
RAB
21
is
a
cytosolic
protein.
However,
definitive
proof
of
its
cellular
location
remains
to
be
established
by
immuno-electron
microscopy.
2282

ABA
and
water-stress
responsive
rice
gene
A
_o
-)_-.
O
)
co
.0
.C
CU
coocu
_
_
oc
C
.0
x
x
co)
(-
coZC
o
Z
COZ
X
51
1
I
131
I
I
_
kB
1.0
2.0
3.0
4.0
B
TGCAG
1501
AGAGGATGACCCTTGTCACCACCGTCATGTAC(;AGGCTCCTTCACCACTGCCTCACTGCCACCAGCGTCTCCCG
CCGCGTGCAATACAAGAAGAAACATC
-
1401
GAACGGTCATATAAGGTAAGACCCACTACCGATTTAACCTATCATTCCCACAATCTAATCCACTTATTTCTCTTCCCATGATC
TTATC
CTCTCATTTCTC
-
13
01
CTCACTACTTTTG
CATTTGTAGGAAACACAATGACAC
CGTCGAAGAAAGCTGG
TGGAG CACCGTAG
CCAGCAATCAC
CAAAACACAGAGGGGAGGAGGTC
-
12
01
GG
CAGCGGC
CATG
CGGACGG
CGATGAGACAACGCGACGCAAAGAGGGAGGAGGACGTTGG
CGATCATG
CTGGTGTTG
G
CGGAGGAGGTCACTGG
CCATG
C
-
1101
GAATGACAG
CGGGGCAGCGCAACACAAAAAGGGGGGAGGATGCCGGCGACCACGCTAGTACCATGAAGCAAGATGATG
TGAAAGGGAGGACCGGACGAGG
-
10
01
GTTGGACCTCTGCCGCCGACGTGAAGACCGTGATGTGTAGAAGGAGATGTTAGACCAGATGCCGACGCAACTTAGCCCTG
CAAGTCACCCGACTGCATAT
-
901
CGCTGCTTGCCCTCGTCCTCATGTACACAATCAGCTTGCTTATCTCTCCATACTTGTCGTTTGTTTCCCGTGGC
CGAAATAGAAGAAGACAGAGGTGGGT
-
801
TTTGTTGGAGAGTTTTAGTGG
TATTGTAGG
CCTATTTGTAATTTTGTTGTACTTTATTGTAT-TAATCAATAAAG
GTGTTT
CATTCTATTTTGACTCAATG
-
701
TTGAATCCATTGATCTCTTGGTGTTG
CACTCAGTATGTTAGAATATTCATTC
CGTTGAAACAATCTTGGTTAAG
GGTTG
GAACATTTTTATCTGTTCGGT
-
601
GAAACATC
CGTAATATTTTCGTTGAAACAATTTTTATCCGACAG
CACCGTC
CAACAATTTACACCAATTTGGAC
GTGTGATACATAG
CAGTC
CC
CAAGTG
-
501
AAACTGAC
CACCAGTTGAAAGGTATACAAAGTGAACTTATTCATCTAAAAGAC
CG
CAGAGATGGG
CCGTGGC
CG
TGG
CTG
CGA!
ACGACAG
CGTTCAGG
C
-
401
C
CATGAGC
CATTTATTTTTTAAAAAAATATTTCAACAAAAAAGAGAACGGATAAAATC
CATCGAAAAAAAAAAACTTT
CCTACG
CATC
CTCTCCTATCTC
-
301
CATCCACGGCGAGCACTCATCCAAACCGTCCATCCACGCGCACAGTACACACACATAGTTATCGTCTCTCCCCCCGATGAGTCACCACCCGTGTCTTCGA
-
201
GAAACGCCTCGCCCGACACCGTACGTGCGCCAC
CGCCGCGCCTGCCGCCTGGACACGTCCGGCTCCTCTCCCGCCGCGCTGGCCACCGTCCACCGGCTCC
-
101
CGCACACGTCTCCCTGTCTCCCTCCACCCATGCCGTGGCAATCGAGCTCATCTCCTCGCCTCCTCCGGCTTATAAATGGCGGCCACCACCTTCACCTGCT
-
1
TG
CACAC
CACAG
CAAGAG
CTAAGTGAG
CTAG
CCAC
TGATCAGAAGAACAC
CTC
GATC
T
CTGAGAG
TGTTTTTTCAG
CTTTAG
CTTAAG
CAG
GATGGAsGCA
10
0
CCAGGGG
CAGCACGG
CCACGTGAC
CAG
CCGCGTCCACGAGTACCGCAACCCGGTCGGCAC
CGGCGCCCGGACACGCGCAGAkTGGG
CACCGCCGGCCATGGGG
20
0
ACGCACCGCACCGCCCGGCACCGGCCGCCAGTTC
CAGC
CCATGAGGGAGGAGCACAAGAC
CGGCGCCGTCCTGCAACGCTCCGGCAGCTC
CAGCTCAAGCT
30
0
CGgtacaacattt
tgacccccaat
tc
tttaccccccac
taaaaccttgcgtacaattcgttgaaaattttaatgtct
tgitgacag,TCTGAGGAiTGATGGA
400
ATGGGAGGGAGGAGGAAGAAGGGGATCAAGGAGAAGATCAAGGAGAAGCTCCCCGG
CGGCAACAAGGG
CGAG
CAGCCAG
CATGC
CATGGGCGGCACCGGCA
50
0
CCGGCAC
CGGCACCGGCACCGGAACCGGCGG
CG
CCTACGGG
CAGCAGGGCCACGGCAC
CGGGATGAC
CACCGG
CAC
CAC
CGG
CG
CACACGG
CACCAC
CAC
60
0
CACCGACACCGGCGAGAAGAAGGGCATCATGGACAAGATCAAGGAGAAGCTGCCCGGCCAGCACTGAG
CT
CGACACA
C
CAC
CACAC
CATGTGTG
C
CTG
CG
70
0
CCGGACGGC
CGCCAC
GTCAC
CTTC
CTGAATAATAAGATGAGCTAGCCGAGCGCAATAAAAGGAAAAAAAAATGTTAC
TGT
CGTGTGATGAGTGTGAATGT
80
0
GTGATGGCG
TTCT
CCAGTACGGAC
CGTGT
CTG
CGTG
TTGTTTGTACTGTGAAAGTACG
CTGTGTATGTACGTCGTATG
TGTACAATTC
CG
CTTATACTGT
900O
*poly
A+
..*poly
A+
TGTTGAAATACGTATAAATATGTGTACATTATACGGTGTATATCTCATCGTG
CATATGTACACAACGTTTTGGTGAT
CGTTATAATG
TTCATTTTTTTC
C
100
0
TTATTCTGATCAATCTGGATCATAGGAGCTC
c
EX-
ENS;ION
PRODUCT
(81
b)
1
-~~~~~~~~~mRNA
2
_
*
f--
-
-
*t
+
mRNA
3
+
S
1
NUCLEASE
4~~~~~~~~~~~~~~~~~~~~~~~~~~~10
..
If
SE
^tflf
w
QUENCE
f
J
(GATC)
%
tHOMOPOLYMER
G-TAIL
Fig.
5.
Structure
of
the
RAB
21
cDNA
and
gene.
(A)
Restriction
map
of
the
4.3-kb
BamHl
genomic
fragment
containing
the
RAB
21
gene.
The
map
was
generated
by
multiple
enzyme
restriction
analysis.
The
two
Sacl
sites
are
at
-52,
downstream
of
CAAT,
and
at
+670,
downstream
of
the
TGA
termlinating
the
ORF
(see
Figure
2B).
(B)
Nucleotide
sequence
of
the
RAB
21
gene.
The
2.4-kb
PstI-AoI
genomic
fragment
(see
A)
and
the
cDNA
were
sequenced
as
described
in
Materials
and
methods.
The
ORF
is
in
bold-faced
type,
the
single
intron
is
in
lower
case.
Putative
transcription
and
the
start
and
stop
codons
of
the
ORF
are
underlined.
Polyadenylation
sites
found
in
the
cDNAs
are
marked
with
an
asterisk.
Various
GC-rich
repeats
in
the
proximal
promoter
region,
including
five
sequences
which
resemble
the
consensus
binding
sites
of
the
SPI
animal
transcription
factor
(Briggs
et
al.,
1986)
are
discussed
in
Results.
(C)
Primer
extension
U-sing
a
synthetic
oligonucleotide
(+
84
to
+
108).
The
end-
labelled
oligo
was
used
as
a
primer
to
reverse
transcribe
without
RNA
(panel
1)
and
with
1
jog
of
poly(A)+
RNA
from
ABA-treated
half-seeds
without
(panel
2)
and
with
subsequent
incubation
with
SI
nuclease
(panel
3).
Panel
4
shows
the
5'
sequence
(GATC
top
to
bottom)
of
the
pRAB
21
cDNA
obtained
with
the
same
oligo
as
primer.
2283

J.Mundy
and
N.-H.Chua
Accumulation
of
RAB
21
mRNA
is
induced
in
different
tissues
by
ABA
and
water-stress
Northern
blot
analysis
was
used
to
analyze
the
steady-state
levels
of
RAB
21
mRNA
in
various
rice
tissues
during
different
treatments.
Experiments
with
total
RNA
from
developing
seed
tissues
(Figure
3A)
confirmed
the
patterns
of
RAB
21
mRNA
accumulation
as
assayed
by
in
vitro
translation
(Figure
1A).
During
seed
development,
low
levels
of
RAB
21
mRNA
are
found
in
both
endosperm
and
embryo
seed
halves
(Figure
3A,
lanes
1
-4).
High
levels
of
this
mRNA
accumulate
in
embryos
between
20
DAF
and
ma-
turity
(30
DAF)
and
survive
as
long-lived
mRNAs
in
the
resting
grain
(Figure
3A,
lane
5).
Homologous
signals
were
also
seen
among
long-lived
RNAs
extracted
from
mature
seeds
of
maize,
barley,
wheat
and
millet
(not
shown),
sug-
gesting
that
proteins
homologous
to
RAB
21
are
functionally
conserved
among
the
Graminae.
As
was
shown
for
barley
(Mundy
et
al.,
1986),
mature
rice
endosperm
does
not
contain
detectable
mRNA
levels.
Therefore
RAB
21
mRNA
was
not
assayed
in
this
tissue.
The
long-lived
RAB
21
mRNA
in
rice
embryos
is
rapidly
turned
over
at
the
onset
of
germination
(Figure
3A,
lane
6).
However,
accumulation
of
this
mRNA
can
be
recapitulated
by
incubating
the
germinating
embryos
in
10
AM
ABA
(Figure
3A,
lane
7).
This
indicates
that
ABA
affects
the
steady-state
level
of
RAB
21
mRNA.
RAB
21
is
not
present
at
detectable
levels
in
the
roots
and
shoots
of
hydroponically-grown,
1-month-old
rice
plants
(Figure
3B,
lanes
1
and
5).
However,
upon
addition
of
ABA
to
the
solution
bathing
the
roots,
the
mRNA
accumulates
in
both
organs
and
reaches
a
steady-state
level
after
12
h
(Figure
3B,
lanes
2-4,
6-
8).
At
this
time
the
level
of
RAB
21
mRNA
is
induced
at
least
20-fold
over
the
control.
Therefore
the
increase
of
the
RAB
21
mRNA
levels
in
the
presence
of
ABA
is
not
restricted
to
seed
tissues.
To
determine
if
the
expression
of
RAB
21
could
also
be
induced
by
water-stress,
NaCl
was
added
to
the
hydroponic
solution
to
a
final
concentration
of
200
mM.
Figure
3B
(lanes
9-11)
indicates
that
RAB
21
mRNA
levels
increase
in
response
to
prolonged
growth
in
solutions
containing
elevated
salt
concentrations.
The
time
course
of
induction
in
response
to
NaCl
is
similar
to
that
obtained
with
ABA;
the
maximal
level
of
accumulation
(20-
to
30-fold
induction)
being
at-
tained
after
12-24
h.
The
pattern
of
RAB
21
mRNA
accumulation
was
unchanged
when
both
ABA
and
salt
were
added
to
the
hydroponic
solution.
RAB
21
mRNA
levels
are
also
induced
by
desiccation
of
rice
plants
(Figure
3B,
lanes
12,
13).
These
results
strongly
suggest
that
the
accumulation
of
RAB
21
mRNA
is
regulated
by
the
water
status
of
the
plant.
Detectable
RAB
21
mRNA
accumulation
in
whole
rice
plants
generally
requires
3-4
h
of
ABA
treatment.
This
long
response
time
may
reflect
a
low
rate
of
exogenous
hormone
uptake
by
roots
and
translocation
to
the
leaves.
To
examine
this
possibility,
RAB
21
mRNA
accumulation
was
measured
in
cultured
suspension
cells
derived
from
embryogenic
calli.
In
contrast
to
whole
plants,
the
increased
RAB
21
mRNA
levels
are
easily
detected
in
suspension
culture
cells
after
only
15
min
of
ABA
treatment
(Figure
4,
lanes
1
and
2).
During
prolonged
incubation
with
ABA,
RAB
21
mRNA
levels
increase
steadily
to
a
maximum
of
-20
times
the
control
level
at
3-6
h
(Figure
4,
lanes
3-6).
The
rapidity
of
the
ABA
response
in
cell
cultures
suggests
that
protein
synthesis
is
not
required
for
RAB
21
gene
expression.
To
obtain
direct
evidence
on
this
point,
cells
were
treated
with
a
combination
of
protein
synthesis
inhibit-
ors
affecting
70S
and
80S
ribosomes
which
reduced
total
protein
synthesis
by
>90%.
The
protein
synthesis
inhibitors
neither
induce
the
expression
of
RAB
21
(Figure
4,
lanes
1
and
3)
nor
block
its
induction
by
ABA
(lanes
2
and
4).
A
longer
transcript
(1000-1200
bases)
that
hybridizes
to
the
RAB
21
cDNA
probe
accumulates
in
the
presence
of
the
inhibitors.
These
results
have
also
been
seen
in
leaf
tissues
treated
with
ABA
plus
inhibitors
(results
not
shown).
This
longer
transcript
has
not
yet
been
characterized
but
it
may
correspond
to
the
primary
transcript
of
a
homologous
gene
which
contains
an
intron
of
300-400
bp.
Structure
of
RAB
21
gene
The
nucleotide
sequence
of
the
pRAB
21
cDNA
was
used
to
deduce
the
primary
structure
of
the
encoded
polypeptide.
The
clone
was
also
used
to
isolate
a
corresponding
genomic
clone,
gRAB
211.
Analysis
of
sequences
upstream
of
the
RAB
21
gene's
coding
region
identified
putative
regulatory
elements.
The
restriction
map
and
nucleotide
sequence
of
the
cDNA
and
the
corresponding
region
of
the
genomic
clone
(2.5-kb
restriction
fragment)
are
presented
in
Figure
5A
and
B.
The
major
ORF
of
the
transcribed region
is
489
bp
(93
-665)
encoding
the
162
amino
acids
of
RAB
21
(Figure
2A).
Flanking
the
ORF
is
a
92-bp
5'
leader
containing
three
stop
codons
and
a
261-bp
3'
tail
containing
the
putative
polyadenylation
sequence
ATAAA
12
bp upstream
of
the
site
of
poly(A)
addition.
The
3'-non-coding
regions
of
six
homologous
cDNAs
were
sequenced.
Five
of
them
were
identical
to
pRAB
21,
being
polyadenylated
at
position
+
926.
This
correlates
well
with
the
size
of
the
RAB
21
mRNA
(800
nucleotides).
The
sequence
of
the
sixth
clone
was
identical
to
that
of
the
genomic
DNA
but
contained
a
poly(A)+
tail
farther
downstream
at
position
+991.
The
original
824-bp
cDNA
was
shown
by
primer
extension
to
lack
only
18
nucleotides
from
the
5'-untranslated
leader
(Figure
SC).
The
transcription
start
site
is
shown
in
Figure
2B
(nucleotide
+
1).
In
the
genomic
DNA,
the
ORF
is
interrupted
by
a
83-bp,
AT-rich
intron
flanked
by
the
consensus
border
sequences
GT
and
AG.
The
sequence
of
gRAB
211
was
identical
to
the
cDNA RAB
21,
indicating
that
this
gene
is
transcribed
in
vivo.
The
proximal
GC-rich
promoter
(-200
to
-1)
contains
a
putative
TATAA
box
(-30)
and
a
putative
CAAT
box
(-62).
This
region
contains
the
following
four
types
of
GC-rich
repeats
(see
Figure
SB):
(1)
TGCGCCACCG
at
-175
and
-
121;
(2)
CGCCGCGC
at
-
167
and
-129;
(3)
TCCGGCTCC
at
-143,
-108
and
-37;
(4)
GTC-
TCCCT
at
-93
and
-85.
This
region
also
contains
five
other
GC-rich
repeats
at
-200,
-195,
-166,
-133
and
-48
whose
opposite
strand
sequence
shows
80%
homology
to
the
decanucleotide
(G/
TG/AGGCGG/TG/AG/AC/T)
binding
site
of
the
SPI
tran-
scription
factor
(Briggs
et
al.,
1986).
Discussion
We
are
interested
in
studying
the
molecular
mechanism
of
action
of
plant
hormones
on
gene
expression.
To
begin
this
work
we
isolated
a
cDNA
encoding
a
major
transcript
that
2284

ABA
and
water-stress
responsive
rice
gene
is
inducible
in
rice
tissues
by
the
plant
hormone
ABA.
This
cDNA
was
shown
by
hybridization
and
immunoassay
to
en-
code
a
prominent
member
of
a
group
of
basic
polypeptides
of
mol.
wt
23-20
kd.
We
call
this
protein
RAB
21.
The
different
RAB
polypeptides
may
be
post-translational
modifi-
cations
of
a
single
gene
product
or
products
of
closely
related
genes.
The
rice
genome
contains
at
least
three
closely-linked
(within
30
bp)
genes
homologous
to
RAB
21
(K.Yamaguchi-
Shinozaki,
unpublished
results).
These
genes
encode
pro-
teins
of
slightly
different
amino
acid
sequence
that
may
ac-
count
for
the
different
groups
of
polypeptides
immunoreactive
to
RAB
21
antibodies.
Northern
blot
hybridizations
show
that
RAB
21
gene
expression
is
not
tissue
specific,
as
shown
by
the
accumu-
lation
of
its
transcript
in
seeds,
roots,
leaves
and
in
un-
differentiated
suspension
cells.
Since
there
at
least
three
other
rice
genes
closely
related
to
RAB
21,
gene-specific
probes
are
needed
to
ascertain
whether
these
genes
are
differentially
expressed.
Physiological
experiments
show
that
RAB
21
mRNA
accumulates
not
only
in
ABA-treated
tissues
but
also
in
leaves,
roots
and
suspension
cells
under
conditions
of
water
deficit.
These
results
suggest
that
RAB
21
gene
ex-
pression
is
dependent
upon
the
water
status
of
plants
and
that
ABA
may
act
as
a
signal
in
this
response.
The
pathway
of
this
response
is
different
to
that
mediating
the
more
general
heat
shock
response
(Heikkila
et
al.,
1984),
because
RAB
21
mRNA
is
undetectable
in
rice
tissues
after
heat
shock
for
various
periods
of
time
(not
shown).
Experiments
with
cultured
cells
show
that
accumulation
of
RAB
21
mnRNA
following
ABA
treatment
is
very
rapid
and
that
it
is
insensitive
to
inhibitors
of
protein
synthesis.
These
data
strongly
suggest
that
ABA-induced
gene
ex-
pression
does
not
require
protein
synthesis
but
probably
involves
modification
of
pre-existing
factors,
as
is
the
case
for the
heat
shock
response
(Zimarino
and
Wu,
1987).
NaCl
also
induces
RAB
21
mRNA
rapidly
in
cultured
cells
(not
shown).
This
response
is
not
additive
to
that
attributed
to
ABA:
addition
of
NaCl
and
hormone
together
does
not
'superinduce'
mRNA
accumulation
at
any
point
in
the
time
course.
These
results
corroborate
the
findings
of
physio-
logical
(Jones
et
al.,
1986)
and
genetic
(Chandler
et
al.,
1988)
studies
which
show
that
the
response
of
plants
to
water-stress
is
mediated
by
ABA
at
the
level
of
specific
gene
expression.
To
initiate
studies
on
the
molecular
mechanism
of
ABA
action,
the
gene
encoding
RAB
21
was
isolated
and
its
nucleotide
sequence
determined.
This
is
the
first
published
genomic
sequence
of
a
strongly
ABA-responsive
gene.
The
proximal
promoter
region
is
GC-rich
and
contains
numerous
repeats
detailed
in
Results.
Another
group
of
repeats
is
closely
related
to
to
the
GC
element
found
in
various
cellular
and
viral
genes
in
mammalian
cells.
This
cis-acting
element
promotes
the
expression
of
genes
by
binding
the
trans-acting
protein
factor
SPI
(Briggs
et
al.,
1986).
Similar
sequences
have been
noted
in
the
promoter
of
the
oa-subunit
of
3-
conglycinin
(Chen
et
al.,
1986)
and
in
that
of
oat
phyto-
chrome
(Hershey
et
al.,
1987),
plant
genes
which
are
not
known
to
be
responsive
to
ABA
treatment.
The
regulatory
roles
of
these
different
GC-rich
repeats
remain
to
be
estab-
lished
by
functional
assays.
These
experiments,
now
in
progress
in
our
laboratory,
may
elucidate
the
molecular
mechanism
by
which
ABA
regulates
gene
expression
and
mediates
the
adaptation
of
plants
to
water-stress.
Materials
and
methods
Plant
materials
Seeds
of
rice
(Oryza
sativa,
var.
Indica,
cv.
IR
36)
were
obtained
from
the
International
Rice
Research
Institute,
Philippines.
Plants
for
developmen-
tal
studies
were
grown
in
soil
at
27°C
and
a
day
length
of
11
h.
Plants
flowered
10-11
weeks
after
planting.
Seeds
were
then
collected
after
10
days
(milk
stage),
20
days
(starchy,
green
pericarp)
and
30
days
(dry,
brown
pericarp,
mature).
Plants
were
also
grown
hydroponically
in
Hoagland's
solution
supplemented
with
20
mM
NH4NO3.
Preparation
and
treatment
of
embryo-containing
half-seeds
at
27°C
was
performed
as
described
previously
for
barley
(Mundy
et
al.,
1986).
Suspension
cells
derived
from
embryonic
callus
of
IR
36
were
grown
at
25'C
in
Kao's
medium
(Kao,
1977)
containing
2.5
mg/l
2-4
D
and
0.2
mg/l
kinetin
with
subculturing
at
1-week
intervals.
Cells
for
mRNA
isolation
were
subcultured
5-7
days
prior
to
harvest.
Protein
analysis
Proteins
were
analyzed
by
NEpHGE,
SDS-PAGE
and
Western
blots
according
to
Tingey
et
al.
(1987).
Isolation
of
mRNA,
in
vitro
translation
using
reticulocyte
lysate
and
immunoprecipitation
with
Protein
A
Sepharose
4B
were
as
described
previously
(Mundy
et
al.,
1986).
Leaf
cell
fractions
were
prepared
from
7-day-old
plantlets
sprayed
three
times
with
100
jtM
aqueous
ABA
solutions
during
the
24
h
prior
to
harvest.
Chloroplasts
and
mitochondria
were
prepared
from
leaf
tissue
after
Boutry
and
Chua
(1985)
while
nuclei
were
prepared
after
Green
et
al.
(1987).
The
National
Bio-
medical
Research
Foundation
Protein
Sequence
Databank
carried
in
the
Rockefeller
University
7000/40
computer
was
screened
for
sequences
related
to
the
RAB
21
ORF
with
the
SEARCH
program
of
Dayhoff
et
al.
(1983).
Detailed
comparisons
of
protein
sequences
thought
to
be
related
to
the
RAB
21
sequence
were
made
with
ALIGN,
utilizing
the
mutation
data
matrix
[250
PAMs
=
6,
and
a
gap
penalty
of
8
(Dayhoff
et
al.,
1983)].
Isolation
of
cDNA
and
genomic
clones
Double-stranded
cDNA
synthesized
by
the
RNase
H
method
(Gubler
and
Hoffman,
1983)
was
size-fractionated
on
a
column
of
Bio-gel
ASOm
(Bio-
Rad
Laboratories).
Molecules
of
450-4500
bp
were
then
cloned
by
homo-
polymer
GC-tailing
into
pEMBL
12
plasmid
(Dente
et
al.,
1983).
Six
thousand
recombinant
clones,
replica-transferred
onto
nitrocellulose
filters
from
96-well
microtiter
plates,
were
screened
with
single-stranded
cDNA
probes
synthesized
from
mRNA
isolated
from
control
and
ABA-treated
half-
seed
mRNAs.
Hybridizations
were
performed
in
50%
formamide,
6
x
SSC,
1
x
Denhardt's,
0.1
%
SDS,
100
pgg/ml
denatured
salmon
sperm
DNA
at
42°C
with
denatured
32P-labelled
DNA
probes
(sp.
act.
1
x
108
c.p.m./
,umg
DNA,
final
concentration
1
x
106
c.p.m./ml).
Sixty
ABA-responsive
clones
were
then
tested
on
Northern
blots
of
the
same
RNAs.
Hybridiz-
ation
selection
after
Tingey
et
al.
(1987)
was
used
to
identify
polypeptides
encoded
by
specific
clones.
Standard
protocols
and
conditions
were
used
for
agarose
electrophoresis,
hybridization
in
50%
formamide,
DNA
fragment
isolation
with
DE-81
paper,
and
plasmid
DNA
preparation
(Maniatis
et
al.,
1982).
Genomic
DNA
was
isolated
from
10-day-old
etiolated
leaves
by
CsCl
centrifugation
(Maniatis
et
al.,
1982).
Southern
hybridization
of
restricted
DNA
revealed
a
4.3-kb
XbaI
fragment
which
hybridized
strongly
to
the
selected
RAB
21
cDNA.
This
genomic
fragment
was
partially
purified
by
size
fractionation
in
agarose
gels
and
then
cloned
by
insertion
into
XbaI-
digested
lambda
ZAP
(Stratagene).
A
total
of
75
000
recombinant
plaques
were
screened
on
duplicate
filters
and
three
clones
containing
identical
4.3-kb
XbaI
inserts
were
identified
by
hybridization
to
pRAB
21.
Excision
and
recircularization
of
pBLUESCRIPT
SK(ml3)
plasmid
following
super-
infection
of
lambda
ZAP
infected
cells
with
IR
408
helper
phage
(Russel
et
al.,
1986)
was
performed
according
to
the
manufacturer's
instructions
(Stratagene).
DNA
sequencing
and
primer
extension
The
pRAB
21
cDNA
insert
and
a
2.5-kb
PstI-AoI
genomic
fragment
taining
the
RAB
21
sequences
were
sequenced
from
overlapping
deletions
created
by
Bal31
exonuclease
(Misra,
1985).
Single-stranded
templates
were
prepared
by
superinfecting
pEMBL
containing
recombinants
with
the
IR
408
helper
phage
(Russel
et
al.,
1986).
Sequencing
reactions
were
performed
according
to
Biggin
et
al.
(1983)
and
the
products
separated
on
6%
poly-
acrylamide/urea
gels.
Inosine
was
used
to
resolve
GC-compressions.
More
than
90%
of
the
sequence
was
obtained
for
both
strands
of
DNA.
At
least
two
overlapping
clones
were
used
when
only
one
strand
was
sequenced.
A
24-base
oligonucleotide
corresponding
to
5'
sequences
of
pRAB
21
was
synthesized
on
an
Applied
Biosystems
model
380A
DNA
synthesizer
2285

J.Mundy
and
N.-H.Chua
after
the
manufacturer's
instructions.
The
gel-purified
oligonucleotide
was
used
for
primer
extension
according
to
Shelness
and
Williams
(1984).
Fusion
protein
and
antibody
production
The
SalI
fragment
of
RAB
21
was
ligated
into
the
SalI
site
of
expression
vectors
pUR
288
(in
frame)
and
pUR
278
and
289
(out
of
frame
controls,
Ruther
and
Muller-Hill,
1983).
The
pUR
288
fusion
plasmid
encodes
amino
acids
15-
163
of
the
RAB
21
ORF
fused
to
the
C
terminus
of
E.coli
,B-galactosidase.
Fusion
proteins
purified
from
cell
extracts
on
anti-,B-
galactosidase-Sepharose
columns
according
to
the
manufacturer's
instruc-
tions
(Promega
Biotech)
were
used
for
immunization
of
rabbits.
RNA
blot
analysis
Total
RNA
was
prepared
by
a
miniprep
procedure
(Nagy
et
al.,
1988).
RNAs
were
separated
in
formaldehyde
gels,
blotted
onto
nitrocellulose
and
hybridized
to
random-primed
cDNA
probes
after
standard
protocols
(Maniatis
et
al.,
1982).
Replicate
gels
were
stained
with
ethidium
bromide
to
ensure
that
samples
contained
approximately
equal
amounts
of
rRNA.
A
commercial
RNA
ladder
was
used
as
size
marker
(BRL).
Acknowledgements
We
thank
Irene
Roberson
for
excellent technical
help
and
Dr
Brian
Keith
for
expert
advice.
J.M.
is
on
leave
from
the
Department
of
Biotechnology,
Carlsberg
Research
Laboratory,
Copenhagen,
Denmark.
This
work
was
supported
by
a
grant
from
the
Rockefeller
Foundation.
Raikel,N.V.
and
Wilkins,T.A.
(1987)
Proc.
Natl.
Acad.
Sci.
USA,
84,
6745-6749.
Ramagopal,S.
(1987)
Proc.
Natl.
Acad.
Sci.
USA,
84,
94-98.
Richardson,M.,
Valdez-Rodriguez,S.
and
Blanco-Labra,A.
(1987)
Nature,
327,
432-434.
Rogers,J.C.,
Dean,D.
and
Heck,G.R.
(1985)
Proc.
Natl.
Acad.
Sci.
USA,
82,
6512-6516.
Russel,M.,
Kidd,S.
and
Kelley,M.R.
(1986)
Gene,
45,
333-338.
Ruther,U.
and
Muller-Hill,B.
(1983)
EMBO
J.,
2,
1791-1794.
Seshu,D.V.
and
Sorrells,M.E.
(1986)
In
Rice
Genetics.
Proceedings
of
the
International
Rice
Genetics
Symposium,
I.R.R.I.,
pp.
369-382.
Shelness,G.S.
and
Williams,D.L.
(1984)
J.
Bio.
Chem.,
259,
9929-9935.
Singh,N.K.,
Braker,C.A.,
Hasegawa,P.M.,
Handa,A.K.,
Buckel,S.,
Hermodson,M.A.,
Pfankuch,E.,
Regnier,F.E.
and
Bressan,R.A.
(1987)
Plant
Physiol.,
85,
529-536.
Suzuki,Y.,
Kurogochi,S.,
Murofushi,N.,
Ota,Y.
and
Takahashi,N.
(1981)
Plant
Cell
Physiol.,
22,
1085-
1093.
Tingey,S.V.,
Walker,E.L.
and
Corruzi,G.M.
(1987)
EMBO
J.,
6,
1-9.
Zimarino,V.
and
Wu,C.
(1987)
Nature,
237,
727-720.
Received
on
March
21,
1988;
revised
on
April
25,
1988
Note
added
in
proof
These
sequence
data
will
appear
in
the
EMBL/GenBank/DDBJ
Sequence
Databases
under
the
accession
number
Y00842.
References
Biggin,M.D.,
Gibson,T.J.
and
Hong,G.F.
(1983)
Proc.
Natl.
Acad.
Sci.
USA,
80,
3963-3965.
Boutry,M.
and
Chua,N.-H.
(1985)
EMBO
J.,
4,
2159-2165.
Bray,E.A.
and
Beachy,R.N.
(1985)
Plant
Physiol.,
79,
746-750.
Briggs,M.R.,
Kadonaga,J.T.,
Bell,S.P.
and
Tjian,R.
(1986)
Science,
234,
47-52.
Chandler,P.M.,
Walker-Simmons,M.,
King,R.W.,
Crouch,M.
and
Close,
T.J.
(1988)
J.
Cell.
Biol.,
Suppi.
12C.
Abs.
#L033,
143.
Chang,T.T.,
Armenta-Soto,J.L.,
Mao,C.X.,
Peiris,R.
and
Loresto,G.C.
(1986)
In
Rice
Genetics.
Proceedings
of
the
International
Rice
Genetics
Symposium,
I.R.R.I.,
pp.
387-398.
Chen,Z.-L.,
Schuler,M.A.
and
Beachy,R.N.
(1986)
Proc.
Natl.
Acad.
Sci.
USA,
83,
8560-8564.
Dayhoff,M.O.,
Barker,W.C.
and
Hunt,L.T.
(1983)
Methods
Enzymol.,
91,
524-545.
Dente,L.,
Cesarini,G.
and
Cortese,R.
(1983)
Nucleic
Acids
Res.,
11,
1645-1655.
Finklestein,R.R.
and
Crouch,M.L.
(1986)
Plant
Physiol.,
81,
907-912.
Finklestein,R.R.,
Tenbarge,K.M.,
Shumway,J.E.
and
Crouch,M.L.
(1985)
Plant
Physiol.,
78,
630-636.
Galau,G.A.,
Hughes,D.W.
and
Dure,L.
(1986)
Plant
Mol.
Biol.,
7,
155-170.
Green,P.J.,
Kay,S.A.
and
Chua,N.-H.
(1987)
EMBO
J.,
6,
2543-2549.
Gubler,U.
and
Hoffman,B.
(1983)
Gene,
25,
263-269.
Heikkila,J.J.,
Papp,J.E.T.,
Schultz,G.A.
and
Bewley,J.D.
(1984)
Plant
Physiol.,
76,
270-274.
Hennessy,K.
and
Kieff,E.
(1983)
Proc.
Natl.
Acad.
Sci.
USA,
80,
5665-
5669.
Henson,I.E.
(1984)
Ann.
Bot.,
54,
569-582.
Hershey,H.P.,
Barker,R.F.,
Idler,K.B.,
Murray,M.G.
and
Quail,P.H.
(1987)
Gene,
61,
339-348.
Hornberg,C.
and
Weiler,E.W.
(1984)
Nature,
310,
321-324.
Jacobsen,J.V.
and
Beach,L.R.
(1985)
Nature,
316,
275-277.
Jones,H.,
Leigh,R.A.,
Tomos,A.D.
and
Jones,R.G.W.
(1987)
Planta,
170,
257-.262.
Kao,K.N.
(1977)
Mol.
Gen.
Genet.,
150,
225-230.
King,R.W.
(1976)
Planta,
132,
43-51.
Litts,J.C.,
Colwell,G.W.,
Chakerian,R.L.
and
Quatrano,R.S.
(1987)
Nucleic
Acids
Res.,
15,
3607-3618.
Maniatis,T.,
Fritsch,E.F.
and
Sambrook,J.
(1982)
Molecular
Cloning.
A
Laboratory
Manual.
Cold
Spring
Harbor
Laboratory
Press,
Cold
Spring
Harbor,
NY.
Misra,T.K.
(1985)
Gene,
34,
263-268.
Mozer,T.J.
(1980)
Cell,
20,
479-485.
Mundy,J.,
Hejgaard,J.,
Hansen,A.,
Hallgren,L.,
Jorgensen,K.G.
and
Munck,L.
(1986)
Plant
Physiol.,
81,
630-626.
Nagy,F.,
Kay,S.A.
and
Chua,N.-H.
(1988)
In
Plant
Molecular
Biology
Manual.
Martinus
Nijhoff,
Dordrecht,
in
press.
2286