Plant Molecular Biology 28: 145-153, 1995.
© 1995 Kluwer Academic Publishers. Printed in Belgium.
145
A GCC element and a G-box motif participate in ethylene-induced
expression of the PRB-lb gene
Guido Sessa, Yael Meller and Robert Fluhr *
Department of Plant Genetics, P.O. Box 26, Weizmann Institute of Science, Rehovot, 76100, Israel
(* author for correspondence)
Received 8 December 1994; accepted in revised form 28 February 1995
Key words: ethylene, G-box, GCC element, nuclear DNA-binding protein, pathogenesis-related
proteins, transgenic plant
Abstract
The PRB-lb gene codes for a basic-type pathogenesis-related protein and is activated at the transcriptional level by the plant hormone ethylene. To identify cis-acting DNA elements essential for ethylene
induction, deleted and mutant forms of the PRB-1 b promoter, fused to the fl-glucuronidase (GU S) coding
region, were introduced in transgenic tobacco plants. A 73 bp fragment (X1 region) of the PRB-lb
promoter, located between positions - 2 1 3 and -141, was sufficient to confer ethylene responsiveness
to the reporter gene. The X1 region contains a TAAGAGCCGCC motif (GCC-box) well conserved in
several ethylene-inducible genes. A substitution mutation in this sequence, in the context of a 213 bp
PRB-lb promoter, completely abolished ethylene induction in transgenic tobacco, defining this conserved
motif as part of a cis-acting element responsive to ethylene. Three other mutations in the X 1 region caused
a pronounced decrease in the PRB-lb promoter activity in transgenic plants, but did not affect ethylene inducibility. One of them, localized in a G-box like motif (CACGTG), disrupted the binding site for
a nuclear factor, as observed in gel-shift analysis. Interestingly, the mobility of the complex formed on
the G-box element was dependent on its phosphorylation state. These results suggest that a cis-acting
element involved in the perception of the ethylene signal resides in a GCC motif and acts in concert with
additional elements in the regulation of ethylene-induced PRB-lb expression.
Introduction
The plant hormone ethylene affects development
and mediates in part plant responses to environmental stress [ 1]. Its biosynthetic pathway is well
characterized [20] and steps in the signal transduction and perception of this hormone have
begun to be elucidated [6, 21, 22, 29, 30]. A
number of mutants in ethylene biosynthesis and
response have been isolated from tomato and
Arabidopsis [21]. The characterization of two response mutants from Arabidopsis allowed the isolation of the ETRI and CTRI genes, which participate in different steps of the ethylene signal
transduction pathway [6, 22]. ETR1 encodes for
a putative transmembrane protein kinase and
shows similarity to prokaryotic signal transducers, known as the two component system [6].
CTR1 is a negative regulator of the ethylene response pathway and encodes for a Raf-like pro-
146
tein kinase [22]. Pathogenesis-related protein
accumulation in response to ethylene has been
shown to require the presence of calcium [29]
and to be mediated by phosphorylation events
[30].
The expression of several genes is regulated by
ethylene at the transcriptional level [4, 9, 12, 18,
26, 40, 41]. Regulatory regions necessary for ethylene induction and their interactions with nuclear
factors have been characterized in the promoters
of some genes expressed during plant development and in the pathogenesis response. Sequences required for ethylene response were defined by stable and transient expression studies in
the tomato fruit ripening genes E4 and E8 [26, 9]
and in the carnation glutathione S-transferase
(GST1) gene [ 18]. The ethylene responsive region
of the E4 gene was shown to bind nuclear factors
in an ethylene- and stage-dependent manner [26].
Promoter regions necessary for ethylene responsiveness have been identified by deletion analysis
in pathogenesis-related genes including a bean
chitinase [4,31] and tobacco fl-l,3-glucanase
genes [40, 41 ]. Moreover, a 61 bp element of the
fl-l,3-glucanase GLB promoter behaves as a constitutive enhancer in a transient expression assay
and interacts with a binding activity regulated by
ethylene [ 16]. We have recently characterized the
ethylene- and dark-induced expression pattern of
the PRB-lb gene, which encodes a basic-type
pathogenesis related protein [ 11, 35]. We chose it
as a model system to elucidate the molecular
events responsible for transcription regulation by
ethylene [12, 24]. Analysis of the PRB-lb promoter in transgenic tobacco plants revealed that
deletion of a fragment of 71 bp (X1 fragment),
between position -213 and -142, abrogates
ethylene-induced accumulation of a fl-glucuronidase reporter gene [24]. By gel-shift analysis of
the -213 bp promoter and of oligonucleotides
from parts of the X1 region, different sequencespecific protein-DNA complexes were observed
[24]. In this study we show that the X1 fragment
is sufficient to confer ethylene inducibility to a
reporter gene in transgenic tobacco. Furthermore,
point mutations in the X1 region affected in vivo
the ethylene inducibility of a reporter gene driven
by the -213 promoter. The same mutations altered in vitro complex formation between nuclear
proteins and the PRB-lb promoter.
Materials and methods
Constructs preparation and plant transformations
Constructs containing 213 bp (minimal ethyleneinducible promoter) and 67 bp (TATA box construct) of the PRB-lb promoter, fused to the
fl-glucuronidase (GUS) reporter gene, were prepared as previously described [24]. The mutant
forms of the -213 minimal promoter, Gml,
Gm2, Yml and Ym2 (Fig. 1), were obtained by
site-directed mutagenesis utilizing synthetic oligonucleotides as follows: for Gml, 5'-TTATCTCAATTGATGTGA-3', for Gm2, 5'-GACTAATGGGTTAACTTATCTC-3'; for Yml, 5'CGTGATGTCAACTTGAAATT-3'; for Ym2,
5' -GTGACATTACTACATCTTGACTTT-3 '
The mutant promoters were inserted as Xba IBarn HI fragments upstream of the B-glucuronidase reporter gene (Bam HI-Eco RI sites) in a
BlueScript plasmid (Stratagene). The X1 construct was obtained by deletion of the region be-
Gin2
- - -GTTAA ...............
Gml
................
G
AT .....
.......................
Ym2
......... ACT-CATC
Yml
---C-AC ...................
y
-213 A G T A T G A C T A A ~
.........
..........................
~C~TGTGACATTGAAATT
CTTTGAC TTTA
-154 A C A C T A A T G T C A T A T G C T T T C A A A T T A A T A A T C C G A T A A A G T C T G C T A A C A T G T G A C T T
-95 T C C A A T T T T T T T C T T T T A C A A A T T G C A G A
-36 A T C C A T A C T A T T C C T T G T T T C T C A C C A A A A C C C A A A A T G
ATTCCCTATTAAAACCC
+I
Fig. 1. Nucleotide sequence of the minimal ethylene-inducible
-213 P R B - l b promoter. Numbering corresponds to nucleotide positions relative to the translation initiation site. Oligonucleotide G and Y and their mutated forms Gml, Gm2, Yml
and Ym2 are indicated above the sequence. The consensus
core for a putative ethylene-responsive element (GCC element) and a G-box-like motif are boxed. The X 1 region, spanning from position -213 to -141, the putative TATA box and
the translation initiation codon are underlined. The transcription start site is marked by an asterisk.
147
tween positions -140 (Nde I site) and -68 (engineered Bgl II site) from the -213 bp promoterGUS construct (Fig. 1). Constructs were then
subcloned in the binary vector pGA492 at the
Xba I site [2], transferred by electroporation to
Agrobacterium tumefaciens, strain EHA101, and
utilized to transform Nicotiana tabacum cv. Samsun NN plants by cocultivation.
begun by the addition of total nuclear extract
(10 #g) to the binding mix. After 30 min incubation at room temperature, reactions were loaded
on a prerun 5 ~o acrylamide/bisacrylamide gel in
glycine buffer (40 mM Tris-HC1 pH 8.5, 195 mM
glycine). After 2 h electrophoresis (18 V/cm), the
gel was dried and exposed to X-ray film.
Results
Ethylene treatment
For ethylene treatment, a constant stream of air
with 20 ppm ethylene was applied in a sealed glass
box containing the potted plants. Light was provided by a mixture of 'cool white' and 'Grolux'
fluorescent lamps (25-30 #E m "2 s-l).
Analysis of GUS activity
Leaf disks were homogenized in 200 #1 of GUS
lysis buffer containing 50 mM sodium phosphate
buffer, pH7.0, 10mM EDTA and 10mM
2-mercaptoethanol. The extracts were tested fluorometrically for GUS activity using the substrate
4-methylumbelliferyl glucuronide (MUG), as described by Jefferson et al. [ 19].
Extraction of nuclear proteins from tobacco leaves
and gel shift assay
Nuclear proteins from tobacco leaves of ethylene
treated or untreated plants, were extracted as previously described [24]. Solutions utilized for
preparation of extracts in the presence of phosphatase inhibitors contained 0.2 mM NaVanadate, 10 mM fl-glycerophosphate, and 1 mM levamisole. Oligonucleotides utilized in gel-shift
analysis were synthesized on both strands and
annealed. Their sequences are shown in Fig. 1.
Binding reactions (15 #1) contained 10 to 30 fmol
radiolabeled probe, 125mM Hepes pH7.9,
50 mM MgCI2, 1 mM CaCI2, 5 mM DTT, 50%
glycerol, 3 #g poly(dI-dC)-(dI-dC) and competitor DNA sequences, as indicated. Reactions were
A 73 bp region of the PRB-1 promoter is sufficient
for ethylene induction in transgenic plants
Regulatory sequences necessary for ethylene activation of the PRB-lb gene were previously located in the promoter sequence between positions
-213 and - 142 [ 24 ]. To further characterize this
promoter region (X1 region), we tested its ability
to behave as a cis-acting element responsive to
ethylene. To this aim tobacco plants were transformed with a construct containing the X 1 fragment directly fused at position -67 to the putative PRB-lb TATA box, followed by the GUS
reporter gene. The level of expression directed by
the X 1 segment was compared to that driven by
the -213 bp minimal ethylene-inducible promoter
and by the -67 bp inactive promoter [24]. Transgenic plants harboring the different constructs
were treated with ethylene for 48 h, and leaf protein extracts were tested for GUS activity by a
fluorometric assay. As shown in Fig. 2, in seven
transgenic plants tested, the X1 region conferred
to the reporter gene an average 30-fold lower GU S
activity than the -213 promoter segment, but
higher ethylene inducibility (25-fold induction).
Both the GUS activity and fold induction in
ethylene-treated X1 plants were significantly
higher than in plants containing the inactive -67
deleted promoter. The ethylene inducibility of the
X1 construct was tested both in the primary
transformants and in their progeny, yielding similar results. The drop in absolute transgene activity
may be due to the presence of complementary
enhancer elements in the deleted fragment between position -140 and -68, or to a change in
the nominal spacing between interacting factors.
Taken together, these results provide evidence for
148
Construct
Plants
tyre
GUS activity.
Fold
induction
[~.Mu~m~/~~ 1
l
I213
10
Deletions
IX'71
167
Point
Gin2
8
1
11
LYm2
S
2
i~
4
I
,
6
I
k
~4J---i
mutations
¢/////////]////////////////]//.a--;
I
~///////////////////Z
~]ll]llll/llllll/lll/.d~a
,
8
I
3
36
2
178
48
• Untreated
[] Ethylenetreated
Fig. 2. EthyleneinductionofGUS activityintransgenicplants
transformed with wild-type, deleted and mutated forms of the
-213 bp P R B - l b promoter fused to the GUS-coding region.
Horizontal columns represent the average GUS activity of
total protein extracts from independent transgenic plants,
treated with ethylene for 48 h or untreated. Numbers at right
indicate the fold-induction by ethylene. Standard deviation of
the mean and the number of plants tested are indicated for
each construct. GUS activity is expressed in pmol 4-methylumbelliferone (4-MU) produced in 1 h assay by 1 #g total
protein extract.
the presence in the 73 bp X1 segment of a cisacting element which mediates stimulation of
transcription by ethylene.
Mutations in the X1 region significantly affect
ethylene responsiveness in transgenic tobacco plants
In an attempt to identify critical sites necessary
for ethylene responsiveness, substitution mutations were inserted in the X1 region by sitedirected mutagenesis. The mutations were tested
in the context of the -213 PRB-lb minimal active
promoter (Fig. 1). In the G m l mutant a G-boxlike motif (CACGTG) [ 15] is abolished by a twobase substitution. The Gm2 mutant contains
modifications in a T A A G A G C C G C C sequence,
present in reverse orientation in PRB-lb gene and
conserved in several ethylene-responsive genes
(GCC-box) [12, 16]. In the Yml and Ym2 mutants three and seven bases, respectively, were
arbitrarily substituted. The mutated promoters
were fused to the G U S reporter gene and introduced in tobacco plants. The resulting transgenic
plants, at least 8 independent transformants for
each construct, were treated with ethylene for
48 h. G U S activity driven by the different constructs was then tested in protein extracts of leaf
disks sampled before and after ethylene treatment. As shown in Fig. 2, all the mutations caused
a pronounced reduction in the basal and induced
level of reporter gene expression, in comparison
to the wild-type -213 construct. Nonetheless, the
mutant constructs G m l , Yml and Ym2 maintained ethylene inducibility exhibiting 36-, 178and 48-fold induction, respectively. The fact that
the mutant constructs showed a fold induction
even higher than that of the -213 construct, is
due to their very low level of basal expression
detected in untreated plants. On the other hand,
the Gm2 mutation completely abrogated response
of the -213 PRB-lb promoter to ethylene induction (Fig. 2). Plants harboring the Gm2 mutation
construct showed a neglectable GUS activity,
which was not significantly affected by ethylene
treatment, in a manner similar to plants containing the inactive -67 deletion construct. The finding that all the four mutations affected the activity of the PRB-lb promoter shows once again the
central role of the X1 region in the activation of
PRB-lb transcription and suggests that more than
one factor is involved in the control of PRB-lb
expression. Moreover, the results concerning the
Gm2 mutation define in a very specific manner
the GCC-box as a sequence region essential for
ethylene inducibility.
Mutations in oligonucleotides from the X1 region
modify DNA-protein interactions
Point mutations G m l , Gm2, Yml and Ym2 were
inserted in the context ofoligonucleotides G (from
positions -179 to -201; Fig. 1), and Y (from
positions -180 to -155; Fig. 1). These oligonucleotides were shown to interact specifically in
gel-shift analysis with tobacco nuclear factors
[24]. When the wild-type G sequence was tested
by gel-shift analysis with extracts from ethylenetreated plants, at least three complexes of retarded
migration were resolved (G1, G2 and G3;
149
Fig. 3A). The G1 complex, previously detected in
Meller et al. [24], was not consistently seen in
different extracts reacted with oligonucleotide G.
No significant difference was observed between
binding of ethylene-treated and untreated extracts
(data not shown). In gel-shift analysis the G m l
mutation (G-box-like) completely abolished the
G2 complex formation, did not alter the G3 complex and increased the intensity of the G1 complex, that is not consistently seen when using the
wild-type G oligonucleotide (Fig. 3A). The Gm2
mutation (GCC-box) altered the apparent affinities of the binding activities for the respective sites
on the G segment, increasing the intensity of the
G1 complex and decreasing that of G3 (Fig. 3A).
Consistent with these observations, when a competition analysis was performed, the G m l oligonucleotide failed to compete with the G2 complex, while the Gm2 oligonucleotide was less
efficient for both the G2 and G3 complexes than
the wild-type oligonucleotide G (Fig. 3B). The
observations relative to the G m l mutation reveal
that the G-box-like motif is an essential site for
the interaction between the G oligonucleotide and
the G2 binding activity. The Gm2 mutation, that
in transgenic plants affected ethylene inducibility,
did not prevent complex formation, but altered
the binding affinities of nuclear factors to their
binding sites. When the labeled Y sequence was
tested in gel-shift assay with the same protein
extracts, at least two specific complexes, Y1 and
Y2, were observed (Fig. 4A). No significant differences were observed in the complex formation
pattern obtained using extracts from ethylenetreated or untreated plants (data not shown). The
mutant oligonucleotides Yml and Ym2 maintained the same qualitative pattern of complex
formation of the wild-type Y, but mutations affected the apparent affinities of the Y1 and Y2
complexes (Fig. 4A). Oligonucleotide Yml
slightly diminished the intensity of the Y1 complex and increased the intensity of Y2. On the
other hand, oligonucleotide Ym2 determined the
formation of a weaker Y2 complex. When oligo'
nucleotide Y, Yml and Ym2 were used as competitors to the wild-type Y, Yml and Y behaved
similarly, while the capability of Ym2 as competitor was reduced (Fig. 4B). Therefor, the Yml and
Fig. 3. A. Gel-shift assay of wild-type and mutated G ofigo-
Fig. 4. A. Gel-shift assay of wild-type and mutated Y oligo-
nucleotides with total nuclear extract. Oligonucleotides G,
G m l and Gm2 (10 fmol) were used as probes, as indicated,
in gel-shift assay with 10 #g nuclear proteins from ethylenetreated plants in the presence of 3/~g poly(dI-dC)-(dI-dC).
G 1, G2 and G3 indicate protein-DNA complexes. P indicates
migration of free probe. B. Gel shift analysis of the G oligonucleotide in the presence of G, G m l and Gm2 competitor
oligonucleotides. Competitors and fold-molar excess used are
indicated. G1, G2 and G3 indicate protein-DNA complexes.
P indicates migration of free probe.
nucleotides with total nuclear extract. Oligonucleotides Y,
Yml and Ym2 (10 fmol) were used as probes, as indicated, in
gel-shift assay with 10/lg nuclear proteins from ethylene
treated plants in the presence of 3 #g poly(dI-dC)-(dI-dC). Y 1
and Y2 indicate protein-DNA complexes. P indicates migration of free probe. B. Gel shift analysis of the Y oligonucleotide in the presence of Y, Yml and Ym2 competitor oligonucleotides. Competitors and fold-molar excess used are
indicated. Y1 and Y2 indicate protein-DNA complexes. P
indicates migration of free probe.
150
Ym2 mutations, that in transgenic plants reduced
activity of the promoter without affecting its responsiveness to ethylene, caused in vitro an alteration in the apparent binding affinity of Y1 and
Y2 complexes.
The phosphorylation state of the G2 complex affects
its mobility
Phosphorylation events are involved in the ethylene signal transduction pathway of pathogenesis-related proteins elicitation [30]. To test
whether the observed binding activities, forming
complexes with the G and Y oligonucleotides,
were affected by phosphorylation, we performed
all the stages of nuclear protein extraction in the
presence or absence of phosphatase inhibitors.
These extracts were used in gel-shift analysis and
when reacted with the Y oligonucleotide they gave
a similar pattern of complex formation (data not
shown). However, when the G fragment was used
as a probe, the extract prepared in the presence
of phosphatase inhibitors formed with G a faster
migrating G2* complex compared to the G2
formed by extract prepared in the absence of
phosphatase inhibitors (Fig. 5, lanes 1 and 2). To
test whether dephosphorylation of the G2 complex also affects the amount of complex formation, the extract prepared in the presence of
phosphatase inhibitors was then incubated with
shrimp alkaline phosphatase prior to the reaction
with the G fragment. Pretreatment with shrimp
alkaline phosphatase determined the formation of
a complex G2 of both slower mobility and higher
intensity compared to the extract prepared in the
presence of phosphatase inhibitors but untreated
with alkaline phosphatase (Fig. 5; lanes 2 and 3).
We therefore conclude that dephosphorylation
events can modify the mobility of the G2 complex
and the amount of G2 complex formation. Alternatively, different binding proteins are involved.
The same results were obtained whether ethylenetreated or untreated tobacco nuclear extracts were
used. Thus, no ethylene dependence for the
phosphorylation-dependent shift of the G2 complex could be demonstrated in vitro.
Fig. 5. Effect of dephosphorylation on the mobility of the G2
complex. Labeled oligonucleotide G (10 fmol) was tested in
gel-shift assay with 10/~g of total nuclear proteins extracted
with or without phosphatase inhibitors. Lane 1, gel-shift assay
of G oligonucleotide with nuclear proteins extracted in the
absence of phosphatase inhibitors; lane 2, gel-shift assay of G
oligonucleotide with nuclear proteins extracted in the presence
of phosphatase inhibitors NaVanadate (0.2 mM), fl-glycerophosphate (10 mM), and levamisole (1 mM); lane 3, gel-shift
assay of G oligonucleotide with nuclear proteins extracted in
the presence of phosphatase inhibitors. The binding reaction
was carried out in the absence of phosphatase inhibitors and
was preceeded by a 30 min preincubation of the extract with
5 units of shrimp alkaline phosphatase. G2 and G3 indicate
protein-DNA complexes, G2*, the shifted phosphorylated G2
complex. P indicates migration of free probe.
Discussion
The plant hormone ethylene induces transcription of several components of the pathogenesisrelated proteins (PR) family [36]. We have previously shown that ethylene treatment determines
accumulation of PRB-lb transcript, and that ethylene induction is lost in transgenic tobacco plants
by deletion of a 71 bp segment (X1 segment) of
the PRB-lb promoter, between position -213 and
-142 [ 11, 24]. Here we further characterized in
transgenic tobacco plants the X1 fragment and
show it to be sufficient to direct ethylene-inducible
expression of a GUS reporter gene. In addition,
by substitution mutations, a cis-acting dement essential for ethylene induction was functionally localized in a TAAGAGCCGCC motif (GCCbox). The GCC-box is a conserved element
present in promoters of several ethylene-induced
pathogenesis-related genes, such as PRB-lb,
151
chitinase fl-l,3-glucanase of different plant species [ 12, 16, 28]. For instance, it is found in a
bean chitinase gene, as part of the minimal promoter responsive to ethylene induction [4, 31].
Two copies of the GCC box are present in a 61 bp
element derived from the promoter of the fl-l,3glucanase GLB gene, which has constitutive enhancer activity in Nicotiana plumbaginifolia protoplasts [ 16]. Mutations in the GCC elements of
the GLB gene reduce its enhancer activity and
capacity to bind nuclear proteins in a developmental and ethylene-dependent fashion [16].
However, deletion analysis of the same promoter
revealed that the 61 bp enhancer element, containing the two copies of the GCC box, is important for the overall activity of the promoter but it
is not essential for ethylene induction of the gene
[41]. A third GCC box is found in reverse orientation in the GLB gene, but it is located in a region which is not sufficient to confer ethylene
responsiveness. Our results concerning the GCC
box of the PRB-lb promoter provide the first evidence that this sequence is directly involved in the
response to ethylene induction in the context of
PRB-Ib expression. Since other ethylene regulated genes, such as those expressed in climateric
fruits and in carnation senescence, do not contain
the T A A G A G C C G C C sequence [5, 7, 9, 18], it
is likely that different cis-acting elements are involved in transcriptional activation by ethylene.
Indeed, gene regulation by ethylene in fruit, as
opposed to leaf, appears to be stage-dependent
[39].
By gel-shift analysis we previously characterized binding activities on two oligonucleotides
(G and Y) which cover almost entirely the
X1 segment (Fig. 1) [24]. The G and Y oligonucleotides interacts with three and two binding
activities, respectively, which are present in both
ethylene treated and untreated extracts. The Gm2
mutation in the G C C box, that prevents ethylene
induction in transgenic plants, when tested in the
context of the G oligonucleotide, caused a reduction in the apparent binding affinity of the G3
complex. This effect observed in vitro may represent an alteration in protein-DNA interactions,
that is sufficient to determine in vivo the loss of
ethylene inducibility of the mutated promoter.
Additional mutations in the X 1 region reduced, in
transgenic plants, the overall activity of the gene
without affecting its responsiveness to ethylene.
This suggests that more than one cis-acting element is involved in the control of PRB-lb expression. The X 1 region contains a G-box like motif
which is part of a binding site for a nuclear factor. Alteration in this motif significantly decreased
PRB-lb expression in transgenic plants, and in
gel-shift assay abolished the G2 complex formation in the context of the G oligonucleotide. The
G-box motif is required for expression of differently regulated promoter s of unrelated genes, such
as light-regulated [10, 33], abscisic acid-induced
[27] and stress-induced genes [13, 14]. Several
DNA-binding proteins specific for G-box-like
motifs have been isolated and characterized as
members of the transcription factors bZIP family [25, 34, 38]. The physical vicinity of the G-box
motif and the GCC-box on the PRB-lb promoter
and their distinct functional properties, revealed
by site-specific mutagenesis, suggest combinatorial interactions between the two elements in the
regulation of PRB-lb expression. We can hypothesize that factors, which transduce the ethylene
signal binding to the GCC-box, act in concert
with other factors which bind to the G-box motif
or to other sequences in the X1 region and increase the level of expression of the gene. Interactions of GCC-motifs with other elements are
likely to take place in other genes since in different instances the conserved element is located in
promoter regions not sufficient to respond to
ethylene induction [40, 41, 31]. The G-box like
motifs have been shown to participate in combinatorial interactions. In the Arabidopsis rbc-lA
promoter two I-boxes that flank a G-box element
are bound by nuclear factors and are necessary
for full light-regulated expression of the gene [ 10].
Similarly, a G-box in the parsley chalcone synthase promoter requires the presence of another
element at a defined distance to correctly mediate
light responsiveness [3].
A characteristic of the G2 complex formed by
the G-box motif of the PRB-lb promoter with
nuclear binding activities, is that its mobility and
152
intensity are modified by changes in the phosphorylation state of the complex. The dephosphorylated complex migrates more slowly than the
phosphorylated one and appears to have increased binding affinity. It is possible that the
phosphorylated complex consists of oligonucleotide G interacting with the monomer form of a
protein, that when dephosphorylated, binds with
a higher affinity to DNA in a multimeric form or
in association with additional proteins. In contrast the binding activity of the G-box-binding
factor GB F 1 of Arabidopsis is stimulated by phosphorylation and migrates more slowly in the
phosphorylated form [23]. In animal systems
phosphorylation has been shown to regulate transcription by affecting nuclear translocation, by
inhibition or stimulation of DNA binding by transcription factors, or interfering in the interactions
of transcription factors transactivating domains
with the transcriptional machinery [ 17 ]. In plants,
the phosphorylation state of binding activity has
been shown to affect formation and mobility of
complexes in gel-shift analysis [e.g. 8, 32, 37],
however the functional effect of this phenomenon
has yet to be elucidated. The phosphorylation
state of transcription factors interacting with the
PRB-lb promoter may be part of a regulating
mechanism related to ethylene-inductive processes.
Acknowledgements
We thank Medi Kan for her excellent technical
assistance. This work was supported by grants
from the Academy of Science, Jerusalem, the
Ministry of Science, Technology and Arts,
Jerusalem (MOSTA), the German Bundesministerium far Forschung und Technologie, Braunschweig (BMFT), and the Leo and Julia Forchheimer Center for Molecular Genetics. R.F. is a
recipient of the Jack and Florence Goodman
Career Development Chair. G.S. is a recipient of
a Scholarship from the Feinberg Graduate
School.
References
1. Abeles FB, Morgan PW, Saltveit Jr ME: Ethylene in
Plant Biology, 2nded. Academic Press, New York
(1992).
2. An G: Binary Ti vectors for plant transformation and
promoter analysis. Meth Enzymol 153:292-305 (1987).
3. Block A, Dangl JL, Hahlbrock K, Shulze-Lefert P: Functional borders, genetic free structure, and distance requirements of cis elements mediating fight responsiveness
of the parsley chalcone synthase promoter. Proc Natl
Acad Sci USA 87:5387-5391 (1990).
4. Brogfie KE, Biddle P, Cressman R, Broglie R: Functional
analysis of DNA sequences responsible for ethylene regulation of a bean chitinase gene in transgenic tobacco.
Plant Cell 1:599-607 (1989).
5. Cass KG, KirvenA, Chrisoffersen RE: Isolation and
characterization of a cellulases gene family member expressed during avocado fruit ripening. Mol Gen Genet
223:76-86 (1990).
6. Chang C, Kwok SF, Bleecker AB, Meyerowitz EM: Arabidopsis ethylene-response gene ETR 1: similarity of product to two-component regulators. Science 262:539-544
(1993).
7. Cordes S, Deikman J, Margosian LJ, Fisher RL: Interaction of a developmentally regulated DNA-binding factor with sites flanking two different fruit-ripening genes
from tomato. EMBO J 7:3315-3320 (1988).
8. Datta N, Cashmore AR: Binding of a pea protein to promoters of certain photoregulated genes is modulated by
phosphorylation. Plant Cell 1:1069-1077 (1989).
9. Deikmann J, Kline R, Fischer RL: Organization of ripening and ethylene regulatory regions in a fruit-specific
promoter from tomato (Lycopersicon esculentum). Plant
Physiol 100:2013-2017 (1992).
10. Donald RGK, Cashmore AR: Mutation of either G box
or I box sequences profoundly affects expression from the
Arabidopsis rbcS-1A promoter. EMBO J 9:1717-1726
(1990).
11. Eyal Y, Sagee O, Fluhr R: Dark-induced accumulation of
a basic pathogenesis-related (PR-1) transcript and a light
requirement for its induction by ethylene. Plant Mol Biol
19:589-599 (1992).
12. Eyal Y, Meller Y, Lev-Yadun S, Fluhr R: A basic-type
PR-1 promoter directs ethylene responsiveness, vascular
and abscission zone-specific expression. Plant J 4: 225234 (1993).
13. Ferl RJ, Nick HS: In vivo detection of regulatory factor
binding sites in the 5' flanking region of maize Adhl. J
Biol Chem 262:7947-7950 (1987).
14. Ferl RJ, Laughner BH: In vivo detection of regulatory factor binding sites of Arabidopsis thaliana Adh. Plant Mol
Biol 12:357-366 (1989).
15. Gilmartin PM, Sarokin L, Memelink J, Chua N-H: Molecular light switches for plant genes. Plant Cell 2: 369378 (1990).
153
16. Hart CM, Nagy F, Meins Jr F: A 61 bp enhancer element
of the tobacco fl-1,3-glucanase B gene interacts with one
or more regulated nuclear proteins. Plant Mol Biol 21:
121-131 (1993).
17. Hunter T, Karin M: The regulation of transcription by
phosphorylation. Cell 70:375-387 (1992).
18. ItzhakiH, MaxsonJM, WoodsonWR: An ethyleneresponsive enhancer element is involved in the senescence-related expression of the carnation glutathione-Stransferase (GST1) gene. Proc Natl Acad Sci USA 91:
8925-8929 (1994).
19. Jefferson RA, Kavanagh TA, Bevan MW: GUS fusions:
fl-glucuronidase as a sensitive and versatile gene fusion
marker in higher plants. EMBO J 6:3901-3907 (1987).
20. Kende H: Ethylene biosynthesis. Annu Rev Plant Physiol
Plant Mol Biol 44:283-307 (1993).
21. Kieber JJ, Ecker JR: Ethylene gas: it's not just for ripening anymore! Trends Genet 9:356-362 (1993).
22. Kieber J J, Rothenberg M, Roman G, Feldmann KA,
Ecker JR: CTR1, a negative regulator of the ethylene response pathway in Arabidopsis, encodes a member of the
Raf family of protein kinases. Cell 72:427-441 (1993).
23. Klimczak LJ, Schindler U, Cashmore AR: DNA binding
activity of the Arabidopsis G-box binding factor GBF 1 is
stimulated by phosphorylation by casein kinase II from
broccoli. Plant Cell 4:87-98 (1992).
24. Meller Y, Sessa G, Eyal Y, Fluhr R: DNA-protein interactions on a cis-DNA element essential for ethylene regulation. Plant Mol Biol 23:453-463 (1993).
25. Menkens AE, Cashmore AR: Isolation and characterization of a fourth Arabidopsis thaliana G-box-bindingfactor,
which has similarities to Fos oncoprotein. Proc Natl Acad
Sci USA 91:2522-2526 (1994).
26. Montgomery, Goldman S, DeikmanJ, Margossian L,
Fisher L: Identification of an ethylene-responsive region
in the promoter of a fruit ripening gene. Proc Natl Acad
Sci USA 90:5939-5943 (1993).
27. Mundy J, Yamaguchi-Shinozaki K, Chua N-H: Nuclear
proteins bind conserved elements in the abscisic acidresponsive promoter of a rice rab gene. Proc Natl Acad
Sci USA 87:1406-1410 (1990).
28. Ohme-Takagi M, Shinshi H: Structure and expression of
a tobacco fl-l,3-glucanase gene. Plant Mol Biol 15: 941946 (1990).
29. RazV, FluhrR: Calcium requirement for ethylenedependent responses. Plant Cell 4:1123-1130 (1992).
30. Raz V, Fluhr R: Ethylene responsiveness is transduced
via phosphorylation events in the plants. Plant Cell 5:
523-530 (1993).
31. Roby D, Broglie K, Gynor J, Broglie R: Regulation of a
chitinase gene promoter by ethylene and elicitors in bean
protoplasts. Plant Physiol 97:433-439 (1991).
32. Sarokin LP, Chua N-H: Binding sites for two novel phosphoproteins, 3AF5 and 3AF3, are required for rbc-3A
expression. Plant Cell 4:473-483 (1992).
33. Schindler U, Cashmore AR: Photoregulated gene expression may involve ubiquitous DNA binding proteins.
EMBO J 9:3415-3427 (1990).
34. Sehindler U, Menkens AE, Beckmann H, Ecker JR,
Cashmore AR: Heterodimerization between light-regulated and ubiquitously expressed Arabidopsis GBF bZIP
proteins. EMBO J 11:1261-1273 (1992).
35. Sessa G, Yang X-Q, Raz V, Eyal Y, Fluhr R: Dark induction and subcellular localization of the pathogenesisrelated PRB-lb protein. Submitted.
36. Somssich IM: Regulatory elements governing PR gene
expression. In: Nover L (ed) Plant Promoters and Transcription Factors, pp. 163-179. Springer-Verlag, Berlin/
Heidelberg/New York (1994).
37. Sun L, Doxsee RA, Harel E, Tobin EM: CA-l, a novel
phosphoprotein, interacts with the promoter of the cab140
gene in Arabidopsis and is undetectable in detl mutant
seedlings. Plant Cell 5:109-121 (1993).
38. Tabata T, Takase H, Takayama S, Mikami K, Nakatsuka A, Kawata T, Nakayama T, Iwabuchi M: A protein
that binds to a cis-acting element of wheat histone genes
has a leucine zipper motif. Science 245:965-967 (1989).
39. Theologis A: One rotten apple spoils the whole bushel:
the role of ethylene in fruit ripening. Cell 70:181-184
(1992).
40. van de Rhee M, Lemmers R, Bol JF: Analysis of regulatory elements involved in stress-induced and organspecific expression of tobacco acidic and basic fl-l,3glucanase genes. Plant Mol Biol 21:451-461 (1993).
41. V0geli-Lange RC, Hart CM, Nagy F, Meins F Jr: Developmental, hormonal, and pathogenesis-related regulation
of the tobacco class I fl-l,3-glucanase B promoter. Plant
Mol Biol 25:299-311 (1994).