The Plant Cell, Vol. 5, 1049-1062, September 1993 O 1993 American Society of Plant Physiologists
Positive and Negative Regulatory Regions Control the
Spatial Distribution of Polygalacturonase Transcription in
Tomato Fruit Pericarp
Julie Montgomery, Vicky Pollard, Jill Deikman,' and Robert L. Fischer *
Department of Plant Biology, University of California, Berkeley, California 94720
The tomato fruit consists of a thick, fleshy pericarp composed predominantly of highly vacuolated parenchymatous cells,
which surrounds the seeds. During ripening, the activation of gene expression results in dramatic biochemical and phys-
iological changes in the pericarp. The polygalacturonase (PG) gene, unlike many fruit ripening-induced genes, is not
activated by the increase in ethylene hormone concentration associated with the onset of ripening. To investigate ethyl-
ene concentration-independent gene transcription in ripe tomato fruit, we analyzed the expression of chimeric PG pmmoter-
B-glucuronidase (GUS) reporter gene fusions in transgenic tomato plants. We determined that a 1.4-kb PG promoter directs
ripening-regulated transcription in outer pericarp but not in inner pericarp cells, with a sharp boundary of PG promoter
activity located midway through the pericarp. Promoter deletion analysis indicated that a minimum of three promoter
regions influence the spatial regulation of PG transcription. A positive regulatory region from -231 to -134 promotes
gene transcription in the outer pericarp of ripe fruit. A second positive regulatory region from -806 to -443 extends
gene activity to the inner pericarp. However, a negative regulatory region from -1411 to -1150 inhibits gene transcription
in the inner pericarp. DNase I footprint analysis showed that nuclear proteins in unripe and ripe fruit interact with DNA
sequences within each of these three regulatory regions. Thus, temporal and spatial control of PG transcription is medi-
ated by the interaction of negative and positive regulatory promoter elements, resulting in gene activity in the outer pericarp
but not the inner pericarp of ripe tomato fruit. The expression pattern of PG suggests that, although they are morphologi-
cally similar, there is a fundamental difference between the parenchymatous cells within the inner and outer pericarp.
The seeds of many angiosperms are borne within a special-
ized organ, the fruit, which develops from the gynoecium. As
shown in Figure 1, tomato fruit consists of a pericarp, formed
from the ovary walls, which surrounds the placenta1 tissue and
the seeds. A waxy cuticle on the outside of the fruit covers
the epidermis and thickens as the fruit ages. Vascular traces
extend through the pericarp from the stem end to the blos-
som end of the fruit, roughly parallel to the outer edge of the
pericarp, and are visible end-on as small dots forming a ring
within the pericarp, as shown in the cross-section depicted
in Figure 1. The pericarp can be subdivided into the exocarp,
the mesocarp, and the endocarp. According to Ho and Hewitt
(1986), the exocarp consists of the outer epidermal layer and
two to four layers of small, thick-walled cells immediately be-
neath the epidermis, the endocarp comprises the single cell
layer adjacent to the locular region, and the mesocarp with
vascular bundles encompasses the layer of large, highly vacuo-
lated parenchymatous cells located between the exocarp and
nia State University, University Park, PA 16802
To whom correspondence should be addressed.
Current address: Biology Department, 208 Mueller Lab, Pennsylva-
Early fruit development is characterized by a period of cell
division followed by cell expansion resulting in the formation
of a mature unripe fruit that is generally hard, tasteless, and
unappealing to eat. Ripening is the final phase of fruit devel-
opment when fruits undergo a series of ultrastructural,
biochemical, and physiological changes to become attractive
for consumption by animals, thereby promoting seed disper-
sal. Tomato fruit ripening is characterized by an increase in
respiration, autocatalytic ethylene biosynthesis, chlorophyll
degradation, carotenoid biosynthesis, conversion of starch to
sugars, production of essential oils, and softening (Brady et
al., 1987). Many of these ripening-related changes result
from the induction of new enzyme activities (Brady, 1987),
which in turn result from the activation of gene expression
(Mansson et al., 1985; Slater et al., 1985; Biggs et al., 1986;
Lincoln et al., 1987).
Molecular analysis of the control of fruit ripening has led
to the isolation of a number of cDNAs that represent mRNAs
more abundant in ripe fruit than in unripe fruit (for review, see
Gray et áI., 1992). Although the identity of the proteins encoded
by many of the ripening-related cDNAs remains unknown, sev-
era1 of them have been shown to encode enzymes known to
1050 The Plant Cell
(Sheehy et ai., 1988; DellaPenna et al., 1989). These results
indicate that the increase in polygalacturonase enzyme activ-
ity in ripe fruit is due to transcriptional activation of PG gene
expression during ripening.
The plant hormone ethylene plays an important role in con-
trolling the ripening of tomato and other fruits. Ethylene
biosynthesis increases dramatically just prior to the first visi-
ble signs of ripening in tomato fruit. In addition, exposure of
unripe tomato fruit to exogenous ethylene hastens the onset
of ripening, and the inhibition of ethylene biosynthesis or eth-
ylene action prevents ripening (Rhodes, 1980; Yang and
Hoffman, 1984; McGlasson, 1985; Oeller etal., 1991). Although
PG transcription increases during ripening concomitantly with
the rise in ethylene biosynthesis, it is not regulated by the in-
crease in ethylene concentration. Exposure of unripe fruit to
exogenous ethylene induces expression of ethylensresponsive
ripening-related genes, but does not induce PG gene expres-
sion (Lincoln et al., 1987). Moreover, the PG gene is expressed
in fruit from tomato plants in which ethylene biosynthesis has
been inhibited through expression of antisense RNAs that block
accumulation of ACC synthase (Oeller et al., 1991). These
results demonstrate that PG gene transcription during ripen-
ing is not controlled by the increase in ethylene concentration
and suggest that expression of ripening-related genes in tomato
fruit is controlled by both ethylene concentration-dependent
and ethylene concentration-independent signal transduction
pathways (Theologis, 1992).
To investigate the molecular basis for ethylene concentra-
tion-independent gene expression during tomato fruit ripening,
we have analyzed the PG gene to identify cis-acting DNA se-
quences and putative ffans-acting nuclear factors that are
important for ripening-specific expression. Previous studies
have shown that 1.4 kb of PG 5‘flanking sequences are suffi-
cient to confer fruit-specific, ripening-regulated expression of
a chloramphenicol acetyltransferase reporter gene (Bird et al.,
1988). In this study, we describe the effects of 5’deletions in
the 1.4-kb PG promoter on the transcription of a Pglucuronidase
(GUS) reporter gene in transgenic tomato plants. In addition,
we have characterized the spatial pattern of PG gene activity
in tomato fruit and the effects of promoter deletions on that
pattern through histochemical localization of P-glucuronidase
vascular bundles ‘I
Figure 1 . Diagram of Tomato Fruit Anatomy.
A schematic representation o f a cross-section through a bilocular tomato
fruit is shown. Various anatomical structures are labeled.
function during ripening. Among these are genes for enzymes
involved in ethylene biosynthesis, l-aminocyclopropane-l-
carboxylate (ACC) synthase (van der Straeten et al., 1989;
Rottman et ai., 1991) and ACC oxidase (Hamilton et al., 1990,
1991); the gene for phytoene synthase, an enzyme involved
in lycopene biosynthesis (Bird et al., 1991; Ray et al., 1992);
and the gene encoding polygalacturonase, a cell wall hydro-
lase (DellaPenna et ai., 1986; Grierson et a1.;1986).
Polygalacturonase accumulates during fruit ripening and is
responsible for the degradation of polyuronides, or pectin, in
the fruit cell wall (Smith et ai., 1988, 1990; Giovannoni et al.,
1989). There is a rough correlation between the level of poly-
galacturonase activity and the degree of softening among
different cultivars and mutants of tomato (Brady et ai., 1983;
Speirs et al., 1989; Ahrens and Huber, 1990). However, experi-
ments designed to directly address the role of polygalact-
uronase during ripening indicate that polygalacturonase activity
alone is not sufficient for fruit softening (Giovannoni et al., 1989;
Schuch et al., 1991). The softening of fruit during ripening is
most likely due to the action of a number of cell wall hydrolytic
enzymes present in ripe fruit, including polygalacturonase
(Fischer and Bennett, 1991). Polygalacturonase enzyme ac-
tivity may contribute to more subtle changes in fruit texture
that affect fruit quality (Schuch et al., 1991).
A single gene encodes the abundant polygalacturonase pro-
tein present in ripe tomato fruit (Bird et al., 1988; Giovannoni
et al., 1989), and its expression is tightly regulated during ripen-
ing. Polygalacturonase (PG) mRNA is not detected in immature
or unripe fruit, but its level increases sharply at the onset of
ripening and accumulates to more than lO/o of the mal
poly(A)+ FINA in ripe fruit (DellaPenna et al., 1986; Sheehy et
al., 1987). Nuclear run-on transcription experiments show that
the transcriptional activity of the PG gene increases at the on-
set of ripening and closely parallels the rise in PG mRNA
Effect of 5’ Deletions on PG Promoter Activity
in Pansgenlc Plants
To identify regions of the PG promoter that are involved in
ripening-regulated gene expression, we constructed a set of
chimeric genes consisting of PG promoters of various lengths
ligated to the GUS reporter gene. These chimeric genes were
constructed using a PG genomic clone, pPGmutHHl.5, that
had been mutagenized at the ATG translation initiation codon
Spatial Regulation of Polygalacturonase 1051
to create an Ncol restriction endonuclease site (Giovannoni
et al., 1989). From this clone, a Hindlll-Ncol restriction frag-
ment containing 1.4 kb of PG 5'flanking sequences, the entire
67-bp untranslated mRNA leader sequence, and the PG ATG
initiation mdon was ligated in frame to GUS coding sequences
and nopaline synthase 3' sequences derived from p61101.3
(Jefferson et al., 1987). This PG-GUS chimeric gene, shown
in Figure 2A, was then used to generate a series of 5' pro-
moter deletions, and the resulting PG-GUS deletion constructs
were introduced into tomato plants by Agrobacterium-mediated
plant transformation as described in Methods.
To determine the level of chimeric PG-GUS gene expres-
sion during ripening, pglucuronidase enzyme activity was
measured in unripe and ripe fruit harvested from each indepen-
dently transformed plant. The average level of P-glucuronidase
activity in plants transformed with each deletion construct is
presented in two graphs with different scales in Figure 26 to
display the large changes in enzyme activity. Our results show
that 1411 bp of PG 5' flanking sequences directed ripening-
regulated expression of a PG-GUS chimeric gene, with 60-
fold more P-glucuronidase activity in ripe fruit than in unripe
fruit (Figure 26). Removal of PG DNA sequences to -1150
increased the amount of P-glucuronidase activity in ripe fruit
40-fold, but had little effect on 0-glucuronidase activity in un-
ripe fruit (Figure 26). Only twofold lower levels of P-glucuron-
idase activity was observed in ripe fruit upon additional dele-
tion of PG sequences to -806. However, further deletion of
DNA sequences to -443 reduced f3-glucuronidase activity 60-
fold in ripe fruit, and removal of sequences from -231 to -134
resulted in equivalent amounts of B-glucuronidase activity in
both unripe and ripe fruit. Background levels of P-glucuronidase
activity were measured in fruit from plants transformed with
a PG-GUS gene containing no PG 5' flanking sequences or
with the pMWl shuttle vector alone.
These results indicate that three regions in the PG promoter
control PG-GUS gene expression during tomato fruit ripen-
ing. The increase in 6-glucuronidase activity in ripe fruit upon
removal of PG DNA sequencesfrom -1411 to -1150 suggests
that this promoter region negatively regulates PG-GUS gene
expression. Deletion of PG sequences from -806 to -443
resulted in a decrease in P-glucuronidase activity in ripe fruit,
indicating that this region positively regulates PG-GUS gene
expression. Finally, ripening-inducible PG-GUS gene expres-
sion was lost upon removal of PG DNA sequences from -231
and -134, indicating that this region also positively regulates
PG-GUS gene expression and is critical for ripening regulation.
Histochemical Localization of PG Gene Expression
in Tomato Fruit
In situ hybridization experiments have shown that PG mRNA
is expressed primarily in the outer cell layers of the tomato
fruit pericarp and in cells surrounding the vascular regions
(Pear et al., 1989). To determine if the pattern of PG-GUSgene
Flgure 2. Effect of Promoter Deletions on PG-GUS Gene Expression
during Tomato Fruit Ripening.
(A) PG-GUS chimeric gene construct. The DNA sequence of the trans-
lational fusion between PG 5' flanking sequences and GUS coding
sequences is indicated. Underlined nucleotides were derived from
pB1101.3 polylinker sequences and GUS coding sequences (Jefferson
et al., 1987). Boldface nucleotides denote PG and GUS A T G transla-
tion initiation codons. H, Hindlll; N, Ncol; S, Sstl; R, EcoRI; the line
represents PG 5'flanking sequences; the filled box represents 67 bp
of PG untranslated mRNA leader sequence; the filled rectangle
represents nopaline synthase 3' sequences.
(B) P-Glucuronidase activity in unripe and ripe fruit harvested from
PG-GUS transgenic tomato plants. To accommodate large changes
in P-glucuronidase activity, the data is plotted against two different
scales, O to 8000 and O to 180 pmol of 4-methylumbelliferyl (4-MU)
per milligram of protein per minute, respectively. PGlucuronidase ac-
tivity (picomoles of 4-methylumbelliferyl per milligram of protein per
minute) was measured as described in Methods. PG promoter dele-
tion end points relative to the start of transcription are indicated in base
pairs and are shown on the abscissa. Error bars indicate standard er-
rors. The number of independent transformants analyzed individually
to determine the mean for each deletion are as follows: -1411, 10;
-1150, 11; -806, nine; -443, nine; -231, six; -134, five; +5, five;
pMW1, five. Unripe, mature green stage 1 fruit; ripe, orange or firm
The Plant Cell
expression is the same as that of the intact PG gene, we local-
ized p-glucuronidase activity in fruit bearing a PG-GUS gene
containing 1411 bp of 5'flanking sequence. Longitudinal slices
were cut from the center of unripe and ripe fruit harvested from
transgenic plants and stained for p-glucuronidase enzyme ac-
tivity as described in Methods. As shown in Figure 3A, ripe
fruit exhibited intense staining in the collumella and in the outer
regions of the pericarp, but not in the inner layer of pericarp
cells. Staining was also detected in the vascular regions (data
not shown). No staining was visible in unripe fruit, pictured
in Figure 3B. Thus, (3-glucuronidase activity is localized in a
pattern similar to the distribution of PG mRNA. This result in-
dicates that the spatial regulation of a PG-GUS gene with 1411
bp of PG 5' flanking sequences corresponds to the spatial regu-
lation of the endogenous PG gene during fruit ripening.
To determine whether the spatial distribution of PG gene ex-
pression in fruit pericarp is a general characteristic of fruit
ripening-related genes, we investigated the expression pat-
tern of another ripening-induced gene, E4 (Lincoln et al., 1987;
Lincoln and Fischer, 1988;Cordesetal., 1989). Fruit harvested
from tomato plants transformed with an £4-GL/Schimeric gene
consisting of 1421 bp of E4 5' flanking sequences fused to GUS
coding sequences (Montgomery et al., 1993) was stained for
P-glucuronidase enzyme activity. As shown in Figure 3C, ripe
fruit containing the E4-GUS gene showed uniform staining
throughout the pericarp, and no color was detected in unripe
fruit (Figure 3D). These results demonstrate that the pattern
of 3-glucuronidase enzyme activity observed in ripe fruit con-
taining the PG-GUS chimeric gene is specific to the PG
promoter and is not due to experimental artifacts such as differ-
ences in the size and vacuolization of cells in the inner and
outer pericarp. Furthermore, these results indicate that differ-
ent ripening-regulated genes have distinct patterns of
expression in tomato fruit pericarp.
Figure 3. Histochemical Localization of 0-Glucuronidase Activity in Fruit from Transgenic PG-GUS and E4-GUS Plants.
Fruit was harvested from plants transformed with a PG-GUS gene containing 1411 bp of 5' flanking DMA (Figure 2) or an E4-GUS gene containing
1421 bp of 5' flanking DMA sequences (Montgomery et al., 1993) and incubated with 5-bromo-4-chloro-3-indolyl glucuronide for 22 hr at room
temperature to stain for p-glucuronidase activity as described in Methods. Blue color indicates p-glucuronidase activity.
(A) and (B) Fruit harvested from PG-GUS transgenic plants, with (A) showing orange fruit and in (B), mature green stage 1 fruit.
(C) and (D) Fruit harvested from E4-GUS transgenic plants. In (C), 50% red fruit is shown, and in (D), mature green stage 1 fruit.
Spatial Regulation of Polygalacturonase 1053
Figure 4. Effect of Promoter Deletions on the Spatial Pattern of PG-GUS Gene Expression in Ripe Tomato Fruit.
Orange or firm red fruit was harvested from plants transformed with the various PG-GUS deletion constructs and incubated with 5-bromo-4-chloro-
3-indolyl glucuronide for 22 hr as described in Figure 3.
(A) Deletion construct -1411.
Effects of 5' Deletions on the Distribution of PG-GUS
Gene Expression in Ripe Fruit
The effects of PG promoter deletions on the distribution of
PG-GUS gene expression was investigated by histochemical
localization of p-glucuronidase activity in ripe tomato fruit. To
determine the expression patterns of the various PG promoter
deletion constructs, cross-sectional slices were cut from the
center of ripe fruit harvested from representative transgenic
plants bearing the PG-GUS chimeric genes and stained for
P-glucuronidase activity. Ripe fruit expressing a PG-GUS gene
with 1411 bp of PG 5' flanking sequences, shown in Figure
4A, again showed staining only in the outer cell layers of the
pericarp. The cross-sectional view displayed a sharp bound-
ary of p-glucuronidase activity within the mesocarp portion
of the pericarp at the position of the vascular bundles. The
outer pericarp cells between the epidermis and the ring of vas-
cular bundles stained strongly, whereas no staining was
detected in the inner pericarp cells between the vascular bun-
dles and the (ocular region (Figure 4A). However, fruit bearing
a PG-GUS gene with 1150 bp of PG 5' flanking sequences
(Figure 4B) stained strongly and uniformly over all regions of
the pericarp. These results show that the 1411-bp PG-GUS
gene is expressed only in the outer pericarp, and removal of
PG DNA sequences from -1411 to -1150 results in PG-GUS
gene expression in both the inner and the outer pericarp. This
change in the distribution of PG-GUS gene expression within
the fruit pericarp suggests that the increased p-glucuronidase
activity measured in ripe fruit extracts upon deletion of the nega-
tive regulatory region from -1411 and -1150 (Figure 2B) is
due, at least in part, to the activation of PG-GUS gene expres-
sion in the inner pericarp cells.
Histochemical localization of P-glucuronidase activity in ripe
fruit containing a PG-GUS gene with 806 bp of 5'flanking se-
quence also showed a uniform distribution of gene expression
(Figure 4C). However, ripe fruit expressing a PG-GUS gene
with 433 bp of 5' flanking sequences stained more strongly
in the outer pericarp than in the inner pericarp, although there
was not a sharp boundary in p-glucuronidase activity at the
vascular bundles (Figure 4D). Thus, removal of PG DNA se-
quences from -806 to -443 reduces PG-GUS gene expres-
sion in the inner pericarp. These results suggest that the
1054 The Plant Cell
decrease in B-glucuronidase activity measured in ripe fruit ex-
tracts upon deletion from -806 to -443 (Figure 28) is due,
at least in part, to the loss of a positive regulatory region that
activates PG-GUS gene expression in the inner pericarp cells.
Fruit bearing a PG-GUS gene with 231 bp of PG sequence
also stained more strongly in the outer pericarp than in the
inner pericarp (Figure 4E). No staining was visible in ripe fruit
from plants transformed with a PG-GUS gene containing just
134 bp of 5'flanking sequence (Figure 40. These results sug-
gest that a second positive ripening regulatory region from
-231 to -134 (Figure 26) activates PG-GUS gene transcrip-
tion in the outer pericarp, but not in the inner pericarp.
T o address the possibility that prolonged incubation in the
presence of saturating amounts of the Pglucuronidase sub-
strate might mask an uneven distribution of enzyme in ripe
fruit with either the 1150-bp PG-GUS or the 806-bp PG-GUS
gene constructs, we monitored the appearance of blue color
during the staining procedure. Toobserve the pattern of color
development, cross-sectional slices cut from the center of ripe
fruit were stained for &glucuronidase activity and photographed
after 0 . 5 , 2, and 4 hr of incubation. As shown in Figure 5, the
blue color in ripe fruit containing a E-GUS gene with 1411
bp of PG 5 ' flanking sequences was faintly visible after 2 hr
of incubation in the outermost cells of the pericarp and after
4 hr was visible in most of the cells in the outer pericarp. By
contrast, color development was more rapid in fruit express-
ing PG-GUS genes containing either the 1150 bp or 806 bp
of 5'flanking sequences, which is consistent with the increased
leve1 of P-glucuronidase activity in these fruit (Figure 28). Sig-
nificantly, the pattern of color development in these fruit was
quite different from the pattern observed in fruit with the 1411-
bp PG-GUS gene construct. After 2 hr of incubation, ripe fruit
bearing a PG-GUS gene with 1150 bp of PG 5 ' sequences
showed intense staining around the vascular regions and an
even distribution of staining in the rest of the pericarp. Fruit
containing a PG-GUS gene with 806 bp of 5'flanking sequence
showed a similar pattern of color development after just 30
min of incubation. By contrast, pglucuronidase enzyme ac-
tivity was restricted to the outer pericarp in fruit containing a
PG-GUS gene with 231 bp of 5' flanking sequence.
Footprint Analysis of the lnteraction of Fruit Nuclear
Factors with PG Regulatory Sequences
DNase I protection experiments were performed to determine
whether tomato fruit nuclear proteins interact with DNA se-
quences included within these functionally defined regulatory
regions. To this end, fragments of the PG promoter were end
labeled, incubated with nuclear protein extracts isolated from
unripe and ripe fruit, and subjected to DNase I digestion as
described in Methods. lncubation of a PG promoter DNA frag-
ment spanning the negative regulatory region from -1411 to
-1150 with nuclear proteins isolated from unripe and ripe fruit
showed two regions of protection, designated A and 6, that
are depicted in Figure 6. DNase I footprints of nuclear protein
factors that interact with PG DNA sequences in the positive
regulatory regions, from -806 to -443 and from -231 to -134,
are shown in Figure 7 . Three regions of protection, designated
C, D, and E, were detected upon incubation of a PG promoter
fragment spanning the positive regulatory region from -806
and -443 with nuclear protein extracts isolated from unripe
and ripe fruit (Figure 7A). Analysis of the interaction of fruit
nuclear proteins with sequences included in the minimal231-
bp ripening-regulated PG promoter identified one region of pro-
tection designated F (Figure 78). This footprint was located
within the positive regulatory region from -231 to -134 (Fig-
ure 26). The results of these footprinting experiments indicate
that nuclear proteins present in both unripe and ripe fruit nuclei
interact with DNA sequences located within regions of the PG
promoter that are important for PG-GUS gene transcription
during fruit ripening.
A Positive Regulatory Region 1 s Required for Temporal
Regulation of PG Transcription
Analysis of the effects of 5'promoter deletions on the expres-
sion of a chimeric E-GUS gene in unripe and ripe fruit
harvested from transgenic tomato plants demonstrated that
231 bp of PG 5' flanking sequences comprise a rninimal
ripening-inducible promoter (Figure 28). DNA sequences be-
tween -231 and -134 were necessary for ripening-regulated
transcription of this minimal promoter (Figure ZB), indicating
that this 104-bp DNA sequence defines a fruit ripening con-
trol region. Comparison of this region with promoter DNA
sequences required for ethylene concentration-independent
ripening-regulated expression of the tomato €8 gene (Deikman
et ai., 1992) showed no significant hornology. In addition, no
regions of homology were identified when these sequences
were compared to the 5'flanking sequences of ethylene con-
centration-dependent genes induced during ripening (Cordes
et ai., 1989; Cass et al., 1990; Rottman et al., 1991).
As a first step toward identifying regulatory proteins involved
in ethylene concentration-independent gene expression dur-
ing ripening, fruit nuclear proteins that interact with the fruit
ripening control region from -231 to -134 were detected using
DNase I protection analysis. We identified a 35-bp AT-rich se-
quence between -227 and -193 that was protected by proteins
in unripe and ripe fruit nuclear extracts (Figure 78) but did not
bind proteins in leaf nuclear extracts (J. Montgomery and R.L.
Fischer, unpublished data). The promoters of a number of plant
genes contain similar AT-rich DNA sequence elements that
interact with nuclear proteins (Jofuku et ai., 19a7; Jensen et
al., 1988; Bustos et al., 1989; Datta and Cashmore, 1989;
Jordano et al., 1989; Jacobsen et al., 1990; Bruce et al., 1991;
Manzara et al., 1991; Pedersen et al., 1991; Czarnecka et al.,
Spatial Regulation of Polygalacturonase 1055
Figure 5. Time Course of Histochemical Staining of Fruit from PG-GUS Transgenic Plants.
Orange or firm red fruit was harvested from plants transformed with the indicated PG-GUS deletion construct and incubated with 5-bromo-4-
chloro-3-indolyl glucuronide as described in Methods. Each fruit slice was photographed after 0.5, 2, and 4 hr of incubation. The incubation time
is indicated at the top, and the PG promoter deletion end points are shown at left, vb, vascular bundles.
1056 The Plant Cell
TOP STRAND BOTTOM STRAND
C U C R C A
Figure 6. DNase I Protection Analysis of PG Promoter Sequences
within the Negative Regulatory Element That Interact with Factors Pres-
ent in Fruit Nuclear Protein Extracts.
(A) DNase I footprint patterns defined by nuclear extracts isolated from
unripe fruit and ripe fruit. A PG promoter DNA fragment from the Hindlll
site (-1411) to the Xmnl site (-967) was subcloned into pUC119, digested
with Hindlll and Smal, and labeled at the -967 end (top strand) or
the -1411 end (bottom strand). DNase I reactions were performed as
described in Methods and run on a 5% (top strand) or 8% (bottom
strand) denaturing polyacrylamide gel. All lanes contain equal amounts
of labeled DNA and nuclear protein. The extent of the DNA sequence
visible on the gel is indicated by the solid lines, and regions protected
from DNase I digestion in the presence of nuclear protein are desig-
nated by the open boxes and letters at left. Maxam-Gilbert G and G+A
sequencing reactions (Maxam and Gilbert, 1980) were run on the same
gel to determine the location of the protected regions (data not shown
1992). Some of these AT-rich binding proteins have been iden-
tified as high-mobility group (HMG)-like proteins (J acobsen et
al., 1990; Pedersen et al., 1991; Czarnecka et al., 1992). In mam-
malian cells, the HMG1 protein and similar HMG-like proteins,
which contain the HMG domain DNA binding motif, bind AT-
rich DNA sequences with relatively little base specificity and
have been shown to play a role in gene transcription (Solomon
et al., 1986; Travis et al., 1991; Thanos and Maniatis, 1992).
The HMG-like lymphoid enhancer binding factor-1 (LEF-1) is
thought to regulate transcription by inducing a sharp DNA
bend, which may facilitate interactions between regulatory pro-
teins bound at distal sites (Giese et al., 1992). Binding of a
fruit nuclear factor to the minimal 231-bp PG promoter may
serve a similar function. The fact that binding is detectable
in unripe fruit before PG transcription increases suggests that
other proteins are probably necessary to activate the PG gene
in ripe fruit or that the binding factor may be modified to be-
come transcriptionally active.
PG Gene Transcription in Ripe Fruit Is Spatially
Expression of a 1411-bp PG-GUS gene in transgenic tomato
plants is similar to the expression of the endogenous polygalact-
uronase gene. Histochemical localization of p-glucuronidase
activity showed that the 1411-bp PG-GUS gene is transcribed
predominantly in the outer pericarp, between the epidermis
and the ring of vascular bundles, and in the vascular regions
of ripe tomato fruit (Figures 3, 4, and 5). This pattern of gene
expression is consistent with the results of in situ hybridiza-
tion experiments, which showed high concentrations of PG
mRNA in the outer cell layers of the pericarp and in cells sur-
rounding the vascular regions of ripening tomato fruit (Pear
etal., 1989). Moreover, immunocvtolocalization of polygalact-
uronase protein on tissue blots of ripening tomato fruit showed
that polygalacturonase appears first in the collumella and the
outer pericarp (Tieman and Handa, 1989). However, in over-
ripe fruit, polygalacturonase protein was also detected in the
inner pericarp. By contrast, expression of the 1411-bp PG-GUS
gene was never detected in the inner pericarp region, even
in overripe fruit (data not shown). We currently do not know
whether the different patterns of polygalacturonase protein ac-
cumulation and p-glucuronidase activity observed in overripe
fruit reflect differences in the experimental techniques or
differences in the expression of endogenous and chimeric
for the top strand). C, DNase I cleavage pattern without nuclear pro-
tein; U, unripe (mature green stage 1) fruit nuclear extract; R, ripe (30%
red) fruit nuclear extract; G+A, Maxam and Gilbert G +A DNA sequenc-
(B) Delineation of the sequences between -1395 and -1160 on the
PG promoter that are protected by the fruit nuclear proteins in the foot-
print analysis. Bracketed sequences on either the top or bottom strand
correspond to the regions of protection designated A and B in (A).
C U C R C
Figure 7. DNase I Protection Analysis of PG Promoter Sequences within the Positive Regulatory Regions from -806 to -443 and from -231
to -134 That Interact with Factors Present in Fruit Nuclear Protein Extracts.
(A) and (B) show DNase I footprint patterns defined by nuclear extracts isolated from unripe fruit and ripe fruit. DNase I reactions were performed
as described in Methods. All lanes contain equal amounts of labeled DNA and nuclear protein. The extent of the DNA sequence visible on the
gel is indicated by the solid lines, and regions protected from DNase I digestion in the presence of nuclear protein are designated by the open
boxes and letters at left. Maxam-Gilbert G and G+A sequencing reactions (Maxam and Gilbert, 1980) were run on the same gel to determine
the location of the protected regions (data not shown in [B]). C, DNase I cleavage pattern without nuclear protein; U, unripe (mature green stage
1) fruit nuclear extract; R, ripe (30% red) fruit nuclear extract; G+A, Maxam and Gilbert G+A DNA sequencing reactions.
(A) Footprint pattern in the region between -806 and -375. A PG promoter DNA fragment from -806 (Exolll-digested) to the Taql site (-375)
was subcloned into the Smal site of pUC119 and labeled at the -806 end (bottom strand). DNase I reactions were run on a 6% denaturing poly-
(B) Footprint pattern in the region between -231 and +5. A PG promoter DNA fragment from the Mboll site (-231) to the Mnll site (+5) was
subcloned into pUC119 at the Smal site and labeled at the -231 end (bottom strand). DNase I reactions were run on an 8% denaturing polyacryl-
(C) Delineation of the DNA sequences protected by fruit nuclear proteins in the footprint analysis. Schematic representations of the DNase I
footprint analysis of PG promoter sequences from -720 to -400 (upper) and from -231 to +5 (lower) are shown. The DNA sequences protected
by nuclear factors are displayed above the open boxes, and letters correspond to the regions of protection shown in (A) and (B).
1058 The Plant Cell
polygalacturonase promoters. Because polygalactumnase, un-
like P-glucuronidase, is secreted to the cell wall, pretreatment
of overripe fruit to release polygalacturonase protein from the
cell wall in the immunocytolocalization experiments may have
resulted in some diffusion of protein to the inner pericarp.
Nevertheless, the results of in situ hybridization experiments,
immunocytolocalization experiments, and histochemical local-
ization of j3-glucuronidase activity in PG-GUS transgenic
tomato plants demonstrated that the PG gene is expressed
most strongly in the outer cell layers of the pericarp and in
vascular regions of ripening tomato fruit.
The expression pattern of PG in tomato fruit pericarp sug-
gests that, although they are morphologically similar, there is
a fundamental difference between the highly vacuolated, par-
enchymatous cells in the inner pericarp and the outer pericarp.
According to the anatomy of tomato fruit described by Ho and
Hewitt (1986) and shown in Figure 1, polygalacturonase is tran-
scribed in the exocarp and in the outer mesocarp cells, but
is not transcribed in the inner mesocarp cells nor in the en-
docarp. We cannot determine, based upon these results,
whether the difference between outer pericarp cells and in-
ner pericarp cells that causes the differential expression of
polygalacturonase may be established during fruit develop-
ment or may be a consequence of ripening.
The biological significance of localized PG gene expression
in the outer pericarp is not known. Polygalacturonase has been
shown to be the primary enzyme for degrading pectin, the ma-
jor constituent of the middle lamellar region of the cell wall
(Smith et al., 1988, 1990; Giovannoni et al., 1989). Although
polygalacturonase is not necessary for fruit softening, its ac-
tivity appears to increase the susceptibility of ripe fruit to
cracking and subsequent colonization by microorganisms
(Schuch et al., 1991). We speculate that localization of poly-
galacturonase activity in the outer pericarp of ripe fruit facilitates
microbial infection, resulting in fruit decomposition and the
release of seeds.
SpatSal Regulation of PG Transcription 1 s lnfluenced by
Multiple Regulatory Regions
The spatial distribution of PG gene transcription in ripe fruit
was influenced by three regulatory regions defined by 5’ pro-
moter deletion analysis. The results are summarized in Table
1. Removal of a negative regulatory region between -1411 and
-1150 resulted in a uniform distribution of PG-GUSgene ex-
pression throughout the fruit pericarp (Figures 4 and 5). Thus,
at a minimum, this regulatory region functions to inhibit PG
gene transcription in the inner pericarp of ripe fruit. Because
we have not quantitatively determined the effect of this dele-
tion on D-glucuronidase enzyme activity specifically in the outer
pericarp, it is possible that the region from -1411 to -1150
also functions in the outer pericarp to reduce gene transcrip-
tion. Ultimately, this issue can best be resolved by performing
experiments in which the -1411 to -1150 regulatory region
alone is used to regulate transcription of a minimal promoter
~ ~~ ~
Table 1. Summary of Regulatory Regions That lnfluence the
Spatial Distribution of PG-GUS Gene Expression in Ripe
Site of Action
-1411 to -1150
-806 to -443
-231 to -134
a ND, not determined.
in tomato fruit harvested from transgenic plants. It is interest-
ing to note that no P-glucuronidase activity was detected in
unripe fruit (Figure 2B), leaves (J. Montgomery and R.L.
Fischer, data not shown), or roots (M.
munication) upon deletion of the negative regulatory region.
This result suggests that the absence of PG gene transcrip-
tion in unripefruit or in nonfruit tissues is not due to the negative
regulatory region at -1450 to -1150.
Two classes of positive regulatory elements were defined
in these experiments. Deletion of the positive regulatory re-
gion from -806 to -443 reduced PG-GUS transcription
specifically in the inner pericarp (Figure 4D). Thus, at a mini-
mum, this region functions to activate PG gene transcription
in the inner pericarp of ripe fruit. Because we have not
quantitatively determined the effect of this deletion on
P-glucuronidase enzyme activity in the outer pericarp, it is pos-
sible that the region from -806 to -443 also increases gene
transcription in the outer pericarp. By contrast, the positive
regulatory region from -231 to -134 activated PG gene tran-
scription specifically in the outer pericarp and vascular regions,
and not the inner pericarp of ripe fruit (Figures 4 and 5). Taken
together, our results showed that the expression pattern of PG
in ripe fruit pericarp arises out of complex interactions between
both positive and negative regulatory elements on the PG
Spatial regulation of gene expression in plants by the com-
binatorial interaction of regulatory promoter elements has been
described for the bean p-phaseolin gene (Bustos et al., 1991;
Burow et al., 1992) and the cauliflower mosaic virus-35s gene
(Benfey and Chua, 1990). Histochemical localization of GUS
reporter gene expression in transgenic tobacco embryos
demonstrated that two positive regulatory regions in the bean
P-phaseolin promoter, UAS7 and UAS2, conferred spatial regu-
lation of gene expression within different regions of the embryo
(Bustos et al., 1991). Specifically, expression of the GUS
reporter gene in the cotyledons and the shoot apex was de-
pendent upon UAS7, but expression in the hypocotyl required
the addition of UAS2. In a separate experiment, a negative
regulatory element in the bean p-phaseolin promoter was
shown to inhibit expression of p-phaseolin mRNA in the stems
and roots of transgenic tobacco plants (Burow et al., 1992).
Hawes, personal com-
Spatial Regulation of Polygalacturonase 1059
Nuclear factors that bind in vitro to DNA sequences within
each of the three regulatory regions of the PG promoter were
detected in unripe and ripe fruit (Figures 6 and 7). The distri-
bution of these nuclear proteins within the fruit pericarp was
not determined; therefore, we cannot assign a direct role for
these DNA binding proteins in the spatial regulation of PG gene
expression. Because all of these factors bind predominantly
AT-rich DNA sequences, it is possible that they are HMG-like
proteins and may function to facilitate protein-protein interac-
tions by inducing DNA bending (Giese et al., 1992; Thanos
and Maniatis, 1992). Most likely, other DNA binding factors not
detected under these binding conditions are involved in acti-
vation or suppression of PG gene expression in ripe fruit
pericarp. Nevertheless, based on the deletion analysis
presented here, we can begin to speculate on a mechanism
for the spatial regulation of PG gene expression in ripe fruit
pericarp. One possibility is that factors acting at the negative
regulatory region from -1411 to -1150 interfere with factors
acting at the positive regulatory region from -806 to -443
(Levine and Manley, 1989). As a result of this antagonistic in-
teraction, transcription of the PG-GUS gene in the outer
pericarp would be controlled primarily by the positive regula-
tory region from -231 to -134. Alternatively, factors acting
at positive and negative regulatory regions might exert their
effects directly, by influencing the formation of the basal tran-
scription complex (Levine and Manley, 1989).
Expression Pattern of the Ethylene Concentration-
lndependent PG Gene 1 s Different from the Ethylene
Concentration-Dependent E4 Gene
The spatial distribution of PG gene expression during fruit
ripening is strikingly different from that of another fruit ripen-
ing-related gene, €4, that is ethylene responsive (Lincoln et
al., 1987; Lincoln and Fischer, 1988). Whereas PG gene ex-
pression is restricted to the outer pericarp and vascular regions,
an €4-GUS gene containing 1421 bp of E4 5' flanking se-
quences is expressed uniformly throughout the pericarp and
collumella of ripe fruit (Figure 3 ) . A similar pattern of €4-GUS
gene expression was observed in ethylene-treated unripe fruit
(J. Deikman and R.L. Fischer, unpublished results). These
results are consistent with the rapid diffusion of ethylene
through the fruit and suggest that all cells in the pericarp are
able to induce E4 gene expression in response to an increase
in ethylene concentration. By contrast, expression of the PG
gene only in the outer pericarp suggests that the signals that
regulate ethylene concentration-independent gene expression
are not uniformly distributed throughout the fruit. Therefore
in ripening fruit, ethylene concentration-dependent regulation
may function to express genes throughout the fruit, whereas
ethylene concentration-independent regulation may be neces-
sary for spatially restricted gene expression in the pericarp.
These experiments demonstrate that the spatial control
of gene transcription in ripe tomato fruit is complex. The
cell-specific transcription of the ethylene concentration-
independent PG gene, mediated by both positive and nega-
tive regulatory elements, is in sharp contrast to the uniform
transcription of the ethylene concentration-dependent E4 gene,
which is controlled by a positive ethylene-responsive element
(Montgomery et al., 1993). Study of the mechanisms that are
involved in the ripening-regulated transcription of these genes
will provide insight into the developmental and molecular sig-
nals that induce ripening.
Tomato (Lycopersicon esculentum cv VFNT Cherry and cv Alcia Craig)
plants were grown under standard greenhouse conditions. Fruit matu-
rity stage was determined as described in Lincoln et al. (1987). Unripe
fruit were at mature green stage 1 in which the locular tissue was firm,
and ripe fruit were orange or firm red fruit.
Reporter Gene Constructs
The genomic clone pPGmutHH1.5 contains 1.4 kb of polygalact-
uronase (PG) 5' flanking sequences and an introduced Ncol site at
the ATG translation initiation codon (Giovannoni et al., 1989). A DNA
fragment containing PG 5' flanking sequences was isolated from
pPGmutHHl.5 by digestion with Hindlll and Ncol restriction endo-
nucleases, and the Ncol site was made blunt using the Klenow fragment
of DNase polymerase 1. To construct a PG-p-glucuronidase (GUS)
chimeric gene, this Hindlll-Ncol (blunt) PG promoter fragment was
ligated to a Smal-EcoRI restriction fragment containing GUS coding
sequences and nopaline synthase 3' polyadenylation sequences iso-
lated from pBllOl.3(Jefferson et al., 1987) and pUCl18digested with
Hindlll and EcoRl restriction enzymes. The promoter deletions to
- 1150, - 806, and - 443 were generated with Exolll as described
by Henikoff (1984) and deletion end points were determined by com-
parison of the DNA sequence to the published PG promoter sequence
(Birdetal., 1988;Roseetal., 1988).The -231, -134,and +5dele-
tions were created using Maelll, Mboll, and Mnll restriction sites,
Chimeric PGGUS genes were subcloned into the shuttle vector pMW1
(de Block et al., 1984) at the EcoRl site or between the EcoRl site
and the Smal site and then transferred into the disarmed Agrobec-
terium tumefaciens pGV3850 Ti plasmid (van Haute et al., 1983). Plant
transformation was performed using sterile tomato (cv Alcia Craig)
cotyledons as described by Deikman and Fischer (1988), and trans-
genic plants were selected by rooting in the presence of 25 mglL of
kanamycin. The presence of the PG-GUS transgene in individual trans-
formed plants was confirmed by neomycin phosphotransferase II
activity assays (McDonnell et al., 1987). Transgenic plants regener-
ated from separate calli were considered to represent independent
transformation events. DNA gel blot analysis was performed on in-
dividual transgenic plants regenerated from a single callus to determine
if they represented independent transformation events.
1060 The Plant Cell
Determination of fl-Glucuronidase Activity
Tissue was ground in liquid nitrogen, lysis buffer (Jefferson, 1987)
was added, and the tissue was homogenized until thawed. Extracts
were centrifuged in a microcentrifuge for 15 min at 4%. and the su-
pernatants were removed to a clean tube. Protein concentration of
extracts was measured using a BCA Protein Assay Reagent (Pierce
Chemical Co.). B-Glucuronidase activity was determined using 4-me-
thylumbelliferyl P-o-glucuronide (Clontech, Palo Alto, CA) according
to the method of Jefferson (1987), except that 20% methanol was in-
cluded in the assay buffer (Kosugi et al., 1990). Fluorescence of
4-methylumbelliferone was measured on a luminescence spec-
trophotometer (LS 30; Perkin-Elmer) at an excitation wavelength of
365 nm and an emission wavelength of 455 nm. For histochemical
staining, fruit was sectioned with a sharp knife and equilibrated in
50 mM sodium phosphate, pH 8, 50 pM potassium ferricyanide, 50
pM potassium ferrocyanide, 0.2% Triton X-100, 20% methanol for
1 hr at room temperature. The slices were then transferred to the same
buffer with the addition of 1 mg/mL 5-bromo-4-chloro-3-indolyl
glucuronide (Molecular Probes, Inc., Eugene, OR) and incubated at
room temperature. Stained fruit was photographed under tungsten
light with Kodak Ektachrome 160T film.
Nuclear Protein lsolation and DNase I Footprinting Reactions
Tomato fruit (cv VFNT cherry) was harvested and nuclear protein ex-
tracts were prepared from unripe (mature green stage 1) fruit pericarp
and ripe (30% red) fruit pericarp essentially as described by Manzara
et al. (1991), except that frozen tissue was ground in liquid nitrogen,
the homogenization buffer was pH 7.5, and the nuclei lysis buffer and
dialysis buffer were pH 8. Protein concentrations were determined
using the BCA Protein Assay Reagent. Preparation of 32P-end-
labeled PG promoter fragments and DNase I footprinting reactions
were performed as described by Manzara et al. (1991). Footprinting
reactions contained 2 fmol of 3zP-labeled DNA and 30 pg of nuclear
protein, were digested with DNase I ata final concentration of I pglmL,
and were loaded onto either 6 or 8% denaturing acrylamide gels in
1 x Tris-borate-EDTA.
We thank Dr. Thianda Manzara for guidance on nuclear protein ex-
traction and DNase I footprinting. We express our gratitude to Barbara
Rotz, John Franklin, Thom Koupal, and Louise Hancock for providing
excellent greenhouse services. We thank Dr. John Harada for criti-
cally reading the manuscript. This research was supported by United
States Department of Agriculture Grant No. 9103085.
Received April 23, 1993; accepted July 7, 1993.
Ahrens, M.J., and Huber, D.J. (1990).
termination of ripening tomato fruit.
Physiology and firmness de-
Physiol. Plant. 78, 8-14.
Benfey, P.N., and Chua, N.-H. (1990). The cauliflower mosaic virus
35s promoter: Combinatorial regulation of transcription in plants.
Science 250, 959-966.
Biggs, M.S., Harriman, R.W., and Handa, A.K. (1986). Changes in
gene expression during tomato fruit ripening. Plant Physiol. 81,
Bird, C.R., Smith, C.J.S., Ray, J.A., Moureau, P . , Bevan, M.W., Bird,
AS., Hughes, S . , Morris, P.C., Grierson, D., and Schuch, W . (1988).
The tomato polygalacturonase gene and ripening-specific expres-
sion in transgenic plants. Plant MOI. Biol. 11, 651-662.
Bird, C.R., Ray, J.A., Fletcher, J.D., Boniwell, J.M., Bird, AS.,
Teulieres, C., Blain, I., Bramley, M., and Schuch, W . (1991). Using
antisense RNA to study gene function: lnhibition of carotenoid bio-
synthesis in transgenic tomatoes. BioTTechnology 9, 635-639.
Brady, C . J . (1987). Fruit ripening. Annu. Rev. Plant Physiol. 38,155-178.
Brady, C.J., Meldrum, S.K., McGlasson, W.B., and Ali, Z.M. (1983).
Differential accumulation of the molecular forms of polygalacturonase
in tomato mutants. J. Food Biochem. 7, 7-14.
Brady, C.J., McGlasson, W.B., and Speirs, J . (1987). The biochem-
istry of fruit ripening. In Tomato Biotechnology, D. Nevins and R.
Jones, eds (New York: Alan R. Liss), pp. 279-288.
Bruce, W.B., Deng, X.-W., and Quail, P.H. (1991). A negativelyacting
DNA sequence element mediates phytochrome-directed repression
of phyA gene transcription. EMBO J. 10, 3015-3024.
Burow, M.D., Sen, P . , Chlan, C.A., and Murai, N. (1992). Develop-
mental control of the P-phaseolin gene requires positive, negative,
and temporal seed-specific transcriptional regulatory elements and
a negative element for stem and root expression. Plant J. 2,537-548.
Bustos, M.M., Guiltinan, M.J., Jordano, J., Begum, D., Kalkan, F.A.,
and Hall, T.C. (1989). Regulation of P-glucuronidase expression in
transgenic tobacco plants by an ATT-rich, cis-acting sequence found
upstream of a French bean P-phaseolin gene. Plant Cell 1,839-853.
Bustos, M.M., Begum, D., Kalkan, F.A., Battraw, M.J., and Hall,
T.C. (1991). Positive and negative cis-acting DNA domains are re-
quired for spatial and temporal regulation of gene expression by
a seed storage protein promoter. EMBO J. 10, 1469-1479.
Cass, L.G., Kirven, K.A., and Christoffersen, R.E. (1990). lsolation
and characterization of a cellulase gene family member expressed
during avocado fruit ripening. MOI. Gen. Genet. 223, 76-86.
Cordes, S . , Deikman, J., Margossian, L.J., and Fischer, R.L. (1989).
lnteraction of a developmentally regulated DNA-binding factor with
sites flanking two different fruit-ripening genes from tomato. Plant
Cell 1 , 1025-1034.
Czarnecka, E., Ingersoll, J.C., and Gurley, W.B. (1992). AT-rich pro-
moter elements of soybean heat shock gene GmhsplZSE bind two
distinct sets of nuclear proteins in vitm. Plant MOI. Biol. 19,985-1000.
Datta, N., and Cashmore, A.R. (1989). Binding of a pea nuclear pro-
tein to promoters of certain photoregulated genes is modulated by
phosphorylation. Plant Cell I, 1069-1077.
de Block, M., Herrera-Estrella, L., Van Montagu, M., Schell, J., and
Zambryski, P . (1984). Expression of foreign genes in regenerated
plants and in their progeny. EMBO J. 3, 1681-1689.
Deikman, J., and Fischer, R.L. (1988). lnteraction of a DNA-binding
factor with the 5’-flanking region of an ethylene-responsive fruit ripen-
ing gene from tomato. EMBO J. 7 , 3315-3320.
Deikman, J., Kllne, R., and Fischer, R.L. (1992). Organization of ripen-
ing and ethylene regulatory regions in a fruit-specific promoter from
tomato (Lycopersicon esculentum). Plant Physiol. 100, 2013-2017.
Spatial Regulation of Polygalacturonase
DellaPenna, D., Kates, D.S., and Bennett, A.B. (1986). Molecular
cloning of tomato fruit polygalacturonase: Analysis of polygalac-
turonase mRNA levels during ripening. Proc. Natl. Acad. Sci. USA
DellaPenna, D., Lincoln, J.E., Fischer, R.L., and Bennett, A.B.
(1989). Transcriptional analysis of polygalacturonase and other ripen-
ing associated genes in Rutgers, rin, nor, and Nrtomato fruit. Plant
Physiol. 90, 1372-1377.
Fischer, R.L., and Bennett, A.B. (1991). Role of cell wall hydrolases
in fruit ripening. Annu. Rev. Plant Physiol. Plant MOI. Biol. 42,
Giese, K., Cox, J., and Grosschedl, R. (1992). The HMG domain of
lymphoid enhancer factor 1 bends DNA and facilitates assembly
of functional nucleoprotein structures. Cell 69, 185-195.
Giovannoni, J.J., DellaPenna, D., Bennett, A.B., and Fischer, R.L.
(1989). Expression of a chimeric polygalacturonase gene in trans-
genic rin (ripening inhibitor) tomato fruit results in polyuronide
degradation but not fruit softening. Plant Cell 1 , 53-63.
Gray, J., Picton, S., Shabbeer, J., Schuch, W., and Grierson, D.
(1992). Molecular biology of fruit ripening and its manipulation with
antisense genes. Plant MOI. Biol. 19, 69-87.
Grlerson, D., Tucker, G.A., Keen, J., Ray, J., Bird, C.R., andschuch,
W . (1986). Sequencing and identification of acDNA clone for tomato
polygalacturonase. Nucl. Acids Res. 14, 8595-8603.
Hamilton, A.J., Lycett, G.W., and Grierson, D. (1990). Antisense gene
that inhibits synthesis of the hormone ethylene in transgenic plants.
Nature 346, 384-387.
Hamilton, A.J., Bouzayen, M., and Grierson, D. (1991). Identifica-
tion of a tomato gene for the ethylene forming enzyme by expression
in yeast. Proc. Natl. Acad. Sci. USA 88, 7434-7437.
Henikoff, S. (1984). Unidirectional digestion with exonuclease III cre-
ates targeted breakpoints for DNA sequencing. Gene 28,351-359.
Jacobsen, K., Laursen, N.B., Jensen, E.& Marcker, A., Poulsen,
C., and Marcker, K.A. (1990). HMG 1-like proteins from leaf and
nodule nuclei interact with different AT motifs in soybean nodulin
promoters. Plant Cell 2, 85-94.
Jefferson, R.A. (1987). Assaying chimeric genes in plants: The GUS
gene fusion system. Plant MOI. Biol. Rep. 5, 387-405.
Jefferson, R.A., Kavanagh, T.A., and Bevan, M.V. (1987). GUS fu-
sions: p-Glucuronidase as a sensitive and versatile gene fusion
marker in higher plants. EM60 J. 6, 3901-3907.
lnteraction of a nodule specific, trans-acting factor with distinct DNA
elements in the soybean leghaemoglobin lbc3 5' upstream region.
EM60 J. 7 , 1265-1271.
Marcker, K.A., Schell, J., and de Bruijn, F.J. (1988).
Jofuku, K.D., Okamuro, J.K., and Goldberg, R.B. (1987). Interac-
tion of an embryo DNA binding protein with a soybean lectin gene
upstream region. Nature 328, 734-737.
Jordano, J., Almoguera, C., and Thomas, T.L. (1989). A sunflower
helianthinin gene upstream sequence ensemble contains an en-
hancer and sites of nuclear protein interaction. Plant Cell 1,855-866.
Kosugi, S., Ohashi, Y., Nakajlma, K., and Arai, Y. (1990). An improved
assay for &glucuronidase in transformed cells: Methanol almost com-
pletely suppresses a putative endogenous p-glucuronidase activity.
Plant Sci. 70, 133-140.
Levine, M., and Manley, J.L. (1989). Transcriptional repression of eu-
karyotic promoters. Cell 59, 405-408.
Lincoln, J.E., and Fischer, R.L. (1988). Diverse mechanisms for the
regulation of ethylene-inducible gene expression. MOI. Gen. Genet.
Lincoln, J.E., Cordes, S., Read, E., and Fischer, R.L. (1987). Regu-
lation o f gene expression by ethylene during Lycopersicon esculentum
(tomato) fruit development. Proc. Natl. Acad. Sci. USA 84,2793-2797.
Mansson, P.E., Hsu, D., and Stalker, D. (1985). Characterization of
fruit specific cDNAs from tomato. MOI. Gen. Genet. 200, 356-361.
Manzara, T., Carrasco, P., and Gruissem, W . (1991). Developmental
and organ-specific changes in promoter DNA-protein interactions
in the tomato rbcS gene family. Plant Cell 3, 1305-1316.
Maxam, A., and Gilbert, W . (1980). Sequencing end-labeled DNA
with base-specific chemical cleavages. Methods Enzymol. 65,
McDonnell, R.E., Clark, R.D., Smith, W.A., and Hinchee, M.A. (1987).
A simplified method for the detection of neomycin phosphotrans-
ferase II activity in transformed plant tissues. Plant MOI. Biol. Rep.
McGlasson, W.B. (1985). Ethylene and fruit ripening. HortScience 20,
Montgomery, J., Goldman, S., Deikmann, J., Margosslan, L., and
Fischer, R.L. (1993). ldentification of an ethylene-responsive region
in the promoter of a fruit ripening gene. Proc. Natl. Acad. Sci. USA
Oeller, P.W., Min-Wong, L., Taylor, L.P., Plke, D.A., and Theologis,
A. (1991). Reversible inhibition of tomato fruit senescence by an-
tisense RNA. Science 254, 437-439.
Pear, J.R., Rldge, N., Rasmussen, R., Rose, R.E., and Houck, C.M.
(1989). lsolation and characterization of a fruit-specific cDNA and
the corresponding genomic clone from tomato. Plant MOI. Biol. 13,
Pedersen, T.J., Arwood, L.J., Spiker, S., Guiltinan, M.J., and
Thompson, W . F . (1991). High mobility group chromosomal proteins
bind to AT-rich tracts flanking plant genes. Plant MOI. Biol. 16,95-104.
Ray, J., Mouteau, P., Bird, C., Bird, A., Grlerson, D., Maunders,
M., ltuesdale, M., Bramley, P . , and Schuch, W . (1992). Cloning
and characterization of a gene involved in phytoene synthesis from
tomato. Plant MOI. Biol. 19, 401-404.
Rhodes, M.J.C. (1980). The maturation and ripening of fruits. In Senes-
cence in Plants, K.V. Thimann, ed (Boca Raton, FL: CRC Press),
Rose, R.E., Houck, C.M., Monson, E.K., DeJesus, C.E., Sheehy,
R.E., and Hiatt, W.R. (1988). The nucleotide sequence of the 5'
flanking region of a tomato polygalacturonase gene. Nucl. Acids
Res. 16, 7191.
Rottman, W.H., Peter, G.F., Oeller, P . W . , Keller, J.A., Shen, N.F.,
Nagy, B.P., Taylor, L.P., Campbell, A.D., and Theologis, A. (1991).
1-Aminocyclopropane-1-carboxylate synthase in tomato is encoded
by a multigene family whose transcription is induced during fruit
and floral senescence. J. MOI. Biol. 222, 937-961.
Schuch, W., Kanczler, J., Robertson, D., Hobson, G., Tucker, G.,
Grierson, D., Bright, S., and Bird, C . (1991). Fruit quality charac-
teristics of transgenic tomato fruit with altered polygalacturonase
activity. HortScience. 26, 1517-1520.
Sheehy, R.E., Pearson, J., Brady, C.J., and Hiatt, W.R. (1987). Mo-
lecular characterization of tomato fruit polygalacturonase. MOI. Gen.
Genet. 208, 30-36.
The Plant Cell
Sheehy, R.E., Kramer, M., and Hiatt, W.R. (1988). Reductjon of poly-
galacturonase activity in tomato fruit by antisense RNA. Proc. Natl.
Acad. Sci. USA 85, 8805-8809.
Slater, A., Maunders, M.J., Edwardq K., Schuch, W . , and Grierson,
D. (1985). lsolation and characterization of cDNA dones for tomato
polygalacturonase and other ripening related proteins. Plant MOI.
Biol. 5, 137-147.
Smith, C.J., Watson, C.F., Ray, J., Bird, C.R., Morris, P.C., Schuch,
W . , and Grierson, D. (1988). Antisense RNA inhibition Of polyga-
lacturonase gene expression in transgenic tomatoes. Nature 334,
Smith,C.J.S., Watson, C.F., Morrls, P.C., Bird,CR.,Seymour, G.B.,
Gray, J.E., Arnold, C., Tucker, G.A., Schuch, W., Harding, S.,
and Grierson, D. (1990). lnheritance and effect on ripening of an-
tisense polygalacturonase genes in transgenictomatoes. Plant MOI.
Biol. 14, 369-379.
Solomon, M.J., Strauss, F., and Varshavsky, A. (1986). A mammalian
high mobility group protein recognizes any stretch of six AT base
pairs in duplex DNA. Proc. Natl. Acad. Sci. USA 83, 1276-1280.
Speirs, J., Lee, E., Brady, D.J., Robertson, J., and McGlasson, W.B.
(1989). Endopolygalacturonase: Messenger RNA, enzyme and
softening in the fruit of a range of tomato genotypes. J. Plant Phys-
iol. 135, 576-582.
Thanos, D., and Maniatis, T. (1992). The high mobility group protein
HMG(Y) is required for NF-KB-dependent virus induction of the hu-
man IFN-P gene. Cell 71, 7i7-789.
Theologis, A. (1992). One rotten apple spoils the whole bushel: The
role of ethylene in fruit ripening. Cell 70, 181-184.
Tieman, D.M., and Handa, A.K. (1989). lmmunocytolocalization of
polygalacturonase in ripening tomato fruit. Plant Physiol. 90, 17-20.
Travis, A., Amsterdam, A., Belanger, C., and Grosschedl, R. (1991).
LfF-7, a gene enccding a lymphoid-specific protein with an HMG
domain, regulatesT-cell receptor a enhancer function. Genes Dev.
van der Straeten, D., van Wiemeersch, L., Goodman, H.M., and
Vsn Montagu, M. (1989). Purification and partia1 characterization
of I-aminocyclopropane-l-carboxylate synthase from tomato pericarp.
Eur. J. Biochem. 182, 639-647.
van Haute, E., Joos, H., Maes, M., Warren, G., Van Montagu, M.,
and Schell, J. (1983). lntergenic transfer and exchange recombi-
nation of restriction fragments cloned in pBR322: A nove1 strategy
for the reversed genetics of the Ti plasmids of Pgrobacterium tumefa-
ciens. EMBO J. 2, 411-417.
Yang, S.F., and Hoffman, N.E. (1984). Ethylene biosynthesis and its
regulation in higher plants. Annu. Rev. Plant Physiol. 35, 155-189.
DOI 10.1105/tpc.5.9.1049 Download full-text
J. Montgomery, V. Pollard, J. Deikman and R. L. Fischer
Transcription in Tomato Fruit Pericarp
Positive and Negative Regulatory Regions Control the Spatial Distribution of Polygalacturonase
This information is current as of July 15, 2011
Sign up for eTOCs at:
Sign up for CiteTrack Alerts at:
is available at:
The Plant Cell
Subscription Information for
ADVANCING THE SCIENCE OF PLANT BIOLOGY
© American Society of Plant Biologists