Plant Molecular Biology 53: 151–162, 2003.
© 2003 Kluwer Academic Publishers. Printed in the Netherlands.
151
A 2.5-kb insert eliminates acid soluble invertase isozyme II transcript in
carrot (Daucus carota L.) roots, causing high sucrose accumulation
Yuan-Yeu Yau1,2 and Philipp W. Simon1,∗
1 USDA-ARS
Vegetable Research Crops Unit and Department of Horticulture, University of Wisconsin-Madison,
1575 Linden Drive, Madison, WI 53706, USA; 2 Current address: Plant Gene Expression Center, USDA-ARS, Albany, CA94710, USA and Plant & Microbial Biology, University of California, Berkeley, CA94720, USA (∗ author
for correspondence, e-mail [email protected])
Received 6 June 2003; accepted in revised form 21 August 2003
Key words: carrot, Daucus carota, insert, invertase, knockout mutant, sugar metabolism
Abstract
The predominant storage carbohydrates of mature carrot (Daucus carota L.) storage roots typically are the free
sugars glucose and fructose. This trait is conditioned by the Rs allele. A naturally occurring recessive mutation,
rs/rs, conditions a shift from these reducing sugars to sucrose. RT-PCR and sequencing revealed a unique 2.5 kb
insert in the first and largest intron near the 5 end of the acid soluble invertase isozyme II gene of rs/rs carrots. This
insert was not totally spliced out during mRNA processing. While the wild-type acid-soluble invertase isozyme II
transcript (ca. 2 kb) was detected in Rs/Rs roots and leaves, none was observed in rs/rs roots throughout development. RT-PCR of rs/rs leaves revealed two novel transcripts (2.7 kb and 3.2 kb). A comparison of enzyme activity
between the near-isogenic Rs/Rs and rs/rs carrot lines revealed very low acid-soluble invertase activity in rs/rs
roots whereas neutral invertase, sucrose synthase and sucrose phosphate synthase levels were comparable. Those
results and linkage analysis indicate that Rs is a candidate locus for carrot vacuolar acid-soluble invertase isozyme
II. Although the 2.5 kb insert does not occur in the Rs wild-type acid-soluble invertase isozyme II allele, it does
occur elsewhere in the genome of Rs/Rs plants.
Introduction
Starch, proteins and lipids are the primary storage
compounds in seeds while storage roots, tubers, and
rhizomes usually store complex carbohydrates such as
starch and fructans as energy reserves. However, some
plants, like carrot, sugar beet and sugar cane, accumulate significant amounts of free sugar in vacuoles. Key
enzymes in storage organs that regulate sucrose metabolism are sucrose-phosphate synthase (EC 2.4.1.14),
sucrose phosphatase (EC 3.1.3.24), sucrose synthase
(EC 2.4.1.13) and invertase (or β-fructofuranosidase;
EC 3.2.1.26) (Tang et al., 1999). Of these, sucrosephosphate synthase and sucrose phosphatase are involved in sucrose synthesis while sucrose synthase
and invertase take part in sucrose breakdown (Copeland, 1990). An understanding of enzyme expression
involved in carbon metabolism is necessary to manip-
ulate allocation of photoassimilates to sink organs and
to thus modify the production of carbohydrates to meet
our future food needs (Muller-Rober et al., 1992).
Sucrose is the major end product of leaf photosynthesis and is the major sugar transported in the
phloem of most higher plants. However, sucrose cannot be used directly for most metabolic processes and
must be cleaved into hexoses by invertase or sucrose
synthase before use. Invertases are present in most
plant tissues and irreversibly catalyze the breakdown
of disaccharide sucrose into fructose and glucose, and
similar reactions with related sugars. Isoforms of invertases are characterized and classified according to
pH optima (acid, neutral and alkaline), subcellular
locations (vacuole or cell wall), and solubility properties (soluble or insoluble) in several plants including
carrot (e.g. Stommel and Simon, 1990; Sturm and
Chrispeels, 1990; Sturm et al., 1995). Acid invertase
152
has a pH optimum between 3 and 5, whereas neutral
and alkaline invertases have pH optima of about 7 and
8 respectively (Lee and Sturm, 1996). Acid invertases
are divided into vacuolar or soluble and extracellular or insoluble forms, with an acidic (Unger et al.,
1994) or basic (Laurière et al., 1988) pI. It has been
suggested that acid invertases are involved in phloem
unloading, control of sugar type in storage organs
(Klann et al., 1993), normal development of endosperm (Cheng et al., 1996), and response to pathogen
infection or wounding (Sturm and Chrispeels, 1990;
Zhang et al., 1996).
To study the roles of invertases and their isozyme
expression in different subcellular compartments,
transgenic plants over-expressing yeast-derived invertases and repressing antisense constructs of invertase
isozymes have been produced in tobacco, Arabidopsis, potato and carrots (von Schaewen et al., 1990;
Sonnewald et al., 1991, 1997; Heineke et al., 1994;
Tang et al., 1999). Expression of a yeast-derived invertase in the cell wall of tobacco and Arabidopsis plants
leads to accumulation of carbohydrate and inhibition
of photosynthesis, and strongly influences the growth
and phenotype of the transgenic tobacco plants (von
Schaewen et al., 1990). Compared to wild type plants
with an average leaf-to-root ratio of 1:3, transgenic
carrots over-expressing cell-wall invertase antisense
gene had a ratio of 17:1, whereas transgenic carrot
plants expressing antisense vacuolar acid soluble invertase isozyme I gene developed normally with an
average leaf-to-root ratio of 1.5:1 (Tang et al., 1999).
While naturally-occurring mutations for genes in the
carbohydrate biosynthetic pathway have been identified in other plants, none have been characterized at
the molecular level in carrot.
In carrots, free sugars (sucrose, glucose, fructose)
are the major reserves in mature roots (Alabran and
Mabrouk, 1973). The type of sugar accumulated in
carrot roots is conditioned by a single dominant gene,
Rs (Freeman and Simon, 1983). Carrots with the Rs/genotype accumulate predominantly reducing sugars
glucose and fructose, while rs/rs carrots accumulate
predominantly sucrose. This finding is similar to those
in tomato where sucrose accumulation is a monogenic
recessive trait (sucr) (Yelle et al., 1991). In nature,
the Rs allele occurs in nearly all carrots with only
rare incidence of the rs allele (Freeman and Simon,
1983). However, the molecular basis of the Rs gene is
unknown.
Wild-type carrot acid-soluble invertase isozymes I
and II, cell-wall invertase and sucrose synthase have
been purified, characterized, and their cDNAs have
been sequenced (Sturm and Chrispeels, 1990; Unger et al., 1994). We recently have developed nearisogenic inbred lines of Rs/Rs and rs/rs in the carrot inbred B4367 (B4367Rs and B4367rs, respectively). The availability of these lines in combination with the sequence information of carrot wildtype invertases has allowed us to use comparative
reverse transcriptase-polymerase chain reaction (RTPCR) (Simpson et al., 1992) and molecular cloning
techniques to determine the probable molecular basis
of the Rs locus and mutant rs allele in carrot. In this
paper, we describe the discovery of a 2.5 kb insert in an
intron of the vacuolar acid-soluble invertase isozyme
II (Sturm, 1996) in carrots homozygous for the rs
locus and its association with a low isozyme II activity
in carrot roots.
Materials and methods
Plant materials and growth conditions
Near-isogenic carrot lines B4367Rs and B4367rs were
developed by selecting for sugar types and progeny testing over four generations from a segregating
progenitor B4367 population (Simon and Freeman,
1985). Seeds of B4367Rs and B4367rs were sown in
a mixture of 2 parts field soil, 1 part peat moss, 1
part sand and 1 part vermiculite, and grown in clay
pots, with a 12–16 h photoperiod at 22–25 ◦ C in the
greenhouse, at the University of Wisconsin-Madison.
Sugar metabolic enzyme activity assay
Enzyme extraction, desalting, and assays for invertases and sucrose synthase were performed as described by Wang et al. (1993) except invertase assays
were performed at pH 4.5 and pH 7.5. Sucrose phosphate synthase was assayed as described by Stommel
(1992).
Total RNA extraction method
Total RNA was extracted with TRIzol Reagent
(Gibco-BRL Life Technologies, Gaithersburg, MD).
Tissues of near-isogenic lines B4367Rs and B4367rs
were collected at 1, 4, 6, 8, 17, 18 and 60 weeks
after planting. Due to their small size, entire plants
(leaves and roots) were evaluated at the 1- and 4week stages. At later stages (6, 8, 17, 18 and 60
weeks), plants were separated into leaves and roots for
153
total RNA extraction. One gram tissue samples were
analyzed. Protocols for RNA extraction followed instructions provided by the manufacturer, except 5 µl
of RNase OUT ribonuclease inhibitor (Gibco-BRL
Life Technologies) was added to each sample.
RT-PCR analysis
For first-strand full-length cDNA synthesis, the Superscript Pre-amplification System (Gibco-BRL Life
Technologies) was used to synthesize the first-strand
cDNA of carrot acid invertases and sucrose synthase. Oligo(dT)12−18 and gene-specific primers were
used to synthesize the first-strand cDNA of carrot
acid-soluble invertase isozyme I, II, cell-wall invertase and sucrose synthase. For preparing RNA/primer
tubes, 1 µl (0.5 µg) of oligo(dT)12−18 or 2 µM
gene-specific primer and 4–5 µg of total RNA was
added to a 0.2 ml, thin-walled micro reaction tube
(USA/Scientific Plastics, Ocala, FL). Nuclease-free
water was added to bring the final volume to 12 µl.
The tubes were incubated at 70 ◦ C for 10 min. Then
7 µl of reaction mixture (2 µl 10× PCR buffer, 2 µl
25 mM MgCl2 , 1 µl 10 mM dNTP mix, 2 µl 0.1 M
DTT) was added to each RNA/primer tube, the tubes
were incubated at 42 ◦ C for 5 min, 1 µl of Superscript II RNase H− reverse transcriptase (no RNase H
activity) was then added to each tube, and the tubes
were incubated at 42 ◦ C for 50 min. The reaction
was terminated at 70 ◦ C for 15 min, 1 µl of RNase
H was added to each tube, and tubes were incubated at 37 ◦ C for 30 min to digest the RNA template
from the cDNA:RNA hybrid molecule. A positive control RNA from the chloramphenicol acetyltransferase
(CAT; offered by the manufacturer) and a negative
control, without reverse transcriptase, were included
as well. For polymerase chain reaction (PCR) of firststrand cDNA, for each sample 2 µl of the first-strand
cDNA solution was incubated in 1× PCR buffer (as
described in instructions), dNTP mixture (0.25 mM),
1 µM of each primer (INV-1 and INV-2; INV-5 and
INV-6; CW-3 and CW-2; SS-1 and SS-2 or SS-3 and
SS-4), 1 U Ex-Taq DNA polymerase (Panvera Corporation, Madison, WI), and water for a total volume
of 20 µl. Two supplied PCR buffers, 10× LA buffer
and 10× Ex-Taq buffer, were used for amplification.
The PCR conditions for the Ex-Taq enzyme were: 1
cycle at 94 ◦ C for 3 min, 35 cycles at 94 ◦ C (30 s),
65 ◦ C (1 min) and 72 ◦ C (3.5 min), then finally 1
cycle at 72 ◦ C for 10 min. The amplification products
were analyzed in 1% TAE agarose gels stained with
ethidium bromide. The primers used for the amplification of the full-length acid invertase and sucrose
synthase cDNAs were based on the published carrot
sequences (Table 1). They included, for full-length
carrot acid-soluble invertase isozyme I cDNA, INV1 and INV-2; for full-length acid-soluble invertase
isozyme II cDNA, INV-5 and INV-6; for full-length
cell-wall invertase cDNA, CW-3 and CW-2; for 5
half-length sucrose synthase cDNA, SS-1 and SS-2;
for 3 half-length sucrose synthase cDNA, SS-3 and
SS-4. Three to four bases were added to the ends of
restriction sites XbaI and SacI, to ensure the restriction
sites could be cut during the digestion reaction for future cloning purpose. Primers INV-1 and INV-2 were
synthesized by the Biotechnology Center at the University of Wisconsin-Madison. INV-5, INV-6, CW-3,
CW-2, SS-1, SS-2, SS-3 and SS-4 were synthesized
from Gibco-BRL. Primers were desalted before use.
Gel extraction, cloning and sequencing
Target bands were cut out from gels and purified with
QIAquick Gel Extraction Kit and QIAquick PCR Purification Kit (QIAgen, Valencia, CA) following the
instructions provided by the manufacturer. Purified
DNA fragments were then ligated into either pGEMT vector (Promega, Madison, WI) or pCR4-TOPO
vector (TOPO TA Cloning kit, Invitrogen, Carlsbad,
CA) for sequencing. ABI PRISM BigDye PCR (Applied Biosystems, Foster City, CA) was used for
sequencing with an ABI Prism 377 DNA sequencer
in the University of Wisconsin-Madison Biotechnology Center. DNA sequences were analyzed with
the GCG (Wisconsin Package Version 9.0, Genetics
Computer Group, Madison, WI) sequence analysis
packages and ‘multiple alignment’ in CURATOOLS
package (http://curatools.curagen.com/). The resulting sequences were then compared to the published
cDNA sequence with ‘BLAST 2 sequences’ function
in the BLASTN program from Basic Local Alignment Search Tool (BLAST). DNA sequences were
further manipulated using on-line Curatools package.
All sequences were deposited in GenBank.
Genomic DNA extraction, Southern and northern
hybridization
Total genomic DNA extraction from carrot leaves for
Southern analysis was carried out according to the
method described by Murray and Thompson (1980),
with some modifications. 2× CTAB was used for extraction. For digestion, 10 µg DNA for each sample
154
Table 1. Primers used for comparative RT-PCR of carrot acid-soluble invertase isozyme I, II and cell-wall invertase, inserts ‘a746’ and
‘b498’ amplification, probe generation for Southern hybridization and PCR amplification of ends of the 2.5 kb insert for sequencing. The
underlined regions indicate restriction sites XbaI and SacI created for cloning purposes.
Primer name
Direction
Primer sequence
Product
INV-1
INV-2
INV-5
INV-6
INV-18
INV-22
INV-27
CW-2
CW-3
SS-1
SS-2
SS-3
SS-4
rs-INVII-copy-1
rs-INVII-copy-2
rsINVIImut-1
rsINVIImut-2
rsINVIImut-3
rsINVIImut-4
rsINVIImut-5
rsINVIImut-8
Sense
Antisense
Sense
Antisense
Antisense
Sense
Antisense
Antisense
Sense
Sense
Antisense
Sense
Antisense
Sense
Antisense
Sense
Antisense
Sense
Antisense
Sense
Antisense
5 -TTTTCTAGAGCTATTTAATTTTCCATCCAATGGATACCTACCA-3
5 -TTTGAGCTCGGTAGCTATATTACAAGGTTACAACTGATCGAAT-3
5 -TTTTTCTAGAGAATTCCATTCTTCATATAA TTAAAATGGAGCATCCAATC-3
5 -TTTTGAGCTCAAATTAAAATATACATGTCATATAACAAGATCTGCAAAATG-3
5 -TAAATGGTACCATCCCTTGTAGAAT-3
5 -GAATGCGGAGCCGCCGGCTAATT-3
5 -CCATCAGGGAGTATTGTCGCGGACCCG-3
5 -TTTGAGCTCCTTCCTGCATTGGTTAGTTCATTCGCAAGGGCTTC-3
5 -TTTTCTAGAAATTCTCTGAAATTCCAGGGATGGGTGTAACAAT-3
5 -TGATCACAATGGGTGAACCT-3
5 -TGGTACCTACTGCATCCGGC-3
5 -GCCGGATGCAGTAGGTACCA-3
5 -TGGGTCGATATGAAAACCAG-3
5 -ATATTGTTTGATCCTAAAACAGGG-3
5 -CATTGTTTTTACAATCATACACCTC-3
5 -GGAATTTAAGGAACTTCCAAAAC-3
5 -GTTAGAATTATCATCTGAAATTATAT-3
5 -GTATCCTCTCAATAACTATGTATAT-3
5 -AATAAAACCAAGCAAGAAAAATTTG-3
5 -CTAGGGAAAAAGGATCAAAATGAGGA-3
5 -GTAATCAATATAGATGTTTCAGAGTC-3
Aa
A
Bb
B
B
B
B
Cc
C
Dd
D
D
D
B
B
Ee
E
Ff
F
Gg
G
a PCR primers for amplification of carrot acid-soluble invertase isozyme I gene.
b PCR primers for amplification of carrot acid-soluble invertase isozyme II gene.
c PCR primers for amplification of carrot cell wall invertase gene.
d PCR primers for amplification of carrot sucrose synthase gene.
e PCR primers for amplification of insert ‘a746’.
f PCR primers for amplification of insert ‘b498’.
g PCR primers for amplifying fragments of the 2.5 kb insert.
was used. Digested DNA was separated on a 1%
TAE agarose gel at 15 mA for 18 h. The gel was
then soaked in denaturation buffer (0.4 M NaOH,
0.6 M NaCl). After being washed with 2× SSC,
the DNA was transferred onto a Zetaprobe membrane (BioRad, Hercules, CA) by capillary blotting
at room temperature for 12 h. The blots were then
washed with 2× SSC buffer, vacuum-baked at 80 ◦ C
for 1 h, and pre-hybridized overnight in hybridization buffer (6.4 ml deionized formamide, 32 µl
0.5 M EDTA, 4 ml 1 M Na2 HPO4 and 5.6 ml 20%
SDS) at 42 ◦ C. Two DNA probes were prepared
by PCR amplifying acid-soluble invertase isozyme
II gene in B4367Rs and insert in B4367rs with the
gene-specific primers rs-INVII-copy-1 and rs-INVIIcopy-2, and insert-specific primers rs-INVIImut-1 and
rs-INVIImut-2 respectively (Table 1). Probe generated from primers rs-INVII-copy-1and rs-INVII-copy-
2 was used for Southern and northern blotting to detect
carrot acid-soluble invertase isozyme II gene signal,
while probe generated from primers rs-INVIImut-1
and rs-INVIImut-2 was used for Southern blotting to
detect insert ‘a746’ fragments. The PCR product was
cut out of a 1% TAE agarose gel and purified with a
Geneclean II kit (Bio101, Vista, CA), and the concentration was measured with a fluorometer. The PCR
product was labeled with 5 -[α-32P]-triphosphate triethylammonium salt (Redivue deoxycitidine) (Amersham Pharmacia Biotech, Piscataway, NJ) by using a
DECAprime II kit (Ambion, Austin, TX). The probe
was passed through a Sephadex G-50-150 (particle
size 50–150 µm) (Sigma, St. Louis, MO) column to
remove the unincorporated nucleotides, and help eliminate background signal on the autorad. Hybridization
was done at 42 ◦ C overnight. Blots were washed as
follows: (1) 2× SSC for 10 min at room temperature,
155
(2) 0.2× SSC + 0.25% SDS at 60–65 ◦ C for 20 min.
Blots were visualized by autoradiography with Kodak
XAR X-ray film or the Cytoclone Storage Phosphor
System.
The GenBank accessions numbers for the sequences mentioned in this article are AY368232 (fulllength insert in carrot acid-soluble invertase isozyme
II gene), AY368233 (insert a746 sequence in carrot
acid-soluble invertase isozyme II cDNA), AY368234
(insert b498 sequence in carrot acid-soluble invertase
isozyme II cDNA).
ured according to Stommel (1992) and Wang et al.
(1993) in leaves, petioles, and three regions of the root
in B4367Rs and B4367rs carrot plants 14–28 weeks
old (Table 2). Enzyme activities of sucrose synthase
and sucrose phosphate synthase were comparable in
Rs and rs tissues throughout development whereas invertase activity was less in all tissues of rs plants than
Rs plants, with greater differences noted at pH 4.5 than
pH 7.5.
Acid-soluble invertase isozyme II transcript in
B4367Rs and B4367rs roots and leaves throughout
development
Results
Comparative RT-PCR of near-isogenic carrot lines
B4367Rs and B4367rs
Transcripts of carrot acid-soluble invertase isozyme I
and II, cell-wall invertase and sucrose synthase genes
in near-isogenic carrot lines B4367Rs and B4367rs
were compared by RT-PCR with those tissues and developmental stages where expression had been found
to be maximal, based on earlier northern analysis
(Sturm, 1996). Five pairs of gene-specific primers
for amplifying full-length transcripts by RT-PCR were
synthesized for these enzymes, based on their wildtype cDNA sequences (Table 1): INV-1/ INV-2 for
carrot acid-soluble invertase isozyme I cDNA; INV5/ INV-6 for carrot acid-soluble invertase isozyme II
cDNA; and CW-2/ CW-3 for carrot insoluble cell-wall
invertase. For sucrose synthase gene amplification,
two pairs of primers were synthesized: SS-1/ SS-2
for the 5 end, and SS-3/ SS-4 for the 3 end. Transcripts of carrot acid-soluble invertase isozyme I (Figure 1A), cell-wall invertase (Figure 1B) and sucrose
synthase (Figure 1D) were comparable and of expected size in tissues evaluated from the near-isogenic
lines B4367Rs and B4367rs as measured by RT-PCR.
We also found acid-soluble invertase isozyme II transcript in B4367Rs roots by RT-PCR (Figure 1C) as had
been observed by Sturm (1996) with northern blotting. However, surprisingly, no transcript of the carrot
acid-soluble invertase isozyme II gene was observed
when measured by RT-PCR for B4367rs carrot roots
(Figure 1C).
Acid-soluble invertase activity in B4367Rs and
B4367rs roots
Activities of pH 4.5 and pH 7.5 invertases, sucrose
synthase, and sucrose phosphate synthase were meas-
To investigate the expression patterns of carrot vacuolar acid-soluble invertase isozyme II throughout
plant development in B4367rs, plants 1, 4, 6, 8, 17,
18 and 60 weeks old were used for RT-PCR studies.
Consistently different expression patterns and RT-PCR
products of acid-soluble invertase isozyme II gene
were observed between B4367Rs and B4367rs plants
(Table 3).
Total RNA from carrot roots 1, 4, 6, 9 and 17
weeks old from both B4367Rs and B4367rs were also
isolated and used for northern analysis. Abundant carrot acid-soluble invertase isozyme II mRNA signal of
ca. 2 kb was detected in the 9-week old B4367Rs carrot roots, but not in roots 1, 4, 6 or 17 weeks old
(data not shown). This is in agreement with the results observed by Sturm, where abundant acid-soluble
invertase isozyme II mRNA was detected only in wildtype carrot roots 10–16 weeks old, but no significant
mRNA signal was detected in carrot roots 2, 4, 6,
8, 18, and 20 weeks old (Sturm, 1996). No wildtype transcript (2 kb band) was detected in northern
analysis of B4367rs roots of any age.
In leaves, no recognizable signal was detected with
northern hybridization by Sturm (1996) or by us; however, we observed abundant RT-PCR product of acidsoluble invertase isozyme II (ca. 2 kb) in the leaves of
B4367Rs with the same primers as for root analysis,
perhaps by virtue of to the higher detection sensitivity
of RT-PCR. Two novel transcripts of ca. 2.7 kb and
3.2 kb, present in B4367rs leaf tissue, were larger
than the ca. 2 kb product of the wild-type B4367Rs
acid-soluble invertase isozyme II (Unger et al., 1994)
(Figure 2A). These two novel RT-PCR products did
not appear for wild-type B4367Rs.
156
Figure 1. Sugar enzyme gene expression in near-isogenic B4367Rs (‘Rs’) and B4367rs (‘rs’) carrots. A. Total RNA prepared from 1-week old
whole plants was used as a template in RT-PCR assays with (+) or without (−) addition of reverse transcriptase with DNA primers (INV-1 and
INV-2) derived from carrot acid-soluble invertase isozyme I cDNA (see text). Chloramphenicol acetyltransferase served as an RT-PCR positive
control (RTP). P, cloned carrot acid-soluble invertase isozyme I cDNA in a vector. B. Total RNA prepared from 1-week old whole plants was
used as a template in RT-PCR assays with (+) or without (−) addition of reverse transcriptase with DNA primers (CW-3 and CW-2) derived
from carrot cell-wall invertase (see text). C. Total RNA prepared from 18-week old roots was used as a template in RT-PCR assays with (+) or
without (−) addition of reverse transcriptase with DNA primers (INV-5 and INV-6) derived from carrot acid-soluble invertase isozyme II (see
text). P, cloned carrot acid-soluble invertase isozyme II cDNA in a vector. D. Total RNA prepared from 1-week old whole plants was used as a
template in RT-PCR assays with (+) or without (−) addition of reverse transcriptase with DNA primers SS-1 and SS-2 (5 end), and SS-3 and
SS-4 (3 end) derived from carrot sucrose synthase to amplify the gene transcripts.
Genomic DNA of carrot acid soluble invertase
isozyme II gene in B4367Rs and B4367rs
With carrot acid-soluble invertase isozyme II genespecific primers INV-5 and INV-6 for amplification of
genomic DNA, PCR products of around 3.8 kb and
6.3 kb were generated for the B4367Rs and B4367rs
acid-soluble invertase isozyme II gene, respectively.
These data suggested a 2.5 kb insert in the rs allele
(Figure 2B and Table 4). Southern hybridization also
confirmed existence of a 2.5 kb insert (Figure 2C).
Acid-soluble invertase isozyme II cDNA inserts in
carrot B4367rs leaves
While genomic PCR and Southern analysis indicated
a 2.5 kb insert in the acid-soluble invertase isozyme
II gene of B4367rs plants, multiple RT-PCR products
were observed in B4367rs leaves (but not roots) and
they were all larger than the single product observed in
B4367Rs leaves (Figure 2A). To characterize these unusual B4367rs leaf transcripts, primers covering overlapping segments of the complete full-length wildtype acid-soluble invertase isozyme II cDNA were
synthesized (Table 1) for comparative RT-PCR. RT-
PCR products from near-isogenic counterparts were
the same size for the entire gene except for the 5 end
between primers INV-5 and INV-18. In contrast to RTPCR with primers INV-5 and INV-6 to cover the entire
gene, where two products (2.7 kb and 3.2 kb) were
amplified, we observed three cDNA variants, ‘b’, ‘c’
and ‘d’ with primers INV-5 and INV-18 (Figure 3A),
and in rare cases, a fourth band was amplified (termed
‘a’) (Figure 3B). Bands ‘a’ (484 bp), ‘b’ (1230 bp) and
‘d’ (1728 bp) were cloned and sequenced to demonstrate 49 bp, 746 bp and 1244 bp (Figure 4) portions
of the insertion integrated into the coding region of
carrot acid-soluble invertase isozyme II gene near the
5 end of the gene. Interestingly, the 1244 bp fragment
(from band ‘d’) comprised a 498 bp fragment (termed
‘b498’) within the 746 bp fragment (termed ‘a746’)
found in band ‘b’ (Figure 4).
Features of the 2.5 kb genomic insert in the B4367rs
acid-soluble invertase isozyme II gene
To further investigate the full-length insert sequence
at the genomic DNA level, two primer pairs, INV22 (sense primer) + rsINVIImut-8 (antisense primer)
and rsINVIImut-5 (sense primer) + INV-27 (antis-
157
Figure 2. cDNA and genomic DNA analyses of an acid-soluble invertase isozyme II gene in near- isogenic B4367Rs (‘Rs’) and B4367rs (‘rs’)
carrot plants. A. Total RNA prepared from B4367Rs and B4367rs plant leaves was used as template in RT-PCR assays with DNA primers
(INV-5 and INV-6) derived from carrot acid-soluble invertase isozyme II cDNA (see text). Resulting cDNA was separated on an 1% agarose
gel and stained with ethidium bromide. B. Genomic DNA prepared from B4367Rs and B4367rs carrot roots and leaves was used as template
for PCR amplification with DNA primers (INV-5 and INV-6) derived from carrot acid-soluble invertase isozyme II. Resulting PCR products
were separated on an 1% agarose gel and stained with ethidium bromide. C. Genomic DNA prepared from B4367Rs and B4367rs plants
was digested with both EcoRI and BglII and the blot was probed with 32 P-labeled PCR product amplified from primers rs-INVII-copy-1 and
rs-INVII-copy-2, which generate a portion of the wild-type carrot acid-soluble invertase isozyme II gene.
Table 2. Activities of sugar metabolic enzymes from tissues of B4367Rs (Rs) and B4367rs (rs) carrots during weeks 14–28
of growth.
Tissue
Age (weeks)
Enzymesa
pH 4.5 invertase
Rs rs
pH 7.5 invertase
Rs rs
sucrose synthase
Rs rs
sucrose-phosphate synthase
Rs rs
Leaf
14
16
20
28
2
1
0
1
2
0
1
0
7
5
1
0
5
3
0
0
6
7
0
0
6
11
0
0
0
0
0
0
0
0
0
0
Petiole
14
16
20
28
3
4
4
1
2
3
2
0
2
4
4
1
1
0*
1
0
10
15
1
0
10
12
0
0
0
0
1
0
0
0
1
0
Root crown
14
16
20
28
10
14
19
8
6
3∗
2∗
1∗
3
7
4
2
1
1∗
1
0
8
11
24
5
9
11
30
3
1
2
2
2
2
2
2
2
Mid root
14
16
20
28
17
18
13
6
5∗
2∗
1∗
0∗
4
6
4
0
1
1∗
1
0
14
20
37
4
17
23
30
7
3
14
3
1
5
16
4
1
Root tip
14
16
20
28
19
20
14
2
3∗
4∗
1∗
1
5
4
4
0
0∗
1
1
0
19
25
34
9
13
27
31
6
0
1
0
0
0
2
0
0
a Activities were measured as described by Stommel (1992) and Wang et al. (1993) and expressed as µmol reducing sugar
per gram tissue fresh weight per hour.
∗ Significant difference in Rs and rs enzyme activity (LSD
0.05 ).
158
Table 3. Acid-soluble invertase isozyme II mRNA expression in
the leaves and roots of the isogenic carrot lines B4367Rs and
B4367rs based on RT-PCR.
Agea
(weeks)
Carrot line
Whole plant
1
B4367Rs
B4367rs
B4367Rs
B4367rs
B4367Rs
B4367rs
B4367Rs
B4367rs
B4367Rs
B4367rs
B4367Rs
B4367rs
B4367Rs
B4367rs
+b
mc
+
m
4
6
8
17
18
60
Roots
–d
–
–
–
N/A
Leaves
+
m
+
m
+
m
+
m
N/Ae
m
+
+
+
+
+
Figure 3. Expression of an acid-soluble invertase isozyme II gene
from 18-week old B4367Rs (‘Rs’) and B4367rs (‘rs’) leaf tissue
measured as RT-PCR with primers INV-5 and INV-18. A. Three
bands (‘b’, ‘c’ and ‘d’) of ca. 1.2 kb, ca. 1.5 kb and ca. 1.7 kb
were amplified from carrot B4367rs. Lanes: 1, no DNA (negative
control); 2, B4367rs; 3, B4367Rs. B. In rare cases, four bands, ‘a’,
‘b’, ‘c’, and ‘d’ (463 bp) were amplified from carrot B4367rs. Only
one band (463 bp) was amplified from B4367Rs. Lanes: 1, no DNA
(negative control); 2, B4367Rs; 3, B4367rs.
a Due to small size, whole plants at weeks 1 and 4 were collected
and used. At later times, plants were separated into leaves and roots
for evaluation.
b Wild-type carrot acid-soluble invertase isozyme II mRNA
(2.0 bp) was expressed and detected in RT-PCR or northern analysis (also see Figure 2A).
c Two novel acid soluble invertase isozyme II transcripts (2.7 kb
and 3.2 kb) and no wild-type transcripts were observed in RT-PCR
(also see Figure 2A).
d No wild-type carrot acid-soluble invertase isozyme II mRNA was
expressed and detected in RT-PCR or northern analysis.
e No data collected.
Table 4. Sizes of RT-PCR and genomic DNA PCR products of
carrot acid-soluble invertase isozyme II with primer pairs INV-5
and INV-6 were used for RT-PCR and PCR amplification from
the leaf and root total RNA and genomic DNA, respectively, of
near-isogenic carrot lines B4367Rs and B4367rs.
B4367Rs
leaf
root
B4367rs
leaf
root
2.0 kb
3.8 kb
2.7 kb
3.2 kb
6.3 kb
RT-PCR band(s)
Genomic band(s)
2.0 kb
3.8 kb
no product
6.3 kb
ense primer) (Table 1), were synthesized based on
the B4367Rs acid-soluble invertase isozyme II gene
and insert (‘a746 + b498’; 1244 bp) sequences, and
used to amplify the B4367rs genomic DNA 5 - and
3 -end regions of the 2.5 kb insert respectively. The
PCR products synthesized from these two primer pairs
were cloned into a TOPO cloning vector for sequencing. The 2389 bp PCR product which amplified with
primers INV-22 and rsINVIImut-8 contained partial
carrot acid-soluble invertase isozyme II gene (green
bars in Figure 4), and its junction with the 5 end of
the 1244 bp cDNA insert (Figure 4). The sequence
of the PCR product amplified with rsINVIImut-5 and
INV-27 includes the 3 end of the 1244 bp cDNA
insert and part of the carrot acid-soluble invertase
isozyme II gene (Figure 4). The resulting sequences
were assembled with the known insert sequences. The
final full-length insert in carrot acid-soluble invertase
isozyme II gene is 2498 bp plus the 3 bp putative target
site duplication (TSD).
Portions of the 1244 bp cDNA insert elsewhere in the
B4367Rs genome
Full-length wild-type acid-soluble invertase isozyme
II of ca. 3.8 kb was obtained from B4367Rs genomic DNA by PCR amplification with primers INV-5
and INV-6 (Table 4). To determine if insert ‘a746’ or
‘b498’ occurs in this 3.8 kb product, insert-specific
primers rsINVIImut-1, rsINVIImut-2, rsINVIImut3 and rsINVIImut-4 (Table 1) were used with this
product as template to perform ‘nested PCR’. No PCR
product of either ‘a746’ or ‘b498’ was amplified from
the wild-type (B4367Rs) carrot acid-soluble invertase
isozyme II genomic DNA (data not presented). This
confirmed the absence of these inserts in acid-soluble
invertase isozyme II of B4367Rs. However, when total
genomic DNA of B4367Rs was used as a template,
159
inserts ‘a746’ and ‘b498’ were amplified to indicate
that the inserts exist somewhere else in the B4367Rs
carrot genome (data not presented).
Discussion
Invertase isozymes have been reported in carrots, including acid-soluble invertase isozymes I, II, cell-wall
invertase, neutral and alkaline invertases. Although
many of the isozymes have been isolated, characterized and sequenced, the functions of each of these
invertase isozymes and their interactions are not entirely clear. Through genetic study, sugar type (sucrose
or reducing sugars) stored in carrot storage roots was
found to be conditioned by the Rs locus. Although the
enzyme corresponding to Rs was unknown, an association between the Rs locus and vacuolar acid-soluble
invertase isozyme II function might be predicted since
this is the only invertase enzyme found to be well
expressed during carrot tap root development (weeks
10–16) measured by northern hybridization (Sturm,
1996), although both acid-soluble invertase isozyme
I and isozyme II locate in the vacuoles. Furthermore,
antisense DNA structure of carrot acid-soluble invertase isozyme I gene was introduced and expressed
in carrot and the tap roots of those transgenic plants
had only a 25% lower reducing sugar (glucose +
fructose) level than that of untransformed plants (Tang
et al., 1999). From comparative RT-PCR, we have now
found that little or no wild-type (ca. 2.0 kb) carrot
acid-soluble invertase isozyme II transcript was expressed from B4367rs while abundant the transcript
occurred in B4367Rs.
Somewhat parallel observations have been noted
for naturally occurring mutations affecting storage
carbohydrates in tomato fruits to ours in carrot root
roots. Studies in tomato revealed that low levels
of acid invertase (vacuolar invertase) are associated
with high levels of sucrose accumulation in Lycopersicon hirsutum (Miron and Schaffer, 1991), Lycopersicon peruvianum (Stommel, 1992) and Lycopersicon
chmielewskii (Yelle et al., 1988). Klann et al. (1993)
also reported that L. chmielewskii accumulates high
levels of sucrose, unlike the domesticated tomato
species, L. esculentum, which accumulates glucose
and fructose. They found the only significant enzymatic difference between the sucrose-accumulating
and hexose-accumulating fruit was the lack of acid invertase activity in sucrose-accumulating fruit (Klann
et al., 1993). All these results indicate that vacuolar
invertase plays a critical role in sugar accumulation
in storage organs. We speculate that similar mutations
may be found to affect storage carbohydrates in beet
roots and sugar cane.
The genomic DNA sequence of the rs carrot acidsoluble invertase isozyme II allele demonstrated a
2.5 kb insert integrated into the gene. The insertion
occurred in the first and largest intron near the 5
end of the B4367rs (rs/rs) carrot acid-soluble invertase
isozyme II gene.
Considering that the insertion occurs in the carrot near-isogenic line B4367rs acid-soluble invertase
isozyme II gene, but not in line B4367Rs acid-soluble
invertase isozyme II gene, and that the insert occurred
elsewhere in the B4367Rs genome, we speculate that
this DNA fragment was derived from a mobile genetic
element. Characterization of this insert is under study.
The origin, copy number and evidence for mobility
of the transposable element described above is being
further investigated.
The insertion of a 2.5 kb DNA fragment may
account for the failure of acid invertase isozyme II
transcription in carrot B4367rs roots and consequently
little to no acid-soluble invertase activity (Table 2).
Relatively early in development (week 14) there was
some pH 4.5 invertase activity in B4367rs roots, especially in the crown of the storage root. Since we
found no transcript for this isozyme in B4367rs roots,
we speculate that this may reflect isozyme I activity.
By week 16 pH 4.5 invertase activity diminished significantly and remained very small throughout the rest
of root development. The enzyme activity of pH 7.5
invertase followed similar developmental and genetic
trends to those for pH 4.5 invertase suggesting that
invertase activity at both pH 4.5 and pH 7.5 reflect
the collective activity of both isozymes. The significant reduction in invertase activity in turn may account
for the accumulation of sucrose (rather than reducing
sugars) in the development of B4367rs roots. Thus,
the carrot acid-soluble invertase isozyme II gene is
the best candidate for Rs locus. Our parallel genetic
studies indicate complete linkage between the rs phenotype and presence of the insert in the enzyme in
populations segregating for Rs/rs (Vivek and Simon,
1999) which strengthens this proposition. The disruption of acid-soluble invertase isozyme II gene function
by the 2.5 kb insert is surprising, since it occurs in an
intron.
Differences in tissue-specific RNA processing
were also observed in the carrot acid-soluble invertase
isozyme II gene of B4367Rs and B4367rs. Two novel
160
Figure 4. Genomic DNA and cDNA of wild-type and mutant carrot acid-soluble invertase isozyme II genes in near-isogenic lines B4367Rs
and B4367rs (diagrams are not drawn to scale). A. Diagram of the cDNA of 1953 bp from the wild-type acid-soluble invertase isozyme II
gene from carrot line B4367Rs amplified by RT-PCR. Primers INV-5 and INV-6 contain the start and stop codons of the gene respectively. The
position of primers INV-5, INV-6, and INV-18 are indicated. B. Diagram of genomic DNA of 3821 bp from wild-type acid-soluble invertase
isozyme II gene from carrot line B4367Rs. Symbols ‘i’ and ‘e’ stand for intron and exon respectively. The first intron is in bright green and the
insertion site for the 2.5 kb insert is labeled with a red oval. Positions of primers INV-5, INV-6, INV-18, and INV-22 are noted. C. Diagram
of genomic DNA of the mutated acid-soluble invertase isozyme II gene from carrot line B4367rs with the first intron highlighted. The green
(including bright green) bars represent the wild-type carrot acid-soluble invertase isozyme II gene. Bright green and yellow bars represent the
sequences spliced out of the first intron during mRNA processing. Yellow, orange and red bars comprise the 2.5 kb insert. Positions of primers
INV-5, INV-6, INV-18, INV-22, INV-27, rsINVIImut-1 through -5, and rsINVIImut-8 are indicated. D–F. Diagrams of cDNA fragments ‘a’,
‘b’, and ‘d’ (from Figure 3) amplified by RT-PCR with primers INV-5 and INV-18.
RT-PCR products of 2.7 kb and 3.2 kb, were observed
in the leaves, but not in the roots, of B4367rs (rs/rs)
plants.
Tissue-specific RNA processing has recently reported in transposable element-inserted maize waxy
gene (Marillonnet and Wessler, 1997) where a
retrotransposon-inserted allele, wxG, has a tissuespecific phenotype with 30-fold more Wx enzymatic
activity in pollen than in the endosperm. Quantification of wxG-encoded transcripts in pollen and the
endosperm demonstrated that this difference can be
accounted for by tissue-specific differences in RNA
processing. Specifically, there is about 30-fold more
correctly spliced RNA in pollen than in the endosperm
(Marillonnet and Wessler, 1997). However, the effects
of the 2.5 kb insert-containing gene in carrot differ
from that of transposable element-inserted wxG allele
in maize. While the inefficient splicing of the 2.5 kb
insert out of the coding region of carrot acid-soluble
invertase is similar to the maize wxG allele, the novel
transcripts for the rs allele were either present or absent in different tissues and, unlike the transposable
element-inserted wxG allele in maize, very little wildtype (Rs) transcript (the ‘a’ band) was detected in any
tissues analyzed in rs plants. Furthermore, the novel
transcripts for the rs allele were observed in leaves,
not roots of rs plants.
161
Based on observations of (1) the comparable DNA
sequences between the genomic DNA and its cDNA
for the rs allele, (2) the fact that only one PCR product
(ca. 6 kb from the start codon to the stop codon)
was observed from B4367rs genomic DNA amplification with primers INV-5 and INV-6, and (3) the fact
that only a single copy of rs acid-soluble invertase
isozyme II gene shown on the Southern blot (data not
presented), we conclude that these two novel transcripts (ca. 3.2 kb and ca. 2.7 kb RT-PCR products
amplified with primers INV-5 and INV-6) from leaves
of rs plants were derived through mRNA processing.
The ‘b498’ fragment is recognized as an intron and
spliced out from the 3.2 kb fragment, while the ‘a746’
fragment remaining in the 2.7 kb fragment contains
almost whole ORF identified in the genomic sequence
of the insert.
Co-dominant markers derived from the 2.5-kb insert to distinguish wild-type and mutant (with the
2.5 kb insert) carrot acid-soluble invertase isozyme II
gene were established and were used to determine
the genotypes of this gene in carrot (Yau and Simon,
submitted). Tests from several mapping populations
revealed that carrot individuals with two copies of
wild type or one copy of wild type plus one copy
of mutated (with insert) carrot acid-soluble invertase
alleles have the Rs/- phenotype (i.e. they accumulate
reducing sugar in storage root) while carrot roots with
two copies of mutated carrot acid-soluble invertase alleles (rs/rs) accumulate predominantly sucrose (Yau
and Simon, submitted). These results also suggest that
carrot acid-soluble invertase isozyme II gene is the
likely candidate for the Rs locus.
Acknowledgements
Support of the vegetable industry is gratefully acknowledged. The discussions and capable assistance
of Dr Douglas Senalik and Dr Dariusz Grzebelus in
initiating and pursuing this research is greatly appreciated.
References
Alabran, D.M. and Mabrouk, A.F. 1973. Carrot flavor, sugars,
and free nitrogenous compounds in fresh carrots. J. Agric. Food
Chem. 21: 205–208.
Casa, A.M., Brouwer, C., Nagel, A., Wang, L., Zhang, Q., Kresovich, S. and Wessler, S.R. 2000. The MITE family Heartbreaker
(Hbr): molecular markers in maize. Proc. Natl. Acad. Sci. USA
97: 10083–10089.
Cheng, W.H., Taliercio, E.W. and Chourey, P.S. 1996. The Miniature 1 seed locus of maize encodes a cell-wall invertase required
for normal development of endosperm and maternal cells in the
pedicel. Plant Cell 8: 971–983.
Copeland, L. 1990. Enzymes of sucrose metabolism. In: P.J. Lea
(Ed.) Methods in Plant Biochemistry: Enzymes of Primary
Metabolism Vol. 3, Academic Press, New York, pp. 73–84.
Freeman, R.E. and Simon, P.W. 1983. Evidence for simple genetic
control of sugar type in carrot (Daucus carota L.). J. Am. Soc.
Hort. Sci. 108: 50–54.
Heineke, D., Wildenberger, K., Sonnewald, U., Willmitzer, L.
and Heldt, H.W. 1994. Accumulation of hexose in leaf vacuole:
studies with transgenic tobacco plants expressing yeast-derived
invertase in the cytosol, vacuole or apoplasm. Planta 194: 29–33.
Klann, E.M., Chetelat, R.T. and Bennett, A.B. 1993. Expression
of acid invertase gene controls sugar composition in tomato
(Lycopersicon) fruit. Plant Physiol. 103: 863–870.
Laurière, C., Laurière, M., Sturm, A., Faye, L. and Chrispeels,
M.J. 1988. Characterization of β-fructosidase, an extracellular
glycoprotein of carrot cells. Biochimie 70: 1483–1491.
Lee, H.S. and Sturm, A. 1996. Purification and characterization of
neutral and alkaline invertase from carrot. Plant Physiol. 112:
1513–1522.
Marillonnet, S. and Wessler, S.R. 1997. Retrotransposon insertion into the maize waxy gene results in tissue-specific RNA
processing. Plant Cell 9: 967–978.
Miron, D. and Schaffer, A.A. 1991. Sucrose phosphate synthase,
sucrose synthase, and invertase activities in developing fruit
of Lycopersicon esculentum Mill. and sucrose accumulating
Lycopersicon hirsutum Humb. and Bonpl. Plant Physiol. 95:
623–627.
Muller-Rober, B., Sonnewald, W. and Willmitzer, L. 1992. Inhibition of the ADP-glucose pyrophosphorylase in transgenic
potatoes leads to sugar-storing tubers and influences tuber formation and expression of tuber storage protein genes. EMBO J. 11:
1229–1238.
Murray, M. and Thompson, W. 1980. Rapid isolation of highmolecular weight plant DNA. Nucl. Acids Res. 8: 4321–4325.
Simon, P.W. and Freeman, R.E. 1985. A rapid method for screening
reducing sugar in carrot roots. HortScience 20: 133–134.
Simpson, C.G., Sinibaldi, R. and Brown, J.S. 1992. Rapid analysis
of plant gene expression by a novel reverse transcriptase-PCR
method. Plant J. 2: 835–836.
Sonnewald, U., Hajirezaei, M.R., Kossman, J., Heyer, A., Tretheway, R.N. and Willmitzer, L. 1997. Increased potato tuber size
resulting from apoplastic expression of a yeast invertase. Nature
Biotech. 15: 794–797.
Sonnewald, U., Stitt, M., Schaewen, A.V., Brauer, M. and Willmitzer, L. 1991. Transgenic tobacco plants expressing yeastderived invertase in either the cytosol, vacuole or apoplast: a
powerful tool for studying sucrose metabolism and sink/source
interaction. Plant J. 1: 95–106.
Stommel, J.R. 1992. Enzymatic components of sucrose accumulation in the wild tomato species Lycopersicon peruvianum. Plant
Physiol. 99: 324–328.
Stommel, J.R. and Simon, P.W. 1990. Multiple forms of invertase
from Daucus carota cell cultures. Phytochemistry 29: 2087–
2089.
Sturm, A. 1996. Molecular characterization and functional analysis
of sucrose-cleaving enzymes in carrot (Daucus carota L.). J. Exp.
Bot. 47: 1187–1192.
Sturm, A. and Chrispeels, M.J. 1990. cDNA cloning of carrot
extracellular β-fructosidase and its expression in response to
wounding and bacterial infection. Plant Cell 2: 1107–1119.
162
Sturm, A., Šebková, V., Lorenz, K., Hardegger, M., Lienhard, S.
and Unger, C. 1995. Development- and organ-specific expression
of the genes for sucrose synthase and three isozymes of acid βfructofuranosidase in carrot. Planta 195: 601–610.
Tang, G.Q., Lüscher, M. and Sturm, A. 1999. Antisense repression
of vacuolar and cell wall invertase in transgenic carrot alters early
plant development and sucrose partitioning. Plant Cell 11: 177–
189.
Unger, C., Hardegger, M., Liehard, S. and Sturm, A. 1994.
cDNA cloning of carrot (Daucus carota) soluble acid βfurctofuranosidases and comparison with the cell wall isoenzyme. Plant Physiol. 104: 1351–1357.
Vivek, B.S. and Simon, P.W. 1999. Linkage relationships among
molecular markers and storage root traits of carrot (Daucus
carota L. ssp. sativus). Theor. Appl. Genet. 99: 58–64.
von Schaewen, A., Stitt, M., Schmidt, R., Sonnewald, U. and Willmitzer, L. 1990. Expression of a yeast-derived invertase in the
cell wall of tobacco and Arabidopsis plants leads to accumulation
of carbohydrate and inhibition of photosynthesis and strongly
influences growth and phenotype of transgenic tobacco plants.
EMBO J. 9: 3033–3044.
Wang, F., Amparo, S., Brenner, M.L. and Smith, A. 1993. Sucrose
synthase, starch accumulation, and tomato fruit sink strength.
Plant Physiol. 101: 321–327.
Yelle, S., Chetelat, R.T., Dorais, M., DeVerna, J.W. and Bennett,
A.B. 1991. Sink metabolism in tomato fruit. IV. Genetic and
biochemical analysis of sucrose accumulation. Plant Physiol. 95:
1026–1035.
Yelle, S., Hewitt, J.D., Robinson, N.L., Damon, S. and Bennett,
A.B. 1988. Sink metabolism in tomato fruit. III. Analysis of
carbohydrate assimilation in a wild species. Plant Physiol. 87:
737–740.
Zhang, L., Cohn, N.S. and Mitchell, J.P. 1996. Induction of a pea
cell-wall invertase gene by wounding and its localized expression
in phloem. Plant Physiol. 112: 1111–1117.