全 文 ：Received 29 Mar. 2004 Accepted 5 Jul. 2004
Supported by the National Natural Science Foundation of China (30170648, 30370998).
* Author for correspondence. Tel: +86 (0)571 86971009; E-mail: .
http://www.chineseplantscience.com
Acta Botanica Sinica
植 物 学 报 2004, 46 (12): 1408-1415
Isolation and Expression Analysis of Fructokinase Genes from Citrus
QIN Qiao-Ping1, 2, ZHANG Shang-Long1*, CHEN Jun-Wei2, XIE Ming2
JIN Yong-Feng3, CHEN Kun-Song1, Syed ASGHAR1
(1. Department of Horticulture, Zhejiang University, Hangzhou 310029, China;
2. Institute of Horticulture, Zhejiang Academy of Agricultural Sciences, Hangzhou 310021, China;
3. Institute of Biochemistry, Zhejiang University, Hangzhou 310029, China)
Abstract: Fructokinase (FRK) is of primary importance in phosphorylation of fructose in plants. Two
genomic DNA fragments encoding putative fructokinase gene were isolated from Citrus unshiu Marc. using
PCR, which were named Cufrk1 and Cufrk2. A cDNA sequence same to the exons’ sequences of Cufrk1
was amplified by RT-PCR and the full-length cDNA sequences of this gene was isolated using rapid amplifi-
cation of cDNA ends (RACE) from mature fruit, named CuFRK1 (GenBank number: AY561840). Sequenc-
ing analysis showed that the amino acid sequences were 68% identical between Cufrk1 and Cufrk2. The
full-length cDNA of CuFRK1 was 1 459 bp in length and contained a complete ORF from 168 to 1 220 bp,
encoding 350 amino acids with a calculated molecular weight of 37.5 kD and a predicted pI of 5.03. The
deduced amino acids of CuFRK1 possessed two sugar-binding domains, three ATP-binding domains and
were 62% to 78% identical to the previously characterized fructokinase genes of other plants. Northern
analysis showed that transcripts of CuFRK1 (Cufrk1) and Cufrk2 were detected at a high level in leaves,
fruit at early developmental stages, but undetectable in peels and stems. Their expression patterns were
distinct in petals and mature fruit. Enzymatic analysis showed that the activity of citrus fructokinase was
decreasing with the fruit development, coincided with the accumulation of fructose in fruit. There was a
significantly negative correlation between the decreasing fructokinase activity and the increasing fructose
content during fruit development.
Key words: citrus; fructokinase; gene cloning; expression analysis; fructose accumulation
Sugar content is an important factor in deterring citrus
fruit quality and is directly related to its taste and quality
(Izumi et al., 1990; Chen et al., 2001; Zhao et al., 2001;
2002). Most sugars in fruit are derived from leaf
photosynthates. In most plants including citrus, sucrose
represents the major transport form of photoassimilates
and is translocated through phloem from source to differ-
ent sinks (Kriedmann, 1969). In sink tissues, sucrose may
either be stored directly or cleaved into hexose or hexose
phosphate by sucrose synthase or invertase. Fructose
derived from sucrose cleavage reactions must be phos-
phorylated by hexokinases (HXK, EC 2.7.1.1) or
fructokinases (FRK, EC 2.7.1.4) for further metabolism.
Because affinities of fructokinase for fructose are much
higher than those of hexokinase (Renz and Stitt, 1993),
fructokinase are likely to be of primary importance in fruc-
tose metabolism. Therefore, studies on the fructose
metabolism, fructokinase gene expression and its regula-
tion would contribute to elucidating the mechanism of sugar
accumulation and supply basis for controling the sugar
composition in citrus fruit.
Fructokinase has been purified and characterized from
several plants, and most studies suggest the presence of
at least two fructokinase isoforms (Chaubron et al.,1995;
Martinez-Barajas and Randall, 1996). Till now, only a few
fructokinase genes have been partially characterized from
potato (Smith et al.,1993; Taylor et al.,1995; Dai et al., 1997),
tomato (Kanayama et al., 1997; 1998; German et al., 2002;
2004), rice (Jiang et al., 2003), and maize (Zhang et al.,
2003). These fructokinase genes have different expression
patterns. However, fructokinase genes from woody plants
and the relationship of fructokinase to sugar accumula-
tion have not been reported. It was important to isolate
fructokinase genes and research their expression pattern
in order to elucidate the sugar accumulation mechanism
regulated by fructokinase. In this study, we isolated two
fructokinase genes and investated their expression
patterns, enzymatic activity, and fructose accumulation in
different stage of satsuma mandarin fruit.
1 Materials and Methods
1.1 Plant materials
Ten-year-old satsuma mandarin (Citrus unshiu Marc.)
trees grown in the orchard of Department of Horticulture,
QIN Qiao-Ping et al.: Isolation and Expression Analysis of Fructokinase Genes from Citrus 1409
Zhejiang University, Hangzhou, China were selected as
experimental material. Leaves, petals, stems, peels from
mature fruit and fruits from early developmental stage
(starting from May) to maturity stage (late in October) were
sampled for sugar content analysis, fructokinase activity
determination and Northern blotting analysis. All sampled
materials were frozen immediately in liquid nitrogen and
stored at -70 ℃.
1.2 DNA extraction and gene cloning
Genomic DNA was extracted from young leaves of cit-
rus according to Chen et al. (2000). Conserved domains
were identified by alignment of the amino acid sequences
of several fructokinase genes. The conserved domains were
used to design the forward degenerate primer UP1: 5-AA
(A/G)CTCGGTGACGATGAGTT(C/T) GG-3 and the re-
verse primer DP1: 5-TCACCAGC(T/A)CCAGTGGT(G/A)
TCAAC-3. The 50 mL PCR mixtures were set up as follows:
50 ng genomic DNA, 10 mmol/L UP1 and DP1, 1× PCR
buffer, 3 mmol/L MgCl2, 0.25 mmol/L dNTPs, and 2 U Taq
DNA polymerase (TaKaRa, Japan). Amplifications were for
30 cycles, each consisting of 94 ℃ for 1 min, 40 s at 52 ℃
and 1 min at 72 ℃. Amplified PCR products were cloned
into the pGEM-T Easy vector (Promega). Two positive
clones obtained were sequenced in Gene Company,
Shanghai, China. Analysis of the sequences was carried
out using CLUSTALX, SEQAID and DNASTAR programs.
1.3 RT-PCR
Total RNA was isolated from mature fruit of satsuma
mandarin according to a modification of single-step method
(Chomczynski and Sacchi, 1987). One mg of total RNA was
used to reverse-transcribe the first strand cDNA using
RevertAid First Strand cDNA synthesis kit (Fermentas).
Fifty ng of the cDNA was added to the PCR mixtures to
amplify the cDNA fragment according to the PCR steps
described above.
1.4 5- and 3- RACE (rapid amplification of cDNA ends)
Using 5-Full Race Core Set (TaKaRa, Japan), 5-RACE
was carried out. Four specific primers were designed ac-
cording to the sequenced fragment obtained from RT-PCR,
S1: 5- GGAGAGGATGCTACCAGACTCTTTGGCC-3; S2:
5 -CAGT CCT TGCAGT GGAAT CA-3 ; A1 : 5 -
ATCAGAAGAGGCTGCTCGCGAGGGC-3; A2: 5-
TTAGAGAAGCTGTTTCATCCT-3. All the four primers
were used to amplify the cDNA 5-terminal regions. Using
A1 and A2 as primens, 3-RACE was also carried out by
the method of 3-Full Race Core Set (TaKaRa, Japan).
The amplified PCR products were cloned to T-vector
and sequenced. The full-length cDNA of fructokinase gene
was cloned using gene-specific primers designed as the
sequences of the two ends.
1.5 Northern blotting analysis
Total RNA isolated was separated by electrophoresis
on a 1.0% (W/V) formaldehyde agarose gel and transferred
to a Hybond-N+ membrane (Amersham, UK). The two spe-
cific DNA probes were obtained from digestion of recom-
bined plasmids by EcoRⅠ and HindⅢ. The probes were
labeled with [a-32P]dCTP by the random primer method
(TaKaRa, Japan). The resulting blots were hybridized at 65
℃ in hybridization mixture of 0.25 mol/L sodium phosphate,
1 mmol/L EDTA, 1% BSA (W/V) and 7% (W/V) SDS. Mem-
branes were washed in 2× SSC (1× SSC is 0.15 mol/L
NaCl and 0.015 mol/L sodium citrate), 0.1% SDS for 15 min
and 1×SSC, 0.1% SDS for 10 min at 55 ℃. The x-ray films
were exposed at -70 ℃ for 5 d.
1.6 Determination of sugar content
The quantity and types of sugars were determined as
described by Chen et al. (2001). Soluble sugars were ex-
tracted by grinding fresh tissues in five volumes (W/V) of
methanol: chloroform: water (E:C:W) as 12:5:3 (V/V). Ex-
tracts were centrifuged at 5 000g for 5 min. The extraction
was performed three times. Water and chloroform were then
added to bring the final E:C:W ratio to 10:6:5. Chloroform
layer was removed. The remaining aqueous-alcohol phase
was adjusted to pH 7.0 using 0.1 mol/L NaOH, then dried
in a vacuum and re-dissolved with distilled water. The sug-
ars in water solution were analyzed by HPLC (Beckman,
USA) equipped with NUCLEOSIL 5 mn NH2 column and RI
detector (Shodex Tokyo, Japan). Column temperature was
90 ℃ and 80% acetonitrile was used as an elute at a flow
rate of 1 mL/min. Fructose, glucose and sucrose were iden-
tified and quantified by comparing the retention and inte-
grated peak areas of external standards.
1.7 Extraction and assay of fructokinase
Fructokinases were extracted as described by Schaffer
and Petreikov (1997). All operations were carried out at 4
℃. Approximately 1 g fresh weight of tissue was homog-
enized with mortar and pestle in 3 mL ice cold extraction
buffer containing 50 mmol/L Hepes-NaOH (pH 7.5), 1 mmol/L
Na2EDTA, 1 mmol/L MgCl2, 2.5 mmol/L DTT, 10 mmol/L
KCl, 1% (W/V) insoluble PVP, and 3 mmol/L
diethyldithiocarbamic acid (DIECA). The homogenate was
centrifuged at 18 000g for 20 min. The supernatant was
precipitated with 80% ammonium sulfate and centrifuged
for 10 min at 18 000g. The precipitate was resuspended in 1
mL of extraction buffer and desalted on a Sephadex G-25
column (Roche) with washing buffer containing 50 mmol/
L Hepes-NaOH (pH 7.5), 1 mmol/L Na2EDTA, and 1 mmol/
L DTT. The desalted extract was used as a crude enzyme.
Acta Botanica Sinica 植物学报 Vol.46 No.12 20041410
Fructokinase activity was measured by an enzyme-
linked assay. Assays contained, in a total volume of 400
mL, 30 mmol/L Hepes-NaOH (pH 7.5), 0.6 mmol/L Na2EDTA,
1 mmol/L MgCl2, 9 mmol/L KCl, 1 mmol/L NAD, 1 mmol/L
ATP, 1 unit of NAD-dependent Glc-6-PDH (Sigma), and 1
unit of PGI (Sigma). The reaction was initiated with 2 mmol/L
fructose. Reactions were carried out at 25 ℃ and A340 was
monitored continuously.
2 Results
2.1 Cloning of two fructokinase genomic fragments
Two gene fragments were amplified using degenerate
primers (UP1 and DP1) from satsuma mandarin genomic
DNA. They were named Cufrk1 and Cufrk2 respectively
after sequencing analysis. Cufrk1 (GenBank: AY118083)
with 1 213 bp, encoded a deduced protein of 205 amino
acids, having four exons and three introns; CufrkK2
(GenBank: AF521003) with 793 bp, encoded 204 amino acids,
having two exons and one intron. All the introns had the
boundary sequences 5-GT/AG-3 at the splicing sites. There
was 68% identical between their polypeptides encoded by
Cufrk1 and Cufrk2. Both deduced amino acid sequences
possessed two sugar-binding domains, which are highly
conserved in previously characterized fructokinase
genes (Pego and Smeekens, 2000). Their deduced amino
acid sequences were 64% to 78% identical to the
fructokinase sequences from tomato, rice, potato, maize
or sugar beet.
2.2 Isolation of full-length fructokinase cDNA from
mature fruit
A cDNA fragment of fructokinase gene was cloned from
mature citrus fruit using primers of UP1 and DP1 by RT-
PCR. Because the sequences of this fragment were con-
sistent with the exons’ sequences of Cufrk1, therefore it
was named CuFRK1. Four specific primers (A1, A2, S1
and S2) were designed according to the sequenced cDNA
fragment. The 5- and 3-regions of the gene were amplified
using these primers and the full-length cDNA sequences
of CuFRK1 (GenBank: AY561840) was obtained. CuFRK1
was 1 459 bp in length and included an ORF that encoded
a putative protein of 350 amino acids with a calculated
molecular weight of 37.5 kD and a predicted pI of 5.03. The
features of CuFRK1 protein resembled previously de-
scribed tomato fructokinases (German et al., 2004). The
ATG triplet beginning at nucleotide 168 was expected to
be the likely site of translation initiation because six nucle-
otides surrounding that ATG were identical to the consen-
sus sequence (AATGGC) for plant initiation codons
(XATGXC) (Lutche et al., 1987). The putative CuFRK1
protein possessed two sugar-binding domains and three
ATP-binding domains (Fig.1), which are conserved in all
of the characterized plant fructokinase genes (Pego and
Smeekens, 2000). The deduced amino acids of CuFRK1
were 62% to 78% identical to the previously characterized
fructokinase genes of other plants.
2.3 Phylogenetic analysis of deduced amino acid se-
quences of the full-length fructokinase cDNA
Phylogenetic analysis based on fructokinase amino acid
sequences availably analyzed with CLUSTALX software
suggested that the CuFRK1 sequences were most close
to tomato fructokinase gene of LeFRK1 (75% identity) and
LeFRK4 (72% identity), the third closely related group was
LeFRK3, which was distinct from tomato LeFRK2, potato
StFRK, sugar beet BvFRK, maize ZmFRK1 and rice
OsFRKI, and distinct from another group of rice OsFRKII
and maize ZmFRK2 (Fig.2). Two conserved functional do-
mains have been found in all isolated fructokinase genes
(Pego and Smeekens, 2000), and the same domains also
found in CuFRK1 (Fig.1). The proposed ATP-binding mo-
tif (Domain A in Fig.1), which contained the GD motif es-
sential for activity, was conserved in the CuFRK1 and
other fructokinase gene sequences. Domain B, which was
also found in the CuFRK1 sequences, was unique to
fructokinases and has been thought to represent a fruc-
tose substrate recognition site. These sequence compari-
son strongly suggested that the CuFRK1 cDNA identi-
fied here encode fructokinases.
2.4 Expression analysis of citrus fructokinase genes
The spatial-temporal expression patterns of CuFRK1
(Cufrk1) and Cufrk2 were analyzed by Northern blot us-
ing gene-specific probes (fragments were 218 bp and 442
bp in length respectively; the 218 bp was the common
sequences of CuFRK1 and Cufrk1).
The analysis showed that the transcript levels of
CuFRK1 (Cufrk1) were high in young leaves, fruit juice
sacs at early developmental stages (10, 20, and 35 d after
anthesis (DAA)), and mature fruit (175 DAA). Petals could
be detected the transcript level of CuFRK1 (Cufrk1), but
the expression level of that was very low. The transcripts
of CuFRK1 (Cufrk1) were undetectable both in stems and
peel of mature fruit (Fig.3).
The expression pattern of Cufrk2 was similar to
CuFRK1 (Cufrk1) in some tissues. Transcripts of Cufrk2
were detected at a high level in young leaves, petals and
fruit juice sacs at early developmental stages (10, 20, and
35 DAA). However, no transcript was detectable in juice
sacs of fruit from 90 to 175 DAA, stems and peel of mature
fruit (Fig.3).
QIN Qiao-Ping et al.: Isolation and Expression Analysis of Fructokinase Genes from Citrus 1411
Fig.1. Nucleotide and deduced amino acid sequences showed beneath the nucleotide sequence of the sense strand of the citrus
fructokinas cDNA CuFRK1 isolated from mature fruit. Three signature patterns for pfkB family of carbohydrate kinase (A1, A2 and
A3) and a region specific to fructokinases (B) is shown. The A1 motif is involved in ATP binding and the B motif contains two sugar-
binding domains (Pego and Smeekens, 2000). The small letters belong to non-coding sequences and the capital letters belong to coding
sequences.
Acta Botanica Sinica 植物学报 Vol.46 No.12 20041412
2.5 Sugar content
In order to elucidate the relationship between
fructokinase activity and sugar accumulation, the sugar
contents were determined in the fruit juice sacs at 20, 35,
66, 96, 135, 175 DAA using HPLC (Fig.4). The results
showed that sucrose, glucose and fructose were accumu-
lated during fruit development. The sucrose content was
found 31.0 mg/g FW, and it was about 59.3% of the total
sugars in the ripe fruit; whereas fructose content was 10.8
mg/g FW and accounted for 20.8% of the total sugars in
the mature fruit; glucose content was 10.4 mg/g FW and
accounted for 19.9% of the total sugars in the mature fruit.
The sucrose content was clearly higher than that of hex-
oses (fructose and glucose) in the mature fruit.
2.6 Enzymatic activity of fructokinase during fruit de-
velopment
The activity of fructokinase in juice sacs of citrus fruit
at different developmental stages was measured (Fig.5).
The maximum fructokinase activity was found in fruit of
early developmental stages (20 DAA) (about 1 622.1
nmol.min-1.g-1 FW) and then decreased sharply, at 66
DAA, the activity was 602.3 nmol.min-1.g-1 FW, which
possessed only 37.1% of the maximum activity. After 66
DAA, the fructokinase activity gradually decreased and
reached the lowest level (226.0 nmol.min-1.g-1 FW) at
175 DAA. There was a turning point of fructokinase activ-
ity at 66 DAA, before which sharp decrease happened and
after which gradual decrease was characterized. It is indi-
cated in Fig.5 that fructose content in the fruit was in-
creased rapidly after 66 DAA and accumulated continu-
ally up to mature. There was a significantly negative
Fig.2. Phylogenetic tree of the deduced amino acid sequences
of eleven plant fructokinases with their respective accession
numbers using CLUSTALX. Bv, Beta vulgaris; Le, Lycopersicon
esculentum; Os, Oryza sativa; St, Solanum tuberosum; Zm, Zea
mays; Cu, Citrus unshiu.
Fig.3. Northern blotting analysis of total RNA from various
citrus organs and fruit juice sacs at different developmental stages
indicated by days after anthesis (DAA) using gene-specific
probes of CuFRK1 (Cufrk1) and Cufrk2. Total RNA was frac-
tionated by electrophoresis on 1.0% agarose gel, including 2.2
mol/L formamide, and was transferred to Hybond-N+ membrane.
The RNA gel blot was hybridized with a-32P-labeled CuFRK1
(Cufrk1) and Cufrk2 gene-specific probes. In the lower panel,
ribosomal RNA stained with ethidium bromide was shown to
indicate the amount of RNA used in each sample.
Fig.4. Accumulation of sucrose, glucose and fructose
(mg/g FW) in fruit juice sacs at different developmental stages of
satsuma mandarin. Fruit at different developmental stages were
indicated by days after anthesis (DAA). Each point represents
the mean of the three values. Vertical bars indicate SD.
Fig.5. Activity of fructokinase (nmol.min-1.g-1 FW) and fruc-
tose content in juice sacs during the fruit development of sat-
suma mandarin. Fruit at different developmental stages were
indicated by days after anthesis (DAA). Each point represents
the mean of the three values. Vertical bars indicate SD.
QIN Qiao-Ping et al.: Isolation and Expression Analysis of Fructokinase Genes from Citrus 1413
correlation between the decreasing fructokinase activity
and the increasing fructose content during fruit develop-
ment and the correlation coefficient (r) was -0.828 33 (Fig.
6), which suggested that fructokinase activity had a close
relationship with fructose accumulation. The high level of
fructokinase activity might be a limit factor for fructose
accumulation in citrus fruit. These results suggested that
the enzymatic activity of FRK was important in fructose
accumulation in citrus fruit.
Northern blotting analysis showed that the expression
patterns of the two genes differed to a certain extent. Tran-
scripts of Cufrk2 could be detected only in young tissues
such as young leaves or young fruit, but those of CuFRK1
(Cufrk1) could also be detected in mature fruit (Fig.3).
These results suggest that CuFRK1 (Cufrk1) and Cufrk2
could encode distinct fructokinase isoforms and play the
function of fructose metabolism in different developmen-
tal stage or tissues. No transcripts of CuFRK1 (Cufrk1)
and Cufrk2 had been detected in fruit from 66 to 136 DAA,
however, the fructokinase activity was detectable during
these stages (Fig.5). These facts suggest that other un-
known isoenzymes of fructokinase may exist in citrus. Till
now four fructokinase genes have been isolated from to-
mato cDNA library and the chromosome locations expres-
sion patterns and enzymatic traits of four fructokinase genes
are different. German et al. (2004) suggest that each
fructokinase may play different physiological roles in
tomato. We isolated CuFRK1 (Cufrk1) and Cufrk2 from
citrus and their expression in various organs and fruit de-
velopmental stages were different to a certain extent. Ac-
cording to the case of tomato we considered that several
fructokinase genes may exist in citrus and the role of each
fructokinase might be different based on our results.
The enzyme activity of fructokinase analysis showed
developmental losses of activity in satsuma mandarin. We
had observed an increase in the fructose accumulation
with the declining fructokinase activities (Figs.4, 5). There
was a significantly negative correlation between the
fructokinase activity and the fructose content in fruit de-
velopment (Fig.6). The turning points of fructokinase ac-
tivity and fructose content during fruit development were
at 66 DAA (Fig.5), and there were no transcripts of CuFRK1
(Cufrk1) or Cufrk2 after 66 DAA except 175 DAA by North-
ern blots (Fig.3), which might be related to the sharply
decreasing of fructokinase activity. The high enzymatic
activity was consistent with the high mRNA level of
fructokinase and very low fructose content in young cit-
rus fruit. The fact that the fructokinase activity and mRNA
level decreased concomitant the increasing of fructose
content suggested that the expression of fructokinase might
play an important role on the accumulation of fructose in
citrus fruit.
References:
Chaubron F, Harris N, Ross H A, Daives H V. 1995. Partial
purification and characterization of fructokinase from devel-
oping taproots of sugar beet (Beta vulgaris). Plant Sci, 110:
3 Discussion
Fructokinase genes have been isolated from several
plants, with one to four genes per species (Qin et al., 2003).
However, there is still no report on fructokinase gene from
woody plants. In this study we first isolated one full-length
cDNA of fructokinase gene from woody plant of satsuma
mandarin. The deduced amino acids of CuFRK1 cDNA
showed high homology with fructokinase genes isolated
from other plants by BLAST analysis in NCBI (data not
shown). The full-length cDNA of CuFRK1 had three ATP-
binding domains (contained the GD motif essential for
activity) and two sugar-binding domains, which are unique
to fructokinase and conserved in other fructokinase genes
(Pego and Smeekens, 2000). The sugar-binding domains
are likely to be the fructose substrate recognition site.
Therefore, CuFRK1 would encode the proteins of
fructokinase (Fig.1). Alignment analysis showed that amino
acid sequences of CuFRK1 were most close to tomato
fructokinase gene LeFRK1 (75% identity) and LeFRK4
(72% identity) (Fig.2). The property of protein encoded by
CuFRK1 needed to be characterized.
In this study, two genomic DNA fragments were iso-
lated and the predicted amino acid sequences of Cufrk1
and Cufrk2 possessed two sugar-binding domains.
Fig.6. The relationships between fructokinase activity and
fructose content during satsuma mandarin fruit development.
Acta Botanica Sinica 植物学报 Vol.46 No.12 20041414
181-186.
Chen J-W , Zhang S-L , Zhang L-C , Xu C-J , Chen K-S. 2001.
Characteristics of photosynthate translocation and portioning
and sugar accumulation in developing satsuma mandarin
(Citrus unshiu Marc.) fruit. Acta Phytophysiol Sin , 27: 186-
192. (in Chinese with English abstract)
Chomczynski P, Sacchi N. 1987. Single-step method of RNA iso-
lation by acid guanidinium-thicyanate-phenol-chloroform
extraction. Anal Biochem, 162: 156-159.
Dai N, Schaffer A A, Petreikov M, Granot D. 1997. Potato
(Solanumtuberosum L.) fructokinase expressed in yeast ex-
hibits inhibition by fructose of both in vitro enzyme activity
and rate of cell proliferation. Plant Sci, 128: 191-197.
German M A, Dai N, Chmelnitsky I, Sobolev I, Salts Y, Barg R,
Schaffer A A, Granot D. 2002. LeFRK4, a novel tomato
(Lycopersicon esculentum Mill.) fructokinase specifically
expressed in stamens. Plant Sci, 163: 607-613.
German M A, Asher I, Petreikov M, Dai N, Schaffer A A, Granot
D. 2004. Cloning, expression and characterization of LeFRK3,
the fourth tomato (Lycopersicon esculentum Mill.) gene en-
coding fructokinase. Plant Sci, 166: 285-291.
Izumi H, Ito T, Yoshida Y. 1990. Sugar and ascorbic acid con-
tents of Satsuma mandarin fruits harvested from exterior and
interior canopy of trees during fruit development. J JPN
Hort Sci, 58: 877-883.
Jiang H, Dian W, Liu F, Wu P. 2003. Isolation and characteriza-
tion of two fructokinase cDNA clones from rice.
Phytochemistry, 62: 47-52.
Kanayama Y, Dai N, Granot D, Petreikov M, Schaffer A,
Bbennett A B. 1997. Divergent fructokinase genes are differ-
entially expressed in tomato. Plant Physiol, 113: 1379-1384.
Kriedmann P E. 1969. 14C translocation in orange plants. Aust J
Agric Res, 20: 291-300.
Martinez-Barajas E, Randall D D. 1996. Purification and char-
acterization of fructokinase from developing tomato
(Lycopersicon esculentum Mill.) fruits. Planta, 199: 451-
458.
(Managing editor: ZHAO Li-Hui)
Pego J V, Smeekens S C M. 2000. Plant fructokinases: a sweet
family get-together. Trends Plant Sci, 5: 531–536.
Qin Q-P, Zhang S-L, Xu C-J. 2003. Hexokinase and develop-
ment of plants. Plant Physiol Commun , 39: 1-8. (in Chinese)
Renz A, Stitt M. 1993. Substrate specificity and product inhibition
of different forms of fructokinases and hexokinases in devel-
oping potato tubers. Planta, 190: 166-175.
Schaffer A A, Petreikov M. 1997. Sucrose-to-starch metabolism
in tomato fruit undergoing transient starch accumulation.
Plant Physiol, 113: 739-746.
Smith S B, Taylor M A, Burch L R, Davies H V. 1993. Primary
structure and characterization of a cDNA clone of
fructokinase from potato (Solanum tuberosum L. cv. Record).
Plant Physiol, 102: 1043.
Taylor M A, Ross H A, Gardner A, Davies H V. 1995. Charac-
terization of a cDNA encoding fructokinase from potato
(Solanum tuberosum L.). J Plant Physiol, 145: 253–256.
Zhang S, Nichols S E, Dong J G. 2003. Cloning and characteriza-
tion of two fructokinases from maize. Plant Sci, 165: 1051-
1058.
Zhao Z-Z, Zhang S-L, Xu C-J, Chen K-S, Liu S-T. 2001. Roles of
sucrose-metabolizing enzymes in accumulation of sugars in
satsuma mandarin fruit. Acta Hortic Sin，28: 112-118. (in
Chinese with English abstract)
Zhao Z-Z, Zhang S-L, Chen J-W, Tao J, Wu Y-J. 2002. The physi-
ological mechanism on the difference of sugar accumulation
in citrus varieties. Sci Agric Sin ，35: 541-545. (in Chi-
nese with English abstract)
Zhao Z-Z, Zhang S-L, Liu S-T, Chen J-W, Tao J. 2003. Effect of
extra nitrogenous fertilizer on sugar accumulation in juice sacs
of Satsuma mandarin fruit. Acta Agric Nucleatae Sin, 17:
119-122. (in Chinese with English abstract)