Isolation and characterization of a novel glycosyltransferase that ...

Isolation and characterization of a novel glycosyltransferase that converts phloretin to phlorizin, a potent antioxidant in apple He´le`ne Jugde´1, Danny Nguy1, Isabel Moller1, Janine M. Cooney2 and Ross G. Atkinson1 1 The Horticulture and Food Research Institute of New Zealand Ltd, HortResearch, Auckland, New Zealand 2 The Horticulture and Food Research Institute of New Zealand, Ruakura, Hamilton, New Zealand

Keywords antioxidant; apple; phloretin; phlorizin; uridine diphosphate glycosyltransferase Correspondence R. Atkinson, Mt Albert Research Centre, HortResearch, Private Bag 92 169, Auckland, New Zealand Fax: +64 9 925 7001 Tel: +64 9 925 7000 E-mail: [email protected] Note Nucleotide sequence data are available in the DDBJ ⁄ EMBL ⁄ GenBank databases under the accession number EU246349 (Received 27 February 2008, revised 22 May 2008, accepted 28 May 2008)

The dihydrochalcone phlorizin (phloretin 2¢-glucoside) contributes to the flavor, color and health benefits of apple fruit and processed products. A genomics approach was used to identify the gene MdPGT1 in apple (Malus x domestica) with homology to the UDP-glycosyltransferase 88 family of uridine diphosphate glycosyltransferases that show specificity towards flavonoid substrates. Expressed sequence tags for MdPGT1 were found in all tissues known to produce phlorizin including leaf, flower and fruit. However, the highest expression was measured by quantitative PCR in apple root tissue. The recombinant MdPGT1 enzyme expressed in Escherichia coli glycosylated phloretin in the presence of [3H]-UDP-glucose, but not other apple antioxidants, including quercetin, naringenin and cyanidin. The product of phloretin and UDP-glucose co-migrated with an authentic phlorizin standard. LC ⁄ MS indicated that MdPGT1 could glycosylate phloretin in the presence of three sugar donors: UDP-glucose, UDP-xylose and UDP-galactose. This is the first report of functional characterization of a UDP-glycosyltransferase that utilizes a dihydrochalcone as its primary substrate.

doi:10.1111/j.1742-4658.2008.06526.x

The dihydrochalcone phlorizin (phloretin 2¢-glucoside, synonyms: phlorhizin, phloridzin, phlorrhizin; Fig. 1) is the major phenolic glucoside found in apple trees. Phlorizin has a bitter taste that contributes to the characteristic flavor of cider [1], and its dimerized oxidation products contribute to the color of apple juices [2]. However, subsequent to its isolation from the bark of the apple tree in 1835 [3], phlorizin has attracted most scientific interest through its use as a pharmaceutical and as a tool for human physiology research. Its principal pharmacological action is to produce renal glycosuria and block intestinal glucose adsorption by inhibition of the sodium-linked glucose transporters [4]. Phlorizin and its derivatives have also been shown to be antioxidants in vitro [5], and to have a range of

bioactive functions, such as inhibition of lipid peroxidation [6,7], prevention of bone loss [8], enhancement of memory [9] and inhibition of cancer cell growth [10]. Until recently, phlorizin was believed to exist only in Malus species. However phloretin glycosides have been reported in the leaves of Australian native sarsaparilla (Smilax glyciphylla) [11], sweet tea (Lithocarpus polystachyus) [12] and at very low levels in strawberry fruit [13]. In apple trees, phlorizin is found primarily in the young shoots, roots, leaves and bark. In fruit, phlorizin is most abundant in the seeds, with intermediate levels in both the core and the skin, and the lowest level in the cortex. Variation has been assessed within apple trees, between orchards, between different

Abbreviations DAFB, days after full bloom; EST, expressed sequence tag; MdPGT1, Malus x domestica UDP-glucose:phloretin glycosyltransferase 1; qPCR, quantitative PCR; UGT, UDP-glycosyltransferase.

3804

FEBS Journal 275 (2008) 3804–3814 ª 2008 The Authors Journal compilation ª 2008 FEBS

H. Jugde´ et al.

Phloretin glycosyltransferase from apple

5

OH

6 5´

HO

OH 1 1´

3’

OH HO

3

OH

2

OH O

OH O MdPGT1 UDP-glc

Phloretin

HO

O

O

OH OH

Phlorizin

Fig. 1. Structure of phlorizin and its aglycone phloretin. The conversion of phloretin to phlorizin (phloretin 2¢-glucoside) is mediated by M. x domestica uridine diphosphate glycosyltransferase (MdPGT1) in the presence of uridine diphosphate glucose (UDP-glc).

cultivars and among mutants [14,15]. Despite this information, little is known of the in planta function of phlorizin in apple tree physiology, although it has been suggested that it might act in apple tree growth and development [16] or be an inhibitor of bacterial [17] or fungal growth [18]. The molecular basis for production of phlorizin in planta has not been described. Phloretin is a product of the phenylpropanoid pathway [19], with conversion to its glucoside phlorizin likely to be catalysed by the action of a UDP-glycosyltransferase (UGT). UGTs mediate the transfer of a sugar residue from an activated nucleotide sugar to acceptor molecules. Plants contain large families of UGTs, with over 100 genes being described in Arabidopsis [20]. These genes have a common signature motif of 44 amino acids thought to be involved in binding of the UDP moiety of the activated sugar [21]. A phylogenetic analysis established the presence of distinct groups (A–N) and families (UGT71-92) of UGT genes in Arabidopsis [20] and this facilitated the characterization of many new activities [22–27]. Although initially thought to be promiscuous enzymes, recent evidence suggests that their broad substrate specificity is limited by regio-specificity [28,29] and, in some cases, UGTs have been shown to be highly specific [30]. Using a functional genomics approach, we have identified and characterized a novel UGT from apple belonging to the UGT88 family. We have established that the MdPGT1 enzyme utilizes the dihydrochalcone phloretin as its primary substrate.

Results The MdPGT1 gene and its predicted protein Gene mining identified over 60 different genes with homology to known UGT sequences amongst the

approximately 270 000 apple expressed sequence tag (EST) sequences in GenBank (as of October 2007). His6-tagged recombinant proteins for eight full-length apple UGT sequences were tested for their ability to glycosylate phloretin in the presence of [3H]-UDP-glucose (data not shown). Only one of the eight apple UGT proteins was shown to possess phloretin glycosyltransferase activity and this clone was tentatively designated MdPGT1 (for Malus x domestica UDP-glucose:phloretin glycosyltransferase 1). The cDNA clone for MdPGT1 was 1745 nucleotides in length and contained an ORF of 483 amino acids with 64 nucleotides of 5¢ UTR and 232 nucleotides of 3¢ UTR. The predicted MdPGT1 protein had a molecular mass of 53.5 kDa and a pI of 5.8 and contained the ‘PSPG’ consensus sequence of 44 amino acids found in all UGTs that is thought to be involved in binding of the UDP moiety of the activated sugar [21]. Within this consensus sequence, MdPGT1 contained two highly conserved motifs ‘WXPQ’ and ‘HCGWNS’ found in 95% of UGT sequences [31]. A framework phylogenetic tree was constructed using MdPGT1 and representative members of the Arabidopsis UGT tree provided by Ross et al. [20]. This framework tree indicated that MdPGT1 clustered with the sole Arabidopsis UGT88 family sequence AtUGT88A1 (data not shown). MdPGT1 shared significant sequence homology with five biochemically characterized members of the UGT88 family: AmC4¢GT, Antirrhinum majus UDP-glucose:chalcone 4¢-O-glucosyltransferase (accession number AB198665; 39.8% amino acid identity) [32]; GmIF7GT, Glycine max UDP-glucose:isoflavone 7-O-glucosyltransferase (AB292164; 40.9% identity) [33]; LvC4¢GT, Linaria vulgaris UDPglucose:chalcone 4¢-O-glucosyltransferase (BAE48240; 40.6% identity) [34]; Rosa hybrida UDP-glucose:anthocyanidin 5,3-O-glucosyltransferase (BAD99560; 44.8% identity) [35]; and SbB7GAT, Scutellaria baicalensis UDP-glucuronate:baicalein 7-O-glucuronosyltransferase (BAC98300; 40.0% identity). MdPGT1 and these five sequences are shown aligned in Fig. 2. MdPGT1 also showed high sequence identity (approximately 34–54%) to other UGT88 family members of unknown biological function from Arabidopsis thaliana (AAM65752), Stevia rebaudiana (AAR06919) [36], Oryza sativa (BAC10743), Trifolium pratense (AAR06919), Vigna mungo, (BAA36412) [37] and Vigna angularis (BAB86919, BAB8692, BAB86923). A more detailed phylogenetic tree was then constructed using these sequences aligned with selected members of Arabidopsis UGT71 and 72 families (i.e. the UGT families most similar to the UGT88 family) [20]. This tree indicates that apple MdPGT1 clusters


3805

H. Jugde´ et al.


Fig. 2. Amino acid alignment of MdPGT1 and other biochemically characterized members of the UGT88 family. AmC4¢GT, A. majus UDPglucose:chalcone 4¢-O-glucosyltransferase (UGT88D3, accession number AB198665); GmIF7GT, G. max UDP-glucose:isoflavone 7-O-glucosyltransferase (UGT88E3, AB292164); LvC4¢GT, L. vulgaris UDP-glucose:chalcone 4¢-O-glucosyltransferase (BAE48240); MdPGT1, M. x domestica UDP-glucose:phloretin 2¢-O-glucosyltransferase (UGT88F1, EU246349); RhA53GT, R. hybrida UDP-glucose:anthocyanidin 5,3-Oglucosyltransferase (BAD99560); and SbB7GAT, S. baicalensis UDP-glucuronate:baicalein 7-O-glucuronosyltransferase (BAC98300). Black and grey boxes contain residues that are identical and similar, respectively. The underlined region indicates the ‘PSPG’ motif conserved in all plant glycosyltransferases.

with members of the UGT88 family and separately from other Arabidopsis UGT sequences (Fig. 3). MdPGT1 separates from other UGT88 family members with putative glycosyltransferases from T. pratense (AAR06919) and V. mungo (BAA36412). However, MdPGT1 shares less than 60% amino acid identity with either of these sequences and therefore MdPGT1 is designated MaldoUGT88F1 according to 3806

the systematic glycosyltransferase nomenclature of Ross et al. [20]. Expression of MdPGT1 In silico expression profiling indicated that MdPGT1 ESTs were highly abundant (255 ESTs) and could be found in all tissues known to produce phlorizin,


H. Jugde´ et al.


VaGT5 GmIF7GT VaGT3

Sb7GAT LvC4´GT

VaGT1 AmC4´GT

RhA53GT AtUGT88A1

1000

1000

1000

SrGT88B1

1000

Fig. 3. Phylogenetic comparison of MdPGT1 and other selected plant glycosyltransferase sequences. The gene identifiers are as shown in Fig. 2 plus AtUGT72B1 (At4g01070), AtUGT72C1 (At4g36770), AtUGT72D1 (At2g18570), AtUGT72E1 (At3g50740), AtUGT71B1 (At3g21750), AtUGT71C1 (At2g29750), AtUGT71D1 (At2g29730), AtUGT88A1 (At3g16520), Arabidopsis thaliana; OsGT (BAC10743) Oryza sativa; SrGT88B1 (AAR06919), Stevia rebaudiana; TpGT (BAE71309), Trifolium pratense; VaGT1 (BAB86919), VaGT3 (BAB86921), VaGT5 (BAB86923), Vigna angularis; and VmUFGT4 (BAA36412), V. mungo. Bootstrap values of > 80% (1000 bootstrap replicates) are given. Members of the UGT88 family are enclosed in a grey circle.

UGT71D1

MdPGT1 1000 1000

984

1000 1000

1000

TpGT

969

UGT71C1

VmUFGT4 OsGT

UGT71B1 UGT72B1

UGT72D1

including leaf, seed and fruit (Fig. 4A). ESTs were most abundant in libraries constructed from flower, phloem and shoot tissue, but absent in libraries constructed from senescing leaf and developing fruit harvested 24–87 days after full bloom (DAFB). In ripe fruit, ESTs were most abundant in core tissue, although ESTs were also detected in the cortex and skin. As in silico expression profiling can be influenced by library sampling depth and cloning bias, MdPGT1 gene expression was also measured by quantitative PCR (qPCR) in four apple tissues: expanding leaf, ripe fruit (150 DAFB), open flower and root tip (Fig. 4B). The qPCR expression data indicated that MdPGT1 was expressed in all four apple tissues tested, with the highest expression being detected in root tip tissue. Together, the expression data indicate that the apple MdPGT1 transcript is abundant (0.1% of total transcripts), is expressed in a wide range of tissues, and can be found in all tissues where phlorizin has been reported. Substrate preference of MdPGT1 Purified recombinant MdPGT1 protein was isolated by Ni2+ affinity and gel filtration chromatography. Purification of the fusion protein was indicated by SDS ⁄ PAGE as demonstrated by the appearance of a protein of the expected 59 kDa size (53.5 kDa for

UGT72E1

UGT72C1

MdPGT1 + 5.5 kDa for the His6 tag and associated residues) in MdPGT1 extracts but not in equivalent vector control extracts (Fig. 5, lanes 2–5). Western analysis using His6 monoclonal antibodies confirmed that the 59 kDa was a His-labelled recombinant protein (Fig. 5, lane 6). For functional characterization of MdPGT1, aliquots of gel filtration-purified recombinant protein were assayed for glycosyltransferase activity using [3H]-UDPglucose as the sugar donor. Fifteen substrates (caffeic acid, catechin, chlorogenic acid, 2-coumaric acid, 3-coumaric acid, 4-coumaric acid, cyanidin, 3-3,4 dihydroxyphenyl) propionic acid, epicatechin, 3-hydroxybenzoic acid, naringenin, phloretin, protocatechuic acid, quercetin and rutin) that are natural constituents of apple fruit [38] or commonly available phenolic compounds were screened for activity. Of these substrates, only phloretin was utilized by MdPGT1 as an acceptor. The vector control showed no activity towards phloretin (or other substrates), indicating the glycosylation reaction was specific to the MdPGT1 enzyme. MdPGT1 enzyme activity was tested over a pH range of 5.0–10.0. The enzyme showed significant activity from pH 7.0–8.0 with maximum activity at pH 7.7. Activity decreased to 20% at pH 6.0 and 9.0 and less than 5% at pH 5.0 and 10.0 (see supplementary Fig. S1). The enzyme showed a broad temperature


3807

H. Jugde´ et al.

A

2.5

Relative expression


2.0 1.5 1.0 0.5

Bu

d Ro ot Fl ow er Sh oo t Se ed O th er

Fr u

it [ Fr 1] ui t[ 2] Fr u Fr it [ 3 ui t[ ] 3Fr sk ui t[ ] 3 Fr -cx ] ui t[ 3 Fr -cr ] ui t[ 1 Le -3] af [in Le f] af [ Le 1] af [ P h 2] lo em Xy le m

0.0

Tissue type Expression corrected for actin

B 1.2 1.0 0.8 0.6 0.03 0.02 0.01 0.00

Root

Leaf Flower Tissue type

Fruit

Fig. 4. Expression of MdPGT1 in apple tissues. (A) Relative expression of MdPGT1 in silico. Data are derived from the supplementary Table S1 where the distribution of MdPGT1 ESTs by apple library in GenBank is compared. The number of MdPGT1 ESTs in each tissue was divided by the total number of ESTs sequenced for that tissue and expressed relative to the phloem sample. Tissue types: Fruit [1], small fruit harvested 10 DAFB [days after full bloom]; Fruit [2], developing fruit harvested 24–87 DAFB; Fruit [3], ripe fruit harvested 126–150 DAFB; Fruit [3-sk], ripe fruit skin; Fruit [3-cr], ripe fruit cortex; Fruit [3-cr], ripe fruit core; Fruit [1–3], ripe fruit combining skin, cortex and core; Leaf [inf], leaf tissue pathogen challenged; Leaf [1], young and expanding leaf; Leaf [2], senescing leaf. (B) Relative expression of MdPGT1 using quantitative RT-PCR. MdPGT1-specific primers were used to measure transcript levels in the four apple tissues indicated. Expression was corrected against apple actin and is given relative to the root sample (value set at 1). Standard errors of the means are indicated for each tissue.

range (15–50 C) with maximum activity at 25 C (supplementary Fig. S1). Enzyme activity was not affected by the addition of divalent cations (Mg2+ or Mn2+, 0–25 mm) or monovalent cations (Na+, 0–100 mm or K+, 0–10 mm) (data not shown). Kinetic parameters were determined for MdPGT1 with respect to phloretin and UDP-glucose. The enzyme showed a Km of 0.62 ± 0.1 lm for phloretin with a turnover rate of 9.72 · 10)4 molÆs)1Æmol)1. The observed Km for UDP-glucose was approximately 13 lm. These kinetic values are similar to those reported for other glucosyltransferases that utilize anthocyanidin and flavonol substrates [39–41]. LC ⁄ MS analysis Products of the reaction between MdPGT1, phloretin and UDP-glucose were analyzed by LC ⁄ MS, and the base peak plots obtained (Fig. 6c) were compared with 3808

those for phloretin and phlorizin standards (Fig. 6A). The product of the MdPGT1 enzyme reaction was observed at 38.0 min, which is the same retention time as the phlorizin standard. No phloretin was observed at 47.0 min (Fig. 6C). An LC ⁄ MS run where the phlorizin standard was spiked into the MdPGT1 reaction mixture further confirmed that that the products of the reaction had identical retention times (Fig. 6F). The MdPGT1 enzyme was tested with phloretin in the presence of two additional activated sugar donors: UDP-xylose and UDP-galactose. Base peak plots indicated that a single glycosylated product was formed with UDP-xylose with a retention time of 40.5 min (Fig. 6D). Two peaks were observed in the presence of UDP-galactose: a small peak at 36.5 min and a larger peak at 38.0 min (Fig. 6B). The larger peak co-eluted with the phlorizin standard at 38.0 min when the products of the UDP-galactose reaction were spiked with authentic phlorizin (Fig. 6E).


H. Jugde´ et al.


sponding formate adducts, m ⁄ z 481 [M+formate])1 for the glucoside (peak 3; Fig. 6C) and the galactosides (peaks 1 and 2; Fig. 6B) and m ⁄ z 451 [M+formate])1 for the xyloside (peak 4; Fig. 6D). MS2 on the formate adducts, identified the expected pseudo-molecular ion at m ⁄ z 435 [M-1]) for the glucoside (Fig. 6K) and galactosides (Fig. 6I,J) and m ⁄ z 405 [M-1]) for the xyloside (Fig. 6L). MS3 on the m ⁄ z 435 [M-1]) glucoside and galactoside ions and the m ⁄ z 405 [M-1]) xyloside ion all identified the m ⁄ z 273 [M-1]) of the phloretin aglycone.

kDa 60 50 40 30 20

1

2

3

4

5

6

Fig. 5. Purification of recombinant MdPGT1 produced in E. coli. SDS ⁄ PAGE separation of proteins purified by Ni2+ affinity chromatography (His-Trap) and gel filtration (G200 Superdex). 1, Benchmark ladder (Invitrogen, Auckland, New Zealand); 2 and 3, fractions of purified MdPGT1; 4 and 5, equivalent fractions of purified pET30a(+) vector control; 6, immunodetection of MdPGT1 using a His6 monoclonal antibody. Arrow indicates the position of the recombinant MdPGT1 band.

Discussion A genomics approach was used to identify the gene MdPGT1 in apple ‘Royal Gala’ with homology to UGT88 family members from snapdragon, soybean, rose, toadflax and the Chinese medicinal herb S. baicalensis that show specificity towards flavonoid substrates. Glycosyltransferase assays and LC ⁄ MS revealed that the recombinant MdPGT1 enzyme could specifically glycosylate phloretin in the presence of UDP-glucose, UDP-xylose and UDP-galactose. This is the first report of functional characterization of a UGT that utilizes a dihydrochalcone as its primary substrate.

Full scan and MS ⁄ MS mass spectral data were used to further characterize the products of the MdPGT1 reactions. Phloretin (Fig. 6H) was detected as its pseudo-molecular ion m ⁄ z 273 [M-1]), whereas phlorizin (Fig. 6G) and the phloretin glycoside reaction products were detected predominately as the corre-

[M + formate]–1

Phloretin

Phlorizin

100 A

Pz

100

[M – 1]–

481

0 100 B

100 P Relative abundance

Relative abundance

3

0 100 D

4

0 100 E

0 100

MS2 481

MS2 481

435

4

0 50

54

0

273

K

273

L

MS 435 3

435 [M – 1]– 405

[M + formate]–1 451

100

J

MS3 435

273 435 481

273

273

435

481

100 2

MS2 481

46

I

MS3 435

435

0

0

38 42 Time (min)

H 167 273

435

1

1

34

273 481

100 3

30

MS 167 3

319 387

0

0 100 F

123

167

MS 273

1

26

273 [M – 1]– 2

0 100 C

G

MS3 435

435

0 2

273

435 MS2 481

MS3 405

MS 451 2

273 405

273

m/z

Fig. 6. LC ⁄ MS analysis of the products of the reaction between MdPGT1, phloretin and UDP-sugars. Base peak plots: (A) mixed standard of phlorizin (Pz) and phloretin (P); (B) phloretin + UDP-galactose; (C) phloretin + UDP-glucose; (D) phloretin + UDP-xylose; (E) phloretin + UDPgalactose spiked with phlorizin; (F) phloretin + UDP-glucose spiked with phlorizin; MS spectra: (G) fullscan, MS2 and MS3 data for phlorizin; (H) fullscan, MS2 and MS3 data for phloretin; (I) fullscan, MS2 and MS3 data for peak 1; (J) fullscan, MS2 and MS3 data for peak 2; (K) fullscan, MS2 and MS3 data for peak 3; and (L) fullscan, MS2 and MS3 data for peak 4.


3809

H. Jugde´ et al.


EST profiling and qPCR data indicated that the MdPGT1 gene was expressed in leaf, root, flower and fruit, which is consistent with the presence of phlorizin in these apple tissues. The high level of MdPGT1 expression in root tissue was not unexpected because phlorizin is currently produced commercially from the root bark of apple trees. The etymology of the word phlorizin (from the Greek phloios bark + rhiza root) also suggests that MdPGT1 would be expressed in root. In silico expression profiling did not indicate that the MdPGT1 transcript was abundant in root; however, this may reflect a sampling bias due to the low number of libraries made for this tissue. Other apple UGT enzymes may also be required to glycosylate phloretin in other apple tissues (e.g. fruit). These enzymes may be identified in the remaining 52 uncharacterized UGT genes that we identified amongst the apple EST sequences currently found in GenBank or in the full apple genome sequence when it becomes available. Phylogenetic analysis indicated that MdPGT1 shared significant homology with 13 existing members of the UGT88 family. Five sequences within this family have been ascribed a definitive function to date. AmC4’GT (A. majus UDP-glucose:chalcone 4¢-O-glucosyltransferase), LvC4’GT (L. vulgaris UDP-glucose:chalcone 4¢-O-glucosyltransferase), SbB7GAT (S. baicalensis UDP-glucuronate:baicalein 7-O-glucuronosyltransferase) and GmIF7GT (G. max UDP-glucose:isoflavone 7-Oglucosyltransferase) are flavonoid-specific UGTs that transfer the glycosyl moiety to equivalent positions in their target substrates (the 4¢ position of chalcones for AmC4¢GT [32] and LvC4¢GT [34]; the seven-position of flavones for SbB7GAT [42] and isoflavones for GmIF7GT [33]). The exception is RhA53GT, (UDPglucose:anthocyanidin 5,3-O-glucosyltransferase) which catalyzes the successive glucosylation of anthocyanidins at the 5¢ and 3¢ position [35]. It is possible that some of these enzymes may also accept dihydrochalcones as substrates but this has not been reported. This may be unlikely given that dihydrochalcone glycosides are found almost exclusively in Malus spp. and have been used as markers for the detection of apple admixtures in other fruit juices and purees [43]. The glycosylated product of phloretin and UDP-glucose co-migrated with a known phlorizin standard, indicating that MdPGT1 is likely to be an enzyme that glycosylates phloretin to phlorizin in planta. The recombinant MdPGT1 protein also produced glycosides with UDP-xylose and UDP-galactose, with one of the products glycosylated with UDP-galactose having the same retention time as phlorizin. Apples have been reported to produce two phloretin glycosides in 3810

addition to phloretin 2¢-glucoside (phlorizin), namely phloretin 2¢-galactoside [44] and phloretin 2¢-xylosyl (1 fi 6) glucoside [45]. Although the exact nature of the glycosylation in these reactions has not been determined (because no authentic apple compounds were available for comparison), the MS spectra clearly indicated that the sugar molecules were attached to phloretin. It is likely the regio-specificity of MdPGT1 is to the 2¢-hydroxyl position because substrates (e.g. naringenin and quercetin) with hydroxyl groups in alternative positions were not utilized as substrates. NMR could be used to confirm the regio-specificity of glycosylation and to clarify the nature of the two glycosylated products obtained using UDP-galactose as the sugar donor. In conclusion, our results have identified a key enzyme in the biochemical pathway to phlorizin production in apples. The ability to manipulate phlorizin levels in fruit is an important target in apple breeding programmes as ‘the use of phlorizin may provide the molecular basis for the clinical observation that ‘‘an apple a day keeps the doctor away’’’ [4].

Experimental procedures Sequence identification and phylogenetics GenBank apple EST sequences were automatically parsed through two rounds of contig building and a set of nonredundant contig sequences was derived [46]. Contig sequences were then blast searched (expect value of exp)05) using previously published UGT genes from GenBank. A full-length sequence in each contig was selected for complete double strand sequencing. Amino acid alignments of predicted proteins were constructed using clustalx (version 1.8). All proteins were checked for the presence of the common signature motif of 44 amino acids found in plant UGTs [21]. For phylogenetic analysis, all amino acid sequences were initially aligned using clustalx then manually edited. Arabidopsis UGT sequences were obtained from the website http:// www.p450.kvl.dk/UGT.shtml. Phylogenetic trees were constructed using phylip and visualized in treeview (version 1.6.6). Confidence values for groupings in trees were obtained using a bootstrap neighbour-joining tree using 1000 bootstrap trials.

Plant material, RNA extraction and qPCR Total RNA was isolated from apple tissues by a modified silica RNA extraction method [47]. The RNA concentration of each sample was measured using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies,


H. Jugde´ et al.

Inc., Wilmington, DE, USA). RNA samples (approximately 3 lg in 30 lL reactions) were treated for 30 min at 37 C with DNase I (Ambion, Inc., Austin, TX, USA) to remove any minor genomic DNA contamination. The DNase was heat inactivated for 30 min at 50 C. Reverse transcription was performed in 20 lL reactions as per the manufacturer’s instructions using SuperScript III RNase H-reverse transcriptase (Invitrogen, Auckland, New Zealand), 500 ng of RNA and the primer NotI-PA 5¢-GACTAGTTCTAGAT CGCGAG CGGCCGCCCT(15)-3¢. Relative expression of MdPGT1 was assessed in apple root, leaf, flower and fruit by quantitative PCR experiments repeated twice. Four quantitative PCR reactions (10 lL) were run on a LightCycler 480 (Roche Diagnostics, Mannheim, Germany) using 3 lL of diluted cDNA, 0.5 lm primers specific for MdPGT1 (forward: 5¢-GAAGGGTGTGT TGCCAGAAGGGT-3¢; reverse: 5¢-GTCACGAACCCAC CAACCGACT-3¢) and 5 lL of LightCycler 480 SYBR Green Master (Roche Diagnostics) following the manufacturer’s instructions. Cycling conditions included an initial hot start at 95 C for 5 min, followed by 45 cycles of 95 C for 10 s, 60 C for 10 s and 72 C for 12 s. Fluorescence was measured at the end of each annealing step. Each PCR reaction was followed by a melting curve program to check that only single products were amplified, starting with denaturation at 95 C for 5 s before cooling to 65 C for 1 min then increasing at 0.1 CÆs)1 with continuous fluorescence measurement until 97 C was reached. Negative controls consisted of water in place of cDNA and were run with all reactions. Data were analyzed using lightcycler 480 software, version 1.2.0.169. A standard curve was generated using a serial dilution of plasmid containing MdPGT1, and the resultant efficiencies were used to calculate expression relative to apple actin (accession number CN938023) [48] to minimize variation in cDNA template levels.

Expression of MdPGT1 in Escherichia coli The ORF of MdPGT1 was amplified using primers RA335 5¢-ACGGGATCCATGGGAGACGTCATTGTACTG-3¢ and RA336 5¢-CCCAAGCTTTTATGTAATGCTACTAA CAAAGTTGAC-3¢. Amplified bands were purified using Qiaquick PCR cleanup columns (Qiagen GmbH, Hilden, Germany), digested with BamHI and HindIII (underlined in the primers above), and ligated into the corresponding sites of the pET30a(+) vector (Novagen, Madison, WI, USA). The clone was sequence verified against the original EST. Recombinant N-terminal His6-tagged protein was expressed from pET-30a(+) plasmids in E. coli BL21Codon-Plus-RIL cells. Cultures were grown in a ZYM5052 autoinducible media [49] at 37 C for 4 h at 300 r.p.m. The temperature was then lowered to 16 C and incubation continued for a further 60 h. Recombinant proteins were purified on 5 mL His-Trap chelating HP columns


(Amersham Biosciences, Little Chalfont, UK) and eluted using a continuous 0–250 mm imidazole gradient as described in Green et al. [50]. The concentrate was then applied to a 1.6 · 40 cm G200 Superdex gel filtration column (Pharmacia Biotech, Auckland, New Zealand) preequilibrated with 50 mm Tris–HCl (pH 7.5), 500 mm NaCl, 5 mm dithiothreitol at a flow rate of 1 mlÆmin)1. Highest purity fractions were pooled, adjusted to 15% glycerol and stored at )80 C. A pET-30(+) vector-only control was expressed and purified as above. Recombinant protein was analyzed on 12% (w ⁄ v) SDSTris-Tricine gels, electroblotted onto polyvinyldifluoride membrane, and blocked as described by Nieuwenhuizen et al. [47]. Proteins were immunolocalized with a His6 monoclonal antibody (Roche Diagnostics), 1 : 1000, (w ⁄ v), diluted in NaCl ⁄ Tris buffer containing milk powder. Membranes were incubated with anti-rabbit alkaline phosphatase conjugated secondary serum (Sigma-Aldrich, St Louis, MO, USA) and binding visualized using BCIPNBT (nitro-blue tetrazolium) Liquid Substrate System (Sigma-Aldrich).

UGT activity assays UGT activity assays were performed in triplicate in 50 lL reactions using approximately 1–2 lg of recombinant protein purified on His-Trap and G200 Superdex columns. Reactions were performed in glycosyltransferase assay buffer (50 mm Tris–HCl, pH 7.5, 2 mm dithiothreitol) with 2 lm substrate and 27 lm [3H]-UDP-glucose (uridine diphospho-d-[6-3H] glucose, 13.6 CiÆmmol)1; GE Heathcare, Little Chalfont, UK). Reactions were performed at 30 C for 30 min and terminated by addition of 10 lL of 2 m HCl. The reaction mixtures were extracted twice with 100 lL of ethyl acetate and 20 lL of the organic phase were combined with 1 mL of non-aqueous scintillation fluid and analyzed by liquid scintillation counting (Tri-Carb 2900TR; PerkinElmer, Boston, MA, USA). Boiled enzyme and pET30a(+) vector controls were run in parallel with all enzyme reactions. Reactions were shown to be linear with respect to time and enzyme concentration under standard conditions. The apparent Km value for phloretin was determined by varying the phloretin concentration from 4 lm to 0.01 lm with a fixed [3H]-UDP-glucose concentration of 27 lm. The Km value for UDP-glucose was determined by varying UDP-glucose concentration from 127 lm to 0.135 lm ([3H]UDP-glucose mixed with cold UDP-glucose) with a fixed phloretin concentration of 2 lm. For LC ⁄ MS analysis, scaled up reactions were performed containing approximately 10 lg enzyme, 10 lm phloretin and UDP-glucose (Sigma Aldrich), UDP-galactose (Sigma Aldrich) or UDP-xylose (CarboSource Services, University of Georgia, Athens, GA, USA) at a final concentration of 250 lm. Reactions were performed for 16 h, and stopped by addition of 10 lL of 10% glacial acetic acid. The products of 3 · 100 lL reactions were combined for LC ⁄ MS


3811

H. Jugde´ et al.


analysis. Authentic phloretin and phlorizin standards were purchased from Sigma Aldrich.

LC ⁄ MS analysis of in vitro reaction mixtures LC ⁄ MS employed an LTQ linear ion trap mass spectrometer fitted with an ESI interface (ThermoQuest, Finnigan, San Jose, CA, USA) coupled to an Ettan MDLC (GE Healthcare Bio-Sciences, Uppsala, Sweden). Phenolic compound separation was achieved using a Prodigy 5 lm ODS(3) 100 A˚ (Phenomenex, Torrance, CA, USA), 150 · 2 mm analytical column maintained at 35 C. A 0.2 lm inline filter (Alltech, Deerfield, IL, USA) was installed before the column. Solvents were (A) acetonitrile + 0.1% formic acid and (B) water + 0.1% formic acid and the flow rate was 200 lLÆmin)1. The initial mobile phase, 5% A ⁄ 95% B, was held for 5 min then ramped linearly to 10% A at 10 min, 17% A at 25 min, 23% A at 30 min, 30% A at 40 min, 97% A between 48–53 min before resetting to the original conditions. The sample injection volume was 50 lL. MS data were acquired in the negative mode using a datadependent LC ⁄ MS3 method. This method isolates and fragments the most intense parent ion to give MS2 data, then isolates and fragments the most intense daughter ion (MS3 data). To maximize sensitivity, the full scan range was set to m ⁄ z 420–490 from 0–42.9 min for the detection of phloretin glycosides and then to m ⁄ z 270–280 from 43 min for the detection of unreacted phloretin. The ESI voltage, capillary temperature, sheath gas pressure and sweep gas were set at )10 V, 275 C, 40 psi and 5 psi, respectively.

Acknowledgements We would like to thank all members of the apple and kiwifruit genomics programme team at HortResearch. Special thanks to Sarah Johnston for performing the qPCR reactions, William Laing and Richard Newcomb for critically reviewing the manuscript and Tim Holmes for photography. This work was funded by the New Zealand Foundation for Research, Science and Technology and through HortResearch Internal Capability Funding.

References 1 Whiting GC & Coggins RA (1975) Estimation of the monomeric phenolics from ciders. J Sci Food Agric 26, 1833–1838. 2 Ridgway T & Tucker G (1997) Phloridzin derivatives; food additives ⁄ chemopreventative drugs of the future. Biochem Soc Trans 25, 109S. 3 Petersen C (1835) Analyse des phloridzins. Ann Acad Sci Fr 15, 178.

3812

4 Ehrenkranz JR, Lewis NG, Kahn CR & Roth J (2005) Phlorizin: a review. Diabetes Metab Res Rev 21, 31–38. 5 Ridgway T, O’Reilly J, West G, Tucker G & Wiseman H (1996) Potent antioxidant properties of novel applederived flavonoids with commercial potential as food additives. Biochem Soc Trans 24, 391S. 6 Ridgway T, O’Reilly J, West G, Tucker G & Wiseman H (1997) Antioxidant action of novel derivatives of the apple-derived flavonoid phloridzin compared to oestrogen: relevance to potential cardioprotective action. Biochem Soc Trans 25, 106S. 7 Rezk BM, Haenen GR, van der Vijgh WJ & Bast A (2002) The antioxidant activity of phloretin: the disclosure of a new antioxidant pharmacophore in flavonoids. Biochem Biophys Res Commun 295, 9–13. 8 Puel C, Quintin A, Mathey J, Obled C, Davicco MJ, Lebecque P, Kati-Coulibaly S, Horcajada MN & Coxam V (2005) Prevention of bone loss by phloridzin, an apple polyphenol, in ovariectomized rats under inflammation conditions. Calcif Tissue Int 77, 311–318. 9 Boccia MM, Kopf SR & Baratti CM (1999) Phlorizin, a competitive inhibitor of glucose transport, facilitates memory storage in mice. Neurobiol Learn Mem 71, 104– 112. 10 Veeriah S, Kautenburger T, Habermann N, Sauer J, Dietrich H, Will F & Pool-Zobel BL (2006) Apple flavonoids inhibit growth of HT29 human colon cancer cells and modulate expression of genes involved in the biotransformation of xenobiotics. Mol Carcinog 45, 164–174. 11 Cox SD, Jayasinghe KC & Markham JL (2005) Antioxidant activity in Australian native sarsaparilla (Smilax glyciphylla). J Ethnopharmacol 101, 162–168. 12 Dong H, Ning Z, Yu L, Li L, Lin L & Huang J (2007) Preparative separation and identification of the flavonoid phlorhizin from the crude extract of Lithocarpus polystachyus Rehd. Molecules 12, 552–562. 13 Hilt P, Schieber A, Yildirim C, Arnold G, Klaiber I, Conrad J, Beifuss U & Carle R (2003) Detection of phloridzin in strawberries (Fragaria x ananassa Duch.) by HPLC-PDA-MS ⁄ MS and NMR spectroscopy. J Agric Food Chem 51, 2896–2899. 14 Hunter MD & Hull LA (1993) Variation in concentrations of phlorhizin and phloretin in apple foliage. Phytochemistry 34, 1251–1254. 15 Awad MA, de Jager A & van Westing LM (2000) Flavonoid and chlorogenic acid levels in apple fruit: characterisation of variation. Sci Hortic 83, 249–263. 16 Zhang XZ, Zhao YB, Li CM, Chen DM, Wang GP, Chang RF & Shu HR (2007) Potential polyphenol markers of phase change in apple (Malus domestica). J Plant Physiol 164, 574–580.


H. Jugde´ et al.

17 MacDonald RE & Bishop CJ (1952) Phloretin: an antibacterial substance obtained from apple leaves. Can J Bot 30, 486–489. 18 Gessler C, Patocchi A, Sansavini S, Tartarini S & Gianfranceschi L (2006) Venturia inaequalis resistance in apple. Crit Rev Plant Sci 25, 473–503. 19 Watts KT, Lee PC & Schmidt-Dannert C (2004) Exploring recombinant flavonoid biosynthesis in metabolically engineered Escherichia coli. Chembiochem 5, 500–507. 20 Ross J, Li Y, Lim E & Bowles DJ (2001) Higher plant glycosyltransferases. Genome Biol 2, 3004. 21 Li Y, Baldauf S, Lim EK & Bowles DJ (2001) Phylogenetic analysis of the UDP-glycosyltransferase multigene family of Arabidopsis thaliana. J Biol Chem 276, 4338– 4343. 22 Lim EK, Doucet CJ, Li Y, Elias L, Worrall D, Spencer SP, Ross J & Bowles DJ (2002) The activity of Arabidopsis glycosyltransferases toward salicylic acid, 4-hydroxybenzoic acid, and other benzoates. J Biol Chem 277, 586–592. 23 Jackson RG, Lim EK, Li Y, Kowalczyk M, Sandberg G, Hoggett J, Ashford DA & Bowles DJ (2001) Identification and biochemical characterization of an Arabidopsis indole-3-acetic acid glucosyltransferase. J Biol Chem 276, 4350–4356. 24 Lim EK, Jackson RG & Bowles DJ (2005) Identification and characterisation of Arabidopsis glycosyltransferases capable of glucosylating coniferyl aldehyde and sinapyl aldehyde. FEBS Lett 579, 2802–2806. 25 Lim EK, Ashford DA, Hou B, Jackson RG & Bowles DJ (2004) Arabidopsis glycosyltransferases as biocatalysts in fermentation for regioselective synthesis of diverse quercetin glucosides. Biotechnol Bioeng 87, 623–631. 26 Jones P, Messner B, Nakajima J, Schaffner AR & Saito K (2003) UGT73C6 and UGT78D1, glycosyltransferases involved in flavonol glycoside biosynthesis in Arabidopsis thaliana. J Biol Chem 278, 43910–43918. 27 Messner B, Thulke O & Schaffner AR (2003) Arabidopsis glucosyltransferases with activities toward both endogenous and xenobiotic substrates. Planta 217, 138– 146. 28 Lim EK, Higgins GS, Li Y & Bowles DJ (2003) Regioselectivity of glucosylation of caffeic acid by a UDPglucose:glucosyltransferase is maintained in planta. Biochem J 373, 987–992. 29 Hansen KS, Kristensen C, Tattersall DB, Jones PR, Olsen CE, Bak S & Moller BL (2003) The in vitro substrate regiospecificity of recombinant UGT85B1, the cyanohydrin glucosyltransferase from Sorghum bicolor. Phytochemistry 64, 143–151. 30 Fukuchi-Mizutani M, Okuhara H, Fukui Y, Nakao M, Katsumoto Y, Yonekura-Sakakibara K, Kusumi T, Hase T & Tanaka Y (2003) Biochemical and molecular characterization of a novel UDP-glucose:anthocyanin


31

32

33

34

35

36

37

38

39

40

41

3¢-O-glucosyltransferase, a key enzyme for blue anthocyanin biosynthesis, from gentian. Plant Physiol 132, 1652–1663. Paquette S, Moller BL & Bak S (2003) On the origin of family 1 plant glycosyltransferases. Phytochemistry 62, 399–413. Ono E, Fukuchi-Mizutani M, Nakamura N, Fukui Y, Yonekura-Sakakibara K, Yamaguchi M, Nakayama T, Tanaka T, Kusumi T & Tanaka Y (2006) Yellow flowers generated by expression of the aurone biosynthetic pathway. Proc Natl Acad Sci USA 103, 11075–11080. Noguchi A, Saito A, Homma Y, Nakao M, Sasaki N, Nishino T, Takahashi S & Nakayama T (2007) A UDP-glucose:isoflavone 7-O-glucosyltransferase from the roots of soybean (Glycine max) seedlings. Purification, gene cloning, phylogenetics, and an implication for an alternative strategy of enzyme catalysis. J Biol Chem 282, 23581–23590. Ono E & Nakayama T (2007) Molecular breeding of novel yellow flowers by engineering the aurone biosynthetic pathway. Transgenic Plant J 1, 66–80. Ogata J, Kanno Y, Itoh Y, Tsugawa H & Suzuki M (2005) Anthocyanin biosynthesis in roses. Nature 435, 757–758. Richman AS, Gijzen M, Starratt AN, Yang Z & Brandle JE (1999) Diterpene synthesis in Stevia rebaudiana: recruitment and up-regulation of key enzymes from the gibberellin biosynthetic pathway. Plant J 19, 411–421. Mato M, Ozeki Y, Itoh Y, Higeta D, Yoshitama K, Teramoto S, Aida R, Ishikura N & Shibata M (1998) Isolation and characterization of a cDNA clone of UDP-galactose: flavonoid 3-O-galactosyltransferase (UF3GaT) expressed in Vigna mungo seedlings. Plant Cell Physiol 39, 1145–1155. Sanchez-Rabaneda F, Jauregui O, Lamuela-Raventos RM, Viladomat F, Bastida J & Codina C (2004) Qualitative analysis of phenolic compounds in apple pomace using liquid chromatography coupled to mass spectrometry in tandem mode. Rapid Commun Mass Spectrom 18, 553–563. Ford CM, Boss PK & Hoj PB (1998) Cloning and characterization of Vitis vinifera UDP-glucose:flavonoid 3-O-glucosyltransferase, a homologue of the enzyme encoded by the maize Bronze-1 locus that may primarily serve to glucosylate anthocyanidins in vivo. J Biol Chem 273, 9224–9233. Ishikura N & Mato M (1993) Partial-purification and some properties of flavonol 3-O-glycosyltransferases from seedlings of Vigna mungo, with special reference to the formation of kaempferol 3-O-galactoside and 3-Oglucoside. Plant Cell Physiol 34, 329–335. Brazier M, Cole DJ & Edwards R (2003) Partial purification and characterisation of a 2,4,5-trichlorophenol detoxifying O-glucosyltransferase from wheat. Phytochemistry 64, 419–424.


3813

H. Jugde´ et al.


42 Nagashima S, Hirotani M & Yoshikawa T (2000) Purification and characterization of UDP-glucuronate: baicalein 7-O-glucuronosyltransferase from Scutellaria baicalensis Georgi. cell suspension cultures. Phytochemistry 53, 533–538. 43 Dragovic-Uzelac V, Pospisil J, Levaj B & Delonga K (2005) The study of phenolic profiles of raw apricots and apples and their purees by HPLC for the evaluation of apricot nectars and jams authenticity. Food Chem 91, 373–383. 44 Barth SW, Fahndrich C, Bub A, Dietrich H, Watzl B, Will F, Briviba K & Rechkemmer G (2005) Cloudy apple juice decreases DNA damage, hyperproliferation and aberrant crypt foci development in the distal colon of DMH-initiated rats. Carcinogenesis 26, 1414– 1421. 45 Lommen A, Godejohann M, Venema DP, Hollman PC & Spraul M (2000) Application of directly coupled HPLC-NMR-MS to the identification and confirmation of quercetin glycosides and phloretin glycosides in apple peel. Anal Chem 72, 1793–1797. 46 Newcomb RD, Crowhurst RN, Gleave AP, Rikkerink EHA, Allan AC, Beuning LL, Bowen JH, Gera E, Jamieson KR, Janssen BJ et al. (2006) Analyses of expressed sequence tags from apple. Plant Physiol 141, 147–166. 47 Nieuwenhuizen NJ, Beuning LL, Sutherland PW, Sharma NN, Cooney JM, Bieleski LRF, Schro¨der R, MacRae EA & Atkinson RG (2007) Identification and characterisation of acidic and novel basic forms of

3814

actinidin, the highly abundant cysteine protease from kiwifruit. Funct Plant Biol 34, 946–961. 48 Espley RV, Hellens RP, Putterill J, Stevenson DE, Kutty-Amma S & Allan AC (2007) Red colouration in apple fruit is due to the activity of the MYB transcription factor, MdMYB10. Plant J 49, 414–427. 49 Studier FW (2005) Protein production by auto-induction in high-density shaking cultures. Protein Expr Purif 41, 207–234. 50 Bowles CE, Wilkinson I, Smith RA, Moir AJ, Montgomery H & Ross RJ (2007) Membrane reinsertion of a myristoyl-peptidyl anchored extracellular domain growth hormone receptor. Endocrinology 148, 824–830.

Supplementary material The following supplementary material is available online: Fig. S1. Effect of pH and temperature on the activity of MdPGT1. Table S1. Distribution of MdPGT1 ESTs by apple library in GenBank. This material is available as part of the online article from http://www.blackwell-synergy.com Please note: Wiley-Blackwell is not responsible for the content or functionality of any supplementary materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.
