Practical preparation of UDP-apiose and its applications for studying apiosyltransferase

Article abstract of DOI:10.1016/j.carres.2019.03.011

UDP-apiose, a donor substrate of apiosyltransferases, is labile because of its intramolecular self-cyclization ability, resulting in the formation of apiofuranosyl-1,2-cyclic phosphate. Therefore, stabilization of UDP-apiose is indispensable for its availability and identifying and characterizing the apiosyltransferases involved in the biosynthesis of apiosylated sugar chains and glycosides. Here, we established a method for stabilizing UDP-apiose using bulky cations as counter ions. Bulky cations such as triethylamine effectively suppressed the degradation of UDP-apiose in solution. The half-life of UDP-apiose was increased to 48.1 ± 2.4 h at pH 6.0 and 25 °C using triethylamine as a counter cation. UDP-apiose coordinated with a counter cation enabled long-term storage under freezing conditions. UDP-apiose was utilized as a donor substrate for apigenin 7-O-β-D-glucoside apiosyltransferase to produce the apiosylated glycoside apiin. This apiosyltransferase assay will be useful for identifying genes encoding apiosyltransferases.

Full text of DOI:10.1016/j.carres.2019.03.011

Accepted Manuscript
Practical preparation of UDP-apiose and its applications for studying
apiosyltransferase
Tae Fujimori, Ryoko Matsuda, Mami Suzuki, Yuto Takenaka, Hiroyuki Kajiura, Yoichi
Takeda, Takeshi Ishimizu
PII:
S0008-6215(19)30056-4
https://doi.org/10.1016/j.carres.2019.03.011
CAR 7693
DOI:
Reference:
To appear in: Carbohydrate Research
Received Date: 29 January 2019
Revised Date: 19 March 2019
Accepted Date: 21 March 2019
Please cite this article as: T. Fujimori, R. Matsuda, M. Suzuki, Y. Takenaka, H. Kajiura, Y. Takeda,
T. Ishimizu, Practical preparation of UDP-apiose and its applications for studying apiosyltransferase,
Carbohydrate Research (2019), doi: https://doi.org/10.1016/j.carres.2019.03.011.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
O
CH2OH
OH
O
NH
O
CH2OH
O
O
O
O
O
ApiT
OH
O
O
P
O
P
O
CH2
N
O
+
OH
OH
OH
OH
OH
O
O
OH
CH2OH
OH
OH
OH OH
7-O-(b-Glucosyl)apigenin
UDP-Apiose
Apiin

ACCEPTED MANUSCRIPT
Practical preparation of UDP-apiose and its applications for studying apiosyltransferase
Tae Fujimoria,1, Ryoko Matsudaa,1, Mami Suzukia, Yuto Takenakab, Hiroyuki Kajiuraa, Yoichi
Takedaa, Takeshi Ishimizua,b,*
aCollege of Life Sciences, Ritsumeikan University, Kusatsu, Shiga 525-8577, Japan
bRitsumeikan Global Innovation Research Organization, Ritsumeikan University, Kusatsu, Shiga
525-8577, Japan
*Corresponding author. Takeshi Ishimizu, College of Life Sciences, Ritsumeikan University,
1-1-1 Noji-Higashi, Kusatsu, Shiga, 5258577, Japan,
Tel.: +81-77-561-5973; Fax: +81-77-599-4358; E-mail: ishimizu@fc.ritsumei.ac.jp
1These authors contributed equally to this study.
Abbreviations: Api, apiose; AXS1, UDP-D-apiose/UDP-D-xylose synthase; GlcUA, glucuronic acid;
PcApiT, Petroselinum crispum UDP-apiose: apigenin 7-O-β-D-glucoside apiosyltransferase; RG-II,
rhamnogalacturonan II; Xyl, xylose
1

ACCEPTED MANUSCRIPT
Abstract
UDP-apiose, a donor substrate of apiosyltransferases, is labile because of its intramolecular
self-cyclization ability, resulting in the formation of apiofuranosyl-1,2-cyclic phosphate. Therefore,
stabilization of UDP-apiose is indispensable for its availability and identifying and characterizing the
apiosyltransferases involved in the biosynthesis of apiosylated sugar chains and glycosides. Here, we
established a method for stabilizing UDP-apiose using bulky cations as counter ions. Bulky cations
such as triethylamine effectively suppressed the degradation of UDP-apiose in solution. The half-life
of UDP-apiose was increased to 48.1 ± 2.4 h at pH 6.0 and 25 °C using triethylamine as a counter
cation. UDP-apiose coordinated with a counter cation enabled long-term storage under freezing
conditions. UDP-apiose was utilized as a donor substrate for apigenin 7-O-β- -glucoside
D
apiosyltransferase to produce the apiosylated glycoside apiin. This apiosyltransferase assay will be
useful for identifying genes encoding apiosyltransferases.
Key words: apiin, apiose, glycoside, glycosyltransferase, nucleotide sugar, UDP-apiose
2

ACCEPTED MANUSCRIPT
1. Introduction
Apiose, a branched-chain pentose found in a wide variety of plants as a secondary metabolite [1], was
originally discovered as a sugar component of the flavone glycoside apiin [2]. Apiin,
apigenin-7-O-β-
D
-apiofuranosyl-(1→2)-β- -glucopyranoside, is found in winter-hardy plants such as
D
parsley and celery. Nearly 1,200 glycosides containing apiose residues have been identified [1].
Their aglycons include phenolics, terpens, terpenoids, cyanogenic glycosides, and alcohols.
Although the role of apiosylation is unclear, some studies suggested that apiosylated compounds
serve as cryoprotectants [3,4] and protect against oxidative stress [5]. Apiose residues are also found
in apiogalacturonan and rhamnogalacturonan II (RG-II), which are pectic domains of the plant cell
wall [6,7]. The apiose residue in RG-II provides scaffold for crosslinking with boron, resulting in the
formation of a rigid cell wall by pectin dimerization [8,9]. Defects in the dimerization of RG-II
resulting from a low-boron supply [10] or alterations in the structure of the side chain containing the
apiosyl residue in RG-II [11,12] caused growth defects in Arabidopsis. Recently, apiose was detected
in the acid hydrolysate of methanolic extracts from some bacteria species, although the structure of
the apiosylated glycosides has not been determined [13,14].
UDP-α- -apiose (UDP-Api) serves as an apiose donor for apiosylated compounds. This
D
compound was first identified in parsley, as the acid hydrolysate of its sugar nucleotides fraction
contained apiose [15]. UDP-Api was synthesized from UDP-glucuronic acid (UDP-GlcUA) by
UDP-Api/UDP-xylose (UDP-Xyl) synthase (AXS) [16,17]. This enzyme catalyzes both
decarboxylation of UDP-GlcUA and the subsequent re-arrangement of the carbon skeleton and ring
contraction to form UDP-Api [18,19]. The synthesis of UDP-Api and UDP-Xyl occurs
simultaneously, and these molecules cannot be interconverted [17]. Arabidopsis contains two AXS
genes (AXS1 and AXS2) and both are expressed in all plant tissues, suggesting that AXS genes have
housekeeping function in Arabidopsis [17]. Knockdown plant of NbAXS1 from Nicotiana
benthamiana lacking the AXS2 homolog resulted in severe defects in body development, abnormal
3

ACCEPTED MANUSCRIPT
structure of the cell walls, and reduction of apiose, suggesting that an aberrant structure of RG-II
causes severe phenotypical changes [20].
Apigenin 7-O-β-D-glucoside apiosyltransferase (EC 2.4.2.25) uses UDP-Api as a donor
substrate to synthesize apiin [21]. Polygalacturonic acid apiosyltransferase, which is involved in the
synthesis of RG-II or apiogalacturonan, is thought to use UDP-Api as a donor substrate, although
their activities have not been detected so far. Biochemical characterization or gene identification of
these apiosyltransferases has not progressed because UDP-Api is not commercially available.
UDP-Api is quite unstable in solution and easily degraded to apiofuranosyl-1,2-cyclic phosphate and
UMP [19, 22] by intramolecular nucleophilic substitution of phosphate with the hydroxyl group at
the 2-position of apiose (Fig. 1A). The half-life of UDP-Api is 97.2 min at 25 °C and pH 8.0 [23],
which is the optimum pH for the AXS reaction. Thus, UDP-Api synthesized enzymatically began to
degrade spontaneously. At acidic pH, UDP-Api is degraded to α- -apiose and UDP with a half-life
D
of 4.67 min (pH 3.0 and 40 °C) [23]. Previous studies showed that apiosyltransferase activity [14,21]
depends on UDP-Api instantaneously formed by AXS. The detection of such UDP-Api highly
depends on real-time nuclear magnetic resonance (NMR) spectroscopy [24]. The NMR spectra
observed for the AXS reaction mixture were derived from resulting compounds such as UMP,
α- -apio-furanosyl-1,2-cyclic phosphate, UDP-Xyl, and undegraded UDP-Api [24]. Therefore,
D
establishing a method for stabilizing UDP-Api is important for understanding the dynamics of apiose
both in vivo and in vitro.
In this study, we aimed to determine the effects of bulky counter cations on UDP-Api stability
and half-life, to utilize it as a donor substrate for apiosyltransferases. Our method will improve the
availability of apiose, which contributes not only to the characterization of apiosyltransferases and
identification of their corresponding genes, but also to clarifying the roles of apiose residues in
secondary metabolites or pectins.
4

ACCEPTED MANUSCRIPT
2. Results and Discussion
2.1 Enzymatic synthesis of UDP-Api
UDP-Api is immediately degraded to UMP and apiofuranosyl-1,2-cyclic phosphate under alkaline
conditions [23]. To prevent the degradation of UDP-Api, the effect of counter cations oriented to
UDP-Api was investigated. The counter cations were predicted to bind the phosphate group of
UDP-Api and escape from nucleophilic attack by the 2′′-OH group on -apiose to the adjacent
D
phosphate group (Fig. 1A), resulting in degradation of UDP-Api.
AtAXS1, AXS from Arabidopsis thaliana, catalyzes the generation of UDP-Api and
UDP-Xyl from UDP-GlcUA [17] (Fig. 1A). The recombinant AtAXS1 protein was reacted with
UDP-GlcUA in 100 mM sodium phosphate buffer, pH 8.0. The reaction was stopped after incubation
with the enzyme at 25 °C for 4 h and then further incubated for 4 h at the same temperature. This
reaction mixture was analyzed by reverse phase HPLC. The reaction yielded only UDP-Xyl and
UMP, which was a degradation product of UDP-Api (Fig. 1B, upper panel). This agreed with
previous results showing that UDP-Api produced by AtAXS1 is immediately degraded in this
solution [17,19]. When the reaction was performed in 100 mM triethylamine phosphate buffer, pH
8.0, an additional peak, which appeared to be UDP-Api, was detected (Fig. 1B, lower panel). The
molecular mass of the compound in this peak was m/z 535.1, which agreed with the calculated mass
of UDP-Api (m/z 535.0) (Fig. 1C). However, its molecular mass also corresponded to that of
UDP-Xyl and thus this peak was further analyzed. UDP-Api degrades to UDP and Api under acidic
conditions [23]. Incubation of this compound in 4 M trifluoroacetic acid at 95 °C for 30 min
generated Api but not Xyl (Fig. 1D). The triethylamine salt of the compound dissolved in D2O was
analyzed by 1H and 2D 1H-1H-COSY NMR spectroscopy (Fig. 1E and Fig. S1). The chemical shift
and coupling constant values of the ribose and uracil rings agreed well with those of typical
UDP-sugars [25]. The chemical shifts of H2′′, H3′′a, H3′′b, H4′′a, and H′′4b of this compound (Table
1) were similar to those of free D-apiose [26]. COSY analysis of the compound (Fig. S1) revealed the
connectivity from H1′′ (5.61 ppm, apparent t, J1′′,2′′ 4.7, J1′′,P 5.4 Hz) to H2′′ (3.91 ppm, dd, J1′′,2′′ 4.7,
5

ACCEPTED MANUSCRIPT
J2′′,P 2.2 Hz). However, another connectivity pattern was not observed, indicating that no proton
attached to the carbon atom between C2′′ and C4′′. The remaining peaks reflect the geminal protons
of H3′′a/b (J3′′a,3′′b 11.6 Hz) and H4”a/b (J4′′a,4′′b 10.3 Hz). The chemical shift of H1′′ and the values of
J1′′,P and J1′′,2′′ show the linkage between the anomeric apiose residue and phosphate of UDP. These
assignments of UDP-Api do not contradict those of the tentatively generated UDP-Api in the
reaction mixture containing UDP-Xyl, UMP, and α-
No signals derived from the degraded compounds of UDP-Api (UMP and
α- -apio- -furanosyl-1,2-cyclic phosphate) were observed in these NMR spectra (Fig. 1E). These
D
-apio- -furanosyl-1,2-cyclic phosphate [24].
D
D
D
analyses demonstrate that this product was UDP-Api, showing that its triethylamine salt form was
relatively stable in aqueous solution.
2.2 Effect of counter cations on the stability of UDP-Api
Triethylamine in the reaction mixture for enzymatic production of UDP-Api suppressed the
degradation of UDP-Api, suggesting that a counter cation for UDP-Api may protect from a
nucleophilic attack of 2′′-OH to the phosphorus atom. Therefore, other various amines were
investigated to determine their effects on the stability of UDP-Api. Triethylamine, a counter cation of
phosphate buffer, was substituted for each amine, and UDP-Api was produced enzymatically under
the conditions shown in Fig. 1B. The undegraded UDP-Api was quantified after a 4-h incubation at
25 °C in the presence of each counter cation (Fig. 2A). When sodium ion was used as a counter
cation of phosphate buffer in the reaction mixture, only 6.2 ± 1.1% of UDP-Api remained. In
contrast, a substantial amount of UDP-Api remained in the reaction mixture when bulky tertiary
amines such as N-ethyldiisopropylamine (90.6 ± 2.1%), N,N-diethylcyclohexylamine (89.9 ± 2.9%),
and triethylamine (82.7 ± 5.7%) were used as counter cations of phosphate buffer. Non-bulky amines
such as ethylamine were less effective. Comparison of the effects among triethylamine (82.7 ± 5.7%),
diethylamine (61.2 ± 1.5%), and ethylamine (10.2 ± 1.7%) showed that the bulkiness and/or
hydrophobicity of the amines affected the stabilization of UDP-Api.
6

ACCEPTED MANUSCRIPT
Triethylamine was selected as a counter cation of UDP-Api for further analysis because of
its high efficiency for stabilizing UDP-Api and because of its convenience. The optimal
concentration of triethylamine phosphate buffer, pH 8.0, in the reaction mixture for UDP-Api
production was 100–200 mM. Triethylamine is empirically known to stabilize UDP-arabinofuranose
[27],which is comparatively unstable because it has an axial 2’-OH similar to UDP-Api does. The
preference of counter cations for stabilization of UDP-arabinofuranose would be similar to that for
stabilization of UDP-Api.
To investigate the efficiencies of triethylamine on stabilizing UDP-Api at different buffer
pHs, the time course of the remaining UDP-Api in each buffer at 25 °C was plotted (Fig. 2B).
UDP-Api was declined logarithmically in the buffer investigated. These results were used to estimate
the pseudo first-order rate constant for the degradation of UDP-Api and the half-life of UDP-Api
(Table 2). The half-life of UDP-Api in 100 mM sodium phosphate buffer, pH 8.0, was 1.0 ± 0.1 h.
This value is consistent with a previously reported value (1.6 h in 50 mM potassium phosphate
buffer, pH 8.0) [23]. The substitution of sodium ion with triethylamine extended the half-life of
UDP-Api to 22.5 ± 1.4 h. Lowering the pH to 6.0 extended the half-life to 30.5 ± 0.5 h. Simultaneous
substitution of sodium ion with triethylamine and the lowering the pH to 6.0 prolonged the half-life
of UDP-Api to 48.1 ± 2.4 h. The substitution of the sodium ion for triethylamine and changing the
pH from 8.0 to pH 6.0, increased the half-life of UDP-Api. Both effects independently contributed to
the stabilization of UDP-Api. Thus, the half-life of UDP-Api was more than 30-fold longer at pH 6.0
with triethylamine than that at 8.0 without triethylamine (Fig. 2B and Table 2). The enzymatic
synthesis of UDP-Api could be carried out at pH 6.0: however, the AXS enzyme activity at pH 6.0
was around 40% of the activity at pH 8.0 under the presence of 100 mM triethylamine phosphate
buffer. The maximum final yield of UDP-Api was observed when the enzymatic synthesis was
carried out at pH 8.0 and the product was stored at pH 6.0 in the presence of triethylamine. Therefore,
the enzymatic synthesis of UDP-Api was carried out at pH 8.0 in the presence of triethylamine in this
study.
7

ACCEPTED MANUSCRIPT
The stabilization of UDP-Api was temperature-dependent (Fig. 2C). In the sodium
phosphate buffer (pH 8.0), nearly all UDP-Api was degraded at 25 °C after a 4-h incubation. In
contrast, UDP-Api was relatively stable in the triethylamine phosphate buffer (pH 6.0) at less than
25 °C. Even when triethylamine was used as a counter cation, nearly all UDP-Api was degraded at
50 °C over 4 h of incubation. Storage at low temperatures led to the stabilization of UDP-Api and the
compound could be stored at -80 °C for at least 2 months without degradation. This is a practical
storage condition for UDP-Api.
2.3 Detection of apiosyltransferase activity
The presence of apiosylated sugar chains and glycosides in plants suggests that some
apiosyltransferases are encoded by plant genomes. The amount of UDP-Api prepared in this study
was sufficient for use as a donor substrate of apiosyltransferase. The activity of parsley Petroselinum
crispum apigenin 7-O-β- -glucoside apiosyltransferase (PcApiT), which catalyzes the formation of
D
apiin, can be detected using transiently generated UDP-Api [21]. This PcApiT activity was
investigated using the UDP-Api prepared in this study. The crude extract of parsley young leaves
[28] was used as an enzyme source. The reaction mixture containing 50 µM apigenin
7-O-β- -glucoside, 1 mM UDP-Api, and crude enzyme solution was incubated at 25 °C and pH 7.0
D
[29]. The 2-h incubation of the reaction mixture generated new compounds observed on a
revered-phase chromatogram (Fig 3A). The elution time corresponded to that of authentic apiin. The
molecular mass of this compound [M-H]- was m/z 563.7 (Fig. 3B), which agreed with its calculated
molecular mass. These results show that a bulky counter cation for UDP-Api had no fatal inhibitory
effect on apiosyltransferase activity. The Km and apparent Vmax values for apigenin
7-O-β-
-glucoside were 118 ± 6 µM and 3.28 ± 0.81 nmol min-1 mg-1 protein, respectively (Fig. 3C).
D
This Km value is equivalent to those reported previously [29]. The Km and apparent Vmax values for
UDP-Api were 920 ± 230 µM and 7.5 ± 1.1 nmol min-1 mg-1 protein, respectively (Fig. 3D).
8

ACCEPTED MANUSCRIPT
More than 20 nucleotide sugars have been identified in plant cells and are required for the
biosynthesis of various glycan compounds including secondary metabolites and plant cell wall
polysaccharides [30]. Each nucleotide sugar is utilized as a donor substrate for specific
glycosyltransferases. However, numerous sugar nucleotides are not commercially available, partially
because of the instability of some nucleotide sugars. The method for stabilizing nucleotide sugars
with counter cations presented in this study can be applied to stabilize and isolate other unstable
sugar nucleotides such as UDP-arabinofuranose [27,31] and CMP-Kdo [32], which are critical
compounds for the biosynthesis of plant cell wall polysaccharides.
Developing an assay for evaluating apiin-synthetic apiosyltransferase activity will be useful
for further biochemical characterization of this enzyme. Additionally, UDP-Api is used for
biochemical assays involving pectin rhamnogalacturonan II-synthetic apiosyltransferase. These
biochemical assays can be used to identify genes encoding apiosyltransferases based on the progress
of plant genome science.
4. Materials and Methods
4.1 Materials
cDNA (pda01652) encoding AtAXS1 was obtained from the RIKEN BioResource Center (Ibaraki,
Japan) [33]. β-NAD+ and UDP-GlcUA were purchased from Sigma (St. Louis, MO, USA) and
Yamasa (Tokyo, Japan), respectively. N,N-Dimethylcyclohexylamine, N,N-diethylcyclohexylamine,
diisopropylamine, N,N-diisopropylethylamine, trimethylamine, ethylamine, diethylamine,
triethylamine, hexamethylenediamine, bis(2-aminoethyl)amine, and isophoronediamine were
purchased from Fujifilm Wako Pure Chemical Corporation (Osaka, Japan). Apigenin
7-O-β- -glucoside and apiin were obtained from Ark Pharm (Arlington Heights, IL, USA). UDP-Xyl
D
was synthesized as described previously [25].
4.2 Enzymatic synthesis of UDP-Api
9

ACCEPTED MANUSCRIPT
The recombinant AtAXS1 protein was prepared as described previously [17]. The enzymatic
synthesis of UDP-Api was carried out in a reaction mixture comprising 100 mM triethylamine
phosphate buffer, pH 8.0, 50 mM UDP-GlcUA, 0.5 mM NAD+ and 10 µg/µL of the purified protein.
The reaction mixtures were incubated at 25°C for 4 h and the reaction was terminated by adding
phenol/chloroform/isoamyl alcohol (25:24:1, v/v/v), followed by vigorous shaking. The water phase
containing the reaction products was obtained by centrifugation (22,000 ×g at 4°C for 5 min) and
subjected to analysis.
The reaction products were separated and monitored by a reversed phase (RP)-HPLC
equipped with a UV detector. RP-HPLC was performed using an Inertsil ODS-3 column (4.6 × 250
mm; GL Sciences, Tokyo, Japan) with an L-6200 intelligent pump and L-4000 UV detector (Hitachi
High-Technologies, Tokyo, Japan) at a flow rate of 1.2 mL/min. The mobile phase was composed of
100 mM N,N-dimethylcyclohexylamine phosphate buffer, pH 6.5. The eluted fractions were
monitored by measuring the absorbance at 262 nm.
4.3 Effect of counter cations on the stability of UDP-Api
UDP-Api generated by enzymatic reaction was incubated at 25°C for a further 4 h in the presence of
various amine compounds. The undegraded UDP-Api was quantified by RP-HPLC. The reported
values are the averages of three independent experiments with standard errors. To determine the
half-life of UDP-Api in the presence of counter cations, the AtAXS1 reaction was carried out and
stopped as described above. Next, the reaction mixture was incubated at 25°C. The residual
UDP-Api was quantified from the RP-HPLC peak areas of UDP-Api at each time point.
4.4 Structural analysis of UDP-Api
The UDP-Api produced enzymatically was purified and collected by RP-HPLC. The molecular mass
of UDP-Api was determined by flow injection into an LCMS-2020 mass spectrometer equipped with
a LC-20AD HPLC pump (Shimadzu, Kyoto, Japan) at a flow rate of 0.2 mL/min of 50% acetonitrile.
10

ACCEPTED MANUSCRIPT
The lyophilized UDP-Api was dissolved in D2O (D, 99.9%), freeze dried, and then dissolved in D2O
again for NMR analysis. 1H-NMR spectra (1D, 2D-COSY) were recorded with an ECA-600HR
NMR spectrometer (JEOL, Tokyo, Japan) at ambient temperature. Acid hydrolysis of UDP-Api was
conducted in 4 M trifluoroacetic acid at 90°C for 30 min. The acid hydrolysate was separated using
an UnisonUK-Amino column (4.6 × 250 mm; Imtakt, Kyoto, Japan) and detected with an
ELSD-LTII evaporative light scattering detector (Shimadzu). The column was equilibrated with 95%
acetonitrile in distilled water at a flow rate of 1.0 mL/min at 45°C. Linear gradient elution was
performed by decreasing the concentration of acetonitrile to 70% over 25 min.
4.5 Apiosyltransferase assay
The young leaves (1 g) of parsley (Apium petroselinum L.) were pulverized with a pestle on ice in 1
mL of extraction buffer containing 150 mM Tris-HCl buffer, pH 7.0, 30 mM ascorbic acid, and 0.1%
2-mercaptoethanol. The homogenate was centrifuged at 20,000 ×g for 10 min at 4°C. The enzyme in
this supernatant was concentrated using Amicon Ultra-15 centrifugal filter units (Millipore, Billerica,
MA, USA) and used as the crude enzyme solution. The apigenin 7-O-β-
D-glucoside
apiosyltransferase assay was conducted with 50 µM apigenin 7-O-β- -glucoside, 1 mM UDP-Api,
D
and 8.0 µg/µL of a crude enzyme in 100 mM Tris-HCl buffer, pH 7.0, containing 50 mM NaCl for 2
h at 23°C. The reaction products were separated by RP-HPLC using an Inertsil ODS-3 column at a
flow rate of 1.0 mL/min with an isocratic flow of 20% acetonitrile containing 0.1% trifluoroacetic
acid for initial 5 min, followed by a linear gradient from 20 to 40% acetonitrile for 20 min. The
eluates were monitored by absorbance at 330 nm. The Km and Vmax values of the enzyme were
determined by assaying with various concentrations of apigenin 7-O-β- -glucoside (50–200 µM) or
D
UDP-Api (125–1,000 µM). The mean values of triplicate experimentals with standard deviation were
plotted on a Michaelis-Menten plot, and the kinetic parameters were calculated. Kinetic data were
calculated by non-linear regression analysis using Prism version 8 (GraphPad Software, San Diego,
CA).
11

ACCEPTED MANUSCRIPT
Author contribution statement
TI designed and supervised the research. MS initiated the research and TF and RM performed most
of the experiments. YT, HK, YT, and TI analyzed the data. TF, YT, HK, YT, and TI wrote the
manuscript.
Conflicts of interest
The authors declare that they have no conflicts of interest.
Funding: This work was supported by Grants-in-Aid for Scientific Research (Grant Numbers
JP15K07016 and JP18H05495) from the Ministry of Education, Culture, Sports, Science and
Technology of Japan. The study was also supported by the Program for the Third-Phase R-GIRO
Research from the Ritsumeikan Global Innovation Research Organization, Ritsumeikan University.
12

ACCEPTED MANUSCRIPT
References
[1] M. Pičmanová, B.L. Møller, Apiose: one of nature’s witty games, Glycobiology 26 (2016) 430–
442. https://doi.org/10.1093/glycob/cww012.
[2] E. Vongerichten, Uber das Apiin und Apiose, Liebigs Ann. Chem. 318 (1901) 121–136.
https://doi.org/10.1002/jlac.19013180106.
[3] R.R. Watson, N.S. Orenstein, Chemistry and biochemistry of apiose, Adv. Carbohydr. Chem.
Biochem. 31 (1975) 135–184. https://doi.org/10.1016/S0065-2318(08)60296-6
[4] E. Beck, H. Hopf, Branched-chain sugars and sugar alcohols, Methods Plant Biochem. 2 (1990)
235–289. https://doi.org/10.1016/B978-0-12-461012-5.50013-6.
[5] H. Eckey-Kaltenbach, W. Heller, J. Sonnenbichler, I. Zetl, W. Schäfer, D. Ernst, H. Sandermann,
Oxidative stress and plant secondary metabolism: 6”-O-malonylapiin in parsley, Phytochemistry
34 (1993) 687–691. https://doi.org/10.1016/0031-9422(93)85340-W.
[6] D.A. Hart, P.K. Kindel, Isolation and partial characterization of apiogalacturonans from the cell
wall of Lemna minor, Biochem. J. 116 (1970) 569–579. https://doi.org/10.1042/bj1160569.
[7] A.G. Darvill, M. McNeil, P. Albersheim, Structure of plant cell walls: VIII. A new pectic
polysaccharide, Plant Physiol. 62 (1978) 418–422. https://doi.org/10.1104/pp.62.3.418.
[8] M. Kobayashi, T. Matoh, J. Azuma, Two chains of rhamnogalacturonan II are cross-linked by
borate-diol ester bonds in higher plant cell walls, Plant Physiol. 110 (1996) 1017–1020.
https://doi.org/10.1104/pp.110.3.1017.
[9] M.A. O’Neill, D. Warrenfeltz, K. Kates, P. Pellerin, T. Doco, A.G. Darvill, P. Albersheim,
Rhamnogalacturonan-II, a pectic polysaccharide in the walls of growing plant cell, forms a dimer
that is covalently cross-linked by a borate ester. In vitro conditions for the formation and
hydrolysis of the dimer, J. Biol. Chem. 271 (1996) 22923–22930.
https://doi.org/10.1074/jbc271.37.22923.
[10] K. Miwa, S. Wakuta, S. Takada, K. Ide, J. Takano, S. Naito, H. Omori, T. Matsunaga, T.
Fujiwara, Roles of BOR2, a boron exporter, in cross linking of rhamnogalacturonan II and root
13

ACCEPTED MANUSCRIPT
elongation under boron limitation in Arabidopsis, Plant Physiol. 163 (2013) 1699–1709.
https://doi.org/10.1104/pp.113.225995.
[11] M.A. O’Neill, S. Eberhard, P. Albersheim, A.G. Darvill, Requirement of borate cross-linking of
cell wall rhamnogalacturonan II for Arabidopsis growth, Science 294 (2001) 846–849.
https://doi.org/10.1126/science.1062319.
[12] J. Sechet, S. Htwe, B. Urbanowicz, A. Agyeman, W. Feng, T. Ishikawa, M. Colomes, K.S.
Kumar, M. Kawai-Yamada, J.R. Dinneny, M.A. O’Neill, J.C. Mortimer, Suppressing Arabidopsis
GGLT1 affects growth by reducing the L-galactose content and borate cross-linking of
rhamnogalacturonan II, Plant J. 96 (2018) 1036–1050. https://doi.org/10.1111/tpj.14088.
[13] J.A. Smith, M. Bar-Peled, Synthesis of UDP-apiose in bacteria: The marine phototroph
Geminicoccus roseus and the plant pathogen Xanthomonas pisi, PLoS One 12 (2017) e0184953.
https://doi.org/10.1371/journal.pone.0184953.
[14] J.A. Smith, M. Bar-Peled, Identification of an apiosyltransferase in the plant pathogen
Xanthomonas pisi, PLoS One 13 (2018) e0206187. https://doi.org/10.1371/journal.pone.0206187.
[15] H. Sandermann, H. Grisebach, Occurrence of UDP-apiose in parsley (Apium petroselium L.),
Biochim. Biophys. Acta 156 (1968) 435–436. https://doi.org/10.1016/0304-4165(68)90282-1.
[16] H. Sandermann, G.T. Tisue, H. Grisebach, Biosynthesis of
D-apiose: IV. Formation of
UDP-apiose from UDP- -glucuronic acid in cell-free extracts of parsley (Apium petroselinum L.)
D
and Lemna minor, Biochim. Biophys. Acta 165 (1968) 550–552.
https://doi.org/10.1016/0304-4165(68)90239-0.
[17] M. Mølhøj, R. Verma, W.D. Reiter, The biosynthesis of the branched-chain sugar D-apiose in
plants: functional cloning and characterization of a UDP-D-apiose/UDP-D-xylose synthase from
Arabidopsis, Plant J. 35 (2003) 693–703. https://doi.org/10.1046/j.1365-313X.2003.01841.x.
[18] C. Gebb, D. Baron, H. Grisebach, Spectroscopic evidence for the formation of a 4-keto
intermediate in the UDP-apiose/UDP-xylose synthase reaction, Eur. J. Biochem. 54 (1975) 493–
498. https://doi.org/10.1111/j.1432-1033.1975.tb04161.x.
14

ACCEPTED MANUSCRIPT
[19] T. Eixelsberger, D. Horvat, A. Gutmann, H. Weber, B. Nidetzky, Isotope probing of the
UDP-apiose/UDP-xylose synthase reaction: evidence of a mechanism via a coupled oxidation and
aldol cleavage, Angew. Chem. Int. Ed. 56 (2017) 2503–2507.
https://doi.org/10.1002/anie.201609288.
[20] J.W. Ahn, R. Verma, M. Kim, J.Y. Lee, Y.K. Kim, J.W. Bang, W.D. Reiter, H.S. Pai, Depletion
of UDP-D-apiose/UDP-D-xylose synthases results in rhamnogalacturonan-II deficiency, cell wall
thickening, and cell death in higher plants, J. Biol. Chem. 281 (2006) 13708–13716.
https://doi.org/10.1074/jbc.M512403200.
[21] R. Ortmann, H. Sandermann, H. Grisebach, Transfer of apiose from UDP-apiose to
7-O-(β-D-glucosyl)-apigenin and 7-O-(β-D-glucosyl)-chrysoeriol with an enzyme preparation
from parsley, FEBS Lett. 7 (1970) 164–166. https://doi.org/10.1016/0014-5793(70)80146-6.
[22] D.L. Gustine, P.K. Kindel, Biosynthesis of
J. Biol. Chem. 244 (1969) 1382–1385.
D-apiose in a cell-free system from Lemna minor L.,
[23] P.K. Kindel, R.R. Watson, Synthesis, characterization and properties of uridine
5’-(α- -apio- -furanosyl pyrophosphate), Biochem. J. 133 (1973) 227–241.
D
D
https://doi.org/10.1042/bj1330227.
[24] P. Guyett, J. Glushka, X. Gu, M. Bar-Peled, Real-time NMR monitoring of intermediates and
labile products of the bifunctional enzyme UDP-apiose/UDP-xylose synthase, Carbohydr. Res.
344 (2009) 1072–1078. https://doi.org/10.1016/j.carres.2009.03.026.
[25] T. Ishimizu, T. Uchida, K. Sano, S. Hase, Chemical synthesis of uridine
5’-diphospho-α-xylophyranose, Tetrahedron Asymm. 16 (2005) 309–311.
https://doi.org/10.1016/j.tetasy.2004.11.073.
[26] J.R. Snyder, A.S. Serianni, DL-Apiose substituted with stable isotopes: synthesis,
N.M.R.-spectral analysis, and furanose anomerization, Carbohydr. Res. 166 (1987) 85–99.
https://doi.org/10.1016/0008-6215(87)80046-0.
15

ACCEPTED MANUSCRIPT
[27] Q. Zhang, N.J. Shirley, R.A. Burton, J. Lahnstein, M. Hrmova, G.B. Fincher, The genetics,
transcriptional profiles, and catalytic properties of UDP-α-D-xylose 4-epimerases from barley,
Plant Physiol. 153 (2010), 555–568. https://doi.org/10.1104/pp.110.157644
[28] K. Hahlbrock, A. Sutter, E. Wellmann, R. Ortmann, H. Grisebach, Relationship between organ
development and activity of enzymes involved in flavone glycoside biosynthesis in young parsley
plants, Phytochemistry 10 (1971) 109–116. https://doi.org/10.1016/ S0031-9422(00)90257-7.
[29] R. Ortmann, A. Sutter, H. Grisebach, Purification and properties of UDPapiose:
7-O-(β-D-glucosyl)-flavone apiosyltransferase from cell suspension cultures of parsley, Biochim.
Biophys. Acta 289 (1972) 293–302. https://doi.org/10.1016/0005-2744(72)90080-0.
[30] M. Bar-Peled, M.A. O’Neill, Plant nucleotide sugar formation, interconversion, and salvage by
sugar recycling, Annu. Rev. Plant Biol. 62 (2011) 172–155.
https://doi.org/10.1146/annurev-arplant-042110-103918.
[31] T. Konishi, T. Takeda, Y. Miyazaki, M. Ohnishi-Kameyama, T. Hayashi, M.A. O’Neill, T. Ishii,
A plant mutase that interconverts UDP-arabinofuranose and UDP-arabinopyranose, Glycobiology
17 (2007) 345–354. https://doi.org/10.1093/glycob/cwl081.
[32] C.H. Lin, B.W. Murray, I.R. Ollmann, C.H.Wong, Why is CMP-ketodeoxyoctonate highly
unstable? Biochemistry 36 (1997) 780–785. https://doi.org/10.1021/bi962055c.
[33] M. Seki, M. Narusaka, A. Kamiya, J. Ishida, M. Satou, T. Sakurai, M. Nakajima, A. Enju, K.
Akiyama, Y. Oono, M. Muramatsu, Y. Hayashizaki, J. Kawai, P. Carninci, M. Itoh, Y. Ishii, T.
Arakawa, K. Shibata, A. Shinagawa, K. Shinozaki, Functional annotation of a full-length
Arabidopsis cDNA collection, Science 296 (2002) 141–145.
https://doi.org/10.1126/science.1071006.
16

ACCEPTED MANUSCRIPT
Figure legends
Fig. 1. Enzymatic synthesis of UDP-Api using triethylamine as a counter cation. A) The
scheme of enzymatical preparation of UDP-Api and degradation of UDP-Api in solution. AtAXS1
catalyzes conversion from UDP-GlcUA to UDP-Xyl and UDP-Api. UDP-Api is degraded to
apiofuranosyl-1,2-cyclic phosphate and UMP. B) Analysis of the reaction products of AtAXS1 in
the presence of triethylamine. UDP-GlcUA was reacted with AtAXS1 in the absence (upper panel)
and presence (lower panel) of triethylamine as a counter cation at 25 °C for 4 h, stopped, and then
incubated at the same temperature for an additional 4 h. The products were separated by reversed
phase HPLC. The peaks were identified by comparison with the elution time (black wedges) of
authentic UMP and UDP-Xyl. UDP-Api was identified by the following analyses. C) Mass
spectrum of UDP-Api determined by ESI-IT MS in negative ion mode. The calculated molecular
mass of [M-H]- ions of UDP-pentose was 535.1. D) Monosaccharide analysis of the acid
hydrolysate of UDP-Api. The acid hydrolysates of UDP-Xyl (upper panel) and UDP-Api (lower
panel) were analyzed with an UnisonUK-Amino column and detected with an evaporative light
scattering detector. The peaks were identified by comparison with the elution time (black wedges)
of authentic Xyl and Api. E) 1H-NMR spectrum of UDP-Api. Each peak was assigned from
2D-COSY spectrum of UDP-Api. The chemical shifts of protons are shown in Table 1.
Fig. 2 Effect of counter cations on UDP-Api stability. A) Percentage of undegraded UDP-Api in
the presence of various amine compounds. The UDP-Api was incubated in 100 mM phosphate
buffer, pH 8.0, containing each amine compound as counter cation at 25°C for 4 h. The values are
presented as the average of three independent experiments with standard errors. B) Time-course of
degradation of UDP-Api in various buffers. The percentage of undegraded UDP-Api in the buffers
at 25 °C was plotted against time. The buffers used were 100 mM sodium phosphate buffer, pH 8.0
(closed diamond), 100 mM sodium phosphate buffer, pH 6.0 (closed circle),100 mM triethylamine
phosphate buffer, pH 8.0 (open diamond), and 100 mM triethylamine phosphate buffer, pH 6.0
17

ACCEPTED MANUSCRIPT
(open circle). The mean values with standard errors for three independent experiments are plotted.
The pseudo-first order rate constants for the degradation of UDP-APi and half-lives of UDP-Api in
the buffers are shown in Table 2. C) Thermal stability of UDP-Api. The percentage of undegraded
UDP-Api in the buffers at various temperature for 4 h were plotted. The buffers used were 100 mM
sodium phosphate buffer, pH 8.0 (closed diamond), 100 mM sodium phosphate buffer, pH 6.0
(closed circle),100 mM triethylamine phosphate buffer, pH 8.0 (open diamond), and 100 mM
triethylamine phosphate buffer, pH 6.0 (open circle). The mean values with standard errors for
three independent experiments are plotted.
Fig. 3 Apiin-synthetic apiosyltransferase activity. A) The activity of apigenin 7-O-β-
apiosyltransferase from parsley, Petroselinum crispum (PcApiT). The crude enzyme containing
PcApiT was reacted with UDP-Api and apigenin 7-O-β- -glucoside used as a donor and acceptor
substrate, respectively, at 23°C for 0 h (upper panel) and 2 h (lower panel). The peaks were
identified by comparison with the elution time of the authentic apigenin 7-O-β- -glucoside (white
D
-glucoside
D
D
wedge) and apiin (black wedge). B) The molecular mass of the PcApiT product determined by
ESI-IT MS in positive ion mode. The calculated molecular mass of [M-H]- ions of apiin is 563.7.
C) Michaelis-Menten kinetics of PcApiT on apigenin 7-O-β- -glucoside. D) Michaelis-Menten
D
kinetics of PcApiT on UDP-Api. The mean values of triplicate experimentals with standard
deviation were plotted. The solid lines represent the theoretical non-linear regression fits of the
data.
18

ACCEPTED MANUSCRIPT
Supplementary data
Fig. S1. The 2D 1H-1H COSY NMR analysis of UDP-Api. Relevant cross-peaks are labeled as in
Fig. 1E. Proton chemical shifts are reported in Table 1.
19

ACCEPTED MANUSCRIPT
Table 1. 1H signal assignment of UDP-Api. The value for proton chemical shifts of the UDP-Api
triethylamine salt in D2O at 25°C are shown.
Apiose
H1”
5.61
H2”
3.91
H3”a/3”b
3.50
H4”a/4”b
4.00
δ (ppm)
Ribose
Uracil
H1’
5.88
H5
H2’
4.27
H6
H3’
H4’
H5’a/5’b
4.10
δ (ppm)
δ (ppm)
4.27
4.16
5.87
7.86
Chemical shifts are in ppm relative to internal H2O signal set at 4.72 ppm.
1

ACCEPTED MANUSCRIPT
Table 2. Pseudo first-order rate constant for the degradation of UDP-Api and half-life of UDP-Api
in the various buffers.
Pseudo first-order rate
Buffer
Half-life (h)
1.0 ± 0.1
constant (h-1)
100 mM Sodium phosphate buffer, pH 8.0
0.29 ± 0.03
100 mM Triethylamine phosphate buffer, pH 8.0 0.013 ± 0.001
100 mM Sodium phosphate buffer, pH 6.0 0.0099 ± 0.0001
22.5 ± 1.4
30.5 ± 0.5
48.1 ± 2.4
100 mM Triethylamine phosphate buffer, pH 6.0 0.0063 ± 0.0003
1

ACCEPTED MANUSCRIPT
(A)
O
N
O
NH
OH
O
O
O
P
O
CH2
O
P
O
OH
O
N
COOH
O
OH
OH
OH
O
NH
OH
O
O
UDP-Xylose OH
AXS1
OH
O
P
O
P
O
CH2
O
OH
O
OH
OH
OH
O
N
O
OH OH
NH
O
CH2OH
O
O
CH2OH
UDP-Glucuronic acid
O
O
O
P
O
P
O
CH2
O
OH OH
PO2H
OH
OH
OH
O
UMP
Apiofuranosyl-
OH OH
1,2-cyclic phosphate
UDP-Apiose
(B)
(C)
UDP-Xyl UDP-Api
UMP
[M-H]−
535.1
0
20
Retention time (min)
40
200
600
400
m/z
O
O
3”
NH
5
(D)
(E)
1”
4”
CH2OH
O
O
5’
CH2
6
O
O
O
P
O
P
O
N
2”
OH
OH
OH
OH
1’
2’
OH
4’
3’
Api
Xyl
OH
H5
H1’
H2’
H3’
H3”
H4”
H6
H4’
H5’
H1”
H2”
8.0
5.0
Chemical shift (ppm)
6.0
4.0
10
20
Retention time (min)
Fig. 1

ACCEPTED MANUSCRIPT
(A)
100
50
0
e
e
e
e
e
l
e
e
o
e
e
e
m
n
ne
n
i
n
n
i
n
n
i
n
n
i
n
n
i
i
i
i
i
i
u
i
i
m
m
m
m
m
m
d
So
m
m
m
m
a
am
)
l
a
i
a
a
l
a
l
a
l
a
a
a
a
i
l
l
l
l
l
d
y
y
y
y
y
y
d
y
y
x
e
p
h
h
x
h
h
e
p
t
t
t
hy
e
n
o
e
n
o
e
e
e
r
r
Et
h
i
o
h
e
r
p
p
o
l
m
i
Di
l
y
o
o
Tr
o
noet
c
c
h
i
s
h
s
t
i
i
i
y
Tr
y
p
e
c
l
c
l
d
o
Di
l
am
y
m
y
-
y
Is
h
a
h
2
h
Et
-
t
t
(
x
e
e
s
m
He
Di
Bi
-
N
Di
N
-
,
N
,
N
N
(B)
(C)
100
100
50
10
5
50
Sodium phosphate buffer, pH 8.0
1
150
20
60
0
0
80
100
Time (hour)
50
40
Temperature (°C)
Fig. 2

ACCEPTED MANUSCRIPT
(A)
(B)
[M-H]-
563.7
20
10
0
600
800
200
400
Retention time(min)
m/z
(C)
(D)
4
2
2
1
0
0.1
0.2
0
0.5
1.0
[S] (mM)
[S] (mM)
Fig. 3

ACCEPTED MANUSCRIPT
Highlights
• UDP-apiose was stabilized in the presence of triethylamine.
• The half-life of UDP-apiose was increased to 48.1 h in solution.
• Apiin-producing apiosyltransferase activity was detected using UDP-apiose.
1