Selective profiling of steviol-catalyzing UDP-glycosyltransferases with a metabolically synthesized probe

Article abstract of DOI:10.1039/d0cc04948d

Selective profiling of steviol-catalyzing UDP-glycosyltransferases in plants was accomplished with a probe metabolically synthesized from two substrate-derived components comprising an alkynylated sugar receptor (steviol) module and a diazirine-modified sugar donor (UDP-glucose) module, thereby illustrating a facile approach for harnessing biosynthetic enzymes of natural glycosides in plants for synthetic biology.

Full text of DOI:10.1039/d0cc04948d

View Article Online
View Journal
ChemComm
Chemical Communications
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: N. Wong, S.
Zhong, W. Li, F. Zhou, Z. Deng and Y. Zhou, Chem. Commun., 2020, DOI: 10.1039/D0CC04948D.
Volume 54
Number
January 2018
Pages 1-112
This is an Accepted Manuscript, which has been through the
1
4
Royal Society of Chemistry peer review process and has been
accepted for publication.
ChemComm
Chemical Communications
rsc.li/chemcomm
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 1359-7345
COMMUNICATION
S. J. Connon, M. O. Senge et al.
Conformational control of nonplanar free base porphyrins:
towards bifunctional catalysts of tunable basicity
rsc.li/chemcomm

Page 1 of 5
Please d oC hn eo mt Ca do mj u ms t margins
View Article Online
DOI: 10.1039/D0CC04948D
COMMUNICATION
Selective profiling of steviol-catalyzing UDP-glycosyltransferases
with a metabolically synthesized probe†
‡
ac
‡a
d
ab
ba
Nai-Kei Wong, Suyun Zhong, Weichao Li, Fugui Zhou, Zhangshuang Deng,* and Yiqing
Zhou*
Received 00th January 20xx,
Accepted 00th January 20xx
ad
DOI: 10.1039/x0xx00000x
Selective profiling of steviol-catalyzing UDP-glycosyltransferases in
plants was accomplished with a probe metabolically synthesized
from two substrate-derived components comprising an alkynylated
sugar receptor (steviol) module and a diazirine-modified sugar
donor (UDP-glucose) module, thereby illustrating a facile approach
to harnessing biosynthetic enzymes of natural glycosides in plants
for synthetic biology.
Natural products are ubiquitous secondary metabolites found
in animals, plants, and microbes. Their extraordinary structural
diversity and scope of bioactivities have inspired numerous
discoveries of lead compounds and therapeutics in the
1
development pipeline. Synthetic biology has rapidly emerged
as a promising approach to producing natural products on an
industrial scale, allowing high precision and control over quality
without drawbacks inherent in natural extraction or total Fig. 1 (A) Structures of steviol and ST-Dayne, a previously reported steviol-derived fully
2
functionalized photoaffinity probe used in profiling of the steviol glycosides biosynthesis
pathway; (B) Structures of bi-substrate probe modules ST-N-yne and UDPG-DA used in
current study; (C) A bi-substrate probe strategy for identifying biological parts with UGT
activities for steviol.
synthesis. In the case of plant natural products, functional
elucidation of biosynthetic genes typically represents a first and
rate-limiting step toward practical utilization of biosynthetic
3
pathways. Nevertheless, technical complexities vary in mining
biosynthetic pathways across different classes of organisms. In
prokaryotes such as bacteria, due to the presence of gene
clusters of secondary metabolites, conventional “genome
mining” strategies can be readily applied to discover and
4
annotate biosynthetic genes. In contrast, in eukaryotes such as
plants, biosynthetic genes are not as compactly clustered as in
prokaryotes, making it extra time-consuming to identify the
functional enzymes in plants by classical genomic or
a. School of Biotechnology and Food Engineering, Changshu Institute of Technology,
Changshu 215500, China. *E-mail: zhyq2012@gmail.com
b.
Hubei Key Laboratory of Natural Products Research and Development, College of
Biological and Pharmaceutical Sciences, China Three Gorges University, Yichang,
4
43002, China. *E-mail: d.zhangshuang@gmail.com
Scheme 1 Synthesis of bi-substrate probe modules (A) ST-N-yne and (B) UDPG-DA.
c.
Department of Pharmacology, Shantou University Medical College, Shantou
15041, China.
5
transcriptomic approaches contingent on biochemically
d.
CAS Key Laboratory of Synthetic Biology, Institute of Plant Physiology and Ecology,
Chinese Academy of Sciences, Shanghai 200032, China.
5
focused characterization. Thus in any plants, the task of
†
Electronic
Supplementary
Information
(ESI)
available.
See
determining the biosynthetically pivotal biological parts
DOI: 10.1039/x0xx00000x
These authors contributed equally to this work.
6
remains technically daunting. Recent progress in chemical
‡
Please do not adjust margins

Please d oC hn eo mt Ca do mj u ms t margins
Page 2 of 5
COMMUNICATION
Journal Name
View Article Online
DOI: 10.1039/D0CC04948D
Fig. 2 Proof of concept with recombinant UGTs of known activities and in Stevia leaf extracts. (A) Labeling of two steviol-catalyzing UGTs, namely SrUGT85C2 and AtUGT73C1, by
combinatorial use of bi-substrate probe modules. The steviol-noncatalyzing UGTs (SrUGT76G1 and SrUGT91D2) could not be labeled (FL: fluorescence; CBB: Coomassie blue
staining); (B) Enzymatic production of the SM-yne by recombinant SrUGT85C2; (C) Labeling of SrUGT91D2 by combinatorial use of SM-yne (crude) and UDPG-DA; (D) Labeling of
Stevia leaf extracts by combinatorial use of ST-N-yne and UDPG-DA; (E) Pull-down and identification of the probe labeled band in (D) as SrUGT85C2. See Table S1 in ESI† for full list
of identified proteins.
proteomics has enabled an robust analytical modality for essential properties of a fully functionalized probe, so that their
identifying binding proteins of bioactive small molecules binding proteins would not be captured in chemical proteomic
7
-9
including natural products. Typically, such binding proteins experiments. However, when catalyzed by a specific UGT, the
are classifiable into three categories with respect to the natural diazirine-modified sugar moiety in UDPG-DA is predictably
products, namely: their functional receptors, transporters, and transferred to the receptor module ST-N-yne, generating a
biosynthetic enzymes, which may be generalized as biological more specific product (i.e. steviolmonoside)-derived probe for
parts.10 In the past decades, biosynthetic pathways or specific further affinity-based labeling, capturing, and identification of
biosynthetic enzymes of multiple classes of natural products corresponding UGTs (Fig. 1C).
including sterols, catechins, and nonribosomal peptides have In previous studies undertaken by our group, we have shown
been successfully profiled by utilizing the affinity-based protein that plant glycosyltransferases can tolerate modifications on
profiling approach.1
0-13
More recently, Morus alba Diels- the C-13 carboxyl group of steviol. Therefore, the sugar
15
Alderase (MaDA), a novel FAD-dependent [4+2] cycloaddition receptor module ST-N-yne was synthesized by attaching an
enzyme involved in chalcomoracin biosynthesis, has been alkyne reporter group to this position (Scheme 1A). Meanwhile,
discovered with full mechanistic details by using a target a combinatorial synthetic route was designed to prepare a
identification strategy based on
biosynthetic intermediate probe (BIP).
a
fully functionalized diazirine-containing sugar donor module, UDPG-DA by using
1
4
two reported enzymes for the synthesis of acetylated UDP-
As one of the most critical modification steps during glucosamine (UDPGNAc).2 Specifically, diazirine-modified
biosynthesis of diverse natural products, glycosylation glucosamine (Glc-DA) was chemically synthesized by coupling a
enhances their solubility and stability, while boosting their carboxylated diazirine to glucosamine, and phosphorylated by
0
1
5
storage and accumulation in cells. In plants, glycosylation is N-acetylhexosamine 1-kinase (NahK) to generate Glc-DA-1-P,
catalyzed by a class of enzymes called UDP-glycosyltransferases which was subsequently pyrophosphorylated by N-
(
UGTs), which utilize UDP-glucose (UDPG) as donors and acetylglucosamine uridyltransferase (GlmU) to give UDPG-DA
1
6
transfer the sugar moiety of UDPG to various acceptors. We (Scheme 1B). With both modules in hand, we then verified by
have previously established a robust chemoproteomics-based LC-MS whether a fully function probe can be biosynthesized in
platform for rapid identification of steviol-catalyzing UGTs in vitro. ST-N-yne (200 M) and UDPG-DA (1 mM) were
plants by using a substrate-derived photoaffnity probe, ST- sequentially added to recombinant Stevia rebaudiana UGT85C2
Dayne (Fig. 1A).1
7,18
Recently, Wang et al. reported the (Fig. S1, ESI†), a specific UGT glycosylating steviol on its C-13
2
1
2
chemoproteomic profiling of protein-lipid interactions by using hydroxyl group, followed by addition of MgCl (10 mM) to
a metabolically synthesized phospholipid probe. Inspired by initiate the reaction. As shown by LC-MS, a steviolmonoside
1
9
this seminal work, we advanced herein an approach based on (SM)-based fully functionalized probe P1 (Fig. S2, ESI†) was
“bi-substrate” probe modules to label, enrich, and identify the generated, suggesting that SrUGT85C2 can tolerate
biosynthetic UGTs with improved selectivity, sensitivity, and modification in UDPG with the diazirine moiety.
efficiency. Unlike our previously reported fully functionalized With the successful detection of generation of P1, we next
probe ST-Dayne, an alkyne-tagged sugar receptor module (ST- performed photoaffinity labeling in the same samples to test
N-yne) and a diazirine-tagged sugar donor module (UDPG-DA) whether the fully functionalized probe was able to label the
were prepared (Fig. 1B). Neither of these modules contains all corresponding glycosyltransferase, namely, SrUGT85C2.
2
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 5
Please d oC hn eo mt Ca do mj u ms t margins
Journal Name
COMMUNICATION
View Article Online
DOI: 10.1039/D0CC04948D
Fig. 3 Discovery of AtUGT73C5 as a novel steviol-catalyzing UGT in plant model organism Arabidopsis thaliana. (A) Labeling of Arabidopsis seeding extracts by combinatorial use of
ST-N-yne and UDPG-DA; (B) Pull-down and identification of probe-labeled band in (A) as a combination of AtUGT73C1 and AtUGT73C5. See Table S2 in ESI† for full list of identified
proteins. (C) Labeling of recombinant AtUGT73C5 by combinatorial use of bi-substrate probe modules. (D) Competitive labeling of AtUGT73C5 by either UDPG or steviol. (E) LC-MS
analysis of the glycosylated product of steviol, steviolmonoside (SM), and 19-O-β glucopyranosol steviol (S19G) catalyzed by AtUGT73C5: (i) authentic steviol; (ii) AtUGT73C5 +
steviol; (iii) authentic SM; (iv) AtUGT73C5 + SM; (v) authentic S19G; (vi) AtUGT73C5 + S19G; or (vii) co-injection of (ii), (iv), and (vi); (F) AtUGT73C5 catalyses the production of
rubusoside from either of steviol, SM, and S19G.
Solution of the enzymatic reaction was irradiated by UV (365 concomitantly to examine the labeling profile of the native
nm) for 15 min and concentrated by ultracentrifugation, proteome extracted from fresh Stevia leaves, in which large
followed by click conjugation with TAMRA-azide and in-gel amounts of steviol glycosides were produced. ST-N-yne, in the
fluorescence scanning. As anticipated, recombinant SrUGT85C2 presence of UDPG-DA concomitantly, significantly labeled a
was able to
band of around 50 kDa upon UV irradiation (Fig. 2D and Fig. S5,
be labeled only when both modules were present, and UV ESI†). To identify the fluorescent band, photo-labeled samples
irradiation also proved indispensable for the labeling (Fig. 2A). were click conjugated to biotin-azide, and biotinylated proteins
To optimize labeling conditions, a series of labeling experiments were enriched by streptavidin sepharose. Following sufficient
were performed with various reaction durations and probe and rigorous washing, bound proteins were boiled in SDS-PAGE
modules’ concentrations. It was determined that 100 M ST-N- loading buffer for elution, resolved by SDS-PAGE, and visualized
yne and 500 M UDPG-DA for 1 h reaction time are optimal for via silver staining. Stained bands were excised and subjected to
labeling experiments (Fig. S3, ESI†). Similarly, Arabidopsis in-gel tryptic digestion, followed by LC-MS/MS analysis. Our
UGT73C1 (Fig. S1, ESI†), a reported glycosyltransferase with the results show that SrUGT85C2 was the sole glycosyltransferase
same activity as SrUGT85C2 towards steviol, was also labeled significantly enriched (Fig. 2E), which suggests much improved
(
Fig. 2A).15 On the other hand, other UGTs involved in stevioside selectivity as compared to results based on the photoaffinity
1
7
biosynthesis but with non-steviol substrates, such as probe ST-Dayne.
SrUGT91D2 and SrUGT76G1 (Fig. S1, ESI†), failed to be labeled As mentioned, in previous work we have also identified a
Fig. 2A).2 To further show the superior selectivity of our bi- heterogeneously glycosyltransferase, UGT73C1, in the leaf
2,23
(
substrate probe strategy, we prepared crude alkynylated extracts of Arabidopsis thaliana as a steviol-catalyzing enzyme
1
8
steviolmonoside (SM-yne) from ST-N-yne with SrUGT85C2 (Fig. by using ST-Dayne. Here, we employed the bi-substrate probe
B and Fig. S2, ESI†) and evaluated its labeling effects on the modules in Arabidopsis seedling proteomes to profile other
2
abovementioned UGTs. Predictably, only SrUGT91D2, the UGT candidates of steviol-catalyzing UGTs. In gel-based labeling
catalyzing the formation of steviobioside from steviolmonoside, experiments, similar to the case of Stevia proteome, fluorescent
was significantly labelled (Fig. 2C and Fig. S4, ESI†). Indeed, signals around 50 kDa were evident (Fig. 3A and Fig. S5, ESI†).
SrUGT91D2 could not be labeled by ST-Dayne and had not been Following pull-down experiment and mass spectrometry, two
enriched from Stevia proteome in previous studies (Fig. S4, glycosyltransferases including AtUGT73C1 and another
ESI†).17 Taken together, the bi-substrate probe strategy is previously unidentified glycosyltransferase AtUGT73C5, were
notably more selective and sensitive than previous strategies.
significantly enriched and identified (Fig. 3B). In planta,
Having proved the applicability of the new strategy with UGT73C5 regioselectively glycosylates the C23−OH of
recombinantly expressed proteins, we set out to profile UGTs in brassinosteroids and exhibits acceptor plasticity toward diverse
plant extracts. First, we used ST-N-yne and UDPG-DA aglycones including flavonoids, phenols, mycotoxins and
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please d oC hn eo mt Ca do mj u ms t margins
Page 4 of 5
COMMUNICATION
Journal Name
oleanane-type pentacyclic triterpenes.23 Remarkably, our
results are good evidence that AtUGT73C5 could be a potential
steviol-catalyzing enzyme within Arabidopsis thaliana. Next, we
prepared recombinant AtUGT73C5 to validate its interactions
with both probe modules and assess its enzymatic activity. As
anticipated, recombinant AtUGT73C5 was successfully labeled
in the same manner as SrUGT85C2 and AtUGT73C1 (Fig. 3C).
The labeling could be dose-dependently competed by either
UDPG (Fig. 3D) or steviol (Fig. 3E), indicating that the interaction
is specific. Finally, in vitro enzymatic assays were carried out to
further confirm the catalytic products. When ST-N-yne and
UDPG-DA coordinately served as substrates, P1 was predictably
detected (Fig. S6, ESI†). Unexpectedly, however, when
unmodified substrates (steviol and UDPG) were applied, no
steviolmonoside signals were detected despite catalytic actions
of AtUGT73C5. Instead, a more polar product with a molecular
weight corresponding to steviol plus two glucose moieties was
detected (Fig. S7, ESI†), suggesting that AtUGT73C5 might
additionally catalyze the glycosylation of steviolmonoside. As
the C19-carboxyl group represents another potential
glycosylation site, we then investigated the catalytic product of
AtUGT73C5 using either steviolmonoside or 19-O-β
glucopyranosol steviol (S19G) as substrate. Interestingly, both
substrates were consumed following catalysis, and the resultant
products registered the same retention time as that of steviol.
Products for the three AtUGT73C5-catalyzed reactions with
distinct substrates were combined and purified by preparative
LC, which turned out to be rubusoside in NMR analysis (Fig. S8,
Conflicts of interest
There are no conflicts to declare.
View Article Online
DOI: 10.1039/D0CC04948D
Notes and references
1
2
3
4
5
6
F. E. Koehn and G. T. Carter, Nat. Rev. Drug Discov., 2005, 4,
06-220.
P. Xu, N. Bhan, and M. Koffas, Curr. Opin. Biotechnol., 2013,
24, 291-299.
B. Wilkinson and J. Micklefield, Nat. Chem. Biol., 2007, 3, 379-
2
3
86.
M. H. Medema, and A. Osbourn, Nat. Prod. Rep., 2016, 33,
51-962.
9
V. De Luca, V. Salim, S. M. Atsum and F. Yu, Science, 2012, 336,
1658-1661.
F. Kellner, J. Kim, B. J. Clavijo, J. P. Hamilton, K. L. Childs, B.
Vaillancourt, J. Cepela, M. Habermann, B. Steuernagel, L.
Clissold, K. McLay, C. R. Buell and S. E. O’Connor, Plant J., 2015,
8
2, 680-692.
7
8
9
1
1
1
B. F. Cravatt, A. T. Wright and J. W. Kozarich, Annu. Rev.
Biochem., 2008, 77, 383-414.
M. H. Wright and S. A. Sieber, Nat. Prod. Rep., 2016, 33, 681-
7
08.
S. Pan, H. Zhang, C. Wang, S. C. Yao and S. Q. Yao, Nat. Prod.
Rep., 2016, 33, 612-620.
0 S. Zhuang, Q. Li, L. Cai, C. Wang and X. Lei, ACS Cent. Sci., 2017,
3, 501-509.
1 J. J. Hulce, A. B. Cognetta, M. J. Niphakis, S. E. Tully and B. F.
Cravatt, Nat. Method. 2013, 10, 259-266.
2 H. Carrie, D. T. Tran, S. Rousseau, S. Chaignepain, J.-M.
Schmitter, D. Deffieux and S. Quideau, Chem. Commun., 2014,
5
0, 9387-9390.
ESI†). Rubusoside is a natural sweetener and a solubilizing agent 13 J. L. Meier and M. D. Burkart, Chem. Soc. Rev., 2009, 38, 2012-
045.
2
with diverse biological activities, whose de novo production in
1
4 L. Gao, C. Su, X. Du, R. Wang, S. Chen, Y. Zhou, C. Liu, X, Liu, R.
Tian, L. Zhang, K. Xie, S. Chen, Q. Guo, L. Guo, Y. Hano, M.
Shimazaki, A. Minami, H. Oikawa, N. Huang, K. N. Houk, L.
Huang, J. Dai and X. Lei, Nat. Chem., 2020, 12, 620-628.
2
5
genetically engineered E. coli. required at least two UGTs. Our
evidence demonstrates that AtUGT73C5 alone can catalyze the
glycosylation of steviol at both C-13 and C-19, which represents
an alternative route for producing rubusoside.
15 D. Bowles, J. Isayenkova, E. K. Lim and B. Poppenberger, Curr.
Opin. Plant Biol., 2005, 8, 254-263.
In conclusion, we have developed a facile and efficient method
for profiling novel UDP-glycosyltransferases potentially
applicable to production of natural products via synthetic
biological approaches. Compared with previous works, the
1
6 L. Lairson, B. Henrissat, G. Davies and S. Withers, Annu. Rev.
Biochem., 2008, 77, 521-555.
7 W. Li, Y. Zhou, W. You, M. Yang, Y. Ma, M. Wang, Y. Wang, S.
Yuan and Y. Xiao, ACS Chem. Biol., 2018, 13, 1944-1949.
1
current strategy based on bi-substrate probe modules permit 18 Y. Zhou, W. Li, W. You, Z. Di, M. Wang, H. Zhou, S. Yuan, N.-K.
Wong and Y. Xiao, Chem. Commun., 2018, 54, 7179-7182.
9 D. Wang, S. Du, A. Cazenave-Gassiot, J. Ge, J.-S. Lee, M. R.
Wenk and S. Q. Yao, Angew. Chem. Int. Ed. 2017, 56, 1-6.
0 G. Zhao, W. Guan, L. Cai and P. G. Wang, Nat. Protoc., 2010,
significant improvements in selectivity and sensitivity toward
UDP-glycosyltransferases in native plant proteomes. As an
example, AtUGT73C5 was unequivocally identified for the first
time as a steviol-catalyzing enzyme in Arabidopsis thaliana. We
1
2
5
, 636-646.
anticipate that our current strategy is applicable to mining of 21 A. Richman, A. Swanson, T. Humphrey, R. Chapman, B.
McGarvey, R. Pocs and J. Brandle, Plant J., 2005, 41, 56-67.
2 J. Wang, S. Li, Z. Xiong and Y. Wang, Cell Res., 2016, 25, 258-
other multi-substrate biosynthetic enzymes in bioinformatically
challenging organisms such as plants. It should be noted that
multi-staged synthetic processes based on bi-substrate probe
modules could nonetheless reduce the catalytic efficiency of
native enzymes. Further optimization of probe structures and
module combinations is imperative to help realize the full
application potential of this promising method.
2
2
2
61.
3 B. Poppenberger, S. Fujioka, K. Soeno, G. L. George, F. E.
Vaistij, S. Hiranuma, H. Seto, S. Takatsuto, G. Adam, S. Yoshida
and D. Bowles, Proc. Natl. Acad. Sci. U.S.A., 2005, 102, 15253-
1
5258.
2
2
4 E. K. Lim, D. A. Ashford, B. Hou, R. G. Jackson, D. J. Bowles,
Biotechnol. Bioeng., 2004, 87, 623-631.
5 Y. Sun, Z. Chen, J. Li, J. Li, H. Lv, J. Yang, W. Li, D. Xie, Z. Xiong,
P. Zhang and Y. Wang, Mol. Plant, 2018, 11, 1308-1311.
This work was supported by the Natural Science Foundation of
Jiangsu Province of China (BK20191477), the Natural Science
Foundation of the Higher Education Institutions of Jiangsu
Province of China (19KJB150001) and the Natural Science
Foundation of China (21502206, 21807102).
4
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 5
ChemComm
View Article Online
DOI: 10.1039/D0CC04948D
TOC
Selective profiling of steviol-catalyzing UDP-glycosyltransferases via “bi-substrate probe” strategy