The in vitro characterization of the iterative avermectin glycosyltransferase AveBI reveals reaction reversibility and sugar nucleotide flexibility

Article abstract of DOI:10.1021/ja065950k

The glycosyltransferase AveBI, which is involved in the biosynthesis of the macrolide antihelmintic avermectin (AVM), was characterized in vitro. AveBI was confirmed to catalyze two separate iterative additions of l-oleandrose, and the reversibility of AveBI-catalyzed reaction was also demonstrated. Investigation of sugar nucleotide specificity revealed 10 unique sugar nucleotide substrates which, in combination with five distinct aglycones, led to the production of 50 differentially glycosylated AVM variants. Copyright

Full text of DOI:10.1021/ja065950k

Published on Web 12/05/2006
The in Vitro Characterization of the Iterative Avermectin Glycosyltransferase
AveBI Reveals Reaction Reversibility and Sugar Nucleotide Flexibility
Changsheng Zhang, Christoph Albermann, Xun Fu, and Jon S. Thorson*
Laboratory for Biosynthetic Chemistry, Pharmaceutical Sciences DiVision, School of Pharmacy,
National CooperatiVe Drug DiscoVery Program, UniVersity of Wisconsin-Madison,
777 Highland AVenue, Madison, Wisconsin 53705
Received August 23, 2006; E-mail: jsthorson@pharmacy.wisc.edu
Scheme 1. (A) Tandem Sugar Assembly by AveBI-Catalyzed
Aglycone-Exchange Reaction; (B) Library of AVM Analogues
Constructed via AveBI-Catalyzed Glycorandomization
Avermectins (AVMs, e.g., Scheme 1, 1) are 16-membered
macrocyclic lactones produced by Streptomyces aVermitilis. The
avermectins, and the related C22-C23-reduced ivermectin (IVM,
e.g., Scheme 1, 7), target the γ-aminobutyric acid (GABA)-related
chloride ion channels unique to nematodes, insects, ticks, and
arachnids, with little or no mammalian toxicity.1 The widespread
commercial use of these remarkable anthelmintic agents began ∼25
years ago as veterinary antiparasitic agents and has more recently
expanded to clinical applications for the control of onchocerciasis,
stongyloidiasis, and lymphatic filariasis. From a biosynthetic per-
spective, the AVMs are one of only a few known natural products
postulated to derive from iterative glycosylation.2 Specifically, a
single glycosyltransferase (GT) is required for the attachment of
the AVM oleandrosyl-disaccharide (AveBI), proposed to proceed
in a stepwise, tandem manner (Scheme 1A). Evidence in support
of iterative glycosylation includes the existence of a single gly-
cosyltransferase gene (aVeBI) within the AVM gene locus,3 S. aVer-
mitilis crude extract studies suggestive of TDP-oleandrose (Scheme
1A, 4) as an immediate precursor to the AVM oleandrose moiety,4
and the production of a variety of glycosylated AVMs via in vivo
pathway engineering.5 Herein we describe, the first definitive in
vitro biochemical verification of AveBI-catalyzed tandem glyco-
sylation. Furthermore, consistent with the recent illumination of
the reversibility of natural product GT-catalyzed reactions,6 this
study reveals the AveBI-catalyzed reaction to also be reversible,
the utility of which is demonstrated by generating 50 AVM variants.
The aVeBI gene was amplified from pWHM4735 and assessed
in several expression systems. However, the functional expression
of aVeBI was only achieved in S. liVidans TK64 by the use of
expression vectors pPWW49 and pPWW50.7 The N-His6-AveBI
fusion was subsequently purified to greater than 90% purity from
this recombinant strain via affinity purification (Figure S1)8 and
used directly for these studies. Aglycons 2, 3, 5, 6, and 8 (Scheme
1B) were prepared for this study via selective acid-mediated
hydrolysis of AVM B1a (1) and IVM (7).9 Given the lack of
availability of TDP-â-L-oleandrose,10 we opted to first examine the
reversibility of the AveBI reaction using commercially available 1
and TDP based upon the recent precedent of the reversibility of
natural product glycosyltransferase-catalyzed reactions.6
3 (63%) from 1 and the subsequent transfer of oleandrose to 5, to
provide 6 (28%) and trace amounts of 7 (7%) (Scheme 1A and
Figure S2).11b Consistent with recent studies,6 this facile TDP-
dependent aglycone exchange supports the in situ intermediacy of
TDP-â-L-oleandrose (4). Cumulatively, these studies unequivocally
establish AveBI as the GT responsible for the stepwise tandem
assembly of the AVM oleandrosyl disaccharide and reveal the
AveBI-catalyzed reaction to be readily reversible and amenable to
aglycone exchange strategies.6
The AveBI sugar nucleotide specificity was subsequently probed
with 22 NDP-sugars (generated chemically or chemoenzymatically,
Figure S3, Supporting Information).12 As a representative example,13
IVM aglycone (5) with TDP-6-deoxy-R-D-glucose led to a new
product (99% conversion, Figure 1A), the LC-MS of which was
consistent with the anticipated product 5a (Scheme 1B). Substitution
of TDP-6-deoxy-R-D-glucose with UDP-6-deoxy-R-D-glucose in the
same assay gave 5a in only 10% yield, indicating a preference for
RP-HPLC analysis of an in vitro assay containing 100 µM 1, 2
mM TDP, and 12 µM AveBI revealed the formation of 3 from 1
(30%, Scheme 1A and Figure S2), while 1 remained unchanged in
control assays lacking either TDP or AveBI.11a This notable
requirement of TDP is inconsistent with a hydrolytic event to
provide free sugar and alternatively implicates a corresponding
AveBI-catalyzed production of TDP-â-L-oleandrose (4). To confirm
this hypothesis and assess whether AveBI was capable of catalyzing
an “aglycone exchange” reaction,6 a reaction containing 100 µM
1, 100 µM 5, 2 mM TDP, and 12 µM AveBI was subsequently
analyzed. Examination of this reaction revealed the production of
9
16420
J. AM. CHEM. SOC. 2006, 128, 16420-16421
10.1021/ja065950k CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
was supported in part by National Institutes of Health Grants
AI52218, CA84374, and GM70637 (to J.S.T.) and National
Cooperative Drug Discovery Group Grant U19 CA113297 from
the National Cancer Institute. J.S.T. dedicates this communication
to Professor Wayland E. Noland on the occasion of his 80th
birthday.
Supporting Information Available: Experimental procedures and
compound characterization data. This material is available free of charge
via the Internet at http://pubs.acs.org.
References
(1) For recent reviews see: (a) Geary, T. G. Trends Parasitol. 2005, 21, 530-
532. (b) Omura, S.; Crump, A. Nat. ReV. Microbiol. 2004, 2, 984-989.
(c) Dourmishev, A. L.; Dourmishev, L. A.; Schwartz, R. A. Int. J.
Dermatol. 2005, 44, 981-988. (d) Ikeda, H.; Omura, S. Chem. ReV. 1997,
97, 2591-2610. (e) Yoon, Y. J.; Kim, E. S.; Hwang, Y. S.; Choi, C. Y.
Appl. Microbiol. Biotechnol. 2004, 63, 626-634.
(2) Some other potential examples include members of the angucycline,
aureolic acid and orthosomycin families (e.g., see Luzhetskyy, A.;
Fedoryshyn, M.; Durr, C.; Taguchi, T.; Novikov, V.; Bechthold, A. Chem.
Biol. 2005, 12,725-729.).
(3) Ikeda, H.; Nonomiya, T.; Usami, M.; Ohta, T.; Omura, S. Proc. Natl.
Acad. Sci. U.S.A. 1999, 96, 9509-9514.
(4) Schulman, M. D.; Acton, S. L.; Valentino, D. L.; Arison, B. H. J. Biol.
Chem. 1990, 265, 16965-16970.
(5) a) Wohlert, S.; Lomovskaya, N.; Kulowski, K.; Fonstein, L.; Occi, J. L.;
Gewain, K. M.; MacNeil, D. J.; Hutchinson, C. R. Chem. Biol. 2001, 8,
681-700. (b) Liao, J; Lomovskaya, N.; Fonstein, L.; Wohlert, S.;
Hutchinson, C. R.; Thorson, J. S. Unpublished.
(6) (a) Minami, K.; Kakinuma, K.; Eguchi, T. Tetrahedron Lett. 2005, 46,
6187. (b) Zhang, C.; Griffith, B. R.; Fu, Q.; Albermann, C.; Fu, X.; Lee,
I.-K.; Li, L.; Thorson, J. S. Science 2006, 313, 1291.
Figure 1. RP-HPLC analysis of representative AveBI reactions. Panels
A-E depicted the formation of glycosides of 5a-5e in AveBI reactions
with 5 as an acceptor. Panels F-I represented the attachment of xylose to
aglycons 2, 3, 5, and 8 to form 2c, 3c, 5c, and 8c by AveBI, respectively.
Conversion rates for each reaction were indicated in parentheses. Assay
and HPLC conditions are available in Supporting Information.
(7) (a) Doumith, M.; Weingarten, P.; Wehmeier, U. F.; Salah-Bey, K.;
Benhamou, B.; Capdevila, C.; Michel, J. M.; Piepersberg, W.; Raynal,
M. C. Mol. Gen. Genet. 2000, 264, 477-485. (b) Zhang, C. S.; Stratmann,
A.; Block, O.; Bruckner, R.; Podeschwa, M.; Altenbach, H. J.; Wehmeier,
U. F.; Piepersberg, W. J. Biol. Chem. 2002, 277, 22853-22862.
(8) The sequence-confirmed aVeBI PCR product was inserted into vector
pPWW50 to give expression plasmid pCAM4.10, which was introduced
into Streptomyces liVidans TK64. The cells expressing N-(His)6-AveBI
were resuspended in 30 ml of buffer A (20 mM NaH2PO4, pH 7.5, 500
mM NaCl, 10 mM imidazole) supplemented with 1 mg/ml of lysozyme.
Cells were lysed by three rounds of French-press (1200 psi). The
supernatant was loaded onto a HisTrap HT column (1 ml) and the N-(His)6-
tagged AveBI eluted with a linear gradient of imidazole (10-500 mM)
in buffer A via FPLC. After desalting through PD-10 column the purified
AveBI was stored in the buffer containing 25 mM Tris-HCl (pH 8.0),
100 mM NaCl, and 10 % glycerol. This purified recombinant AveBI was
TDP-sugars. Further AveBI-IVM assays revealed that nine ad-
ditional TDP-sugar substrates were converted to their corresponding
IVM glycosides 5b-5j (Scheme 1B). In a similar fashion, the same
set sugars were transferred to aglycones 2, 3, 6, and 8, producing
glycosides 2a-2j, 3a-3j, 6a-6j, and 8a-8j (Scheme 1, Figure
1), respectively. The conversion rates for a-e glycosides ranged
from 18% to 99% while only trace production (1-10%) of f-j
glycosides was observed, with the exception of 6h (25%) and 6g
(19%). All products were confirmed by LC-MS (Supporting Infor-
mation, Tables S1 and S2), and controls lacking AveBI or sugar
nucleotide gave no reaction. Consistent with the previous in vivo
studies,5 tandem additions of D-configured sugars to aglycone 2
and 5, or trisaccharide AVM derivatives, were not observed in this
study. While this study suggests AveBI to be particularly tolerant
of C-6 and/or C-4 sugar modifications, the attachment of unnatural
sugar appendages appears to inhibit subsequent disaccharide
formation.
In summary, this study is noteworthy for a number of reasons.
First, this work provides direct biochemical evidence of the AveBI-
catalyzed tandem sugar addition within AVM biosynthesis. Second,
this study greatly extends the repertoire of known AveBI D-sugar
nucleotide substrates and provides a rapid one-pot strategy for the
generation of 50 differentially glycosylated AVMs. Third, in
contrast to the in vitro macrolide GT studies to date,14 this study
reveals AveBI does not require an “auxiliary/activator” protein for
activity. Finally, this study demonstrates the recently established
“sugar/aglycone exchange” strategies, based upon the reversibility
of GT-catalyzed reactions,6 are also applicable to macrolides.
50-80% active in the absence of exogenous Mg+2
.
(9) (a) Wei, G. H.; Du, Y. G.; Linhardt, R. J. Tetrahedron. Lett. 2004, 45,
6895-6898. (b) Wei, L.; Wei, G.; Zhang, H.; Wang, P. G.; Du, Y.
Carbohydr. Res. 2005, 340, 1583-1590.
(10) The synthesis of TDP-â-L-oleandrose has not been reported, and the
chemoenzymatic synthesis of the closest related sugar nucleotide (TDP-
â-L-olivose, which lacks the sugar 3′-OMe) required six linear steps with
an overall reported yield of 20% (Amann, S.; Drager, G.; Rupprath, C.;
Kirschning, A.; Elling, L. Carbohydr. Res. 2001, 335, 23-32.).
(11) (a) Generally, AveBI assays were performed in a total volume of 100 µL
in Tris-HCl buffer (50 mM, pH 8.0) containing 2 mM MgCl2. Reversibility
of AveBI reaction was assayed by co-incubation of 100 µM avermectin
B1a (1) or ivermectin (7) and 2 mM TDP with 12 µM AveBI at 30 °C
overnight. (b) The AveBI-catalyzed aglycone exchange reaction was
assayed by co-incubation of 100 µM 1, 100 µM 5, and 2 mM TDP with
12 µM AveBI at 30 °C overnight.
(12) The sugar nucleotides for this study were generated as previously
described. (a) Barton, W. A.; Biggins, J. B.; Jiang, J.; Thorson, J. S.;
Nikolov, D. B. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 13397-13402.
(b) Jiang, J.; Biggins, J. B.; Thorson, J. S. Angew. Chem., Int. Ed. 2001,
40, 1502-1505. (c) Barton, W. A.; Lesniak, J.; Biggins, J. B.; Jeffrey, P.
D.; Jiang, J.; Rajashankar, K. R.; Thorson, J. S.; Nikolov, D. B. Nat. Struct.
Biol. 2001, 8, 545-551. (d) Jiang J.; Biggins J. B.; Thorson J. S. J. Am.
Chem. Soc. 2000, 122, 6803-6804. (e) Albermann, C.; Soriano, A.; Jiang,
J.; Vollmer, H.; Biggins, J. B.; Barton, W. A.; Lesniak, J.; Nikolov, D.
B.; Thorson, J. S. Org. Lett. 2003, 5, 933-936. (f) Fu, X.; Albermann,
C.; Jiang, J.; Liao, J.; Zhang, C.; Thorson, J. S. Nat. Biotechnol. 2003,
21, 1467-1469. (g) Borisova, S. A.; Zhang, C.; Takahashi, H.; Zhang,
H.; Wong, A. W.; Thorson, J. S.; Liu, H. W. Angew. Chem., Int. Ed.
2006, 45, 2748-2753.
(13) A typical reaction to assess sugar nucleotide specificity contained 50 µM
algycon (1-3, 5-8.), approximately 300 µM TDP-sugar and 12 µM
AveBI incubated at 30 °C overnight.
(14) (a) Borisova, S. A.; Zhao, L.; Melancon, I. C.; Kao, C. L.; Liu, H. W. J.
Am. Chem. Soc. 2004, 126, 6534-6535. (b) Yuan, Y.; Chung, H. S.;
Leimkuhler, C.; Walsh, C. T.; Kahne, D.; Walker, S. J. Am. Chem. Soc.
2005, 127, 14128-14129.
Acknowledgment. We thank the University of Wisconsin-
Madison School of Pharmacy Analytical Facility for analytical
support. We are grateful to Dr. U. F. Wehmeier and Prof. Dr. W.
Piepersberg (Bergische University, Wuppertal, Germany) for gener-
ous gifts of vectors pUCPU21, pPWW49, pPWW50. This research
JA065950K
9
J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006 16421