High throughput discovery of heteroaromatic-modifying enzymes allows enhancement of novobiocin selectivity

Article abstract of DOI:10.1039/c1cc13552j

Glycosylated analogues of novobiocin, discovered using a broad library of enzymes, have 100-fold improved activity against breast, brain, pancreatic, lung and ovarian cancers and ablated associated off-target activity leading to an up to 2.7 × 10⁴ fold increase in selectivity.

Full text of DOI:10.1039/c1cc13552j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 10569–10571
www.rsc.org/chemcomm
COMMUNICATION
High throughput discovery of heteroaromatic-modifying enzymes allows
enhancement of novobiocin selectivityw
Sital M. Patel,a Maria de la Fuente,a Song Ke,a Andreia M. R. Guimaraes,a
Adeola O. Oliyide,a Xiaoyun Ji,b Paul Stapleton,a Anne Osbourn,c Yi Pan,b
Dianna J. Bowles,d Benjamin G. Davis,e Andreas Schatzleina and Min Yang*a
Received 15th June 2011, Accepted 29th July 2011
DOI: 10.1039/c1cc13552j
Glycosylated analogues of novobiocin, discovered using a broad
library of enzymes, have 100-fold improved activity against
breast, brain, pancreatic, lung and ovarian cancers and ablated
associated off-target activity leading to an up to 2.7 ꢀ 104 fold
increase in selectivity.
late stage intermediate with often very good efficiency and
selectivity. Previous studies have successfully demonstrated
the power of such glycorandomization/glycodiversification
either through the use of wild-type enzymes8,9 or structure-
based enzyme engineering.10 Strategies for the rapid discovery
of synthetically useful GT activities against a given target are
therefore of high value. Based on a previous MS-based High
Throughput Screening (HTS) method,11 we have now
developed and describe here a nano-LC ion trap (NLCIT)
MS method. The novel application of MSn detection can be
used to rapidly screen not only for formation efficiency but
also the product structure.
Novobiocin is a member of the aminocoumarin family that
acts as an inhibitor by binding to the GyrB subunit of DNA
gyrase (type II topoisomerase)1 and also possesses weak anti-
cancer activity by binding to the Hsp90 chaperone protein.2
The clinical use of this class of compound has been restricted
due to their low water solubility3 and toxicity in vivo.4
Modifications of aminocoumarins have allowed some
improvements in activity,5 however, to date such modifica-
tions have required redesign of the novobiocin scaffold from
scratch and cannot be performed at a late synthetic stage.
Glycosyltransferases (GTs) catalyse the transfer of sugar
moieties from active donor molecules to specific acceptor
molecules forming glycosidic bonds. GTs are important tools
in the synthesis of possible drugs.6 The exact identity and
pattern of glycosyl moieties can influence pharmacology/
pharmacokinetics, invoke biological specificity at the
molecular/tissue/organism level and even define the precise
mechanism of action.7 In nature, the attachment of sugars to
small molecules is used to mediate targeting, or modify the
mechanism of action and/or pharmacology.8 The complexity
of natural products often makes traditional chemical glycosy-
lation strategies impractical for such ‘‘glycodiversification’’.
The use of enzymes such as GTs can avoid the need for
protecting groups and allow glycosides to be formed from a
As part of a programme to generate novel anticancer
compounds we have explored strategies to enhance or rescue
the activity of representative compounds such as the model
target novobiocin. GTs are known to display activities that
modify planar, cyclic and hydrophobic molecules similar to
these targets.9,12 We reacted 107 Arabidopsis thaliana GTs and
18 Oat GTs using a-UDP-glucose as donor against 81 acceptors
(Fig. 1, SI Fig. 1w).
A total of 81 ꢀ 125 4 10 000 screening events with 125 plant
GTs allowed the identification of a number of compounds that
processed modified aromatics and heteroaromatics (See SI)
such as flavonoids, coumarins, benzoic acids, cinnamic acid
and zeatin analogues. Following detailed analysis, we were
delighted to find eleven GTs that glycosylated novobiocin
(SI Fig. 2w): 78D2 (4%), 71B8 (4%), 71B1 (15%), 88A1
(1%), 73C5 (62%), 73C6 (5%), 73C1 (8%), 73B3 (4%),
73B4 (1%), 76E1 (4%) from A. thaliana and 20n10 (3%)
from oat. NLCIT (Fig. 1) simultaneously revealed exquisite
regioselectivity from glucosylated-novobiocin for modification
of the 40-OH of the aminocoumarin ring. Glycosylation
of the precursor of novobiocin has been studied extensively.13
However, to date, no glycosylation on novobiocin has
been reported. After isolation from synthetic scale reactions
(SI Scheme 1w), all structures were unambiguously confirmed.
Many reports have shown that GTs can have broad
promiscuity with respect to both donor and acceptor.9,10 It
was also found through the screen that identified GTs could
use a range of donors (up to 9, data not shown). However,
only a-UDP-glucose could be used by GTs when novobiocin
acted as an acceptor; other donors like a-UDP-galactose,
a Department of Pharmaceutical & Biological Chemistry, The School
of Pharmacy, University of London, 29/39 Brunswick Square,
London, WC1N 1AX, UK. E-mail: min.yang@pharmacy.ac.uk;
Fax: + 44(0)2077535964; Tel: + 44(0)2077535876
b School of Chemistry and Chemical Engineering, Nanjing University,
No. 22 Hankou Road, Gulou District, Nanjing, 210093, China
c Department of Metabolic Biology, John Innes Centre, Colney Lane,
Norwich NR4 7UH, UK
d Department of Biology, University of York, York, YO105YW, UK
e Department of Chemistry, University of Oxford, CRL, Mansfield
Road, Oxford, OX1 3TA, UK
w Electronic supplementary information (ESI) available. See DOI:
10.1039/c1cc13552j
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 10569–10571 10569

View Article Online
a-UDP-GlcNAc, a-GDP-mannose, and b-GDP-fucose did not
show activity.
Unmodified novobiocin binds to the Hsp90 C-terminal
domain weakly and has poor activity.2 Although they are
not yet in a position to compete with their N-terminal binding
counterparts, e.g. 17-AAG, novobiocin based C-terminal
inhibitors have shown superior activities against various
cancer cell lines. For example, F4 demonstrated superior
activity to 17-AAG in inducing apoptosis in prostate cancer
cell lines;5a KU135 inhibits Jurkat T lymphocyte cell prolif-
eration and promotes apoptosis more potently than
17-AAG.14 Using the products of the synthetic enzyme screen,
we found that a single glycosylation at the 40-hydroxy moiety
of the coumarin ring increased significantly the anti-proliferative
activity against human ovarian, lung, breast, brain and
pancreatic cancer cell lines (Fig. 2, SI Table 1w). Glucosylation
of novobiocin (ꢁ2) increased the activity against breast cancer
by 100-fold, lung cancer by 17-fold, brain cancer by 29-fold
and pancreatic cancer by 32-fold respectively. Prompted by
this exciting result, a variant of ꢁ2, b-galactosyl novobiocin (ꢁ3)
was prepared with a subtle configurational change at just one
stereogenic centre 40 00-OH (SI Scheme 1w). The modification in
3 increased those activities to a lesser extent by 25-fold,
ꢁ
10-fold, 19-fold and 11-fold, respectively, suggesting an impor-
tant role for glycosylation in both enhancing and modulating
activity (up to 4-fold). Interestingly, both compounds only
showed a limited increase of activity against the A2780
ovarian cancer, with increases of only 1.4-fold and 1.9-fold.
The exact mechanisms of how an individual sugar leads
to these changes are yet to be understood; early protein
modeling suggested enhanced binding mediated by sugar
residues to the Hsp90 C-terminal domain peptide residues
(SI Fig. 7 and 8w). The in vitro activities of novobiocin (ꢁ1),
Fig. 1 NLCIT HTS method, glycosylation of novobiocin and struc-
ture elucidation. Top: HTS screen plate results of 73C1 indicated in
colour: Green, positive; Red, negative and Amber, intermediate.
Green spot at position 45 (novobiocin) indicates that novobiocin
can be glycosylated by 73C1. Scheme shows the glycosylation reaction
and possible MS fragments (MS1–MS4); Bottom: Structure elucida-
tion of b-glucosyl-novobiocin (ꢁ2) using ion trap MS. Peak (773)
indicates the formation of 2. The glycosylation position can be
ꢁ
determined by MSn. The MS4 (773-4730-4556-4368, 394) at 368
(cycle) delineates regioselectivity and indicated that the D-glucose
residue was added onto the 40-OH of the aminocoumarin ring.
and glycosyl novobiocins (ꢁ2 & 3) were evaluated against
ꢁ
several bacterial strains that are sensitive to topoisomerase
inhibition (SI Table 2w). Glycosylation reduced this activity
up to 266-fold. Glycosylation therefore separates the anti-
cancer activity and antibacterial (as a proxy for topo-
isomerase) activities up to 2.7 ꢀ 104 fold (Fig. 2) as compared
to novobiocin.
In conclusion, the NLCIT HTS method disclosed here
rapidly elucidated substrate specificity and product structure
via MSn fragments. Eleven catalysts were discovered that
allowed modification and ‘‘reclamation’’ of novobiocin as a
model target. Previously, poor affinity to Hsp90 and higher
affinity for type II topoisomerases has prevented novobiocin
from being evaluated as a clinically useful Hsp90 inhibitor.15
A 2.7 ꢀ 104 fold separation of wanted activities from
unwanted was achieved efficiently and rapidly, suggesting a
useful alternative strategy for anticancer drug discovery based
on the use of sugars as prosthetics.
Fig. 2 Activity and selectivity of glycosylated novobiocins compared
with novobiocin. Left: anticancer activity (IC50), (a)–(e) represent
ovarian, lung, breast, brain and pancreatic cancer, respectively.
Centre: MIC result. (f), S. aureus NCTC 12981; (g), S. aureus NCTC
13373; (h), E. faecalis NCTC 12697 and (i), E. faecalis NCTC 13379 as
proxy of topoisomerase. Each MIC was determined independently on
three occasions and, where an endpoint was determined, varied no
more than two-fold between values. Right: Relative selectivity of the
anticancer activity of ꢁ2 and ꢁ3 compared to ꢁ1 (MIC with S. aureus as a
reference).
References
1 M. Gellert, M. H. O’Dea, T. Itoh and J. I. Tomizawa, Proc. Natl.
Acad. Sci. U. S. A., 1976, 73, 4474.
2 M. G. Marcu, T. W. Schulte and L. Neckers, J. Natl. Cancer Inst.,
2000, 92, 242.
3 A. Maxwell and D. M. Lawson, Curr. Top. Med. Chem., 2003,
3, 283.
4 A. Maxwell, Mol. Microbiol., 1993, 9, 681.
c
10570 Chem. Commun., 2011, 47, 10569–10571
This journal is The Royal Society of Chemistry 2011

View Article Online
5 (a) S. B. Matthews, G. A. Vielhauer, C. A. Manthe,
V. K. Chaguturu, K. Szabla, R. L. Matts, A. C. Donnelly, B. S.
J. Blagg and J. M. Holzbeierlein, Prostate, 2010, 70, 27;
(b) D. C. Hooper, J. S. Wolfson, G. L. McHugh, M. B. Winters
and M. N. Swartz, Antimicrob. Agents Chemother., 1982, 22, 662;
(c) C. Anderle, M. Stieger, M. Burrell, S. Reinelt, A. Maxwell,
M. Page and L. Heide, Antimicrob. Agents Chemother., 2008,
52, 1982; (d) J. A. Burlison, C. Avila, G. Vielhauer,
D. J. Lubbers, J. Holzbeierlein and B. S. J. Blagg, J. Org. Chem.,
2008, 73, 2130; (e) U. Galm, S. Heller, S. Shapiro, M. Page,
S. M. Li and L. Heide, Antimicrob. Agents Chemother., 2004,
48, 1307; (f) H. Zhao, A. C. Donnelly, B. R. Kusuma, G. E. L.
Brandt, D. Brown, R. A. Rajewski, G. Vielhauer, J. Holzbeierlein,
M. S. Cohen and B. S. Blagg, J. Med. Chem., 2011, 54(11), 3839;
(g) A. C. Donnelly, J. R. Mays, J. A. Burlison, J. T. Nelson,
G. Vielhauer, J. Holzbeierlein and B. S. J. Blagg, J. Org. Chem.,
2008, 73, 8901; (h) J. A. Burlison, L. Neckers, A. B. Smith,
A. Maxwell and B. S. J. Blagg, J. Am. Chem. Soc., 2006,
128, 15529.
8 X. Fu, C. Albermann, J. Jiang, J. Liao, C. Zhang and J. S.
Thorson, Nat. Biotechnol., 2003, 21, 1467.
9 M. Yang, M. R. Proctor, D. N. Bolam, J. C. Errey, R. A. Field,
H. J. Gilbert and B. G. Davis, J. Am. Chem. Soc., 2005, 127, 9336.
10 (a) R. W. Gantt, R. D. Goff, G. J. Williams and J. S. Thorson,
Angew. Chem., Int. Ed., 2008, 47, 8889; (b) J. S. Thorson, B. W. A.,
D. Hoffmeister, C. Albermann and D. B. Nikolov, ChemBioChem,
2004, 5, 16.
11 M. Yang, M. Brazier, R. Edwards and B. G. Davis, ChemBioChem,
2005, 6, 346.
12 W. Offen, C. Martinez-Fleites, M. Yang, E. Kiat-Lim, B. G. Davis,
C. A. Tarling, C. M. Ford, D. J. Bowles and G. J. Davies,
EMBO J., 2006, 25, 1396.
13 (a) C. L. F. Meyers, M. Oberthur, J. W. Anderson, D. Kahne and
C. T. Walsh, Biochemistry, 2003, 42, 4179; (b) G. J. Williams,
R. D. Goff, C. Zhang and J. S. Thorson, Chem. Biol., 2008, 15, 393;
(c) C. Albermann, A. Soriano, J. Jiang, H. Vollmer, J. B. Biggins,
W. A. Barton, J. Lesniak, D. B. Nikolov and J. S. Thorson,
Org. Lett., 2003, 5, 933.
6 A. Luzhetskyy, C. Mendez, J. A. Salas and A. Bechthold, Curr.
Top. Med. Chem., 2008, 8, 680.
14 S. N. Shelton, M. E. Shawgo, S. B. Matthews, Y. Lu,
A. C. Donnelly, K. Szabla, M. Tanol, G. A. Vielhauer,
R. A. Rajewski, R. L. Matts, B. S. Blagg and J. D. Robertson,
Mol. Pharmacol., 2009, 76, 1314.
15 (a) J. Trepel, M. Mollapour, G. Giaccone and L. Neckers, Nat.
Rev. Cancer, 2010, 10, 537; (b) A. C. Donnelly and B. S. Blagg,
Curr. Med. Chem., 2008, 15, 2702.
7 (a) A. Ahmed, N. R. Peters, M. K. Fitzgerald, J. A. J. Watson,
F. M. Hoffmann and J. S. Thorson, J. Am. Chem. Soc., 2006,
128, 14224; (b) J. M. Langenham, N. R. Peters, I. A. Guzei,
F. M. Hoffmann and J. S. Thorson, Proc. Natl. Acad. Sci. U. S. A.,
2005, 102, 12305.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 10569–10571 10571