Heterocyclic acyl-phosphate bioisostere-based inhibitors of Staphylococcus aureus biotin protein ligase

Article abstract of DOI:10.1016/j.bmcl.2014.08.030

Inhibitors of Staphylococcus aureus biotin protein ligase (SaBPL) are generated by replacing the acyl phosphate group of biotinyl-5′-AMP with either a 1,2,3-triazole (see 5/10a/10b) or a 1,2,4-oxadiazole (see 7) bioisostere. Importantly, the inhibitors are inactive against the human BPL. The nature of the 5-substituent in the component benzoxazolone of the optimum 1,2,3-triazole series is critical to activity, where this group binds in the ATP binding pocket of the enzyme.

Full text of DOI:10.1016/j.bmcl.2014.08.030

Bioorganic & Medicinal Chemistry Letters xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Heterocyclic acyl-phosphate bioisostere-based inhibitors
of Staphylococcus aureus biotin protein ligase
William Tieu a, , , Angie M. Jarrad b, ,§, Ashleigh S. Paparella b, Kelly A. Keeling a,
⇑
Tatiana P. Soares da Costa b,à, John C. Wallace b, Grant W. Booker b, Steven W. Polyak b, Andrew D. Abell a,
⇑
a School of Chemistry and Physics, University of Adelaide, Adelaide, South Australia 5005, Australia
b School of Molecular and Biomedical Science, University of Adelaide, Adelaide, South Australia 5005, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
Inhibitors of Staphylococcus aureus biotin protein ligase (SaBPL) are generated by replacing the acyl phos-
phate group of biotinyl-50-AMP with either a 1,2,3-triazole (see 5/10a/10b) or a 1,2,4-oxadiazole (see 7)
bioisostere. Importantly, the inhibitors are inactive against the human BPL. The nature of the 5-substitu-
ent in the component benzoxazolone of the optimum 1,2,3-triazole series is critical to activity, where this
group binds in the ATP binding pocket of the enzyme.
Received 29 May 2014
Revised 7 August 2014
Accepted 11 August 2014
Available online xxxx
Ó 2014 Elsevier Ltd. All rights reserved.
Keywords:
Antibiotics
Bioisosteres
Inhibitors
Ligases
Drug design
Biotin protein ligase (BPL) is an adenylate forming enzyme that
catalyses the reaction of biotin and ATP to form an acyl AMP inter-
mediate known as biotinyl-50-AMP (1). This intermediate is then
employed in the biotinylation and subsequent activation of acetyl
CoA carboxylase; a key metabolic enzyme that is central to mem-
brane biogenesis and, hence, the viability of all organisms.1,2 Thus,
the inhibition of BPL has been identified as a viable drug target for
pathogens resistant to existing chemotherapies.3–8 Recent efforts
in this area have focused on developing mimics of biotinyl-50-
AMP 1, where the reactive acyl phosphate group is replaced with
a stable bioisosteres,9–24 but to date, only a handful of acyl phos-
phate bioisosteres have been reported that mimic biotinyl-5’-
AMP.5–8 For example, biotinol-50-AMP 2 with its phosphodiester
bioisostere, is a potent inhibitor of BPL from Staphylococcus aureus
(SaBPL), Escherichia coli and Homo sapiens (HsBPL).5,24 Importantly,
this compound also inhibits the growth of Staphylococcus aureus
Sulfomylamide isosteres, as found in 3, have also been reported
to be active against Mycobacterium tuberculosis BPL, with no data
reported on other BPLs.6,7
We also recently reported a 1,2,3-triazole as an effective bioiso-
stere of the hydrolytically unstable acyl phosphate of 1, for exam-
ple, see 4, Figure 1.5 A 1,2,3-triazole heterocycle as in 4 offers
significant advantages over other reported acyl phosphate bioisos-
teres in that it allows for both facile synthesis by Huisgen cycload-
dition and also combinatorial in situ approaches to inhibitor
discovery and optimization.5,25 This work identified 1,2,3-triazole
5 as the most potent (Ki = 0.09 0.01 lM) and selective (>1100 fold
in Ki SaBPL vs HsBPL) inhibitor of SaBPL reported to date.5 The tri-
azole 5 inhibits the growth of S. aureus, while being devoid of cyto-
toxicity against cultured human liver cells.5 X-ray crystal
structures of SaBPL in complex with 5, in combination with muta-
genesis studies, identified a key role for active site amino acids
Arg122, Arg125 and Asp180 in selective binding to SaBPL (see
Fig. 3). X-ray crystallography also confirmed that the benzoxazo-
lone group of 1,2,3-triazole 5 binds into the ATP pocket of SaBPL
thereby functioning as a replacement of the adenine group present
in 1–4.
with
a minimal inhibitory concentration (MIC) of 8 lg/l
L.5
⇑
Corresponding authors. Tel.: +61 88 3135360 (W.T.); tel.: +61 88 313 5652; fax:
+61 88 303 4358 (A.D.A.).
E-mail addresses: william.tieu@adelaide.edu.au (W. Tieu), andrew.abell@
adelaide.edu.au (A.D. Abell).
This paper reports the synthesis of analogues of 1,2,3-triazole 5
and their inhibition against SaBPL and HsBPL. The inverted
1,2,3-triazole 6, 1,2,4-oxadiazole 7, 1,2,4-triazole 8 and 1,3,4-oxa-
diazole 9 heterocycles were compared as possible bioisosteres of
the reactive acyl phosphate group of 1 (see Fig. 2). All compounds
These authors contributed equally.
Current address: School of Biomedical Sciences, Charles Sturt University, Booroma
à
St, Wagga Wagga, New South Wales 2678, Australia.
§
Current address: Institute for Molecular Bioscience, The University of Queensland,
Brisbane 4072, Australia.
http://dx.doi.org/10.1016/j.bmcl.2014.08.030
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Tieu, W.; et al. Bioorg. Med. Chem. Lett. (2014), http://dx.doi.org/10.1016/j.bmcl.2014.08.030

2
W. Tieu et al. / Bioorg. Med. Chem. Lett. xxx (2014) xxx–xxx
N
O-
O
HN
O
NH
O
H
O
O
NH2
NH2
N
O
P
O
O
N
N
N
R
OH
S
H
HO
HO
OH
Biotinyl-5'-AMP 1
ATP
PPi
N
O-
O
NH
H
O
N
P
O
HN
O
N
N
O
S
H
O
P
NH2
O
OH
N
O
R
O
Biotinol-5'-AMP 2
N
O-
N
OH
HO
N
O
acyl-AMP
O
O
S
NH
O
H
O
NH2
N
N
H
HN
R2-XH
N
N
H
N
S
H
HO
OH
AMP
Sulfonmyl amide 3
O
N
O
NH
H
O
NH2
N
R2
N
N
R
X
HN
N
N
N
X = NH, O, S
S
H
HO
OH
1,2,3-triazole 4
Figure 1. General reaction mechanism of adenylate forming enzymes (left); acyl AMP analogues of biotin protein ligase (right).
Figure 2. Heterocyclic analogues derived from biotinyl-50-AMP 1.
compromising inhibitory activity. A range of substituents on the
benzoxazolone group of 5 were also investigated in order to begin
to explore interactions with the ATP binding pocket of SaBPL, see
compounds 10 Figure 2.
The 1,2,3-triazole 6 was prepared by alkylation of benzoxazo-
lone 11,5 followed by copper-catalyzed cycloaddition with biotin
azide 133 in the presence of copper nano powder. Heterocycles
7–9 were each prepared in two steps from a common starting
nitrile 17 as shown in Scheme 1. In particular, oxime 18, prepared
on reaction of 17 with hydroxylamine, was treated with biotin
14b26 and EDCI with subsequent dehydration under reflux to give
1,2,4-oxadiazole 7b. Conversely, conversion of nitrile 17 to imidic
ester 19, followed by microwave reaction with hydrazide 15b in
the presence of K2CO3, gave 1,2,4-triazole 8b in 32% yield. 1,3,4-
Oxadiazole 9b was also isolated from this reaction in 17% yield.
Interestingly, reaction of 14 with 19 under acidic conditions (acetic
acid) gave solely the 1,3,4-oxadiazole 9b and not the 1,2,4-triazole
8b based on analysis by analytical HPLC and mass spectrometry.
Truncated analogues 7a, 8a and 9a were prepared in the same
fashion as shown in Scheme 1. The key building blocks 15a,27
15b and 17 used in these syntheses were prepared as shown in
Scheme 1.
Figure 3. (a) X-ray crystal structure of 5 bound to SaBPL. Interactions between the
1,2,3-triazole ring and residues are highlighted with black dashes. The edge to face
pi interaction between 1,2,3-triazole ring and W127 is not shown. (b) An overlay of
X-ray crystal structure of biotinyl-50-AMP 1 and 1,2,3-triazole 5. The hydrogen
bonding interactions between amine of 1 and D211 (3.06 Å) and S128 (3.41 Å) are
highlighted with black dashes. Methyl substituent of 5 is 4.00 Å and 3.02 Å away
from D211 and S128, respectively.
The synthesis of the 1,2,3-triazoles 10a–g is summarized
in Scheme 2. The key benzoxazolone azides 22a–e were obtained
on reacting the respective 2-amino phenol (20a–e) with
share the benzoxazolone group and optimum tether linkers either
side of the bioisostere as found in 5, where a simple methylene-
based tether can replace the ribose group of 1 and 4 without
Please cite this article in press as: Tieu, W.; et al. Bioorg. Med. Chem. Lett. (2014), http://dx.doi.org/10.1016/j.bmcl.2014.08.030

W. Tieu et al. / Bioorg. Med. Chem. Lett. xxx (2014) xxx–xxx
3
Scheme 1. Reagents and conditions: (a) 5-hexynyl tosylate, K2CO3, DMF; (b) Cu nano powder, MeCN/H2O; (c) (i) SOCl2, MeOH, (ii) NH2NH2ꢀH2O,
D
; (d) KCN, DMF; (e) NH2OH,
EtOH,
D
; (f) (i) 14, EDCIꢀHCl, DIPEA, DMF, (ii) PhMe,
D; (g) AcCl, EtOH; (h) 15, K2CO3, 1:1 DMF/PhMe, 150 °C, MW, 2 h (gave 9 (18%) and 8 (32%)) or AcOH, EtOH, D, 4 h (gave 9
(21%)).
1,1-carbonyldiimidazole, followed by alkylation with dibromobu-
tane and subsequent conversion of the bromide to azide. Ben-
zoxazolone 22f was prepared by demethylation of 21a with NaI
in aqueous HBr, followed by treatment with sodium azide. Copper
catalyzed cycloaddition of biotin alkyne 23 and benzoxazolone
azides 22a–f in the presence of copper nano powder gave the cor-
responding 1,2,3-triazole 10a–f. Reduction of 1,2,3-triazole 10c
with aqueous titanium trichloride gave 10g.
The complete activity profiles of heterocycles 6–9 and the new
derivatives of 5 (i.e., 10a–g) were determined against SaBPL and
HsBPL using an in vitro biotinylation assay5,28 that measures the
enzymatic incorporation of radiolabelled biotin onto an acceptor
protein, see Tables 1 and 2. A 1,2,4-oxadiazole heterocycle as in
7b provided an effective acyl phosphate bioisostere for the gener-
results are somewhat surprising given the close similarities of
the constituent heterocycles (i.e., 5 vs 6 and 9b vs 7b). We suggest
that the relative ability of each heterocycle to hydrogen bond with
SaBPL residues Arg125, Asp180 and Lys187 (see Fig. 3a) may
account for the observed difference. These three residues undergo
significant conformational changes on inhibitor binding and are
responsible for creating the tight association between ligand and
enzyme.3,5,29 Any disruption to these key interactions, by a partic-
ular heterocycle, would be expected to significantly impact on
binding and hence potency, as in 5 and 6. In support, we know that
certain biotin-based analogues are more effective than others at
inducing the conformational change.3,29 Specifically, biotin acety-
lene 23, the synthetic precursor for 5, is a more potent inhibitor
of SaBPL than is biotin azide 13, the precursor for 6 (Ki of 0.3
lM
ation of inhibitors of SaBPL (7b had a Ki = 1.2 0.4
compound is somewhat less active than the parent 1,2,3-triazole 5
(Ki of 0.09 0.01
M)5 it does provide an important new lead for
l
M). While this
and >10
l
M respectively).3 Based on this observation alone it is
clear that the position of the nitrogen relative to the biotin hetero-
cycle is critical for activity. Whilst 1,2,3-triazole 5 and the 1,2,4-
oxadiazole 7b were active against SaBPL, both were inactive
against human BPL (HsBPL). This is an important observation for
l
further optimisation. Interestingly, 1,2,3-triazole 6, 1,3,4-triazole
8b and 1,3,4-oxadiazole 9b were all inactive against SaBPL. These
Please cite this article in press as: Tieu, W.; et al. Bioorg. Med. Chem. Lett. (2014), http://dx.doi.org/10.1016/j.bmcl.2014.08.030

4
W. Tieu et al. / Bioorg. Med. Chem. Lett. xxx (2014) xxx–xxx
Scheme 2. Reagents and conditions: (a) CDI, DCM; Br(CH2)4Br, K2CO3, DMF; (b) NaI, HBr (aq); (c) NaN3, DMF; (d) Cu nano powder, MeCN/H2O; (e) 15% TiCl3 (aq).
binding pocket of SaBPL, see Figure 3b.30 The position of the methyl
substituent on the 1,2,3-triazole heterocycle is critical, where
methyl at the alternative R2 position gives rise to an inactive com-
pound (see 10e). Interestingly the unsubstituted derivative (10d) is
also inactive.
Importantly the two new active 1,2,3-triazoles 10a and 10b
both displayed bacteriostatic activity against S. aureus ATCC
49755 in a antibacterial microdilution broth assay (see Fig. 4). At
Table 1
Inhibition assay results of 4–12 against SaBPL and HsBPLa,b,c
Heterocycle
SaBPL Ki (
lM)
HsBPL Ki (lM)
5
6
1,2,3-Triazole
0.09 0.01
NI
NI
1.2 0.4
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
1,2,3-Triazoled
1,2,4-Oxadiazole
1,2,4-Oxadiazole
1,2,4-Triazole
7a
7b
8a
8b
9a
9b
1,2,4-Triazole
1,3,4-Oxadiazole
1,3,4-Oxadiazole
8
l
g/mL inhibitor concentration, S. aureus growth was reduced to
55% for 10a and 60% for 10b relative to 58% for 5 and the strepto-
mycin control (MIC = 8 g/mL). This compares to a value of 24% for
l
a
b
c
See Figure 2 for structures and see Supplementary material for procedure.
All compounds possessed no inhibition against HsBPL at P80 M.
NI: No inhibition at concentrations P80 M.
This heterocycle possesses a different configuration than 5, see Figure 2.
biotin acetylene 23. MIC values could not be obtained for these
compounds because of limited solubility at concentrations greater
than 64 lg/ml.
l
l
d
In summary, heterocyclic-based bioisosteres of the acyl phos-
phate of biotinyl-50-AMP 1 are reported. 1,2,3-Triazole (see 5,
10a and 10b) and 1,2,4-oxadiazole (see 7b) heterocycles provide
potent and selective inhibitors of SaBPL These heterocycles are,
in general, easier to prepare than classic phosphodiester and sul-
fomylamides bioisosteres and provide improved selectivity for
SaBPL over HsBPL. A second series of analogues containing the opti-
mum 1,2,3-triazole isostere, were also prepared, to investigate
binding of the terminal benzoxazolone to the ATP binding pocket
of SaBPL. A hydrophobic substituent at R1 on the benzoxazolone
Table 2
Inhibition assay results of 5 and 10a–g against SaBPL and HsBPLa,b,c
R1
R2
SaBPL Ki (
l
M)
HsBPL Ki (lM)
5
Me
OMe
Cl
NO2
H
H
OH
NH2
H
H
H
H
H
Me
H
0.09 0.01
1.87 0.11
0.60 0.05
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
10a
10b
10c
10d
10e
10f
10g
H
a
b
c
See Figure 2 for structures and see Supplementary material for procedure.
All compounds possessed no inhibition against HsBPL at P80 M.
NI: No inhibition at concentrations P80 M.
l
l
ongoing development of new antibiotics based on the inhibition of
BPL. Finally, the length of the tether between the biotin bicycle and
the bioisostere is critical for activity. The corresponding truncated
1,2,4-oxadiazole 7a (cf. 7b) was devoid of inhibitory activity
against SaBPL.
The results for the second series of compounds revealed that for
the best base heterocycle (1,2,3-triazole), both methyl (5, Ki =
0.09 0.01
were well tolerated at R1, as was methoxy to a lesser extent
(10a, Ki = 1.87 0.11 M) (Table 2). Interestingly, neither a hydro-
lM) and chloro substituents (10b, Ki = 0.60 0.05 lM)
l
xyl nor amino group were tolerated at R1 (see 10f and 10g), despite
molecular modeling suggesting these groups could potentially
hydrogen bond with Asn212 and Ser128 residues within the ATP
Figure 4. Inhibition of S. aureus growth in vitro. Compounds
(orange), 10a (lime), 10b (green), 10c (fuchsia), 10d (purple), 10f (azure), 10g (blue)
and positive control streptomycin (red). See Supplementary data for conditions.
5 (yellow), 23
Please cite this article in press as: Tieu, W.; et al. Bioorg. Med. Chem. Lett. (2014), http://dx.doi.org/10.1016/j.bmcl.2014.08.030

W. Tieu et al. / Bioorg. Med. Chem. Lett. xxx (2014) xxx–xxx
8. Brown, P. H.; Cronan, J. E.; Grøtli, M.; Beckett, D. J. Mol. Biol. 2004, 337, 857.
5
is favored, with chloro being particularly active (Ki = 0.6 lM). Work
9. Lu, X.; Zhou, R.; Sharma, I.; Li, X.; Kumar, G.; Swaminathan, S.; Tonge, P. J.; Tan,
D. S. ChemBioChem 2012, 13, 129.
10. Somu, R. V.; Boshoff, H.; Qiao, C.; Bennett, E. M.; Barry, C. E.; Aldrich, C. C. J.
Med. Chem. 2005, 49, 31.
11. Qiao, C.; Wilson, D. J.; Bennett, E. M.; Aldrich, C. C. J. Am. Chem. Soc. 2007, 129,
6350.
12. Yu, X. Y.; Hill, J. M.; Yu, G.; Wang, W.; Kluge, A. F.; Wendler, P.; Gallant, P.
Bioorg. Med. Chem. Lett. 1999, 9, 375.
on optimization the 1,2,3-triazoles with a more expansive SAR
study and developing these isosteres to other clinically relevant
adenylate forming enzyme targets is currently underway.
Acknowledgments
13. Bernier, S.; Akochy, P.-M.; Lapointe, J.; Chênevert, R. Biooorg. Med. Chem. 2005,
13, 69.
14. Auld, D. S.; Lovell, S.; Thorne, N.; Lea, W. A.; Maloney, D. J.; Shen, M.; Rai, G.;
Battaile, K. P.; Thomas, C. J.; Simeonov, A.; Hanzlik, R. P.; Inglese, J. Proc. Natl.
Acad. Sci. U.S.A. 2010, 107, 4878.
15. Lun, S.; Guo, H.; Adamson, J.; Cisar, J. S.; Davis, T. D.; Chavadi, S. S.; Warren, J.
D.; Quadri, L. E. N.; Tan, D. S.; Bishai, W. R. Antimicrob. Agents Chemother. 2013,
57, 5138.
Thank you to Lim Jing Ting Vernise for synthesizing compounds
21e and 22e. We are grateful to the National Health and Medical
Research Council (NHMRC application APP1011806 and APP1068885)
for financial support and to Australian National Fabrication Facility
for providing access to analytical HPLC equipment.
16. Patrone, J. D.; Yao, J.; Scott, N. E.; Dotson, G. D. J. Am. Chem. Soc. 2009, 131,
16340.
Supplementary data
17. Tian, Y.; Suk, D.-H.; Cai, F.; Crich, D.; Mesecar, A. D. Biochemistry 2008, 47,
12434.
18. Ciulli, A.; Scott, D. E.; Ando, M.; Reyes, F.; Saldanha, S. A.; Tuck, K. L.; Chirgadze,
D. Y.; Blundell, T. L.; Abell, C. ChemBioChem 2008, 9, 2606.
19. Tuck, K. L.; Saldanha, S. A.; Birch, L. M.; Smith, A. G.; Abell, C. Org. Biomol. Chem.
2006, 4, 3598.
20. Ferreras, J. A.; Ryu, J.-S.; Di Lello, F.; Tan, D. S.; Quadri, L. E. N. Nat. Chem. Biol.
2005, 1, 29.
21. Schmelz, S.; Naismith, J. H. Curr. Opin. Struct. Biol. 2009, 19, 666.
22. Duckworth, B. P.; Nelson, K. M.; Aldrich, C. C. Curr. Top. Med. Chem. 2012, 12,
766.
Supplementary data (details on chemical synthesis, enzyme
inhibition assay and antimicrobial broth assay) associated with
this article can be found, in the online version, at http://
dx.doi.org/10.1016/j.bmcl.2014.08.030.
References and notes
1. Paparella, A. S.; Soares da Costa, T. P.; Yap, M. Y.; Tieu, W.; Wilce, M. C. J.;
Booker, G. W.; Abell, A. D.; Polyak, S. W. Curr. Top. Med. Chem. 2014, 14, 4.
2. Polyak, S. W.; Abell, A. D.; Wilce, M. C. J.; Zhang, L.; Booker, G. W. Appl.
Microbiol. Biotechnol. 2012, 93, 983.
3. Soares da Costa, T. P.; Tieu, W.; Yap, M. Y.; Zvarec, O.; Bell, J. M.; Turnidge, J. D.;
Wallace, J. C.; Booker, G. W.; Wilce, M. C. J.; Abell, A. D.; Polyak, S. W. ACS Med.
Chem. Lett. 2012, 3, 509.
23. Vannada, J.; Bennett, E. M.; Wilson, D. J.; Boshoff, H. I.; Barry, C. E.; Aldrich, C. C.
Org. Lett. 2006, 8, 4707.
24. Xu, Z.; Yin, W.; Martinelli, L. K.; Evans, J.; Chen, J.; Yu, Y.; Wilson, D. J.; Mizrahi,
V.; Qiao, C.; Aldrich, C. C. Bioorg. Med. Chem 2014, 22, 1726.
25. Tieu, W.; Soares da Costa, T. P.; Yap, M. Y.; Keeling, K. L.; Wilce, M. C. J.;
Wallace, J. C.; Booker, G. W.; Polyak, S. W.; Abell, A. D. Chem. Sci. 2013, 4, 3533.
26. Wilbur, D. S.; Chyan, M.-K.; Pathare, P. M.; Hamlin, D. K.; Frownfelter, M. B.;
Kegley, B. B. Bioconjugate Chem. 2000, 11, 569.
4. Payne, D. J.; Gwynn, M. N.; Holmes, D. J.; Pompliano, D. L. Nat. Rev. Drug. Disc.
2007, 6, 29.
5. Soares da Costa, T. P.; Tieu, W.; Yap, M. Y.; Pendini, N. R.; Polyak, S. W.; Sejer
Pedersen, D.; Morona, R.; Turnidge, J. D.; Wallace, J. C.; Wilce, M. C. J.; Booker, G.
W.; Abell, A. D. J. Biol. Chem. 2012, 287, 17823.
6. Duckworth, B. P.; Geders, T. W.; Tiwari, D.; Boshoff, H. I.; Sibbald, P. A.; Barry Iii,
C. E.; Schnappinger, D.; Finzel, B. C.; Aldrich, C. C. Chem. Biol. 2011, 18, 1432.
7. Shi, C.; Tiwari, D.; Wilson, D. J.; Seiler, C. L.; Schnappinger, D.; Aldrich, C. C. ACS
Med. Chem. Lett. 2013, 4, 1213.
27. Wilchek, M.; Bayer, E. A. In Meir, W., Edward, A. B., Eds.; Methods in
Enzymology; Academic Press, 1990; 184, p 123.
28. Polyak, S. W.; Chapman-Smith, A.; Brautigan, P. J.; Wallace, J. C. J. Biol. Chem.
1999, 274, 32847.
29. Wood, Z. A.; Weaver, L. H.; Brown, P. H.; Beckett, D. J. Mol. Biol. 2006, 357, 509.
30. See Supporting information.
Please cite this article in press as: Tieu, W.; et al. Bioorg. Med. Chem. Lett. (2014), http://dx.doi.org/10.1016/j.bmcl.2014.08.030