Determining the necessity of phenyl ring π-character in warfarin

Article abstract of DOI:10.1016/j.bmcl.2019.05.039

Despite the difficulty in administering a safe dose regimen and reports of emerging resistance, warfarin (1) remains the most widely-used oral anticoagulant for the prevention and treatment of thrombosis in humans globally. Systematic substitution of the warfarin phenyl ring with either 1,3,5,7-cyclooctatetraene (COT) (2), cubane (3), cyclohexane (4) or cyclooctane (5) and subsequent evaluation against the target enzyme, vitamin K epoxide reductase (VKOR), facilitated interrogation of both steric and electronic properties of the phenyl pharmacophore. The tolerance of VKOR to further functional group modification (carboxylate 14, PTAD adduct 15) was also investigated. The results demonstrate the importance of both annulene conferred π-interactions and ring size in the activity of warfarin.

Full text of DOI:10.1016/j.bmcl.2019.05.039

Accepted Manuscript
Determining the necessity of phenyl ring π-character in warfarin
Hui Xing, Sevan D. Houston, Xuejie Chen, Da-Yun Jin, G. Paul Savage, Jian-
Ke Tie, Craig M. Williams
PII:
S0960-894X(19)30336-1
https://doi.org/10.1016/j.bmcl.2019.05.039
BMCL 26459
DOI:
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
12 February 2019
18 May 2019
18 May 2019
Please cite this article as: Xing, H., Houston, S.D., Chen, X., Jin, D-Y., Savage, G.P., Tie, J-K., Williams, C.M.,
Determining the necessity of phenyl ring π-character in warfarin, Bioorganic & Medicinal Chemistry Letters (2019),
doi: https://doi.org/10.1016/j.bmcl.2019.05.039
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.

Bioorganic & Medicinal Chemistry Letters
Determining the necessity of phenyl ring π-character in warfarin
Hui Xing,a,§ Sevan D. Houston,a,§ Xuejie Chenb, Da-Yun Jinb, G. Paul Savage,c Jian-Ke Tieb and Craig M.
Williamsa,*
aSchool of Chemistry and Molecular Biosciences, University of Queensland, Brisbane, 4072, Queensland, Australia
E-mail: c.williams3@uq.edu.au
bDepartment of Biology, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, United States of America (USA)
cCSIRO Manufacturing, Ian Wark Laboratory, Melbourne, 3168, Victoria, Australia
§These authors contributed equally as co-first authors
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
Despite the difficulty in administering a safe dose regimen and reports of emerging resistance,
warfarin (1) remains the most widely-used oral anticoagulant for the prevention and treatment of
thrombosis in humans globally. Systematic substitution of the warfarin phenyl ring with either
1,3,5,7-cyclooctatetraene (COT) (2), cubane (3), cyclohexane (4) or cyclooctane (5) and
subsequent evaluation against the target enzyme, vitamin K epoxide reductase (VKOR),
facilitated interrogation of both steric and electronic properties of the phenyl phar-macophore.
The tolerance of VKOR to further functional group modification (carboxylate 14, PTAD adduct
15) was also investigated. The results demonstrate the importance of both annulene conferred π-
interactions and ring size in the activity of warfarin.
Keywords:
warfarin
VKOR
π-interaction
cyclohexane
cyclooctane
2019 Elsevier Ltd. All rights reserved.
Warfarin (1) was discovered in the USA 30 years ago and is
the most widely-used oral anticoagulant for the prevention and
treatment of thrombosis in humans globally.1 Despite its
prevalence in society, warfarin has a number of drawbacks which
have yet to be adequately addressed. Although the introduction of
monitoring systems and improved dose management have made
it easier for practitioners to navigate its narrow therapeutic
index,2 warfarin is still ranked among the top 10 pharmaceuticals
with serious adverse drug events.3 Its continued popularity can be
attributed to problems with developments in warfarin
pharmacogenetics4 associated with its target enzyme, vitamin K
epoxide reductase (VKOR),5 as well as shortcomings of a
number of warfarin replacements.6 The controversy regarding the
usefulness of treating individual VKOR genotypes7 and reports
of emerging resistance to warfarin8 compound these unmet
medical needs and make the drug an obvious candidate for
further scrutiny by the medicinal chemistry community.
1,3,5,7-Cyclooctatetraene (COT) was postulated as a viable
(bio)motif in this regard and was subsequently evaluated against
its cubane and phenyl counterparts.14 It was in this context that
warfarin (1) first came to our attention, where it was
demonstrated that a racemic mixture of COT-warfarin (2) was a
2-fold more active inhibitor of VKOR than (S)-warfarin (the
more active enantiomer of the two), whereas (S)-cubane warfarin
(3) was 10-fold less active than racemic warfarin. Encouraged by
the increased activity of COT-warfarin relative to warfarin itself
(i.e. 3-phenyl substituent), a study to further investigate the
importance of annulene conferred π-interactions was initiated
with the corresponding saturated systems [e.g. cyclohexyl-(4)
and cyclooctyl-warfarin (5) (Figure 1)] and analogues. The
synthesis and evaluation of these analogues against VKOR is
herein disclosed.
O
O
OH
OH
O
O
O
OH
Since its seminal synthesis in 1964, cubane has enjoyed a
wealth of attention in the areas of medicinal chemistry, material
design and fundamental bonding and reactivity.9 A recent line of
enquiry in our laboratories has been into Eaton’s 1992 conjecture
that cubane can act as a phenyl ring (bio)isostere.10 Having
successfully validated the relationship between the two scaffolds
using a portfolio of known pharmaceutical and agrochemical
templates,11 an obvious question arose as to the ability of cubane
to confer suitable π-character, an electronic property important
for both pharmacodynamic12 and pharmacokinetic13 interactions.
O
O
O
O
1
2
3
O
O
OH
O
OH
Annulene
-character
removed.
O
O
5
O
4

Figure 1: Warfarin [phenyl- (1)] along with cyclooctatetraenyl- (2),
cubyl- (3), cyclohexyl- (4) and cyclooctyl-warfarin (5).
To further delineate the relationship between structure and
VKOR inhibition of the annulene pharmacophore, the ring
system of the most active analogue [i.e. COT-warfarin (2)] was
elaborated to incorporate additional polar functionality. COT-
warfarin carboxylate 14 was chosen as a suitable target, and was
reached in a synthetic sequence beginning with the commercially
available dimethyl 1,4-cubanedicarboxylate (11).18 Half-
hydrolysis19 of 11 followed by borane reduction gave cubane
alcohol 12. Application of our recent developments20 in the Ley-
Griffith oxidation21 and tandem Wittig22 reaction to 12 gave an
intermediate α,β-unsaturated ketone, which underwent L-proline
catalysed23 conjugate addition of 4-hydroxycoumarin to give
coumarin 13. Finally, rhodium(I) norbornadiene chloride dimer
(i.e. [Rh(nbd)Cl]2) mediated valence isomerisation15 followed by
saponification gave the target COT-warfarin carboxylate 14.
PTAD adduct 15, synthesised as part of our previous study,14 was
also included for testing (Scheme 2). COT-warfarin carboxylate
existed solely as the ketone in solution (1H NMR, CDCl3) and
was isolated and evaluated as a racemic mixture. PTAD adduct
15 was obtained and used as a single enantiomer.14
Warfarin (1) was purchased from Sigma-Aldrich, whilst COT-
warfarin (2) was readily accessible from cubyl-warfarin (3),14 via
valence isomerisation methodology.15 Synthesis of cyclohexyl-
warfarin (4) began with a Wittig olefination between the
commercially available cyclohexanecarbaldehyde (6) and 1-
(triphenylphosphoranylidene)-2-propanone
completed using D-proline mediated conjugate addition between
the resulting Wittig alkene and 4-hydroxycoumarin.
(7),
and
was
8
Cyclooctyl-warfarin (5) was accessed in a similar manner starting
from the known cyclooctanecarbaldehyde (9)16 (Scheme 1).
Phenyl-(1), cubyl-(3), cyclohexyl-(4) and cyclooctyl-warfarin (5)
all existed partially in the hemi-ketal form in solution (1H NMR,
CDCl3), whereas COT-warfarin (2) existed solely as the hemi-
ketal form (1H NMR, CDCl3 or DMSO:D2O 3:7). All analogues
were isolated and evaluated as racemic mixtures.
O
O
OH
a
b
O
O
MeO2C
OH
6
8
HO
O
4
O
a, b
CO2Me
c, d
CO2Me
CO2Me
O
O
11
12
O
13
O
OH
O
c
d
O
O
5
O
10
O
9
N
O
O
N
OH
OH
N
Scheme 1: Synthesis of cyclohexyl- (4) and cyclooctyl-warfarin (5); (a) 7,
DCM, 40 ℃, 66%; (b) 4-hydroxycoumarin, D-proline, DMSO, 72%; (c) 7,
toluene, 110 ℃, 53%; (d) 4-hydroxycoumarin, D-proline, DMSO, 60%.
e, f
O
O
O
O
14
CO2H
15
The half-maximal inhibition concentration (IC50) values for
VKOR of cyclohexyl-(4) and cyclooctyl-warfarin (5) were
measured using a mammalian cell-based assay17 and plotted
against previously acquired values for phenyl-(1) and COT-
warfarin (2) (Figure 2).
Scheme 2: Synthesis of additional warfarin analogue 14, and previously
disclosed PTAD adduct 15: (a) MeOH/NaOH, THF, 95%; (b) BH3•SMe2,
THF, 0 ℃ → rt 91%; (c) i) TPAP, NMO, 4Å MS, DCM; ii) 7, DCM, 40 ℃,
40% (two steps); (d) 4-hydroxycoumarin, L-proline, DMSO, 70%; (e)
[Rh(nbd)Cl]2, PhMe, 110 ℃, 80%; (f) MeOH/NaOH, THF, 60%.
1 2 0
1 0 0
8 0
6 0
4 0
2 0
0
The IC50 values for VKOR of warfarin carboxylate 14 and
PTAD adduct 15 were measured and plotted against COT-
warfarin (2) (Figure 3).
3 0 0 0
1 5 0 0
8 0
6 0
4 0
2 0
0
1
2
4
5
Figure 2: IC50 values of phenyl- (1), COT- (2), cyclohexyl- (4) and
cyclooctyl-warfarin (5) against VKOR using FIXgla-PC/HEK293 reporter
cells.
Comparing phenyl-(1) to cyclohexyl-warfarin (4), complete
saturation of the phenyl ring resulted in a 10-fold decrease in
VKOR inhibition. This is consistent with the prior observation
that removing π-character decreases VKOR inhibition [i.e. (±)-
cubyl-warfarin (3) was 15-fold less active than (±)-phenyl-
warfarin (1)].14 The same trend was observed by comparing
COT-(2) to cyclooctyl-warfarin (5), where complete saturation of
the COT ring resulted in a 13-fold decrease in VKOR inhibition
between the two analogues.
2
1 4
1 5
Figure 3: IC50 values of warfarin carboxylate 14, PTAD adduct 15 and
COT-warfarin (2) against VKOR using FIXgla-PC/HEK293 reporter cells.
Incorporation of additional polar functionality in the form of
COT-warfarin-carboxylate (14) resulted in a 430-fold decrease in
VKOR inhibition compared to COT-warfarin (2). Interestingly,
this observation is contrary to that observed with acenocoumarol,

J. Guay, M. Callejo, M. M. Miller, R. M. Lawrence, J.
Banville, J. Guy, B. D. Maxwell, E. S. Priestley, A.
Marinier, R. R. Wexler, M. Bouvier, D. A. Gordon, W. A.
Schumacher, J. Yang, Sci. Transl. Med. 2017, 9, eaaf5294.
a) S. E. Kimmel, B. French, S. E. Kasner, J. A. Johnson, J.
L. Anderson, B. F. Gage, Y. D. Rosenberg, C. S. Eby, R.
A. Madigan, R. B. McBane, S. Z. Abdel-Rahman, S. M.
Stevens, S. Yale, E. R. Mohler, M. C. Fang, V. Shah, R. B.
Horenstein, N. A. Limdi, J. A. S. Muldowney, J. Gujral, P.
Delafontaine, R. J. Desnick, T. L. Ortel, H. H. Billett, R. C.
Pendleton, N. L. Geller, J. L. Halperin, S. Z. Goldhaber,
M. D. Caldwell, R. M. Califf, J. H. Ellenberg, N. Engl. J.
Med. 2013, 369, 2283-2293; b) T. I. Verhoef, G. Ragia, A.
de Boer, R. Barallon, G. Kolovou, V. Kolovou, S.
Konstantinides, S. Le Cessie, E. Maltezos, F. J. M. van der
Meer, W. K. Redekop, M. Remkes, F. R. Rosendaal, R. M.
F. van Schie, A. Tavridou, D. Tziakas, M. Wadelius, V. G.
Manolopoulos, A. H. Maitland-van der Zee, N. Engl. J.
Med. 2013, 369, 2304-2312.
a) J.-K. Tie, D.-Y. Jin, K. Tie, D. W. Stafford, J. Thromb.
Haemost. 2013, 11, 1556-1564; b) S. Rost, A. Fregin, V.
Ivaskevicius, E. Conzelmann, K. Hörtnagel, H.-J. Pelz, K.
Lappegard, E. Seifried, I. Scharrer, E. G. D. Tuddenham,
C. R. Müller, T. M. Strom, J. Oldenburg, Nature 2004,
427, 537.
a) T. A. Reekie, C. M. Williams, L. M. Rendina, M.
Kassiou, J. Med. Chem., 2019, 62, 1078–1095; b) K. F.
Biegasiewicz, J. R. Griffiths, G. P. Savage, J. Tsanaktsidis,
R. Priefer, Chem. Rev. 2015, 115, 6719-6745; c) G. W.
Griffin, A. P. Marchand, Chem. Rev. 1989, 89, 997-1010;
d) G. M. Locke, S. S. R. Bernhard, M. O. Senge, Chem.
Eur. J. 2018, DOI: 10.1002/chem.201804225.
P. E. Eaton, Angew. Chem. Int. Ed. 1992, 31, 1421-1436.
a) B. A. Chalmers, H. Xing, S. Houston, C. Clark, S.
Ghassabian, A. Kuo, B. Cao, A. Reitsma, C.-E. P. Murray,
J. E. Stok, G. M. Boyle, C. J. Pierce, S. W. Littler, D. A.
Winkler, P. V. Bernhardt, C. Pasay, J. J. De Voss, J.
McCarthy, P. G. Parsons, G. H. Walter, M. T. Smith, H.
M. Cooper, S. K. Nilsson, J. Tsanaktsidis, G. P. Savage, C.
M. Williams, Angew. Chem. Int. Ed. 2016, 55, 3580-3585;
b) S. D. Houston, B. A. Chalmers, G. P. Savage, C. M.
Williams, Org. Biomol. Chem., 2019, 17, 1067–1070.
A. S. Mahadevi, G. N. Sastry, Chem. Rev. 2013, 113,
2100-2138.
J. E. Stok, S. Chow, E. H. Krenske, C. Farfan Soto, C.
Matyas, R. A. Poirier, C. M. Williams, J. J. De Voss,
Chem. Eur. J. 2016, 22, 4408-4412.
H. Xing, S. D. Houston, X. Chen, S. Ghassabian, T.
Fahrenhorst-Jones, A. Kuo, C.-E. P. Murray, K.-A. Conn,
K. N. Jaeschke, D.-Y. Jin, C. Pasay, P. V. Bernhardt, J. M.
Burns, J. Tsanaktsidis, G. P. Savage, G. M. Boyle, J. J. De
Voss, J. McCarthy, G. H. Walter, T. H. J. Burne, M. T.
Smith, J.-K. Tie, C. M. Williams, Chem. Eur. J., 2019, 25,
2729 – 2734.
which only differs from warfarin by the inclusion of a p-nitro
group on the phenyl ring, but is much more active than
warfarin.24 Furthermore, preventing the 4-hydroxycoumarin and
ketone moieties from undergoing enolization was not tolerated,
as demonstrated by the PTAD adduct 15, which was 13-fold less
active than COT-warfarin (2).
7.
In summary, π-character contained within the eastern portion,
whether aromatic of alkenic, seems essential to mimic the
activity of warfarin, as demonstrated by the decreased VKOR
inhibition observed for both warfarin analogues with full
annulene saturation. Activity against VKOR was enhanced by
replacing the phenyl ring for COT as previously reported, a
modification that both increased steric bulk and modulated π-
character from aromatic to non-aromatic, an improvement which
is likely also attributed to the dynamic equilibrium of COT.14
Overall, modifications of the warfarin scaffold continue to reveal
inhibition improvements and subtilties associated with VKOR,
and underpin the importance of developing phenyl ring
(bio)isosteres, or (bio)motifs, capable of conferring suitable π-
character.
8.
9.
Acknowledgements
The authors thank the University of Queensland (UQ) and the
CSIRO (Melbourne) for financial support. C.M.W. gratefully
acknowledges the Department of Agriculture and Water Resources
for financial support (CT-06), and The Australian Research Council
for funding (DP180103004) and a Future Fellowship award (grant
number FT110100851). Jian-Ke Tie gratefully acknowledges
financial support from the National Heart, Lung and Blood Institute,
National Institutes of Health, USA (HL131690).
10.
11.
Conflicts of Interest
G. P. S. and C. M. W. have formal and informal commercial
relationships with companies developing and supplying cubane
intermediates.
Supplementary Data
Supplementary data for this article, which includes
experimental procedures and copies of NMR spectra, can be
found online at XXXXXX.
12.
13.
References
1.
2.
3.
a) D. Wardrop, D. Keeling, Br. J. Haematol. 2008, 141,
757-763; b) K. P. Link, Circulation 1959, 19, 97-107.
H. Moualla, D. Garcia, Thromb. Haemost. 2011, 128, 210-
215.
a) Institute for Safe Medication Practices (2012)
Anticoagulants the Leading Reported Drug Risk in 2011.
http://www.ismp.org/QuarterWatch/pdfs/2011Q4.pdf; b) J.
L. Navarro, J. M. Cesara, M. A. Fernàndez, J. Fontcuberta,
J. C. Reverter, J. Gol-Freixa, Rev. Esp. Cardiol. 2007, 60,
1226-1232; c) K. Lund, D. Gaffney, R. Spooner, A. M.
Etherington, P. Tansey, R. C. Tait, Br. J. Haematol. 2012,
158, 256-261.
14.
15.
S. D. Houston, H. Xing, P. V. Bernhardt, T. J. Vanden
Berg, J. Tsanaktsidis, G. P. Savage, C. M. Williams, Chem.
Eur. J., 2019, 25, 2735 – 2739.
H. Kawamoto, H. Nakashima, T. Kato, S. Arai, K. Kamata,
Y. Iwasawa, Tetrahedron 2001, 57, 981-986.
J. K. Tie, D. W. Stafford, in Methods in Enzymology, Vol.
584 (Ed.: M. H. Gelb), Academic Press, 2017, pp. 349-
394.
16.
17.
4.
5.
I. Zineh, M. Pacanowski, J. Woodcock, N. Engl. J. Med.
2013, 369, 2273-2275.
a) K. J. Czogalla, A. Biswas, K. Höning, V. Hornung, K.
Liphardt, M. Watzka, J. Oldenburg, Nat. Struct. Mol. Biol.
2016, 24, 77; b) G. Shen, W. Cui, H. Zhang, F. Zhou, W.
Huang, Q. Liu, Y. Yang, S. Li, G. R. Bowman, J. E.
Sadler, M. L. Gross, W. Li, Nat. Struct. Mol. Biol. 2016,
24, 69; c) T. Li, C.-Y. Chang, D.-Y. Jin, P.-J. Lin, A.
Khvorova, D. W. Stafford, Nature 2004, 427, 541.
P. C. Wong, D. Seiffert, J. E. Bird, C. A. Watson, J. S.
Bostwick, M. Giancarli, N. Allegretto, J. Hua, D. Harden,
18.
19.
20.
M. J. Falkiner, S. W. Littler, K. J. McRae, G. P. Savage, J.
Tsanaktsidis, Org. Process Res. Dev. 2013, 17, 1503-1509.
P. E. Eaton, N. Nordari, J. Tsanaktsidis, S. P. Upadhyaya,
Synthesis 1995, 501-502.
a) P. W. Moore, C. D. G. Read, P. V. Bernhardt, C. M.
Williams, Chem. Eur. J. 2018, 24, 4556-4561; b) T. J.
Zerk, P. W. Moore, J. S. Harbort, S. Chow, L. Byrne, G. A.
Koutsantonis, J. R. Harmer, M. Martínez, C. M. Williams,
6.

P. V. Bernhardt, Chem. Sci. 2017, 8, 8435-8442; c) T. J.
Zerk, P. W. Moore, C. M. Williams, P. V. Bernhardt,
Chem. Commun. 2016, 52, 10301-10304; d) P. W. Moore,
Y. Jiao, P. M. Mirzayans, L. N. Q. Sheng, J. P. Hooker, C.
M. Williams, Eur. J. Org. Chem. 2016, 3401-3407; e) P.
W. Moore, P. M. Mirzayans, C. M. Williams, Chem. Eur.
J. 2015, 21, 3567-3571.
21.
a) M. Pagliaro, S. Campestrini, R. Ciriminna, Chem. Soc.
Rev. 2005, 34, 837-845; b) S. V. Ley, J. Norman, W. P.
Griffith, S. P. Marsden, Synthesis 1994, 639-666; c) W. P.
Griffith, Chem. Soc. Rev. 1992, 21, 179-185; d) W. P.
Griffith, S. V. Ley, Aldrichim. Acta. 1990, 23.
22.
23.
C. D. G. Read, P. W. Moore, C. M. Williams, Green
Chem. 2015, 17, 4537-4540.
D- and L-proline were used interchangably and have no
effect on the ee of the conjugate addition. See: T. E.
Kristensen, K. Vestli, F. K. Hansen, T. Hansen, Eur. J.
Org. Chem. 2009, 2009, 5185-5191.
24.
X. Chen, D.-Y. Jin, D. W. Stafford, J.-K. Tie, Blood 2018,
132, 1974-1984.

To create your abstract, type over the instructions in the
template box below.
Graphical Abstract
Fonts or abstract dimensions should not be changed or altered
Determining the necessity of phenyl ring π-
Leave this area blank for abstract info.
character in warfarin
H. Xing, S. D. Houston, X. Chen, D.-Y. Jin, G. P. Savage,
J.-K. Tie, C. M. Williams
O
O
OH
OH
O
O
O
O