Synthesis of annulated pyridines as inhibitors of aldosterone synthase (CYP11B2)

Article abstract of DOI:10.1039/c6ob00848h

A series of cyclopenta[c]pyridine aldosterone synthase (AS) inhibitors were conveniently accessed using batch or continuous flow Kondrat'eva reactions. Preparation of the analogous cyclohexa[c]pyridines led to the identification of a potent and more selective AS inhibitor. The structure-activity-relationship (SAR) in this new series was rationalized using binding mode models in the crystal structure of AS.

Full text of DOI:10.1039/c6ob00848h

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
COMMUNICATION
Synthesis of annulated pyridines as inhibitors of
aldosterone synthase (CYP11B2)†
Rainer E. Martin,*a Johannes Lehmann,a Thibaut Alzieu,a Mario Lenz,a
Marjorie A. Carnero Corrales,a Johannes D. Aebi,a Hans Peter Märki,a Bernd Kuhn,a
Kurt Amrein,b Alexander V. Maywega and Robert Britton*c
Cite this: Org. Biomol. Chem., 2016,
14, 5922
Received 20th April 2016,
Accepted 25th May 2016
DOI: 10.1039/c6ob00848h
www.rsc.org/obc
A series of cyclopenta[c]pyridine aldosterone synthase (AS) inhibi- enzyme that is mainly present in adrenal gland zona glomeru-
tors were conveniently accessed using batch or continuous flow losa cells and to a lower extent in other tissues and is exclu-
Kondrat’eva reactions. Preparation of the analogous cyclohexa[c] sively involved in the synthesis of aldosterone.10,11 CYP11B2 is
pyridines led to the identification of a potent and more selective therefore considered to be a suitable target for the develop-
AS inhibitor. The structure-activity-relationship (SAR) in this new ment of selective aldosterone synthesis inhibitors, although
series was rationalized using binding mode models in the crystal targeting this enzyme presents certain challenges. Specifically,
structure of AS.
human CYP11B2 shows a high sequence identity (∼93%) to
cortisol synthase (CYP11B1, 11β-hydroxylase), an enzyme that
is critical for the synthesis of cortisol and mainly located in
the zona fasciculata/reticularis.12 In addition, the difference
between human and rodent CYP11B2 enzymes is high. For
instance, rat and human CYP11B2 display an overall sequence
identity of only 68%,12 thus further complicating the develop-
ment of aldosterone synthase inhibitors with comparable
potency on both species.
Aldosterone, a steroid hormone belonging to the mineralo-
corticoid family, is part of the Renin-Angiotensin-Aldosterone
System (RAAS),1 which is a pathway that has been targeted by
many important drugs, including angiotensin II receptor
blockers (ARBs), angiotensin-converting-enzyme (ACE) inhibi-
tors, renin inhibitors,2 and mineralocorticoid receptor (MR)
blockers.3 Unfortunately, while several of these drugs block
the action of aldosterone and show an impressive increase in
survival in post infarct patients, their prolonged use can also
result in massive up-regulation of aldosterone levels by com-
pensatory mechanisms4 and aldosterone escape.5 This is a par-
ticular concern considering that substantial evidence exists
suggesting a pathological role of aldosterone via non-MR-
mediated pathways.6 Consequently, the inhibition of aldo-
sterone synthesis may prove to be a superior therapeutic
approach to treating hypertension.7,8
Recently, the outcome of the clinical trial of imidazole-type
inhibitor LCI699 (1, Fig. 1), an orally active CYP11B2 inhibitor,
in patients with primary aldosteronism (PA) has been
Aldosterone is synthesized from cholesterol through
a linear cascade of reactions mainly catalyzed by enzymes of
the cytochrome P450 family.9,10 The last three steps in this
sequence are mediated by aldosterone synthase (CYP11B2), an
aMedicinal Chemistry, Roche Pharmaceutical Research and Early Development
(pRED), Roche Innovation Center Basel, F. Hoffmann-La Roche Ltd,
Grenzacherstrasse 124, CH-4070 Basel, Switzerland.
E-mail: rainer_e.martin@roche.com
bDiscovery Infectious Disease, Roche Pharma Research and Early Development
(pRED), Roche Innovation Center Basel, F. Hoffmann-La Roche Ltd,
Grenzacherstrasse 124, CH-4070 Basel, Switzerland
Fig. 1 Structures of aldosterone synthase inhibitors 1–3. In vitro aldo-
sterone synthase inhibition was tested in renal leiomyoblastoma cells,
ectopically expressing human (h), cynomolgus monkey (c) or mouse (m)
CYP11B2 or CYP11B1. The selectivity factor (SF) = IC50 (hCYP11B2)/IC50
(hCYP11B1).
cDepartment of Chemistry, Simon Fraser University, 8888 University Drive, Burnaby,
British Columbia, Canada, V5A1S6. E-mail: rbritton@sfu.ca
†Electronic supplementary information (ESI) available: Experimental details and
characterization data. See DOI: 10.1039/c6ob00848h
5922 | Org. Biomol. Chem., 2016, 14, 5922–5927
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Communication
reported.8,15 This study revealed that administration of LCI699
at doses of up to 1.0 mg b.i.d. safely inhibited CYP11B2 in
patients with PA and resulted in significantly reduced levels of
aldosterone in both plasma and urine, and consequently the
rapid correction of hypokalemia and a modest decrease in
blood pressure.9 However, due to the low selectivity of LCI699
for CYP11B2 over CYP11B1, partial inhibition of cortisol syn-
thesis resulted in dose-dependent increases in both plasma
adrenocorticotropic hormone (ACTH) and 11-deoxycortisol
(the biosynthetic precursor to cortisol).8 Very recently, Novartis
published an inhibitor with 100-fold selectivity over CYP11B1,
that reduced plasma and urinary aldosterone in healthy volun-
teers and had no concomitant effect on cortisol, 11-deoxy-
cortisol, or ACTH in humans.16 These clinical data provide
support for the inhibition of CYP11B2 as a means to lower
inappropriately high aldosterone levels and the use of aldo-
sterone synthase inhibitors as leads for the treatment of hyper-
tension and renal disorders. However, the identification of
aldosterone synthase inhibitors with high levels of selectivity
for CYP11B2 over CYP11B1 is critical in advancing this thera-
peutic approach. Towards this goal, several groups have
reported pyridine,17 imidazole,18 and isoquinoline13,14 inhibi-
tors of CYP11B2 (e.g., 2).
We have recently reported that the acylated amino tetra-
hydroisoquinoline (+)-(R)-3 is a potent inhibitor of human and
cynomolgus monkey aldosterone synthase and is remarkably
selective for CYP11B2 over CYP11B1 (selectivity factor (SF) =
160) in the latter species (Fig. 1).19 The tetrahydroisoquinoline
(+)-(R)-3 was made available following a 6-step sequence of
reactions that also provided access to various analogues.19
Notably, tetrahydroisoquinoline (+)-(R)-3 significantly lowered
aldosterone plasma levels in a dose-dependent manner in
mice, rat and cynomolgus monkeys, and protected kidney
Scheme 1 Application of the inverse electron demand Kondrat’eva
reaction to the synthesis of annulated pyridines 6–12 (yields reported in
parentheses). aCompound 11 was prepared in a microwave reactor, see
ESI† for details. IC50 values were measured against hCYP11B2 and are
structure and function in a ZSF1 rat model of chronic kidney the mean value of ≥3 experiments with a standard deviation <25%. The
disease.19 Here we describe our efforts aimed at identifying
selectivity factor (SF) = IC50 (hCYP11B2)/IC50 (hCYP11B1).
further optimized CYP11B2 inhibitors with an improved
selectivity profile and suitability for proof-of-concept studies in
translational animal models of hyper-aldosteronism.
appendage on activity and selectivity, the N-methyl dihydro-
As depicted in Scheme 1, we have recently reported20 a con- quinolinone 14 was prepared directly from the C–H arylation22
tinuous flow inverse-electron-demand Kondrat’eva reaction21 of oxazole and reacted20 with cyclopentene or cycloheptene to
and applied this process to the rapid synthesis of several annu- produce the annulated pyridines 11 and 12, respectively. Both
lated pyridines. Considering that these annulated pyridines of these later compounds proved to be potent inhibitors of
(e.g., 6–10) are produced in a single step from commercially hCYP11B2, with selectivity comparable to the corresponding
available starting materials and bear clear structural resem- cyclopenta[c]pyridines 9 and 10. Moreover, the 2- or 3-fluoro-4-
blance to established aldosterone synthase inhibitors (e.g., 2 trifluoromethylphenyl analogues also displayed improved
and (+)-(R)-3, Fig. 1), we endeavored to expand the scope of selectivity for hCYP11B2/hCYP11B1 when compared to the
this convenient process for the production of new inhibitors of annulated pyridine 7 (SF = 8), which incorporates the 4-cyano-
hCYP11B2. Towards this goal, biological testing of the readily phenyl function found in the lead candidate (+)-(R)-3.19 Un-
available annulated pyridines 6–10 revealed that these com- fortunately, the physicochemical and metabolic properties
pounds were indeed potent inhibitors of hCYP11B2, and that of these lipophilic annulated pyridines were not optimal.
the 3-fluoro and 2-fluoro-4-trifluomethylphenyl analogues 9 For example, the cyclopenta[c]pyridine 9 has an aqueous solu-
and 10 displayed selectivity for hCYP11B2 over hCYP11B1 that bility of only 1.0 µg mL−1 (Lysa assay23) and high clearance in
inspired further efforts in this area. The 1-methyl-3,4-dihydro- both human and rat microsomes (CL = 300 and 164 µL min−1
quinolin-2-one side-chain, originally introduced by the mg−1 protein). Moreover, the log D value for 9 could not
Hartmann group, confers particularly high potency and selecti- be measured due to compound precipitation.24 However,
vity for CYP11B2.13 To further explore the influence of this aryl passive permeability as measured by the Pampa assay25
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem., 2016, 14, 5922–5927 | 5923

View Article Online
Communication
Organic & Biomolecular Chemistry
(Pe = 1.09 × 10−6 cm s−1) as well as cytochrome P450 (CYP) profile (CYP3A4, CYP2D6 and CYP2C9: all >50 µM). The ter-
inhibition data for this compound (CYP3A4, CYP2D6, and tiary alcohol 20 demonstrated similar CYP11B2 inhibitory
CYP2C9: all >45 µM) were encouraging.
activity (IC50 = 13 nM, SF = 28) and a promising physico-
In order to reduce lipophilicity and increase aqueous solu- chemical and metabolic profile (log D = 3.63, solubility =
bility of the most promising aldosterone synthase inhibitors 317 µg mL−1, hCL/rCL = 10/10 µL min−1 mg−1 protein,
(i.e., 9 and 10), we investigated the incorporation of an alcohol CYP3A4, CYP2D6, and CYP2C9: all >50 µM). Interestingly,
function onto the saturated annulated ring. Toward this goal, despite the slightly increased lipophilicity of the tertiary
pyridylic oxidation26 gave the regioisomeric ketones 15–18 in alcohol 20, this compound demonstrated solubility similar
modest yield (∼50%).27,28 Reduction of the carbonyl function to that of alcohol 19 and reduced microsomal turnover
with NaBH4 or 1,2-addition of methyl lithium then afforded a (rat microsomes).
small collection of pyridylic alcohols 19–23. As summarized
Considering the improved properties of the alcohols 19
in Scheme 2, introduction of an alcohol function at the and 20, we endeavored to access additional 7-functionalized
7-position had little effect on potency or selectivity (e.g., 19: cyclopenta[c]pyridines. Exploiting the inverse-electron-demand
IC50 = 5 nM, SF = 14). Conversely, the corresponding 5-hydroxy- Kondrat’eva reaction,20 2-cyclopentene-1-acetic acid (25) was
cyclopenta[c]pyridines 21–23 were less active (e.g., 23: IC50
=
reacted with oxazole 24 at high temperatures to provide
739 nM, SF > 41). Gratifyingly, the physicochemical and meta- the regioisomeric cyclopenta[c]pyridines 26 and 27
bolic profile of the 7-hydroxy cyclopenta[c]pyridine 19 was
improved considerably when compared to the parent com-
pound 9. For example, the pyridin-7-ol 19 has a measureable
log D value of 3.54, enhanced solubility (46.0 µg mL−1), passive
permeability (Pe = 4.44 × 10−6 cm s−1), improved clearance
(hCL/rCL = 10/36 µL min−1 mg−1 protein) and a clean CYP
Scheme 3 Synthesis and hCYP11B2 inhibitory activity of the amides
28–33. The selectivity factor (SF) = IC50 (CYP11B2)/IC50 (CYP11B1). Con-
ditions: (a) TFA, 200 °C, microwave reactor, 10 h (34%, ca. 1 : 1 mixture
of 26 and 27 separated by MPLC); (b) RNH2, HATU, DIEA, DMF, rt, 16 h
(amides 28–33 purified by preparative HPLC). Yields for compounds
Scheme 2 Synthesis and hCYP11B2 inhibitory activity of the hydroxyl-
ated annulated cyclopenta[c]pyridines 19–23. The selectivity factor (SF)
= IC50 (CYP11B2)/IC50 (CYP11B1). Conditions: (a) dirhodium(II)tetrakis-
(caprolactam), TBHP, NaHCO3, CH2Cl2, rt, 48 h; (b) NaBH4, MeOH, 0 °C;
(c) MeLi, LiCl, Et2O, 0 °C to rt, 2 h. TBHP = tert-butyl hydroperoxide.
28–33 after preparative HPLC purification: 28 (30%); 29 (21%); 30 (25%);
31 (34%); 32 (19%); 33 (20%). TFA = trifluoroacetic acid, HATU = 1-[bis(di-
methylamino)-methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium
3-oxid
hexafluorophosphate, DIEA
=
diisopropyl ethylamine, DMF =
dimethylformamide.
5924 | Org. Biomol. Chem., 2016, 14, 5922–5927
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Communication
(Scheme 3).29,30 Notably, due to the involatile nature of 2-cyclo- bearing a 3-fluoro or 2-fluoro-4-trifluoromethylphenyl group
pentene-1-acetic acid 25, this reaction could be performed in a (see above), we elected to couple only 2-fluoro-4-trifluoro-
microwave reactor. Standard amide formation followed by pre- methylphenyl boronic acid (36) to the bromopyridines 35,
parative HPLC then afforded the amides 28–32, which were which provided the corresponding enoate. Hydrogenation
thus accessed in only 2 steps from oxazole 24. For comparison afforded the ester 37 that was subsequently reacted with a
purposes, the regioisomeric amide 33 was also isolated follow- small collection of primary or secondary amines afforded the
ing coupling of the mixture of cyclopenta[c]pyridines 26 amides 38–42. As highlighted in Scheme 4, within this series of
and 27 with (aminomethyl)cyclopropane. As indicated in amides the N,N-dimethylacetamide 42 (IC50 = 32 nM, SF = 96)
Scheme 3, the amides 28–32 were all potent inhibitors of and N-methylacetamide 38 (IC50 = 11 nM, SF = 95) proved to be
hCYP11B2 and showed modest selectivity (SF ∼ 5–18) over both the most potent and selective inhibitors of hCYP11B2
hCYP11B1. As expected, the regioisomeric amide 33 proved to and the former compound was selected for chiral separation.
be a less potent inhibitor of hCYP11B2.
The isolated enantiomers of 42 displayed significantly
The promising inhibitory activity of the 6,7-dihydro-5H- different affinities for the hCYP11B2 receptor, with the
cyclopenta[c]pyridines 28–32 prompted us to explore the analo- (+)-enantiomer (IC50 = 36 nM, SF = 42) being ∼4-fold more
gous series of cyclohexa[c]pyridines, which possess a tetra- potent than the (−)-enantiomer (IC50 = 133 nM, SF = 69). The
hydroisoquinoline core similar to that of the reported log D value for (+)-42 was measured to be 3.63 and the solubi-
aldosterone synthase inhibitor (+)-(R)-3. We have reported pre- lity was found to be 191 µg mL−1. Passive permeability of this
viously a significantly reduced yield for the inverse electron compound was high (Pe = 3.12 × 10−6 cm s−1), the CYP profile
demand Kondrat’eva reaction involving cyclohexene,20 and was clean (CYP3A4, CYP2D6 and CYP2C9: all >50 µM)
elected instead to prepare this series of compounds following and clearance in both human and rat microsomes (hCL/rCL =
a linear sequence of reactions that initiated with the bromo- 10/61 µL min−1 mg−1 protein) was much improved over the
pyridine 34 (Scheme 4).19 Appending the appropriate carboxy- unfunctionalized annulated pyridine 10 (Scheme 1).
methyl group involved
a
Horner–Wadsworth–Emmons
In order to better rationalize the structure activity relation-
reaction, which afforded a 3 : 2 mixture of E : Z alkenes 35. ships among these annulated pyridines, we modeled represen-
Considering the similar inhibitory activity of compounds tative ligands into the recently solved X-ray crystal structure of
hCYP11B2 with the bound 4-chloro-3-fluorophenyl analogue of
compound (+)-(R)-3.19 As depicted in Fig. 2, there is good
spatial overlap between the cyclohexa[c]pyridine 42 and the
anaologue of this previously reported inhibitor (+)-(R)-3. Both
compounds have a close contact of the pyridine nitrogen atom
to the heme iron. Additionally, both the propanamide appen-
dage of this (+)-(R)-3 analogue and the N,N-dimethylacetamide
of 42 are preferentially oriented perpendicular to the tetra-
Scheme 4 Synthesis and hCYP11B2 inhibitory activity of the amides 38
− 42. The selectivity factor (SF) = IC50 (CYP11B2)/IC50 (CYP11B1). Con-
ditions: (a) (i) ethyl 2-(diethoxyphosphoryl)acetate, NaHMDS, THF,
−50 °C to 0 °C, 1 h; (ii) 34, reflux, 4 h (ca. 3 : 2 mixture of E : Z isomers);
(b) 36, Pd(PPh3)4, Na2CO3, EtOH : H2O (6 : 1), 85 °C, microwave reactor,
6 h; (c) H2 (3 bar), 10% Pd/C, MeOH, rt, 16 h; (d) NHR1R2, Al(CH3)3, THF,
microwave reactor, 120 °C, 1 h (yields over two steps: 38 (95%); 39
(67%); 40 (89%); 41 (59%); 42 (89%)). NaHMDS = sodium bis(trimethyl-
silyl)amide.
Fig. 2 Model of compound 42 (cyan) in the X-ray structure of
hCYP11B2 with the 4-chloro-3-fluorophenyl analogue of (+)-(R)-3 (PDB
code: 4ZGX)19 in yellow. The protein is colored by increasing B-factor
on a blue to red color scale with red indicating B-factors >50 Å2. The
heme is shown in green. The four amino acid residues that are in close
contact (≤4.5 Å) with a potential substituent at C5 of the annulated
6-membered ring system are labeled.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem., 2016, 14, 5922–5927 | 5925

View Article Online
Communication
Organic & Biomolecular Chemistry
hydroisoquinoline core to minimize steric strain. These model- support with acquisition and analysis of NMR spectra,
ing data further suggest that the more potent enantiomer of 42 Mr Christian Bartelmus for MS and HRMS spectra, Mrs
(i.e., (+)-42) should possess an (R)-configuration at C8. As Isabelle Parrilla, Mr Björn Wagner and Mr Severin Wendel-
discussed above, in general, the 5-substituted cyclopenta[c] spiess for physicochemical characterization of compounds,
pyridines (e.g., 21–23 and 33) are significantly less active than Mrs Sandrine Simon for microsomal stability data, Dr Stephen
the corresponding 7-substituted isomers. Our binding mode Fowler for CYP inhibition measurements, Dr Doris Roth,
model indicates that C5 appendages point into a hydrophobic Dr Remo Hochstrasser, and Mr Henry Cabrière and Mrs
pocket formed by the side chains of amino acid residues Trp Angelika Schuler for running the CYP11B1 and CYP11B2
116, Phe 130, Phe 231, and Phe 487, while substituents at the in vitro assays. R. B. gratefully acknowledges support from the
7-position are more exposed to solvent. The small size and Michael Smith Foundation for Health Research through a
necessary desolvation penalty involved with the placement of Career Investigator Award.
polar functional groups (e.g., hydroxyl or carbonyl) into this
pocket may well be responsible for the diminished inhibitory
activity of the C5-substituted compounds. Furthermore, the
opposite side of the hCYP11B2 binding pocket (i.e., location
References
of C7 appendages) shows higher crystallographic B-factors,
indicating greater intrinsic flexibility and less preorganization.
This inherent flexibility in combination with partial solvent
exposure may also explain the fact that inhibitory activities
varied only slightly amongst the series of 7-functionalized
cyclopenta[c]pyridines (e.g., 28–32) and cyclohexa[c]-pyridines
(e.g., 38–42).
1 B. J. Hoogwerf, Am. J. Cardiol., 2010, 105(suppl), 30A–35A.
2 C. A. Romero, M. Orias and M. R. Weir, Nat. Rev. Endo-
crinol., 2015, 11, 242.
3 (a) B. Pitt, F. Zannad, W. J. Remme, R. Cody, A. Castaigne,
A. Perez, J. Palensky and J. Wittes, N. Engl. J. Med., 1999,
341, 709; (b) F. Zannad, J. J. McMurray, H. Krum, D. J. van
Veldhuisen, K. Swedberg, H. Shi, J. Vincent, S. J. Pocock
and B. Pitt, N. Engl. J. Med., 2011, 364, 11; (c) B. Pitt,
L. Kober, P. Ponikowski, M. Gheorghiade, G. Filippatos,
H. Krum, Ch. Nowack, P. Kolkhof, S.-Y. Kim and F. Zannad,
Eur. Heart J., 2013, 34, 2453; (d) M. J. Meyers,
G. B. Arhancet, S. L. Hockerman, X. Chen, S. A. Long,
M. W. Mahoney, J. R. Rico, D. J. Garland, J. R. Blinn,
J. T. Collins, S. Yang, H.-C. Huang, K. F. McGee,
J. M. Wendling, J. D. Dietz, M. A. Payne, B. L. Homer,
M. I. Heron, D. B. Reitz and X. Hu, J. Med. Chem., 2010, 53,
5979.
4 J. Staessen, P. Lijnen, R. Fagard, L. J. Verschueren and
A. Amery, J. Endocrinol., 1981, 91, 457.
5 A. D. Struthers, Eur. Heart J., 1995, 16, 103.
6 J. S. Williams, Curr. Opin. Endocrinol., Diabetes Obes., 2013,
20, 198.
7 (a) G. H. Williams, Endocrinol. Metab. Clin. North Am.,
1994, 23, 429; (b) Q. Hu, L. Yin and R. W. Hartmann,
J. Med. Chem., 2014, 57, 5011.
Conclusions
In conclusion, we have exploited a recent advance in the
inverse electron demand Kondrat’eva reaction to prepare a
series of 7-substituted 6,7-dihydro-5H-cyclopenta[c]pyridines
and have evaluated their activity as aldosterone inhibitors.
These compounds demonstrated promising activity and
selectivity over the structurally related hCYP11B1, prompting
us to explore the synthesis of the analogous cyclohexa[c]
pyridines. The tetrahydroisoquinoline 42 proved to be the
most selective compound in this latter series and following
separation of the enantiomers by chiral HPLC, the (+)-enantio-
mer was found to be approximately 4-fold more potent than
the (−)-enantiomer against aldosterone synthase and twice
as selective against the humans enzyme CYP11B2 than our
previously reported inhibitor (+)-(R)-3. In order to rationalize
the relationship between structure and CYP11B2 inhibitory
activity, binding mode models for representative ligands were
constructed using the crystal structure of human CYP11B2
with the structurally close analogue of the known ligand
8 C. D. Schumacher, R. E. Steele and H. R. Brunner, J. Hyper-
tens., 2013, 31, 2085.
9 M. Azizi, L. Amar and J. Ménard, Nephrol., Dial., Trans-
plant., 2013, 28, 36.
(+)-(R)-3 as a template. Despite some topological differences, 10 R. Bernhardt, Expert Opin. Ther. Targets, 2016, DOI:
the novel aldosterone inhibitor 42 and 4-chloro-3-fluorophenyl 10.1517/14728222.2016.1151873.
tetrahydroisoquinoline analogue (+)-(R)-3 were found to 11 M. H. Bassett, P. C. White and W. E. Rainey, Mol. Cell.
occupy a very similar binding site region with close contact
of the pyridine nitrogen atom to the heme iron.
Endocrinol., 2004, 217, 67.
12 M. A. Cerny, Curr. Top. Med. Chem., 2013, 13, 1385.
13 S. Lucas, R. Heim, C. Ries, K. E. Schewe, B. Birk and
R. W. Hartmann, J. Med. Chem., 2008, 51, 8077.
14 C. Ries, S. Lucas, R. Heim, B. Birk and R. W. Hartmann,
J. Steroid Biochem. Mol. Biol., 2009, 116, 121.
Acknowledgements
We thank Mr Christoph Kuratli, Mr Ronny Hardegger and 15 (a) L. Amar, M. Azizi, J. Ménard, S. Peyrard, C. Watson and
Mrs Jennifer Steiner for technical support, Mr Daniel Zimmerli
P.-F. Plouin, Hypertension, 2010, 56, 831; (b) J. Ménard,
for chiral separation of compound 42, Dr Inken Plitzko for
D. F. Rigel, C. Watson, A. Y. Jeng, F. Fu, M. Beil, J. Liu,
5926 | Org. Biomol. Chem., 2016, 14, 5922–5927
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Communication
W. Chen, C.-W. Hu, J. Leung-Chu, D. LaSala, G. Liang,
S. Rebello, Y. Zhang and W. P. Dole, J. Transl. Med., 2014,
12, 340.
1986, 86, 781; (c) D. L. Boger, Tetrahedron, 1983, 39, 2869;
(d) I. J. Turchi, Ind. Eng. Chem. Prod. Res. Dev., 1981, 20, 32;
(e) I. J. Turchi and M. J. S. Dewar, Chem. Rev., 1975, 75, 389;
(f) A. Padwa and L. A. Cohen, J. Org. Chem., 1984, 49, 399;
(g) L.-A. Jouanno, V. Tognetti, L. Joubert, C. Sabot and
P.-Y. Renard, Org. Lett., 2013, 15, 2530; (h) V. P. Singh,
J.-F. Poon and L. Engman, J. Org. Chem., 2013, 78, 1478;
(i) S. N. Goodman, D. M. Mans, J. Sisko and H. Yin, Org.
Lett., 2012, 14, 1604; ( j) C. Sabot, E. Oueis, X. Brune and
P.-Y. Renard, Chem. Commun., 2012, 48, 768; (k) C. Lalli,
M. J. Bouma, D. Bonne, G. Masson and J. Zhu, Chem. – Eur.
J., 2011, 17, 880; (l) J. I. Levin and S. M. Weinreb, J. Am.
Chem. Soc., 1983, 105, 1397; (m) E. E. Harris,
R. A. Firestone, K. Pfister, R. R. Boettcher, F. J. Cross,
R. B. Currie, M. Monaco, E. R. Peterson and W. Reuter,
J. Org. Chem., 1962, 27, 2705.
16 J. P. N. Papillon, Ch. Lou, A. K. Singh, Ch. M. Adams,
G. M. Ksander, M. E. Beil, W. Chen, J. Leung-Chu, F. Fu,
L. Gan, Ch.-W. Hu, A. Y. Jeng, D. LaSala, G. Liang,
D. F. Rigel, K. S. Russell, J. A. Vest and C. Watson, J. Med.
Chem., 2015, 58, 9382.
17 (a) Q. Hu, L. Yin, A. Ali, A. J. Cooke, J. Bennett, P. Ratcliffe,
M. M. Lo, E. Metzger, S. Hoyt and R. W. Hartmann, J. Med.
Chem., 2015, 58, 2530; (b) S. B. Hoyt, W. Petrilli, C. London,
G.-B. Liang, J. Tata, Q. Hu, L. Yin, C. J. van Koppen,
R. W. Hartmann, M. Struthers, T. Wisniewski, N. Ren,
Ch. Bopp, A. Sok, T.-Q. Cai, S. Stribling, L.-Y. Pai, X. Ma,
J. Metzger, A. Verras, D. McMasters, Q. Chen, E. Tung,
W. Tang, G. Salituro, N. Buist, J. Clemas, G. Zhou,
J. Gibson, C. A. Maxwell, M. Lassman, T. McLaughlin, 22 (a) N. A. Strotman, H. R. Chobanian, Y. Guo, J. He and
J. Castro-Perez, D. Szeto, G. Forrest, R. Hajdu,
M. Rosenbach and Y. Xiong, ACS Med. Chem. Lett., 2015, 6,
861; (c) S. B. Hoyt, M. K. Park, C. London, Y. Xiong, J. Tata,
J. E. Wilson, Org. Lett., 2010, 12, 3578; (b) S. A. Ohnmacht,
P. Mamone, A. J. Culshaw and M. F. Greaney, Chem.
Commun., 2008, 1241.
D. J. Bennett, A. Cooke, J. Cai, E. Carswell, J. Robinson, 23 Aqueous solubility was measured from lyophilized DMSO
J. MacLean, L. Brown, S. Belshaw, T. R. Clarkson, K. Liu,
G.-B. Liang, M. Struthers, D. Cully, T. Wisniewski, N. Ren,
C. Bopp, A. Sok, T.-Q. Cai, S. Stribling, L.-Y. Pai, X. Ma,
stock solutions spectrophotometrically at pH = 6.5 in a
50 mM phosphate buffer (Lyophilisation solubility assay,
Lysa).
J. Metzger, A. Verras, D. McMasters, Q. Chen, E. Tung, 24 log D values were measured spectrophotometrically at pH =
W. Tang, G. Salituro, N. Buist, J. Kuethe, N. Rivera,
J. Clemas, G. Zhou, J. Gibson, C. A. Maxwell, M. Lassman,
7.4 in an 1-octanol/50 mM TAPSO buffer system containing
5% (v/v) DMSO.
T. McLaughlin, J. Catro-Perez, D. Szeto, G. Forrest, 25 Pampa (Parallel Artificial Membrane Permeation Assay):
R. Hajdu, M. Rosenbach and A. Ali, ACS Med. Chem. Lett.,
2015, 6, 573; (d) L. Yin, Q. Hu, J. Emmerich, M. M.-C. Lo,
E. Metzger, A. Ali and R. W. Hartmann, J. Med. Chem.,
low: Pe < 0.1, medium: 0.1 < Pe < 1.0, high: Pe > 1.0. See:
M. Kansy, F. Senner and K. Gubernator, J. Med. Chem.,
1998, 41, 1007.
2014, 57, 5179; (e) Q. Hu, L. yin and R. W. Hartmann, 26 M. P. Doyle, L. J. Westrum, W. N. E. Wolthuis, M. M. See,
J. Med. Chem., 2012, 55, 7080.
18 J. P. N. Papillon, C. M. Adams, Q.-Y. Hu, Ch. Lou,
W. P. Boone, V. Bagheri and N. M. Pearson, J. Am. Chem.
Soc., 1993, 115, 958.
A. K. Singh, Ch. Zhang, J. Carvalho, S. Rajan, A. Amaral, 27 J. Aebi, K. Amrein, B. Hornsperger, B. Kuhn, Y. Liu,
M. E. Beil, F. Fu, E. Gangl, Ch.-W. Hu, A. Y. Jeng, D. LaSala,
G. Liang, M. Logman, W. M. Maniara, D. F. Rigel,
S. A. Smith and G. M. Ksander, J. Med. Chem., 2015, 58,
4749.
H. P. Märki, R. E. Martin, A. V. Mayweg, P. Mohr
and X. Tan, WO 2013/156423, 2013 (F. Hoffmann-La
Roche AG).
28 The structures of regioisomeric ketones 15–18 were
determined unambiguously by analysis 2D NMR spectra
(HMBC, HSQC).
19 R. E. Martin, J. D. Aebi, B. Hornsperger, H.-J. Krebs,
B. Kuhn, A. Kuglstatter, A. M. Alker, H. P. Märki, S. Müller,
D. Burger, G. Ottaviani, W. Riboulet, P. Verry, X. Tan, 29 J. Aebi, K. Amrein, R. Britton, B. Hornsperger, B. Kuhn,
K. Amrein and A. V. Mayweg, J. Med. Chem., 2015, 58, 8054.
20 J. Lehmann, T. Alzieu, R. E. Martin and R. Britton, Org.
Lett., 2013, 15, 3550.
21 For reviews and examples, see: (a) G. Y. Kondrat’eva, Khim.
Nauka Prom-st., 1957, 2, 666; (b) D. L. Boger, Chem. Rev.,
H. P. Märki, R. E. Martin, A. V. Mayweg and X. Tan, WO
2015/055602, 2015 (F. Hoffmann-La Roche AG).
30 J. Aebi, K. Amrein, R. Britton, B. Hornsperger, B. Kuhn,
H. P. Märki, R. E. Martin, A. V. Mayweg and X. Tan, WO
2015/055604, 2015 (F. Hoffmann-La Roche AG).
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem., 2016, 14, 5922–5927 | 5927