Evaluation of tert-butyl isosteres: Case studies of physicochemical and pharmacokinetic properties, efficacies, and activities

Article abstract of DOI:10.1002/cmdc.201402502

The tert-butyl group is a common motif in medicinal chemistry. Its incorporation into bioactive compounds is often accompanied by unwanted property modulation, such as increased lipophilicity and decreased metabolic stability. Several alternative substituents are available for the drug discovery process. Herein, physicochemical data of two series of drug analogues of bosentan and vercirnon are documented as part of a comparative study of tert-butyl, pentafluorosulfanyl, trifluoromethyl, bicyclo[1.1.1]pentanyl, and cyclopropyl-trifluoromethyl substituents.

Full text of DOI:10.1002/cmdc.201402502

DOI: 10.1002/cmdc.201402502
Full Papers
Evaluation of tert-Butyl Isosteres: Case Studies of
Physicochemical and Pharmacokinetic Properties,
Efficacies, and Activities
Matthias V. Westphal,[a] Bernd T. Wolfstꢀdter,[a, b] Jean-Marc Plancher,[c] John Gatfield,[d] and
Erick M. Carreira*[a, b]
The tert-butyl group is a common motif in medicinal chemistry.
Its incorporation into bioactive compounds is often accompa-
nied by unwanted property modulation, such as increased
lipophilicity and decreased metabolic stability. Several alterna-
tive substituents are available for the drug discovery process.
Herein, physicochemical data of two series of drug analogues
of bosentan and vercirnon are documented as part of a compa-
rative study of tert-butyl, pentafluorosulfanyl, trifluoromethyl,
bicyclo[1.1.1]pentanyl, and cyclopropyl-trifluoromethyl sub-
stituents.
Introduction
The tert-butyl group is a substituent often encountered in
drugs. Examples of pharmaceutically active ingredients bearing
this entity include salbutamol,[1] GlaxoSmithKline’s b2-adrener-
gic receptor agonist, remikiren,[2] Roche’s renin inhibitor, and
terfenadin,[3] Sanofi-Aventis’ antihistamine. The incorporation
of a tert-butyl group onto a scaffold of interest, however, is
often accompanied by a decrease in metabolic stability, a phe-
nomenon known to correlate with lipophilicity (logP, logD).[4]
Such issues are addressed early on in the modeDunning Jr.rn
drug discovery process through the identification of replace-
ments or scaffold hopping to ensure the viability of a drug
candidate throughout the discovery process and to minimize
dropout rates at later stages. There are a number of surrogate
groups that could be considered as tert-butyl isosteres, namely
pentafluorosulfanyl (SF5),[5,6] bicyclo[1.1.1]pentanyl (BCP),[7,8]
and cyclopropyl-trifluoromethyl (cyclopropyl-CF3).[9] Surprising-
ly, these have not yet found widespread use in this context,
which may result from the lack of comparative physicochemi-
cal data available to those active in the discipline. Herein, we
present two case studies (Figure 1) focused on tert-butyl surro-
[a] M. V. Westphal, B. T. Wolfstꢀdter, Prof. E. M. Carreira
Laboratorium fꢁr Organische Chemie, ETH Zꢁrich
Vladimir-Prelog-Weg 3, 8093 Zꢁrich (Switzerland)
E-mail: carreira@org.chem.ethz.ch
[b] B. T. Wolfstꢀdter, Prof. E. M. Carreira
Competence Center for Systems Physiology and Metabolic Diseases
Schorenstrasse 16, 8603 Schwerzenbach (Switzerland)
[c] Dr. J.-M. Plancher
Pharmaceutical Research & Early Development
Roche Innovation Center Basel, F. Hoffmann-La Roche Ltd.
Grenzacherstrasse 124, 4070 Basel (Switzerland)
Figure 1. Summary of selected parameters investigated in this work for bo-
sentan and vercirnon analogues: logD=intrinsic distribution coefficient be-
tween octanol and aqueous buffer (pH 7.4); pKa =acidity constant deter-
mined spectrophotometrically at 23ꢀ18C; Cl=intrinsic clearance
(mmminꢁ1 mgꢁ1); aqueous solubility was determined by using the lyophiliza-
tion solubility assay (LYSA) at pH 6.5; PAMPA=membrane permeability as
derived from the parallel artificial membrane permeability assay.
[d] Dr. J. Gatfield
Drug Discovery Biology, Actelion Pharmaceuticals Ltd.
Gewerbestrasse 16, 4123 Allschwil (Switzerland)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201402502.
ChemMedChem 0000, 00, 0 – 0
1
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
gates that provide fundamental information for use in the
drug discovery process. The work includes assessment of bio-
activities of drug analogues of bosentan (1a) and vercirnon
(2a) as well as important in vitro data on chemical and biologi-
cal stability (pH and metabolic liabilities), cytochrome P450
enzyme inhibition, membrane permeability, distribution coeffi-
cients, ionization constants, and aqueous solubility (at pH 6.5).
To the best of our knowledge, this is the only comprehensive
study that allows a direct comparison of tert-butyl isosteres in
terms of these important parameters.
ment for tert-butyl. According to the investigators, this group
possesses significantly improved metabolic stability over its
tert-butyl counterpart in simple model compounds.[9] In sepa-
rate work, researchers at Gilead confirmed this trend in the
context of NS5B inhibitors as promising anti-HIV drugs.[26]
The bicyclo[1.1.1]pentane motif has been incorporated by
Novo Nordisk researchers as a phenyl ring isostere in com-
pounds that interact with metabotropic glutamate receptors[27]
and later by Pfizer scientists focusing on g-secretase inhibi-
tors.[8] In addition, it was used in a study on quinolone antibac-
terials and shown to be more active than structurally closely
related ciprofloxacin.[28] Compounds incorporating bicy-
clo[1.1.1]pentane were also compared with counterparts in-
cluding fluorinated cyclic and open-chain alkyl groups, and
this group was found to be a reasonable surrogate with re-
spect to calculated logP, clearance rates, and Caco-2 permea-
bility.[29] Despite the wide ranging studies, there has been no
direct comprehensive comparison of the various groups.
We became interested in a comparative examination of all
five groups in a wide range of assays and chose two promi-
nent biologically active components for this work. Bosentan
(1a) is a dual endothelin receptor antagonist used for the
treatment of pulmonary arterial hypertension.[30] Vercirnon (2a)
is a CCR-9 receptor antagonist originally developed for the
treatment of inflammatory bowel disease.[31] Both compounds
share a para-substituted tert-butylphenyl sulfonamide unit
(Figure 1). Their straightforward synthesis and the availability
of robust bioassays to determine functional activity of the de-
rived analogues made them ideal starting points for an investi-
gation of tert-butyl isosteres.
In the context of our ongoing efforts to access novel struc-
tures in uncharted chemical space, we recently introduced
small saturated heterocycles[10,11] and spirocycles[12,13] for use as
novel building blocks in discovery medicinal chemistry pro-
grams. In line with these interests, we and others have focused
increased attention on the pentafluorosulfanyl group (SF5).[14–17]
It was originally described by Sheppard in 1960[18] and has
found selected applications recently in materials science, crop
protection, and pharmaceuticals.[19] The few reports on SF5
drug analogues describe its use for the replacement of CF3
and tert-butyl substituents. However, these examples are, with
two exceptions described below, strictly limited to the compar-
ison of bioactivity. For example, Wipf et al. synthesized two SF5
analogues of the CF3-containing antimalarial mefloquine and
demonstrated the analogues to display equivalent antimalarial
activity.[5] Additionally, Diederich and co-workers published
a study on typanothione reductase inhibitors in which the SF5-
and CF3-substituted analogues proved superior to the tert-
butyl counterpart.[20] The finding of improved activity was at-
tributed to the smaller size of these two relative to tert-butyl.
Pentafluorosulfanyl analogues of 5-hydroxytryptamine receptor
inhibitors fluoxetine, fenfluramine, and norfenfluoramine were
synthesized by Welch and Lim. These investigators noted that
SF5 exhibits conventional substituent influences in radioligand
binding assays and that the selectivity for certain 5-HT receptor
subtypes was enhanced by the switch from CF3 to SF5.[21] The
O’Hagan research group published a work in which the CF3
group of the allosteric calcium receptor agonist cinacalcet was
replaced by SF5 and documented similar biological profiles for
both compounds.[22] In 2011, Phillips and Rathod disclosed lead
optimization efforts of triazolopyrimidine-based analogues
with antimalarial activity, in which CF3 and SF5 analogues were
among the most potent compounds and exhibited favorable
physicochemical properties.[23,24] Recently, Zanda and col-
leagues disclosed work on cannabinoid receptor ligands in
which the SF5 group was compared with CF3 and tert-butyl.
Evaluating experimentally derived logP values and radioligand
binding assays, they concluded that SF5 is a bioisostere of CF3
and possibly of tert-butyl.[6] Herein, we substantiate this finding
by providing additional, directly comparable physicochemical
data. The results we describe also complement recent work
from GlaxoSmithKline that was limited to the specific compari-
son of CF3 and SF5 analogues of a dopamine antagonist with
respect to clearance rates, CYP inhibition constants, and a col-
lection of in vivo pharmacokinetic (PK) properties.[25]
We examined the following parameters to derive a full phys-
icochemical profile of all compounds:
1. pH: stability of compounds in aqueous media was deter-
mined at pH 1, pH 4, pH 6.5, pH 8, and pH 10.
2. CYP inhibition: compounds were tested for inhibition of
three P450 enzyme isoforms (2D6, 3A4, and 2C9).
3. logD: partition coefficients between an aqueous buffer
(pH 7.4) and octanol were determined.
4. Membrane permeability was determined by using the par-
allel artificial membrane permeability assay (PAMPA).
5. pKa: ionization constants were determined by spectropho-
tometric analysis at different pH values.
6. Solubility was determined according to the lyophilization
solubility assay (LYSA)[32] at pH 6.5.
7. Metabolic stability: intrinsic clearance rates were deter-
mined in human, mouse, and rat liver microsome prepara-
tions.
A simple web-based calculation using the freely available
molinspiration software revealed the volumes of the substitu-
ents used in this study relative to tert-butyl, namely DV=
[33]
VanalogueꢁVt-butyl
.
Their size increases in the order CF3 (DV=
ꢁ34.9 ꢂ3) < SF5 (DV=ꢁ11.1 ꢂ3) < BCP (DV=ꢁ4.1 ꢂ3) < t-
butyl < cyclopropyl-CF3 (DV= +4.4 ꢂ3) (Figure 2). With CF3
being significantly smaller, the similar volumes of the other
substituents paired with their different shapes and electronic
A recent publication from Novartis describes the cycloprop-
yl-CF3 group as a useful substituent, particularly as a replace-
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
2
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Figure 2. Relative volumes of substituents in this study.[33]
properties offer valuable opportunities for structure–activity
studies in the context of any discovery program.
Results and Discussion
Synthesis
Scheme 2. Synthesis of bosentan and vercirnon analogues.
We commenced the synthetic efforts with the preparation of
sulfonyl chlorides 4b–c and sulfonyl amides 6b–e (Scheme 1).
um carbonate, bispyrimidine 7, and the respective sulfonamide
6b–e in acetonitrile was heated at 808C until the starting ma-
terial was consumed. Subsequent addition of ethylene glycol
and prolonged heating at 808C afforded bosentan analogues
1b–e in 37–60% yield. The reaction of 8 with sulfonyl chlor-
ides 4a–e in pyridine at room temperature afforded the corre-
sponding sulfonamides, which were transformed into the re-
spective vercirnon analogues 2a–e by mCPBA oxidation in di-
chloromethane in 53–96% yield over the two steps. Interest-
ingly, reactions with CF3- and SF5-substituted sulfonylchlorides
showed considerable amounts of the sulfonimide byproducts
during the first step of the sequence. We attribute this to the
strong electron-withdrawing properties of the two groups, ren-
dering sulfonyl chlorides 4b and 4e highly reactive.
Biological activity
Both drug analogue series are antagonists of ETA, ETB, and
CCR9 G-protein-coupled receptors. In both cases bioactivity
could be measured by a fluorescence imaging plate reader
(FLIPR) calcium assay using cells expressing the respective re-
ceptor, a fluorescent calcium indicator, the naturally occurring
agonist, and drug analogues 1a–e/2a–e.
For bosentan derivatives 1a–e, Chinese hamster ovary (CHO)
cells expressing the human endothelin receptor subtype A (ETA
receptor) or the ETB receptor, respectively, the cytosol of which
had been loaded with the fluorescent calcium indicator fluo-4,
were incubated with a dilution series of antagonists 1a–e. Ad-
dition of the natural agonist endothelin 1 (ET-1) induces Ca2+
efflux from the endoplasmic reticulum into the cytosol, where
it can be quantified by measuring the resulting fluorescence
upon binding to fluo-4.
Scheme 1. Synthesis of sulfonyl chlorides and sulfonamides.
Previously unknown SF5-substituted sulfonyl chloride 4b was
prepared from diazonium tetrafluoroborate 3.[34,35] When a solu-
tion of 3 in acetonitrile was added to a mixture of thionyl chlo-
ride, 2 mol% copper(I) chloride and concentrated hydrochloric
acid in water at 08C, 4b was isolated in 68% yield after simple
filtration. This material was sufficiently pure to be used in the
next step. Further purification by column chromatography on
silica led to decomposition (40% yield). It is important to note
that 3!4 represents the first instance describing the synthesis
of this SF5-arene sulfonyl chloride. We anticipate it will find
wide use for the preparation of novel arene sulfonamide deriv-
atives. Cyclopropyl-CF3-substituted sulfonyl chloride 4c was
prepared by chlorosulfonation of 5 in dichloromethane in 87%
yield as a single regioisomer. Subsequent aminolysis of arene
sulfonyl chlorides 4b–e using excess ammonia in water or
methanol furnished the corresponding sulfonamides 6b–e in
79–88% yield (Scheme 1).
Experiments at various antagonist concentrations allow the
determination of IC50 values (Table 1). At ETA, SF5- and CF3-bo-
sentan (1b, 1e) exhibit IC50 values ~10-fold higher than bosen-
tan 1a (112 and 130 nm versus 11 nm for 1a). The cyclopropyl-
CF3 and BCP analogues show about the same IC50 values as
tert-butyl-bosentan (9 and 4 nm). A similar trend could be ob-
Analogues of bosentan and vercirnon were synthesized ac-
cording to known routes (Scheme 2).[36,37] A mixture of potassi-
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
3
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Table 1. Bioassay of bosentan analogues.
Compound
IC50 [nm]
[a]
[b]
ETA
ETB
1a
1b
11 ꢀ1.7[c]
179[d]
994[d]
112 ꢀ12.1[c]
1c
9[d]
66[d]
1d
1e
4[d]
64[d]
130[d]
2720[d]
Figure 3. CCR9 Ca2+ response to TECK and TECK+antagonists: all measure-
ments were conducted at a concentration of 50 nm for each component.
DF=FmaxꢁF0; n=1, measured in triplicates. Bar graphs shown are
meanꢀSD.
[a] IC50 value at human endothelin receptor subtype A. [b] IC50 value at
human endothelin receptor subtype B. [c] n=3, measured in duplicates
per experiment. [d] n=1, measured in duplicates.
served at ETB. Accordingly, SF5- and CF3-bosentan exhibit 5-
and 15-fold higher IC50 values than the parent drug (994 and
2720 nm versus 179 nm). However, at this receptor cycloprop-
yl-CF3- and BCP-bosentan are significantly more active (66 and
64 nm). These results are in agreement with findings reported
by Neidhart and colleagues, who found that analogues bearing
lipophilic electron-donating para substituents in the sulfon-
amide group are most potent.[38]
CF3-substituted analogues showing slight CYP inhibition of the
2C9 subtype in the double-digit micromolar range (IC50: 17
and 13 mm). In contrast, inhibition of this subtype seems to be
a general issue in the vercirnon series, with IC50 values be-
tween 3.9 and 11 mm. Compounds 1a–d show similar permea-
bility values from 41 to 47 nmsꢁ1, whereas the corresponding
value for 1e (CF3) is lower (28 nmsꢁ1). Interestingly, this order
is reversed in compounds 2a–e, in which the CF3 analogue ex-
hibits the highest permeability (96 nmsꢁ1) exceeding SF5
(65 nmsꢁ1) and BCP (19 nmsꢁ1). In accordance with recent
For the vercirnon series, Ready-to-Assay CCR9 chemokine re-
ceptor expressing cells (Merck-Millipore) were incubated with
fluo-4. Addition of the thymus-expressed chemokine (TECK),
which acts as a CCR9 agonist, induces Ca2+ efflux from the en-
doplasmic reticulum into the cytosol, leading to a peak in fluo-
rescence. For the readout, baseline fluorescence was recorded
for 30 s (F0), then an aqueous solution of TECK or TECK+an-
tagonist was added to the cells, resulting in a final concentra-
tion of 50 nm for both TECK agonist and antagonist. Biological-
ly active ligands would be expected to abolish the calcium re-
lease. The resulting fluorescence was recorded, and the maxi-
mum value (Fmax) was normalized to baseline fluorescence
before the addition of ligands to the cells (FmaxꢁF0/F0; see Sup-
porting Information for details).
[6]
work comparing tert-butyl with CF3 and SF5 the lipophilicity
increases in the order CF3 < SF5 < cyclopropyl-CF3 < t-butyl <
BCP in both series. The electronic properties of the various
substituents are reflected in the pKa values. It comes as no sur-
prise that SF5-substituted analogues bear the most acidic sulfo-
namide proton in both series. The pKa values increase in the
order SF5 < CF3 < cyclopropyl-CF3 ꢂ tert-butyl < BCP.
For bosentan and its analogues, the introduction of SF5
proves favorable for solubility (604 versus 520 mgLꢁ1 for tert-
butyl analogue 1a). This is especially interesting when com-
pared with the CF3 analogue 1e (128 mgLꢁ1). Surprisingly, in-
troduction of the cyclopropyl-CF3 group appears to be detri-
mental to solubility (31 mgLꢁ1), whereas the BCP analogue 1d
is similar to the parent compound 1a in this regard, with a sol-
ubility of 484 mgLꢁ1. Ionization constants influence both logD
and LYSA values. However, with pKa values being at least 1.7
units below the buffer pH value, all analogues should be pres-
ent mainly as anions, so that differences in solubility could be
attributed mainly to the polarities of the substituents. Never-
theless, these findings should be considered with care due to
the limited precision of the LYSA. In contrast, the vercirnon an-
alogues generally showed low solubilities; however, it is inter-
esting to note that the SF5 vercirnon analogue 2b is the most
soluble in the series.
Figure 3 shows the results obtained. Sole addition of TECK
induces a significant fluorescence response (+58%). As shown,
this response is inhibited by every antagonist tested when ap-
plied in equimolar concentration of 50 nm. Thus, we can con-
clude that functional activity is retained throughout the series
upon replacement of the original tert-butyl group in 2a by the
substituents b–e.[39]
MDO parameters
Table 2 summarizes experimentally derived multidimensional
optimization (MDO) parameters of the drug analogues that
formed the basis of the study. In aqueous media, all com-
pounds are stable in the range pH 1–10. The bosentan series
behaves well in the CYP inhibition assay, with only SF5- and
All replacements tend to be metabolically more stable than
the tert-butyl group itself with few exceptions. SF5- and CF3-
substituted analogues are metabolically the most stable in
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
4
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Table 2. Physicochemical data of bosentan, vercirnon, and their analogues.
Compd
Aq.Stab.[a]
CYPIC50 [mm]
2C9; 2D6; 3A4[b]
logD[c]
PAMPA[d]
PAMPA
Details[e]
pKa[f]
LYSA[g]
hCl[h]
mCl[h]
rCl[h]
Bosentan
1a
stable
stable
>50;>50;>50
0.99ꢀ0.04
0.37ꢀ0.06
46ꢀ8
41ꢀ5
82; 0; 18
81; 4; 15
4.6ꢀ0.4
3.9ꢀ0.4
520ꢀ50
604ꢀ30
<10
<10
13ꢀ9
<10
37ꢀ9
<10
1b
1c
17ꢀ13;>50;>50
stable
>50;>50;>50
0.81ꢀ0.03
47ꢀ5
77; 7; 16
4.6ꢀ0.02
31ꢀ2
<10
15ꢀ5
<10
1d
stable
stable
>50;>50;>50
13ꢀ6;>50;>50
1.27ꢀ0.03
0.14ꢀ0.04
43ꢀ3
28ꢀ4
71; 15; 14
87; 1; 11
4.8ꢀ4
484ꢀ29
218ꢀ50
<10
<10
16ꢀ5
<10
16ꢀ5
<10
1e
4.1ꢀ0.4
Vercirnon
[j]
[j]
2a
2b
stable
stable
11ꢀ6;>50;>50
2.84ꢀ0.08
1.73ꢀ0.02
–
–
6.8ꢀ0.4
5.8ꢀ0.4
<0.6
14ꢀ3
<10
68ꢀ4
18ꢀ4
34ꢀ3
<10
3.9ꢀ0.5;>50;>50
65ꢀ8
40; 49; 11
3.6ꢀ0.3
[i]
[j]
[j]
2c
stable
5.1ꢀ1.4;>50;>50
–
–
–
6.5ꢀ0.03
<0.6
<10
50ꢀ11
20 ꢀ10
2d
2e
stable
stable
8.6ꢀ2.8;>50;>50
5.4ꢀ2.3;>50;>50
3.01ꢀ0.02
1.39ꢀ0.04
19ꢀ7
19; 80; 1
6.9ꢀ0.02
6.1ꢀ0.02
<0.6
<0.6
<10
<10
230ꢀ13
<10
52ꢀ11
<10
96
40; 44; 16
[a] Compound stability measured in aqueous buffer at pH 1.0; 4.0; 6.5; 8.0; 10.0. [b] Cytochrome P450 inhibition constants of the isoforms indicated.
[c] logD=intrinsic distribution coefficient between octanol and aqueous buffer (pH 7.4). [d] Membrane permeability (nmsꢁ1) as derived from the parallel
artificial membrane permeability assay (PAMPA). [e] PAMPA retention values (%): in donor compartment; in membrane; in acceptor compartment. [f] Acidi-
ties determined spectrophotometrically at 23ꢀ18C. [g] Solubility (mgLꢁ1) determined by lyophilization solubility assay (LYSA) at pH 6.5. [h] Metabolic sta-
bility; values describe intrinsic clearance (Cl, [mmminꢁ1 mgꢁ1]) in human (h), mouse (m), and rat (r) microsomes. [i] Poor data quality. [j] Precipitation during
assay.
both series, with clearance rates <10 mmminꢁ1 mgꢁ1 in human,
mouse, and rat microsomes (except for 2b in mouse micro-
somes: 18 mmminꢁ1 mgꢁ1). The study confirms slightly higher
stability for the cyclopropyl-CF3 group in comparison with tert-
butyl, as recently claimed for simple test compounds,[9] al-
though attenuated from what would have been expected on
the basis of the model compounds. The largest difference was
found between 1a and 1c in rat microsomes (change from 37
to <10 mmminꢁ1 mgꢁ1). Interestingly, BCP vercirnon 2d is sig-
nificantly more susceptible to metabolism in mouse and rat
microsomes than other analogues.
biological activity. The compounds that form the basis of the
study remained biologically active. In the bosentan series, for
which appropriate IC50 values could be derived, two exhibited
even higher potency than the parent drug. Most importantly,
the derived data on metabolic stability, solubility, logP, pKa,
and permeability render these heretofore atypical substituents
as candidates for the future standard arsenal available to me-
dicinal chemists. Thus, we expect these structures to be con-
sidered with increased frequency in drug discovery programs.
Experimental Section
Commercial chemicals were purchased from ABCR, Acros, Alfa-
Aesar, Apollo Scientific, Combi-Blocks, Fluka, Fluorochem, Merck,
Sigma–Aldrich, or TCI and were used without further purification.
Reactions were magnetically stirred and monitored by thin-layer
chromatography on Merck silica gel 60 F254 TLC glass plates (visual-
ized with UV fluorescence quenching at 254 nm and by KMnO4 or
ceric ammonium nitrate (CAN) stain). Flash column chromatogra-
phy was performed with Fluka silica gel (230–400 mesh, 60 ꢂ) elut-
ing with distilled technical-grade solvents. The yields refer to puri-
fied compounds. NMR data were recorded on Bruker Ascend
(400 MHz), Bruker AV (400, 500, 600 MHz), or Bruker DRX (400, 500,
Conclusions
In summary, we have synthesized a collection of several ana-
logues of bosentan and vercirnon, replacing the tert-butyl
group by CF3, SF5, cyclopropyl-CF3, and bicyclo[1.1.1]pentane.
It is important to note that the last three of these substituents
are relatively new in drug discovery. The study provides full
physicochemical characterization as well as data on bioactivity
across the series, complementing previous studies that largely
include pairwise comparisons or that are limited to reporting
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
5
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
600 MHz) spectrometers at 258C. Chemical shifts (d) are reported
in ppm with the residual solvent signal as internal standard. The
data are reported as s=singlet, d=doublet, t=triplet, m=multip-
let, br=broad signal, coupling constant(s) J in Hz. 13C and 19F NMR
spectra were recorded with broad-band 1H decoupling. Service
measurements were performed by the NMR service team of the
Laboratorium fꢃr Organische Chemie at ETH Zꢃrich by Philipp
Zumbrunnen, Rainer Frankenstein, and Renꢄ Arnold under the di-
rection of Dr. Marc-Olivier Ebert. IR spectra were recorded on a Per-
kinElmer Spectrum Two FT-IR (UATR) instrument as thin films. Ab-
sorptions are given in wavenumbers (cmꢁ1). Mass spectrometric
analyses were performed as high-resolution ESI measurements on
a Bruker Daltonics maXis ESI-QTOF instrument or as high-resolution
EI measurements on a Waters Micromass AutoSpec Ultima instru-
ment by the mass spectrometry service of the Laboratorium fꢃr Or-
ganische Chemie at ETH Zꢃrich by Louis Bertschi, Rolf Hꢀfliger, and
Oswald Greter under the direction of Dr. Xiangyang Zhang.
4-(Bicyclo[1.1.1]pentan-1-yl)-N-(6-(2-hydroxyethoxy)-5-(2-me-
thoxyphenoxy)-[2,2’-bipyrimidin]-4-yl)benzenesulfonamide (1d).
Synthesized according to procedure A from 7 (77 mg, 0.222 mmol)
and 6d (51 mg, 0.228 mmol). After completion of the reaction,
water was added and it was extracted with CH2Cl2. The organic
phase was concentrated, taken up in MeOH and precipitated with
Et2O. Filtration and washing with Et2O afforded the product as col-
orless solid (72 mg, 0.128 mmol, 58%). 1H NMR (400 MHz,
[D6]DMSO): d=9.00 (d, J=4.9 Hz, 2H), 7.94–7.70 (br, 2H), 7.63 (t,
J=4.9 Hz, 1H), 7.19–7.05 (br, 2H), 7.03 (m, 1H), 6.91 m, 1H), 6.75–
6.63 (m, 1H), 6.34 (br, J=9.3 Hz, 1H), 4.86 (br, 1H), 4.32–4.29 (m,
2H), 3.83 (s, 3H), 3.62–3.50 (m, 2H), 2.52 (s, 1H), 2.01 ppm (s, 6H);
13C NMR (101 MHz, [D6]DMSO): d=162.3, 162.3, 160.7, 157.9, 148.4,
146.8, 143.7, 142.8, 127.2, 124.9, 122.6, 121.3, 120.4, 113.3, 113.0,
67.8, 62.8, 59.7, 55.8, 51.6, 46.4, 26.3 ppm; IR (n˜): 3376, 2967, 2871,
1560, 1524, 1500, 1469, 1440, 1400, 1382, 1348, 1294, 1251, 1208,
1177, 1131, 1081, 1026, 887, 837, 804, 740, 697, 641, 610, 571,
424 cmꢁ1; HRMS m/z (ESI) [M+]: calcd (C28H27N5O6S+H+) 562.1755,
found 562.1752.
N-(6-(2-Hydroxyethoxy)-5-(2-methoxyphenoxy)-[2,2’-bipyrimi-
din]-4-yl)-4-(pentafluorosulfanyl)benzenesulfonamide (1b). Pro-
N-(6-(2-Hydroxyethoxy)-5-(2-methoxyphenoxy)-[2,2’-bipyrimi-
din]-4-yl)-4-(trifluoromethyl)benzenesulfonamide (1e). Synthe-
sized according to procedure A from 7 (90 mg, 0.258 mmol) and
6e (59 mg, 0.262 mmol). After completion of the reaction, water
was added and it was extracted with CH2Cl2. The solids in the or-
ganic phase were filtered off, taken up in MeOH and precipitated
with Et2O. Filtration and washing with Et2O afforded the product as
colorless solid (78 mg, 0.138 mmol, 54%). 1H NMR (600 MHz,
CD3OD): d=9.09 (d, J=5.0 Hz, 2H), 7.89 (d, J=8.1 Hz, 2H), 7.62 (t,
J=5.0 Hz, 1H), 7.56 (d, J=8.2 Hz, 2H), 7.02–6.88 (m, 2H), 6.70 (m,
1H), 6.39–6.30 (m, 1H), 4.41 (m, 2H), 3.79 (s, 3H), 3.67–3.59 ppm
(m, 2H); 13C NMR (151 MHz, CD3OD): d=163.2, 162.29, 160.5, 160.1,
155.3, 150.5, 148.9, 148.4, 133.7 (q, J=32 Hz), 128.3, 127.8, 126.3,
125.1 (q, J=272 Hz), 123.7, 123.1, 121.9, 116.3, 114.3, 70.3, 61.9,
56.9 ppm; 19F NMR (471 MHz, CD3OD): d=ꢁ64.41 ppm; IR (n˜):
3497, 3072, 2923, 2853, 1618, 1558, 1501, 1455, 1403, 1349, 1322,
1255, 1191, 1142, 1120, 1107, 1080, 1061, 1027, 1017, 656, 875, 832,
cedure A. 4,6-Dichloro-5-(2-methoxyphenoxy)-2,2’-bipyrimidine
7
(170 mg, 0.49 mmol, 1 equiv), 4-pentafluorosulfanylbenzenesulfo-
namide 6b (145 mg, 0.51 mmol, 1.05 equiv) and potassium carbon-
ate (572 mg, 4.14 mmol) were mixed in MeCN (1.7 mL) and heated
at 808C. After 2 h, LC–MS analysis showed clean formation of sulfo-
namide intermediate. Ethylene glycol (1.7 mL, 30.7 mmol) was
added via syringe and it was stirred at 808C for 19 h. Water was
added, and the pH was adjusted to pH 2 with aq. HCl. Extraction
with EtOAc, washing with brine, drying (Na2SO4), filtration and con-
centration afforded a white solid. NMR showed residual ethylene
glycol. The material was suspended in Et2O. It was sonicated and
filtered (washing with Et2O) to yield the product as colorless solid
1
(113 mg, 0.18 mmol, 37%). H NMR (400 MHz, CD3OD): d=8.95 (dd,
J=5.0, 1.5 Hz, 2H), 8.03 (d, J=8.4 Hz, 2H), 7.83–7.66 (m, 2H), 7.56
(t, J=4.9 Hz, 1H), 7.12–6.84 (m, 2H), 6.80–6.64 (m, 1H), 6.52 (dd,
J=7.9, 1.3 Hz, 1H), 4.56–4.39 (m, 2H), 3.85 (s, 3H), 3.78–3.64 ppm
(m, 2H); 13C NMR (101 MHz, CD3OD): d=163.0, 159.1, 155.8, 150.6,
149.4, 148.5, 129.0, 127.0, 123.6, 122.8, 121.9, 116.1, 114.3, 70.1,
62.2, 56.8 ppm; 19F NMR (282 MHz, CD3OD): d=85.99–79.48 (m),
61.23 ppm (d, J=148.1 Hz); IR (n˜): 3402, 2965, 1561, 1522, 1500,
1470, 1440, 1402, 1381, 1351, 1293, 1252, 1208, 1178, 1135, 1075,
1028, 830, 745, 666, 644, 599, 585, 566 cmꢁ1; HRMS m/z (ESI) [M+]:
calcd (C23H20F5N5O6S2 +H+) 622.0848, found 622.0835.
806, 788, 739, 716, 688, 650, 636, 610, 596, 550, 479, 428, 403 cmꢁ1
;
HRMS m/z (ESI) [M+]: calcd (C24H20F3N5O6S+H+) 564.1159, found
564.1161.
4-(2-(4-(tert-Butyl)phenylsulfonamido)-5-chlorobenzoyl)pyridine
1-oxide (2a). This compound was prepared as described.[33]
N-(6-(2-Hydroxyethoxy)-5-(2-methoxyphenoxy)-[2,2’-bipyrimi-
din]-4-yl)-4-(1-(trifluoromethyl)cyclopropyl)benzenesulfonamide
N-(4-Chloro-2-isonicotinoylphenyl)-4-(pentafluorosulfanyl)benze-
nesulfonamide (2b-SI). Procedure B. 4b (90 mg, 0.30 mmol,
1 equiv) and 8[40] (69 mg, 0.85 mmol, 1 equiv) were mixed in pyri-
dine (1.0 mL) and it was stirred at room temperature overnight.
The reaction mixture was concentrated. Toluene (2 mL) was added,
and it was concentrated again (3ꢅ). Column chromatography
(silica, CH2Cl2 +0.6% MeOH) yielded the product as a light-yellow
wax. Trituration with Et2O or recrystallization from EtOH afforded
the product as colorless solid (66 mg, 45%); mp (EtOH): 1858C.
Note: side product is the doubly sulfonylated product. 1H NMR
(400 MHz, CDCl3): d=10.16 (s, 1H), 8.87–8.74 (m, 2H), 7.90–7.72
(m, 5H), 7.58 (dd, J=8.9, 2.5 Hz, 1H), 7.33 (d, J=2.5 Hz, 1H), 7.23–
7.13 ppm (m, 2H); 13C NMR (101 MHz, CDCl3): d=196.7, 157.1 (m),
150.8, 143.6, 142.1, 137.5, 135.4, 133.0, 130.2, 128.0 (m), 127.3,
125.4, 124.4, 122.2 ppm; 19F NMR (376 MHz, CDCl3): d=82.46–80.42
(m), 62.45 ppm (d, J=150.8 Hz); IR (n˜): 1650, 1497, 1396, 1347,
1328, 1293, 1249, 1173, 1102, 912, 840, 770, 736, 669, 655, 601,
(1c). Synthesized according to procedure A from
7 (63 mg,
0.18 mmol) and 6c (49 mg, 0.185 mmol). Colorless solid (66 mg,
1
0.11 mmol, 60%). H NMR (400 MHz, [D6]DMSO): d=11.52 (br, 1H),
9.09 (m, J=4.7 Hz, 2H), 8.48–8.35 (m, 2H), 7.73–7.56 (m, 3H), 7.14–
6.94 (m, 2H), 6.80 (m, J=7.2 Hz, 1H), 6.71 (m, J=7.8 Hz, 1H), 4.68
(br, 1H), 4.38–4.30 (m, 2H), 3.79 (br, 3H), 3.47 (br, 2H), 1.42–1.32
(m, 2H), 1.19–1.12 ppm (m, 2H); 13C NMR (101 MHz, [D6]DMSO):
d=161.3, 157.9, 155.3, 151.0, 149.0, 145.7, 140.2, 131.0, 129.2,
126.2 (q, J=272 Hz), 123.5, 121.7, 120.5, 115.9, 113.1, 68.4, 59.0,
55.8, 28.9, 27.5 (q, J=32 Hz), 9.8 ppm; 19F NMR (377 MHz,
[D6]DMSO) d=ꢁ68.15 ppm; IR (n˜): 3404, 3328, 2928, 1585, 1557,
1500, 1482, 1443, 1433, 1408, 1386, 1377, 1359, 1359, 1332, 1311,
1297, 1278, 1258, 1246, 1230, 1202, 1183, 1165, 1149, 1123, 1114,
1094, 1079, 1059, 1040, 1030, 1020, 996, 960, 939, 885, 853, 831,
806, 786, 755, 743, 724, 705, 695, 668, 635, 608, 691, 581, 554, 536,
511, 465, 445, 416 cmꢁ1
;
HRMS m/z (ESI) [M+]: calcd
480 cmꢁ1
;
HRMS m/z (ESI) [M+]: calcd (C18H12ClF5N2O3S2 +H+)
(C27H24F3N5O6S+H+) 604.1472, found 604.1472.
498.9971, found 498.9970.
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
6
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
4-(5-Chloro-2-(4-(pentafluorosulfanyl)phenylsulfonamido)ben-
zoyl)pyridine 1-oxide (2b). Procedure C. 2b-SI (88 mg, 0.176 mmol,
1.0 equiv) and mCPBA (52.2 mg, 0.212 mmol, 1.2 equiv) were
mixed in CH2Cl2 (1 mL) and it was stirred at room temperature
until TLC showed complete conversion. The mixture was concen-
trated and subjected to column chromatography (silica, CH2Cl2 +
2% MeOH) to yield 2b (83 mg, 0.182 mmol, 91%) as a light-yellow
solid; mp (EtOH): 2258C; Rf (CH2Cl2 +3% MeOH): 0.17. It can be re-
crystallized from EtOH as colorless solid. 1H NMR (400 MHz,
[D6]DMSO): d=10.43 (s, 1H), 8.30 (d, J=6.6 Hz, 2H), 8.08 (d, J=
8.5 Hz, 2H), 7.79 (d, J=8.5 Hz, 2H), 7.71–7.40 (m, 4H), 7.03 ppm (d,
J=8.4 Hz, 1H); 13C NMR (101 MHz, [D6]DMSO): d=189.6, 155.2 (m),
142.7, 139.1, 135.5, 132.9, 132.2, 131.4, 131.3, 129.8, 128.3, 128.0,
127.2, 126.8 ppm; 19F NMR (282 MHz, [D6]DMSO): d=84.89 (m, 1F),
63.49 ppm (d, J=151.4 Hz, 4F); IR (n˜): 2658, 1683, 1610, 1446, 1395,
1351, 1286, 1234, 1178, 1169, 1154, 1101, 1033, 967, 879, 836, 785,
1H NMR (400 MHz, CDCl3): d=10.00 (s, 1H), 8.81–8.75 (m, 2H), 7.78
(d, J=8.9 Hz, 1H), 7.67–7.60 (m, 2H), 7.53 (dd, J=8.9, 2.5 Hz, 1H),
7.29 (d, J=2.4 Hz, 1H), 7.23–7.13 (m, 4H), 2.56 (s, 1H), 2.03 ppm (s,
6H); 13C NMR (101 MHz, CDCl3): d=196.3, 150.5, 147.4, 143.8,
138.2, 136.4, 134.9, 132.5, 129.3, 127.2, 126.9, 125.5, 124.4, 122.4,
52.2, 46.5, 27.0 ppm; IR (n˜): 2968, 2910, 2872, 1648, 1596, 1566,
1477, 1407, 1386, 1340, 1327, 1292, 1247, 1209, 1182, 1165, 1116,
1092, 1063, 1017, 958, 597, 836, 769, 743, 721, 682, 653, 594, 567,
525, 476 cmꢁ1; HRMS m/z (ESI) [M+]: calcd (C23H19ClN2O3S+H+)
439.0878, found 439.0879.
4-(2-(4-(Bicyclo[1.1.1]pentan-1-yl)phenylsulfonamido)-5-chloro-
benzoyl)pyridine 1-oxide (2d). Prepared according to procedure C
from 2d-SI (40 mg, 0.091 mmol, 1.0 equiv) and mCPBA (29.2 mg,
0.118 mmol, 1.3 equiv) in CH2Cl2 (1 mL) to yield 2d (41 mg,
0.090 mmol, 99%) as pale-yellow solid; mp (EtOH): 2098C; Rf
(CH2Cl2 +3% MeOH): 0.16. It can be recrystallized from EtOH as col-
735, 665, 626, 610, 600, 585, 567, 535, 520, 496, 483, 467 cmꢁ1
;
1
HRMS m/z (ESI) [M+]: calcd (C18H12ClF5N2O4S2 +H+) 514.9920, found
orless solid. H NMR (400 MHz, CDCl3): d=9.39 (s, 1H), 8.16 (d, J=
6.6 Hz, 2H), 7.74 (d, J=8.9 Hz, 1H), 7.59–7.51 (m, 3H), 7.37–7.29
(m, 3H), 7.14–7.09 (m, 2H), 2.56 (s, 1H), 2.01 ppm (s, 6H); 13C NMR
(101 MHz, CDCl3): d=192.1, 147.5, 139.5, 137.2, 136.4, 134.5, 131.8,
131.0, 130.0, 127.3, 127.0, 126.9, 126.7, 125.9, 77.48, 52.3, 46.5,
27.1 ppm; IR (n˜): 2972, 2872, 1648, 1607, 1477, 1443, 1388, 1338,
1282, 1247, 1210, 1166, 1092, 906, 838, 736, 683, 626, 610, 568,
514.9920.
N-(4-Chloro-2-isonicotinoylphenyl)-4-(1-(trifluoromethyl)cyclo-
propyl)benzenesulfonamide (2c-SI). Prepared according to proce-
dure B from 8 (70 mg, 0.30 mmol, 1 equiv) 4c (110 mg, 0.39 mmol,
1.3 equiv) in pyridine (1 mL). Purified twice on silica (CH2Cl2 +1.8%
MeOH+1.2% HCOOH; then hexane/EtOAc 6:4). Isolated as
a yellow foam (145 mg, 99%). It could be precipitated from cold
toluene/heptane as a pale solid; mp (toluene/heptane): 1398C.
1H NMR (400 MHz, CDCl3): d=10.17 (s, 1H), 8.79–8.71 (m, 2H), 7.77
(d, J=8.9 Hz, 1H), 7.75–7.68 (m, 2H), 7.52 (dd, J=8.9, 2.5 Hz, 1H),
7.49–7.43 (m, 2H), 7.30 (d, J=2.5 Hz, 1H), 7.22–7.18 (m, 2H), 1.39–
1.33 (m, 2H), 0.96–0.90 ppm (m, J=1.7 Hz, 2H); 13C NMR (101 MHz,
CDCl3): d=196.7, 150.6, 143.9, 141.9, 138.9, 138.2, 135.2, 132.9,
132.0, 129.3, 127.3, 125.8 (q, J=273.9 Hz), 124.9, 123.8, 122.2, 28.1
(q, J=33.8 Hz), 10.2 ppm; 19F NMR (377 MHz, CDCl3): d=
ꢁ69.63 ppm; IR (n˜): 1648, 1598, 1567, 1478, 1408, 1388, 1359, 1328,
1296, 1247, 1169, 1154, 1133, 1095, 1078, 1040, 1019, 958, 906, 828,
770, 753, 730, 695, 673, 654, 590, 579, 528, 477, 439 cmꢁ1; HRMS
m/z (ESI) [M+]: calcd (C22H16ClF3N2O3S+H+) 481.0595, found
481.0596.
519 cmꢁ1
;
HRMS m/z (ESI) [M+]: calcd (C23H19ClN2O4S+H+)
455.0827, found 455.0827.
N-(4-Chloro-2-isonicotinoylphenyl)-4-(trifluoromethyl)benzene-
sulfonamide (2e-SI). Prepared according to procedure B from 8
(70 mg, 0.30 mmol, 1 equiv) and 4-(trifluoromethyl)benzene-1-sul-
fonyl chloride (88 mg, 0.36 mmol, 1.2 equiv) in pyridine (1 mL). Pu-
rified twice on silica (CH2Cl2 +1.8% MeOH+1.2% HCOOH; then
hexane/EtOAc 6:4) to yield a colorless solid (73 mg, 55%); mp
(EtOH): 1968C. It can be recrystallized from EtOH. 1H NMR
(400 MHz, CDCl3): d=10.15 (s, 1H), 8.84–8.73 (m, 2H), 7.92–7.85
(m, 2H), 7.80 (d, J=8.9 Hz, 1H), 7.64 (dt, J=7.3, 1.0 Hz, 2H), 7.57
(dd, J=8.9, 2.4 Hz, 1H), 7.32 (d, J=2.4 Hz, 1H), 7.21–7.16 ppm (m,
2H); 13C NMR (101 MHz, CDCl3): d=196.7, 150.8, 143.6, 142.4,
137.7, 135.3, 135.2 (q, J=33.5 Hz), 133.0, 130.0, 128.0, 126.5, 125.3,
124.2, 123.0 (q, J=273.4 Hz), 122.2 ppm; 19F NMR (377 MHz, CDCl3):
d=ꢁ63.26 ppm; IR (n˜): 2643, 1688, 1601, 1559, 1476, 1414, 1404,
1344, 1324, 1277, 1252, 1226, 1191, 1158, 1110, 1093, 1061, 1017,
1007, 965, 922, 893, 860, 836, 812, 783, 729, 710, 698, 669, 648,
4-(5-Chloro-2-(4-(1-(trifluoromethyl)cyclopropyl)phenylsulfona-
mido)benzoyl)pyridine 1-oxide (2c). Prepared according to proce-
dure C from mCPBA (30 mg, 0.122 mmol) and 2c-SI (47 mg,
0.098 mmol) in CH2Cl2 (1 mL). Column chromatography (silica,
CH2Cl2 +2% MeOH) yielded 2c (47 mg, 0.095 mmol, 97%) as
a pale-yellow solid. It can be recrystallized from EtOH as colorless
solid; mp (EtOH): 2118C; Rf (CH2Cl2 +3%MeOH): 0.15. 1H NMR
(400 MHz, CDCl3): d=9.77 (s, 1H), 8.24–8.16 (m, 2H), 7.76 (d, J=
8.8 Hz, 1H), 7.73–7.67 (m, 2H), 7.55 (dd, J=8.9, 2.5 Hz, 1H), 7.51–
7.41 (m, 2H), 7.41–7.34 (m, 3H), 1.44–1.35 (m, 2H), 1.00–0.92 ppm
(m, J=1.7 Hz, 2H); 13C NMR (101 MHz, CDCl3): d=192.7, 142.0,
139.6, 139.0, 137.5, 134.8, 132.1, 132.0, 131.7, 129.9, 127.4, 126.6,
126.0, 125.9 (q, J=273.9 Hz) 124.7, 28.2 (q, J=33.8 Hz), 10.3 ppm;
19F NMR (376 MHz, CDCl3): d=ꢁ69.65 ppm; IR (n˜): 2682, 1682,
1610, 1446, 1342, 1287, 1235, 1153, 1131, 1080, 967, 878, 841, 826,
783, 711, 673, 625, 607, 593, 577, 519, 495, 484, 461.6 cmꢁ1; HRMS
m/z (ESI) [M+]: calcd (C22H16ClF3N2O4S+H+) 497.0544, found
497.0544.
597, 557, 514, 493, 466 cmꢁ1
;
HRMS m/z (ESI) [M+]: calcd
(C19H12ClF3N2O3S+H+) 441.0282, found 441.0282.
4-(5-Chloro-2-(4-(trifluoromethyl)phenylsulfonamido)benzoyl)-
pyridine 1-oxide (2e). Synthesized according to procedure C from
2e-SI (39 mg, 0.088 mmol, 1.0 equiv) and mCPBA (26 mg,
0.106 mmol, 1.2 equiv) in CH2Cl2 (0.3 mL) to give 2e (39 mg,
0.085 mmol, 96%) as pale-yellow solid; mp (EtOH): 2158C; Rf
(CH2Cl2 +3% MeOH): 0.13. It can be recrystallized from EtOH as col-
orless solid. 1H NMR (400 MHz, CDCl3): d=9.69 (s, 1H), 8.24–8.16
(m, 2H), 7.90–7.82 (m, 2H), 7.74 (d, J=8.8 Hz, 1H), 7.66–7.59 (m,
2H), 7.57 (dd, J=8.8, 2.4 Hz, 1H), 7.41–7.33 ppm (m, 3H); 13C NMR
(101 MHz, CDCl3): d=192.3, 142.4, 139.6, 136.9, 135.0, 134.8, 131.7,
131.6, 130.5, 128.0, 126.6, 126.5, 126.5, 125.2 ppm; 19F NMR
(376 MHz, CDCl3): d=ꢁ63.25 ppm; IR (n˜): 2660, 1679, 1607, 1481,
1442, 1404, 1340, 1323, 1304, 1286, 1243, 1166, 1154, 1108, 1091,
1061, 1016, 966, 876, 847, 830, 785, 745, 716, 708, 678, 668, 612,
604, 556, 512, 490, 481, 463 cmꢁ1; HRMS m/z (ESI) [M+]: calcd
(C19H12ClF3N2O4S+H+) 457.0231, found 457.0233.
4-(Bicyclo[1.1.1]pentan-1-yl)-N-(4-chloro-2-isonicotinoylphenyl)-
benzenesulfonamide (2d-SI). Prepared according to procedure B
from 8 (70 mg, 0.30 mmol, 1.0 equiv) and 6d (102 mg, 0.42 mmol,
1.4 equiv) in pyridine (1 mL). Purified twice on silica (CH2Cl2 +1.8%
MeOH+1.2% HCOOH; then hexane/EtOAc 6:4). Isolated as
a yellow foam (127 mg, 96%); Rf (CH2Cl2 +3% MeOH): 0.17.
4-(Trifluoromethyl)benzenediazonium tetrafluoroborate (3). 4-
Pentafluorosulfanylaniline hydrochloride (2 g, 7.8 mmol, 1.0 equiv)
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
7
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
was dissolved in a mixture of EtOH (78 mL) and HBF4 (48% in
water, 3.11 mL, 23.5 mmol) and cooled to 08C (ice bath). Then iso-
amyl nitrite (1.8 mL, 13.3 mmol) was added slowly and it was
stirred for 1 h. Et2O (5ꢅ volume, ~400 mL) was added to precipi-
tate the product (stirring for ~25 min at 08C). The product was iso-
lated by filtration (leaving the mother liquor at 48C gives a second
crop of product). The filter cake was dissolved in a minimum
amount of acetone, and the diazonium salt was precipitated by
the addition of Et2O. Filtration afforded the product as colorless
solid (1501 mg+380 mg (second crop), 5.9 mmol, 76%). 1H NMR
(400 MHz, CD3CN): d=8.73 (appd, J=9.2 Hz, 2H), 8.47–8.27 ppm
(m, 2H); 13C NMR (101 MHz, CD3CN): d=161.7 (m), 134.9, 130.7 (m),
120.3 ppm; 19F NMR (377 MHz, CD3CN): d=78.59–77.00 (m, 1F),
60.92 (d, J=149.6 Hz, 4F), ꢁ150.85 ppm (s, 4F); IR (n˜): 3119, 2307,
4-(Pentafluorosulfanyl)benzenesulfonamide (6b). Compound 4b
(68 mg, 0.23 mmol) was dissolved in 7m NH3 in MeOH (1 mL) and
stirred for 30 min at 08C. The solvent was evaporated and the resi-
due purified on silica (hexane/EtOAc 7:3–6:4) to yield the product
as colorless solid (50 mg, 0.17 mmol, 79%); Rf (hexane/EtOAc 6:4):
1
0.45. H NMR (400 MHz, CD3CN): d=8.11–7.91 (m, 4H), 5.86 ppm (s,
2H); 13C NMR (101 MHz, CD3CN): d=156.5 (m), 147.6, 128.1,
128.0 ppm (m); 19F NMR (377 MHz, CD3CN): d=83.48–80.95 (m),
61.86 ppm (d, J=148.2 Hz); IR (n˜): 3378, 6260, 3108, 1562, 1399,
1332, 1312, 1291, 1161, 1104, 935, 881, 820, 732, 668, 598, 582,
551, 504, 428 cmꢁ1; HRMS m/z (ESI) [M+]: calcd [M+Na+] 288.0277,
found 288.0283.
4-(1-(Trifluoromethyl)cyclopropyl)benzenesulfonamide
(6c).
Compound 4c (183 mg, 0.64 mmol) was suspended in a 25% solu-
tion of NH3 in water (3 mL) and stirred for 2 h at room tempera-
ture. Extraction with EtOAc, drying (MgSO4), filtration and concen-
tration followed by flash chromatography (silica, hexane/EtOAc
8:2–7:3) yielded the product as colorless solid (150 mg, 0.57 mmol,
1575, 1420, 1303, 1042, 838, 755, 675, 606, 579, 546, 520 cmꢁ1
;
HRMS m/z (ESI) [M+]: calcd (F5SPhN2+) 231.0010, found 231.0021.
4-(Pentafluorosulfanyl)benzene-1-sulfonyl chloride (4b). Thionyl
chloride (310 mL, 4.25 mmol) was added dropwise to water (1.7 mL)
at 08C (ice bath). After 10 min it was allowed to stir at room tem-
perature for 40 min. After cooling to 08C, conc. HCl (430 mL) was
added followed by CuCl (2 mg, 0.02 mmol) and it was stirred for
5 min. Then a solution of 3 (270 mg, 0.85 mmol) in MeCN (300 mL)
was added dropwise (rinsing the syringe with additional 100 mL)
and it was stirred at 08C for 1 h. The solution turned yellow and
a precipitate formed. Filtration afforded a yellow solid (175 mg,
68%) which can be used as such or further purified by flash chro-
matography (silica, hexanes) to give a colorless solid (101 mg,
1
88%); Rf (hexane/EtOAc 6:4): 0.40. H NMR (400 MHz, CD3CN): d=
7.96–7.77 (m, 2H), 7.74–7.60 (m, 2H), 5.69 (s, 2H), 1.50–1.35 (m,
2H), 1.22–1.11 ppm (m, J=1.7 Hz, 2H); 13C NMR (101 MHz, CD3CN):
d=144.4, 141.3, 133.0, 127.4 (q, J=272 Hz), 127.1, 28.7 (q, J=
33 Hz), 10.6 ppm (q, J=2 Hz); 19F NMR (376 MHz, CD3CN): d=
ꢁ70.24 ppm; IR (n˜): 3358, 3286, 3255, 1599, 1531, 1496, 1423, 1399,
1361, 1332, 1317, 1285, 1163, 1150, 1135, 1124, 1097, 1076, 1037,
1019, 948, 914, 849, 828, 823, 781, 752, 732, 671, 600, 578, 548,
496, 459, 436, 426, 404 cmꢁ1
;
HRMS m/z (ESI) [M+]: calcd
(C10H10F3NO2S+H+) 288.0277, found 288.0283.
1
40%); Rf (hexane): 0.18. H NMR (400 MHz, CDCl3): d=8.21–8.14 (m,
2H), 8.08–8.01 ppm (m, 2H); 13C NMR (101 MHz, CDCl3): d=
158.90–157.73 (m), 146.67, 128.41–127.57 ppm (m); 19F NMR
(377 MHz, CDCl3): d=81.49–79.57 (m), 62.37 ppm (d, J=150.1 Hz);
IR (n˜): 1403, 1382, 1299, 1201, 1181, 1097, 831, 821, 731, 665, 598,
584, 573, 560, 524, 456, 421 cmꢁ1; HRMS m/z (EI) [M+]: calcd
(C6H4ClF5O2S2) 301.9261, found 301.9256.
4-(Bicyclo[1.1.1]pentan-1-yl)benzenesulfonamide (6d). Com-
pound 4d (276 mg, 1.14 mmol) was suspended in 25% aq. NH3
(3 mL) and stirred for 5 h at room temperature. Extraction with
EtOAc, drying (MgSO4), filtration and concentration followed by
flash chromatography (silica, hexane/EtOAc 8:2–5:5) yielded the
product as colorless solid (200 mg, 0.90 mmol, 79%); Rf (hexane/
1
EtOAc 7:3): 0.27. H NMR (400 MHz, CD3CN): d=7.93–7.66 (m, 2H),
4-(1-(Trifluoromethyl)cyclopropyl)benzene-1-sulfonyl
chloride
7.48–7.26 (m, 2H), 5.61 (s, 3H), 2.56 (s, 1H), 2.12 ppm (s, 6H);
13C NMR (101 MHz, CD3CN): d=147.0, 142.3, 127.5, 127.0, 52.8,
47.3, 27.6 ppm; IR (n˜): 3346, 3268, 2961, 2910, 2875, 1598, 1538,
1450, 1403, 1326, 1286, 1213, 1190, 1172, 1096, 1065, 1018, 958,
(4c). 1-(Trifluoromethyl)cyclopropyl)benzene (purchased from
Apollo Scientific; 431 mg, 2.3 mmol, 1 equiv) was dissolved in
CH2Cl2 (3.3 mL) and cooled to ꢁ5–08C using a salt/ice bath. Chlor-
osulfonic acid (388 mL, 5.8 mmol, 2.5 equiv) was then added drop-
wise. It was stirred overnight at room temperature. The mixture
was poured onto ice, followed by extraction with CH2Cl2. The com-
bined organics were washed with satd NaHCO3 and brine. Drying
(MgSO4), filtration and concentration followed by flash chromatog-
915, 885, 834, 776, 741, 666, 631, 606, 560, 541, 458, 426, 407 cmꢁ1
;
HRMS m/z (ESI) [M+Na+]: calcd (C11H13NO2S+Na+) 246.0559,
found 246.0565.
raphy (silica, hexane/EtOAc 9:1) yielded
3 as colorless solid
(572 mg, 2.0 mmol, 87%); Rf (hexane): 0.13. 1H NMR (400 MHz,
CDCl3): d=8.11–7.95 (m, 2H), 7.79–7.65 (m, 2H), 1.53–1.44 (m, 2H),
1.15–1.07 ppm (m, 2H); 13C NMR (101 MHz, CDCl3): d=144.1, 143.9,
132.5, 127.1, 126.7 (q, J=274 Hz), 28.3 (q, J=34 Hz), 10.2 ppm (q,
J=2 Hz); 19F NMR (376 MHz, CDCl3): d=ꢁ69.62 ppm; IR (n˜): 1596,
1402, 1378, 1360, 1333, 1290, 1192, 1168, 1144, 1128, 1087, 1073,
1045, 1016, 949, 827, 779, 751, 667, 575, 551, 509, 437 cmꢁ1; HRMS
m/z (EI) [M+]: calcd (C10H8ClF3O2S+) 283.9886, found 283.9881.
Acknowledgements
We thank Cꢂlia Mueller Grandjean (Actelion Pharmaceuticals
Ltd.) for technical assistance with measurement of the bosentan
analogues. SpiroChem is acknowledged for the generous gift of
compound 4d. Financial support from ETH Zꢁrich (ETH 26 13-1)
is acknowledged. We are also grateful for support from F. Hoff-
mann-La Roche.
4-(Bicyclo[1.1.1]pentan-1-yl)benzene-1-sulfonyl chloride (4d).
1
Provided by SpiroChem. Colorless solid; Rf (hexane): 0.20. H NMR
Keywords: drug analogues · isosterism · metabolic stability ·
pentafluorosulfanyl · structure–property relationships
(400 MHz, CDCl3): d=7.95 (app d, J=8.5 Hz, 2H), 7.42 (app d, J=
8.5 Hz, 2H), 2.62 (s, 1H), 2.15 ppm (s, 8H); 13C NMR (101 MHz,
CDCl3): d=149.7, 142.4, 127.4, 127.1, 46.7, 27.2 ppm; IR (n˜): 2984,
2916, 2879, 1589, 1400, 1377, 1363, 1325, 1293, 1210, 1186, 1164,
1143, 1101, 1081, 1059, 1011, 966, 944, 863, 842, 832, 775, 738, 718,
577, 557, 543, 504, 421 cmꢁ1; HRMS m/z (EI) [M+]: calcd ([MꢁH]+)
241.0085, found 241.0088, calcd ([MꢁCl]+) 207.0475, found
207.0478.
[1] V. A. Cullum, J. B. Farmer, D. Jack, G. P. Levy, Br. J. Pharmacol. 1969, 35,
141–151.
[2] J. P. Clozel, W. Fischli, Arzneim.-Forsch. 1993, 43, 260–262.
[3] P. Rafferty, S. T. Holgate, Am. Rev. Respir. Dis. 1987, 135, 181–184.
[4] D. A. Smith, B. C. Jones, D. K. Walker, Med. Res. Rev. 1996, 16, 243–266.
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
8
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
[5] P. Wipf, T. Mo, S. J. Geib, D. Caridha, G. S. Dow, L. Gerena, N. Roncal, E. E.
Milner, Org. Biomol. Chem. 2009, 7, 4163–4165.
Matthews, F. S. Buckner, S. A. Charman, M. A. Phillips, P. K. Rathod, J.
Med. Chem. 2011, 54, 3935–3949.
[6] S. Altomonte, G. L. Baillie, R. A. Ross, J. Riley, M. Zanda, RSC Adv. 2014, 4,
20164–20176.
[7] K. D. Bunker, N. W. Sach, Q. Huang, P. F. Richardson, Org. Lett. 2011, 13,
4746–4748.
[8] A. F. Stepan et al., J. Med. Chem. 2012, 55, 3414–3424.
[9] D. Barnes-Seeman, M. Jain, L. Bell, S. Ferreira, S. Cohen, X.-H. Chen, J.
Amin, B. Snodgrass, P. Hatsis, ACS Med. Chem. Lett. 2013, 4, 514–516.
[10] G. Wuitschik, M. Rogers-Evans, K. Mꢃller, H. Fischer, B. Wagner, F. Schu-
ler, L. Polonchuk, E. M. Carreira, Angew. Chem. Int. Ed. 2006, 45, 7736–
7739; Angew. Chem. 2006, 118, 7900–7903.
[11] J. A. Burkhard, G. Wuitschik, M. Rogers-Evans, K. Mꢃller, E. M. Carreira,
Angew. Chem. Int. Ed. 2010, 49, 9052–9067; Angew. Chem. 2010, 122,
9236–9251.
[24] J. M. Coteron et al., J. Med. Chem. 2011, 54, 5540–5561.
[25] F. Micheli, D. Andreotti, S. Braggio, A. Checchia, Bioorg. Med. Chem. Lett.
2010, 20, 4566–4568.
[26] S. E. Lazerwith et al., J. Med. Chem. 2014, 57, 1893–1901.
[27] R. Pellicciari, M. Raimondo, M. Marinozzi, B. Natalini, G. Costantino, C.
Thomsen, J. Med. Chem. 1996, 39, 2874–2876.
[28] M. R. Barbachyn, D. K. Hutchinson, D. S. Toops, R. J. Reid, G. E. Zurenko,
B. H. Yagi, R. D. Schaadt, J. W. Allison, Bioorg. Med. Chem. Lett. 1993, 3,
671–676.
[29] L. Zehnder et al., J. Med. Chem. 2011, 54, 3368–3385.
[30] L. J. Rubin et al., N. Engl. J. Med. 2002, 346, 896–903.
[31] K. O. Arseneau, F. Cominelli, Expert Opin. Invest. Drugs 2013, 22, 907–
913.
[12] J. A. Burkhard, C. Guꢄrot, H. Knust, M. Rogers-Evans, E. M. Carreira, Org.
Lett. 2010, 12, 1944–1947.
[32] J. T. Reeves, K. Camara, Z. S. Han, Y. Xu, H. Lee, C. A. Busacca, C. H. Sena-
nayake, Org. Lett. 2014, 16, 1196–1199.
[13] J. A. Burkhard, B. Wagner, H. Fischer, F. Schuler, K. Mꢃller, E. M. Carreira,
Angew. Chem. Int. Ed. 2010, 49, 3524–3527; Angew. Chem. 2010, 122,
3603–3606.
[33] Calculated with the online property calculation service, Molinspiration:
www.molinspiration.com.
[34] T. Okazaki, K. K. Laali, S. D. Bunge, S. K. Adas, Eur. J. Org. Chem. 2014,
1630–1644.
[14] A. Joliton, E. M. Carreira, Org. Lett. 2013, 15, 5147–5149.
´ˇ
[15] T. Pastyrꢆkovꢇ, G. Iakobson, N. Vida, R. Pohl, P. Beier, Eur. J. Org. Chem.
[35] P. J. Hogan, B. G. Cox, Org. Process Res. Dev. 2009, 13, 875–879.
[36] N. C. Niphade, K. M. Jagtap, C. T. Gaikawad, M. N. Jachak, V. T. Mathad,
Org. Process Res. Dev. 2011, 15, 1382–1387.
2012, 2123–2126.
´ˇ
[16] P. Beier, T. Pastyrꢆkovꢇ, N. Vida, G. Iakobson, Org. Lett. 2011, 13, 1466–
1469.
[37] S. Ungashe, J. J. Wright, A. Pennell, Z. Wei, A. Melikan (Chemocentryx
Inc.), US Pat. No. US7420055 B2, 2008.
[38] W. Neidhart, V. Breu, D. Bur, K. Burri, M. Clozel, G. Hirth, M. Mꢃller, H. P.
Wessel, H. Ramuz, Chimia 1996, 50, 519–524.
[39] For a general overview of Vercirnon’s SAR, see: A. Pennell, S. Ugashe,
J. J. Wright, W. Zheng (Chemocentryx Inc.), Int. PCT Pub. No.
WO2004046092A2, 2004.
[17] O. S. Kanishchev, W. R. Dolbier, Jr., Angew. Chem. Int. Ed. 2015, 54, 280–
284; Angew. Chem. 2015, 127, 282–286.
[18] W. A. Sheppard, J. Am. Chem. Soc. 1960, 82, 4751–4752.
[19] V. Gouverneur, K. Mꢃller, Fluorine in Pharmaceutical and Medicinal
Chemistry, Imperial College Press, London, 2012.
[20] B. Stump, C. Eberle, W. B. Schweizer, M. Kaiser, R. Brun, R. L. Krauth-
Siegel, D. Lentz, F. Diederich, ChemBioChem 2009, 10, 79–83.
[21] J. T. Welch, D. S. Lim, Bioorg. Med. Chem. 2007, 15, 6659–6666.
[22] P. W. Chia, S. C. Brennan, A. M. Z. Slawin, D. Riccardi, D. O’Hagan, Org.
Biomol. Chem. 2012, 10, 7922–7927.
[40] J. V. Earley, N. W. Gilman, Synth. Commun. 1985, 15, 1271–1276.
[23] R. Gujjar, F. El Mazouni, K. L. White, J. White, S. Creason, D. M. Shackle-
ford, X. Deng, W. N. Charman, I. Bathurst, J. Burrows, D. M. Floyd, D.
Received: November 24, 2014
Published online on && &&, 0000
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
9
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

FULL PAPERS
M. V. Westphal, B. T. Wolfstꢀdter,
The alternative scene: Given how
common the tert-butyl group is in me-
dicinal chemistry, we set out to explore
some isosteric alternatives in hopes of
identifying motifs that circumvent the
common drawbacks imparted by tert-
butyl groups, such as increased lipophi-
licity and lower metabolic stability.
J.-M. Plancher, J. Gatfield, E. M. Carreira*
&& – &&
Evaluation of tert-Butyl Isosteres: Case
Studies of Physicochemical and
Pharmacokinetic Properties, Efficacies,
and Activities
&
ChemMedChem 0000, 00, 0 – 0
www.chemmedchem.org
10
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!