A new series of bicalutamide, enzalutamide and enobosarm derivatives carrying pentafluorosulfanyl (SF5) and pentafluoroethyl (C2F5) substituents: Improved antiproliferative agents against prostate cancer

Article abstract of DOI:10.1016/j.ejmech.2019.07.001

SAR studies on bicalutamide, enobosarm and enzalutamide analogues, functionalised with polyfluorinated groups, is presented. Among the novel bicalutamide and enobosarm derivatives synthesised, several displayed significantly improved in vitro anticancer activity, with IC₅₀ values in the low micromolar range against four different prostate cancer cell lines (LNCaP, VCaP, DU-145 and 22Rv1), showing up to 48-fold increase in comparison with the parent structures. In particular, SF₅ enobosarm analogues were found to be most potent compounds, full AR antagonists and with favourable ADME properties. The most promising compound (48a) was evaluated for its in vivo efficacy in PC xenograft mouse model (22Rv1) with results comparable to the standard-of-care docetaxel.

Full text of DOI:10.1016/j.ejmech.2019.07.001

Accepted Manuscript
A new series of bicalutamide, enzalutamide and enobosarm derivatives carrying
pentafluorosulfanyl (SF ) and pentafluoroethyl (C F ) substituents: Improved
5
2 5
antiproliferative agents against prostate cancer
Fabrizio Pertusati, Salvatore Ferla, Marcella Bassetto, Andrea Brancale, Sahar
Khandil, Andrew D. Westwell, Christopher McGuigan
PII:
S0223-5234(19)30622-1
DOI:
https://doi.org/10.1016/j.ejmech.2019.07.001
EJMECH 11498
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 4 April 2019
Revised Date: 30 June 2019
Accepted Date: 1 July 2019
Please cite this article as: F. Pertusati, S. Ferla, M. Bassetto, A. Brancale, S. Khandil, A.D. Westwell,
C. McGuigan, A new series of bicalutamide, enzalutamide and enobosarm derivatives carrying
pentafluorosulfanyl (SF ) and pentafluoroethyl (C F ) substituents: Improved antiproliferative agents
2 5
5
against prostate cancer, European Journal of Medicinal Chemistry (2019), doi: https://doi.org/10.1016/
j.ejmech.2019.07.001.
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.

ACCEPTED MANUSCRIPT
A new series of bicalutamide, enzalutamide and enobosarm derivatives carrying
pentafluorosulfanyl (SF ) and pentafluoroethyl (C F ) substituents: improved
5
2 5
antiproliferative agents against prostate cancer.
‡
‡
‡
Fabrizio Pertusati,* Salvatore Ferla , Marcella Bassetto , Andrea Brancale, Sahar Khandil,
Andrew D. Westwell and Christopher McGuigan
School of Pharmacy and Pharmaceutical Sciences, Redwood Building, King Edwards VII Avenue, CF10
3
NB, Cardiff, Wales, UK.
Keywords: pentafluorosulfanyl, pentafluoroethyl, prostate cancer, androgen receptor,
antiproliferative activity, enobosarm, bicalutamide, enzalutamide.
Abstract: SAR studies on bicalutamide, enobosarm and enzalutamide analogues, functionalised
with polyfluorinated groups, is presented. Among the novel bicalutamide and enobosarm
derivatives synthesised, several displayed significantly improved in vitro anticancer activity, with
IC50 values in the low micromolar range against four different prostate cancer cell lines (LNCaP,
VCaP, DU-145 and 22Rv1), showing up to 48-fold increase in comparison with the parent
structures. In particular, SF enobosarm analogues were found to be most potent compounds,
5
full AR antagonists and with favourable ADME properties. The most promising compound (48a)
was evaluated for its in vivo efficacy in PC xenograft mouse model (22Rv1) with results
comparable to the standard-of-care docetaxel.
Introduction
In the recent years, incorporation of fluorine and fluorine-containing groups into organic
molecules to perturb the chemical, physical and biological properties of the parent compounds
1-6
has proved to be a very successful strategy in drug design.
Fluorine atom in organic
molecules, can impart a unique set of properties due to the distinctive combination of
7-9
electronegativity, size and lipophilicity. These factors combined have a substantial impact on
molecular conformation of the drug, which may affect the binding affinity to the target protein.
For these reasons, organo-fluorine structures have received and they are still receiving
1

ACCEPTED MANUSCRIPT
1
0-15
increasing attention in medicinal chemistry,
and currently, fluorine is found in about 20% of
13
16-18
19-22
23
pharmaceuticals. The trifluoromethyl,
difluoromethyl
and fluoromethyl groups (in
24
decreasing order) are the most exploited fluorinated groups in medicinal chemistry. In
contrast, polyfluoroalkyl groups have been little exploited in medicinal chemistry compared to
25, 26
the above groups.
However, recent years have witnessed an increased use of higher
polyfluorinated groups in drug candidate molecules. Examples of these are the pentafluoroethyl
27
(
C F ) and the pentafluorosulfanyl groups (SF ). The C F group shows a broad synthetic
2 5 5 2 5
access, whereas incorporation of SF , group is more challenging. The intrinsic difficulty of its
5
introduction into a variety of substrates due to the harsh conditions needed has often precluded
the investigation of its use in medicinal chemistry. Nevertheless, the last few years have
witnessed an increased use of this functional group, especially in material chemistry, due to the
2
8, 29
appearance of novel and safe synthetic methodologies.
blocks functionalised with the SF5 group are commercially available making the
Nowadays more and more building
30-34
pentafluorosulfanyl group increasingly used even among medicinal chemists.
on drugs functionalised with SF and C have showed an improvement in their physiological
and / or biological properties compared to the parent drugs.
agent fulvestrant (Faslodex®, Figure.1), is the only approved drug containing a C F group.
A few reports
5
2 5
F
34-40
The breast cancer therapeutic
2
5
Studies on its mode of action have shown that its unique C-7 side chain with a terminal C2F5
4
1
group is responsible for the increased metabolic stability during the receptor binding. In 2017,
the first clinical candidate containing a SF group (DSM265) reached Phase IIa trials for the
5
42
treatment of malaria (Figure.1). During these studies, an aniline ring with a 4-SF motif
5
provided the optimal balance between parasite activity, metabolic stability, and good
43
pharmacokinetic and safety profiles in patients.
Figure 1: Structure of the anticancer drug Fulvestrant and the antimalarial agent DSM265.
2

ACCEPTED MANUSCRIPT
Introduction of fluorine and polyfluorinated units such as the trifluoromethyl group has been a
successful strategy in the development of efficacious drugs for the treatment of Prostate Cancer
44
(
PC) (Figure 2). PC is one of the major causes of death of men in worldwide. The initial
treatment of PC is often androgen deprivation therapy (ADT). Unfortunately, after an initial
response to ADT, the vast majority of patients will go on to develop the most aggressive form of
this disease called castration resistant (CRPC) disease within a median timespan of 2-3
45, 46
years.
CRPC still depends on androgen stimulation for its growth and as such the use of
androgen receptor (AR) inhibitors is still useful either alone or in combination with other
47
therapies.
Figure 2: Structures of non-steroidal anti-androgens (1-3) and abiraterone acetate (4), drugs
approved for the treatment of various stages of Prostate Cancer.
One of the problems associated with the use of bicalutamide, and similar compounds is that
their prolonged use not only causes resistance but theyir use enhance the tumor growth. Efforts
to contrast this resistance resulted in the development of novel and high affinity anti-androgens,
which extended the median survival time of CRPC patients. Molecules such as enzalutamide
(
Xtandi®, MDV3100, 2) and the most recently approved apalutamide (ARN-509, 3, for non-
®
metastatic PC) and abiraterone acetate (Zytiga 4) are now extensively used in PC therapy.
3

ACCEPTED MANUSCRIPT
Unfortunately, despite a favorable profile of these drugs, patients with CPRC and those who are
initially sensitive eventually begin to develop secondary resistance also to these drugs as for
50
example reported for enzalutamide. Despite the resistance problem, ASCO recommends “men
with metastatic CRPC to continue hormone therapy to keep androgen levels low regardless of
51
the other treatments used”. In this scenario it is evident that the search for more potent agent
52, 53
against prostate cancer is still imperative. In our recent communications
we have
demonstrated that the introduction of fluorinated groups into selected positions of the aromatic
rings of the bicalutamide (1) and enzalutamide (2) allowed us to obtain new compounds with
improved in vitro anti-proliferative activity in the most common PC cell lines. We were the first to
demonstrate that depending on the position of these fluorinated groups we could convert the
partially AR agonist enobosarm into a full AR antagonist molecule with the potential to become
5
3
a PC drug. Moreover, structure–activity relationship (SAR) studies on bicalutamide-like
molecules have shown that electron-poor ring A, (e.g bearing CN, NO or CF ) is required for a
2
3
54
strong ligand–AR binding. These findings prompted us to explore the effects of the
polyfluorinated groups SF and C on the antiproliferative activity of bicalutamide, enobosarm
and enzalutamide in PC cell lines.
5
2 5
F
55
5 3
The SF group, considered as a CF , tert-butyl, halogen, or nitro groups bioisoster, is stable
56
under physiological conditions and possess unique physical and chemical properties such as
a great electronegativity coupled with an unusual lipophilicity, which might have an impact in the
activity considering the very hydrophobic environment of the AR binding pocket). SF has also a
5
different electron density profile (pyramidal for SF
larger molar volume than CF that may be beneficial for biological activity. On other hand the
C F group is reported to offer similar properties to SF such as comparable lipophilicity, while
5 3
opposite to spherical for CF ) as well as a
55
3
2
5
5
slightly less electronegative. It shows also high chemical and metabolic stability with a size
intermediate between CF and tert-butyl. Therefore, selected enobosarm candidates, were also
3
functionalised with the bulkier, but electronically similar to CF
their antiproliferative activity.
3 2 5 ,
, C F group and evaluated for
In this article, we would like to report our findings on the synthesis and the biological evaluation
including antiproliferative activity on four PC cell lines, of SF5 and C2F5 substituted
bicalutamide/enobosarm and enzalutamide analogues (Figure 3). For selected structures the
AR antagonist/agonist behaviour as well as the in vitro ADME properties are reported. The lead
compound was evaluated for its ability to impair tumor growth in PC xenograft mouse model.
4

ACCEPTED MANUSCRIPT
Figure 3. General structures of the compounds investigated in this study
Results and discussion
Chemistry
Synthesis of SF5-substituted bicalutamide and enobosarm derivatives.
The synthesis of bicalutamide and enobosarm-based derivatives have been performed
57-59
according to our previously reported synthetic strategy (Scheme 1).
As previously reported
by us, racemic and R-bicalutamide analogues showed no significant difference in
antiproliferative activity and as such, all our enobosarm and bicalutamide-derived structures in
this article have been prepared as racemic mixtures. However, R-Bicalutamide was prepared
and included in the biological assays as reference.
Reactions of the appropriate functionalised anilines 6-14 (SF5, CF
combination) with methacryloyl chloride 5 in dimethyl acetamide provided methacrylamides 15-
3 in good yields (77-95%). These compounds in turn were converted into the epoxides 24-32
3 2
, CN, NO and their
2
by oxidation with a hydrogen peroxide/trifluoroacetic anhydride mixture in dichloromethane
solution at room temperature for 24 hours. Opening of the epoxide with sodium
phenolates/thiophenolates generated by deprotonation with sodium hydride, furnished the
desired sulfide/ether derivatives 33-53. Further oxidation of the thioether derivatives with meta-
chloroperbenzoic acid in dichloromethane solution provided the bicalutamide derivatives 49-53
in good yields (57-97%)
5

ACCEPTED MANUSCRIPT
Scheme 1. Synthesis of bicalutamide (33-39,49-53) and enobosarm analogues (40-48)
Reagents and conditions: (a) Anilines 6-14 (3.4 mmol scale), 5 (8 equiv.), DMA, rt, 3h; (b) 15-
2
3, H
2 2 3 2
O (4 equiv.)/(CF CO) O (5 equiv.), DCM, rt, 24h; (c) NaH (1.2 equiv.), 24-32 (1.2 equiv.),
ArXH (a-p), (1.1 equiv. ), THF, rt, 24h; (d) m-CPBA (1.4 equiv.), DCM, rt, 4-6h.
Synthesis of SF
5
substituted enzalutamide derivatives
Enzalutamide analogues bearing SF group either on B or A ring, were synthesised following
5
60
two different procedures, both derived by appropriate modification of reported methodologies.
Compounds 65-68 a,b and 69-70d-e were synthesised according to a three-step literature
61
procedure highlighted in Scheme 2.
Briefly, anilines 6-11 were converted into the
corresponding isothiocyanates 54-59 by treatment with thiophosgene in dichloromethane
solution in presence of sodium bicarbonate. These intermediates were successively reacted
with amino-cyanide 61-64, (in turn prepared from anilines 10-12 and 60 via Strecker reaction),
to obtain final enzalutamide analogues 65-68 a-b and 69-70d-e. We were also interested in
keeping the enzalutamide B ring intact while introducing the pentafluorosulfanyl group on the A
ring. The synthesis of these compounds (75a and b) was accomplished by a literature route
62
reported (Scheme 3). Ullmann type reaction between 4-bromo-2-fluoro-N-methylbenzamide 72
prepared from the corresponding acid 71) and 2-aminoisobutyric acid, catalysed by copper
chloride, afforded the functionalized aniline 73 in 67% yield. This was converted into the
(
corresponding methyl ester 74 (methyl iodide, potassium carbonate, in
a
6

ACCEPTED MANUSCRIPT
water/dimethylformamide mixture), which was subsequently cyclised into the desired
enzalutamide derivatives 75a-b with pentafluorosulfanyl anilines 58 and 59 in hot
dimethylsulfoxide/isopropyl acetate mixture for 14 hours.
61
Scheme 2. Synthesis of enzalutamide derivatives 65-68a-b and 69-70d-e.
Reagents and conditions: a) 6-11, CSCl (1.5-2 equiv.), NaHCO (4.1 equiv.), H O, DCM, rt,
2
3
2
2
4h; b) TMSCN (5.2 equiv.), acetone, 80 °C, 12 h; c) 54-59 (1.0 mmol), 61-64 (0.84 mmol)
DMF, rt, 48 h, then 2N HCl (3 mL), MeOH (10 mL), reflux, 6 h.
62
Scheme 3. Synthesis of pentafluorosulfanyl enzalutamide derivatives 75a-b.
7

ACCEPTED MANUSCRIPT
Reagents and conditions: a) SOCl , iPAc, DMF, 72 °C, 4 h; b) MeNH , DMF, rt, 15 min; c) 2-
2
2
aminoisobutyric acid, K CO , CuCl, 2-acetylcyclohexanone, DMF, H O, 105 °C, 14 h; d) MeI,
2
3
2
2 3 2
K CO , H O, DMF, 40 °C, 1h; e) 58-59, DMSO, iPAc, 84 °C, 14 h.
Synthesis of pentafluoroethyl bicalutamides and enobosarms.
As part of our interest in exploring the use of polyfluorinated groups in bicalutamide and
enobosarm series, we then moved to investigate the effect of the pentafluoroethyl functional
group (C F , on the anticancer activity of these new derivatives. For the synthesis of these
2
5)
derivatives, we envisaged that an introduction of the desired fluorinated group at the late stage
of the synthetic pathway (ideally in the last step) would be the best approach. In the recent
years, a great deal of research has been devoted to the introduction of fluoroalkyl group into
aromatic rings. Introduction of pentafluoroethyl group onto aromatic compounds have been
subjected to a deep investigation and several methods are available in literature, mainly by
means of organometallic reagents. These methodologies often require the in-situ preparation of
63-65
63
the pentafluoethyl-metal reagent
high temperature (180 ˚C), and careful control of reaction
conditions. To avoid these problems, we looked at the work of Hartwig group who have
developed a copper-based reagent (pentafluoroethylator®) capable of replacing aromatic iodide
66, 67
or bromide with a CF or a C F groups under relatively mild conditions.
When we attempted
3
2 5
the pentafluorethylation of compounds 46o, 48o,p, (see Scheme 1), the reaction worked very
8

ACCEPTED MANUSCRIPT
well and the desired polyfluorinated derivatives, 77-79 were obtained in good yields after
chromatographic separation on silica gel (Scheme 4).
Scheme 4. Synthesis of enobosarm’s pentafluoroethyl derivatives via
pentafluoroethylation of iodides 46o, 48o, 48p.
Reagents and conditions: a) Cu(Phen)CF CF 76 (2 equiv.), DMF, 50 ˚C, 24 h.
2
3
53
In our previous studies, we demonstrated that Enobosarm analogues with the bis-3,5-CF
3
motif on the ring B were quite active in the antiproliferative assay. We were then interested into
investigate if the introduction of the C F functionality on ring A of these derivatives could further
2
5
improve their antiproliferative activity.
However, the synthetic approach proposed in Scheme 1 was unsuccessful for the synthesis
these derivatives. In particular, when we attempted the oxidation of 3-iodo alkene 22 with the
mixture of trifluoroacetic anhydride and hydrogen peroxide, we were only able to observe trace
of the desired product 31 in a complex mixture with other unidentified product likely derived from
the formation of hypervalent iodine compounds like bis(trifluoroacetoxy)iodoarene (Scheme
68
5
A). In addition, the presence of the highly electronegative trifluoromethyl group in the ortho
position of anilines 8 and 9 made these compounds scarcely reactive, and low yields of the
acrylamides 17, 18 and epoxides 26, 27 were obtained. We envisaged that this low reactivity
would have been even more severe in the case of a C F group in this position.
2
5
9

ACCEPTED MANUSCRIPT
Scheme 5. A) Unsuccessful epoxidation of methacrylamide 22; B) synthesis of compounds 83
and 82 via pentafluoroethylation of 22.
Reagents and conditions: a) 76 (2 equiv.), DMF, 50˚C, 48h; b) H O (4 equiv.)/(CF CO) O (5
2
2
3
2
equiv.), DCM, rt, 24h; (c) NaH (1.2 equiv.) 3,5-bis-trifluoromethyl thiophenol (1.2 equiv.), THF, rt,
4h; d) NaH (1.2 equiv.) 3,5-bis-trifluoromethyl phenol (1.2 equiv.), THF, rt, 24 h.
2
We therefore decided to attempt the pentafluorethylation directly on compound 22. When a dry
DMF solution of 22 was then treated under argon with pentafluoroethylator 76 (2 equiv.) at 55
˚
C for 48 hours, the desired pentafluoroethyl methacryl amide 80 was obtained in quantitative
yield. Epoxidation with trifluoroacetic anhydride and hydrogen peroxide proceeded smoothly
affording pentafluoroethyl epoxide 81 which was then coupled with 3,5-bis-trifluoromethyl
thiophenol and 3,5-bis-trifluoromethyl phenol to afford compounds 83 and 82 in 27% and 36%
yield respectively.
1
0

ACCEPTED MANUSCRIPT
Biological Evaluation
In vitro 2D monolayer antiproliferative assay.
All newly synthesised compounds were evaluated for their antiproliferative effect in an in vitro
2
D monolayer assay in four human prostate cancer cell lines (LNCaP, 22Rv1, VCaP, and
DU145). LNCaP, VCaP and 22Rv1 exhibit androgen sensitivity, whereas DU145 was selected
as hormone-insensitive cell line.
Vcap cell line was selected because they show positive response for androgen sensitivity with
69
wild-type AR mRNA/protein, and expresses PSA mRNA/ protein as well. To establish if our
compounds are effective also on cells with mutated AR, we selected LNcaP cell line that
70, 71
express endogenous T877A mutation
in the AR coding sequence. Finally, 22Rv1 cell line
was selected because of endogenously expressed AR splice variants. These have been
72
identified as one of the main players in hormone refractory tumor progression. Two such
variants have been identified as specifically a full length isoform with an exon 3 duplication and
73
a C terminal domain truncations with aberrant exon 2b expression.
The antiproliferative activity (absolute IC50 in µM) of the most active compounds are reported in
table 1 for the bicalutamide/enobosarm series and in table 2 for the enzalutamide series
including also the antiproliferative activities of four reference compounds (racemic bicalutamide
7
4
1
and its R- isomer , enzalutamide 2, enobosarm 40j), which appeared consistent with previous
reported data for these specific cell lines (see supporting information for the complete table of
IC50s).
Table 1. Antiproliferative activity of selected polyfluorinated bicalutamide and enobosarm
derivatives in four prostate cancer cell lines
.
Entry Cpd
X = O Enobosarm derivatives
X = SO Bicalutamide derivatives
2
1
1

ACCEPTED MANUSCRIPT
a
R (A ring)
X
Ar (B ring)
Absolute IC (µM)
50
VCap LNcap 22Rv1 Du-145
1
2
3
4
5
6
7
8
9
44f
48a
44n
79
4-SF5
3,5-CF3
4-SF5
O
O
O
O
O
O
O
S
2-CF3
4-SF5
3.1
2.4
1.2
1.9
1.4
3.2
7
7.1
6.5
3,5-CF3
4-C2F5
3,5-CF3
2.6
2.3
8.6
3,5-CF3
3-SF5
3.0
2.7
2.5
6.5
2.3
5.4
4.2
3.6
4.1
24.8
49.6
46.3
6.8
45n
48b
2.8
2.8
9.3
3,5-CF
3
3-SF
5
5
2.7
3.0
5.7
41b 4-NO2, 3-CF
3
3-SF
5.3
3.3
7.4
83
77
78
3-C2F5
4-CF3
3,5-CF3
3-C2F5
3-C2F5
4-CN
4-F
8.6
3.9
8.9
O
O
O
8.1
3.9
7.6
1
0
1
2
3
3,5-CF3
9.5
4.4
8.5
1
1
1
40J 4-CN, 3-CF3
24.5
68.4
51.6
20.9
45.3
45.2
44.6
49.2
45.4
(1)
4-CN, 3-CF3 SO2
R-(1) 4-CN, 3-CF
3
SO
2
4-F
1
2

ACCEPTED MANUSCRIPT
a
All data are mean values from at least triplicate experiments, with standard deviations of ±
0% of the value quoted unless otherwise stated (mean value ± standard deviations).
1
Almost all new derivatives performed better than bicalutamide, showing concentration
dependent activity in all four PC in vitro models. It is possible to appreciate that the best linkers
between the two portions of the molecule contain either a sulfur or an oxygen atom. When the
sulfur was oxidized to the corresponding sulfoxide (or sulfone) a general loss of activity was
observed. This is consistent with what has been previously observed with other bicalutamide
60, 75
derivatives.
IC when compared with the pentafluoroethyl moiety except when present in the para position
In general, the pentafluorosulfanyl group is slightly more efficient in reducing the
50
of ring B and ring A is functionalised with a 3,5-bistrifluoromethyl moiety (Table 1, entry 4).
However, it is important to remark that both polyfluorinated functional groups are more effective
than fluorine of CF groups. In terms of ring substitution, from our data, it seems that the SF
3
5
group is very effective when it is located in the para position of ring A (Table 1, entries 1 and 3)
and B (Table 1, entry 2). Compound 44f (4-SF on ring A and 2-CF on ring B) is more potent
5
3
than the compound 45f (full table on SI) with the same substitution pattern on ring B (2-CF
3
) but
with the SF on ring A moved to the meta position (table 1). However, when the 3,5-bis CF
5
3
motif is present on either ring A or B the position of the SF (either ortho or meta) on the other
5
ring seem to be not important. Indeed, compounds 48a and 48b (table 1 entry 2 vs 6) do shows
the same activities except in LNCaP cell where 48a seem to perform better. In the
2 5
pentafluoroethyl series it is clear from the data of table 1 that p-C F in the B ring is very
effective for the antiproliferative activity. When the ring A is substituted with the 3,5-bis-
trifluoromethyl moiety, moving the pentafluoroethyl moiety from the 4-position to the 3-position
resulted in the reduction of activity (Table 1 entries 4 vs 10).
Considering the antiproliferative results across the four different cell lines, the new compounds
showed significant activity also on androgen-insensitive DU-145 cells with IC50 values for most
compounds higher in this cell line than in the other three, suggesting the presence of a potential
off-target effect, besides the canonical anti-androgen mechanism. Interestingly, some studies
have demonstrated that even DU-145 and PC3 cell lines are expressing significant levels of AR
76
protein, although at a lower level than LnCaP cell line. In addition, DU-145 cell line usually
77
express NF-kB, a transcription factor involved in prostate cancer progression. It seem
plausible to speculate that our molecules are either blocking the AR expressed by the DU-145
cells or are somewhat able to target other pathways such as the expression of NF-kB.
1
3

ACCEPTED MANUSCRIPT
Within the enzalutamide series only three derivatives (67a, 70e and 67b) showed a better
activity than enzalutamide itself (Table 2) in LnCaP cell line. Unfortunately, in these series, the
solubility of these polyfluorinated compounds posed a serious limitation in their anti-proliferative
evaluation (See SI for complete table of the antiproliferative data).
Table 2. Antiproliferative activity of selected polyfluorinated enzalutamide derivatives in four
prostate cancer cell lines
.
Entry Cpd
a
R (A ring)
Ar (B ring)
Absolute IC (µM)
50
VCap
3
LNcap
5.4
22Rv1
12.7
ND
Du-145
3
1
2
3
7
67a
4-CN, 3-CF3
3-SF5
4-SF5
3-CF3
70e
3.4
4.1
11.5
ND
3.4
67b
2
4-CN, 2-CF3
4-CN, 3-CF3
3-SF5
ND
ND
4.1
3-F,
31.8
32.3
11.5
Enzalutamide
4-CONHMe
a
All data are mean values from at least triplicate experiments, with standard deviations of ± 10% of the
value quoted unless otherwise stated (mean value ± standard deviations).
During the preparation of this manuscript, a similar study appeared in the literature where
pentafluorosulfanyl substituted enzalutamide analogues, were evaluated for their in vitro efficacy
78
in LnCaP cell lines as well as for their AR antagonism. Similar to our findings, the
1
4

ACCEPTED MANUSCRIPT
antiproliferative activity of these derivatives, although structurally different from ours, was found
similar to enzalutamide. In addition, antagonism behaviour was proved for these compounds.
Agonist-antagonism assay
Once established the efficacy of these new derivatives in the antiproliferative assay, we
progressed to evaluate their AR antagonist activity.
5
Among all the derivatives prepared, compound 44f with a SF group in the A ring and our best
analogue in terms of antiproliferative activity, along with compound 41b, bearing the
pentafluorosulfanyl substituent on the B ring were investigated in an agonist/antagonist assay
(
Figure 4). We used the GeneBLAzer® beta lactamase reporter technology adapted to the
nuclear receptor, to asses if these compounds are AR antagonists. Indeed, we observed that
both compounds behave as full AR antagonists. It is worth noting that introduction of the SF
motif leads to the switch of the enobosarm scaffold from partial agonist to full antagonist of the
5
5
3, 60
AR receptor, as also previously observed by us with other fluorinated groups.
Figure 4
below shows the dose response curve for enobosarm 40j (A) 41b (B) and 44f (C). Table 3
collects the IC50 values, the antagonistic and agonistic mean effect, expressed as percentage,
for the three compounds in the same assay. This result is of great significance for the future
development of AR antagonist and ultimately for the treatment of PC condition.
1
5

ACCEPTED MANUSCRIPT
NC
CN
O
O
N
CF3
H
OH
0j
4
1
6

ACCEPTED MANUSCRIPT
Figure 4. Dose Response curve (at 10 different concentrations, max 10 µM) for the antagonist
assay for Enobosarm 40j (A) and compounds 41b (B), 44f (C). The mean value from the two
experiments has been plotted in a graph log (antagonist) ([M]) against % of inhibition.
Table 3. AR antagonistic and agonistic activity of enobosarm (40j), compounds 41b, 44f and, as
comparison, Bicalutamide (1).
Antagonistic
Agonistic
1,3
Cpds
IC (µM)
50
1,2,3
1,4
(
mean % inhibition)
(mean % activation)
4
0j
59
90
93
83
0.036
0.540
1.810
0.490
26
2
4
1b
44f
-3
1
5
1
2
Data presented are the means from duplicate experiments; Compounds were considered full
antagonists if at 10µM the reduction of R1881 (standard AR antagonist) effect was greater than
3
4
8
0%; Data from compounds tested at 10 different concentrations; Data from compounds
tested at single (10 µM) concentration in absence of R1881.
In vitro Absorption, Distribution, Metabolism, and Excretion (ADME) screening.
Compounds 44f, 41b and 48a were evaluated for their metabolic stability, binding to protein
and cardiotoxicity (Table 4). First the tested compounds were incubated for 45 min with pooled
liver microsomes, and the intrinsic clearance (CLint) and half-life (t1/2) values were calculated
based on 5 time points for a maximum of 45 minutes (Table 4). All compounds examined,
except 44f (which shows a medium CLint value and short half-life), have low intrinsic clearance
in liver microsome and long half-lives (~7.5 hours). In particular, compounds 48a, 41b, showed
1
7

ACCEPTED MANUSCRIPT
CLint values below 6
µ
L/min/mg microsomal protein, suggesting that these compounds will
undergo much slower rates of hepatic metabolism in vivo. In addition, compounds 48a and 41b
showed a metabolic stability greater than bicalutamide confirming the reported bio-stability of
79
the pentafluorosulfanyl motif. All the selected candidates showed plasma protein binding
greater than 91%. Our most active compound 44f, showed a 98.6% protein binding very similar
80
to the bicalutamide value (96%). The least bound to plasma protein was compound 48a
91%). Moreover, all tested compounds lacked significant hERG inhibition in the patch-clamp
(
assay (at 6 different concentrations). In particular, the most active compounds in antiproliferative
assay, 44f and 48a showed a mean hERG inhibition of 26.3% and 13.4% respectively at 25 µM.
The IC50 calculated over 6-different concentrations were all above 25 µM showing a satisfactory
degree of safety for these derivatives.
Table 4. Metabolic stability in pooled liver microsomes, plasma protein binding and
cardiotoxicity results of selected compounds.
a,b
b
c
Comp
CLint
T
1/2 (min.)
63.1
PPB (% bound)
hERG (% mean inhibition)
4
4f
22.0
3.08
3.21
6.48
98.6
90.9
98.5
26.3
13.4
38.1
4
8a
1b
450
4
431
d
98
e
(R)-1
214
ND
a
b
µL/min/mg protein, from quadruplicate assay; Data expressed as mean values from
c
duplicate experiments; data at 25 µM: Weak or no inhibition from quadruplicate experiments if
d
80 e
IC50 >10µM; From literature data;
Not Determined.
In vivo efficacy evaluation in 22Rv1 prostate cancer model.
With the ADME data in hand we decided to evaluate one of our best performing
pentafluorosulfanyl derivatives in an in vivo efficacy xenograft model in mice to assay their
1
8

ACCEPTED MANUSCRIPT
ability to impair the growth of prostate tumour in comparison to the standard-of-care docetaxel.
Among the most active bicalutamide/enobosarm series in the antiproliferative activity,
compound 48a, although slightly less potent than compound 44f was selected for in vivo
evaluation because its longer t1/2 value, and lower plasma protein binding and because it was
one of the better performing compound in our previously published results on bis-CF3
53
analogues.
At first, the tolerability of 48a were tested in nude, non-tumour bearing mice with the maximum
tolerated dose (MTD) established at 100 mg/Kg and then used later for the xenograft
experiment. Standard-of-care docetaxel was instead given at a dose level of 15mg/Kg. (Table
5
).
Table 5. Summary of the in vivo efficacy experiments Tumor Model PRXF 22Rv1
a
Therapy
Dose Level
Schedule
Route Minimum T/C [%] Efficacy
Td
[Days]
Tq
[Days]
b
c
d
(Day)
[
mg/kg/day]
[dosing
Days]
Rating
e
Vehicle
10
0-18
i.p.
n/a
n/a
2.7
6.2
ml/kg/day
4
8a
100
15
0-16
i.p.
i.v.
43.9 (14)
22.4 (14)
+
6.7
6.3
14.3
43.2
e
Docetaxel
0,7,14
++
n/a, not applicable; n.r., not reached (i.e. group median RTVs always < 200% / 400%); Efficacy rating:
++++, T/C < 5%, +++, T/C: 5 - < 10%; ++, T/C: 10 - <25%; +, T/C: 25 - <50%; +/-, T/C : 50-65%; -, T/C ≥
a
6
5%; Vehicle for 48a and the control group: 10% EtOH, 10% Cremophor EL, 5% dextrose; vehicle for
b
c
docetaxel, 0.9% Saline; Minimum T/C values are calculated based on median values; Tumor volume
d
e
doubling time; Tumor volume quadrupling time; Docetaxel was used as positive control, while DMSO
(
vehicle) was used as negative control.
Then, in vivo efficacy data were obtained selecting 22Rv1 cells to perform this study. Mice were
injected in the hind flank with tissue culture 22Rv1 cells. Once enough tumors have formed, the
1
9

ACCEPTED MANUSCRIPT
tumor-bearing mice were randomised into groups and therapy with 48a and docetaxel were
performed for 4 weeks. Treatment with the investigational compound was given as 16-18 daily
intraperitoneal (i.p.) doses, whereas docetaxel was given as three weekly intravenous (i.v.)
treatments. The treatment period was followed by an observation period until day 40 for 48a
group and until day 49 in the docetaxel group. The first day of dosing was either the day of
randomisation (day 0) or the following day (day 1). Anti-tumor efficacy was evaluated as
inhibition of tumor growth represented by minimum T/C (treated to control) value based on
group median relative tumor volumes (RTVs) in test and control groups (see SI for the pertinent
plot). The body weight was also recorded to assess the animal’s tolerance of the compound.
The group receiving treatment with 48a was compared to the docetaxel and to the control
groups. The highest anti-tumor efficacy was observed in the docetaxel-treated group with a
minimum T/C value of 22.4%. Compound 48a displayed moderate anti-tumor efficacy with a
minimum T/C value of 43.9%. (Figure 5) Statistical significance was observed for both docetaxel
and compound 48a in comparison to the vehicle control group.
Figure 5. Efficacy of compound 48a and docetaxel in R22v1 xerograph mice model.
Minimum T/C values for the two groups are shown. As a turning point, a minimum T/C value of 65% (upper limit for borderline anti-tumor efficacy) was
chosen.
2
0

ACCEPTED MANUSCRIPT
As it can be appreciated by the figure 6A that in the first 10 days of administration, compound
4
8a was able to better control the absolute volume tumor growth with respect to Docetaxel
which become more effective as the time progress to 20 days.Treatment with 48a was well-
tolerated with adjusted survival rates of 100% and no group median body weight losses (BWLs,
Figure 6B). A severe BWL of 19.8% was recorded in the docetaxel-treated group, although the
adjusted survival rate was 100%.
Figure 6. Plot of medians absolute (A) tumor volume and (B) body weight for 48a and docetaxel
as function of time.
2
1

ACCEPTED MANUSCRIPT
Conclusions
In summary, we have demonstrated that functionalisation of bicalutamide and enobosarm
scaffold with pentafluorosulfanyl and pentafluoroethyl functionality, lead to the development of
potent antiproliferative agents against prostate cancer in in vitro models. The IC50 for the
majority of the newly prepared molecules were about 48-times better than the parent drugs.
Among the compounds with improved anticancer activity, enobosarm analogues performed
better with respect to their bicalutamide analogues, pointing out that a less bulky linker
(
compared to SO
2
) between ring A and B lead to a better activity.
Overall SF structures displayed the best in vitro anticancer activity, together with AR full
5
antagonism (a key requirement for a drug against prostate cancer) and ADME properties. One
of the most active compounds (48a) showed also some efficacy in in vivo xenograft experiment,
although its activity was lower than the standard of care docetaxel. Pleasingly, (48a) was better
tolerated with respect to docetaxel as evidenced by the lack of body weight lost in the treated
mice. Pentafluoroethyl derivatives also showed an excellent improvement of the antiproliferative
activity of the parent compound.
In contrast, SF5-functionalised Enzalutamides were hampered by low solubility that precluded
their full evaluation in animal model, although one derivative (67a) showed an improved
antiproliferative activity (in vitro) with respect to enzalutamide. Altogether, our results are
indicative of the fact that both SF and C F are indeed two functionalities that can be used in
5
2 5
place of the trifluoromethyl group and exploited for the design of drug-like molecules.
Having established in this work that the in vitro anticancer activity can be extensively improved,
for a successful progression of these derivatives to clinical studies, further evaluation of their
biological properties (PC protein expression, inhibition of AR-mediated gene transcription,
quantitative AR binding assay) needs to be addressed. These experiments are currently
underway in our laboratories and will be reported in due course.
2
2

ACCEPTED MANUSCRIPT
Experimental
General Chemistry methods
All solvents and reagents were used as obtained from commercial sources unless
otherwise indicated. All reactions were performed under a nitrogen atmosphere. The 1H
and 13C NMR spectra were recorded on a Varian spectrometer operating at 500 MHz
for 1H and 125 MHz for 13C. Deuterated chloroform was used as the solvent for NMR
experiments, unless otherwise stated. 1H chemical shifts values (δ) are referenced to
the residual nondeuterated components of the NMR solvents (δ = 7.26 ppm for CHCl3,
etc.). The 13C chemical shifts (δ) are referenced to CDCl (central peak, δ = 77.0 ppm).
3
Fluorine chemical shifts are referenced to CFCl . Mass spectra were measured in
3
positive mode electrospray ionization (ESI). TLC was performed on silica gel 60 F254
plastic sheets. Column chromatography was performed using silica gel (35–75 mesh) or
on an Isolera Biotage system. Purity of prepared compounds was determined by HPLC-
UV analysis (Thermo HPLC connected with UV detector). The purity of all final
compounds was determined to be >95% by RP-HPLC using the eluents water (eluent
A), acetonitrile (eluent B), and the following conditions: Varian Pursuit XS, 4.6 mm × 150
mm, 5.0 µm, 1.0 mL/min, gradient 30 min 10% → 100% eluent B in eluent A (method 1).
HRMS were acquire on a Waters XEVO-27-Jun-2019 G2XSQTOF instrument under
electrospray condition in both positive and negative modes.
Synthetic Procedures
General method for the preparation of intermediates 15-23, 80
2
3

ACCEPTED MANUSCRIPT
Methacryloyl chloride 5 (2.63 mL, 27.16 mmol) was added over the course of 10
minutes to a stirring solution of the different aniline 6-14 (3.4 mmol) in N,N-
dimethylacetamide (14 mL) at r.t. for 3 h or o.n. After the reaction was complete, the
mixture was diluted with ethyl acetate (100 mL), extracted with sat. aq. NaHCO solution
3
(
3x25 mL) and with cold brine (4x50 mL). The organic layer was dried over Na SO and
2 4
the solvent vas removed at reduced pressure. The crude residue was purified by flash
column chromatography.
N-(4-(Pentafluorosulfanyl)phenyl)methacrylamide 19
Purified by flash column chromatography eluting with n-hexane-EtOAc 100:0 v/v
1
increasing to n-hexane-EtOAc 80:20 v/v. Obtained in 99% yield as a white solid. H
3
NMR (CDCl ), δ: 7.74 (d, J = 9.2 Hz, 2H), 7.70-7.68 (m, 3H), 5.84 (s, 1H), 5.56-5.55 (m,
1
9
1
1
8
H), 2.09 (m, 3H). F NMR (CDCl ), δ: 85.3 (quintet, J = 151.1 Hz, 1F), 63.5 (d, J =
3
13
51.1 Hz, 4F). C NMR (CDCl
3
), δ: 166.7, 140.2 (m), 127.0 (m), 120.8, 119.1, 103.9,
0.7, 30.1, 18.6.
N-(3-iodophenyl)methacrylamide 22
Purified by flash column chromatography eluting with n-hexane-EtOAc 100:0 v/v
1
increasing to n-hexane-EtOAc 90:10 v/v. Obtained in 60% yield as a white solid. H
NMR (500 MHz, CDCl ) δ 8.00 (s, 1H, NH), 7.57 (ddd, J = 0.7, 2.0, 8.2 Hz, 1H), 7.48
3
(ddd, J = 1.0, 1.5, 7.9 Hz, 1H), 7.08 (t, J = 8.0 Hz, 1H), 5.81 (s, 1H), 5.51 (d, J = 1.4
13
Hz, 1H), 2.08 (d, J = 0.4 Hz, 3H). C NMR (126 MHz, CDCl ) δ 166.50, 140.65,
3
2
4

ACCEPTED MANUSCRIPT
1
38.90, 133.45, 130.47, 128.72, 120.21, 119.21, 103.99, 94.08, 80.81, 30.11, 24.18,
8.46.
1
N-(3,5-bis(Trifluoromethyl)phenyl)methacrylamide 23
Purified by flash column chromatography eluting with n-hexane-EtOAc 100:0 v/v
1
increasing to n-hexane-EtOAc 90:10 v/v. Obtained in 99% yield as a white solid. H
NMR (CDCl ), δ: 8.11 (s, 2H), 7.76 (bs, 1H), 7 1H), 5.87 (d, J = 1.1 Hz), 5.59 (m, 1H),
3
19
13
2
.1 (s, 3H). F NMR (CDCl ), δ: -63.0 (s, 3F). C NMR (CDCl ), δ: 166.6, 140.2 (q, J =
3 3
33.6 Hz), 139.2, 137.5 (q, J = 272.8 Hz), 121.0, 117.6 (m), 18.6.
Preparation of N-(3-pentafluoroethyl)phenyl)methacrylamide 80
A solution of iodide 22 (0.8 g, 2.8 mmol), and (1, 1, 2, 2, 2- Pentafluoroethyl) (1, 10-
phenanthroline- κN1, κN10)-copper (Pentafluoroethylator®, 5.6 mmol, 2 equiv) in dry
DMF where heated at 50 ˚C for 48 hours. After this period the mixture was cooled at
room temperature and diluted with diethyl ether and the orange solution filtered through
a pad of celite. The pad was washed with diethyl ether and the combined organic were
washed with 2M HCl (2x 10 mL) sat NaHCO (2x 10 mL) dried and evaporated. The
3
residue was pure enough (90%) to be used in the next step.
1
H NMR (500 MHz, CDCl ) δ 7.85 (t, J = 12.0 Hz, 2H), 7.65 (s, 1H), 7.50 (t, J = 8.0
3
Hz, 1H), 7.38 (d, J = 7.8 Hz, 1H), 5.85 (s, 1H), 5.55 (d, J = 1.4 Hz, 1H), 2.10 (s, 3H).
19
3
F NMR (CDCl ), δ: -84.63 (s), -117.01 (br s); MS [ESI, m/z]: 302.05 [M+Na].
2
5

ACCEPTED MANUSCRIPT
General method for the preparation of intermediates 24-32
To a stirred solution of the different intermediate 15-23 (3 mmol) in DCM (7 mL) was
added 30% hydrogen peroxide (3.6 mL, 32.03 mmol). The reaction mixture was put in a
water bath at r.t. and trifluoroacetic anhydride (3.7 mL, 26.7 mmol) was added slowly to
the mixture, which was then stirred for 24 h. The reaction mixture was transferred to a
separating funnel using DCM (30 mL). The organic layer was washed with distilled
water (20 mL), sat. aq. Na2S O (4x20 mL), sat. aq. NaHCO (3x20 mL) and brine (20
2
3
3
mL), dried over Na SO and concentrated at reduced pressure. These intermediates
2
4
were usually of good purity (90%) to be used in the next step without further purification.
When necessary, column chromatography (EtOAc/Hexane, silica gel) or preparative
TLC were used to purify the compounds.
N-(4-(pentafluorosulfanyl)phenyl)-2-Methyloxirane-2-carboxamide 28
1
Obtained in 73% yield as a white wax. H NMR (CDCl
3
),
δ
: 8.32 (bs, 1H), 8.03-8.02
(m, 1H), 7.72-7.70 (m, 1H), 7.52-7.49 (m, 1H), 7.44-7.39 (m, 1H), 3.01 (d, J = 4.2 Hz,
1
9
1
1
1
H), 2.97 (d, 4,8 Hz, 1H), 1.67 (s, 3H). F NMR (CDCl ),
δ
: 85.1 (quintet, J = 151.1 Hz,
: 169.0, 154.0 (p, J = 18.4 Hz),
37.6, 129.2, 122.6, 121.9 (p, J = 4.8 Hz), 117.3, 67.6, 56.6, 16.6; MS [ESI, m/z]: 326.2
3
13
3
F), 63.0 (d, J = 151.1 Hz, 4F). C NMR (CDCl ), δ
[
M+Na].
N-(3-(pentafluorosulfanylphenyl)-2-Methyloxirane-2-carboxamide 29
1
Obtained in 72% yield as a white solid. H NMR (CDCl ),
δ
: 8.05-7.99 (m, 1H), 7.82-
3
7.77 (m, 1H), 7.72 (bs, 1H), 7.55-7.50 (m, 1H), 7.48-7.40 (m, 1H), 5.85 (s, 1H), 5.55 (m,
2
6

ACCEPTED MANUSCRIPT
1
9
1
1
1
H), 2.09 (m, 3H). F NMR (CDCl ),
δ
: 83.96 (quintet, J = 151.2 Hz, 1F), 62.62 (d, J =
3
13
51.2 Hz, 4F). C NMR (CDCl ),
δ: 166.7, 140.4 (p, J = 18.1 Hz), 138.1, 129.2, 122.9,
3
21.7 (p, J = 4.4 Hz), 120.6, 117.6 (p, J = 4.5Hz), 60.4, 18.6.
N-(3,5-bis(Trifluoromethylphenyl)-2-methyloxirane-2-carboxamide 32
1
Obtained in 82% yield as a white solid. H NMR (CDCl
3
),
δ
: 8.35 (bs, 1H), 8.07 (s, 2H),
19
13
7
.64 (s, 1H), 3.02 (m, 2H), 1.71 (s, 3H). F NMR (CDCl3),
δ: -63.1 (s, 3F). C NMR
(CDCl
3
), : 169.0, 138.4, 132.6 (m), 119.3, 117.8 (m), 56.6, 16.7.
δ
General method for the preparation of compounds 33-48
To a mixture of NaH (60% in mineral oil, 0.050 g, 1.23 mmol) in anhydrous THF (2 mL)
at 0 °C under Ar atmosphere was added a solution of the differently substituted phenol
or thiophenol (1.11 mmol) in 1 mL of anhydrous THF. This mixture was stirred at r.t. for
20 minutes. A solution of the different intermediate 24-32 (0.74 mmol) in anhydrous THF
(3 mL) was added slowly. The reaction mixture was stirred at r.t. overnight. The mixture
was then diluted with ethyl acetate (30 mL), washed with brine (15 mL), water (30 mL),
and dried over Na SO and concentrated under vacuum. The crude residue was purified
2
4
by flash column chromatography or via Biotage Isolera 1.
N-(4-(pentafluorosulfanyl)phenyl)-3-(2-(trifluoromethyl)phenoxy)-2-Hydroxy-2-methyl
propanamide 44f
Purified by flash column chromatography eluting with n-hexane/EtOAc 100:0 v/v increasing to n-
1
hexane/EtOAc 70:30 v/v. Obtained in 59% yield as a white solid. H NMR (CDCl ),
δ
: 8.84 (bs,
3
2
7

ACCEPTED MANUSCRIPT
1
7
3
1
H), 7.75 (d, J = 9.2 Hz, 2H), 7.69 (d, J = 9.2 Hz, 2H), 7.61-7.59 (m, 1H), 7.56-7.53 (m, 1H),
.12-7.09 (m, 1H), 7.03 (d, J = 8.3 Hz, 1H), 4.48 (d, J = 8.7 Hz, 1H), 4.12 (d, J = 8.7 Hz, 1H),
19
.55 (s, 1H), 1.64 (s, 3H). F NMR (CDCl3),
δ
: 84.8 (quintet, J = 150.1 Hz, 1F), 62.9 (d, J =
13
50.1 Hz, 4F), -61.8 (s, 3F). C NMR (CDCl ), δ: 172.2, 155.3, 149.5 (m), 139.8, 133.7, 127.2
3
(m), 127.0 (m), 124.8, 122.6, 121.3, 120.5, 119.2 (m), 118.5, 112.9, 75.1, 72.9, 22.8. MS [ESI,
m/z]: 488.0 [M+Na]. HPLC (method 1): retention time = 24.60 min.
N-(3,5-(bis-trifluoromethyl)phenyl)-3-(4-pentafluorosulfanylphenoxy)-2-hydroxy-2-methyl-
propanamide 48a
Purified by flash column chromatography eluting with n-hexane/EtOAc 100:0 v/v increasing to n-
1
hexane/EtOAc 50:50 v/v. Obtained in 54% yield as a thick transparent oil. H NMR (500 MHz,
CDCl
3
) δ: 9.09 (s, 1H), 8.15 (s, 2H), 7.22-7.66 (m, 3H), 6.96 (d, J = 9.1 Hz, 2H), 4.53 (d, J =
1
9
9
.1 Hz, 1H), 4.08 (d, J = 9.1 Hz, 1H), 3.40 (s, 1H), 1.65 (s, 3H). F NMR (CDCl3),
δ
: 85.36
quintet, J = 151.1 Hz, 1F), 64.08 (d, J = 151.1 Hz, 4F), -62.9 (s, 1F). C NMR (126 MHz,
CDCl : 172.1, 159.5, 147.61 (t, J = 18.0 Hz), 138.60, 132.53 (q, J = 33.6 Hz), 127.83 (d, J =
.9 Hz), 123.02 (q, J = 272.8 Hz), 119.37 (d, J = 3.2 Hz), 117.96 (d, J = 7.6 Hz), 75.69, 72.71,
3.9; MS [ESI, m/z]: 534.05 [M+H], 556.05 [M+Na]. HPLC (method 1): retention time = 26.39
13
(
3
) δ
8
2
min.
General method for the preparation of isothiocyanates 54-59
A solution of NaHCO (0.82 g) in distilled water (4 mL) was stirred for 10 minutes and
3
to it was added DCM (4 mL) followed by the different aniline 7-12 (2.4 mmol). The
reaction mixture was cooled to 0 °C, thiophosgene (0.3 mL, 3.7 mmol) was added
dropwise over 10 minutes and the reaction was then stirred at r.t. overnight. The mixture
was then diluted with DCM (20 mL), washed with brine (20 mL), the organic layer was
2 4
dried over Na SO and concentrated to dryness.
4-Isothiocyanato-3-(trifluoromethyl)benzonitrile 56
2
8

ACCEPTED MANUSCRIPT
Aniline 9 (4.9 mmol) was dissolved in 9 mL toluene. Thiophosgene (10.67 mmol) was
added dropwise and the mixture was stirred at 75 °C for five days. Obtained in 99%
1
yield as a brown wax. H NMR (CDCl ),
δ
: 7.97 (d, J = 1.6 Hz, 1H), 8.86 (dd, J = 8.4
3
1
1
9
13
Hz, J
NMR (CDCl3),
16.6, 110.5.
2
= 1.6 Hz, 1H), 7.53 (d, J = 8.4 Hz, 1H). F NMR (CDCl
3
),
δ
: -63.12 (s, 3F).
C
δ
: 136.4, 130.9 (q, J = 5.5 Hz), 129.3, 122.7 (m), 120.5, 119.1 (m), 117.1,
1
General method for the preparation of nitriles 61-64
The different substituted anilines 10-12 and 60 (2.2 mmol) was added to a TMSCN
solution (11.4 mmol) in acetone (10 mL). The mixture was heated at 80 °C overnight.
The reaction was then concentrated in vacuo and the residue was partitioned between
water (15 mL) and ethyl acetate (15 mL). The water phase was re-extracted with ethyl
acetate (2x15 mL). The combined organic layers were washed with brine (20 mL), dried
over Na SO and concentrated to dryness.
2
4
2-Methyl-2-((4-(pentafluorosulfanyl)phenyl)amino)propanenitrile 61
1
Obtained in 69% yield as white crystals. H NMR (CDCl
3
),
δ
: 7.53 (d, J = 9.0 Hz, 2H),
: 86.4 (quintet,
: 146.0, 145.6 (p, J =
1
9
6
.63 (d, J = 9.0 Hz, 2H), 4.03 (bs, 1H), 1.62 (s, 6H). F NMR (CDCl ), δ
3
13
3
J = 150.1 Hz, 1F), 64.6 (d, J = 150.1 Hz, 4F). C NMR (CDCl ), δ
17.5 Hz), 127.4 (p, J = 4.8 Hz), 120.9, 114.1, 48.0, 28.0.
General method for the preparation of thiohydantoins 65-70
2
9