1-Amino-4,4-difluorocyclohexanecarboxylic acid as a promising building block for drug discovery: Design, synthesis and characterization

Article abstract of DOI:10.1016/j.tet.2013.03.072

1-Amino-4,4-difluorocyclohexanecarboxylic acid has been designed as a fluorinated analogue of the pharmacologically relevant 1- aminocyclohexanecarboxylic acid. The synthesis has been performed in three steps from a commercially available material in 22% overall yield. An impact of fluorine atoms on conformation, lipophilicity, acidity and fluorescent properties of the amino acid has been studied. Various practical applications of the obtained compound are suggested.

Full text of DOI:10.1016/j.tet.2013.03.072

Accepted Manuscript
1-Amino-4,4-difluorocyclohexanecarboxylic acid as a promising building block for
drug discovery: design, synthesis and characterization
Pavel K. Mykhailiuk, Viktoriia Starova, Vladimir Iurchenko, Svitlana V. Shishkina, Oleg
V. Shishkin, Oleksandr Khilchevskyi, Olga Zaporozhets
PII:
S0040-4020(13)00461-4
10.1016/j.tet.2013.03.072
TET 24163
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 3 December 2012
Revised Date: 21 February 2013
Accepted Date: 18 March 2013
Please cite this article as: Mykhailiuk PK, Starova V, Iurchenko V, Shishkina SV, Shishkin OV,
Khilchevskyi O, Zaporozhets O, 1-Amino-4,4-difluorocyclohexanecarboxylic acid as a promising building
block for drug discovery: design, synthesis and characterization, Tetrahedron (2013), doi: 10.1016/
j.tet.2013.03.072.
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
Graphical abstract
Design
Synthesis
X-Ray
Fluorescence
pKa
O
O
NH2
3 steps
HO2C
F
22% yield
O
F
logD
1

ACCEPTED MANUSCRIPT
1-Amino-4,4-difluorocyclohexanecarboxylic acid as a
promising building block for drug discovery: design, synthesis and characterization.
Pavel K. Mykhailiuk,*a,b Viktoriia Starova,b Vladimir Iurchenko,a Svitlana V. Shishkina,c Oleg
V. Shishkin,c,d Oleksandr Khilchevskyi,e Olga Zaporozhetsb
aEnamine Ltd., Vul. Oleksandra Matrosova 23, 01103 Kyiv, Ukraine
bDepartment of Chemistry, Taras Shevchenko National University of Kyiv, Volodymyrska str.,
64, 01601 Kyiv, Ukraine
cSSI, ´Institute for Single Crystalls,´ National Academy of Science of Ukraine, 60 Lenina ave.,
Kharkiv 61001, Ukraine
dDepartment of Inorganic Chemistry V.N.Karazin Kharkiv National University, 4 Svobody sq.,
Kharkiv 61022, Ukraine
e Institute of Bioorganic Chemistry and Petrochemistry, National Ukrainian Academy of Science,
Murmanska 1, Kiev 02660, Ukraine
Pavel.Mykhailiuk@gmail.com
Abstract
1-Amino-4,4-difluorocyclohexanecarboxylic acid has been designed as a fluorinated
analogue of the pharmacologically relevant 1-aminocyclohexanecarboxylic acid. The synthesis
has been performed in three steps from a commercially available material in 22% overall yield.
An impact of fluorine atoms on conformation, lipophilicity, acidity and fluorescent properties of
the amino acid has been studied. Various practical applications of the obtained compound are
suggested.
Key words
Amino acids, 1-aminocyclohexanecarboxylic acid, fluorine, fluorination, morph-DAST.
2

ACCEPTED MANUSCRIPT
Introduction
1-Aminocyclohexanecarboxylic acid (1) is a non-proteinogenic quaternary α-amino acid,
which has gained much application in peptide chemistry1 and drug discovery.2,3 The interest to
this compound is caused by several reasons. First, amino acid 1 is intrinsically conformationally
restricted due to both the cyclic ring and the quaternary α-carbon atom.4 Conformationally
restricted compounds, due to fixation of functional groups in a biologically active conformation,
are often more efficient and selective ligands for various targets compared to their non-restricted
analogues.5 Second, structure 1 is symmetric and therefore achiral, so that there is no need to
additionally prepare the optically active compounds in the medicinal structure-activity
relationship studies. Also, the saturated cyclohexane core ensures that derivatives of 1 are
lipophilic, which is an important property of oral drugs.6 It is not surprising therefore, that ~ 150
pharmacologically relevant derivatives of amino acid 1 are currently known.3 Among them are
the commercialized antibiotic Cyclacillin,7 antineoplastic Spiromustine8 and antiosteoporosis
agents Balicatib9 and L-006235.10
Cl
O
CO2H
N
N
H2N
CO2H
O
Cl
HN
S
N
H2N
HN
O
O
1
Spiromustine
Cyclacillin
antibacterial drug
anticancer drug
S
N
CN
CN
HN
N
N
N
N
O
H
H
N
N
O
O
Balicatib
L-006235
O
O
antiosteoporosis agent
antiosteoporosis agent
Figure 1. 1-Aminocyclohexanecarboxylic acid (1) and its bioactive derivatives:
Cyclacillin (antibacterial drug, Wyeth, 1979), Spiromustine (anticancer drug, National Cancer
Institute, 1987), Balicatib (antiosteoporosis agent, Novartis, 2006), L-006235 (antiosteoporosis
agent, Merck, 2005).7-10
Substituted cyclohexanes are often metabolically labile due to rapid enzymatic
hydroxylation at C(4) (Figure 2).11 This problem, however, can be solved by replacing the 4-CH2
group with a 4-CF2 fragment. Incorporation of fluorine atoms into the cyclohexane ring prevents
its oxidative degradation, since the C-F bond (116 kcal/mol) is significantly stronger than the C-
H bond (99 kcal/mol).12,13
3

ACCEPTED MANUSCRIPT
1
2
6
5
3
4
F
F
H
H
metabolic stability
metabolic oxidation
is blocked
metabolic oxidation
Figure 2. Comparison of metabolic stabilities of cyclohexane- and 4,4-
difluorocyclohexane motifs.
Several examples of this concept are outlined in Figure 3. Neurokinin-2 antagonist 2,14
antiosteoporosis agent 315 and bradykinin B1 antagonist 416 recently reached the
preclinical/clinical trials. In all cases, incorporation of fluorine atoms into the cyclohexane ring
prevented the initially observed metabolic hydroxylation at 4-CH2.17 It is also worth mentioning
that introduction of the difluoro-unit into the launched antiretroviral drug Maraviroc, along with
improving the metabolic profile, additionally reduced the compound`s toxicity.18
F
N N
F
N
O
N
N
N
HN
O
N
HO
N
F
O
MeS
NH
CN
Cl
F
F
Cl
F
F
Cl
F
F
4
2
3
antiosteoporosis agent
bradykinin B1 antagonist
NK2 antagonist
H
N
N
N
O
N
N
Ph
Maraviroc
antiretroviral drug
F
F
Figure 3. Pharmacologically relevant compounds with 4,4-difluorocyclohexane motif:
neurokinin-2 antagonist 2 (Pfizer, 2005); antiosteoporosis agent 3 (Merck, 2006); bradykinin B1
antagonist 4 (Merck, 2008); antiretroviral drug Maraviroc (Pfizer, 2007).
Given the high pharmaceutical potential of amino acid 1 and sensitivity of substituted
cyclohexanes to metabolic hydroxylation at 4-CH2, fluorinated building block 5 (Figure 4) seems
to be conceptually useful for drug discovery projects. Importantly, incorporation of two fluorine
atoms into amino acid 1 at C(4) preserves its achiral structure. Preparation of compound 5,
however, is not described in the open literature. Only the synthesis of a derivative of 5 is briefly
mentioned in a recent patent19 starting from a rather exotic non-commercially available starting
material. Neither the detailed experimental procedures, nor the full compound characterizations
are provided. In the present work, therefore, the practical synthesis of amino acid 520 from a
4

ACCEPTED MANUSCRIPT
commercially available starting compound has been developed. An impact of fluorine atoms on
the conformation, lipophilicity, acidity and fluorescent properties of the amino acid has also been
studied.
H2N
CO2H
F
F
5
Figure 4. 1-Amino-4,4-difluorocyclohexanecarboxylic acid (5).
Results and Discussion
Retrosynthetic analysis. In a search for a convenient starting point for the synthesis of
amino acid 5, we pointed our attention to the substituted cyclohexanone 6. This compound
contained two carbonyl groups at the needed C(1) and C(4) positions of the cyclohexane ring.
Moreover, one carbonyl group was already protected by the ketal moiety, allowing thus the
stepwise transformation of the functional groups. Finally, ketone 6 was commercially available21
and rather cheap.22
Incorporation of fluorine atoms into the cyclohexane skeleton (7) was planned to be
performed by fluorination of the carbonyl group with morpholinosulfur trifluoride (morph-
DAST, O(CH2CH2)2NSF3).23,24 Subsequent construction of the aminocarboxylate moiety was
expected to be realized by Bucherer-Berg reaction25 at the other carbonyl group (8) followed by
hydrolysis of the formed hydantoin 9. Implementation of this strategy led to the retrosynthetic
approach depicted in Scheme 1.
O
Bucherer-Berg
reaction
NH
O
H2N
F
CO2H
F
HN
O
F
F
F
F
5
9
8
O
O
morph-DAST
O
F
O
F
O
7
6
Scheme 1. Retrosynthetic analysis of amino acid 5.
Synthesis. Fluorination of ketone 6 by morph-DAST in dry dichloromethane at room
temperature afforded a 4:1 mixture of the required difluoro compound 7 and the by-product vinyl
fluoride 10 in 92% combined yield (Scheme 2).26 Practical separation of this mixture, however,
was not effective. Optimization studies undertaken to influence the ratio of products by changing
the reaction temperature and stoichiometry of the reagents were unsuccessful.
5

ACCEPTED MANUSCRIPT
O
O
O
O
O
F
O
F
O
F
O
F
O
O
KMnO4
morph-DAST
CH2Cl2
92%
OH
O
F
O
6
7
10
7
17
O
O
NH
(NH4)2CO3
KCN
Boc2O, DMAP
NEt3, CH2Cl2
O
F
O
F
HN
aq. HCl
H2O
74%
O
60%
over 2 steps
96%
F
F
over 2 steps
F
F
7
8
9
O
O
Boc
1. 2N KOH
2. cation exchange
chromatography
NH
F
N
H2N
F
CO2H
N
N
Boc
O
Boc
O
45%
7 steps
18% yield
F
F
F
F
15
5
16 (6 %)
Scheme 2. Synthesis of amino acid 5.
Formation of vinyl fluoride co-product from treatment of ketones with DAST and its
analogues is described in the literature and appears difficult to control.26,27 Fluorination of
substituted cyclohexanones usually affords a mixture of products, which are difficult to separate
(Scheme 3).26,28 Moreover, it seems that no general method on the effective separation of such
compounds is known.
R
F-
R
R
R
R
F
F
SF3NR2
-F-
F-
7
-FSONR2
-F-
F
O
R
+O
O
F
SF2NR2
SF2NR2
6
-H+
F
10
Scheme 3. Mechanism of formation of fluorocyclohexene 10.
In this context, a recent publication from Pfizer on the optimized industrial preparation of
Maraviroc (Figure 3) is worth mentioning.29 Acid 11 is a key intermediate in the synthesis of the
drug (Scheme 4). The authors reported that fluorination of cyclohexanone 12 with DAST
afforded a mixture of product 13 and fluorovinyl 14. Hydrolysis of the ester group gave a
6

ACCEPTED MANUSCRIPT
mixture of the corresponding acids, from which compound 11 was isolated in 51% yield after
column chromatography followed by two crystallizations. The moderate yield of that optimized
transformation aimed at bulk production of the launched drug, indicates that separation of
difluorocyclohexane / fluorocyclohexene derivatives does still represent a significant challenge.
CO2Et
CO2Et
CO2Et
CO2H
DAST
1. NaOH
2. Column chromatography
3. Two crystallizations
51%
CH2Cl2
96%
F
F
F
F
O
12
F
14
13
11
Scheme 4. Literature synthesis of acid 11, a key component of Maraviroc (Ref. 29).
Since a direct separation of the mixture 7/10 was not effective, oxidation of alkene 10 into
compound 17 was attempted next. Alcohol 17 seemed to be much more hydrophilic than 7, so
that their resolution was expected to be easier to perform. A solution of 7/10 in dichloromethane
was treated with aq. solution of KMnO4 at room temperature under vigorous stirring. The
1
heterogeneous oxidation of the double C=C bond was very slow as monitored by H-NMR
spectroscopy. After 24 h, however, the reaction was completed. Unexpectedly, washing the
crude reaction mixture with water afforded the pure crystalline product 7 in 74% yield, whereas
the putative water-soluble alcohol 17 was completely extracted. It is important to note, that the
developed simple protocol on the quantitative isolation of difluoromethyl compound 7 from the
mixture 7/10 seems to be rather general and we believe that it will find further application in
fluoroorganic chemistry, where formation of the by-product vinyl fluorides is common.
Acidic cleavage of the ketal group in compound 7 was next studied. The transformation
smoothly proceeded by stirring a solution of 7 in aq. hydrochloric acid / isopropanol mixture at
room temperature for 96 h. High volatility of compound 8, however, hindered its quantitative
isolation. Therefore, in order not to loose the product during the work-up process, the reaction
mixture with crude ketone 8 was directly used in the next synthesis step. The pH value was
adjusted first to 8-9 by aq. K2CO3 solution, followed by addition of KCN and (NH4)2CO3 for the
Bucherer-Berg transformation. Heating the reaction at reflux for 48 h afforded hydantoin 9 in
60% isolated yield after two steps.
Hydrolysis of the hydantoin ring is usually performed under rather harsh conditions, e. g.
conc. NaOH/Ba(OH)2 at temperatures >100 °C, or refluxing conc. HBr, which is not compatible
with many functional groups.30 This problem was partially solved by Rebek and coworkers,31
who showed that N,N’-bis(tert-butyloxycarbonyl)hydantoins could be hydrolyzed with dilute
aqueous alkali solution at room temperature. Having this transformation in mind, hydantoin 9
was first converted into the bis-Boc-derivative 15 by treating with Boc2O/NEt3/DMAP.
Hydrolysis of 15 with aq. 2N KOH at room temperature for 10 h followed by cation-exchange
chromatography on CU-2 resin afforded the requested amino acid 5 in 45% yield. Interestingly,
along with the target amino acid 5, the side product 16 was also obtained in 6% yield.
Thus, the synthesis of amino acid 5 was performed in seven steps from ketone 6 in 18%
overall yield.
Optimized synthesis. The practical usefulness of building blocks in drug discovery
projects strongly depends on the availability of simple well-validated synthetic approaches to
them.32 The shorter and easier the synthesis of a building block is, the higher are chances of that
compound to be involved in medicinal research. In this context, the synthesis of amino acid 5
was optimized to maximally reduce the number of synthetic manipulations.
7

ACCEPTED MANUSCRIPT
First, synthesis of difluorocyclohexane 7 was performed one-pot from ketone 6 (Scheme
5). Products 7/10, formed by fluorination of ketone 6, were not isolated from the reaction
mixture, but their solution in dichloromethane was directly treated with aq. KMnO4. The pure
crystalline product 7 was obtained in 73% yield without any purification. Next, one-pot synthesis
of hydantoin 9 from ketal 7 was performed following the initially elaborated procedures. Finally,
one-pot synthesis of amino acid 5 from hydantoin 9 was carried out. The key point was not to
isolate compound 15, but to treat the reaction mixture with aq. KOH directly. In order to prevent
formation of the side-product 16, the amount of KOH as well as the reaction time was increased
twice. As a result, amino acid 5 was obtained in 50% yield.
The optimized synthesis of compound 5 from ketone 6 was performed in 22% overall
yield in three steps.
O
1. aq. HCl
2. (NH4)2CO3
1. Boc2O, DMAP
NEt3, CH2Cl2
2. 3N KOH
NH
F
1. morph-DAST
2. KMnO4
3. H2O
O
O
O
F
O
F
H2N
F
CO2H
HN
O
KCN
60%
3 steps
22% yield
73%
3. cation exchange
chromatography
F
F
O
50%
6
7
9
5
Scheme 5. Optimized synthesis of amino acid 5.
Influence of fluorine on conformation. Having developed a short and effective route to
amino acid 5, physico-chemical properties of that compound were studied next. Indeed, when
replacing amino acid 1 by compound 5 in medicinal projects, one should always keep in mind,
that incorporation of fluorine atoms into the cyclohexane ring might not only improve the
compound metabolic stability, but it could also change its conformation, electronic properties,
lipophilicity and pKa of functional groups.13
Recently, O`Hagan et al. showed that the geometry of difluoromethylene-containing
cyclododecanes differed much from that of the non-fluorinated analogues.33 The observed
difference was caused by the distortion of bond angles. The authors studied a set of structures
with CH2CF2CH2 fragment at Cambridge Crystal Structure Database (CCSD) and found that the
average C-CF2-C angle was 118°, whereas the average F-C-F angle was 104° (Figure 5),
significantly wider and narrower, respectively, than the common tetrahedral angle of 109°.
ca. 104
F
F
R
R'
ca. 118
Figure 5. Average bond angles in CH2CF2CH2 motif according to Ref 33.
Following this scenario, it was important to find out whether the fluorine atoms influenced the
conformation of amino acid 5 and its derivatives. Since, only one cyclohexane with CH2CF2CH2
unit was found in CCSD - 1,1,4,4-tetrafluorocyclohexane,34 - single crystals of compounds
5*HCl, 7, 9 and 16 suitable for X-Ray diffractional analysis were obtained (Figure 6).
8

ACCEPTED MANUSCRIPT
NH2
O
F
HO
F
5*HCl
O
O
F
F
7
O
HN
O
NH
F
F
9
H
O
N
O
O
N
F
F
O
16
Figure 6. X-ray crystal structures of compounds 5*HCl, 7, 9 and 16.
9

ACCEPTED MANUSCRIPT
Comparison of geometrical parameters of amino acid 5*HCl and its none-fluorinated
analogue 1*HCl35 in crystal phase revealed a chair conformation of the cyclohexane ring with an
equatorial carboxylic group. In amino acid 5*HCl, angle ω = 105° was in a good agreement with
previous by O`Hagans observation (Table 1). On the other hand, angle θ = 115° was larger than
that of 111° in amino acid 1*HCl (Table 1), but significantly smaller than 118°, observed
previously by O`Hagan et al. The possible explanation relies on intrinsic conformational stability
of the cyclohexane ring, which hampers any valuable geometric deviations of the molecule
imposed by different substituents. As a result, conformations of the cyclohexane ring in
fluorinated compound 5*HCl and non-fluorinated one 1*HCl are very similar, as additionally
confirmed by the characteristic puckering parameters36 (Table 1). From the above data one can
conclude that the presence of fluorine atoms in amino acid 5 does not influence the overall
compound conformation.
Table 1. Geometric characteristics from X-ray diffraction data: C(3)-C(4) and C(4)-C(5)
bond length (Å); X-C-X (ω, deg) and C-CX2-C (θ, deg) bond angles; ring puckering parameters
(S; Θ, deg; Ψ, deg) and deviations of C(1) and C(4) atoms from the mean plane of the remaining
ring atoms in fluorinated compounds 5*HCl, 7, 9, 16 and none-fluorinated ones 1*HCl, 17
R
5
6
1
R
3
2
X
X = F or H
4
θ
ω
X
Parameter
C(3)-C(4), Å
C(4)-C(5), Å
ω, deg
θ, deg
S
5*HCl
1.508(7)
1.476(7)
104.8(4)
114.6(4)
1.09
1*HCla
1.505(3)
1.505(3)
99.0(3)
111.0(2)
1.10
17a
1.504(3)
1.514(3)
106.0(2)
111.1(2)
1.14
7
9
16
1.506(1)
1.506(2)
104.23(8)
114.1(1)
1.11
1.488(6)
1.486(6)
103.9(2)
114.9(3)
1.13
1.494(3)
1.485(2)
104.3(1)
114.6(1)
1.08
Θ, deg
0.9
2.4
0.2
2.8
0.6
1.9
7.9
0.0
9.0
19.4
21.0
18.7
Ψ, deg
dev. C(1), Å
-0.64
-0.61
-0.65
-0.71
-0.66
-0.61
dev. C(4), Å
0.60
0.66
0.62
0.59
0.66
0.61
a
Data for compounds 1*HCl and 17 are taken from refs. 35 and 37, respectively.
The above conclusion can also be made for compound 9. Both hydantoin 9 and its non-
fluorinated analogue 1737 (Table 2) had a chair conformation of the cyclohexane ring (Table 1)
with carboxamide group at the equatorial position. Although θ angles in hydantoins 9, 17 were
rather different (Table 1), no distortion of the overall compound geometry was observed.
Comparison of the compound geometrical parameters based on only X-ray diffraction
data may lead, however, to significant errors because conformational characteristics in the crystal
phase are sensitive to packing effects. Moreover, in crystal phase, amino acid 5 existed as a
hydrochloride salt, which might cause some conformational deformation of the cyclohexane ring
due to strong electrostatic interactions. In order to separate these effects we have optimized
geometry of compounds 5, 7, 9, 16 and their non-fluorinated counterparts 1, 18, 17, 19 using
M06-2X/6-311G(d,p) method. The calculation clearly indicated an influence of fluorine atoms
on the endocyclic C-C bond lengths at C(4) atom (Table 2). The presence of the fluorine
10

ACCEPTED MANUSCRIPT
shortened C(3)-C(4) and C(4)-C(5) bonds compared to those of non-fluorinated analogues. Also,
decrease of ω angle and considerable increase of θ angle were observed (Table 2). Nevertheless,
replacement of hydrogen atoms for fluorine at C(4) did not influence the conformation of the
saturated ring.
Table 2. Geometric characteristics from quantum-chemical calculations using M06-2X/6-
311G(d,p) method: C(3)-C(4) and C(4)-C(5) bond length (Å); X-C-X (ω, deg) and C-CX2-C (θ,
deg) bond angles; ring puckering parameters (S; Θ, deg; Ψ, deg) and deviations of C(1) and C(4)
atoms from the mean plane of the remaining ring atoms in compounds 5, 7, 9, 16 and their non-
fluorinated analogues 1, 18, 17, 19.
O
H
N
O
O
NH2
NH
O
HN
O
HO2C
N
Boc
O
1*HCl
18
17
19
Parameter
5
1
7
18
9
17
16
19
C(3)-C(4), Å
C(4)-C(5), Å
ω, deg
θ, deg
S
1.511
1.513
106.0
114.0
1.15
1.9
1.530
1.531
107.0
111.1
1.16
1.2
1.512
1.512
105.8
113.9
1.14
0.7
1.531
1.531
107.0
110.8
1.15
1.513
1.513
106.2
114.0
1.14
1.0
1.530
1.530
107.0
111.0
1.15
0.1
1.511
1.511
105.9
113.8
1.12
1.4
1.529
1.529
107.1
110.8
1.12
2.6
Θ, deg
0.5
16.7
0.71
-0.63
7.9
0.70
-0.67
28.4
0.67
-0.64
26.2
0.65
-0.68
5.5
0.68
-0.63
28.0
0.67
-0.67
0.0
0.64
-0.64
0.3
0.62
-0.69
Ψ, deg
dev. C(1), Å
dev. C(4), Å
In summary, incorporation of fluorine atoms into amino acid 1 and its derivatives at C(4)
did not have any valuable effect on their conformation. In particular, introduction of fluorine into
1 slightly changed the angle θ, but preserved the chain conformation of the cyclohexane ring, and
its characteristic parameters. Therefore, one can safely replace amino acid 1 in bioactive
compounds with the residue of 5 with no change of the compound geometry.
Impact of fluorine on fluorescence. With the aim to estimate the influence of fluorine
atoms on amino acid 1 comprehensively, fluorescence measurements of the both 1 and 5 were
performed. Excitation spectra of amino acids in aqueous solution (λem = 441 nm) had one
maximum at 347 nm (Supporting information). Under excitation at 347 nm, the fluorescence
emission spectra still had one identical maximum at 441 nm. Importantly, however, that though
the fluorescence and emission spectra of both compounds had the same maxima, the spectra of 5
had ~30 times higher intensity than those of amino acid 1 under identical concentration. This
difference clearly indicated the significant change of the polarity and lipophilic properties of the
amino acid 1 upon incorporation of fluorine atoms. The determined quantum yield of compound
5 in aqueous solution was 0.10. Supplemented with blocked metabolism of 5 at C(4) atom, the
above phenomena suggest the use of the amino acid as a fluorescent label to study, for example,
the mechanism of translocation and/or metabolic degradation of the corresponding bioactive
compounds.
11

ACCEPTED MANUSCRIPT
Impact of fluorine on pKa. In medicinal chemistry, pKa of functional groups is an
important characteristic of drugs, as it strongly effects their binding affinity, selectivity,
lipophilicity and toxicity.38 An influence of fluorine atoms on pKa of aminocarboxylate moiety in
compound 5 was studied next.39 Amino acid 1 was used as a reference point. pKa measurements
were performed by potentiometric titration (Supporting Information).
+H3N
CO2H
+H3N
CO2H
pKa:
9.3
2.6
10.5
2.9
F
F
5*HCl
1*HCl
Figure 7. pKa values for the both carboxylic and amino groups in 1*HCl, 5*HCl.
+
The measured pKa values of amino acid 1 were 2.9 (CO2H) and 10.5 (NH3 ), while those
+
in compound 5 were 2.6 (CO2H) and 9.3 (NH3 ) (Figure 7). As expected, incorporation of
fluorine atoms slightly increased an acidity of the carboxylic function with ∆pKa = -0.3. Basicity
of the nitrogen atom, on the contrast, was significantly reduced by more than one magnitude of
order with ∆pKa = -1.2. The different impact of fluorine atoms on pKa of amino and carboxylic
groups is caused by the dependence of an inductive effect on the number of chemical bonds. In
compound 5 hydrogen atoms at amino group are separated from fluorine by six chemical bonds,
while that of carboxylic function – by seven.
Toxicity is frequently associated with basic nitrogen atoms.40 In drug design therefore
toxicity of bioactive compounds is often reduced by incorporating electronegative substituents
near the basic center. Given the fact, that introduction of fluorine atoms into amino acid 1 does
not have any valuable effect on the compound conformation, but does effectively reduce basicity
of the nitrogen atom, amino acid 5 seems to be a promising replacement for compound 1, once
high toxicity of the corresponding derivatives is observed.
Impact of fluorine on lipophicility. Lipophilicity is another key characteristic of drugs.
Indeed, bioactive compounds must be lipophilic enough to effectively cross the biomembrane.
On the other hand, highly lipophilic compounds have poor water solubility.6 For oral drugs,
therefore, it is important to have the right balance between these opposite effects.
To evaluate the impact of fluorine on lipophilicity; compounds 17 and 9 were compared.
The measured logD (pH = 7.4) of compound 17 was 1.5, whereas the corresponding value of
difluoro hydantoin 9 was 0.5, with a significant decrement of ∆logD7.4 = -1.0 (Figure 8). To
understand this impressive effect, one should remember that LogD value depends on the both
intrinsic lipophilicity (logP) of a compound and the number of neutral molecules at a given pH.
Due to polarization of C-F bond, incorporation of two fluorine atoms into the cyclohexane ring
brings polarity to molecule 9, which decreases its intrinsic lipophilicity and hence the logD
value.41 Second, incorporation of fluorine obviously increases acidity of the imide NH proton.
Therefore, at pH = 7.4, the number of none-dissociated molecules of hydantoin 9 is lower than
that of compound 17, which additionally reduces logD of 9.42
12

ACCEPTED MANUSCRIPT
O
O
NH
NH
F
HN
HN
O
O
logD7.4
:
1.5
0.5
F
17
9
Figure 8. The measured logD7.4 values of compounds 17 and 9.
As previously mentioned, derivatives of amino acid 1 are highly lipophilic because of the
saturated cyclohexane skeleton. Therefore, once a low solubility of bioactive derivatives of 1 is
observed, amino acid 5 can be challenged as a replacement of 1 to decrease the compound LogD
value, which might increase the water solubility.
Conclusion
1-Amino-4,4-difluorocyclohexanecarboxylic acid 5 was designed as an analogue of the
pharmacologically relevant amino acid 1. An optimized synthesis of 5 was performed in three
steps from commercially available ketone 6 in 22% overall yield. Incorporation of fluorine atoms
into amino acid 1 at C(4) did not change its geometry, but did have a profound effect on the
fluorescent properties, pKa and logD characteristics. Therefore, amino acid 5 can be very useful
in those medicinal projects, where biologically active derivatives of amino acid 1 a) are
metabolically unstable; b) toxic; c) have low water solubility. Also, we suggest the use of amino
acid 5 as a fluorescent label to study, for example, the mechanism of translocation and/or
metabolic degradation of the corresponding bioactive compounds. With rapid scalable synthesis
and herein reported studies, we believe that building block 5 will find wide practical application
in drug discovery projects in the future.
Experimental part
Solvents were purified according to standard procedures. Another starting materials were
given by Enamine. Column chromatography was performed using Kieselgel Merck 60 (230-400
1
mesh) as the stationary phase. H-, 19F-, 13C-NMR spectra were recorded either on Bruker
Avance 500 spectrometer (at 500 MHz, 470 MHz and 125 MHz) or on Varian Unity Plus 400
spectrometer (at 400 MHz, 377 MHz and 101 MHz). Chemical shifts are reported in ppm
downfield from TMS (1H, 13C) or CFCl3 (19F) as internal standards. Mass spectra were recorded
on Agilent 1100 LCMSD SL instrument by chemical ionization (CI).
8,8-Difluoro-1,4-dioxaspiro[4.5]decane (7)
Method A:
A solution of ketone 6 (4.40 g, 28.2 mmol) in dry CH2Cl2 (100 mL) was cooled to 0 °C
under argon atmosphere. Morph-DAST (11.85 g, 67.7 mmol, 2.4 equiv) was added dropwise
upon stirring. The reaction mixture was allowed to warm to a room temperature and was stirred
for 72 h. A saturated solution of NaHCO3 in water (50 mL) was added and the reaction mixture
was stirred for 10 min to quench the unreacted fluorinating agent. The organic layer was
separated and the water phase was extracted with CH2Cl2 (2×50 mL). Combined organic phase
was dried over Na2SO4 and evaporated under reduced pressure at 30 °C to afford the mixture
of compounds 7/10 = 4/1 (4.61 g, 25.8 mmol, 92% yield) as a colourless oil. Products are very
volatile! Evaporation time must be minimized. All amount of compounds 7/10 was dissolved in
CH2Cl2 (100 mL). Solution of KMnO4 (5.7 g, 36 mmol, 10 eq) in water (100 mL) was added to
13

ACCEPTED MANUSCRIPT
the combined organic phases and the formed suspension was vigorously stirred for 24 h. The
organic layer was separated and the water phase was extracted with CH2Cl2 (2×100 mL).The
combined organic phases were dried over Na2SO4, and evaporated under vacuum to provide
pure compound 7 (3.38 g, 19.0 mmol, 74% yield; 67% yield over two steps from ketone 6) as
an colourless oil, which crystallized upon storage. M.p. 48-49 °С.
Method B:
A solution of ketone 6 (4.40 g, 28.2 mmol) in dry CH2Cl2 (100 mL) was cooled to 0 °C
under argon atmosphere. Morph-DAST (11.85 g, 67.7 mmol, 2.4 equiv) was added dropwise
upon stirring. The reaction mixture was allowed to warm to a room temperature and was stirred
for 72 h. A saturated solution of NaHCO3 in water (50 mL) was added and the reaction mixture
was stirred for 10 min to quench the unreacted fluorinating agent. The organic layer was
separated and the water phase was extracted with CH2Cl2 (2×50 mL). Solution of KMnO4 (5.7 g,
36 mmol, 10 eq) in water (100 mL) was added to the combined organic phase and the formed
suspension was vigorously stirred for 24 h. The organic layer was separated and the water phase
was extracted with CH2Cl2 (2×100 mL).The combined organic phases were dried over Na2SO4,
and evaporated under vacuum at 30 °C to provide pure compound 7 (3.65 g, 20.5 mmol, 73%
yield) as an colourless oil, which crystallized upon storage. The crystals of 7 were suitable for X-
Ray diffractional study. M.p. 48-49 °С. The product is very volatile! Evaporation time must be
minimized in order to avoid sublimation of the product into a rotary evaporator.
1H NMR (500 MHz; CDCl3; Me4Si), δ: 3.98 (4H, s, OCH2CH2O), 2.07 (4H, tt, J = 13.5, 6.6 Hz,
CH2CF2), 1.82 (4H, t, J = 6.6 Hz, CH2CH2CF2).
1
13C NMR (125 MHz; CDCl3; Me4Si), δ: 122.8 (q, J(C,F) = 241.3 Hz, CF2), 107.3 (s,
2
3
C(OCH2)2), 64.5 (s, OCH2), 31.4 (t, J(C,F) = 25.1 Hz, CH2CF2), 31.0 (t, J(C,F) = 6.3 Hz,
CH2CH2CF2).
19F NMR (376 MHz; CDCl3; CFCl3), δ: -100.16 (s, CF2).
MS (m/z): 178 (M)+.
IR (KBr), cm-1: 2975, 2889 (ν Csp3-H), 1121 (ν C-O).
Anal. Calcd for C8H12F2O2: C, 53.93; H, 6.79. Found: C, 53.55; H, 6.44.
4,4-Difluorocyclohexanone (8)
A suspension of ketale 7 (20.0 g, 112.2 mmol) in 20% aq. HCl (16 ml) was vigorously
stirred at 100 °C for 3 h. After cooling to a room temperature NaCl (ca. 3 g) was dissolved in the
water phase. The water phase was then extracted with CH2Cl2 (3 * 50 ml). The combined organic
phase was washed with water (50 ml), dried over Na2SO4 and slowly evaporated under reduced
pressure at 20 °C. The product is very volatile. The formed material was a 1/1 mixture of product
8 and starting material 5. After repeating the described above cycle, the obtained material was
8/5 = 7/3. The product of 95% purity (8) was obtained after repeating the described above
procedure for five times in total. White crystalline (11.0 g, 71% yield). M. p. 31-32 °С.
1H NMR (500 MHz; CDCl3; Me4Si), δ: 2.50 (4H, t, 3J(H,H) = 6.5 Hz, CH2CH2CF2), 2.28 (4H, tt,
3J(H,F) = 13.0, 3J(H,H) = 6.5 Hz, CH2CF2).
13C NMR (125 MHz; CDCl3; Me4Si), δ: 207.0 (s, C=O), 122.8 (q, 1J(C,F) = 240.0 Hz, CF2), 36.5
(t, 3J(C,F) = 3.0 Hz, CH2CH2CF2), 32.6 (t, 2J(C,F) = 25.0 Hz, CH2CF2).
19F NMR (376 MHz; CDCl3; CFCl3), δ: -100.81 (p, 3J(F,H) = 13.0 Hz, CF2).
MS (m/z): 134 (M)+.
IR (KBr), cm-1: 2981 (ν Csp3-H), 1719 (ν C=O).
Anal. Calcd for C6H8F2O: C, 53.73; H, 6.01. Found: C, 53.42; H, 5.78.
8,8-Difluoro-1,3-diazaspiro[4.5]decane-2,4-dione (9)
1N aq. HCl (25 mL) was added to a solution of compound 7 (2.55 g, 14.3 mmol) in
14

ACCEPTED MANUSCRIPT
isopropanol (50 mL). The reaction mixture was stirred at room temperature for 96 h. 2N aq.
KOH solution was added to adjust pH value of the solution to 8-9. The reaction mixture was
transferred into a glass tube followed by addition of KCN (1.40 g, 21.5 mmol) and (NH4)2CO3
(4.10 g, 42.7 mmol). The tube was closed, and the reaction mixture was heated at 100 °C for 24
h. After cooling to a room temperature, the reaction mixture was transferred into a flask and the
solvent was evaporated to ~ 30 mL. The formed solid was filtered off, washed with water (2×10
mL) on the filter and dried on air to provide hydantoin 9 (1.76 g, 8.6 mmol, 60% yield) as a
white solid. M.p. > 220 °С.
The crystals of 9 suitable for X-Ray diffractional study were obtained by a slow evaporation of a
diluted solution of 9 in isopropanol.
1H NMR (500 MHz; DMSO-d6; Me4Si), δ: 8.51 (1H, br. s, NH), 2.18-2.00 (4H, broad m), 1.80
(2H, m), 1.74 (2H, m).
1
13C NMR (125 MHz; DMSO-d6; Me4Si), δ: 178.2 (s, CO), 156.94 (s, CO), 123.5 (t, J(C,F) =
240.0 Hz, CF2), 60.6 (s, C(NH)CO), 30.7 (s, CH2), 29.0 (t, 2J(C,F) = 23.9 Hz, CH2CF2).
19F NMR (377 MHz; DMSO-d6; CFCl3), δ: -91.31 (br. d, 2J(F,F) = 233.7 Hz, CFF), -100.31 (br.
d, 2J(F,F) = 233.7 Hz, CFF).
MS (m/z): 204 (M)+.
IR (KBr), cm-1: 3199 (broad, ν N-H), 1769 (νassym C=O), 1734 (νsym C=O).
Anal. Calcd for C8H10F2N2O2: C, 47.06; H, 4.94; N, 13.72. Found: C, 47.41; H, 4.74; N, 13.35.
1-Amino-4,4-difluorocyclohexanecarboxylic acid (5)
Method A:
A mixture of compound 9 (935 mg, 4.6 mmol), NEt3 (834 mg, 9.2 mmol) and catalytic
amount of DMAP (12 mg) in 1,2-dimethoxyethane (30 mL) was treated with Boc2O (5.0 g, 22.9
mmol). The reaction mixture was vigorously stirred for 12 h. Another portion of DMAP (12 mg)
was added, and the mixture was stirred for additional 12 h. The solvent was removed under
reduced pressure and EtOAc (50 mL) / water (10 mL) were added to the residue. The mixture
was stirred for 10 min. Organic phase was separated and washed with 1N aq. HCl (2×10 mL)
and with water (10 mL). Organic phase was dried over Na2SO4 and evaporated under reduced
pressure. The residue was additionally dried at 50 °C at 0.1 mm Hg for 1 h to afford di(tert-
butyl)-8,8-difluoro-2,4-dioxo-1,3-diazaspiro[4.5]decane-1,3-dicarboxylate (15) (1.78 g, 4.4
mmol, 96% yield). The obtained compound was of ~ 90% purity and was used in the next
1
synthesis step without additional purification. Yellowish oil. H NMR (500 MHz; DMSO-d6;
Me4Si), δ: 2.90 (2H, m, CH2), 2.55 (2H, m, CH2), 2.11 (2H, m, CH2), 1.83 (2H, m, CH2), 1.61-
1.55 (18H, 2 s, C(CH3)3+ C(CH3)3). All amount of compound 15 was dissolved in 1,2-
dimethoxyethane (30 mL) followed by addition of 2N aq. KOH (23 mL, 46 mmol, 10 eq). The
reaction mixture was vigorously stirred at room temperature for 12 h. The solvents were
removed under vacuum, and water (30 mL) was added. The water phase was washed with
CH2Cl2 (2×5 mL). Combined organic phases were discarded, and the water phase was acidified
with aq. HCl to pH ~ 4. The formed precipitate was filtered off to provide pure compound 16 (84
mg, 0.3 mmol, 6% yield). The crystals of 16 suitable for X-Ray diffractional study were obtained
by slow evaporation of a diluted solution of 16 in isopropanol. M.p. > 220 °С.
1H NMR (500 MHz; DMDO-d6; Me4Si), δ: 2.63 (2H, t, J = 5.0 Hz, CH2), 2.01 (2H, m, CH2),
1.92 (2H, m, CH2), 1.47 (s, 9H, C(CH3)3).
13C NMR (125 MHz; DMDO-d6; Me4Si), δ: 177.2 (s, CO), 156.9 (s, CO), 148.4 (s, CO, Boc),
123.0 (t, 1J(C,F) = 241.3 Hz, CF2), 83.1 (s, C(CH3)3), 62.9 (s, C(NH)CO), 29.7 (t, 2J(C,F) = 25.0
Hz, CH2CF2), 28.2 (s, C(CH3)3), 26.9 (d, 3J(C,F) = 6.4 Hz, CH2CH2CF2).
19F NMR (376 MHz; DMDO-d6; CFCl3), δ: -91.77 (d, 2J(F,F) = 236.9 Hz, CFF), -100.83 (dtt, J =
236.9, 37.7, 11.3 Hz).
MS (m/z): 304 (M)+.
15

ACCEPTED MANUSCRIPT
IR (KBr), cm-1: 3247 (broad, ν N-H), 2987, 2944 (ν Csp3-H), 1826, 1791, 1744, 1739 (ν C=O),
1146, 1109 (ν C-O).
Anal. Calcd for C13H18F2N2O4: C, 51.31; H, 5.96; N, 9.21. Found: C, 51.01; H, 5.65; N, 8.89.
pH value of the collected water phase was again adjusted to ~ 9 with aq. KOH. The solution was
evaporated under reduced pressure to ~ 5 mL, followed by purification on cation-exchange resin
Cu-2. Elution with water was first performed to remove all inorganic salts. Subsequent elution
with 25% NH3 afforded pure amino acid 5 (360 mg, 2.0 mmol, 45% yield from 15. 43% yield
over two steps from 9) as a white solid. M.p. > 220 °С.
1H NMR (500 MHz; D2O + CF3COOD; Me4Si), δ: 2.21 (2H, m), 2.17-2.05 (2H, broad m), 2.05-
1.85 (4H, broad m).
13C NMR (125 MHz; D2O + CF3COOD; Me4Si), δ: 172.28 (s, CO2H), 122.29 (t, 1J(C,F) = 238.8
Hz, CF2), 57.44 (s, C(NH2)CO2H), 28.49 (t, 2J(C,F) = 25.0 Hz, CH2CF2), 28.30 (s, CH2).
19F NMR (470 MHz; D2O; CFCl3), δ: -95.95 (br. d, 2J(F,F) = 233.0 Hz, CFF), -99.56 (br. d,
2J(F,F) = 233.0 Hz, CFF).
Anal. Calcd for C7H11F2NO2: C, 46.93; H, 6.19; N, 7.82. Found: C, 46.63; H, 6.01; N, 7.54.
The crystals of 5*HCl suitable for X-Ray diffractional study were obtained by a slow
evaporation of a diluted solution of 5 in aq. hydrochloric acid.
+
IR (KBr), cm-1: 3383 (broad, ν NH3 ), 2944 (ν Csp3-H), 1748 (ν C=O).
Method B:
A mixture of compound 9 (935 mg, 4.6 mmol), NEt3 (834 mg, 9.2 mmol) and catalytic
amount of DMAP (12 mg) in dimethoxyethane (30 mL) was treated with Boc2O (5.0 g, 22.9
mmol). The reaction mixture was vigorously stirred for 12 h. Another portion of DMAP (12 mg)
was added, and the mixture was stirred for additional 24 h. 2N aq. KOH (34 mL, 69 mmol, 15
eq) was added. The reaction mixture was vigorously stirred at room temperature for 24 h. The
solvents were removed under vacuum, and water (50 mL) was added. The water phase was
washed CH2Cl2 (2×10 mL). Combined organic phases were discarded, and the water phase was
evaporated under reduced pressure to ~ 10 mL. The obtained solution was purified on cation-
exchange resin Cu-2. Elution with water was first performed to remove all inorganic compounds.
Subsequent elution with 25% NH3 afforded pure amino acid 5 (410 mg, 2.3 mmol, 50% yield) as
a white solid.
Acknowledgement
This work was supported by Enamine Ltd. PM is very grateful to Dr. Andrei Bezdudny for the
generous gift of morph-DAST; and to Dr. Sergey Zozulya for the help with determination of
logD parameters, to Dr. Olena Litsis for IR measurements, and to Dr. Karen Tarasenko for the
help in the preparation of manuscript.
Supporting Information
Crystallographic data for compounds 5*HCl, 7, 9, 16 and structure description;
determination of pKa and logD values; fluorescence measurements.
16

ACCEPTED MANUSCRIPT
References
(1) (a) Kowalczyk, W.; Prahl, A.; Derdowska, I.; Dawidowska, O.; Slaninova, J.; Lammek, B. J. Med. Chem. 2004, 47,
6020. (b) Labudda-Dawidowska, O.; Wierzba, T. H.; Prahl, A.; Kowalczyk, W.; Gawinski, L.; Plackova, M.; Slaninova, J.;
Lammek, B. J. Med. Chem. 2005, 48, 8055. (c) Sikorska, E.; Slusarz, Lammek, B. J. Peptide Res. 2006, 66, 30. (d)
Jastrzebska, B.; Derdowska, I.; Kowalczyk, W.; Machova, A.; Slaninova, J.; Lammek, B. J. Peptide Res. 2003, 62, 70. (e)
J. B. Bremner, P. A. Keller, S. G. Pyne, M. J. Robertson, K. Sakthivel, K. Somphol, D. Baylis, J. A. Coates, J. Deadman,
D. Jeevarajah, D. I. Rhodes Beilstein J. Org. Chem. 2012, 8, 1265. (f) Rodrigues-Ropero, F.; Zanuy, D.; Casanovas, J.;
Nussimov, R.; Aleman, C. J. Chem. Inf. Model. 2008, 48, 333. (g) Crisma, M.; Bonora, G. M.; Toniolo, C.; Basovo, A.;
Benedetti, E.; Di Blasio, B.; Pavone, V.; Pedone, C. Macromolecules 1988, 21, 2071. (h) Pavone, V.; Benedetti, E.;
Barone, V.; Di Blasio, B.; Lelj, F.; Pedone, C.; Santini, A.; Crisma, M.; Bonora, G. M.; Toniolo, C. Macromolecules 1988,
21, 2064. (i) Saviano, M.; Isernia, C.; Rossi, F.; Di Blasio, B.; Iacovino, R.; Mazzeo, M.; Pedone, C.; Benedetti, E.
Biopolymers 2000, 53, 189.
(2) (a) Garcia-Escheverria C.; Furet, P.; Gay, B.; Fretz, H.; Rahuel, J.; Schoepfer, J.; Caravatti, G. J. Med. Chem. 1998, 41,
1741. (b) Schoepfer, J.; Fretz, H.; Gay, B.; Furet, P.; Garcia-Echeverria, C.; End, N.; Caravatti, G. Bioorg. Med. Chem.
Lett. 1999, 9, 221.
(3) 147 derivatives. The search was performed in November 2011.
(4) For our activity in this field, see: (a) Kubyshkin, V.; Mykhailiuk, P.; Ulrich, A.; Komarov, I.
Synthesis 2009, 19, 3327. (b) Kubyshkin, V. S.; Mikhailiuk, P. K.; Komarov, I. V. Tetrahedron Lett. 2007, 48, 4061. (c)
Tkachenko, A. N.; Radchenko, D. S.; Mykhailiuk, P. K.; Grygorenko, O. O.; Komarov, I. V. Org. Lett. 2009, 11, 5674. (d)
Denisenko, A. V.; Mityuk, A. P.; Grygorenko, O. O.; Volochnyuk, D. M.; Shishkin, O. V.; Tolmachev, A. A.; Mykhailiuk,
P. K. Org. Lett. 2010, 12, 4372.
(5) (a) Mann, A. Conformational restriction and/or steric hindrance in medicinal chemistry. In The Practice of Medicinal
Chemistry, 3rd ed.; Wermuth, C. G., Ed.; Academic Press/Elsevier: Amsterdam, 2008; p. 363. (b) Martin, S. F. Pure Appl.
Chem. 2007, 79, 193. (c) Cowell, S. M.; Lee, Y. S.; Cain, J. P.; Hruby, V. J. Curr. Med. Chem. 2004, 11, 2785. (d) Harrold,
M. W. Am. J. Pharm. Edu. 1996, 60, 192.
(6) Lipinski, C. A. J. Pharmacol. Toxicol. Methods 2000, 44, 235.
(7) Gold, J. A.; Hegarty, C. P.; Deitch, M.W.; Walker, B. R. Antimicrob. Agents Chemother. 1979, 1, 55.
(8) Pazdur, R.; Redman, B. G.; Corbett, T.; Phillips, M.; Baker, L. H. Cancer Res. 1987, 47, 4213.
(9) Adami, S.; Supronik, J.; Hala, T.; Brown, J. P.; Garnero, P.; Haemmerle, S.; Ortmann, C. E.; Bouisset, F.; Trechsel, U.
J. Bone Min. Res. 2006, 21, S24.
(10) Falgueyret, J.-P.; Desmarais, S.; Oballa, R.; Black, W. C.; Cromlish, W.; Khougaz, K.; Lamontagne, S.; Masse, F.;
Riendeau, D. ; Toulmond, S.; Percival, M. D. J. Med. Chem. 2005, 48, 7535.
(11) Search in MDL Metabolite Database yielded 232 hits for metabolic hydroxylation of 1-monosubstituted and 1,1-
disubstituted cyclohexane derivatives at the 4th position of the cyclohexane ring.
(12) (a) Park, B. K.; Kitteringham, N. R.; ONeill, P. M. Annu. Rev. Pharmacol. Toxicol. 2001, 41, 443. (b) Hodgetts, K. J.;
Combs, K. J.; Elder, A. M.; Harriman, G. C. Ann. Rep. Med. Chem. 2010, 45, 429.
(13) (a) Purser, S.; Moore, P. R.; Swallow, S.; Gouverneur, V. Chem. Soc. Rev. 2008, 37, 320. (b) Hagmann, W. K. J. Med.
Chem. 2008, 51, 4359. (c) Böhm, H.-J.; Banner, D.; Bendels, S.; Kansy, M.; Kuhn, B.; Müller, K.; Obst-Sander, U.; Stahl,
M. ChemBioChem 2004, 5, 637. (d) O`Hagan, D. Chem. Soc. Rev. 2008, 37, 308.
(14) Middleton, D. S.; MacKenzie, A. R.; Newman, S. D.; Corless, M.; Warren, A.; Marchington, A. P.; Jones, B. Bioorg.
Med. Chem. Lett. 2005, 15, 3957.
(15) Crane, S. N.; Black, W. C.; Palmer, J. T.; Davis, D. E.; Setti, E.; Robichaud, J.; Paquet, J.; Oballa, R. M.; Bayly, C. I.;
McKay, D. J.; Somoza, J. R.; Chauret, N.; Seto, C.; Scheigetz, J.; Wesolowski, G.; Masse, F.; Desmarais, S.; Ouellet, M. J.
Med. Chem. 2006, 49, 1066.
(16) Kuduk, S. D.; Chang, R. K.; DiPardo, R. M.; Di Marco, C. N.; Murphy, K. L.; Ransom, R. W.; Reiss, D. R.; Tang, C.;
Prueksaritanont, T.; Pettibone, D. J.; Bock, M. G. Bioorg. Med. Chem. Lett. 2008, 18, 5107.
(17) For another examples on blocking the enzymatic hydroxylation of cyclohexane derivatives by replacing 4-CH2 for 4-
CF2, see: (a) Ontoria, J. M.; Bufi, L. L.; Torrisi, C.; Bresciani, A.; Giomini, C.; Rowley, M.; Serafini, S.; Bin, H.; Hao, W.;
17

ACCEPTED MANUSCRIPT
Steinkühler, C.; Jones, P. Bioorg. Med. Chem. Lett. 2011, 21, 5274. (b) Dong, Y.; Chollet, J.; Vargas, M.; Mansour, N. R.;
Bickle, Q.; Alnouti, Y.; Huang, J.; Keiser, J.; Vennerstrom, J. L. Bioorg. Med. Chem. Lett. 2010, 20, 2481. (c) Moriarty, K.
J.; Takahashi, H.; Pullen, S. S.; Khine, H. H.; Sallati, R. H.; Raymond, E. L.; Woska, J. R.; Jeanfavre, D. D.; Roth, G. P.;
Winters, M. P.; Qiao, L.; Ryan, D.; DesJarlais, R.; Robinson, D.; Wilson, M.; Bobko, M.; Cook, B. N.; Lo, H. Y.; Nemoto,
P. A.; Kashem, M. A.; Wolak, J. P.; White, A.; Magolta, R. L.; Tomczuk, B. Bioorg. Med. Chem. Lett. 2008, 18, 5545.
(18) Price, D. A.; Armour, D.; de Groot, M.; Leishman, D.; Napier, C.; Perros, M.; Stammen, B. L.; Wood, A. Bioorg.
Med. Chem. Lett. 2006, 16, 4633.
(19) Shizuka, N.; Sasatani, T.; Hiramatsu, Y.; Inagaki, M.; Odan, M.; Hashizume, H.; Mitsumori, S.; Fujii, Y.; Fukui, Y.;
Watanabe, H.; Sato, J. EP1806342, 2007.
(20) For another fluorinated analogues of amino acid 1 see: (a) Fustero, S.; Sanchez-Rosello, M.; Rodrigo, V.; Pozo, C.;
Sanz-Cervera, J. F.; Simon, A. OrgLett 2006, 8, 4129. (b) Fustero, S.; Sanchez-Rosello, M.; Rodrigo, Sanz-Cervera, J. F.;
Piera, J.; Simon-Fuentes, A.; Pozo, C. Chem. Eur. J. 2008, 14, 7019. (c) Shoup, T. M.; Goodman, M. M. J. Labelled Comp.
Radiopharm. 1997, 40, 46.
(21) Ketone 6 is commercially available at more than 100 chemical suppliers.
(22) 2012, Aldrich: 100 g / 375 USD.
(23) Morph-DAST was preferred over DAST as a result of its higher thermal stability and better yields of fluorinated
products in certain cases: (a) Messina, P. A.; Mange, K. C.; Middleton, W. J. J. Fluorine Chem. 1989, 42, 137. (b) Box, J.
M.; Harwood, L. M.; Whitehead, R. C. Synlett 1997, 571.
(24) (a) Mykhailiuk, P. K.; Radchenko, D. S.; Komarov, I. V. J. Fluorine Chem. 2010, 131, 221. (b) Mykhailiuk, P. K.;
Shishkina, S. V.; Shishkin, O. V.; Zaporozhets, O. A.; Komarov, I. V. Tetrahedron 2011, 67, 3091.
(25) (a) Bucherer, H. T.; Fischbeck, H. T. J. Prakt. Chem. 1934, 140, 69. (b) Bucherer, H. T.; Steiner, W. J. Prakt.
Chem. 1934, 140, 291.
(26) Fluorination of ketone 10 with commercially none-available diethylaminodifluorosulfinium tetrafuoroborate to afford
the none-inseparable mixture of compounds 11/14 is mentioned in (a) L'Heureux, A.; Beaulieu, F.; Laflamme, F.;
Couturier, M.; Bennett, C.; Clayton, S.; Tovell, D.; Bill, D. R.; Mirmehrabi, M.; Tadayon, S. J. Org. Chem.
2010, 75, 3401. (b) Beaulieu, F.; Beauregard, L.-P.; Courchesne, G.; Couturier, M.; Laflamme, F.; L'Heureux, A. Org.
Lett. 2009, 11, 5050.
(27) (a) Lal, G. S.; Pez, G. P.; Pesaresi, R. J.; Prozonic, F. M.; Cheng, H. J. Org. Chem. 1999, 64, 7048; (b) Middleton, W.
J. J. Org. Chem. 1975, 40, 574.
(28) (a) Allen, D. G.; Coe, D. M.; Cook, C., M.; Dowle, M. D.; Edlin, C. D.; Hamblin, J. N.; Johnson, M. R.; Jones, P. S.;
Knowles, R. G.; Lindvall, M. K.; Mitchell, C. J.; Redgrave, A. J.; Ward, P. WO2004/24728 (2004). (b) Kapoor, R.
Spindola, S. D. US2006/9477 (2006). (c) Denis, R.; Poisson, C.; Das, S. K.; Motorina, I.; Rej, R.; Yannopoulos, C, G.;
Chan Сhun Kong, L. WO2006/119646 (2006). (d) Mitsudera, H. WO2009/5110 (2009). (e) Connors, R. V.; Dai, K.;
Eksterowicz, J.; Fan, P.; Fisher, B.; Fu, J.; Li, K.; Li, Z.; Mcgree, L., R.; Sharma, R.; WO2009/126584 (2009). (f)
Goodfellow, V.; Rowbottom, M.; Dyck, B. P.; Tamiya, J.; Zhang, M.; Grey, J.; Vickers, T. D.; Pontillo, J.; Wu, D.
WO2004/81005 (2004). (g) Pu, Y. M.; Torok, D. S.; Ziffer, H.; Pan, X.-Q.; Meshnick, S. R. J. Med. Chem. 1995, 38, 4120.
(h) Rich, R. H.; Bartlett, P. A. J. Org. Chem. 1996, 61, 3916.
(29) Haycock-Lewandowski, S. J.; Wilder, A.; Ahman, J. Org. Process Res. Dev. 2008, 12, 1094.
(30) Kruger, G.; Houben-Weyl, Methoden der organischen Chemie; Vol. E5, G. Thieme Verlag: Stuttgart, 1985, p. 534 -
546.
(31) Kubik, S.; Meissner, R. S.; Rebek, J. Tetrahedron Lett. 1994, 35, 6635.
(32) (a) Ward, R. A.; Kettle, J. G. J. Med. Chem. 2011, 54, 4670. (b) Roughley, S. D.; Jordan, A. M. J. Med. Chem. 2011,
54, 3451.
(33) Skibinski, M.; Wang, Y.; Slawin, A. M. Z.; Lebl, T.; Kirsch, P.; O'Hagan, D. Angew. Chem. Int. Ed. 2011, 50, 10581.
(34) Dunitz, J. D.; Schweizer, W. B.; Seiler, P. Helv. Chim. Acta 1983, 66, 134.
(35) Belandria, L. M.; Delgado, G. E.; Mora, A. J.; Seijas, L. E.; Gonzalez, T. Acta Cryst. 2009, E65, o245.
(36) Zefirov, N.S.; Palyulin,V.A.; Dashevskaya, E.E. J. Phys. Org. Chem. 1990, 3, 147.
18

ACCEPTED MANUSCRIPT
(37) Gauthier, T. J.; Yokum, T. S.; Morales, G. A.; McLaughlin, M. L.; Liu, Y.-H.; Fronczek, F. R. Acta Crystallogr.,
Sect.C: 1997, 53, 1659.
(38) Morgenthaler, M.; Schweizer, E.; Hoffmann-Roeder, A.; Benini, F.; Martin, R. E.; Jaeschke, G.; Wagner, B.; Fischer,
H.; Bendels, S.; Zimmerli, D.; Schneider, J.; Diederich, F.; Kansy, M.; Muller, K. ChemMedChem 2007, 2, 1100.
(39) Schlosser, M. Angew. Chem. Int. Ed. 1998, 110, 1496.
(40) Jamieson, C.; Moir, M. E.; Rankovic, Z.; Wishart, G. J. Med. Chem. 2006, 49, 5029.
(41) Muller, K.; Faeh, C.; Diederich, F. Science 2007, 317, 1881.
(42) Wuitschik, G.; Carreira, E. M.; Wagner, B.; Fischer, H.; Parrilla, I.; Schuler, F.; Rogers-Evans, M.; Muller, K. J. Med.
Chem. 2010, 53, 3227.
19

ACCEPTED MANUSCRIPT
SUPPORTING INFORMATION
1-Amino-4,4-difluorocyclohexanecarboxylic acid as a
promising building block for drug discovery:
design, synthesis and characterization
Pavel K. Mykhailiuk,*a,b Viktoriia Starova,b Vladimir Iurchenko,a Svitlana V. Shishkina,c Oleg V.
Shishkin,c,d Oleksandr Khilchevskyi,e Olga Zaporozhetsb
aEnamine Ltd., Vul. Oleksandra Matrosova 23, 01103 Kyiv, Ukraine
bDepartment of Chemistry, Taras Shevchenko National University of Kyiv, Volodymyrska str., 64,
01601 Kyiv, Ukraine
cSSI, ´Institute for Single Crystalls,´ National Academy of Science of Ukraine, 60 Lenina ave.,
Kharkiv 61001, Ukraine
d Department of Inorganic Chemistry V.N.Karazin Kharkiv National University, 4 Svobody sq.,
Kharkiv 61022, Ukraine
e
Institute of Bioorganic Chemistry and Petrochemistry, National Ukrainian Academy of Science,
Murmanska 1, Kiev 02660, Ukraine
Pavel.Mykhailiuk@gmail.com
Table of contents
1. Determination of pKa values for amino acids 1 and 5.
2. Fluorescent properties of amino acids 1 and 5.
3. Determination of LogD of compounds 9 and 17.
4. X-Ray diffraction study of the compounds 5*HCl, 7, 9, 16
5. Copies of 1H NMR, 13C NMR and IR spectra
2
4
6
8
13
1

ACCEPTED MANUSCRIPT
1) Determination of pKa values for amino acids 1 and 5.
pKa values were obtained from the pH-metric titration curves using “Hyperquad 2000” program.
Obtained data
Amino acid
+
pK(COOH)
pK(NH3 )
H2N
CO2H
2,92
10,48
1
a
b
pH
12
pH
6
5
4
3
2
1
11
10
9
8
7
1
2
3
4
0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0
V, ml (HCl 0,0056 M)
V, ml (NaOH 0,056 M)
Titration curves of amino acid 1 with 0,056М HCl (a) and with 0,056М NaOH (b). V0= 10 mL.
Ve.p.= 1,6 mL (a); Ve.p.= 1,4 mL (b).
Obtained data
Amino acid
+
pK(COOH)
pK(NH3 )
H2N
F
CO2H
F
2,56
9,31
5
2

ACCEPTED MANUSCRIPT
a
b
pH
12
pH
6
11
10
9
5
4
3
2
8
7
6
1
2
3
4
V, ml (NaOH 0,056 M)
1
2
3
4
V, ml (HCl 0,056 M)
Titration curves of amino acid 5 with 0,056М HCl (a) and with 0,056М NaOH (b). V0= 10 mL.
Ve.p.= 1,4 mL (a); Ve.p.= 1,6 mL (b).
3

ACCEPTED MANUSCRIPT
2) Fluorescent properties of amino acids 1 and 5.
a (1)
b (5)
Excitation (1a, 1b) and fluorescence (2a, 2b) spectra of aqueous solution of amino acids 1 (а) and
5 (b). Сamino acid = 0,001 М, λem= 441 нм (1), λexi=347 нм (2), interval 10, 20. l= 1 см. The
spectra are referenced excluding the signal of water.
Fluorescence Quantum yield
For determination of quantum yield of compound 5, L-phenylalanine was chosen as a standard (ST),
with the known quantum yield in aqueous solution.1
The solutions were prepared by dissolving the compounds in bi-distilled water (pH ≈ 5,5). pI
(phenylalanine) = (2.2 + 9.1) / 2 = 5.6; pI (5) = (2.56+9.31) / 2 = 5.9. Under these conditions the
compounds exist in a zwitter-ionic form. Changing the pH value of the solution by adding bases or
acids decreased the fluorescent intensity; emission and excitation maximum were shifted to a long
wavelength region.
The quantum yield was calculated on the basis of dependence of integrated fluorescence intensity (I)
on the optical density (A):
1 Lakowicz, J. R.; Brouwer, A. M. Pure Appl. Chem., 2011, 83, 2213.
4

ACCEPTED MANUSCRIPT
I
40000
2
30000
20000
10000
0
1
0,01 0,02 0,03 0,04 0,05 0,06 0,07
Α
Linear plots for standard sample of phenylalanine (1) and substance 5. The data of fluorescence
intensity were obtained under excitation wavelength 258 nm.
ΦX = ΦST(GradX/GradST)(η2 /η2ST) = 0,024*(545027/131488)= 0,099
X
where,
ΦX - quantum yield of unknown substance
ΦST - quantum yield of standard
GradST – gradient = tgα of curve 1
GradX - gradient = tgα of curve 2
Similar results were also obtained under excitation at 230 and 347 nm wavelengths. The average
measured quantum yield for 5 in aqueous solution was 0,100±0,005.
5

ACCEPTED MANUSCRIPT
3) Determination of LogD of compounds 9 and 17
Partition ratio (or LogD) is a logarithm of the ratio of drug concentrations in two immiscible
solvents, typically pH-buffered water and n-octanol. It is a measure of hydrophobic/hydrophilic
properties of a given molecule. The partition ratio is determined using shake-flask method, which
consists of dissolving some of the solute of interest in a volume of n-octanol and an aqueous buffer of
choice, then measuring the concentration of the solute in each solvent, by using LC-MS.
Compounds:
Compound
Structure
MW
O
NH
HN
17
168.20
O
O
O
NH
F
HN
9
204.18
F
Materials and reagents:
DMSO (SIGMA-ALDRICH 34869 - CHROMASOLV® Plus, for HPLC, ≥99.7%)
PBS buffer (Helicon, Am-E404-100, pH 7.4)
n-Octanol (SIGMA-ALDRICH 472328 - ACS reagent, ≥99%)
Acetonitrile (SIGMA-ALDRICH 34998 - CHROMASOLV® Plus, for HPLC, ≥99.9%)
Equipment:
Multi Mix MTR 22 rotator
Eppendorf 5417R centrifuge
Shimadzu VP HPLC system and AB Sciex API3000 mass-detector.
Procedure:
Aqueous phase: n-octanol mix was pre-saturated during 24 hours.
10 mM compound DMSO stock was dissolved in aqueous phase to a final concentration of DMSO of
1%. Equal volume of n-octanol was added and mixed in rotator for 60 minutes. Phase separation was
assured by centrifugation for 3 minutes @14000rpm. Both phases were sampled. Each sample was
analyzed by LC-MS and the peak area ratios calculated to produce LogD values.
6

ACCEPTED MANUSCRIPT
Results:
Aq_Analyte Oct_Analyte
Mean
Compound
ID
LogD,
pH7.4
Peak Area Peak
Area D
LogD,
pH7.4
1.51
SD
RSD,%
2.3
(counts)
(counts)
43400
48200
44100
258500
298000
286000
1516600*
1437800*
1451900*
974500
933000
904000
34.94
1.54
1.48
1.52
0.58
0.50
0.50
0.035
17
17
17
9
9
9
29.83
32.92
3.77
3.13
3.16
0.52
0.045
8.7
7

ACCEPTED MANUSCRIPT
4) X-Ray diffraction study of the compounds 5*HCl, 7, 9, 16
Figure S1. Structure of compound 5 according to X-ray diffraction data. Thermal ellipsoids
are shown at the 50 % probability level.
The compound 5 is chloride of organic cation (Fig. S1) which exists in the crystal phase as
hydrate in the ratio 1:1.25. The water molecule with partial occupancy is disordered over two
positions (A and B) with equal populations what does not allow to determine the positions of the
hydrogen atoms at it. The positive charge of the cation is localized on the nitrogen atom. It is
confirmed by protonation of this atom (positions of hydrogen atoms were located from electron
density difference maps and the N1-C1 bond length (1.503(4) Å) which is elongated as compared
with mean value [1] for the Csp3-N bond (1.469 Å)).
The saturated ring adopts a chair conformation (the puckering parameters [2] are: S= 1.09, Θ
= 0.9°, Ψ = 7.9°). Deviations of the C1 and C4 atoms from the mean plane of the remaining atoms of
the ring are -0.64 Å and 0.60 Å, respectively. The protonated amino group has axial orientation (the
C5-C6-C1-N1 torsion angle is 69.7(4)°) what causes the appearance of the shortened intramolecular
contact H5b…H1Nb 2.25 Å (the van der Waals radii sum [3] is 2.34 Å). The carboxyl group is
located in equatorial position and it is almost orthogonal to the C6-C1 endocyclic bond (the C5-C6-
C1-C7 and C6-C1-C7-O1 torsion angles are -175.2(3)° and -101.8(5)°, respectively).
In the crystal phase the cations of 5 are bonded with anions and water molecules by
intermolecular hydrogen bonds: N(1)-H(1Na)…O(1w)’ (1-x, 0.5+y, -z) H…O 2.00 Å N-H...O 171°;
N(1)-H(1Nb)...Cl(1)’ (1+x, 1+y, z) H...Cl 2.39 Å N-H...Cl 162°; N(1)-H(1Nc)...O(1w)’ H...O 1.98 Å
N-H...O 170°; O(1w)-H(1wa)...Cl(1)’ (1+x, y, z) H...Cl 2.43 Å O-H...Cl 149°; O(1w)-
H(1wb)...Cl(1)’ (1-x, 0.5+y, -z) H...Cl 2.24 Å O-H...Cl 176°; O(2)-H...Cl(1)’ H...Cl 2.22 Å O-H...Cl
173°; C(5)-H(5a)...F(2)’ (2-x, -0.5+y, 1-z) H...F 2.50 Å C-H...F 169°; C(5)-H(5b)...Cl(1)’ (1+x, 1+y,
z) H...Cl 2.80 Å C-H...Cl 148°; C(6)-H(6a)...Cl(1)’ (1+x, y, z) H...Cl 2.95 Å C-H...Cl 166°; C(2)-
H(2a)...Cl(1)’ (x, 1+y, z) H...Cl 2.86 Å C-H...Cl 159°; C(3)-H(3a)...Cl(1)’ (1+x, 1+y, z) H...Cl 2.89 Å
C-H...Cl 146°; C(3)-H(3b)...F(1)’ (1-x, 0.5+y, 1-z) H...F 2.38 Å C-H...F 165°.
8

ACCEPTED MANUSCRIPT
Figure S2. Structure of compound 7 according to X-ray diffraction data. Thermal ellipsoids
are shown at the 50 % probability level.
The five-memebered ring of the compound 7 (Fig. S2) adopts an envelope conformation.
The C8 atom deviates from the mean plane of remaining atoms of the ring by -0.53 Å. The
cyclohexane ring adopts a chair conformation (the puckering parameters are: S= 1.11, Θ = 0.2°, Ψ =
9.0°). Deviations of the C1 and C4 atoms from the mean plane of remaining atoms of the ring are -
0.65 Å and 0.62 Å, respectively. Joined cycles are turned in such way that mean planes through-
passing the O1, C1, O2, C7 atoms and the C2, C3, C5, C6 atoms are almost orthogonal (the angle
between two planes is 85°).
In the crystal phase the intermolecular hydrogen bond C8-H8b…F2’ (1-x, 1-y, -z) H…F 2.53
Å C-H…F 131° is observed.
9

ACCEPTED MANUSCRIPT
Figure S3. Structure of compound 9 according to X-ray diffraction data. Thermal ellipsoids
are shown at the 50 % probability level.
The saturated ring of the compound 9 (Fig. S3) adopts a chair conformation (the puckering
parameters are S= 1.13, Θ = 2.8°, Ψ = 19.4°). Deviations of the C1 and C4 atoms from the mean
plane of the remaining atoms of the cycle are -0.71 Å and 0.59 Å, respectively. The five-membered
heterocycle is joined with cyclohexane ring and is turned almost orthogonal to the plane of the C2,
C3, C5, and C6 atoms (the angle between planes is 90°).
In the crystal phase the molecules 9 form the infinite chains (Fig. S4) along the [0 0 1]
crystallographic direction due to intermolecular hydrogen bonds N(1)-H…O(1)’ (1-x, 1-y, -z) H…O
2.09 Å N-H…O 170° and N(2)-H…O(2)’ (1-x, 1-y, 1-z) H…O 2.17 Å N-H…O 163°.
Figure S4. The infinite chains of the molecules 9 along the [0 0 1] crystallographic direction.
The formation of the hydrogen bonds causes the elongation of the C7-O1 (1.228(2) Å) and
C8-O2 (1.225(2) Å) bonds as compare with mean value 1.210 Å.
Figure S5. Structure of compound 16 according to X-ray diffraction data. Thermal ellipsoids are
shown at the 50 % probability level.
10