Synthesis and structure-activity relationship of nitrile-based cruzain inhibitors incorporating a trifluoroethylamine-based P2 amide replacement

Article abstract of DOI:10.1016/j.bmc.2019.115083

The structure-activity relationship for nitrile-based cruzain inhibitors incorporating a P2 amide replacement based on trifluoroethylamine was explored by deconstruction of a published series of inhibitors. It was demonstrated that the P3 biphenyl substituent present in the published inhibitor structures could be truncated to phenyl with only a small loss of affinity. The effects of inverting the configuration of the P2 amide replacement and linking a benzyl substituent at P1 were observed to be strongly nonadditive. We show that plotting affinity against molecular size provides a means to visualize both the molecular size efficiency of structural transformations and the nonadditivity in the structure-activity relationship. We also show how the relationship between affinity and lipophilicity, measured by high-performance liquid chromatography with an immobilized artificial membrane stationary phase, may be used to normalize affinity with respect to lipophilicity.

Full text of DOI:10.1016/j.bmc.2019.115083

Journal Pre-proofs
Synthesis and structure-activity relationship of nitrile-based cruzain inhibitors
incorporating a trifluoroethylamine-based P2 amide replacement
Juliana C. Gomes, Lorenzo Cianni, Jean Ribeiro, Fernanda dos Reis Rocho,
Samelyn da Costa Martins Silva, Pedro Henrique Jatai Batista, Carolina Borsoi
Moraes, Caio Haddad Franco, Lucio H.G. Freitas-Junior, Peter W. Kenny,
Andrei Leitão, Antonio C.B. Burtoloso, Daniela de Vita, Carlos A. Montanari
PII:
S0968-0896(19)31004-1
https://doi.org/10.1016/j.bmc.2019.115083
BMC 115083
DOI:
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
23 June 2019
22 August 2019
26 August 2019
Please cite this article as: Gomes, J.C., Cianni, L., Ribeiro, J., dos Reis Rocho, F., da Costa Martins Silva, S., Batista,
P.H.J., Moraes, C.B., Franco, C.H., Freitas-Junior, L.H.G., Kenny, P.W., Leitão, A., Burtoloso, A.C.B., de Vita,
D., Montanari, C.A., Synthesis and structure-activity relationship of nitrile-based cruzain inhibitors incorporating
a trifluoroethylamine-based P2 amide replacement, Bioorganic & Medicinal Chemistry (2019), doi: https://doi.org/
10.1016/j.bmc.2019.115083
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will
undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing
this version to give early visibility of the article. 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.
© 2019 Published by Elsevier Ltd.

Synthesis and structure-activity relationship of nitrile-based cruzain inhibitors
incorporating a trifluoroethylamine-based P2 amide replacement
Juliana C. Gomesa, Lorenzo Ciannib, Jean Ribeirob, Fernanda dos Reis Rochob, Samelyn da
Costa Martins Silvab, Pedro Henrique Jatai Batistab, Carolina Borsoi Moraesc, Caio Haddad
Francoc, Lucio H. G. Freitas-Juniorc, Peter W. Kennyb, Andrei Leitãob, Antonio C. B.
Burtolosoa, Daniela de Vita*b, Carlos A. Montanari*b
a Instituto de Química de São Carlos, Universidade de São Paulo, São Carlos, São Paulo,
Brazil
b Grupo de Química Medicinal do IQSC/USP, Instituto de Química de São Carlos,
Universidade de São Paulo, São Carlos, São Paulo, Brazil
c Departamento de Microbiologia, Instituto de Ciências Biomédicas, Universidade de São
Paulo, São Paulo, Brazil
Contact details: Carlos.Montanari@usp.br, Department of Chemistry and Molecular Physics
Institute of Chemistry of São Carlos University of São Paulo, Av. Trabalhador Sancarlense,
400, 13566-590 - São Carlos / SP
Abstract
The structure-activity relationship for nitrile-based cruzain inhibitors incorporating a P2 amide
replacement based on trifluoroethylamine was explored by deconstruction of a published series
of inhibitors. It was demonstrated that the P3 biphenyl substituent present in the published
inhibitor structures could be truncated to phenyl with only a small loss of affinity. The effects
of inverting the configuration of the P2 amide replacement and linking a benzyl substituent at
P1 were observed to be strongly non-additive. We show that plotting affinity against molecular

size provides a means to visualize both the molecular size efficiency of structural
transformations and the non-additivity in the structure-activity relationship. We also show how
the relationship between affinity and lipophilicity, measured by high-performance liquid
chromatography with an immobilized artificial membrane stationary phase, may be used to
normalize affinity with respect to lipophilicity.
Keywords: covalent inhibitor, cruzain, cysteine protease, group efficiency, lipophilic
efficiency, non-additivity, structure-activity relationship
1. Introduction
Chagas disease [1-3], also known as American trypanosomiasis, is caused by T. cruzi parasite
infection and is a significant public health problem both in Latin America and internationally
[4]. Cruzipain, the major T. cruzi cysteine protease, is expressed in all stages of the life-cycle
of the parasite and is considered to be an attractive target for therapeutic intervention in the
treatment of Chagas disease [5-7]. Enzyme inhibition studies typically make use of
recombinant cruzain (Cz) which consists of the catalytic domain of cruzipain. Cruzain
inhibitors have shown antichagasic activity in murine disease models [7, 8].
A commonly employed tactic in design of cysteine protease inhibitors is to incorporate an
electrophilic entity, commonly referred to as a “warhead”, in the inhibitor structure [9, 10].
This enables a covalent bond to be formed between an electrophilic atom of the inhibitor and
the thiol of the catalytic cysteine. Michael acceptors such as the vinylsulfone K777 [7] are
typically irreversible inhibitors of cysteine proteases. A nitrile group in an inhibitor structure
can form a covalent thioimidate adduct with the catalytic cysteine thiol and addition is typically
reversible [11, 12]. Nitrile-based cruzain inhibitors have been reported [13-15] and in vivo
activity in a murine disease model has been published [8] for inhibitors based on this warhead.
Anti-trypanosomal activity has also been observed for nitrile-based cysteine protease inhibitors
2

in cell-based assays at concentrations below their cruzain Ki values, suggesting that inhibition
of cysteine proteases other than cruzipain may be responsible for these effects [16].
A published [13] series of nitrile-based cruzain inhibitors that are structurally related to
the cathepsin K inhibitor odanacatib [17] provided the starting point for this study. The
compounds in this series typically feature a substituted biphenyl group at P3. The principal
objective of the study was to establish a structure-activity relationship (SAR) for cruzain
inhibition by deconstruction of the published inhibitor structures. Secondary objectives were
to explore the utility of chromatographically measured logKw for normalization of cruzain
inhibition with respect to lipophilicity and to assess the anti-trypanosomal activity of the
compounds against an intracellular form of the of T. cruzi Y strain.
3

2. Results
2.1 Synthetic chemistry
Chart 1. Compounds described in the present study
Structures of the compounds synthesized for elucidation of the SAR are shown in Chart 1. The
syntheses of the intermediate amino acids, to be used in the peptide coupling, were carried out
as shown in Scheme 1. Methyl esters (1-4) were produced from the commercial D- or L-amino
acids, using thionyl chloride in dry methanol, following the procedure previously described
[16]. After 18 h, the thin layer chromatography (TLC) indicated the total consumption of the
starting material in all the cases. The following step consisted of preparation of the 2,2,2-
4

trifluorophenyl amino acid intermediates (5-12), as previously described with slight
modification [18].
Scheme 1. Reagents and conditions: (a) SOCl2, CH3OH, reflux, 1 hr  r.t., overnight; (b)
ArC(=O)CF3, K2CO3, CH3OH, 50 °C, 18 hrs; (c) 1M Zn(BH4)2 in DME, CH3CN/CH3OH (5:1),
-40/-45 °C, 3 hrs; (d) NaBH4, THF, r.t., 6 hrs.
Briefly, the amino esters (1-4) were reacted with the appropriate 2,2,2-
trifluoroacetophenone, in the presence of potassium carbonate and methanol at 50 °C for 18 h,
to afford the respective imines. The presence of water, resulting from imine formation, in the
basic reaction medium at elevated temperature favors the hydrolysis of the methyl ester group,
providing the potassium salts of the 2,2,2-trifluorophenyl imines. Due to their instability, all
imines were used in the next step (for the diastereoselective reduction) without purification. As
described by Hughes et al. [18], the reduction with NaBH4 afforded, for the L-amino acid
derivatives, the R,S diastereoisomers (anti isomers), reflecting the fact that the re face is
sterically more hindered than the si face for the more stable conformation of these salts. This
is due to the presence of the R1 group of the L-amino acid moiety (phenyl, aryl, isopropyl),
allowing the attack of the hydride ion to be, preferably, on the si side of the imine [18]. On the
contrary, to obtain the S,S diastereoisomers (syn isomers) the imines were reduced with
Zn(BH4)2. In that case, the formation of the syn isomer as a major diastereoisomer can be
5

explained by complexation of Zn2+ with the nitrogen and oxygen atoms at the 1,3-positions of
the imine, respectively, making the si face sterically more hindered. When the starting material
is the D-amino acid, the reduction of the corresponding imine gives the S,R diastereoisomer
(anti isomer) with NaBH4 and the R,R diastereoisomer (syn isomer) with Zn(BH4)2.
In general, we observed that NaBH4 reduction of the imine produced the desired amines
with a diastereomeric excess over 95%, while when using Zn(BH4)2 the diastereomeric excess
was lower than 80%. The diastereomeric excesses in favor of the syn or anti isomer were
measured by HPLC of the final compounds (13-29). If necessary, final products were purified
by HPLC, equipped with a chiral column, to afford the desired product with d.e. > 99% and
e.e. > 99%. In order to confirm the absence of epimerization in the chiral center of the amino
acids during the synthetic steps, phenylalanine was employed as a model in this study.
Therefore, racemic phenylalanine was used as a starting material, and the final compounds,
after reduction (with NaBH4 or Zn(BH4)2) and amine coupling, were analyzed by HPLC. The
chromatograms of the racemic product were compared with those of 16a and 16b that came
from the enantiopure phenylalanine (see SI), confirming the integrity of the stereogenic center
with an enantiomeric excess > 99%.
After imine reduction the amino acids 5-12 were coupled with the appropriate amine, using
HATU as the coupling agent, to afford 13, 14, 16-22, 24-27, 29 in good yield (Scheme 2). All
compounds were fully characterized (characterization data for 16a and 19a are reported
elsewhere [16]). The synthesis of 2‐amino‐3‐phenylpropanenitrile (37), employed in the
coupling to provide compounds 21-27, 32, was carried out from L- or D-phenylalanine by two
methods (see SI). Both methods gave the Boc-derivatives with a high enantiomeric excess (>
99% e.e.), as resulted from the HPLC analysis (see SI). Then, Boc removal was performed
under a mild condition, using HCOOH.
6

Scheme 2. Reagents and conditions: (a) HATU, DIPEA, H2NXCN , r.t. 22hrs
Finally, compounds 15a and 23b were obtained by a Suzuki cross-coupling reaction of
the aryl bromide intermediates (Scheme 3). According to the procedure described by Chen et
al. [19], the bromides reacted in the presence of sodium carbonate, with the appropriate boronic
acid pinacol ester, using PdCl2dppf as catalyst, in DMF, at 80ºC for 3hrs (Scheme 3).
Scheme 3. Reagents and conditions: (a) ArB(OH)2 pinacol ester, PdCl2dppf, DMF, 80°C, 3
hrs, argon atmosphere.
Compounds 30-32 were prepared following the general procedure showed in Scheme
4. The Boc-L-amino acid was coupled with the selected amine by HATU, followed by Boc
removal in a mild condition (HCOOH). Next, the free amino group was coupled with benzoic
acid using HATU as the coupling agent. This procedure gave the final compound with an e.e.
> 99 %. All compounds were fully characterized (characterization data for 31 is reported
elsewhere [14]).
7

Scheme 4. Reagents and conditions: (a) H2NXCN, HATU, DIPEA, DMF, 16 hrs, r.t., argon
atm.; (b) HCOOH, 16 hrs, r.t.; (c) R2CO2H acid, HATU, DIPEA, DMF, 16 hrs, r.t., argon atm.
2. 2 Enzyme inhibition and lipophilicity
Measured pKi values for inhibition of cruzain and cathepsin L and lipophilicity measurements
performed by HPLC using an immobilized artificial membrane column are given in Table 1 for
the compounds in Chart 1. The most potent compound (22b; pKi = 9.2) synthesized during the
course of the current study is the des-methylsulfonyl analog of a cruzain inhibitor for which a
pIC50 value of 9.7 has been reported [13].
8

Table 1. Enzyme inhibition and lipophilicity
a,b
a,d
f
g
Compound Cz pKi
SE Cz pKi CatL pKi
SE CatL pKi logKw
SE logKw
0.02
a,c
a,e
6.7
7.4
7.2
8.1
7.1
5.6
6.0
5.7
5.5
5.1
5.4
< 5
6.4
8.8
6.6
9.2
8.8
< 5
5.4
5.6
< 5
6.4
6.9
< 5
7.3
6.5
7.3
0.01
0.01
1.1
13a
13b
14a
14b
15a
16a
16b
17b
18a
19a
19b
20a
21a
21b
22a
22b
23b
24a
25a
26b
27a
27b
28b
29a
30
0.01
0.01
0.02
0.07
0.01
0.04
0.05
0.05
0.05
6.2
6.1
6.0
0.03
0.03
0.03
2.8
2.6
2.8
1.4
1
1.7
1.7
1.5
1.4
0.01
0.01
0.02
0.02
0.004
0.02
0.02
0.03
0.02
6.8
6.6
0.05
0.02
6
0.03
0.02
0.03
0.02
0.01
0.02
0.01
0.01
6.6
8.3
2.2
2.3
3.1
3.2
0.02
0.02
0.2
5.8
7.6
0.03
0.01
0.2
0.05
0.03
5.8
6.2
< 5
6.2
5.9
5.9
7.7
7.4
7.2
0.07
0.05
2.7
2.6
0.01
0.02
0.02
0.09
0.01
0.04
0.03
0.02
0.03
0.04
2.7
2
0.02
0.01
0.01
0.02
0.02
0.76
0.01
31
32
a pKi = log10(Ki/M)
b Cruzain pKi
c Standard error in cruzain pKi
d Cathepsin L pKi
e Standard error in cathepsin L pKi
f logKw determined by HPLC using IAM stationary phase
g Standard error in logKw
2.3 Anti-trypanosomal assay results
Only two of the compounds listed in Table 1 exhibited concentration-responsive activity
against intracellular T. cruzi amastigotes infecting human U2OS cells [20-22]. EC50 and CC50
values of 11 ( 1) M and 33 ( 5) M respectively were determined for 20a. Compound 29a
was less potent (EC50 /M = 26  0.3) and CC50 could not be determined for this compound.
9

Benznidazole presented potency and efficacy values corresponding to data previously reported
for Y strain [20].
Figure 1. Concentration-response curves for compounds and reference compound
benznidazole after 96 h of exposure to U2OS cells infected with Trypanosoma cruzi Y strain.
Data points are means and error bars represent standard deviations of two replicates. Black dots
and curves represent data of compound activity whereas red dots and curves indicate the cell
ratio parameter.
3. Discussion
3.1 Structure-activity relationships
The principal objective of this study is to deconstruct published trifluoroethylamine-based
cruzain inhibitors to structures of lower molecular complexity [23] in order to establish a
structure-activity relationship (SAR). In medicinal chemistry, SARs are typically specified in
terms of the effects of structural transformations (e.g., substitute chlorine for hydrogen) on
potency or affinity. In this study, we use [X  Y] to specify structural transformations where
the labels X and Y can be defined flexibly by the user to refer to structure number, substructure
(e.g. P3-phenyl), functional group (e.g. aldehyde), or configuration as required [10]. The
affinity difference, pKi, corresponding to the [X  Y] transformation can be written as:
[
]
[ ] [ ]
∆푝퐾 푋 → 푌 = ∆푝퐾 푌 − ∆푝퐾 푋
푖
(1)
푖
푖
10

The basis of the lead-likeness [24] concept is that optimization of leads typically results in
increased lipophilicity and molecular size and these parameters may be regarded as the primary
physicochemical risk factors in drug design [25]. The structural modifications for which the
largest increases in affinity are accompanied by the smallest increases in risk can be regarded
being the most efficient in the context of hit or lead optimization [25]. This is the basis of group
efficiency (GE) [26] and a highly efficient structural transformation can also be seen as a type
of activity cliff [27, 28]. Although GE [26] is frequently presented [29] as a substituent
property, it is actually structural transformations with which values of the GE metric should be
associated [25]. One limitation of GE as a tool for SAR analysis is that it is not defined for
structural transformations such as inversion of configuration, aza-substitution and amide
reversal for which there is no net change in the number of non-hydrogen atoms. The effects of
inverting chiral centers are of considerable interest [30, 31] since the effects on logP are
typically small (zero in the case of compound with a single chiral center in its molecular
structure). The substructural transformations that form the basis for the analyses in this study
are given in Table 2.
Table 2. Key for structural transformations depicted in Figures 2 and 3
Representation
in Figure 2
Representation
in Figure 3
a
Label
Transformation
NnH
A
B
C
D
5
6
0
3
[P1-cyclopropyl  P1-benzyl(S)]
[P3-phenyl  P3-biphenyl]
[P3-S  P3-R]
N/A b
N/A b
[P2-Leu  P2-Phe]
a Change in number of non-hydrogen atoms
b Not applicable
The published [13] cruzain inhibitors were deconstructed to the structural prototype 13b
by truncation of P3 biphenyl substituent to phenyl and by replacement of the benzyl substituted
P1 linker with cyclopropane. Compound 13a, in which the P3 chiral center of 13b is S rather
than R, and the dipeptide nitrile 30 were also used as structural references. The effects of
11

synthetic elaboration are summarized by plotting pKi versus the change (NnH) in the number
of non-hydrogen atoms relative to the reference structure (Figure 2). Plotting against NnH
rather than NnH allows the dipeptide nitrile SAR to be visualized in the same frame of reference
as the other inhibitors. The gradient of the line linking two points quantifies the sensitivity of
affinity to an increase in molecular size [25]. For a structural transformation associated with a
net change in molecular size, the extent to which the different lines representing the structural
transformation are parallel gives a visual indication of the degree to which SAR is additive.
Figure 2. Sensitivity of cruzain pKi to structural modifications A, B and D (see Table 2 for
key)
The [phenyl  biphenyl] transformation at P3 (solid black line) leads to increases in pKi
for trifluoroethylamines regardless of the configuration of the P3 substituent. In contrast, the
[cyclopropane  benzylmethylene] transformation at P1 (dashed blue line) leads to an increase
in pKi when the P3 configuration is S but a decrease in pKi where this is R. The [cyclopropane
12

 benzylmethylene] transformation at P1 (dashed blue line) has no effect the affinity of the
dipeptide nitrile 30. Exchange of the P2 leucine for phenylalanine (solid red line) leads to
decreases in affinity ranging from 0.7 to 1.4 for the three reference compounds.
Substitution of the P3 substituents with a halogen typically leads to small decreases in
affinity and the pKi values for 15a, 17b and 18a are all within 0.3 log units of the respective
values for the compounds from which they were derived. Aza-substitution of the P3-biphenyl
is well tolerated and 23b is only 0.4 log units less potent than 22b. Replacement of the P2
phenylalanine by O-methyltyrosine for 16a and 16b resulted in decreases in pKi of 0.5 and 0.6
respectively. In contrast, the [26b  27b] transformation leads to an increase in pKi of 0.8 log
units which illustrates how the P1-substituent can influence the P2-subsituent SAR. Compound
25a was observed to be at least 0.4 log units more potent than 27a and this preference for the
S configuration at P1 is consistent with what has been reported [13] for analogues substituted
with 2-fluoro-4-cyanobenzyl at P1.
Additivity [32-35] is an essential concept in interpretation of SAR and additivity is
assumed when performing Free-Wilson analysis [36]. For the example of two hydrogen atoms
on a scaffold being substituted with groups X and Y, the requirement for additivity can be
written as:
[
]
[
]
[
]
(2)
∆푝퐾 퐻퐻 → 푋푌 = ∆푝퐾 퐻퐻 → 푋퐻 + ∆푝퐾 퐻퐻 → 퐻푌
푖
푖
푖
Although additivity of SAR is usually discussed in terms of adding substituents, the concept
can be applied to any structural transformation (e.g., inversion of configuration) and the labels
0 or 1 can be used more generally to indicate whether or not a structural transformation has
been applied. Additivity can also be quantified for physicochemical properties, such as
solubility, permeability, partition coefficient and reactivity, provided that differences in values
of the relevant measurable quantities are physically meaningful. The degree of additivity can
13

be quantified for a pair of structural transformations by  which is defined for a biochemical,
biophysical or physicochemical property, X, by:
[
]
[
]
[
]
(3)
훿[00 → 11] = ∆푋 00 → 11 − (∆푋 00 → 10 + ∆푋 00 → 01 )
The effects of applying two structural transformations can be classified as additive ( = 0),
subadditive ( < 0) or superadditive ( > 0). In analysis of non-additivity, structural
transformations are typically defined so that their application leads to increased molecular
complexity and inversion of configuration would typically be regarded as having no effect on
molecular complexity. Subadditive SAR should generally be anticipated for structural changes
that lead to an increase in molecular complexity [23], especially if there is a high degree of
constraint in the system [25]. Factors such as conformational rigidity, covalent protein-ligand
contact and multiple hydrogen bonds between protein and ligand would all be expected to
increase the degree of constraint. Non-additive SAR is particularly valuable for testing and
validation of models for binding affinity of ligands [37].
The cruzain inhibition SAR associated with structural transformations A, B and C (defined
in Table 2) is illustrated in Figure 3. Opposite faces of the cube shown in Figure 3 represent
the application of the two structural transformations to starting points that are connected by the
third structural transformation. The difference between pKi values associated with parallel
sides of each cube face quantifies the extent to which the associated SAR is non-additive while
the dependence of the pKi values on the order in which structural transformations are applied
indicates whether the SAR is subadditive or superadditive. This can be illustrated by the effects
of applying structural transforms A and C to 13a (see front face of the cube in Figure 3). The
difference ( = 1.7) between the of pKi value resulting from simultaneous application of both
structural transformations and the sum of pKi values corresponding to the individual structural
transformations shows that the associated SAR is strongly non-additive. In each case, the pKi
14

value for a structural transformation is greater when it is the second to be applied than if it is
applied first which indicates that the SAR is superadditive for this pair of structural
transformations.
Figure 3. Non-additivity in SAR for structural transformations A, B and C (see Table 2 for
key; Ph = phenyl; Ph2 = 4-biphenyl)
The non-additivity in the SAR depicted in Figure 3 is summarized in Table 3. One
interpretation of the superadditive SAR observed for the A and C transformations is that having
the P3 substituent with the S configuration enables the P1 benzyl substituent to interact more
effectively with the protein than would be the case for the R configuration. Nevertheless,
caution is needed when interpreting non-additivity in terms of molecular interactions because
the phenomenon is inherently non-local in nature.
Table 3. Non-additivity in SAR for structural transformations A, B and C (see Table 2 for key)
15

Initial
structure
13a
13b
13a
Transformations a
pKi[00  10] b pKi[00  01] b pKi[00  11] b
 b
A
A
A
A
B
B
B
B
C
C
C
C
+0.5
0.3
+1.4
0.3
0.6
+0.5
+0.2
0.1
+1.8
+2.1
+2.0
+1.4
+2.8
0.3
0.3
+1.7
+1.7
0
+0.7
+0.7
+0.9
+0.9
+2.4
14a
13a
21a
+0.2
a Defined in Table 2
b Defined in equation (3)
3.2 Cruzain versus cathepsin L selectivity
Compound 22b shows significant selectivity for cruzain over cathepsin L and the pKi values
for inhibition of these cysteine proteases differ by 3.4 log units. This is consistent with what
has been reported for structural analogs of this compound [13]. The corresponding pKi
differences (cruzain  cathepsin L) for compounds 14b (2.0) and 21b (0.5) show that both the
P3 biphenyl and P1 benzyl influence selectivity. The P3 configuration also influences
selectivity as exemplified by the pKi differences (cruzain  cathepsin L) for 14a (1.0) and 21a
(0.2). Aza-substitution of the P3 biphenyl of 22b leads to reduced selectivity as shown by
the pKi values of 8.8 (cruzain) and 7.6 (cathepsin L) measured for 23b.
3.3 Normalization of affinity with respect to measured lipophilicity
Affinity can be normalized with respect to lipophilicity by subtracting logD [38] or logP [39]
from pKi and the resulting quantity is often referred to as ligand lipophilicity efficiency (LLE),
lipophilic ligand efficiency (LLE) or lipophilic efficiency (LipE) [29]. It has been suggested
[25, 40, 41] that it would be more appropriate to use the actual trend in the data for
normalization and this can be done by first establishing a linear free energy relationship (LFER)
[42] between pKi and the relevant measure of lipophilicity. The slope of the LFER quantifies
the sensitivity of affinity to an increase in lipophilicity and, in general, varies with both target
and the scaffold on which the ligands are based [25]. The residuals (equation 4) from fitting
16

the LFER quantify the extent to which individual pKi values beat (or are beaten by) the trend
in the data [25, 40, 41]. This approach to normalization of affinity is analogous to that
introduced in 1984 by Andrews et al [43].
[
]
[
]
(4)
푝퐾 푟푒푠푑 = 푝퐾 푒푥푝푡 − 푝퐾 [푝푟푒푑]
푖
푖
푖
Scott and Waring have emphasized the benefits of using measured, rather than calculated,
lipophilicity for normalization of affinity [44]. In our study, lipophilicity was measured using
high performance liquid chromatography (HPLC) with an immobilized artificial membrane
(IAM) column. This assay is typically used for permeability assessment and lipophilicity
measured in this manner may also be described as phosphoplipophilicity. A plot of pKi versus
logKw is shown in Figure 4A and the correlation between these quantities is weak (R2 = 0.27;
Table 4). Provided that it is not merely due to a narrow range in the data, a weak correlation
between affinity and lipophilicity is actually desirable in optimization projects because it points
to specific SAR rather than affinity being driven solely by lipophilicity [25]. The plot of
pKi[resd] versus logKw (Figure 4B) shows the binding of 21b to be more efficient with respect
to lipophilicity than that of the more potent 22b. An advantage of presenting data in this manner
is that the extent to which the affinity of a compound beats (or is beaten by) the trend in the
data can be easily perceived.
17

A
B
Figure 4. Normalization of affinity with respect to lipophilicity
Table 4. Relationship between pKi and logKw
(
)
푝퐾 = 퐴0 + 퐴1 × log 퐾푤
푖
b
c
N a
18
A0
SE A0
0.82
A1
SE A1
0.36
RMSE d
1.0
R2 e
4.6
0.88
0.27
a Number of observations
b Standard error in A0
c Standard error in A1
d Root mean square error
e Coefficient of determination
18

Small differences in logKw were observed for diastereomers related by inversion of the
configuration of the P3 substituent and these appear to be influenced by the nature of the P1
substituent (Table 5). Variation in logKw values corresponding to a specific structural
transformation can be regarded as non-additivity in the structure-property relationship.
Table 5. Differences in logKw for diastereomers
Transformation
[14a14b]
[16a16b]
[19a19b]
[21a21b]
[22a22b]
P1-group
logKw
0.2
0.4
0.1
+0.1
Cyclopropane
Cyclopropane
Cyclopropane
benzyl
benzyl
+0.1
3.4 Anti-trypanosomal activity
Meaningful SAR cannot be derived for the cell-based assay results because it was only possible
to measure EC50 values for 20a (11 M) and 29a (26 M). In general, failure of enzyme
inhibitory activity to translate to activity in a cell-based assay should be anticipated, especially
for an intracellular parasite, and may reflect poor permeability or an enzyme inhibition model
that is not relevant in the cellular context [16]. An observation that compounds are more potent
in a cell-based assay than in the corresponding enzyme inhibition assay is more significant
(provided that it is not the result of non-specific cytotoxicity) and may indicate that the effects
on the cells are due to engagement of targets other than the cellular equivalent of the enzyme
used in the inhibition assay [16]. Although neither of the two compounds for which it was
possible to measure EC50 shows significant cruzain inhibition (pKi < 5), their potencies in the
cell-based assay are too low to allow any inference to be drawn as to whether or not their effects
in cells are due to inhibition of enzymes other than cruzipain.
4. Conclusion
19

We have deconstructed a published series of potent cruzain inhibitors and show that plotting
affinity against molecular size allows features of the structure-activity relationship to be easily
visualized. Examination of the structure-activity relationship reveals that the P3 biphenyl
substituent can be truncated to phenyl with the loss of only 0.4 log units of affinity. In
combination, the effects of inverting the configuration of the P3 substituent and substituting at
P1 with benzyl were observed to be strongly non-additive. We also show how lipophilicity
measured by HPLC with an immobilized artificial membrane can be used to normalize affinity.
5. Experimental Section
5.1 Synthetic chemistry
5.1.1 General materials and methods
Commercially available reagents and solvents were used as received, without further
purification, unless otherwise stated. All solvents were dried and distilled before use by
standard procedures. The reaction progress was monitored by thin-layer chromatography on
silica gel (aluminum foils) and spotted under UV light (254 nm), followed by staining with
ethanolic 5% phosphomolybdic acid solution or with aqueous KMnO4. Purification by column
chromatography was carried out on silica gel (Merck 60, particle size 0.040 -0.063 mm).
Melting points were determined on a Kofler apparatus. Infrared spectra were obtained using
FTIR at 4.0 cm-1 resolution and are reported in wavenumbers. The analytical HPLC system
consisted of a Shimadzu LC (Kyoto, Japan) equipped with an LC-20AT pump, an LC-20AD
pump, a SIL-20A HT autosampler, a DGU-20A5 degasser, a CBM-20A, SPDM20A DAD
detector and an FRC-10A fraction collector. In the established HPLC protocol, chiral analysis
was carried out at 32 °C (column oven) where not otherwise specified. The most common
mobile phase composition was acetonitrile-water (50:50) (v/v). Volumes of 10 μL (analytical)
and 1000 μL (semipreparative) were injected. Quantification was carried out at 200–800 nm
20

and the chromatographic run time varied according to the sample. Data acquisition was
performed using an LC solution software version 1.26 SP5. The LC system was coupled to an
AmaZon SL ion trap mass spectrometer (Bruker Daltonics, Bremen, Germany) equipped with
electrospray ionization (ESI) interface. Data acquisition was performed with Bruker Daltonics
1
13
19
Data Analysis software (version 4.2.383.1). H-NMR, C-NMR and F NMR spectra were
recorded at 400 or 500, 100 and 376 MHz respectively in CDCl3 or DMSO-d6 at 25°C, in a
Varian Mercury 400 or Bruker Avance DRX 500. Chemical shifts (δ) were reported in ppm
and the coupling constants (J) in Hertz (Hz). 1H-NMR spectra were referenced to DMSO (δ =
2.54 ppm) or CHCl3 (δ = 7.26 ppm) as internal standard, and 13C-NMR were referenced to the
central signal of the DMSO-d6 multiplet (δ = 40.45 ppm) or CDCl3 triplet (δ = 77.00 ppm).
Signal multiplicity was assigned as singlet (s), doublet (d), double doublet (dd), triplet (t),
double triplet (dt), quartet (q), quintuplet (qt), multiplet (m) and broad (br). The high-resolution
mass spectra (HRMS) were recorded using a Q-TOF Micromass equipment (Waters, UK)
employing ESI-MS techniques.
General Procedure for the synthesis of trifluoramino acids 9-12
To a solution of aminoester hydrochloride 1-4 (2.21 mmol) in dry methanol (5 mL), under an
argon atmosphere, potassium carbonate (4.42 mmol) and the appropriate trifluoroacetophenone
(2.43 mmol) were added. The mixture was stirred at 50ºC for 18 h and then, the reaction was
cooled to r.t., filtered and the solvent was removed under vacuum. The collected oil was used
in the next step without further purification.
Reduction with Zn(BH4)2: To a solution of imine/potassium carboxylate (2.21 mmol) in CH3CN
(15 mL) and MeOH (1.5 mL), a freshly prepared solution of Zn(BH4)2 in DME (1 mmol/mL,
1.5 mL) was added at -40/-45 °C. After 3 hrs at -40/-45 °C, the reaction mixture was allowed
to warm until 0°C and 1N HCl was added until pH 2-3. The aqueous mixture was extracted
21

with EtOAc (3x 25mL); the combined organic phases were dried on Na2SO4, evaporated under
vacuum and purified by flash column chromatography (AcOEt: n-Hexane 30:70 v/v) to give
the acid in a good yield (50-70%).
Reduction with NaBH4: The imine/potassium carboxylate (2.21 mmol) was suspended in dry
THF (25 mL) and MeOH (1.5 mL), stirred at 0ºC and then, NaBH4 (12.15 mmol) was added
in only batch. After stirring for 6 hrs, at room temperature, the system was cooled to 0°C and
quenched with 2 M HCl solution until stopping the gas release (final pH 2-3). Next, the organic
solvent was evaporated under vacuum and the crude extracted with ethyl acetate (3 x 10 mL).
The combined organic layers were then dried over Na2SO4, filtered, and evaporated in rotary
evaporator. Purification by flash column chromatography (AcOEt: n-Hexane 30:70 v/v)
afforded the trifluoroamino acid (56-71% yield).
General Procedure for the synthesis of dipeptidylnitriles 13, 14, 16-22, 24-27, 29
The appropriate acid (0.82 mmol), HATU (1.22 mmol, 465 mg) and amine as hydrochloride
salt or a free base (0.90 mmol) were sequentially introduced into a 25 mL round bottomed
flask, provided with a magnetic stirrer and under argon atmosphere. Next, dry DMF (13 mL)
and N,N-diisopropylethylamine (2.05 mmol.) were added into the flask. The reaction was
stirred at room temperature overnight. Ethyl acetate (100 mL) was added to the reaction
mixture and the organic layer was washed with saturated aqueous NaHCO3 (3x), NH4Cl (3x)
and NaCl solution (3x). The organic layer was then dried over Na2SO4, filtered, and evaporated
in rotary evaporator. Purification by flash column chromatography (AcOEt: n-hexane) afforded
the dipeptidyl nitriles. When necessary, compounds were purified by HPLC equipped with a
chiral column.
22

5.1.1
(S)-N-(1-cyanocyclopropyl)-4-methyl-2-(((R)-2,2,2-trifluoro-1-phenyl
ethyl)amino)pentanamide (13a)
Silica/ AcOEt: n-hexane 30:70 v/v (71% yield). Colorless oil. [α]D24= - 108.74º (MeOH, c =
1.03 x 10-3). FT-IR (KBr, cm-1): 3319, 2959, 2253, 1686, 1525, 1331, 1263, 1173, 1107, 887,
704. 1H NMR (400 MHz, CDCl3): δ 7.66 (s, 1H), 7.45-7.39 (m, 3H), 7.33-7.31 (m, 2H), 3.80
(q, J = 7.37 Hz, 1H), 3.07 (dd, J = 3.64 and 9.77 Hz, 1H), 1.99 (s, 1H), 1.64-1.42 (m, 5H),
1.41-1.35 (m, 1H), 1.24-1.19 (m, 2H), 0.85 (d, J = 6.28 Hz, 3H), 0.57 (d, J = 7.17 Hz, 3H). 13C
NMR (100 MHz, CDCl3): δ 174.8, 133.3, 129.6, 129.1, 128.0, 124.9 (q, J = 230 Hz), 119.8,
63.8 (q, J = 30 Hz) 59.0, 42.6, 24.6, 23.2, 23.1, 20.9, 20.1, 16.9, 16.4. HRMS (ESI): m/z calcd
for C18H23F3N3O [M+H]+: 354.17877; found: 354.17926. HPLC: Lux® 5 µm Cellulose-2, 250
x 4.6 mm, Acetonitrile/Water: 50:50, 0.5mL min-1, tR: 13.888 min, 206nm. CHIRALPAK®
IC-5 (5μM), 250 x 4.6 mm, Acetonitrile/Water: 50:50, 0.5mL min-1, tR: 20.850 min, 206nm.
After HPLC purification d.e. = 96.0 %.
5.1.2
(S)-N-(1-cyanocyclopropyl)-4-methyl-2-(((S)-2,2,2-trifluoro-1-phenyl
ethyl)amino)pentanamide (13b)
Silica/ AcOEt: n-hexane 30:70 v/v (71% yield). White solid. M.p. 144-146 ºC. [α]D24 = + 8.00º
(MeOH, c = 1.02 x 10-3). FT-IR (KBr, cm-1): 3319, 2959, 2253, 1686, 1525, 1331, 1263, 1173,
1107, 887, 704. 1H NMR (400 MHz, CDCl3): δ 7.42-7.37 (m, 3H), 7.34-7.32 (m, 2H), 7.15 (s,
1H), 4.03 (q, J = 7.37 Hz, 1H), 3.35 (dd, J = 4.75 and 8.95 Hz, 1H), 2.12 (s, 1H), 1.76 (ddq, J
= 6.53, 8.74 and 13.00 Hz, 1H), 1.60 (ddd, J = 4.78, 8.81 and 13.70 Hz, 1H), 1.49-1.36 (m,
3H), 1.00-0.96 (m, 1H), 0.95 (d, J = 2.50 Hz, 3H), 0.94 (d, J = 2.56 Hz, 3H), 0.83-0.77 (m,
1H). 13C NMR (100 MHz, CDCl3): δ 174.8, 134.1, 129.5, 129,0, 125.3 (q, J = 220 Hz), 119.6,
64.2, 63.8 (q, J = 28.7 Hz), 42.6, 24.8, 23.1, 21.7, 19.9, 16.8, 16.2. HRMS (ESI): m/z calcd for
23

C18H23F3N3O [M+H]+: 354.17877; found: 354.17926. HPLC: Lux® 5 µm Cellulose-2, 250 x
4.6 mm, Acetonitrile/Water: 50:50, 0.5 mL min-1, tR: 12.912 min, 206nm. After HPLC
purification d.e. > 99%.
5.1.3
(S)-2-(((R)-1-([1,1'-biphenyl]-4-yl)-2,2,2-trifluoroethyl)amino)-N-(1-cyano-
cyclopropyl) -4-methylpentanamide (14a)
Silica/ AcOEt: n-hexane 30:70 v/v (78% yield). White solid, Mp 149-151oC. [α]D24= -126º
(MeOH, c = 1.21x 10-3). FT-IR (KBr, ʋmax): 3323, 2958, 2245, 1683, 1519, 1267, 1172, 1103,
1
894, 765, 696 cm-1. H NMR (400 MHz, CDCl3): δ = 7.63-7.66 (m, 3H), 7.59-7.61 (m, 2H),
7.45-7.49 (m, 2H), 7.37-7.42 (m, 3H), 3.88 (m, 1H), 3.12 (dd, J = 2.9 e 9.6 Hz, 1H), 1.50-1.64
(m, 4H), 1.40-1.44 (m,1H), 1.21-1.32 (m, 3H), 0.88 (d, J = 5.9 Hz, 3H), 0.61 (d, J = 6.0 Hz,
13
3H) ppm. C NMR (100 MHz, CDCl3): δ = 174.9, 142.6, 140.0, 132.1, 128.9, 128.5, 127.8,
127.7, 127.1, 125.0 (q, J = 220 Hz), 119.8, 63.5 (q, J = 29.3 Hz), 59.0, 42.7, 24.7, 23.2, 21.0,
20.1, 16.9, 16.4 ppm. HRMS (ESI): m/z calcd for C24H27F3N3O [M + H]+: 430.21007; found:
430.21014. HPLC: Lux® 5 µm Cellulose-2, 250 x 4.6 mm, Acetonitrile/Water: 55:45, 0.5 mL
min-1, tR: 17.631 min, 206nm. After HPLC purification d.e. > 99%.
5.1.4
(S)-2-(((S)-1-([1,1'-biphenyl]-4-yl)-2,2,2-trifluoroethyl)amino)-N-(1-cyano-
cyclopropyl)-4-methylpentanamide (14b)
24
Silica/ AcOEt: n-hexane 10:90 to 30:70 v/v (78% yield). White solid. M.p. 200- 204o C. [α]D
= +38º (MeOH, c = 1.21x 10-3). FT-IR (KBr, ʋmax): 3323, 2958, 2245, 1683, 1519, 1267, 1172,
1103, 894, 765, 696 cm-1. 1H NMR (400 MHz, CDCl3): δ = 7.58-7.64 (m, 4H), 7.37-7.48 (m,
24

5H), 7.20 (s, 1H), 4.11 (q, J = 7.3 Hz, 1H), 3.40 (dd, J = 4.8, 8.8 Hz, 1H), 1.75-1.85 (m, 1H),
1.60-1.66 (m, 1H), 1.39-1.53 (m, 3H), 0.99-1.04 (m, 1H), 0.98 (d, J = 3.2 Hz, 3H), 0.97 (d, J
= 3.3 Hz, 3H), 0.84-0.92 (m, 2H) ppm. 13C NMR (100 MHz, CDCl3): δ = 174.8, 142.5, 140.0,
132.9, 128.9, 128.7, 127.8, 127.7, 127.1, 123.9, 119.6, 63.5 (q, J = 28.9 Hz), 60.0, 42.6, 24.8,
23.0, 21.8, 19.9, 16.8, 16.3 ppm. 19F NMR (376 MHz): δ -73.65 (d, J = 7.1 Hz). HRMS (ESI):
m/z calcd for C24H27F3N3O [M + H]+: 430.21007; found: 430.21014. HPLC: Lux® 5 µm
Cellulose-2, 250 x 4.6 mm, Acetonitrile/Water: 55:45, 0.5 mL min-1, rt: 16.281 min, 206nm;
Chiralpak® IC-5 (5 µm), 250 x 4.6 mm, Acetonitrile/Water: 50:50, rt: 20.850 min, 206nm.
After HPLC purification d.e. > 99%.
5.1.6
S)-N-(1-cyanocyclopropyl)-3-phenyl-2-(((S)-2,2,2-trifluoro-1-phenyl-
ethyl)amino)propanamide (16b)
Silica/ AcOEt: n-hexane 30:70 v/v (92% yield). White solid, M.p. 141-143oC. [α]D24 = +45 º
(MeOH, c = 1.07x 10-3). FT-IR (KBr, ʋmax): 3329, 3246, 3032, 2237, 1649, 1539, 1499, 1456,
1
1263, 1171, 1126, 850, 742, 567 cm-1. H NMR (400 MHz, CDCl3): δ = 7.19-7.43 (m, 10H),
3.99 (q, J = 7.19Hz, 1H), 3.43 (m, 1H), 3.11 (dd, J= 5.43 and 13.89Hz, 1H), 3.06 (dd, J= 6.83
e 13.89Hz, 1H), 2.03 (s, 1H), 1.42-1.51 (m, 2H), 1.00-1.04 (m, 1H), 0.86-0.92 (m, 1H) ppm.
13C NMR (100 MHz, CDCl3): δ = 173.4135.8, 133.1, 129.6, 129.3, 129.1, 129.0, 128.6, 127.5,
124.9 (q, J = 220 Hz), 119.5, 63.5 (q, J = 30 Hz), 60.8, 37.9, 19.9, 16.7, 16.4 ppm. HRMS
(ESI): m/z calcd for C21H21F3N3O [M + H]+: 388.16312; found: 388.16379. HPLC: Lux® 5
µm Cellulose-2, 250 x 4.6 mm, Acetonitrile/Water: 50:50, 0.5mL min-1,rt: 17.500 min, 206nm.
After HPLC purification d.e. > 99%.
25

5.1.7 (S)-2-(((R)-1-(4-bromophenyl)-2,2,2-trifluoroethyl)amino)-N-(1-cyanocyclo propyl)-3-
phenylpropanamide (17a)
Silica/ AcOEt: n-hexane 30:70 v/v (78% yield). White solid. [α]D24= -51.11º (MeOH, c = 1.35
x 10-3). FT-IR (KBr, ʋmax): 3339, 3028, 2243, 1686, 1491, 1263, 1170, 1124, 1010, 814, 727,
702 cm-1.1H NMR (400 MHz, CDCl3): δ = 7.62 (s, 1H), 7.26-7.36 (m, 5H), 6.97-7.03 (m, 2H),
6.64 (d, J = 8.4 Hz, 2H), 3.67-3.73 (m, 1H), 3.15-3.19 (m, 2H), 2.62 (dd, J = 10.9, 14.6 Hz,
1H), 2.07 (d, J = 9.61 Hz, 1H), 1.55-1.65 (m, 2H), 1.15-1.30 (m, 2H) ppm. 13C NMR (100
MHz, CDCl3): δ = 173.4, 135.8, 132.0, 131.2, 129.3, 129.2, 129.0, 127.5, 125.6, 123.5, 119.6,
63.1 (q, J = 29.4 Hz), 61.6, 39.3, 20.1, 16.9, 16.5 ppm. 19F NMR (376 MHz): δ -74.45 (d, J =
7.5 Hz). HRMS (ESI): m/z calcd for C21H19BrF3N3O [M + H]+: 466.07364; found: 466.07471.
HPLC: Lux® 5 µm Cellulose-2, 250 x 4.6 mm, Acetonitrile/Water 60:40, rt: 14.337 min,
206nm. Chiralpak® IC-5 (5 µm), 250 x 4.6 mm, Acetonitrile/Water: 45:55, rt: 30.751 min,
206nm. After HPLC purification d.e. > 99%.
5.1.8 (S)-2-(((R)-1-(4-bromophenyl)-2,2,2-trifluoroethyl)amino)-N-(1-cyanocyclo propyl)-3-
phenylpropanamide (17b)
o
Silica/ AcOEt: n-hexane 30:70 v/v (78% yield). White solid. M.p. 130-133 C. [α]D24= +77º
(MeOH, c = 1.21 x 10-3). FT-IR (KBr, ʋmax): 3339, 3028, 2243, 1686, 1491, 1263, 1170, 1124,
1010, 814, 727, 702 cm-1.1H NMR (400 MHz, CDCl3): δ = 7.50-7.54 (m, 2H), 7.29-7.38 (m,
3H), 7.12-7.19 (m, 5H), 3.97 (m, 1H), 3.39 (q, J = 6.24Hz, 1H), 3.08 (dd, J = 3.6, 1.9 Hz, 1H),
13
3.05 (dd, J = 4.6, 11.8 Hz, 1H), 2.19 (s, 1H), 1.45-1.52 (m, 2H), 0.92-1.06 (m, 2H) ppm. C
NMR (100 MHz, CDCl3): δ = 173.2, 135.7, 132.2, 132.0, 130.2, 129.2, 129.1, 127.5, 124.6 (q,
J = 230 Hz), 123.5, 119.4, 62.82 (q, J= 29.0 Hz), 60.7, 38.0, 20.0, 16.7, 16.5 ppm. HRMS
(ESI): m/z calcd for C21H19BrF3N3O [M + H]+: 466.07364; found: 466.07471. HPLC: Lux® 5
26

µm Cellulose-2, 250 x 4.6 mm, Acetonitrile/Water: 60:40, rt: 12.501 min, 206nm. After HPLC
purification d.e. > 99%.
5.1.9 (S)-2-(((R)-1-(3-chlorophenyl)-2,2,2-trifluoroethyl)amino)-N-(1-cyanocyclo propyl)-3-
phenylpropanamide (18a)
Silica/ AcOEt: n-hexane 30:70 v/v (73% yield). Colorless oil. [α]D24= +77.00º (MeOH, c = 1.13
x 10-3). FT-IR (KBr, ʋmax): 3336, 3057, 3030, 2250, 1683, 1531, 1475, 1354, 1138, 1101, 897,
1
790, 761, 700, 514 cm-1. H NMR (400 MHz, CDCl3): δ = 7.66 (s, 1H), 7.25-7.31 (m, 4H),
7.14 (t, J = 7.9 Hz, 1H), 7.00 (m, 2H), 6.78 (s, 1H), 6.70 (d, J= 7.7 Hz, 1H), 3.72 (s, 1H), 3.20
(m, 2H), 2.63 (dd, J= 9.9, 13.5 Hz, 1H), 2.08 (s, 1H), 1.55-1.65 (m, 2H), 1.17-1.32 (m, 2H)
ppm. 13C NMR (100 MHz, CDCl3): δ = 173.4, 135.5, 134.8, 134.2, 130.0, 129.6, 129.1, 128.8,
127.8, 127.7, 126.2, 125.9, 123.1, 119.6, 63.6, 63.3, 63.0, 62.7, 61.7, 39.2, 20.1, 16.9, 16.4
ppm. HRMS (ESI): m/z calcd for C21H20F3N3ClO [M + H]+: 422.12415; found: 422.12399.
HPLC: Lux® 5 µm Cellulose-2, 250 x 4.6 mm, Acetonitrile/Water 50:50, 0.5 mL min-1, rt:
25.316 min, 206nm. After HPLC purification d.e. > 99%.
5.1.12
(S)-N-(1-cyanocyclopropyl)-3-phenyl-2-(((S)-2,2,2-trifluoro-1-(4-methoxy-
phenyl)ethyl)amino)propanamide (19b)
Silica/ AcOEt: n-hexane 30:70 v/v (90% yield). Colorless oil. [α]D24= +59.35 º (MeOH, c =
1.23x 10-3). FT-IR (KBr, ʋmax): 3273, 3038, 2237, 1672, 1610, 1533, 1512, 1265, 1145, 1039,
1
873, 794, 561 cm-1. H NMR (400 MHz, CDCl3): δ = 7.36-7.42 (m, 3H), 7.27-7.28 (m, 3H),
7.10-7.13 (m, 2H), 6.88-6.91 (m, 2H), 4.02 (q, J = 7.2 Hz, 1H), 3.04 (dd, J = 4.6, 12.4 Hz, 1H),
13
3.00 (dd, J = 4.1, 12.4 Hz, 1H), 2.19 (s, 1H), 1.40-1.50 (m, 2H) ppm. C NMR (100 MHz,
CDCl3): δ = 173.6, 158.9, 133.1, 130.3, 129.6, 129.0, 128.6, 127.5, 126.4, 123.6, 119.5, 114.4,
27

63.31 (q, J= 28.8 Hz), 55.3, 36.9, 19.9, 16.7, 16.4 ppm. HRMS (ESI): m/z calcd for
C22H23F3N3O2 [M + H]+: 418.17369; found: 418.17435. HPLC: Lux® 5 µm Cellulose-2, 250
x 4.6 mm, Acetonitrile/Water: 50:50, 0.5 mL min-1, rt: 18.946 min, 206nm. After HPLC
purification e.e. > 99%.
5.1.13
(S)-N-(1-cyanocyclopropyl)-3-(1H-indol-3-yl)-2-(((R)-2,2,2-trifluoro-1-
phenylethyl)amino)propanamide (20a)
Silica/ AcOEt: n-hexane 30:70 v/v (82% yield). Colourless oil. 1H NMR (400 MHz, CDCl3): δ
= 8.20 (s, 1H), 7.63 (s, 1H), 7.30-7.25 (m, 2H), 7,11 (m, 2H), 6,99-6,90 (m, 3H), 6,85 (d, J =
2.0 Hz, 1H), 6.68 (d, J = 7.2 Hz, 1H), 3.69 (q, J = 7.6 Hz, 1H), 3,30 (dd, J = 4.4, 9.2 Hz, 1H),
3,20 (dd, J = 4.4, 14.4 Hz, 1H), 2.87 (dd, J = 9.2, 14.4 Hz, 1H), 1.78 (br s, 1H), 1.50 (m, 2H);
1.21-1.06 (m, 2H partially overlapped with water signal). ppm. 13C NMR (100 MHz, CDCl3):
δ = 174.7, 136.5, 132.5, 129.2, 128.6, 127.9, 127.1, 127.1, 125.0 (q, J = 279.1 Hz), 123.5,
122.5, 119.9, 119.9, 118.4, 111.5, 109.7, 63.9 (J = 28.9 Hz), 61.0, 29.3, 17.0, 16.5 ppm. HRMS
(ESI): m/z calcd for C23H21F3N4O [M + H]+: 427.1746; found: 427.1741. HPLC: Lux® 5 µm
Cellulose-2, 250 x 4.6 mm, Acetonitrile/Water: 50:50, 0.5 mL min-1, rt: 18.570 min, 206nm.
5.1.14
(S)-N-((S)-1-cyano-2-phenylethyl)-4-methyl-2-(((R)-2,2,2-trifluoro-1-phenyl-
ethyl)amino)pentanamide (21a)
Silica/ AcOEt: n-hexane 30:70 v/v (76% yield). White solid. M.p. 152-154oC. [α]D24= -115 º
(MeOH, c = 1.21x 10-3). FT-IR (KBr, ʋmax): 3331, 3360, 2966, 2934, 2245, 1672, 1516, 1497,
1
1267, 1107, 887, 698 cm-1. H NMR (400 MHz, CDCl3): δ = 7.52 (d, J = 8.8 Hz, 1H), 7.26-
7.41 (m, 10H), 5.12 (dt, J = 7.0, 8.8 Hz, 1H), 3.74-3.77 (m, 1H), 3.15 (d, J = 7.0 Hz, 2H), 3.02
(dd, J = 3.9 e 9.8 Hz, 1H), 1.93 (d, J = 9.1 Hz, 1H), 1.32-1.44 (m, 2H), 1.19-1.25 (m, 1H), 0.80
13
(d, J = 6.4 Hz, 3H), 0.49 (d, J = 6.3 Hz, 3H) ppm. C NMR (100 MHz, CDCl3): δ = 173.4,
133.8, 133.3, 129.5, 129.3, 129.0, 129.0, 128.0, 127.9, 123.4, 117.9, 63.5 (q, J = 29.2 Hz), 58.4,
28

42.7, 41.1, 38.9, 24.6, 23.1, 20.8 ppm. HRMS (ESI): m/z calcd for C23H27F3N3O [M + H]+:
418.21007; found: 418.20877. HPLC: Lux® 5 µm Cellulose-2, 250 x 4.6 mm,
Acetonitrile/Water: 60:40, 0.5 mL min-1, rt: 14.022 min, 206nm. After HPLC purification d.e.
> 99%.
5.1.15
(S)-N-((S)-1-cyano-2-phenylethyl)-4-methyl-2-(((S)-2,2,2-trifluoro-1-phenyl
ethy)amino)pentanamide (21b)
Silica/ AcOEt: n-hexane 30:70 v/v (76% yield). White solid. M.p. 125-129oC. [α]D24 = -26 º
(MeOH, c = 1.26x 10-3). FT-IR (KBr, ʋmax): 3331, 3360, 2966, 2934, 2245, 1672, 1516, 1497,
1267, 1107, 887, 698 cm-1. 1H NMR (400 MHz, CDCl3): δ = 7.30-7.39 (m, 6H), 7.19-7.22 (m,
4H), 7.11 (d, J = 8.77 Hz, 1H), 4.90-5.00 (m, 1H), 4.00 (q, J= 7.20 Hz, 1H), 3,32 (dd, J = 4.73
and 8.87 Hz, 1H), 3.00 (dd, J = 4.37 and 11.50 Hz, 1H), 2.97 (dd, J = 4.28 and 11.50Hz, 1H),
1.99 (s, 1H), 1.70 (ddq, J = 6.54, 8.79 and 13.04 Hz, 1H), 1.51 (ddd, J = 4.76, 8.85 and 13.75
Hz, 1H), 1.34 (ddd, J = 5.58, 8.92 and 14.21 Hz, 1H), 0.91 (dd, J = 6.57 and 9.59 Hz, 6H) ppm.
13C NMR (100 MHz, CDCl3): δ = 173.4, 133.7, 133.5, 129.5, 129.3, 129.1, 129.0, 128.3, 128.0,
124.2, 117.7, 63.6 (d, J = 28.6 Hz), 59.2, 42.6, 40.8, 38.8, 24.8, 23.0, 21.8 ppm. HRMS (ESI):
m/z calcd for C23H27F3N3O [M + H]+: 418.21007; found: 418.20877. HPLC: Lux® 5 µm
Cellulose-2, 250 x 4.6 mm, Acetonitrile/Water: 60:40, 0.5mL min-1, rt: 13.679 min, 206nm.
After HPLC purification d.e. > 99%.
5.1.16 (S)-2-(((R)-1-([1,1'-biphenyl]-4-yl)-2,2,2-trifluoroethyl)amino)-N-((S)-1-cyano -2-
phenylethyl)-4-methylpentanamide (22a)
Silica/ AcOEt: n-hexane 30:70 v/v (81% yield). White solid. M.p. 173-176oC. [α]D24= -113º
(MeOH, c = 1.1x 10-3). FT-IR (KBr, ʋmax): 3313, 2954, 1672, 1531, 1494, 1332, 1265, 1174,
29