Exploring the role of 2-chloro-6-fluoro substitution in 2-alkylthio-6-benzyl-5-alkylpyrimidin-4(3 H)-ones: Effects in HIV-1-infected cells and in HIV-1 reverse transcriptase enzymes

Article abstract of DOI:10.1021/jm500284x

A comparison of the effects of the 6-(2-chloro-6-fluorobenzyl)-2- (alkylthio)pyrimidin-4(3H)-ones (2-Cl-6-F-S-DABOs) 7-12 and the related 6-(2,6-difluorobenzyl) counterparts 13-15 in HIV-1 infected cells and in the HIV-1 reverse transcriptase (RT) assays

Full text of DOI:10.1021/jm500284x

Subscriber access provided by Virgina Commonwealth University Libraries & VIVA (Virtual Library of Virginia)
Article
Exploring the Role of 2-Chloro-6-Fluoro Substitution in 2-
Alkylthio-6-benzyl-5-alkylpyrimidin-4(3H)-ones: Effects in HIV-1-
Infected Cells and in HIV-1 Reverse Transcriptase Enzymes
Dante Rotili, Domenico Tarantino, Maxim B Nawrozkij, Alexandre S. Babushkin, Giorgia
Botta, Biagina Marrocco, Roberto Cirilli, Sergio Menta, Roger Badia, Emmanuele
Crespan, Flavio Ballante, Rino Ragno, Jose Este, Giovanni Maga, and Antonello Mai
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm500284x • Publication Date (Web): 31 May 2014
Downloaded from http://pubs.acs.org on June 2, 2014
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Exploring the Role of 2-Chloro-6-Fluoro Substitution in 2-Alkylthio-6-
benzyl-5-alkylpyrimidin-4(3H)-ones: Effects in HIV-1-Infected Cells
and in HIV-1 Reverse Transcriptase Enzymes
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Dante Rotili,† Domenico Tarantino,† Maxim B. Nawrozkij,^ Alexandre S.
Babushkin,^ Giorgia Botta,† Biagina Marrocco,† Roberto Cirilli,§ Sergio Menta,†
Roger Badia,° Emmanuele Crespan,# Flavio Ballante,¥ Rino Ragno,¥ José A. Esté,*,°
Giovanni Maga,*,# and Antonello Mai*,†,ç
†Dipartimento di Chimica e Tecnologie del Farmaco, Università degli Studi di Roma “La
Sapienza”, P.le A. Moro 5, 00185 Roma, Italy
^Volgograd State Technical University, pr. Lenina, 28, 400131 Volgograd, Russia
§Istituto Superiore di Sanità, Dipartimento del Farmaco, Viale Regina Elena 299, 00161 Rome,
Italy
°IrsiCaixa, Hospital Universitari Germans Trias i Pujol, Universitat Autònoma de Barcelona,
08916 Badalona, Spain
#Istituto di Genetica Molecolare IGM-CNR, via Abbiategrasso 207, 27100 Pavia, Italy
¥Rome Center for Molecular Design, Dipartimento di Chimica e Tecnologie del Farmaco,
Università degli Studi di Roma “La Sapienza”, P.le A. Moro 5, 00185 Roma, Italy
çIstituto Pasteur – Fondazione Cenci Bolognetti, Università degli Studi di Roma “La Sapienza”,
P.le A. Moro 5, 00185 Roma, Italy
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 44
1
2
3
4
5
6
7
*To whom correspondence should be addressed: J.A.E., Phone: +34-934656374; Fax: +34-934653968; e-
mail: iaeste@irsicaixa.es. G.M., Phone: +39 0382 546354; Fax: +39 0382 422286; e-mail: maga@igm.cnr.it.
A.M., Phone: +39 06 49913392; Fax:+39 06 49693268; e-mail: antonello.mai@uniroma1.it.
8
9
Dedicated to Prof. Marino Artico in the occasion of his 80° birthday.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Abstract
A comparison of the effects of the 6-(2-chloro-6-fluorobenzyl)-2-(alkylthio)pyrimidin-4(3H)-ones
(2-Cl-6-F-S-DABOs) 7-12 and the related 6-(2,6-difluorobenzyl) counterparts 13-15 in HIV-1
infected cells and in the HIV-1 reverse transcriptase (RT) assays is here described. The new 2-Cl-6-
F-S-DABOs showed up to picomolar activity against wt HIV-1. Against clinically relevant HIV-1
mutants and in enzyme assays, the simultaneous C5(methyl)/C6(methyl/ethyl) substitution in the 2-
Cl-6-F- and 2,6-F2-benzyl series furnished compounds with the highest, wide-spectrum inhibitory
activity against HIV-1. Three representative 2-Cl-6-F-S-DABOs carrying two (9c, 10c) or one (10a)
stereogenic centres were resolved into their individual stereoisomers, and showed a significant
diastereo- and enantioselectivity in HIV-1 inhibition, the highest antiviral activity well correlating
with the R absolute configuration to the stereogenic centre of the C6-benzylic position in both
cellular and enzymatic tests. Application of previously reported COMBINEr protocol on 9c and 10c
confirmed the influence of the stereogenic centres on their binding modes in the HIV-1 RT.
2
ACS Paragon Plus Environment

Page 3 of 44
Journal of Medicinal Chemistry
1
2
3
Introduction
4
5
6
7
8
9
The dihydro-alkyloxy-benzyl-oxopyrimidines (DABOs)1-3 were disclosed as non-nucleoside reverse
transcriptase inhibitors (NNRTIs)4,5 by our group in 1992.6-8 Since then, many structural
modifications have been performed on the pyrimidine ring with the aim to obtain more potent and
selective compounds. These efforts led to the discovery by us and other groups of some series of
excellent DABO and DABO-related compounds such as S-DABOs,9-19 NH-DABOs,20 N,N-
DABOs,21-23 and DAPY-DABO hybrids24,25 that are endowed with inhibitory potencies in the low
nanomolar or subnanomolar range against wild-type and mutant strains without a significant
cytotoxicity at higher concentrations (up to >100 ꢀM) (Figure 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. SAR graphical summary for the various DABO derivatives.
In previous works we have reported that, in the series of alkyl-S-DABOs, the 2,6-
difluorobenzyl/1-(2,6-difluorophenyl)ethyl substitution at the C6 position (R1 = H/Me in 1, Figure
2) of the pyrimidin-4(3H)-one ring is better than the 2,6-dichloro counterpart in terms of HIV-1
inhibiting activity,14,15 independently on the substitution at C5 (R2 in 1, Figure 2) and C2 (S-R3 in 1,
Figure 2) positions.
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 44
1
2
3
4
5
6
7
8
9
In 2005, some properly substituted 6-(2,6-dichlorobenzyl)-2-(benzylthio)pyrimidin-4(3H)-
ones (Cl2-S-DABOs) 2 (Figure 2) were reported to be more potent than their 2,6-difluorobenzyl
analogues,17 but this trend was inverted when a methyl group was inserted at the benzylic position
of such derivatives (F2-S-DABOs 3, Figure 2).18
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In 2008, we reported the first series of DABO-related compounds bearing a 2-chloro-6-
fuorobenzyl substitution at C6, the oxophenethyl-S-DABOs 4 (Figure 2).19 Such compounds
showed a peculiar SAR profile, deeply different from the (cyclo)alkyl-S-DABOs, with the 2,6-
dichlorobenzyl/2-chloro-6-fluorobenzyl moiety at C6 and the ethyl/iso-propyl group at C5 being
crucial for the broadest spectrum anti-HIV-1 activity (see also Figure 1).19 Also in the structurally
related 2-((4-benzyl-6-oxo-1,6-dihydropyrimidin-2-yl)thio)-N-phenylacetamide 5 (Figure 2), the 2-
chloro-6-fluorobenzyl substitution afforded higher anti-HIV-1 activity than the 2,6-difluorobenzyl
one.26
In a comparative study between 2,6-difluoro- (F2-) and 2-chloro-6-fluoro- (2-Cl-6-F-) N,N-
DABO derivatives, we observed that the latter are endowed with an improved activity against wild-
type and mutated HIV-RT forms and display a peculiar mechanism of action especially toward
mutated (K103N and Y181I) RT enzymes.23 In contrast to F2-N,N-DABOs which behave as slow-
binding/high-affinity RT inhibitors,22 the 2-Cl-6-F-N,N-DABOs show highly improved association
rates toward wild-type/mutated HIV-1 RT enzymes coupled with an high-affinity binding to the
same forms and with a peculiar higher-affinity binding to mutated RT forms than to the wild-type
enzyme (6, Figure 2).23
Thus, in some contexts, in addition to the F2-, also the 2-Cl-6-F-benzyl substitution at the C6
position of the S-DABOs afforded potent and selective anti-HIV-1 agents. Nevertheless, a
systematic study on the effect of the 2-Cl-6-F substitution on the “classical” alkyl/cycloalkyl-S-
DABOs was still missing.
4
ACS Paragon Plus Environment

Page 5 of 44
Journal of Medicinal Chemistry
1
2
3
O
O
R2
4
5
6
7
Me
HN
HN
R3
R1
R1
S
N
F
S
N
F
Cl
Cl
8
R
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
: R1 = H
1
3: R1 = Me
R1, R2 = H, Me; R3 = (cyclo)alkyl
O
O
R
R2
R2
HN
HN
Me
Me
X
S
N
N
N
Cl
F
Me
Cl
F
6
4
5
: X = CO; R2 = Et, i-Pr
R2 = H, Me
: X = NHCO; R2 = Me, Et
Figure 2. Previously reported 2,6-dihalosubstituted DABO compounds (1-6).
Therefore, we prepared a new series of S-DABOs characterized by the best (cyclo)alkylthio
substituents at C2 (S-R3 in Figure 2), as resulted from previous SAR studies,1-3,14,15 by a 2-Cl-6-F-
benzyl group at C6 position, and by an hydrogen, methyl or ethyl substituent at C5 (R2 in Figure 2)
and/or at the C6-benzylic (R1 in Figure 2) moiety (7-12, Figure 2). For comparison purposes, we
also included in the study some previously reported15,27,28 and newly synthesized F2-benzyl
analogues characterized by the same substitutions at C2, and by a methyl or ethyl substituent at both
C5 and C6-benzylic group (13-15, Figure 2).
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 44
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. 2,6-Dihalosubstituted alkyl-S-DABO derivatives (7-15) object of the present study.
Since many derivatives contain one or two stereogenic centres and, therefore, exist as racemic
or stereoisomers’ mixtures, respectively, we decided to perform a systematic investigation on their
potential enantio-/diastereoselective anti-HIV-1 activity. Focusing our attention on three of the most
potent compounds (9c, 10a, 10c), we performed the enantio-/diastereoseparation, the chiroptical
characterization, the absolute configuration assignation (where it was possible), and the enzymatic
and cellular evaluation of the HIV-1 inhibitory activity of the single enantiomers/stereoisomers in
comparison with the corresponding racemic/stereoisomers’ mixtures.
Finally, a structure-based studies by the means of a previously reported COMBINEr protocol
was undertaken on 9c and 10c to gather the influence of the stereogenic centers on their binding
modes in the HIV-1 RT.
Chemistry
The required 2-(2,6-dihalophenyl)propionic/butanoic acids 16a-d were obtained by alkylation
of the corresponding commercially available (2,6-dihalo)phenylacetic acids with the proper alkyl
iodides in the presence of n-butyllithium in anhydrous THF first at -78 °C and then at room
6
ACS Paragon Plus Environment

Page 7 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
temperature.15,23,28 The imidazolides obtained from the acids 16a-d and from the commercial 2-
chloro-6-fluorophenylacetic acid by treatment with N,N-carbonyldiimidazole (CDI) in dry
acetonitrile at room temperature were then treated with potassium ethylmalonate, potassium ethyl-
2-methylmalonate or potassium ethyl-2-ethylmalonate in the presence of the magnesium
dichloride/triethylamine system to yield, after decarboxylation with 12% hydrochloric acid, the
required ethyl β-oxoesters 17a-i.14,15,18,19,21,23,27,28 Condensation of these β-ketoesters with thiourea
in the presence of sodium ethoxide in dry ethanol under reflux conditions gave the intermediates 6-
[(2,6-dihalophenyl)alkyl]-5-alkyl-2-thioxo-2,3-dihydropyrimidin-4(1H)ones 18a-i,15,18,19,28 which
were finally S-alkylated in dry DMF with the proper commercially available alkyl/cycloalkyl
halides in the presence of anhydrous potassium carbonate to yield the final compounds 7-15
(Scheme 1).15,28 Chemical and physical data for compounds 7-18 (only for the novel compounds)
are listed in Table S1 in Supporting Information.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Synthesis of Compounds 7-15.a
a(a) (1) n-BuLi, (Et)2NH, dry THF, -78 °C; (2) proper alkyl iodides, dry THF, -78 °C then room
temp; (b) (1) CDI, dry CH3CN, room temp; (2) KOOCCHR2COOEt, MgCl2, Et3N, dry CH3CN,
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 44
1
2
3
4
5
reflux; (3) 12% HCl, room temp; (c) NH2CSNH2, EtONa, reflux; (d) alkyl/cycloalkyl iodide or
bromide, anhydrous K2CO3, dry DMF, room temp.
6
7
8
9
Results and Discussion
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cytotoxicity and Wild Type HIV-1 Inhibiting Activity. The new alkyl-S-DABOs 7-15
were tested in MT-4 cells to evaluate their cytotoxicity and their capability to inhibit by 50% the
HIV-1-induced cytopathic effect in MT-4 cells (HIV-1 strain: NL4-3). Nevirapine (NVP) and
efavirenz (EFV) were also tested as reference drugs (Table 1).
Table 1. Cytotoxicity and Anti-HIV-1 Activity (wt HIV-1 Strain: NL4-3) of 7-15a
compound
R1
R2
R3
EC50,b µM
CC50,c µM
selectivity indexd
H
H
H
H
i-Pre
n-Pr
0.38
0.26
>80f
>80
>76
>76
>74
>76
>76
>73
>73
>71
>75
>75
>73
>73
>71
>210
>308
7a
7b
7c
7d
7e
8a
8b
8c
8d
8e
9a
9b
9c
9d
9e
H
H
s-Bu
0.16
>475
H
H
n-Bu
0.29
>262
H
H
cyclopentyl
i-Pr
0.15
>493
H
Me
Me
Me
Me
Me
H
0.12
>633
H
n-Pr
0.086
0.047
0.11
>884
H
s-Bu
>1,553
>664
H
n-Bu
H
cyclopentyl
i-Pr
0.18
>394
Me
Me
Me
Me
Me
0.002
0.003
0.001
0.0003
0.0003
>37,500
>25,000
>73,000
>243,333
>236,667
H
n-Pr
H
s-Bu
H
n-Bu
H
cyclopentyl
8
ACS Paragon Plus Environment

Page 9 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
Me
Me
Me
Me
Me
Et
i-Pr
n-Pr
0.00004
0.00005
0.0003
0.0003
0.00003
0.96
>15
>1.5
>1.5
>1.5
>15
>3
>375,000
>30,000
>5,000
>5,000
>500,000
>3
10a
10b
10c
10d
10e
11a
11b
11c
11d
11e
12a
12b
12c
12d
12e
13ag
13b
13cg
13dg
13eg
14a
14b
14c
14d
14e
15ah
15b
15ch
s-Bu
n-Bu
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
cyclopentyl
i-Pr
Et
n-Pr
0.028
0.27
>3
>107
Et
s-Bu
>3
>11
Et
n-Bu
0.064
>3
>3
>47
Et
cyclopentyl
i-Pr
>3
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
0.084
0.026
0.054
0.11
>3
>36
>115
>56
Et
n-Pr
>3
Et
s-Bu
>3
Et
n-Bu
>3
>27
Et
cyclopentyl
i-Pr
0.58
>3
>5
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
0.001
0.028
0.0009
0.050
0.001
0.089
0.24
>3
>3,000
>107
>3,333
>60
n-Pr
>3
s-Bu
>3
n-Bu
>3
cyclopentyl
i-Pr
>3
>3,000
>34
>3
Et
n-Pr
>3
>12
Et
s-Bu
0.14
>3
>21
Et
n-Bu
0.057
0.16
>3
>53
Et
cyclopentyl
i-Pr
>3
>19
Me
Me
Me
0.004
0.012
0.0008
>30
>3
>7,500
>250
>3,750
Et
n-Pr
Et
s-Bu
>3
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 44
1
2
3
4
5
6
7
8
9
Et
Et
Me
Me
n-Bu
0.023
0.011
0.12
>3
30
>7
>3
>130
2,727
>62
15d
15eh
NVP
EFV
cyclopentyl
0.007
>429
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
b
aValues are means determined from at least two experiments. Effective concentration 50, concentration needed to
inhibit by 50% the HIV-1-induced cytopatic effect, evaluated with the MTT method in MT-4 cells (HIV-1 strain: NL4-
c
3). Cytotoxic concentration 50, concentration needed to induce 50% death of non-infected cells, evaluated with the
MTT method in MT-4 cells. dSelectivity index (SI), CC50/EC50. eAbbreviations: i-Pr, iso-propyl; n-Pr, normal-propyl; s-
f
Bu, sec-butyl; n-Bu, normal-butyl. Higher concentrations could not be tested due to precipitation of compounds in the
culture medium. gRef. 15. hRef. 28.
All the tested compounds without solubility problems (7-9) were not cytotoxic up to 71 ꢀM.
Some compounds (10-15) showed >1.5 or >3.0 ꢀM values of CC50 (compound concentration toxic
for 50% of cells) because higher concentration could not be achieved for the precipitation of the
compounds in the culture medium.
From the analysis of inhibition data for wt HIV-1 in MT-4 cells (Table 1), it is clear that an
alkyl substituent at the benzylic portion of the molecule is crucial to increase the anti-HIV-1
potency of the new 2-Cl-6-F-S-DABOs (compare the EC50 values of the benzyl-unsubstituted
uracils and thymines 7 and 8 with the corresponding analogues 9 and 10 carrying a methyl group at
the benzylic position). In analogy to what observed in other DABO and DABO-related
series,15,16,18,21,24 also among the 2-Cl-6-F-S-DABOs 7-12 the double methyl substitution at both
C6(R1) and C5(R2) positions was the most productive in terms of wt HIV-1 inhibition, and led to a
set of derivatives (10a-e) active in the picomolar range (300 to 30 pM), from 23- to 230-fold more
potent than efavirenz, and in some cases (10a and 10e) with impressive selectivity indexes (SI,
CC50/EC50) > 375,000.
10
ACS Paragon Plus Environment

Page 11 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Despite the presence of the C6-benzylic methyl substituent, the replacement of the C5-methyl
with a C5-ethyl group (compounds 11a-e) gave up to 5-magnitude orders decrease of potency of the
derivatives (compare 10a with 11a). Keeping fixed the C5-methyl group, the change of the C6-
methyl with a C6-ethyl group (compounds 12a-e) again led to less potent compounds, also if in this
case the drop of potency seems to be less pronounced (3 or 4-magnitude orders).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As reported in previous S-DABO series,9,10,12-15 the alkyl chain inserted at C2 (branched, such
as iso-propyl and sec-butyl, linear, such as n-propyl and n-butyl, and cyclic, such as cyclopentyl)
has modulating effects on the anti-HIV-1 activity. In particular, in the C6-methyl uracil (9a-e) and
thymine (10a-e) derivatives the C2-alkyl chain gave up to 10-fold differences in EC50 values of the
compounds, the best substituent being the cyclopentyl (9e and 10e) and the iso-propyl ones (9a and
10a), and up to 100-fold differences in EC50 values in the series 11a-e [C5(Et)/C6(Me)] and 12a-e
[C5(Me)/C6(Et)] where, on the contrary, the cyclopentyl substituent was the worst in terms of
potency (11e and 12e, respectively), whereas the n-propyl was the best (11b and 12b, respectively).
The F2-S-DABOs 13-15 never reached the extremely high potency and selectivity of the best
2-chloro-6-fluoro compounds (10a-e), and showed a trend in the wt HIV-1 inhibiting activity
partially different from what observed in the 2-chloro-6-fluoro analogues (10-12). Indeed, in the F2-
S-DABO series there were no significant differences between thymine compounds substituted with
the C6-methyl (13a-e) or C6-ethyl (15a-e) group, all showing EC50 values in the low
nanomolar/picomolar range. Differently, the potency significantly decreased in case of C5-methyl
to C5-ethyl replacement and simultaneous C6-methyl substitution (14a-e, EC50s in the high
nanomolar range).
Inhibitory Activity Against Clinically Relevant HIV-1 Mutant Strains. The alkyl-S-
DABOs 7-15 were also tested against a panel of clinically relevant HIV-1 mutants (K103N, Y181C,
and Y188L) (Table 2). In analogy to what observed against wt HIV-1, among the 2-chloro-6-fluoro
alkyl-S-DABOs 7-12 the inhibitory activity decreased according to the order C6(R1)/C5(R2) Me/Me
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 44
1
2
3
4
5
6
7
8
9
> Me/H > H /Me > H/H ≅ Et/Me ≅ Me/Et. Also in this case, the double methyl substitution at both
C6(R1) and C5(R2) positions led to the most potent derivatives of the 2-Cl-6-F series (10a-e), that
were active in the submicromolar range against the mutants K103N (EC50 values ranging from
0.093 µM for 10e to 0.66 µM for 10d) and Y181C (EC50 values ranging from 0.23 µM for 10e to
0.62 µM for 10d) and in the single digit micromolar range against Y188L, resulting more potent
than nevirapine against all three mutants, and more (vs K103N), equal (vs Y188L), or less (vs
Y181C) potent than efavirenz depending on the HIV-1 mutant strain. The C6-methyl uracil
derivatives 9a-e were in general 1-magnitude order less potent than the corresponding thymines
10a-e, whereas the remaining 2-Cl-6-F-S-DABOs displayed a more severe drop of inhibiting
potency. Since a concentration higher than 3 µM could not be tested due to precipitation of
compounds in the culture medium, it was impossible (with the only exception of compound 12b) to
determine reliable EC50 values for the series 11a-e and 12a-e, and, for this reason, any comparison
with the series 7a-e and 8a-e was precluded. The F2-S-DABOs 13-15 substantially confirmed
against the mutants the same SAR observed against the wt HIV-1. The two most potent series (EC50
values in the submicromolar range) were 13a-e [C6(Me)/C5(Me)] and 15a-e [C6(Et)/C5(Me)], with
a slight predominance of the first (13a-e) against K103N and Y181C, and of the second (15a-e)
against Y188L. At the C2 position, a branched S-alkyl chain (iso-propyl, sec-butyl and cyclopentyl)
assured the highest potency. On the other hand, the C2 linear alkyl (n-propyl and n-butyl) -
substituted derivatives (13b,d and 15b,d) together with all the derivatives of the series 14a-e
[C6(Me)/C5(Et)] were markedly less potent against all of the three tested mutant strains (EC50 > 3
µM).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
12
ACS Paragon Plus Environment

Page 13 of 44
Journal of Medicinal Chemistry
1
2
3
Table 2. Anti-HIV-1 Activity of 7-15 against Clinically Relevant HIV-1 Mutant Strainsa
4
5
EC50,b µM (fold-resistance)c
6
compd
R1
R2
R3
7
8
9
K103N
39.2 (103)
18.8 (72)
25.2 (157)
29.2 (101)
42.1 (281)
10.6 (88)
9.2 (107)
5.4 (115)
20.0 (182)
>71
Y181C
17.8 (47)
8.2 (31)
8.3 (52)
16.7 (58)
18.3 (122)
8.4 (70)
2.3 (27)
5.2 (111)
2.3 (21)
>71
Y188L
>80e
H
H
H
H
i-Prd
n-Pr
7a
7b
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
>80
H
H
s-Bu
>76
7c
H
H
n-Bu
>76
7d
H
H
cyclopentyl
i-Pr
>74
7e
H
Me
Me
Me
Me
Me
H
>76
8a
H
n-Pr
>76
8b
H
s-Bu
>73
8c
H
n-Bu
>73
8d
H
cyclopentyl
i-Pr
>71
8e
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
1.1 (550)
0.89 (297)
1.0 (1,000)
1.1 (3,667)
0.9 (3,000)
0.14 (4,667)
0.35 (7,000)
0.32 (1,067)
0.66 (2,200)
0.093 (3,100)
>3
7.5 (3,750)
2.3 (767)
3.6 (3,600)
3.9 (13,000)
3.0 (10,000)
0.25 (8,333)
0.26 (5,200)
0.39 (1,300)
0.62 (2,067)
0.23 (7,667)
>3
30.7 (15,350)
7.9 (2,633)
7.5 (7,500)
5.6 (18,667)
>71
9a
H
n-Pr
9b
H
s-Bu
9c
H
n-Bu
9d
H
cyclopentyl
i-Pr
9e
Me
Me
Me
Me
Me
Et
3.4 (113,333)
>1.5
10a
10b
10c
10d
10e
11a
11b
11c
11d
11e
n-Pr
s-Bu
>1.5
n-Bu
>1.5
cyclopentyl
i-Pr
2.9 (96,667)
>3
Et
n-Pr
>3
>3
>3
Et
s-Bu
>3
>3
>3
Et
n-Bu
>3
>3
>3
Et
cyclopentyl
>3
>3
>3
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 44
1
2
3
4
5
6
7
8
9
Et
Et
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
i-Pr
n-Pr
>3
2.1 (81)
>3
>3
2.8 (108)
>3
>3
12a
12b
12c
>3
Et
s-Bu
>3
Et
n-Bu
>3
>3
>3
12d
12e
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Et
cyclopentyl
i-Pr
>3
>3
>3
13af
13b
13cf
13df
13ef
14a
14b
14c
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
0.27 (270)
>3
0.12 (120)
>3
>3
n-Pr
>3
s-Bu
0.11 (122)
>3
0.071 (79)
>3
>3
n-Bu
>3
cyclopentyl
i-Pr
0.23 (230)
>3
0.1 (100)
>3
>3
>3
Et
n-Pr
>3
>3
>3
Et
s-Bu
>3
>3
>3
Et
n-Bu
>3
>3
>3
14d
14e
Et
cyclopentyl
i-Pr
>3
>3
>3
15ag
15b
15cg
15d
15eg
NVP
EFV
Me
Me
Me
Me
Me
0.53 (132)
>3
0.75 (187)
>3
0.27 (67)
>3
Et
n-Pr
Et
s-Bu
0.2 (250)
>3
0.3 (375)
>3
0.9 (1,125)
>3
Et
n-Bu
Et
cyclopentyl
1.0 (91)
5.6 (47)
0.34 (49)
1.1 (100)
>7
1.6 (145)
>7
0.013 (2)
1.97 (281)
aValues are means determined from at least two experiments. Effective concentration 50, concentration needed to
b
c
inhibit by 50% the HIV-1-induced cytopatic effect, evaluated with the MTT method in MT-4 cells. Fold resistance:
ratio of EC50 value against drug-resistant strain and EC50 of the wt NL4-3 strain (Table 2). dAbbreviations: i-Pr, iso-
e
propyl; n-Pr, normal-propyl; s-Bu, sec-butyl; n-Bu, normal-butyl. Higher concentrations could not be tested due to
precipitation of compounds in the culture medium. fRef. 15. gRef. 28.
14
ACS Paragon Plus Environment

Page 15 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Enzyme Inhibiting Activity: Effect on HIV-1 RT Wild Type and NNRTI-resistant
Mutants. The most potent 2-Cl-6-F-S-DABOs in cell assays 9-12 as well as their 2,6-difluoro
counterparts 13-15 were tested against recombinant wt HIV-1 RT and against a panel of
recombinant RTs carrying known NNRTI-resistance mutations (K103N, Y181I, and L100I) (Table
3). Similarly to what observed in cellular assays, the double C6/C5-methyl-substituted analogs 10a-
e displayed the highest - from nanomolar (10a,d,e) to picomolar (10b and 10c) - inhibition of wt
HIV-1 RT. Unfortunately, they suffered from a great decrease of potency against the mutant RTs, in
particular against the Y181I mutant. The related compounds 9a-e sharing the sole C6-methyl
substitution showed in general less potency than 10a-e against wt RT, but higher inhibition towards
the Y181I mutant, with lower fold-resistance values respect to 10a-e. The introduction of a C5-ethyl
group in addition to the C6-methyl group (compounds 11a-e) further decreased the potency of
derivatives against wt RT, but maintained the same potency against the mutated RTs as the uracil
counterparts 9a-e (single-digit micromolar against K103N, micromolar against Y181I, and
submicromolar against L100I). All considering, in the enzyme assays the 2-Cl-6-F-S-DABOs
bearing an ethyl group at C6 and a methyl group at C5 (12a-e) showed the most valuable anti-RT
activity, because they displayed a slightly reduced potency against wt RT with respect to 10a-e (no
compounds with picomolar activity among 12a-e), but showed submicromolar inhibitory potencies
against the K103N and L100I mutants, and single-digit micromolar activity against Y181I, the less
sensitive RT mutant.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The tested F2-S-DABOs 13-15, apart their C6/C5 methyl or ethyl substitution, generally
exerted nanomolar inhibition of wt RT and submicromolar (K103N and L100I) or micromolar
(Y181I) potency against mutant RTs. Similarly as the 2-Cl-6-F-S-DABOs 12a-e, also with the 2,6-
difluoro compounds the C6-ethyl/C5-methyl substitution (compounds 15a-e) assured the highest
potency against the Y181I RT mutant, at single digit micromolar level. Globally, both the most
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 44
1
2
3
active 2-Cl-6-F- and F2-S-DABOs 12 and 15 were more potent than the reference drug efavirenz
4
5
against the wt RT and the K103N RT mutant, and less effective against the Y181I RT mutant.
6
7
8
9
Table 3. Inhibitory Activity of 9-15 against HIV-1 RT Wild Type and NNRTI-resistant Mutantsa
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ID50,b µM (fold-resistance)c
compd
R1
R2
R3
wt
K103N
1.3 (162)
1.3 (162)
3.0 (375)
2.2 (220)
1.8 (120)
1.4 (93)
Y181I
L100I
0.34 (42)
0.5 (62)
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
H
H
i-Prd
n-Pr
0.008
0.008
0.008
0.01
10.0 (1,250)
15.0 (1,875)
21.1 (2,638)
15.2 (1,520)
16.6 (1,107)
200.0 (13,333)
80.0 (88,889)
9a
9b
H
s-Bu
0.2 (25)
9c
H
n-Bu
0.6 (60)
9d
H
cyclopentyl
i-Pr
0.015
0.015
0.0009
0.0003
0.003
0.011
0.048
0.010
0.008
0.022
0.024
0.006
0.005
0.002
0.003
0.004
0.003
1.1 (73)
9e
Me
Me
Me
Me
Me
Et
0.29 (19)
0.09 (100)
0.075 (250)
0.156 (52)
0.12 (11)
1.1 (23)
10a
10b
10c
10d
10e
11a
11b
11c
11d
11e
12a
12b
12c
12d
12e
13ae
n-Pr
1.5 (1,667)
s-Bu
1.75 (5,833) 35.0 (116,667)
n-Bu
0.93 (310)
1.3 (118)
2.6 (54)
60.0 (20,000)
32.0 (2,909)
27.0 (562)
4.7 (470)
cyclopentyl
i-Pr
Et
n-Pr
2.6 (260)
2.4 (300)
2.3 (104)
1.8 (75)
0.23 (23)
0.4 (50)
Et
s-Bu
16.2 (2,025)
13.1 (595)
15.6 (650)
7.7 (1,283)
3.5 (700)
Et
n-Bu
0.33 (15)
0.53 (22)
0.12 (20)
0.24 (48)
0.12 (60)
0.26 (87)
0.068 (17)
0.28 (93)
Et
cyclopentyl
i-Pr
Me
Me
Me
Me
Me
Me
0.42 (70)
0.63 (126)
0.45 (225)
1.0 (333)
0.34 (85)
0.58 (193)
Et
n-Pr
Et
s-Bu
5.9 (2,950)
6.1 (2,033)
6.0 (1,500)
47.0 (15,667)
Et
n-Bu
Et
cyclopentyl
i-Pr
Me
16
ACS Paragon Plus Environment

Page 17 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
Me
Me
Me
Me
Et
n-Pr
s-Bu
0.001
0.031
0.003
0.005
0.012
0.003
0.011
0.003
0.002
0.004
0.006
0.006
0.003
0.005
0.4
0.38 (380)
1.7 (55)
5.4 (5,400)
96.0 (3,097)
4.6 (1,533)
0.9 (180)
0.23 (230)
0.67 (22)
0.23 (77)
0.3 (60)
13b
13ce
13de
13ee
14a
n-Bu
0.45 (150)
0.03 (6)
cyclopentyl
i-Pr
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.83 (69)
0.43 (143)
0.45 (41)
0.39 (130)
0.2 (100)
0.31 (78)
0.42 (70)
0.14 (23)
0.28 (93)
0.38 (76)
7.0 (17)
129.4 (10,783)
2.8 (933)
27.2 (2,267)
0.27 (90)
0.11 (10)
0.28 (93)
0.13 (65)
0.21 (53)
0.22 (37)
0.048 (8)
0.23 (77)
0.11 (22)
9.0 (22)
Et
n-Pr
14b
14c
Et
s-Bu
4.6 (418)
Et
n-Bu
3.0 (1,000)
1.8 (900)
14d
14e
Et
cyclopentyl
i-Pr
15af
15b
15cf
15d
15ef
NVP
EFV
Me
Me
Me
Me
Me
4.1 (1,025)
3.6 (600)
Et
n-Pr
Et
s-Bu
4.0 (667)
Et
n-Bu
4.6 (1,533)
4.9 (980)
Et
cyclopentyl
35.0 (87)
0.03
3.0 (100)
0.08 (3)
NDd
aValues are means determined from at least three experiments. Inhibitory dose 50, compound dose required to inhibit
the HIV-1 rRT activity by 50%. cFold resistance: ratio of ID50mut/ID50wt values. dAbbreviations: i-Pr, iso-propyl; n-Pr,
normal-propyl; s-Bu, sec-butyl; n-Bu, normal-butyl. eRef. 15. fRef. 28.
b
Separation and Chiroptical Characterization of 9c, 10a and 10c Single Stereoisomers
Three 2-Cl-6-F-S-DABOs, namely the 6-[1-(2-chloro-6-fluorophenyl)ethyl]-2-(sec-
butylsulfanyl)pyrimidin-4(3H)-one
(9c),
the
6-[1-(2-chloro-6-fluorophenyl)ethyl]-2-(iso-
propylsulfanyl)-5-methylpyrimidin-4(3H)-one (10a), and the 6-[1-(2-chloro-6-fluorophenyl)ethyl]-
2-(sec-butylsulfanyl)-5-methylpyrimidin-4(3H)-one (10c), were resolved by stereoselective HPLC
and the collected single stereoisomers screened in cellular and enzyme assays, to assess the
influence of chirality in their antiviral activity (see below). The direct chromatographic resolution
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 44
1
2
3
4
5
6
7
8
9
was achieved by using the immobilized-type Chiralpak IA chiral stationary phase (CSP) under
normal-phase and polar organic conditions.29 As an example, Figure 3 shows the simultaneous
enantio- and diastereoseparation of 9c and the analytical check of the enantiomeric (ee) and
diastereomeric (de) excess of the stereoisomers collected on a semipreparative scale. By analysis of
the sign of the chiroptical properties and relative peak areas (Figures 3 and 4) it was possible to
identify the first and fourth eluted stereoisomer of compound 9c, and the first and third of
compound 10c, as belonging to the same enantiomeric pairs (compounds 9c′ and 10c′,
respectively), and the second and third one of 9c and the second and fourth one of 10c to the
diastereomeric pairs of enantiomers (compounds 9c′′ and 10c′′, respectively). The absolute
configuration of the enantiomers of 10a and of the stereogenic centre of the C6-benzylic moiety of
9c and 10c was achieved by circular dichroism (CD) correlation method28,30 and is in full agreement
with docking studies (see below). The absolute configuration of the stereogenic centre for the C2-
sec-butylthio chain of 9c and 10c (R or S) was suggested by docking studies (see below).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
18
ACS Paragon Plus Environment

Page 19 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
20
15
10
5
(+)
(+)
(-)
(-)
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
UV at 260 nm
0
6
3
0
CD at 260 nm
-3
ee>99%
400
200
0
de>99%
300
ee>99%
de=99%
150
0
400
ee=99%
de>99%
200
0
ee>99%
de>99%
140
70
0
0
4
8
12
16
20
Elution time (min)
Figure 3. From top to bottom: typical chromatogram illustrating the resolution of 9c using
simultaneous UV (black) and CD (orange) detection; purity control of the single fractions collected
at semipreparative scale; ee, enantiomeric excess; de, diastereomeric excess. Column: Chiralpak IA
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 44
1
2
3
250 mm × 4.6 mm I.D.; detection: UV at 260 nm; eluent: n-hexane-ethanol 95:5 (v/v); flow-rate:
4
5
1.0 mL min−1; column temperature: 25 °C.
6
7
8
9
As previously demonstrated,28,30 the CD properties of 2-(sec-butylthio)-6-[1-(2,6-
dichlorophenyl)propyl]-5-methylpyrimidin-4(3H)-one (Figure 4), which is structurally similar to
the compounds studied in this work, depend only on the absolute configuration of the stereogenic
centre at C6 and they are not greatly influenced by the configuration at the alkyl group linked to the
sulfur atom at C2 position of the pyrimidine ring. As a result, the diastereomers with the same
stereochemistry at the C6 stereogenic centre adjacent to the 2,6-dihalophenyl moiety showed
similar CD profile and the same sign of the diagnostic CD band located at around 245 nm. On this
basis, the CD spectra of the single stereoisomers of 9c, 10a and 10c were recorded in the same
solvent (ethanol) and compared with those of the diastereomers (S,R)- and (R,R)-2-(sec-butylthio)-
6-[1-(2,6-dichlorophenyl)propyl]-5-methylpyrimidin-4(3H)-one.28,30
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As shown in Figure 4, the (-)-10a enantiomer and the couples of the diastereomers (-)-9c′/(-)-
9c′′ and (-)-10c′/(-)-10c′′ displayed bisigned CD curves between 300 and 210 nm with a positive
Cotton effect strikingly similar to those of the reference stereoisomers with R absolute configuration
at C6-benzyl methylene unit. Therefore, as a result of the CD study, the R configuration was
empirically assigned to the stereogenic centre of the benzyl fragment of the levorotatory
stereoisomers of all the three chiral compounds. Differently, the CD analysis not allowed us to
define the absolute configuration of the stereogenic centre of the alkylthio group at C2 pyrimidine
ring position of 9c and 10c, and the reported configuration (R or S) was suggested by molecular
modelling studies (see below).
20
ACS Paragon Plus Environment

Page 21 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
(+)-9c'
(-)-9c'
(+)-9c''
(-)-9c''
60
30
0
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
-30
-60
(-)-10c'
(+)-10c'
(-)-10c''
(+)-10c''
45
30
15
0
-15
-30
-45
(-)-10a
(+)-10a
30
15
0
-15
-30
-45
O
CH3
CH3
150
100
50
HN
(S.R)-(-)
(R,R)-(-)
∗
∗
S
N
CH3
Cl
CH3
Cl
0
-50
-100
210
240
270
300
330
Wavelength (nm)
Figure 4. Comparison between the CD spectra of the stereoisomers of 9c, 10a and 10c, and the S,R
and R,R diastereoisomers of 2-(sec-butylthio)-6-[1-(2,6-dichlorophenyl)propyl]-5-methylpyrimidin-
4(3H)-one recorded in ethanol.
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 44
1
2
3
4
5
6
7
8
Anti-HIV-1 Activity of 9c, 10a and 10c Individual Stereoisomers. After separation, the
four stereoisomers of 9c [(+)- and (-)-9c′, and (+)- and (-)-9c′′], the two enantiomers of 10a [(+)-
and (-)-10a], and the four stereoisomers of 10c [(+)- and (-)-10c′, and (+)- and (-)-10c′′] were
individually tested in a separate sequence of cellular and enzymatic experiments, in comparison
with the corresponding mixtures (hereafter simply referred to as 9c, 10a, and 10c). Nevirapine and
efavirenz were added and tested as reference drugs (Tables 4 and 5).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 4. Cytotoxicity and Anti-HIV-1 Activity against WT (NL4-3) and Clinically Relevant HIV-1
Mutant Strains of 9c, 10a and 10c and their Individual Stereoisomersa
EC50,c µM (fold-resistance)d
compdb
R1
R2
R3
CC50,e µM
SIf
NL4-3
0.004
K103N
1.3 (325)
1.22 (111)
7.64 (53)
0.2 (610)
5.56 (118)
Y181C
Y188L
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
H
H
s-Bug
s-Bu
s-Bu
s-Bu
s-Bu
i-Pr
2.3 (575) 19.2 (4800)
>73h
>39
>18250
>3545
9c
0.011
1.24 (113)
7.4 (673)
(R,S)-(+)-9c′
(S,S)-(+)-9c′′
(R,R)-(-)-9c′′
(S,R)-(-)-9c′
10a
H
0.144
>73.3
>73.3
>73
>507
H
0.0003
0.047
0.58 (1924) 0.86 (2859)
13.2 (281) >51.6
>31
>103333
>1106
H
>52
Me
Me
Me
Me
Me
Me
Me
Me
0.0001
0.00003
0.22
0.22 (2200) 0.99 (9900) 6.3(63000)
0.010 (333) 0.15 (5000) 0.35 (11667)
>73
>730000
>783333
>168.7
>78222
>1173333
>50285
>12.8
i-Pr
>23.5
>36.9
>70.4
>70.4
>70.4
>47
(R)-(-)-10a
(S)-(+)-10a
10c
i-Pr
>36.9
0.06 (67)
0.015 (14)
0.15 (135)
>47
>36.9
>36.9
s-Bu
0.0009
0.41 (456) 2.1 (2333)
0.087 (79) 0.39 (351)
0.89 (811) 3.3 (3006)
s-Bu 0.00006
(R,R)-(-)-10c′
(S,R)-(-)-10c′′ Me
s-Bu
s-Bu
s-Bu
0.0014
3.66
Me
Me
>47
64 (58407)
>7
>47
>70
>7
(S,S)-(+)-10c′
(R,S)-(+)-10c′′
NVP
43.9
>70
>70.4
>7
>1.6
0.12
5.6 (47)
>62
22
ACS Paragon Plus Environment

Page 23 of 44
Journal of Medicinal Chemistry
1
2
3
EFV
0.007
0.34 (49)
0.013 (2)
1.97 (281)
>3
>429
4
5
b
aValues are means determined from at least two experiments. Compounds were listed according to their elution order.
6
7
The underlined R or S indicates the absolute configuration of the sec-butyl chiral centre, as suggested by modelling
8
9
c
studies (see below). Effective concentration 50, concentration needed to inhibit by 50% the HIV-1-induced cytophatic
d
effect, evaluated with the MTT method in MT-4 cells. Fold resistance: ratio of EC50 value against drug-resistant strain
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
e
and EC50 of the wt NL4-3 strain. Cytotoxic concentration 50, concentration needed to induce 50% death of non-
f
g
infected cells, evaluated with the MTT method in MT-4 cells. Selectivity index, CC50/EC50. Abbreviations: i-Pr, iso-
h
propyl; s-Bu, sec-butyl. Higher concentrations could not be tested due to precipitation of compounds in the culture
medium.
In general, for all the three investigated compounds a marked diastereo-/enantioselective effect
in the anti-HIV-1 activity has been observed, with the results and the order of relative potency
against wt and mutant strains always consistent among the different stereoisomers at both cellular
(Table 4) and enzymatic levels (Table 5).
Among the four stereoisomers of 9c, the highest antiviral activity was displayed by the third
eluted stereoisomer (R,R)-(-)-9c′′ (the absolute configuration of the sec-butyl chiral centre is
underlined because it was suggested by modelling studies, see below): it was active at picomolar
level against wt HIV-1 (EC50 = 300 pM), and at submicromolar level against the clinically relevant
mutants. Globally, (R,R)-(-)-9c′′ was up to 13- and 480-fold more potent and selective than 9c
(mixture of four stereoisomers) and (S,S)-(+)-9c′′ (the less active stereoisomer, belonging to the
same enantiomeric pair), respectively (Table 4). In enzyme assays, (R,R)-(-)-9c′′ showed the
highest potency against wt RT as well as against the mutant RTs (Table 5).
The second best-scoring stereoisomer of the 9c mixture, in terms of potency and selectivity
both in cellular (wt and mutant HIV-1 strains) and enzyme (wt and mutant RTs) assays, was the
first eluted one (R,S)-(+)-9c′. This suggests that in addition to the C6 chiral centre, that with the R
configuration drives the highest antiviral potency (see (R,R)-(-)-9c′′ above), also the chiral centre
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 44
1
2
3
4
5
6
7
8
9
due to the C2 sec-butylthio chain plays a role for determining the anti-HIV-1 effects of the
derivatives. Indeed, if the C2 sec-butylthio substituent would display no stereoselective antiviral
activity, the second more potent stereoisomer should be expected to be the diastereoisomer (S,R)-
(-)-9c′, that instead is the third of the 9c series for antiviral potency and selectivity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The different anti-HIV-1 potency observed between the two enantiomers of compound 10a
further confirmed the crucial role of the R stereochemistry at the chiral centre at the C6-benzylic
position of DABOs. Indeed, the eutomer (R)-(-)-10a was active at picomolar level against wt HIV-1
and at submicromolar level against the three tested mutants (Table 4), and showed nanomolar and
(sub)micromolar inhibition against the wt RT and the mutant RTs (Table 5), with improved potency
respect to both the other enantiomer (S)-(+)-10a and the corresponding racemic mixture.
About the four stereoisomers of compound 10c, as expected the two dextrorotatory
stereoisomers (S,S)-(+)-10c′ and (R,S)-(+)-10c′′, with the S configuration at the C6-benzylic chiral
centre, were practically devoid of any significant anti-HIV-1 activity both in cell and in enzyme
assays (Tables 4 and 5). On the other hand, the two levorotatory isomers (R,R)-(-)-10c′ and (S,R)-
(-)-10c′′ were potent anti-HIV-1 agents in cells up to picomolar (wt) and nanomolar (mutant
strains) concentration (Table 4), and displayed nanomolar and (sub)micromolar inhibition against
RT and the mutant RTs, respectively, in enzyme assays (Table 5).
Table 5. Inhibitory Activity of Derivatives 9c, 10a and 10c and their Individual Stereoisomers
against HIV-1 RT Wild Type and NNRTI-Resistant Mutantsa
ID50,c µM (fold-resistance)d
compdb
R1
R2
R3
wt
K103N
10.2 (804)
6.157 (229)
>20
Y181I
>20f
>20
L100I
Me
Me
Me
H
H
H
s-Bue
s-Bu
s-Bu
0.01269
0.02689
0.5467
0.1294 (10)
0.6491 (24)
2.218 (4)
9c
(R,S)-(+)-9c′
(S,S)-(+)-9c′′
>20
24
ACS Paragon Plus Environment

Page 25 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
H
s-Bu
s-Bu
i-Pr
0.003759
0.04052
0.015
0.009
3.348
0.011
0.001
0.014
4.664
5.918
0.4
1.613 (488)
14.8 (365)
3.204 (214)
0.610 (67.7)
>40
14.11 (3753)
>20
0.06871 (18)
0.1481 (4)
0.498 (33)
0.177 (19.7)
1.486 (0.4)
0.377 (34)
0.041 (41)
0.524 (37)
0.913 (0.2)
0.596 (0.1)
9.0 (22)
(R,R)-(-)-9c′′
(S,R)-(-)-9c′
10a
H
Me
Me
Me
Me
Me
Me
Me
Me
>40
i-Pr
21.19 (2354)
>40
(R)-(-)-10a
(S)-(+)-10a
10c
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
i-Pr
s-Bu
s-Bu
s-Bu
s-Bu
s-Bu
12.43 (1130)
0.101 (101)
0.972 (69)
>40
>40
(R,R)-(-)-10c′
(S,R)-(-)-10c′′
(S,S)-(+)-10c′
(R,S)-(+)-10c′′
NVP
3.55 (3550)
>40
>40
>40
>40
7.0 (17)
35.0 (87)
0.08 (3)
EFV
0.03
3.0 (100)
NDf
aValues are means determined from at least three experiments. bCompounds were listed according to their elution order.
The underlined R or S indicates the absolute configuration of the sec-butyl chiral centre, as suggested by modelling
c
studies (see below). Inhibitory dose 50, compound dose required to inhibit the HIV-1 rRT activity by 50%. Fold
d
resistance: ratio of ID50mut/ID50wt values. dAbbreviation: s-Bu, sec-butyl. Higher concentrations could not be tested
e
due to precipitation of compounds in the culture medium. fND, not determined.
Molecular Modeling, Docking Calculations and COMBINEr predictions. A structure-based
study by the means of a previously reported COMBINEr protocol31 was undertaken to further
support experimental evidences on the absolute 9c and 10c stereoisomer configurations. For
comparison purposes, calculations were performed also on the F2-S-DABO 13e.13 The statistically
most robust COMBINEr model CM4 was used to verify the mutation role on the activity profiles of
9c and 10c stereoisomers. The binding modes of the S-DABO derivatives were analyzed by the
means of the Vina program. Similarly as reported in previous studies, the R configuration at the C6-
benzylic chiral centre displays an overall binding profile similar to that described for NH-DABOs,25
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 44
1
2
3
4
5
while the S configuration prevents such behavior (compare binding modes in Figures S1A, S1B and
S1C in Supporting Information).
6
7
8
9
Interestingly, the role of the C5 methyl group is somehow controversial. On one hand its
presence exerts a conformation constraint13,15 pre-selecting a bioactive-like conformation for (R)-C6
stereoisomers (Figure S1B), thus leading to more potent derivatives13 in particular for wt, L100I and
K103N RT isoforms. On the contrary, the enhanced rigidity is undesired for the (R)-C6
stereoisomers when binding to Y181I mutated RT, where the replacement of the Tyr181 with an
isoleucine prevent the binding mode maintenance as for the other RT isoforms (compare
conformations for (R,R)-10c' and (S,R)-10c'' stereoisomers in the last column in Figure S1B). In
fact, for the (R,R)-9c'' and (S,R)-9c' stereoisomers, the lack of the C5 methyl allows the
compounds to adopt all similar binding modes (Figure S1A), although not being always more
potent than their cognate 10c derivatives. Regarding the (S)-C6 configuration, the above described
behavior (presence of C5 methyl group) is even more stressed: in no case (S,S)-10c' and (R,S)-10c''
showed a putative binding conformation close to those of (R,R)-10c' and (S,R)-10c''
configurations. Differently, for (S,S)-9c'' and (R,S)-9c' four out of eight binding modes are close to
those of (R,R)-9c'' and (S,R)-9c' stereoisomers. This scenario is in good agreement with the
experimental activities reported in Table 5. Interestingly, in the 9c stereoisomers the decrease of
IC50 values due to the S configuration at the C6-benzylic position is at certain level alleviated by the
higher conformational flexibility due to the lack of the C5 methyl group. This is correctly
recognized by the subsequent application of the COMBINEr model by predicting, in most cases,
(S,S)-9c'' and (R,S)-9c' more potent than their C5-methyl homologues (S,S)-10c' and (R,S)-10c'',
respectively (Table 6).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For comparison purposes, the docking and binding mode analysis with the same procedure
were also performed on the 13e enantiomers.13 Experimental activities for (R)-13e and (S)-13e were
only determined against wild type RT,13 nevertheless their binding modes (Figure S1C) against the
26
ACS Paragon Plus Environment

Page 27 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
four considered isoforms were fully overlapping with those of the 9c and 10c stereoisomers
(compare Figures S1A, B and C).
6
7
8
9
Once the binding mode of 9c and 10c had been estimated, as above anticipated, the previously
reported COMBINEr model CM4 was readily applied to the eight stereoisomers. Apart the exact
numerical values, the model successfully reproduced the experimental eudismic and RT isoform
selectivity ratio (not shown) of most (R) and (S) enantiomers as well as the general activity trend
(Table 6 and Figure S2 in Supporting Information). In general, the COMBINEr model is slightly
over predictive: the greater error of prediction is obtained for (R,S)-9c', likely due also to some
inaccuracy in the binding mode proposed by Autodock Vina. As reported in Table 6, the
experimental/predicted pIC50 matching supported the indirect assigned C6-benzyl configurations
and aided to hypothesize the full absolute configurations for all the 9c and 10c stereogenic centres.
In particular, regarding the sec-butyl absolute configuration and focusing on the two (R,R)-10c' and
(S,R)-10c'' diastereomers, from the Autodock Vina proposed binding modes into the wild type RT
non-nucleoside binding site (NNBS), the slightly bulkier ethyl group seems to force the sec-butyl
moiety to point towards a wider zone delimited by residues Val106, Phe227, Leu234, His235 and
Pro236 (Figure 5). This zone identifies the opening of the channel that was found to be occupied by
the delavirdine indole group32 (Figure 5). In this context, the (R,R)-10c' (R)-sec-butyl configuration
in the NNBS adopts a relaxed conformation without any particular repulsive interaction. On the
other hand, the (S,R)-10c'' (S)-sec-butyl configuration, to avoid steric clashes with the NNBS,
forces a slight molecular rearrangement (Figure 5) leading to a minor loss of steric interactions (not
shown). Furthermore, considering the ligand binding strain energies13,33 listed in Table 6, the latter
scenario is of particular advantage for the (R)-sec-butyl over the (S)-sec-butyl compound, and is in
good agreement with the fact that (R,R) configurations are predicted more potent than the (S,R) ones
(Table 6 and Table S3 in Supporting Information).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 44
1
2
3
Table 6. COMBINEr model CM4 predicted activities (pIC50) for 9c, 10c and 13e stereoisomers.
4
5
For direct comparison the experimental pID50s are also reported.
6
7
8
9
wt
expermtl
K103N
expermtl
Y181I
expermtl
L100I
expermtl
compda
predct
9.035
9.029
9.277
9.029
9.299
9.323
5.623
5.204
9.194
7.775
predct
4.531
4.690
5.362
4.690
7.182
6.001
6.000
5.457
ND
predct
5.847
6.004
5.793
6.596
5.358
5.665
5.355
5.570
ND
predct
4.822
5.138
6.932
7.134
8.514
6.840
5.137
5.357
ND
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6.262
7.570
8.425
7.392
9.000
7.854
5.331
5.228
8.097b
6.000b
<4.699
5.211
5.792
4.830
6.996
6.012
<4.398
<4.398
NDc
<4.699
<4.699
4.850
<4.699
5.450
<4.398
<4.398
<4.398
ND
5.654
6.188
7.163
6.829
7.387
6.281
6.040
6.225
ND
(R,S)-9c'
(S,S)-9c''
(R,R)-9c''
(S,R)-9c'
(R,R)-10c'
(S,R)-10c''
(S,S)-10c'
(R,S)-10c''
(R)-13e
ND
ND
ND
ND
ND
ND
(S)-13e
aCompounds were listed according to their elution order. The underlined R or S indicates the absolute configuration of
the sec-butyl chiral centre as suggested by modelling studies; bRef. 13; cND, not determined.
Figure 5. Left: (R,R)-10c' (orange) and (S,R)-10c'' (cyan) binding mode comparison in wild type
RT (gray). Right: Delavirdine in complex with RT (PDB entry code 1KLM).
28
ACS Paragon Plus Environment

Page 29 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
Conclusions
6
7
8
9
New alkyl-S-DABOs bearing 2-chloro-6-fluoro substitution at the C6-benzyl position (2-Cl-6-
F-S-DABOs, 7-12) have been reported as NNRTIs active against wt HIV-1 (NL4-3 strain) as well
as against a panel of clinically relevant HIV-1 mutant strains (K103N, Y181C, and Y188L). In
addition, some previously reported15,28 and newly synthesized F2-S-DABOs (13-15) have been
included in the study for comparison purposes and in order to evaluate the effects of the
simultaneous alkyl substitution (Me, Et) at both C5 and C6-benzyl moiety of the pyrimidinone ring.
In general terms, the cellular and enzyme assays here presented confirmed and significantly
expanded the SAR information already known about the S-DABOs.1-3,14,15,17,18 Indeed, among the
four different substitutions studied in the present work (at the C6-benzyl moiety, at the methylene
unit of the C6-benzyl moiety (R1); at C5 position (R2); at the C2-alkylthio chain (R3)), the
simultaneous double Me substitution at C5 and C6 in the 2-chloro-6-fluoro- and 2,6-difluoro-benzyl
series, and the C5(Me)/C6(Et) disubstitution in the 2,6-difluoro-benzyl series joined to branched
(iso-propyl and sec-butyl) and cyclic (cyclopentyl) alkyl substituents at C2 seem to be suitable to
obtain the compounds with the highest, wide-spectrum inhibitory activity against HIV-1, with a
prevalence of the 2-chloro-6-fluoro derivatives 10a,c,e against wt virus and the mutant K103N, and
a slight predominance of 2,6-difluoro compounds against Y181C (13a,c,e) and Y188L (15a,c,e)
mutant strains. The 6-[1-(2-chloro-6-fluorophenyl)ethyl]-5-methyl-2-(cyclopentylthio)pyrimidin-
4(3H)-one 10e and the 6-[1-(2,6-difluorophenyl)propyl]-5-methyl-2-(sec-butylthio)pyrimidin-
4(3H)-one 15c, with EC50 values in the picomolar range against the wt virus [EC50sWT (pM): 30
(10e) and 800 (15c)], and at nanomolar level versus the mutant strains [EC50sK103N (nM): 93 (10e)
and 200 (15c); EC50sY181C: 230 (10e) and 300 (15c); EC50sY188L: 2900 (10e) and 900 (15c)], were
two of the compounds endowed with the highest, broad spectrum HIV-1 inhibitory activity, that
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
29
ACS Paragon Plus Environment