A Broad anti-influenza hybrid small molecule that potently disrupts the interaction of polymerase acidic protein-basic protein 1 (PA-PB1) subunits

Article abstract of DOI:10.1021/acs.jmedchem.5b00012

In continuing our efforts to identify small molecules able to disrupt the interaction of the polymerase acidic protein-basic protein 1 (PA-PB1) subunits of influenza virus (Flu) RNA-dependent RNA polymerase, this paper is devoted to the optimization of a

Full text of DOI:10.1021/acs.jmedchem.5b00012

Subscriber access provided by University of Virginia Libraries & VIVA (Virtual Library of Virginia)
Article
A Broad Anti-influenza Hybrid Small Molecule that Potently Disrupts the
Polymerase Acidic Protein-Basic Protein 1 (PA-PB1) Subunits Interaction
Serena Massari, Giulio Nannetti, Jenny Desantis, Giulia Muratore, Stefano
Sabatini, Giuseppe Manfroni, Beatrice Mercorelli, Violetta Cecchetti, Giorgio
Palù, Gabriele Cruciani, Arianna Loregian, Laura Goracci, and Oriana Tabarrini
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b00012 • Publication Date (Web): 09 Apr 2015
Downloaded from http://pubs.acs.org on April 12, 2015
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 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A Broad Antiꢀinfluenza Hybrid Small Molecule that Potently Disrupts
the Polymerase Acidic ProteinꢀBasic Protein 1 (PAꢀPB1) Subunits
Interaction
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
¥
§
¥
§
Serena Massari,§ ⊥ Giulio Nannetti, ⊥ Jenny Desantis, Giulia Muratore, Stefano Sabatini,
,
,
Giuseppe Manfroni,§ Beatrice Mercorelli,¥ Violetta Cecchetti,§ Giorgio Palù,¥ Gabriele Cruciani,#
Arianna Loregian,¥,*,⊥ Laura Goracci,# * and Oriana Tabarrini.§,*,⊥
Department of Pharmaceutical Sciences, University of Perugia, 06123 Perugia, Italy, Department of
Molecular Medicine, University of Padua, 35121 Padua, Italy, Department of Chemistry, Biology
and Biotechnology, University of Perugia, 06123 Perugia, Italy.
TITLE RUNNING HEAD: Polymerase PAꢀPB1 interaction triazolopyrimidine inhibitors with antiꢀ
influenza activity.
* To whom correspondence should be addressed. For OT: phone, + 39 075 585 5139; fax, +39 075
585 5115; eꢀmail, oriana.tabarrini@unipg.it. For AL: phone, +39 049 8272363; fax, +39 049
8272355; eꢀmail, arianna.loregian@unipd.it. For LG: phone, +39 075 585 5632; fax, +39 075
45646; eꢀmail, laura.goracci@unipg.it.
⊥ Authors contributed equally to this work.
§ Department of Pharmaceutical Sciences.
¥ Department of Molecular Medicine.
# Department of Chemistry, Biology, and Biotechnology.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 43
1
2
3
ABSTRACT
4
5
6
7
8
In continuing our efforts to identify small molecules able to disrupt PAꢀPB1 subunits interaction of
influenza virus (Flu) RNAꢀdependent RNA polymerase, this paper was devoted to the optimization
of a dihydrotriazolopyrimidine derivative, previously identified through a structureꢀbased drug
discovery. The structure modifications performed around the bicyclic core led to the identification
of compounds endowed with both the ability to disrupt PAꢀPB1 subunits interaction and antiꢀFlu
activity with no cytotoxicity. Very interesting results were obtained with the hybrid molecules 36
and 37, designed by merging some peculiar structural features known to impart PAꢀPB1 interaction
inhibition, with compound 36 that emerged as the most potent PAꢀPB1 interaction inhibitor (IC50 of
1.1 µM) among all the small molecules reported so far. Calculations showed a very favored Hꢀ
bonding between the 2ꢀamidic carbonyl of 36 and Q408, which seems to justify its potent ability to
interfere with the polymerase subunits interaction.
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
INTRODUCTION
Influenza is a highly contagious, acute, febrile, respiratory illness. Only two classes of drugs are
currently available to treat influenza infections, i.e. M2 ion channel inhibitors and neuraminidase
inhibitors. Unfortunately, their clinical use is limited by serious side effects, low efficacy, and the
emergence of resistant virus variants.1ꢀ3 These issues have made the M2 ion channel inhibitors even
no longer recommended,4 thus, the neuraminidase inhibitors such as oseltamivir, represent the firstꢀ
line therapy, although their widespread use has been reported to favor the emergence of resistant
viral strains.5
Clearly, new anti influenza virus (Flu) strategies based on alternative mechanisms of action are
strongly needed. On this line, we have chosen the viral RNAꢀdependent RNA polymerase (RdRP)
as a valid antiviral target since it is essential for virus replication and plays an important role in the
viral pathogenesis by contributing to the hostꢀadaptation process. In addition, it is highly conserved
among FluA, B, and C and its mechanism of action significantly differs from that of the human
ACS Paragon Plus Environment

Page 3 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
RNA polymerases.6 Flu RdRP is a complex of three subunits, polymerase acidic protein (PA),
polymerase basic protein 1 (PB1), and polymerase basic protein 2 (PB2), whose correct assembly
into functional RdRP is an essential step for Flu RNA synthesis and replication. The inhibition of
proteinꢀprotein interactions (PPI) between the subunits of the viral RNA polymerase represents an
attractive as well as challenging strategy that could expand the narrow target spectrum of antiꢀFlu
drugs.7ꢀ9 Previous studies have revealed that the Nꢀterminal and the Cꢀterminal regions of PB1
interact with the Cꢀterminus of PA10ꢀ14 and Nꢀterminus of PB2,11,15 respectively. While no small
molecule targeting PB1ꢀPB2 interaction is known, with only a few short peptides reported for this
activity,16,17 various small molecules able to disrupt the interactions between the PA and PB1
subunits have emerged during the last three years.18ꢀ25 Their PPI inhibitory activity correlated to the
inhibition of viral RdRP activity and of virus replication in cellular context, thus validating this
antiviral strategy. Different approaches led to identify the PAꢀPB1 small molecule inhibitors
reported so far (Figure 1). Nitrobenzofurazan derivatives have emerged from a biochemical ELISAꢀ
based screening,21 which was followed by an optimization study that gave compound 123 as one of
the most active. On the other hand, anthracene derivative 220 was discovered by serendipity.
However, the largest number of PAꢀPB1 interaction inhibitors was identified through a structureꢀ
based drug discovery (SBDD) strategy. In particular, compounds 3 (benzbromarone)19 and 4
(diclazuril)19 came from an in silico screening performed on a database of 4,000 drugs, while by
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
screening 3 million small molecules we have discovered the promising hit compounds 5ꢀ8.18
A
successive round of optimization performed on compounds 6 and 7 led to the improved derivatives
922 and 1022 (Figure 1), whose antiꢀFlu activity also encompassed a panel of clinical isolates of
FluA, including an oseltamivirꢀresistant strain, and of FluB, without showing appreciable
cytotoxicity.
In this study, with the aim of broadening the class of PAꢀPB1 interaction inhibitors as alternative
antiꢀFlu agents, the optimization of the dihydrotriazolopyrimidine derivative 8 is reported.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 43
1
2
3
DESIGN OF TRIAZOLOPYRIMIDINE DERIVATIVES
4
5
6
7
8
Among the hit compounds previously identified, the dihydrotriazolopyrimidine derivative 8 showed
a weak PAꢀPB1 interaction inhibitory activity (IC50 = 170.6 µM) and no antiviral activity up to 100
µM, but the lack of any cytotoxic effect (CC50 > 250 µM in MDCK cells)18 made it worthy of
further investigation. The structural modifications were initially focused on the benzamide moiety
at the Cꢀ2 position (Table 1). The methoxy group was run into ortho and meta positions
(compounds 11 and 12) and an analogous set of derivatives was synthesized having a fluorine atom
(compounds 13ꢀ15). Since in these two series of compounds the para derivatives were slightly more
active than the other isomers, the successive substituents were placed only at the para position,
synthesizing compounds 16ꢀ24. Additional derivatives were 25, characterized by a bicyclic
aromatic ring, 26, devoid of any substituent on the phenyl ring, and 27, where an ethyl chain was
inserted between the phenyl ring and the amide moiety. Since some difficulties were encountered
with the purification of this first series of derivatives, the 4,7ꢀdihydroꢀ[1,2,4]triazolo[1,5ꢀ
a]pyrimidine nucleus was oxidized leading to more manageable aromatic compounds. Derivatives
28ꢀ30 were synthesized using the phenyl, the pꢀmethoxyphenyl, and pꢀpropoxyphenyl rings,
respectively, as Cꢀ2 substituents.
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
Previous computational studies indicated that a more linear shape of the compounds could increase
the inhibitory effect.24 Moreover, the alignment of the 1,2,4ꢀtriazolo[1,5ꢀa]pyrimidine derivative 28
to the recently published pharmacophore for PAꢀPB1 inhibitors24 (Figure 2, A and B), indicated that
the 7ꢀphenyl ring was not fitting any hydrophobic pharmacophoric point (green spheres in Figure 2)
but, on the contrary, it was pointing towards polar atomic coordinates (red and blue spheres in
Figure 2). In the light of the above considerations, we decided to further modify the central core by
interchanging the methyl and the phenyl ring, synthesizing compounds 31ꢀ33. Figure 2C reports the
alignment of 31 to the pharmacophore.
Since the Hꢀbond acceptor and Hꢀbond donor pharmacophoric points underneath the amide group
proved to be essential for pharmacophore definition,24 the presence of an inverse amide group
ACS Paragon Plus Environment

Page 5 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
linking the triazolopyrimidine core and the phenyl ring was studied by synthesizing the couple of
positional isomers 34 and 35.
6
7
8
9
Finally, as additional analogues of 8, hybrid molecules were designed by combining some of the
structural features shown to confer PAꢀPB1 inhibitory activity.18,22,24 Among the planned
molecules, the positional isomers 36 and 37 (Table 1) were synthesized by merging the 1,2,4ꢀ
triazolopyrimidine nucleus with the cycloheptathiopheneꢀ3ꢀcarboxamide core that characterizes the
previous hit 7 (Figure 1). The two hybrid molecules were aligned to the pharmacophore, obtaining a
good similarity both for polar and hydrophobic regions (Figure 3), with the 1,2,4ꢀtriazolopyrimidine
nucleus being oriented in the lower hydrophobic region of the pharmacophore, in an opposite way
to compounds 28 and 31 (Figure 2), to gain the better alignment of the polar regions.
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
CHEMISTRY
The synthesis of all the target compounds was accomplished through a carbodiimideꢀmediated
coupling reaction by reacting the 2ꢀamino and 2ꢀcarboxylic acid 1,2,4ꢀtriazolo[1,5ꢀa]pyrimidine
synthones with the appropriate benzoyl chlorides and amines, respectively. Although the
triazolopyrimidines chemistry has been widely explored,26 a very few reports were focused on the
synthesis of 2ꢀamino and 2ꢀcarboxylic acid 1,2,4ꢀtriazolo[1,5ꢀa]pyrimidine derivatives. Thus, the
synthesis of the synthones 38, 41, 42, 45, and 47 has required an inꢀdepth study aimed to determine
the best synthetic pathway and conditions to obtain the desired positional isomers in high yield. In
Schemes 1 and 2, selected procedures among those reported in the literature or appositely setꢀup are
shown, while a thorough comparative study will be reported in the due course. Synthone 3827 was
prepared as reported in the literature, by reacting 3,5ꢀdiaminoꢀ1,2,4ꢀtriazole with
benzylideneacetone in DMF at reflux (Scheme 1). The same reaction was applied to prepare
synthone 4128 (Scheme 1) but adding Ac2O to the reaction mixture after the condensation
completion, in order to protect the amino group in the successive oxidation step. The oxidation of
compound 39 performed with Nꢀbromosuccinimide (NBS) gave intermediate 40, which after acid
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 43
1
2
3
4
5
6
7
8
9
hydrolysis gave the desired synthone 41. An analogous threeꢀstep procedure was planned to
synthesize the unknown synthone 42 by reacting 3,5ꢀdiaminoꢀ1,2,4ꢀtriazole with phenyl 1ꢀpropenyl
ketone in DMF at reflux (Scheme 1). Fortunately, the reaction furnished directly the aromatic
compound 42. The reaction of synthones 38, 41, and 42 with the appropriate benzoyl chloride,
performed in pyridine, furnished the target compounds 8 and 11ꢀ33. All the 4,7ꢀ
dihydrotriazolopyrimidine derivatives were purified by crystallization by EtOH, since the
compounds undergo to an extensive transformation to a main unidentified product when column
chromatography was used as purification method.
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
Similarly to the preparation of synthones 41 and 42, the synthesis of triazolopyrimidineꢀ2ꢀ
carboxylic acids 45 and 47 was accomplished by reacting ethyl 5ꢀaminoꢀ1,2,4ꢀtriazoleꢀ3ꢀ
carboxylate29 with phenyl 1ꢀpropenyl ketone and benzylideneacetone, respectively, in DMF at
reflux (Scheme 2). By using 3ꢀaminoꢀ1,2,4ꢀtriazoleꢀ5ꢀcarboxylic acid as starting material, the
reaction furnished only decarboxylated triazolopyrimidine scaffolds. In agreement with what has
been described above, the reaction with benzylideneacetone gave the saturated compound 43, which
was then oxidized with NBS in EtOH at reflux in the presence of AcONa, to give 44. On the other
hand, the reaction with phenyl 1ꢀpropenyl ketone provided directly the aromatic intermediate 46.
Ethyl triazolopyrimidine esters 44 and 46 were then hydrolyzed under basic conditions, furnishing
the key acids 45 and 47. The carbodiimideꢀmediated coupling reaction was then carried out by
reacting the 1,2,4ꢀtriazolopyrimidineꢀ2ꢀcarbonyl chloride, prepared from the acids 45 and 47 by
reaction
with
oxalyl
chloride
in
CH2Cl2,
with
2ꢀaminoꢀ5,6,7,8ꢀtetrahydroꢀ4Hꢀ
cyclohepta[b]thiopheneꢀ3ꢀcarboxamide or aniline, in CH2Cl2 in the presence of DIPEA, to give the
target compounds 34ꢀ37.
RESULTS AND DISCUSSION
Hit compound 8 (compound 31 in ref 18), previously tested as provided from vendor, was
synthesized and retested, giving comparable results in terms of PAꢀPB1 interaction inhibitory
ACS Paragon Plus Environment

Page 7 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
activity (IC50 = 153 µM) and maintaining the same profile of antiviral activity and cytotoxicity. The
inꢀhouse synthesized 8 was then used as comparative compound in all successive studies.
The whole set of derivatives were evaluated for the ability to inhibit the physical interaction
between PA and PB1 subunits by ELISA including the PB1(1–15)–Tat peptide30 as a positive control.
In parallel, for all the synthesized compounds the antiviral activity was tested by plaque reduction
assays (PRA) in MDCK cells infected with a reference FluA virus, the A/PR/8/34 strain. Ribavirin
(RBV), a known broadꢀspectrum inhibitor of RNA viruses polymerase,31 was included as a positive
control of inhibition, exhibiting an EC50 of 10 ꢁM. To exclude that the observed antiviral activities
could be due to toxic effects in the target cells, all the synthesized compounds were tested by MTT
assays in two cell lines, i.e. MDCK and HEK 293T. The antiviral activity and toxicity data for the
tested compounds are reported in Table 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
For what concern the modifications made on the phenyl ring at the Cꢀ2 position of the
dihydrotriazolopyrimidine core, the shifting of the methoxy group from para (compound 8) to meta
(compound 12) and ortho (compound 11) positions decreased the PAꢀPB1 interaction inhibitory
activity. The same behavior was observed for the fluoro analogues 13ꢀ15. Among the various para-
substituted derivatives, compounds 16ꢀ18 were more active than hit 8 but only for pꢀpropoxy
derivative 16 the ability to inhibit the PAꢀPB1 interaction perfectly translated to antiꢀFlu activity in
cellular context, with an EC50 = 47 ꢁM. On the other hand, compounds 19, 21, and 25 inhibited the
viral growth but did not interfere with the PAꢀPB1 complex formation. For these compounds, all
characterized by a bulky substituent at the Cꢀ2 position, an alternative mechanism of action might
be postulated or simply the antiviral activity could be due to a certain toxicity.
The aromatic triazolopyrimidine compounds 28ꢀ30 were all devoid of activity against the PAꢀPB1
interaction thus resulting less suitable than their 4,7ꢀdihydrotriazolpirimidine counterparts
(compounds 26, 8, and 16). The Fingerprints for Ligands and Proteins (FLAP) binding poses
evaluated in silico for compound 26 and its aromatic analogue 28 were very similar, with the
heterocyclic and heteroaromatic cores interacting with W706 (Figure 4). However, compound 28,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 43
1
2
3
4
5
6
7
8
due to its more linear shape, appears to match the hydrophobic field generated by W706 less
efficiently than 26, and this difference might be responsible for the lower activity of the aromatic
analogues.
9
When in the aromatic scaffold the methyl and phenyl groups were interchanged (compounds 31ꢀ
33), the PAꢀPB1 inhibitory activity was restored, with compound 31 that exhibited a good IC50
value of 26 ꢁM, perfectly comparable to that obtained in inhibiting the viral replication (EC50 = 25
ꢁM). The ability of compound 31 to interfere with PAꢀPB1 complex markedly decreased when the
Cꢀ2 amide linker was inverted as in compound 35, which however maintained the antiviral activity.
Definitely, the most exciting result in this study was achieved with the hybrid molecules 36 and 37,
designed by merging some peculiar structural features of the PAꢀPB1 interaction inhibitors
developed by us. With an IC50 = 1.1 ꢁM, compound 36 emerged as the most potent PAꢀPB1
inhibitor within this series of compounds as well as among all the small molecules published so far.
The other hybrid molecule 37, although endowed with a lower ability to interfere with PAꢀPB1
complex formation (IC50 = 28 ꢁM), emerged as the best in inhibiting the viral growth (EC50 = 8.0
ꢁM). Structurally, the two molecules only differ for the opposite position of the substituents on the
triazolopyrimidine core, with the 5ꢀmethyl group and 7ꢀphenyl ring being the most suitable
substitution pattern for PAꢀPB1 inhibition. To evaluate possible differences in the binding mode of
compounds 36 and 37, their FLAP binding poses were analyzed (Figure 5). Both compounds orient
the triazolpyrimidine ring toward the W706 residue of PA that defines the first hydrophobic pocket,
to give an efficient πꢀπ stacking. In addition, compound 37, due to its elongated shape, seems to be
able to locate the phenyl ring into the second hydrophobic pocket of PA, defined by P710 and L666,
and the cycloheptathiophene ring into the third hydrophobic pocket defined by L640, V636, and
W619, according to the cavity description by Liu and coworkers33 (Figure 5A). This binding mode
is comparable to that previously observed for compound 924 (data not shown). On the other hand,
compound 36 seems to be shifted towards the opposite side of the cavity, thus matching only the
first hydrophobic region (Figure 5B). Changing the perspective (Figures 5C and 5D), we can notice
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
ACS Paragon Plus Environment

Page 9 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
that, based on this model, the shift for compound 36 is due to a favored Hꢀbonding between its 2ꢀ
amidic carbonyl group and Q408, according to the FLAP GRID fields analysis. In our previous
studies, the interaction with W706 and with the second hydrophobic pocket in proximity of P710
were usually detected for the most active compounds; the achievement of more active derivatives
designed to optimize these hydrophobic interactions validated the hypothesis about this favorable
binding mode.22 On the contrary, compound 36 emerges as the first compound within our series of
derivatives for which an interaction with Q408 seems to occur. Although these findings need to be
further investigated, it is known that Q408 is a key residue for PB1 binding through Hꢀbonding, and
this might be the reason for the surprisingly potent activity. The interaction with Q408 has been
recently hypothesized also for other PAꢀPB1 inhibitors, although their inhibitory effect was lower
than that of compound 36.34 From the comparison of the spatial dispositions of the two amidic
carbonyl groups in compounds 36 and 37 in Figures 5 with the ones in Figure 3, minor differences
emerge. These differences are induced by the interaction with the protein structure, and this is a
demonstration that combining a pharmacophore approach with a structureꢀbased approach, when
possible, represents a suitable strategy for the pharmacophore refinement.
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
Compounds 16, 31, 36, and 37, endowed with the best ability to inhibit both PAꢀPB1 interaction
and viral growth, were further characterized. First, we investigated their ability to interfere with the
catalytic activity of FluA RdRP in a cellular context by minireplicon assays performed in
transfected HEK 293T cells. As shown in Table 2, for all the compounds the ability to disrupt the
PAꢀPB1 interaction in vitro well correlated with the polymerase inhibitory activity, with the best
compounds 36 and 37 showing EC50 values of 12 and 16 µM, comparable to that of the reference
compound RBV (EC50 = 18 µM). Then, we tested the ability of the best compound 36 to interfere
with the physical interaction between PA and PB1 in a cellular context. To this end, we investigated
whether the compound could affect PAꢀPB1 binding in the cell cytoplasm and consequently block
the intranuclear translocation of PA, which requires formation of a PAꢀPB1 complex,32 while PB2
enters the cell nucleus independently. Thus, we transfected HEK 293T cells with plasmids
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 43
1
2
3
4
5
6
7
8
9
expressing PB1, PB2 and a PAꢀGFP fusion protein and analyzed the intracellular localization of
PAꢀGFP in the presence or absence of compound 36 (or DMSO as a control). As previously
shown,32 individually expressed PAꢀGFP was largely cytoplasmic, while coexpression of PAꢀGFP
with PB1 and PB2 resulted in marked nuclear accumulation of PA (Figure 6). Treatment of PAꢀ
PB1ꢀPB2ꢀcoexpressing cells with compound 36 reduced PA nuclear localization, while DMSO had
no effect (Figure 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
Finally, the antiviral effects of compounds 16, 31, 36, and 37 against a number of strains of FluA of
both HIN1 and H3N2 subtypes, including an oseltamivirꢀresistant clinical isolate (A/Parma/24/09),
and three strains of FluB, were also evaluated (Table 3). All the compounds showed an interesting
antiviral activity against the whole panel of Flu strains, perfectly comparable to that measured
against the A/PR/8/34 strain. Compound 37 confirmed its antiꢀFlu potency exhibiting the best
antiviral activity, with EC50s of 5 µM against the FluA oseltamivirꢀresistant clinical isolate and Flu
B/Lee/40 strain.
At this stage, the ADME profile was preliminarily determined to assess the drugꢀlike properties of
these new small molecules. Toward this end, for the same set of compounds, 16, 31, 36, and 37
along with the hit 8, the solubility (equilibrium solubility in phosphate buffer at pH 7.4),
permeability (parallel artificial membrane permeation assay (PAMPA)), and human liver
microsome (HLM) metabolic stability were experimentally determined. Concerning the solubility,
compound 8 showed the higher equilibrium solubility (13 µg/mL), followed by compounds 16 and
31 (5 µg/mL) and then by compounds 36 and 37, which showed very low solubility (<1 µg/mL).
This issue did not allow to clearly estimate the cell permeability of compound 37. Based on the
PAMPA, all the tested compounds showed a medium/high permeability (data are reported in the
Experimental Section) and a percentage of compound trapped into the membrane ranging from 41
to 57%, with the exception of compound 31, the sole not retained by the lipid membrane. Notably,
all the studied PAꢀPB1 inhibitors were metabolically stable in HLM since no metabolites were
detected after 30ꢀmin exposure to cytochrome metabolism. Altogether, these studies suggested that
ACS Paragon Plus Environment

Page 11 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the triazolopyrimidineꢀbased compounds are endowed with an adequate ADME profile for a drugꢀ
discovery phase, even though the most promising antiviral derivatives 36 and 37 suffered from low
solubility. Since these hybrid molecules are made of a triazolopyrimidine core merged to a
cycloheptathiophene scaffold, we hypothesize that the latter may be responsible for the low
solubility. Indeed, ADME studies performed on the previous cycloheptathiophene derivatives 7 and
10 showed the same poor equilibrium solubility (<1 µg/mL) and a medium/high permeability,
although they interacted almost entirely with the lipid membrane, hampering the enrichment of the
acceptor phase. Thus, although compound 36 is characterized by a promising ADME profile,
further modifications of the cycloheptathiophene moiety will be required to achieve the best
compromise between solubility and cell permeability.
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
CONCLUSIONS
The limited therapeutic options against Flu infection along with the drug resistance issue make
imperative the search for nextꢀgeneration agents. Although the inhibition of viral polymerases is a
common approach in the identification of antiviral drugs, the development of compounds able to
interfere with the correct assembly of viral polymerase complexes is an innovative strategy.35ꢀ38
In this context, we have been recently interested in the inhibition of Flu polymerase PAꢀPB1
subunits interaction by small molecules. In the present work, starting from the
dihydrotriazolopyrimidine hit 8, a weak PAꢀPB1 interaction inhibitor with no significant antiꢀFlu
activity, and through a hitꢀtoꢀlead optimization, novel potent inhibitors endowed with a broad antiꢀ
Flu activity have been discovered. The best results were achieved with the hybrid molecules 36 and
37, obtained by combining the herein developed triazolopyrimidine and the previously explored
cycloheptathiophene scaffolds. With an IC50 = 1.1 µM, compound 36 exhibits a 150ꢀfold higher
activity than hit compound 8, definitely emerging as the most potent in the scenario of the PAꢀPB1
small molecule inhibitors developed so far. This ability translates in a broad antiꢀFlu activity with
no cytotoxicity. From this study, the triazolopyrimidine core emerges as very suitable to inhibit the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 43
1
2
3
4
5
6
7
8
9
physical interaction between PA and PB1 subunits, probably thanks to an efficient πꢀπ staking
interaction with W706 of the first hydrophobic pocket, as suggested by the FLAP poses. The
computational studies also showed that the different binding mode of the various triazolopyrimidine
derivatives is markedly influenced by the substituents at the Cꢀ5 and Cꢀ7 positions but mainly by
the group placed at the Cꢀ2 position. While the 2ꢀcycloheptanethiophene moiety is definitely able to
ensure potent activity, a wider exploration of the Cꢀ2 position is strongly required, also to meet the
request for compounds with an improved ADME profile. The design of new drugꢀlike analogues,
also based on hydrophilic interactions while decreasing unnecessary hydrophobic moieties, is also
supported by the peculiar ability of the potent compound 36 to establish a very favored Hꢀbonding
with Q408.
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
During the final drafting of this manuscript, novel crystal structures of the influenza virus
polymerase were published,39,40 thus opening new opportunities to refine the SBDD.
EXPERIMENTAL SECTION
Computational methods. The binding poses in the PA cavity were generated using the FLAP
software (Molecular Discovery Ltd., UK; www.moldiscovery.com). Indeed, the FLAP approach41
was previously used to identify inhibitors of the PAꢀPB1 complex by virtual screening18 and proved
also to be useful to design more active compounds.22 The procedure has been extensively described
elsewhere.22 As in the previous studies, the main cavity of the crystallographic structure of a large
Cꢀterminal fragment of PA (aa 257ꢀ716) (pdb code: 3CM8)14 was used as a template. A total of 50
conformers for each ligand were generated to mimic the compound flexibility, and the most
abundant protonation state predicted of each molecule was used, as predicted by MoKa.42 The
probes used to generate the GRID Molecular Interaction Fields were H (shape), DRY (hydrophobic
interactions), N1 (Hꢀbond donor) and O (Hꢀbond acceptor) interactions. The FLAP software was
also used to align the selected compounds into the pharmacophore model. The pharmacophore
ACS Paragon Plus Environment

Page 13 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
generation using FLAPpharm algorithm43 was recently described.24 Here, the pharmacophore was
used as a template to run the FLAP virtual screening module.
6
7
8
9
Chemistry. All reactions were routinely checked by TLC on silica gel 60F254 (Merck) and
visualized by using UV or iodine. Flash column chromatography separations were carried out on
Merck silica gel 60 (mesh 230ꢀ400). Melting points were determined in capillary tubes (Büchi
Electrothermal Mod. 9100) and are uncorrected. HRMS spectra were registered on Agilent
Technologies 6540 UHD Accurate Mass QꢀTOF LC/MS, HPLC 1290 Infinity. Purities of the
compounds were determined by elemental analyses and the data for C, H, and N are within ± 0.4%
of the theoretical values (purity of ≥95%). Elemental analyses were performed on a Fisons
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
1
elemental analyzer, Model EA1108CHN. H NMR and 13C NMR spectra were recorded at 200
MHz (Bruker Avance DPXꢀ200) and 400 MHz (Bruker Avance DRXꢀ400) using residual solvents
such as chloroform (δ = 7.26) or dimethylsulfoxide (δ = 2.48) as an internal standard. Chemical
shifts are given in ppm (δ) and the spectral data are consistent with the assigned structures.
Reagents and solvents were purchased from common commercial suppliers and were used as such.
After extraction, organic solutions were dried over anhydrous Na2SO4, filtered, and concentrated
with a Büchi rotary evaporator at reduced pressure. Yields are of purified product and were not
optimized. All starting materials were commercially available unless otherwise indicated.
General procedure for carbodiimide formation (Method A). A solution of the appropriate
synthones (1.0 equiv) in pyridine was added of the suitable benzoyl chloride (2.0 equiv). The
reaction mixture was maintained at room temperature until no starting material was detected by
TCL. After cooling, the reaction mixture was poured into ice/water, obtaining a precipitate which
was filtered and purified as described below.
Compounds 8 and 11ꢀ27 were prepared starting from synthone 3827 through Method A and purified
by crystallization by EtOH.
4ꢀMethoxyꢀNꢀ(5ꢀmethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide
1
(8). Yield, 100%. Light yellow crystals, mp 228ꢀ230 °C; H NMR (DMSOꢀd6) δ 1.95 (s, 3H, CH3),
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 43
1
2
3
4
5
6
7
8
9
3.75 (s, 3H, OCH3), 4.50 and 6.00 (s, each 1H, CH), 6.95 (d, J = 7.9 Hz, 2H, aromatic CH), 7.20ꢀ
7.35 (m, 5 H, aromatic CH), 7.85 (d, J = 7.9 Hz, 2H, aromatic CH), 9.50 (s, 1H, NH), 10.25 (s, 1H,
NHCO); 13C NMR (DMSOꢀd6) δ 18.7, 55.8, 59.8, 96.1, 113.9, 126.3, 127.2, 128.2, 128.9, 130.1,
131.9, 143.0, 148.9, 154.3, 162.4, 164.9. HRMS: m/z calcd for C20H19N5O2 362.1572 and
C20H19N5O2 H2O 380.1678 (M+H+), found 362.1619 and 380.1724. Anal. (C20H19N5O2 · 0.2 H2O)
C, H, N.
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
2ꢀMethoxyꢀNꢀ(5ꢀmethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide
(11). Yield, 94%. Light yellow crystals, mp 154ꢀ156 °C; 1H NMR (DMSOꢀd6) δ 1.95 (s, 3H, CH3),
3.75 (s, 3H, OCH3), 4.50 and 5.90 (s, each 1H, CH), 6.95 (t, J = 7.4 Hz, 1H, aromatic CH), 7.10 (d,
J = 8.2 Hz, 1H, aromatic CH), 7.20ꢀ7.35 (m, 5H, aromatic CH), 7.45 (t, J = 7.4 Hz, 1H, aromatic
CH), 7.55ꢀ7.65 (m, 1H, aromatic CH), 9.50 (s, 1H, NH), 10.00 (s, 1H, NHCO). Anal. (C20H19N5O2 ·
0.3 H2O) C, H, N.
3ꢀMethoxyꢀNꢀ(5ꢀmethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide
(12). Yield, 75%. Light yellow crystals, mp 159ꢀ161 °C; 1H NMR (DMSOꢀd6) δ 1.95 (s, 3H, CH3),
3.75 (s, 3H, OCH3), 4.50 and 5.90 (s, each 1H, CH), 7.00ꢀ7.45 (m, 9H, aromatic CH), 9.50 (s, 1H,
NH), 10.45 (s, 1H, NHCO). Anal. (C20H19N5O2 · 0.2 H2O) C, H, N.
4ꢀFluoroꢀNꢀ(5ꢀmethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide
1
(13). Yield, 100%. Light yellow crystals, mp 253ꢀ255 °C; H NMR (DMSOꢀd6) δ 1.95 (s, 3H,
CH3), 4.50 and 6.00 (s, each 1H, CH), 7.20ꢀ7.35 (m, 7H, aromatic CH), 7.90ꢀ7.95 (m, 2H, aromatic
CH), 9.50 (s, 1H, NH), 10.50 (s, 1H, NHCO). Anal. (C19H16FN5O · 0.4 H2O) C, H, N.
3ꢀFluoroꢀNꢀ(5ꢀmethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide
1
(14). Yield, 100%. Light yellow crystals, mp 248ꢀ250 °C; H NMR (DMSOꢀd6) δ 1.95 (s, 3H,
CH3), 4.50 and 6.00 (s, each 1H, CH), 7.20ꢀ7.35 (m, 7H, aromatic CH), 7.45ꢀ7.55 (m, 2H, aromatic
CH), 9.50 (s, 1H, NH), 10.60 (s, 1H, NHCO). Anal. (C19H16FN5O · 0.5 H2O) C, H, N.
2ꢀFluoroꢀNꢀ(5ꢀmethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide
1
(15). Yield, 100%. Light yellow crystals, mp 139ꢀ141 °C; H NMR (DMSOꢀd6) δ 1.95 (s, 3H,
ACS Paragon Plus Environment

Page 15 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
CH3), 4.50 and 5.90 (s, each 1H, CH), 7.15ꢀ7.35 (m, 7H, aromatic CH), 7.45ꢀ7.55 (m, 2H, aromatic
CH), 9.50 (s, 1H, NH), 10.50 (s, 1H, NHCO). Anal. (C19H16FN5O · 0.3 H2O) C, H, N.
6
7
8
Nꢀ(5ꢀMethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)ꢀ4ꢀpropoxybenzamide
9
1
(16). Yield, 100%. Light yellow crystals, mp 235ꢀ237 °C; H NMR (DMSOꢀd6) δ 0.95 (t, J = 7.0
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
Hz, 3H, CH2CH3), 1.70 (sex, J = 7.0 Hz, 2H, CH2CH3), 1.90 (s, 3H, CH3), 3.95 (t, J = 7.0 Hz, 2H,
OCH2), 4.50 and 5.95 (s, each 1H, CH), 6.90 (d, J = 8.7 Hz, 2H, aromatic CH), 7.20ꢀ7.35 (m, 5H,
aromatic CH), 7.80 (d, J = 8.7 Hz, 2H, aromatic CH), 9.50 (s, 1H, NH), 10.25 (s, 1H, NHCO).
Anal. (C22H23N5O2 · 0.2 H2O) C, H, N.
4ꢀButoxyꢀNꢀ(5ꢀmethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide
1
(17). Yield, 45%. White crystals, mp 249ꢀ251 °C; H NMR (DMSOꢀd6) δ 0.90 (t, J = 7.4 Hz, 3H,
CH2CH3), 1.40 (sex, J = 7.0 Hz, 2H, CH2CH3), 1.70 (quin, J = 6.5 Hz, 2H, CH2CH2CH3), 1.85 (s,
3H, CH3), 4.00 (t, J = 6.5 Hz, 2H, OCH2), 4.50 and 5.95 (s, each 1H, CH), 6.90 (d, J = 8.5 Hz, 2H,
aromatic CH), 7.20ꢀ7.35 (m, 5H, aromatic CH), 7.80 (d, J = 8.5 Hz, 2H, aromatic CH), 9.50 (s, 1H,
NH), 10.25 (s, 1H, NHCO). Anal. (C23H25N5O2 · 0.2 H2O) C, H, N.
4ꢀButylꢀNꢀ(5ꢀmethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide (18).
1
Yield, 51%. White crystals, mp 261ꢀ263 °C; H NMR (DMSOꢀd6) δ 0.90 (t, J = 7.4 Hz, 3H,
CH2CH3), 1.40 (sex, J = 7.4 Hz, 2H, CH2CH3), 1.70 (quin, J = 7.4 Hz, 2H, CH2CH2CH3), 1.85 (s,
3H, CH3), 2.55 (t, J = 7.4 Hz, 2H, CCH2), 4.50 and 5.95 (s, each 1H, CH), 7.20ꢀ7.25 (m, 5H,
aromatic CH), 7.30ꢀ7.35 (m, 2H, aromatic CH), 7.80 (d, J = 8.1 Hz, 2H, aromatic CH), 9.50 (s, 1H,
NH), 10.25 (s, 1H, NHCO). Anal. (C23H25N5O · 0.4 H2O) C, H, N.
Nꢀ(5ꢀMethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)ꢀ4ꢀphenoxybenzamide
1
(19). Yield, 84%. White crystals, mp 217ꢀ219 °C; H NMR (DMSOꢀd6) δ 1.95 (s, 3H, CH3), 4.50
and 5.90 (s, each 1H, CH), 6.90ꢀ7.40 (m, 12 H, aromatic CH), 7.90 (d, J = 8.4 Hz, 2H, aromatic
CH), 9.55 (s, 1H, NH), 10.40 (s, 1H, NHCO). Anal. (C25H21N5O2 · 0.3 H2O) C, H, N.
4ꢀIsopropoxyꢀNꢀ(5ꢀmethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀ
1
yl)benzamide (20). Yield, 20%. White crystals, mp 229ꢀ231 °C; H NMR (DMSOꢀd6) δ 1.20 (d, J
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 43
1
2
3
4
5
6
7
8
9
= 6.0 Hz, 6H, CH(CH3)2), 1.80 (s, 3H, CH3), 4.50 (s, 1H, CH), 4.60 (quin, J = 6.0 Hz, 1H,
CH(CH3)2), 5.85 (s, 1H, CH), 6.90 (d, J = 8.7 Hz, 2H, aromatic CH), 7.20ꢀ7.35 (m, 5H, aromatic
CH), 7.80 (d, J = 8.7 Hz, 2H, aromatic CH), 9.55 (s, 1H, NH), 10.25 (s, 1H, NHCO). Anal.
(C22H23N5O2 · 0.2 H2O) C, H, N.
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
Nꢀ(5ꢀMethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)biphenylꢀ4ꢀ
carboxamide (21). Yield, 56%. Light brown crystals, mp 257ꢀ259 °C; 1H NMR (DMSOꢀd6) δ 1.85
(s, 3H, CH3), 4.50 and 5.90 (s, each 1H, CH), 7.20ꢀ7.45 (m, 8H, aromatic CH), 7.60ꢀ7.70 (m, 4H,
aromatic CH), 7.90 (d, J = 8.4 Hz, 2H, aromatic CH), 9.55 (s, 1H, NH), 10.50 (s, 1H, NHCO).
Anal. (C25H21N5O · 0.3 H2O) C, H, N.
4ꢀEthoxyꢀNꢀ(5ꢀmethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide
1
(22). Yield, 36%. Pink crystals, mp 227 °C (dec.); H NMR (DMSOꢀd6) δ 1.25 (t, J = 7.0 Hz, 3H,
CH2CH3), 1.80 (s, 3H, CH3), 4.00 (q, J = 7.0 Hz, 2H, CH2CH3), 4.50 and 5.90 (s, each 1H, CH),
6.90 (d, J = 8.8 Hz, 2H, aromatic CH), 7.20ꢀ7.30 (m, 5H, aromatic CH), 7.80 (d, J = 8.8 Hz, 2H,
aromatic CH), 9.55 (s, 1H, NH), 10.25 (s, 1H, NHCO). Anal. (C21H21N5O2 · 0.5 H2O) C, H, N.
Nꢀ(5ꢀMethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)ꢀ4ꢀnitrobenzamide (23).
1
Yield, 60%. Light yellow crystals, mp 272ꢀ274 °C; H NMR (DMSOꢀd6) δ 1.80 (s, 3H, CH3), 4.50
and 5.90 (s, each 1H, CH), 7.15ꢀ7.30 (m, 5H, aromatic CH), 8.05 and 8.25 (d, J = 8.8 Hz, each 2H,
aromatic CH), 9.55 (s, 1H, NH), 10.85 (s, 1H, NHCO). Anal. (C19H16N6O3 · 0.4 H2O) C, H, N.
4ꢀChloroꢀNꢀ(5ꢀmethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide
1
(24). Yield, 31%. White crystals, mp 254ꢀ256 °C; H NMR (DMSOꢀd6) δ 1.80 (s, 3H, CH3), 4.50
and 5.90 (s, each 1H, CH), 7.10ꢀ7.30 (m, 5H, aromatic CH), 7.45 and 7.80 (d, J = 8.8 Hz, each 2H,
aromatic CH), 9.55 (s, 1H, NH), 10.55 (s, 1H, NHCO). Anal. (C19H16ClN5O · 0.2 H2O) C, H, N.
Nꢀ(5ꢀMethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)ꢀ2ꢀnaphthamide
(25).
1
Yield, 42%. Pink crystals, mp 230 °C (dec.); H NMR (DMSOꢀd6) δ 1.80 (s, 3H, CH3), 4.50 and
5.90 (s, each 1H, CH), 7.20ꢀ7.30 (m, 5H, aromatic CH), 7.50ꢀ7.55 (m, 2H, aromatic CH), 7.90ꢀ8.00
ACS Paragon Plus Environment

Page 17 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
(m, 4H, aromatic CH), 8.45ꢀ8.50 (m, 1H, aromatic CH), 9.55 (s, 1H, NH), 10.50 (s, 1H, NHCO).
Anal. (C23H19N5O · 0.3 H2O) C, H, N.
6
7
8
Nꢀ(5ꢀMethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide (26). Yield,
9
1
18%. Light brown crystals, mp 257 °C (dec.); H NMR (DMSOꢀd6) δ 1.80 (s, 3H, CH3), 4.50 and
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
5.90 (s, each 1H, CH), 7.20ꢀ7.50 (m, 8H, aromatic CH), 7.75 and 7.80 (m, 2H aromatic CH), 9.55
(s, 1H, NH), 10.50 (s, 1H, NHCO). Anal. (C19H17N5O · 0.2 H2O) C, H, N.
Nꢀ(5ꢀMethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)ꢀ3ꢀphenylpropanamide
1
(27). Yield, 9%. White crystals, mp 169ꢀ171 °C; H NMR (DMSOꢀd6) δ 1.80 (s, 3H, CH3), 2.50
and 2.75 (t, J = 7.2 Hz, each 2H, CH2), 4.50 and 5.90 (s, each 1H, CH), 7.00ꢀ7.25 (m, 10H,
aromatic CH), 9.55 (s, 1H, NH), 10.00 (s, 1H, NHCO). Anal. (C21H21N5O · 0.3 H2O) C, H, N.
Compounds 28ꢀ30 were prepared starting from synthone 4128 through Method A and purified by
flash chromatography eluting with CHCl3/MeOH (98:2).
Nꢀ(5ꢀMethylꢀ7ꢀphenyl[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide (28). Yield, 19%. White
crystals, mp 183ꢀ185 °C 1H NMR (DMSOꢀd6) δ 2.75 (s, 3H, CH3), 7.05 (s, 1H, Hꢀ6), 7.40ꢀ7.55 (m,
6H, aromatic CH), 8.00 (d, J = 7.0 Hz, 2H, aromatic CH), 8.15ꢀ8.20 (m, 2H, aromatic CH), 9.75 (s,
1H, NHCO); 13C NMR (CDCl3) δ 25.1, 109.4, 127.9, 128.6, 128.9, 129.3, 129.6, 131.8, 132.2,
133.7, 146.8, 154.9, 160.0, 164.7, 165.0. HRMS: m/z calcd for C19H15N5O 330.1310 (M + H+),
found 330.1359.
4ꢀMethoxyꢀNꢀ(5ꢀmethylꢀ7ꢀphenyl[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide (29). Yield,
1
68%. White crystals, mp 190ꢀ192 °C; H NMR (DMSOꢀd6) δ 2.75 (s, 3H, CH3), 3.80 (s, 3H,
OCH3), 6.90 (d, J = 8.8 Hz, 2H, aromatic CH) 7.05 (s, 1H, Hꢀ6), 7.50ꢀ7.55 (m, 3H, aromatic CH),
7.95 (d, J = 8.8 Hz, 2H, aromatic CH), 8.15ꢀ8.20 (m, 2H, aromatic CH), 9.25 (s, 1H, NHCO). Anal.
(C20H17N5O2) C, H, N.
Nꢀ(5ꢀMethylꢀ7ꢀphenyl[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)ꢀ4ꢀpropoxybenzamide (30). Yield,
1
27%. White crystals, mp 180ꢀ182 °C; H NMR (DMSOꢀd6) δ 1.05 (t, J = 7.0 Hz, 3H, CH2CH3),
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 43
1
2
3
4
5
6
7
8
9
1.90 (sex, J = 7.0 Hz, 2H, CH2CH3), 2.75 (s, 3H, CH3), 4.00 (t, J = 7.0 Hz, 2H, OCH2), 6.90 (d, J =
8.8 Hz, 2H, aromatic CH) 7.05 (s, 1H, Hꢀ6), 7.55ꢀ7.65 (m, 3H, aromatic CH), 8.00 (d, J = 8.8 Hz,
2H, aromatic CH), 8.20ꢀ8.25 (m, 2H, aromatic CH), 9.30 (bs, 1H, NHCO). Anal. (C22H21N5O2) C,
H, N.
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
Compounds 31ꢀ33 were prepared starting from synthone 42 through Method A and purified by
flash chromatography eluting with CHCl3/MeOH (98:2).
Nꢀ(7ꢀMethylꢀ5ꢀphenyl[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide (31). Yield, 23%. Light
1
brown crystals, mp 217 °C (dec.); H NMR (DMSOꢀd6) δ 2.75 (s, 3H, CH3), 7.50ꢀ7.65 (m, 6H,
aromatic CH), 7.90 (s, 1H, Hꢀ6), 8.05 (d, J = 7.0 Hz, 2H, aromatic CH), 8.25ꢀ8.30 (m, 2H, aromatic
CH), 11.50 (s, 1H, NHCO); 13C NMR (DMSOꢀd6) δ 17.5, 107.3, 127.8, 128.5, 128.8, 129.5, 131.5,
132.6, 134.1, 136.6, 148.2, 154.5, 159.8, 160.8, 165.3. HRMS: m/z calcd for C19H15N5O 330.1310
(M + H+), found 330.1355.
4ꢀMethoxyꢀNꢀ(7ꢀmethylꢀ5ꢀphenyl[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)benzamide (32). Yield,
1
42%. White crystals, mp 210ꢀ213 °C; H NMR (DMSOꢀd6) δ 2.80 (s, 3H, CH3), 3.85 (s, 3H,
OCH3), 7.05 (d, J = 8.4 Hz, 2H, aromatic CH) 7.55ꢀ7.60 (m, 3H, aromatic CH), 7.90 (s, 1H, Hꢀ6),
8.05 (d, J = 8.4 Hz, 2H, aromatic CH), 8.25ꢀ8.30 (m, 2H, aromatic CH), 11.25 (s, 1H, NHCO).
Anal. (C20H17N5O2) C, H, N.
Nꢀ(7ꢀMethylꢀ5ꢀphenyl[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀyl)ꢀ4ꢀpropoxybenzamide (33). Yield,
1
35%. White crystals, mp 184ꢀ186 °C; H NMR (DMSOꢀd6) δ 1.00 (t, J = 7.0 Hz, 3H, CH2CH3),
1.75 (sex, J = 7.0 Hz, 2H, CH2CH3), 2.85 (s, 3H, CH3), 4.00 (t, J = 7.0 Hz, 2H, OCH2), 7.05 (d, J =
8.8 Hz, 2H, aromatic CH) 7.55ꢀ7.60 (m, 3H, aromatic CH), 7.90 (s, 1H, Hꢀ6), 8.05 (d, J = 8.8 Hz,
2H, aromatic CH), 8.25ꢀ8.30 (m, 2H, aromatic CH), 11.25 (s, 1H, NHCO). Anal. (C22H21N5O2) C,
H, N.
General procedure for carbodiimide formation (Method B). To a solution of the suitable
carboxylic acid (1.0 equiv) in well dry CH2Cl2, oxalyl chloride (3 equiv) was added and after 30
min DMF (2 drops) was added. After 2 h, the reaction mixture was evaporated to dryness to give a
ACS Paragon Plus Environment

Page 19 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
residue that was dissolved in well dry CH2Cl2 and added of the appropriate aniline (1.0 equiv) and
DIPEA (1.0 equiv). After 1 h, the reaction mixture was evaporated to dryness to give a residue that
was treated with ice/water to give a solid that was filtered and purified as described below.
Compounds 34 and 36 were prepared starting from synthone 45 through Method B and purified by
flash chromatography eluting with CHCl3/MeOH (98:2) and then by crystallization by EtOH/DMF
mixture.
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
7ꢀMethylꢀN,5ꢀdiphenyl[1,2,4]triazolo[1,5ꢀa]pyrimidineꢀ2ꢀcarboxamide (34). Yield, 42%.
Yellow crystals, mp 186ꢀ188 °C; 1H NMR (CDCl3) δ 2.85 (s, 3H, CH3), 7.15ꢀ7.20 (m, 2H, Hꢀ6 and
aromatic CH), 7.40 (t, J = 7.9 Hz, 2H, aromatic CH), 7.60ꢀ7.65 (m, 3H, aromatic CH), 7.80 (d, J =
7.9 Hz, 2H, aromatic CH), 8.15ꢀ8.20 (m, 2H, aromatic CH), 9.00 (s, 1H, NHCO). HRMS: m/z calcd
for C19H15N5O 330.1310 (M + H+), found 330.1358. Anal. (C19H15N5O) C, H, N.
5ꢀMethylꢀN,7ꢀdiphenyl[1,2,4]triazolo[1,5ꢀa]pyrimidineꢀ2ꢀcarboxamide (35). Yield, 39%. White
1
crystals, mp 242ꢀ245 °C; H NMR (CDCl3) δ 3.00 (s, 3H, CH3), 7.20 (t, J = 7.3 Hz, 1H, aromatic
CH), 7.40 (t, J = 7.9 Hz, 2H, aromatic CH), 7.60ꢀ7.65 (m, 4H, Hꢀ6 and aromatic CH), 7.85 (d, J =
7.9 Hz, 2H, aromatic CH), 8.25ꢀ8.30 (m, 2H, aromatic CH), 9.50 (s, 1H, NHCO). HRMS: m/z calcd
for C19H15N5O 330.1310 (M + H+), found 330.1360. Anal. (C19H15N5O) C, H, N.
Nꢀ[3ꢀ(aminocarbonyl)ꢀ5,6,7,8ꢀtetrahydroꢀ4Hꢀcyclohepta[b]thienꢀ2ꢀyl]ꢀ7ꢀmethylꢀ5ꢀ
phenyl[1,2,4]triazolo[1,5ꢀa]pyrimidineꢀ2ꢀcarboxamide (36). Yield, 30%. Yellow crystals, mp
1
288 °C (dec.); H NMR (DMSOꢀd6) δ 1.50ꢀ1.65 (m, 4H, cycloheptane CH2), 1.75ꢀ1.80, (m, 2H,
cycloheptane CH2), 2.60ꢀ2.70 (m, 5H, cycloheptane CH2 and CH3), 2.75ꢀ2.80 (m, 2H, cycloheptane
CH2), 7.45ꢀ7.60 (m, 6H, aromatic CH, NH2, and Hꢀ6), 8.20ꢀ8.25 (m, 2H, aromatic CH), 12.00 (s,
1H, NHCO); 13C NMR (DMSOꢀd6) δ 25.34, 27.5, 27.9, 28.6, 28.7, 31.9, 112.3, 122.2, 129.1, 129.7,
130.0, 131.7, 132.2, 135.6, 136.7, 147.1, 155.5, 155.9, 157.5, 167.4, 167.7. HRMS: m/z calcd for
C23H22N6O2S 447.1558 (M + H+), found 447.1611. Anal. (C23H22N6O2S) C, H, N.
Nꢀ[3ꢀ(Aminocarbonyl)ꢀ5,6,7,8ꢀtetrahydroꢀ4Hꢀcyclohepta[b]thienꢀ2ꢀyl]ꢀ5ꢀmethylꢀ7ꢀ
phenyl[1,2,4]triazolo[1,5ꢀa]pyrimidineꢀ2ꢀcarboxamide (37). Yield, 25%. Orange crystals, mp
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 43
1
2
3
1
313 °C (dec.); H NMR (DMSOꢀd6) δ 1.50ꢀ1.65 (m, 4H, cycloheptane CH2), 1.75ꢀ1.80, 2.70ꢀ2.75,
and 2.80ꢀ2.85 (m, each 2H, cycloheptane CH2), 2.90 (s, 3H, CH3), 7.45ꢀ7.60 (m, 5H, aromatic CH
and NH2), 8.05 (s, 1H, Hꢀ6), 8.25ꢀ8.30 (m, 2H, aromatic CH), 12.00 (s, 1H, NHCO); 13C NMR
(DMSOꢀd6) δ 17.4, 27.5, 27.9, 28.6, 28.7, 32.0, 109.2, 122.1, 128.1, 129.6, 131.8, 132.1, 135.6,
136.2, 136.7, 149.7, 155.2, 155.5, 157.9, 161.9, 167.7. HRMS: m/z calcd for C23H22N6O2S
447.1558 (M + H+), found 447.1601. Anal. (C23H22N6O2S) C, H, N.
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
7ꢀMethylꢀ5ꢀphenyl[1,2,4]triazolo[1,5ꢀa]pyrimidinꢀ2ꢀamine (42). To a solution of 3,5ꢀdiaminoꢀ
1,2,4ꢀtriazole (0.1 g, 0.001 mmol) in DMF (5 mL) at reflux, phenyl 1ꢀpropenyl ketone (0.14 mL,
0.001 mmol) was added and the reaction mixture was maintained at reflux for 30 min. After
cooling, the mixture was evaporated to dryness, obtaining a residue that was purified by flash
1
chromatography eluting with CH2Cl2/MeOH (98:2), to give 42 (0.1 g, 44%); mp 225ꢀ227 °C; H
NMR (DMSOꢀd6) δ 2.60 (s, 3H, CH3), 6.35 (s, 2H, NH2), 7.45ꢀ7.50 (m, 3H, aromatic CH), 7.05 (s,
1H, Hꢀ6), 8.05ꢀ8.15 (m, 2H, aromatic CH); 13C NMR (DMSOꢀd6) δ 17.4, 105.4, 127.4, 129.3,
130.9, 137.0, 146.1, 155.1, 157.4, 167.3.
Ethyl 5ꢀmethylꢀ7ꢀphenylꢀ4,7ꢀdihydro[1,2,4]triazolo[1,5ꢀa]pyrimidineꢀ2ꢀcarboxylate (43). The
title compound was prepared starting from ethyl 5ꢀaminoꢀ1,2,4ꢀtriazoleꢀ3ꢀcarboxylate29 and
following the same procedure as used for the synthesis of 42 (2 h) using benzylideneacetone. After
cooling, the reaction mixture was poured into ice/water obtaining a precipitate that was filtered and
1
washed with Et2O, to give 43 in 29% yield, which was used without further purification; H NMR
(DMSOꢀd6) δ 1.20 (t, J = 7.1 Hz, 3H, CH2CH3), 1.80 (s, 3H, CH3), 4.20 (q, J = 7.1 Hz, 2H,
CH2CH3), 4.55 and 6.00 (bs, each 1H, CH), 7.20ꢀ7.30 (m, 5H, aromatic CH), 9.75 (s, 1H, NH).
Ethyl 5ꢀmethylꢀ7ꢀphenyl[1,2,4]triazolo[1,5ꢀa]pyrimidineꢀ2ꢀcarboxylate (44). To a suspension of
43 (1.4 g, 4.92 mmol) in EtOH (30 mL), NaOAc (0.59 g, 7.18 mmol) was added and the reaction
mixture was heated at 40ꢀ50 °C. Then, NBS (0.87 g, 4.88 mmol) was added and the reaction
mixture was refluxed for 30 min. After cooling, the reaction mixture was poured into ice/water
obtaining a precipitate that was filtered and purified by flash chromatography eluting with
ACS Paragon Plus Environment

Page 21 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
1
CH2Cl2/MeOH (98:2), to give 44 (0.45 g, 33%); mp 174ꢀ176 °C; H NMR (CDCl3) δ 1.45 (t, J =
7.1 Hz, 3H, CH2CH3), 2.75 (s, 3H, CH3), 4.50 (q, J = 7.1 Hz, 2H, CH2CH3), 7.15 (s, 1H, Hꢀ6), 7.55ꢀ
7.65 (m, 3H, aromatic CH), 8.05ꢀ8.10 (m, 2H, aromatic CH).
9
5ꢀMethylꢀ7ꢀphenyl[1,2,4]triazolo[1,5ꢀa]pyrimidineꢀ2ꢀcarboxylic acid (45). A mixture of 44
(0.45 g, 1.59 mmol) and NaOH (0.079 g, 1.99 mmol) in MeOH (10 mL) was refluxed for 30 min.
After cooling, the reaction mixture was poured into ice/water and acidified with 2N HCl obtaining a
precipitate that was filtered. The filtrate was extracted with EtOAc and the organic layers were
evaporated to dryness obtaining a residue that was treated with Et2O and filtered, to give 45 (0.13 g,
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
1
32%); mp 172ꢀ174 °C; H NMR (DMSOꢀd6) δ 2.70 (s, 3H, CH3), 7.55ꢀ7.60 (m, 4H, Hꢀ6 and
aromatic CH), 8.05ꢀ8.10 (m, 2H, aromatic CH).
Ethyl 7ꢀmethylꢀ5ꢀphenyl[1,2,4]triazolo[1,5ꢀa]pyrimidineꢀ2ꢀcarboxylate (46). The title
compound was prepared starting from ethyl 5ꢀaminoꢀ1,2,4ꢀtriazoleꢀ3ꢀcarboxylate29 following the
same procedure as used for the synthesis of 42 (1 h). After cooling, the reaction mixture was poured
into ice/water and extracted with EtOAc. The organic layers were evaporated to dryness obtaining a
1
residue that was treated with Et2O, to give 46 in 11% yield; mp 170ꢀ172 °C; H NMR (DMSOꢀd6)
δ1.30 (t, J = 7.1 Hz, 3H, CH2CH3), 2.80 (s, 3H, CH3), 4.50 (q, J = 7.1 Hz, 2H, CH2CH3), 7.50ꢀ7.65
(m, 3H, aromatic CH), 8.05 (s, 1H, Hꢀ6), 8.15ꢀ8.25 (m, 2H, aromatic CH).
7ꢀMethylꢀ5ꢀphenyl[1,2,4]triazolo[1,5ꢀa]pyrimidineꢀ2ꢀcarboxylic acid (47). The title compound
was prepared from 46 following the same procedure as used for the synthesis of 45 (2 h). After
cooling, the reaction mixture was evaporated to dryness obtaining a residue that was treated with
ice/water and acidified with 2N HCl obtaining a precipitate that was filtered and washed with Et2O,
1
to give 47 in 87% yield; mp 168ꢀ169 °C; H NMR (DMSOꢀd6) δ 2.90 (s, 3H, CH3), 7.50ꢀ7.60 (m,
3H, aromatic CH), 8.05 (s, 1H, Hꢀ6), 8.20ꢀ8.25 (m, 2H, aromatic CH).
Compounds 35 and 37 were prepared starting from synthone 47 through Method B and purified by
flash chromatography eluting with CHCl3/MeOH (98:2) and then by crystallization by EtOH/DMF
mixture.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 43
1
2
3
Biology
4
5
6
7
8
Compounds and peptide. RBV (1ꢀDꢀribofuranosylꢀ1,2,4ꢀtriazoleꢀ3ꢀcarboxamide) was obtained
from Roche. Each test compound was dissolved in DMSO 100%. The PB1(1–15)–Tat peptide was
synthesized and purified by the Peptide Facility of CRIBI Biotechnology Center (University of
Padua, Padua, Italy). This peptide contains the first 15 Nꢀterminal amino acids of PB1 protein
conjugated to the Cꢀterminal sequence of HIV Tat protein (amino acids 47−59), which allows
intracellular delivery.44
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
Plasmids. Plasmids pcDNAꢀPB1, pcDNAꢀPB2, pcDNAꢀPA, and pcDNAꢀNP, containing cDNA
copies of the influenza A/PR/8/34 virus PB1, PB2, PA, and NP genes, respectively, were created as
described elsewhere45 and kindly provided by P. Digard (Roslin Institute, University of Edinburgh,
United Kingdom). Plasmid pPolIꢀFluꢀffLuc, which contains an influenza virusꢀbased luciferase
minireplicon vRNA under the control of the human RNA polymerase I promoter, was provided by
L. Tiley (University of Cambridge, United Kingdom). Plasmid pRLꢀSV40 expressing the Renilla
luciferase was purchased from Promega. The pcDNAꢀPAꢀGFP plasmid, which encodes a PAꢀGFP
fusion protein was kindly provided by E. Fodor (University of Oxford, United Kingdom).
Cells and Virus. Human Embryonic Kidney (HEK) 293T and MardinꢀDarby canine kidney
(MDCK) cells were grown in Dulbecco’s modified Eagle’s medium (DMEM, Life Biotechnologies)
supplemented with 10% (v/v) fetal bovine serum (FBS, Life Technologies) and antibiotics (100
U/mL penicillin and 100 µg/mL streptomycin, Life Technologies). The cells were maintained at 37
°C in a humidified atmosphere with 5% CO2. Influenza A/PR/8/34 virus (H1N1, Cambridge
lineage) was obtained from P. Digard (Roslin Institute, University of Edinburgh, United Kingdom).
The FluA strains A/Wisconsin/67/05 and A/Solomon Island/3/06 and the FluB strains
B/Malaysia/2506/4 and B/Bangladesh/333/07 were provided by R. Cusinato (Clinical Microbiology
and Virology Unit, Padua University Hospital, Padua, Italy). The clinical isolate A/Parma/24/09
was kindly provided by I. Donatelli (Istituto Superiore di Sanità, Rome, Italy). Influenza B/Lee/40
ACS Paragon Plus Environment

Page 23 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
virus was obtained from W. S. Barclay (Imperial College, London, United Kingdom). All influenza
viruses were propagated in MDCK cells.
6
7
8
9
PAꢀPB1 interaction enzymeꢀlinked immunosorbent assay (ELISA). The PA−PB1 interaction
was detected by a procedure already described.18 Briefly, microtiter plates (Nuova Aptca) were
coated with 400 ng of 6HisꢀPA(239−716) for 3 h at 37 °C and then blocked with 2% BSA (Sigma) in
PBS for 1 h at 37 °C. The 6HisꢀPA(239−716) protein was expressed in E. coli strain BL21(DE3)pLysS
and purified as described.18 After washing, 200 ng of GSTꢀPB1(1−25), or of GST alone as a
control, in the absence or the presence of test compounds at various concentrations (10, 50, 100,
200 ꢁM), were added and incubated O/N at room temperature. Escherichia coliꢀexpressed, purified
GST and GSTꢀPB1(1−25) proteins were obtained as previously described.18 After washing, the
interaction between 6HisꢀPA(239−716) and GSTꢀPB1(1−25) was detected with an antiꢀGST
monoclonal antibody conjugated to horseradish peroxidase (HRP) (GenScript). Following washes,
the substrate 3,3′,5,5′tetramethylbenzidine (TMB, KPL) was added and absorbance was measured at
450 nm by an ELISA plate reader (Tecan Sunrise™). Values obtained from the samples treated
with only DMSO were used to set as 100% of PA−PB1 interaction.
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 assay. Cytotoxicity of compounds was tested in MDCK and HEK 293T cells by the 3ꢀ
(4,5ꢀdimethylthiazolꢀ2ꢀyl)ꢀ2,5ꢀdiphenyl tetrazolium bromide (MTT) method, as previously
reported.18
Plaque reduction assay (PRA). The experiments were carried out as previously described.18
Briefly, confluent monolayers of MDCK cells were infected with FluA or FluB virus at 40
PFU/well in DMEM supplemented with 1 ꢁg/mL of TPCKꢀtreated trypsin (Worthington
Biochemical Corporation) and 0.14% BSA and incubated for 1 h at 37°C. The influenza viruses
infection was performed in the presence of different concentrations of test compounds or solvent
(DMSO) as a control. Serumꢀfree medium containing 1 ꢁg/mL of TPCKꢀtreated trypsin, 0.14%
BSA, 1.2% Avicel, and DMSO or test compounds at the indicated concentrations was added after 1
h of virus adsorption. After 48 h of incubation, cells were fixed with 4% formaldehyde and stained
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 43
1
2
3
4
5
with 0.1% toluidine blue. Viral plaques were counted, and the mean plaque number in the DMSOꢀ
treated control was set at 100%.
6
Minireplicon assay. HEK 293T cells were seeded at 105 per well on 24ꢀwell plates. After 24 h,
cells were transfected, in the presence of test compounds at different concentrations or DMSO, with
pcDNAꢀPB1, pcDNAꢀPB2, pcDNAꢀPA, pcDNAꢀNP plasmids (100 ng/well of each) along with
pPolIꢀFluꢀffLuc plasmid (50 ng/well) as described elsewhere.18 In addition, a plasmid constitutively
expressing Renilla luciferase, pRLꢀSV40 (50 ng/well), was included in the transfection mixture to
normalize variations in transfection efficiency. After 4 h, the medium was replaced with fresh
DMEM containing DMSO or test compounds. At 24 h postꢀtransfection, cells were lysed, and the
relative firefly and Renilla luciferase activities were determined using the Dual Luciferase Assay
Kit from Promega and a luminescence counter (Victor X2, PerkinElmer). The luciferase activities
of the samples treated with DMSO were set as 100%.
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
Analysis of nuclear accumulation of the PAꢀPB1 complex in the presence of compound 36.
HEK 293T cells were transiently transfected using the ArrestꢀIN™ (Biosystems) with pcDNAꢀPAꢀ
GFP, pcDNAꢀPB1, and pcDNAꢀPB2 in the absence or the presence of compound 36 (50 ꢀM) or
DMSO as a control. At 24 hr postꢀtransfection, cells were fixed for 20 min with 4% formaldehyde
in PBS. After permeabilization with 0.2 % Triton Xꢀ100 in PBS for 5 min at RT, cells were
incubated for 20 min with TOTOꢀ3 iodide (Molecular Probes) in PBS and 4% FBS, mounted using
mounting fluid (70% glycerol in PBS), and imaged using a Leica TCSꢀNT/SP2 confocal
microscope equipped with a 63X oil immersion objective. Images were digitally analyzed with
Leica software.
Preliminary ADME studies
Metabolic stability. The selected substrates were incubated with human liver microsomes (HLM,
0.5 mg protein/ml) (BD Biosciences) according to manufacturer’s recommendations with minor
modification. In summary, substrates at 5 µM final concentration were preꢀincubated in a shaking
water bath for 5 min at 37 °C in 100 mM potassium phosphate buffer (pH 7.4) in a total volume of
ACS Paragon Plus Environment

Page 25 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
250 µL. Thus, 10 µL of a 1 mM NADPH aqueous solution were added to activate the enzymatic
process. After incubation for 0 and 30 min, the addition of 250 µL of cold acetonitrile (containing
0.6 ꢁM labetalol as internal standard) was used to terminate the reaction. Proteins were precipitated
by centrifugation at 12,000 x g for 5 min at 4 °C, and aliquots of supernatants were analyzed by
HPLCꢀMS/MS, on an Agilent 6550 UHD accurate mass QꢀTOF LC/MSMS system (Agilent
Technologies, Palo Alto, CA) governed by Agilent MassHunter software (B.05.01 version). The
system consists of a binary pump, autosampler, thermostated column compartment, DAD detector,
source and QꢀTOF spectrometer. Chromatographic separation of the metabolites (Agilent 1290
UHPLC system) was performed with Agilent RRHD 1.7 ꢁ ꢀC18, 100 x 2.1 mm (Phenomenex USA)
at a constant temperature of 40 °C. The mobile phases consisted of A: H2O/0.1% formic acid and B:
acetonitrile/0.1% formic acid at the flow of 0.45 mL/min with the following gradient: Time 0 min,
B 10%; Time 6 min, B 100%. The DAD Detector stored all the acquired spectra in the 190ꢀ640 nm
range (2 nm spectrum step). The ion source was an Agilent Dual JetStream operating under positive
ionization mode (4000 V), with as nitrogen the desolvating gas (290 °C, 12 L/min, 35 psig). The
fragmentor was set to 375 V, the skimmer to 65 V, and the octrapole RF to 750 V. The
spectrometric data were collected in AutoMS/MS mode in the 100ꢀ1000 mass range, with 3
scans/sec at Collision Energy of 30 V. The TOF operated at 2 GHz.
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
Permeability assay. The PAMPA (Parallel Artificial Membrane Permeation Assay) method46 was
used to study permeability of compounds 8, 16, 31, 36, and 37. In a PAMPA assay drugs permeate
by passive diffusion from the donor compartment (320 µL, Roche Teflon plate), to the acceptor
compartment (280 µl, Millipore MAIPN4550), where the acceptor and the donor compartments are
separated by a phospholipid coated filter (4.5 µL, Millipore MAIPN4550). Donor buffer: 0.05 M
MOPSO at pH 6.5 + 0.5% (w/v) Glycocholic acid. Acceptor buffer: 0.05 M MOPSO at pH 6.5.
Membrane: 10% (w/v) Egg Lecithin in Dodecane + 0.5% (w/v) Cholesterol. Permeation time was
18 h. The permeation constant, Peff [10ꢀ6 x cm/s], and the sample distribution can be retrieved by
UV analytics (SpectraMax Plus 384 Absorbance Microplate Reader, Molecular Devices) of donor
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 43
1
2
3
4
5
6
7
8
9
(tstart), donor (tend) and acceptor (tend) using the pION PAMPA Evolution software. According to this
method, low permeability is expected when Peff < 0.2 x 10ꢀ6 cm/s and Membrane% < 20%, where
Peff represents the effective permeability and Membrane% represents the percentage of compounds
trapped into the membrane. When one or both parameters are above the cutꢀoff then a medium to
high permeability is expected (compound 8: Peff = 3.52 x 10ꢀ6 cm/s and Membrane% = 57%;
compound 16: Peff = 2.93 x 10ꢀ6 cm/s and Membrane% = 41%; compound 31: Peff = 5.49 x 10ꢀ6 cm/s
and Membrane% = 2%; compound 36: Peff = 0.61 x 10ꢀ6 cm/s and Membrane% = 45%; compound
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
37: undetected; compound 7: Peff = 0.00 x 10ꢀ6 cm/s and Membrane% = 99%; compound 10: Peff
=
0.00 x 10ꢀ6 cm/s and Membrane% = 98%).
Equilibrium solubility in aqueous media. An accurately weighted amount of compound (about 1
mg) as dry powder was placed in a glass tube and 330 µL of 0.15 M Phosphate buffer at pH 7.4
were added. Thus, the sample was ultrasonicated for 1 hour, shaken at room temperature for 2 hours
(Branson 1210, Branson Ultrasonics, Danbury, CT) and let stand to equilibrate overnight. The next
day, the pH of the sample was checked and the solution was filtered using a MultiScreenHTSꢀPCF
filter (Millipore, Darmstadt, Germany). Aliquots of the filtrate were used to generate three different
samples (undiluted, diluted as 1:10 and as 1:100) by dilution with the same buffer. Compound
calibration curves were generated from standard 1 mg/mL DMSO solutions diluted in buffer.
Calibration standards and samples were then analyzed by LCꢀUV.MS using an Agilent 1260
Infinity HPLCꢀMS system (Agilent Technologies, Santa Clara, CA) equipped with a DAD detector
and an API single quadruple mass spectrometer (Agilent 6140, Santa Clara, CA). A Kinetex XB 2.6
µm, 2.1 x 50 mm analytical column (Phenomenex, Torrance, CA) operating at 60 °C was used. The
mobile phase consisted of A (water) and B (acetonitrile), both containing 0.1 %/vol formic acid.
The gradient elution was performed with a flow rate of 1.5 mL/min using a generic fast gradient.
After detection with DAD, the eluent entered the electrospray system using the following source
settings: capillary voltage + 3.5 kV, drying gas flow (N2) 13 L/min, nebulizer pressure 60 sig,
ACS Paragon Plus Environment

Page 27 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
drying gas temperature 350 °C. The integral of the pseudo molecular ion intensity, [M+H]+, was
acquired in single reaction monitoring mode (SRM).
6
7
8
9
ASSOCIATED CONTENT
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
Supporting information
Table containing Elemental Analysis data for target compounds 8 and 11ꢀ37. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
For OT: phone, + 39 075 585 5139; fax, +39 075 585 5115; eꢀmail, oriana.tabarrini@unipg.it. For
LG: phone, +39 075 585 5632; fax, +39 075 45646; eꢀmail, laura.goracci@unipg.it. For AL: phone,
+39 049 8272363; fax, +39 049 8272355; eꢀmail, arianna.loregian@unipd.it.
Author contributions
⊥There are two sets of equal contributions: (1) S.M. and G.N. contributed equally to this work; (2)
A.L. and O.T. contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGEMENTS
We thank P. Digard, R. Cusinato, Donatelli, and W. S. Barclay for FluA and FluB viruses; P.
Digard, E. Fodor, and L. Tiley for plasmids. We also thank R. Alvarez, S. Wendelspiess, and V.
Micallef at F. HoffmannꢀLa Roche Ltd. for permeability and solubility assays. This work was
supported by Italian Ministry of Health and Istituto Superiore Sanità, Progetto Finalizzato 2009
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 43
1
2
3
4
5
“Studio e Sviluppo di Nuovi Farmaci Antivirali Contro Infezioni da Virus Influenzale AH1N1” (to
V.C., A.L., G.P., and O.T.), and by ESCMID Research Grant 2013 (to B.M.).
6
7
8
9
ABBREVIATIONS USED
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
ADME, absorption, distribution, metabolism and excretion; CC50, concentration that causes a
decrease of cell viability by 50%; FLAP, Fingerprints for Ligands and Proteins; Flu, influenza
virus; FluA, influenza A virus; FluB, influenza B virus; HEK, human embryonic kidney; HLM,
human liver microsomes; MDCK, MardinꢀDarby canine kidney; PA, polymerase acidic protein;
PAMPA, parallel artificial membrane permeability assay; PB1, polymerase basic protein 1; PB2,
polymerase basic protein 2; PPI, proteinꢀprotein interaction; PRA, plaque reduction assays; RBV,
Ribavirin; RdRP, RNAꢀdependent RNA polymerase.
REFERENCES
1. Ison M. G. Clinical Use of Approved Influenza Antivirals: Therapy and Prophylaxis. Influenza
Other Respir Viruses 2013, 1, 7ꢀ13.
2. Pielak R. M, Chou J. J. Flu Cannel Drug Resistance: a Tale of Two Sites. Protein Cell. 2010, 1,
246ꢀ258.
3. Samson, M.; Pizzorno, A.; Abed, Y.; Boivin, G. Influenza Virus Resistance to Neuraminidase
Inhibitors. Antiviral Res. 2013, 98, 174ꢀ185.
4. Bright, R. A.; Shay, D. K.; Shu, B.; Cox, N. J.; Klimov, A. I. Adamantane Resistance Among
Influenza A Viruses Isolated Early During the 2005–2006 Influenza Season in the United
States. JAMA 2006, 295, 891–894.
5. de Jong, M. D.; Tran, T. T.; Truong, H. K.; Vo, M. H.; Smith, G. J.; Nguyen, V. C.; Bach, V.
C.; Phan, T. Q.; Do, Q. H.; Guan, Y.; Peiris, J. S.; Tran, T. H.; Farrar, J. Oseltamivir Resistance
During Treatment of Influenza A (H5N1) Infection. N. Engl. J. Med. 2005, 353, 2667ꢀ2672.
ACS Paragon Plus Environment

Page 29 of 43
Journal of Medicinal Chemistry
1
2
3
4
5
6. ResaꢀInfante, P.; Jorba, N.; Coloma, R.; Ortin, J. The Influenza Virus RNA Synthesis Machine:
Advances in its Structure and Function. RNA Biol. 2011, 8, 207ꢀ215.
6
7
8
9
7. Loregian, A.; Palù, G. Disruption of ProteinꢀProtein Interactions: Towards New Targets for
Chemotherapy. J. Cell. Physiol. 2005, 204, 750ꢀ762.
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
8. Palù, G.; Loregian, A. Inhibition of Herpesvirus and Influenza Virus Replication by Blocking
Polymerase Subunit Interactions. Antiviral Res. 2013, 99, 318ꢀ327.
9. Loregian, A.; Mercorelli, B.; Nannetti, G.; Compagnin, C.; Palù, G. Antiviral Strategies
Against Influenza Virus: Towards New Therapeutic Approaches. Cell. Mol. Life Sci. 2014, 71,
3659ꢀ3683.
10. Perez, D.R. and Donis, R.O. A 48ꢀAminoꢀacid Region of Influenza A Virus PB1 Protein is
Sufficient for Complex Formation with PA. J Virol. 1995, 69, 6932ꢀ6939
.
11. González S, Zürcher T, Ortín J. Identification of Two Separate Domains in the Influenza Virus
PB1 Protein Involved in the Interaction with the PB2 and PA Subunits: a Model for the Viral
RNA Polymerase Structure. Nucleic Acids Res. 1996, 24, 4456ꢀ4463.
12. Perez, D.R. and Donis, R.O. Functional Analysis of PA Binding by Influenza a Virus PB1:
Effects on Polymerase Activity and Viral Infectivity. J Virol. 2001, 75, 8127ꢀ8136.
13. Obayashi, E.; Yoshida, H.; Kawai, F.; Shibayama, N.; Kawaguchi, A.; Nagata, K.; Tame, J. R.;
Park, S. Y. The Structural Basis for an Essential Subunit Interaction in Influenza Virus RNA
Polymerase. Nature 2008, 454, 1127ꢀ1131.
14. He, X.; Zhou, J.; Bartlam, M.; Zhang, R.; Ma, J.; Lou, Z.; Li, X.; Li, J.; Joachimiak, A.; Zeng,
Z.; Ge, R.; Rao, Z.; Liu, Y. Crystal Structure Of The Polymerase PA(C)ꢀPB1(N) Complex
From An Avian Influenza H5N1 Virus. Nature 2008, 454, 1123ꢀ1126.
15. Sugiyama, K.; Obayashi, E.; Kawaguchi, A.; Suzuki, Y.; Tame, J. R.; Nagata, K.; Park, S. Y.
Structural Insight into the Essential PB1–PB2 Subunit Contact of the Influenza virus RNA
Polymerase. EMBO J. 2009, 28, 1803–1811.
ACS Paragon Plus Environment