Structure-activity relationships in fungal nucleobases transporters as dissected by the inhibitory effects of novel purine analogues

Article abstract of DOI:10.1016/j.ejmech.2018.06.038

We have previously rationally designed, synthesized and tested a number of 3-deazapurine analogues, which inhibit the ubiquitous fungal nucleobase transporter FcyB, through binding in its major substrate binding site, by specifically interacting with Asn163. Here, in an effort to further understand the molecular details of structure-activity relationships in all three major nucleobase transporters of fungi, we extend this study by designing, based on our previous experience, synthesizing and testing further 3-deazapurine analogues. We thus identify seven new compounds with relatively high affinity (19–106 μΜ) for the FcyB binding site. Importantly, four of these compounds can also efficiently inhibit AzgA, a structurally and evolutionary distinct, but functionally similar, purine transporter. Contrastingly, none of the new compounds tested had any effect on the transport activity of the uric acid-xanthine transporter UapA, albeit this being a structural homologue of AzgA. Besides the apparent importance for understanding how nucleobase transporter specificity is determined at the molecular level, our work might constitute a critical step in the design of novel purine-related antifungals.

Full text of DOI:10.1016/j.ejmech.2018.06.038

Accepted Manuscript
Structure-activity relationships in fungal nucleobases transporters as dissected by the
inhibitory effects of novel purine analogues
Efthymios-Spyridon Gavriil, Spyridon Dimitrakis, Georgia Papadaki, Sophia Balaska,
George Lambrinidis, Nikolaos Lougiakis, Panagiotis Marakos, George Diallinas,
Nicole Pouli, Emmanuel Mikros
PII:
S0223-5234(18)30529-4
10.1016/j.ejmech.2018.06.038
EJMECH 10508
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 7 March 2018
Revised Date: 13 June 2018
Accepted Date: 13 June 2018
Please cite this article as: E.-S. Gavriil, S. Dimitrakis, G. Papadaki, S. Balaska, G. Lambrinidis, N.
Lougiakis, P. Marakos, G. Diallinas, N. Pouli, E. Mikros, Structure-activity relationships in fungal
nucleobases transporters as dissected by the inhibitory effects of novel purine analogues, European
Journal of Medicinal Chemistry (2018), doi: 10.1016/j.ejmech.2018.06.038.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Structure-activity relationships in fungal nucleobases
transporters as dissected by the inhibitory effects of novel
purine analogues
a
a
b
Efthymios-Spyridon Gavriil, Spyridon Dimitrakis, Georgia Papadaki,
b
a
a
Sophia Balaska, George Lambrinidis, Nikolaos Lougiakis, Panagiotis
a
,b
a
,a
Marakos, George Diallinas* , Nicole Pouli, Emmanuel Mikros*
Graphical abstract

1
ACCEPTED MANUSCRIPT
Structure-activity relationships in fungal nucleobases
transporters as dissected by the inhibitory effects of novel
purine analogues
a
a
b
Efthymios-Spyridon Gavriil, Spyridon Dimitrakis, Georgia Papadaki,
b
a
a
Sophia Balaska, George Lambrinidis, Nikolaos Lougiakis, Panagiotis
a
,b
a
,a
Marakos, George Diallinas* , Nicole Pouli, Emmanuel Mikros*
a
Division of Pharmaceutical Chemistry, Department of Pharmacy, National and
Kapodistrian University of Athens, Panepistimiopolis, Zografou 15771, Greece.
b
Department of Biology, National and Kapodistrian University of Athens,
Panepistimiopolis, Athens 15784, Greece.
*
Correspondence to:
Emmanuel Mikros, e-mail: mikros@pharm.uoa.gr
George Diallinas, e-mail: diallina@biol.uoa.gr

2
ACCEPTED MANUSCRIPT
Abstract
We have previously rationally designed, synthesized and tested a number of 3-
deazapurine analogues, which inhibit the ubiquitous fungal nucleobase transporter
FcyB, through binding in its major substrate binding site, by specifically interacting
with Asn163. Here, in an effort to further understand the molecular details of
structure-activity relationships in all three major nucleobase transporters of fungi, we
extend this study by designing, based on our previous experience, synthesizing and
testing further 3-deazapurine analogues. We thus identify seven new compounds with
relatively high affinity (19-106 µΜ) for the FcyB binding site. Importantly, four of
these compounds can also efficiently inhibit AzgA, a structurally and evolutionary
distinct, but functionally similar, purine transporter. Contrastingly, none of the new
compounds tested had any effect on the transport activity of the uric acid-xanthine
transporter UapA, albeit this being a structural homologue of AzgA. Besides the
apparent importance for understanding how nucleobase transporter specificity is
determined at the molecular level, our work might constitute a critical step in the
design of novel purine-related antifungals.
Key words: purine isosters / imidazo[4,5-c]pyridine / rational design / Aspergillus
nidulans / purine transporters / Molecular Dynamics

3
ACCEPTED MANUSCRIPT
1
. INTRODUCTION
Early genetic and biochemical studies established the presence of highly specific
nucleobase transporters in fungi. The lack of growth on purines or the use of purine or
pyrimidine toxicity, caused either by an excess of a base (e.g. uracil, uric acid), or by
a cytotoxic analogue (e.g. oxypurinol, allopurinol, 8-azaguanine, 5-fluorouracil, 5-
flurocytosine), provided a powerful tool to select mutants and identify the
corresponding genes.[1,2] Out of work performed mostly in Aspergillus nidulans, but
also in Aspergillus fumigatus, Saccharomyces cerevisiae and Candida albicans, we
now know that there are four specific nucleobase uptake systems that belong to two
structurally and evolutionary distinct protein families.[3] The first, called NAT
(Nucleobase Ascorbate Transporters, also called Nucleobase Cation Symporters 2 or
NCS2) includes two sub-groups, with totally non-overlapping specificities, the
UapA/C-like and the AzgA-like transporters.[4-6] UapA/C transporters are specific
for uric acid and xanthine, whereas AzgA transporters are specific for adenine-
hypoxanthine-guanine. In addition, UapA/C and AzgA transport totally different
purine analogues and drugs, namely oxypurinol and allopurinol versus 8-azaguanine
and 6-mercaptourine, respectively. The second major family is called NCS1
(Nucleobase Cation Symporters 1) and also includes two sub-groups with totally non-
overlapping specificities, the Fcy-like [7,8] and the Fur-like transporters.[9,10] Fcy-
like transporters are specific for adenine-hypoxanthine-guanine-cytosine, whereas
Fur-like transporters are highly specific for either uracil or allantoin, the latter being a
+
purine catabolic metabolite. All these transporter groups function as H - dependent
+
secondary transporters or H symporters. All filamentous fungi, including the major
pathogens, A. fumigatus and A. flavus, possess all four subgroups of the two families,
3
while several yeasts, including S. cerevisiae, lack the UapA/C subgroup. The
distinction of nucleobase transporters in two major families, initially based simply on
functional assays, was confirmed by structural approaches, including x-ray
crystallography and relative modeling approaches of bacterial homologues,[11-15]
but also of the UapA transporter of A. nidulans.[16] The sub-group distinction of the
NAT and NCS1 families is supported by prominent differences in their substrate
specificities and the presence of characteristic amino acid sequence motifs.
What seems to be an apparent paradox is that while UapA and AzgA are
structurally similar (NAT family), they transport entirely different substrates and
drugs, while FcyB which belong to a different major family (NCS1) has very similar
specificity profile and similar binding affinities for purines with AzgA. The only
rigorous way to functionally distinguish AzgA from FcyB is via their relative capacity
to transport cytosine or specific purine or pyrimidine analogues. These observations
suggest that the architecture of the binding site of NAT and NCS1 transporter is
shaped via both divergent and convergent evolution events. While functional details
for substrate binding and transport have been elucidated by genetic and in vivo
transport kinetic approaches for several fungal members of the NCS1 family (see
references above), direct structural information comes only from the bacterial benzyl-
hydantoin Mhp1 transporter.[12-14] Mhp1, but also all studied fungal NCS1
homologues modeled on the Mh1 structure, are characterized by a core of ten
transmembrane helices divided in two symmetric sets oppositely orientated with
respect to the membrane, termed the 5-helix inverted repeat (5HIR) motif. There are
additionally two C-terminal transmembrane helices (TMH11 and TMH12), which
apparently do not participate in transport activity, and their role is unknown.
Importantly, as the structure of Mhp1 has been previously solved in three different

4
ACCEPTED MANUSCRIPT
conformations, this has provided insights into the molecular basis of the alternating
access mechanism of transport. Briefly, sodium ions bind and favor the formation of
an outward-facing state and the consequent substrate binding in a cavity formed
between a “bundle” (TMHs 1, 2, 6 and 7) and a “hash” motif (TMHs 3, 4, 8 and 9).
Upon substrate binding TMH10 bends over the substrate in the occluded state
triggering the rotation of the one motif relative to the other towards the inward-facing
conformation, accompanied by a bend of TMH5 which opens the cavity further to the
intracellular and enables substrate transport.
In A. nidulans we have developed tools to study rigorously structure-function
relationships in nucleobase transporters. These include the construction of strains
expressing a single nucleobase transporter in a genetically ‘clean’ background lacking
all functionally similar transporters due to multiple null mutations, and efficient
cellular and functional assays measuring transporter expression, turnover and
activity.[17, 18] Additionally, all nucleobase transporters of A, nidulans have been
extensively analyzed via classical and reverse genetics and modeling using available
crystal structures of bacterial homologues.[18] Thus, we have defined, at the
molecular level, transporter-substrate interactions and identified the major elements of
substrate binding sites. Furthermore, we have identified elements located outside the
major binding site that affect nucleobase transporter specificity, collectively named
gating elements.[17-19] Most importantly, we obtained direct evidence of structure-
function relationships in UapA by analyzing the crystal structure of a genetically
stabilized version of UapA in complex with xanthine.[19]The crystal structure
validated conclusions drawn by genetic and biochemical approaches and further
showed dimerization is crtical for UapA function and specificity.[16-18]
In a different but complementary approach, we recently developed chemical
biology methodologies to better understand how specificity is determined in different
nucleobase transporters.[20] In particular, we used a virtual screening and semi-
rational approaches to identify compounds that might interact with the substrate
binding site of a specific nucleobase transporter, and then test whether positive-
scoring compounds can also be recognized by a functionally similar transporter. More
specifically, we looked for FcyB ligands and then tested whether they interact with
AzgA (i.e. a functional analogue) or UapA (i.e. a structural homologue). We thus
identified a number of 3-deazapurines substituted in the corresponding 6- and 9-
positions of the original purine scaffold, which proved highly specific solely for
FcyB.[20] One of these compounds proved to be moderately toxic for A. nidulans
growth. In the present work, we exploited findings of our original approach to identify
further purine analogues that might act differentially on the relative A. nidulans
transporters. We identified seven new compounds recognized with high affinity by the
FcyB, and among them, four were also recognized by AzgA, but none by UapA. This
work complements in our understanding of the transporter-substrate interactions and
could also be considered as an important step towards developing purine-related
analogues as potentially novel highly targeted antifungals.
2
. RESULTS
2
.1. Rationale for 3-deazapurine analogue design. In order to design new chemical
probes allowing better understanding the FcyB function, a structural model for the
protein was required. In this study we have de novo constructed an FcyB structure by
homology modeling, using as template the most recent Mhp1 occluded crystal
structure (pdb 4d1d), which is in the outward conformation in complex with an
hydantoin derivative acting as inhibitor.[13] The new model displays only few

5
ACCEPTED MANUSCRIPT
differences within the binding cavity of the transporter when compared to a
previously constructed model based on the occluded Mhp1 topology (pdb 2jlo).[7]
The integrity of the transmembrane helices is improved when compared with the
previous model. The major difference is related to the first part of TMH10 between
Glu401 and Tyr409, which is tilted moving away from the cavity forming the outward
conformation. For exploring possibilities of modifying further the 3-deazapurine
scaffold, we were guided by the structure-activity relationships of the recently
reported series of 1,4-substituted-3-deazapurine analogues,[20] which can be
summarized as follows. The most potent analogue has been found to be the 4-
benzylamine substituted derivative 1 (Figure 1A), interacting with Glu397 on TMH9-
1
0 at the edge of the transporter outward cavity along with the main interaction
between N3 and Asn163.
Figure 1. A. Structures of Compounds 1-4. B. Global minimum energy structure of
compound 4 in complex with FcyB model. In grey the hydrophobic surface of the
protein binding pocket near Pro353 forming a cavity that is targeted by the
synthesized compounds.
Additionally, compound 2 bearing at position 4- a methylpiperazino- group,
containing a quaternary nitrogen positively charged at the physiological pH
conditions, has proved to possess interesting activity as well. The corresponding
monosubstituted parent analogues 3 and 4 respectively (Figure 1A), albeit not highly
potent, maintain a certain degree of inhibitory activity against both AzgA and FcyB.
In order to further investigate the different substitutions of this scaffold, we examined
the possibilities for an additional interaction with an important residue in the vicinity,
namely Asn354 by introducing at position 2- an enolic OH or SH group, maintaining
the optimum substitution pattern at position 4-. With the aim of facilitating the
interpretation of the resulting data, we have decided to preserve only the 4-substituent
of the previously studied compounds, and thus deal with 2,4-disubstituted analogues.
This would also provide the possibility to investigate the effect of both imidazole
nitrogen atoms in the molecular level. Furthermore, no substitution at position 1-
together with the enolic OH at position 2- which is expected to be in its carbonyl form

6
ACCEPTED MANUSCRIPT
will generate correspondingly a hydrogen bond (HB) -donor and -acceptor that could
better match the targeted interaction with amide side chains of Asn163 or Asn354.
Finally, in order to further investigate substitution at position 2- and more specifically
the hydrophobic cavity existing around Pro353 (Figure 1B), a number of analogues
bearing both a hydrophobic surface and a functional group that could interact with
Asn354 have also been designed and synthesized.
2
.2 Chemical synthesis of novel 3-deazapurine analogues The target compounds
were prepared using 4-amino-2-chloropyridine (5, Scheme 1), which was nitrated to
result into a mixture of the nitroderivatives 6 and 7.[21] The appropriate 3-
nitroderivative 6 underwent nucleophilic substitution using N-methylpiperazine or
benzylamine and provided the intermediate nitropyridines 8 and 9.[20] These
derivatives were first reduced and the resulting diamines were cyclized using CDI,
tCDI or selected aromatic aldehydes to result in the desired 2-substituted
imidazopyridines 10a-f and 11a-f. Concerning the benzylamino-derivatives 8, the
cyclization reaction provided the isomeric imidazopyridines 12a-f as well (Scheme
1
). Additionally, during the preparation of compounds 10d and 10f, a second product
was also isolated, that identified to be 13 and 14 respectively (Figure 2). These
compounds should have derived from the concomitant formation of arylimines on
both aminogroups of the intermediate pyridodiamines, followed by ring-closure.
Scheme 1 Reagents and conditions: a) HNO3 (f.), H SO (c.), 90 ˚C; b) N-
2
4
methylpiperazine (for 8), or benzylamine (for 9), EtOH, reflux; c) (1) H , Pd/C,
2
EtOH, 35psi, r.t.; (2) CDI or tCDI, THF, reflux, or R-CHO, toluene, montmorillonite
K-10, reflux.

7
ACCEPTED MANUSCRIPT
N
N
N
N
HO
N
N
N
N
N
N
N
OH
N
1
3
14
Figure 2. Structure of compounds 13-14
.3 Evaluation of novel 3-deazapurine analogues as inhibitors of purine
2
transporters. The new 3-deazapurine analogues were tested as competitive inhibitors
3
3
of FcyB- or AzgA-mediated H-radiolabelled adenine uptake, or UapA-mediated H-
radiolabelled xanthine uptake. In all cases, assays were performed, as already
described,[22] in strains expressing the specific transporter studied in the absence of
other transporters of similar specificity.[20] The competition results for FcyB and
AzgA are summarized in Figure 3, along with the previously reported compounds 3
and 4, as well as, unlabeled adenine or hypoxanthine (physiological substrates of
FcyB and AzgA), as controls.
3
Figure 3. Competition of FcyB- (A) or AzgA- (B) mediated [ H]-adenine uptake by
1
000-fold excess (0.5 mM) of unlabeled 3-deazapurine analogues.
Compounds 12a-f, 13 and 14 did not show any inhibitory activity. Among the
3
designed derivatives, 10a, 10d, 10e, 11d, 11e and 11f reduced FcyB-mediated H-
adenine accumulation rate to <10%, compared to the control (absence of any
inhibitor). The compounds 10c, 11d, 11e and 11f also showed a relative significant

8
ACCEPTED MANUSCRIPT
3
inhibition of AzgA-mediated H-adenine accumulation rate, which however was
lower than the one observed with FcyB. None of the compounds tested showed any
inhibitory activity on UapA-mediated transport ox xanthine (data not shown).
2
.4 Kinetic characterization of competing analogues Compounds that proved to be
active as inhibitors of FcyB were further evaluated by estimating their K , using IC50
i
measurements. Results are shown in Table 1. All compounds exhibited values lower
than 110 µΜ for FcyB, with the methylpiperazine analogue 10a being the most potent
followed by the benzylamine analogue 11e, exhibiting K of 19 and 42 µM
i
respectively. As compound 11d showed the higher inhibitory activity against AzgA
the corresponding K value was also determined and found to be 58 µΜ.
i
Table 1. 3-deazapurine analogue binding profile of FcyB
Compound
K (µM)
i
1
1
1
1
1
1
1
1
1
2
0a
0c
0d
0e
1a
1d
1f
19 ± 3
106 ± 12
49 ± 10
100 ± 5
74 ± 6
82 ± 8
72 ± 4
42 ± 6
5 ± 1
1e
38 ± 6
7 ± 1
Adenine
Hypoxanthine
11 ± 1
Results are averages of at least three independent experiments with three replicates for
each concentration point. SD was < 10%.
2
.5 Molecular Modeling FcyB-ligand interactions In order to further develop a
structure-activity relationship model and gain insight in FcyB-ligand interactions, MD
and docking calculations have been carried out. First, we present the models derived
by docking calculations for the most active compounds namely compounds 10a and
1
1e. Each one of these two compounds contains different substitutions at positions 4-
and 2- thus exemplifying the basic characteristics for the protein-ligand interactions as
derived by theoretical calculations. The synthesized analogues were submitted to
docking calculations using the Induced Fit Docking algorithm as implemented in
Maestro 10 (Schrodinger Inc). The most active compound the piperazine-substituted
1
0a was considered as positively charged, since the pKa value of N-methyl-
piperazino group was calculated to be 8.6 using Jaguar Software. The resulted low
energy structures show that the ligand was stabilized inside the binding pocket mainly
in a similar position as described previously.[20] Upon binding to FcyB, the
protonated tertiary nitrogen forms salt bridge with Glu397. Additionally, two other
hydrogen bonds are formed between N3H and Asn163 amide carbonyl group, while
C2=O interacts with NH2 of Asn354 side chain. Finally, an extra H-bond seems to be
possible between N1H and Ser261. Trp159 shows π-π stacking interactions with 3-
deaza aromatic ring. (Figure 4A, C). In the case of compound 11e, the bulky 2-[(2,5-
dimethoxy)phenyl] group forms the π-π stacking interactions with Trp159 while two
hydrogen bonds are formed one between Glu397 and the N4H and the second
between Ser85 side chain and the methoxy group of the 2-substituent (Figure 4B, D).

9
ACCEPTED MANUSCRIPT
Figure 4: Low energy binding mode of compounds 10a (A) and 11e (B) in complex
with FcyB as resulted by docking calculations. C and D detailed presentation of the
interactions in the binding cavity for 10a and 11e (in green) correspondingly.
Hydrogen bonds are depicted with red dashed lines.
2
.6 Molecular Dynamics of FcyB in complex with analogue 10a. To further address
aspects of molecular recognition and dynamic behavior, a study of the protein-ligand
interactions has been undertaken for FcyB in complex with compound 10a using
molecular dynamics. Initially the ligand was positioned in a location similar to the
corresponding to the hydantoin derivative in the parent Mhp1 crystal structure used
for template (pdb: 4d1d) and a 90 ns simulation was carried out. The MD trajectory
shows that the protein remains relatively stable as depicted by the small variations of
the overall Ca RMSD from the 30th ns till the end of the simulation (Figure S1A).
The residue RMS fluctuations ranged from 0.8 Å for rigid secondary structural
elements to 6 Å for mobile loops (Figure S1B). The most important conformational
change of the protein seems to be the reorientation of TMH10, taking place mainly till

1
0
ACCEPTED MANUSCRIPT
th
the 30 ns. This reorientation of TMH10 displaces the Glu401 of ~4 Å and all
residues 386-400 in the loop between TMH9-TMH10 (Figure 5A).
Figure 5. Molecular Dynamic Analysis of Fcyb in complex with compound 10a.
A. Superposition of the binding site of FcyB in complex with compound 10a: Initial
pose of MD simulation (t=0 ns, yellow) and final pose of MD simulation (t=90ns,
green). B. Superposition of starting (cyan) and ending (red) position of TMH10 after
9
0ns of simulation time. C. Timeline representation of interactions of compound 10a
during the MD simulation. The panel shows which residue interacts with the ligand in
each trajectory frame. Dark shades of orange indicate more than one specific contact.
The ligand 10a remains in the binding cavity although during of the simulation is
displaced by ~3 Å towards the protein surface (Figure 5B). During the initial part till
th
the 40 ns the ligand forms HB mainly with Asn163 and Asn354 while it displays
weaker interactions with Tyr262 and Ser85. It is then relocated losing contact with
Asn354 and Tyr262. During the whole simulation the ligand forms a pi-pi interaction
with Trp159 and till the end it remains between Asn163, Trp159 and Ile89 (Figure
5
C). It is interesting that Trp259 remains parallel to the ligand however interactions
are rare specifically between 0- 30 ns then the ligand moves away and there is no
interaction after. The charged quaternary piperazine nitrogen initially forms contacts
with Glu397 and Glu401 (the later through a water molecule). However its
displacement and the simultaneous shift of TMH10 reorients the residues’ side chains

1
1
ACCEPTED MANUSCRIPT
weakening and finally eliminating those interactions, leaving the piperazine moiety
mainly surrounded by water molecules (data not shown).
3
. DISCUSSION
+
The A. nidulans purine/H symporters FcyB and AzgA belong to evolutionary and
structurally distinct transporter families (NCS1 and NAT/NCS2, respectively), but
display very similar, high-affinity, binding for adenine, hypoxanthine and guanine.
However, their substrate binding sites should be significantly different, as revealed by
their binding affinities for cytosine (recognized only by FcyB) or other, non-
physiological substrates, such as the drugs 5-fluorocytosine (FcyB) or 8-azaguanine
(AzgA). We have previously reported a model of FcyB based on the first crystal
structure of the bacterial homologue Mhp1.[7] In the same report we identified, via
substrate docking approaches and rational mutational analysis, residues essential for
the function and specificity that seem to define the substrate-binding site of FcyB.
Furthermore, we have subsequently shown that 3-deazaadenine derivatives bearing
substitutions on positions 1 and 4, inhibit the transport of physiological substrates.[20]
Here we present a refined model of FcyB structure built upon homology threading to
the updated, outward-facing, crystal structure of Mhp1. This new FcyB model shows
that the side chains of all five residues that have been shown previously to be critical
for transport activity, namely Ser85 (TMH1), Trp159, Asn163 (TMH3), Trp259
(TMH6) and Asn354 (TMH8), protrude to the substrate-binding cavity of the
transporter. Using the improved FcyB model, we evaluated, as inhibitors of FcyB and
two other A. nidulans purine transporters, novel 3-deazaadenine analogues, which
were rationally designed based on our previous relative experience.
The synthesized compounds tested in this work expand the study on FcyB-ligand
interactions. They firstly show that substitution at both positions 2 and 4 can
antagonize adenine uptake, with similar inhibitory activities to some of the previously
tested 3-deazaadenine derivatives. The introduction of a carbonyl group in position 3
of 3-deazaadenine enhances dramatically binding to FcyB, since compounds 10a and
1
1a exhibit 10- and 5-fold higher inhibitory activity when compared to the parent
molecules 4 and 3. This suggests that the carbonyl group participates actively in
protein-ligand interactions, confirming the model proposed in Figure 4A, C, which
shows that the major interaction of FcyB with the purine ring occurs via H-bonding
with residues Asn163 and Asn354. A very similar pattern has been observed in the
crystal structure of hydantoin bound to Mhp1 forming H-bonds with the
corresponding Glu121 and Asn318.[13] This is also in agreement with the observation
that conversion of carbonyl oxygen to sulfur in the thiourea analogues 10b and 11b
decreases the inhibitory activity, as the corresponding interactions formed by sulfur
are expected to be weaker. The carbonyl substitution is also advantageous when
comparing derivative 10a to 2, both bearing a 4-methylpiperazine substituent at
position 4. This trend is inversed in the case of the 11a and 1 pair bearing a
benzylamino substitution at the same position. In this case the carbonyl substitution is
detrimental, decreasing the inhibitory activity ten-fold. According to the models
derived by docking calculations this is probably due to the fact that in 1 the
contribution for the protein-ligand interaction stems from the bidentate H-bonding
formation between the NH at position 4 and N3 with the Asn163 amide, while this not
the case in 11a where interactions involving C2=O and N3H with Asn163 are
apparently weaker. Other substitutions than the OH at position 2 can still produce
interesting activities with most surprising the 2,5-dimethoxyphenyl substitution that
proved to be the most active derivative of the 11a-f series. Compound 11e seems to be

1
2
ACCEPTED MANUSCRIPT
displaced in the outer part of the binding cavity in such orientation that can form
appropriate hydrophobic interactions via the 2,5-dimethoxyphenyl moiety and H-
bonding between all different nitrogen atoms with surrounding residue side chains
(Figure 4B, D) resulting to an increased activity.
A critical point of our present work, but also of our previous relative work, concerns
the specificity of the compounds tested against FcyB. All compounds exhibiting
important inhibitory activity display significantly increased specificity for FcyB,
compared to AzgA (Figure 3). Although AzgA and FcyB fold in completely different
structures, they both transport adenine, hypoxanthine or guanine. The major
interaction in the AzgA cavity is between the substrate and two carboxylates from
Glu339 and Asp394.[6] In this case, interactions are completely different and
specificity is driven by a complex multiplicity of interactions. It is noticeable that the
parent compounds 4 and 3 do not show any significant inhibitory activity to either
AzgA or FcyB, however the urea compounds 10a and 11a do exhibit significant
binding specificity. Interestingly substitution at position 2 by a 4-NO -phenyl moiety
2
abolishes specificity, compound 10c being one of the most actives against AzgA
along with the 2-(2-OH-phenyl)-substituted 11d. Those compounds form the basis for
further development of new optimized structures targeting AzgA. Although from the
point of view of basic science, the distinct FcyB selectivity over AzgA is important as
it shows that highly specific transporter drugs can indeed be rationally designed, for
the purpose of a therapeutic antifungal activity, the discovery or design of dual
inhibitors acting on several fungal-specific transporters might prove even more
valuable. In this direction, analogues 10c or 11d (see Figure 3) seem as promising
starting compounds to inhibit both FcyB and AzgA fungal nucleobase transporters.
Solute transport is associated with major changes in the structure of
transporters. High-resolution structures of the Mhp1 transporter suggest that during
the transport cycle the transition from the outward-open to occluded and then to the
inward-open conformation, a semi-rigid body rotation occurs between hash and
bundle motifs while TMH10 bends as an extracellular gate to close the binding cavity,
accompanied by a bending of TMH5 operating as an intracellular gate. Upon binding
of the substrate TMH10 twists over the substrate closing the extracellular gate and
occluding it in the binding cavity. The importance of the TMH10 reorientation is also
shown in the well-studied 5HIR transporter LeuT, where the extracellular gate is
closed through the formation of a salt-bridge between Arg30 (TMH1b) and Asp404
(TMH10) along with other interactions between the two helices. The importance of
TMH10 in transport has also been shown in the case of FurD, where site-specific
mutations in TMH10 result to enlarged specificity.[10] In the comlplex of FcyB with
the most active compound of these series, namely 10a, MD simulations suggested that
the ligand is stabilized in a position different than the corresponding one in the case of
hydantoin-Mhp1 complex. After 90 ns of simulation the ligand is displaced ~3 Å
towards the outer part of the cavity. In this position, 10a is stabilized by forming HBs
with Asn163 and Ser85 as well as pi-pi stacking with Trp159, while the “rest” of the
analogue is mainly solvated by a sphere of water molecules. Although the starting
structure was constructed upon the outward-facing conformation of Mhp1, the new
position of 10a in the cavity tilts the TMH10 further away. This dynamic behavior
confirms that this compound acts as a competitive inhibitor of transport. In that sense
scaffolds based on 10a are interesting not only as chemical biology tools for the study
of similar purine transporters, but also can be considered for further development of
specific transporter-targeted antifungal drugs.

1
3
ACCEPTED MANUSCRIPT
In conclusion, we have identified seven new compounds with relatively high affinity
for the FcyB binding site. Importantly three of these compounds can also efficiently
inhibit AzgA, a structurally and evolutionary distinct, but functionally similar, fungal
purine transporter. None of the new compounds had any effect on the transport
activity of the uric acid-xanthine transporter UapA, albeit this being a structural
homologue of AzgA. Besides the apparent importance for understanding how
nucleobase transporter specificity is determined at the molecular level, our work
might constitute a critical step in the design of novel purine-related antifungals.
4
. METHODS
4
.1 Chemistry. Melting points were determined on a Büchi apparatus and are
1
uncorrected. H NMR spectra and 2D spectra were recorded on a Bruker Avance III
6
1
3
00 or a Bruker Avance DRX 400 instrument, whereas C NMR spectra were
recorded on a Bruker Avance III 600 or a Bruker AC 200 spectrometer in deuterated
1
13
solvents and were referenced to TMS (δ scale). The signals of H and C spectra
1
1
were unambiguously assigned by using 2D NMR techniques: H H COSY, NOESY,
HMQC, and HMBC. Mass spectra were recorded with a LTQ Orbitrap Discovery
instrument, possessing an Ionmax ionization source. Flash chromatography was
performed on Merck silica gel 60 (0.040–0.063 mm). Analytical thin layer
chromatography (TLC) was carried out on precoated (0.25 mm) Merck silica gel F-
2
4
4
54 plates.
.1.1 Synthesis and characterization of compounds 10a-f, 11a-f, 12a-f
.1.1.1. 4-(4-Methylpiperazin-1-yl)-1,3-dihydroimidazo[4,5-c]pyridin-2-one (10a)
A solution of the nitro derivative 8 (240 mg, 1.0 mmol) in absolute ethanol (60 mL)
was hydrogenated in the presence of 10% Pd/C (90 mg) at 35 psi for 4h. The solution
was filtered through a celite pad to remove the catalyst and the filtrate was evaporated
to dryness. To a solution of the resulting diamine (210 mg, 1.0 mmol) in
tetrahydrofuran (10 mL), 1,1’-carbonyldiimidazole was added (325 mg, 2.0 mmol)
and the mixture was heated at reflux for 24h. Then, the organic solvent was
evaporated under vacuo and the residue was purified using silica gel column
chromatography (dichloromethane/methanol: 8/2). White solid, 230 mg (98%), m.p.
o
1
>
250 C (methanol/ether). H NMR (400 MHz, DMSO-d ) δ 10.97 (s, 1H, NH), 10.75
6
(
s, 1H, NH), 7.75 (d, J = 5.3 Hz, 1H, H-6), 6.60 (d, J = 5.3 Hz, 1H, H-7), 3.20 (m, 4H,
1
3
piperazine H-2,6), 2.45 (m, 4H, piperazine H-3,5), 2.21 (s, 3H, NCH ). C NMR (151
3
MHz, DMSO-d ) δ 154.70 (C-2), 145.26 (C-4), 139.46 (C-6), 136.49 (C-7a), 114.93
6
(
C-3a), 99.55 (C-7), 54.60 (piperazine C-3,5), 47.35 piperazine (C-2,6), 45.84
-
(NCH ). HRMS-ESI(-) (C H N O) [M-H] m/z calcd 232.1193, found 232.1198.
3
11 14
5
4
.1.1.2. 4-(4-Methylpiperazin-1-yl)-1,3-dihydroimidazo[4,5-c]pyridine-2-thione
(
10b)
This compound was synthesized using a procedure analogous to that of 10a, using
,1’-thiocarbonyldiimidazole. The residue was purified using silica gel column
chromatography (dichloromethane/methanol: 9/1). Beige solid, yield 68%, m.p.
1
o
1
>
250 C (methanol/ether). H NMR (400 MHz, DMSO-d ) δ 13.25-12.45 (brs, 2H,
6
NH), 7.87 (d, J = 5.4 Hz, 1H, H-6), 6.75 (d, J = 5.4 Hz, 1H, H-7), 3.33 (m, 4H,
1
3
piperazine H-2,6), 2.50 (m, 4H, piperazine H-3,5), 2.25 (s, 3H, NCH ). C NMR (151
3
MHz, DMSO-d ) δ 168.21 (C-2), 145.87 (C-4), 140.60 (C-6), 139.01 (C-7a), 118.19
6
(
C-3a), 99.29 (C-7), 54.49 (piperazine C-3,5), 47.36 (piperazine C-2,6), 45.75
-
(NCH ). HRMS-ESI(-) (C H N S) [M-H] m/z calcd 248.0964, found 248.0969.
3
11 14
5
4
.1.1.3. 4-(4-Methylpiperazin-1-yl)-2-(4-nitrophenyl)-1H-imidazo[4,5-c]pyridine
(10c)

1
4
ACCEPTED MANUSCRIPT
To a solution of 8 (210 mg, 1.0 mmol) in toluene (10 mL), was added montmorillonite
K-10 (300 mg) followed by 4-nitrobenzaldehyde (166 mg, 1.1 mmol) and the mixture
was heated at reflux for 24h. Then, the organic solvent was evaporated under vacuo
and the residue was purified using silica gel column chromatography
(
dichloromethane/methanol: 9/1) to provide 10c as a brown solid, 200mg (59%),
o
1
m.p.>250 C (ethanol/ether). H NMR (400 MHz, DMSO-d ) δ 13.64-13.34 (brs, 1H,
6
NH), 8.40 (d, J = 8.9 Hz, 2H, H-3’,5’), 8.35 (d, J = 8.9 Hz, 2H, H-2’,6’), 7.84 (d, J =
5
.6 Hz, 1H, H-6), 6.91 (d, J = 5.6 Hz, 1H, H-7), 4.16 (m, 4H, piperazine H-2,6), 2.54
1
3
(m, 4H, piperazine H-3,5), 2.28 (s, 3H, NCH ). C NMR (151 MHz, DMSO-d ) δ
3
6
1
1
51.12 (C-4), 147.69 (C-4’), 145.91 (C-2), 141.26 (C-7a), 140.73 (C-6), 135.51 (C-
’), 128.87 (C-3a), 127.18 (C-2’,6’), 124.34 (C-3’,5’), 98.52 (C-7), 54.61 (piperazine
-
C-3,5), 45.64 (piperazine C-2,6), 45.51 (NCH ). HRMS-ESI(-) (C H N O ) [M-H]
3
17 17
6
2
m/z calcd 337.1408, found 337.1412.
.1.1.4. 2-(2-Hydroxyphenyl)-4-(4-methylpiperazin-1-yl)-1H-imidazo[4,5-
c]pyridine (10d)
This compound was synthesized using a procedure analogous to that of 10c. The
residue was purified using silica gel column chromatography
dichloromethane/methanol: 9/1) and upon successive preparative thin layer
4
(
chromatography (chloroform/methanol: 9/1), compound 10d was isolated as a beige
o
solid, yield 47%, m.p. 224-225 C (ether), together with a small amount of 13. Data for
1
1
0d: H NMR (600 MHz, DMSO-d ) δ 13.47 – 13.02 (brs, 1H, NH), 12.47 – 12.03
6
(brs, 1H, OH), 8.05 (d, J = 7.5 Hz, 1H, H-6’), 7.87 (d, J = 5.2 Hz, 1H, H-6), 7.37 (t, J
=
7.5 Hz, 1H, H-4’), 7.02 (m, 2H, H-3’,5’), 6.99 (d, J = 5.2 Hz, 1H, H-7), 4.00 (m,
1
3
4
H, piperazine H-2,6), 2.57 (m, 4H, piperazine H-3,5), 2.30 (s, 3H, NCH ). C NMR
3
(
151 MHz, DMSO-d ) δ 157.16 (C-2’), 150.39 (C-4), 148.81 (C-2), 140.21 (C-6),
6
1
3
4
3
39.93 (C-7a), 131.54 (C-4’), 126.59 (C-6’), 126.10 (C-3a), 119.35 (C-5’), 116.99 (C-
’), 112.98 (C-1’), 99.14 (C-7), 54.64 (piperazine C-3,5), 46.02 (piperazine C-2,6),
-
5.86 (NCH ). HRMS-ESI(-) (C H N O) [M-H] m/z calcd 308.1506, found
3
17 18
5
08.1510.
Data for 1-[(2-hydroxyphenyl)methyl]-2-(2-hydroxyphenyl)-4-(4-methylpiperazin-1-
o
yl)-1H-imidazo[4,5-c]pyridine (13): beige solid, yield 4%, m.p. 165-166 C
1
(
(
(
methanol). H NMR (600 MHz, DMSO-d ) δ 10.54-10.10 (brs, 1H, OH), 10.05-9.65
6
brs, 1H, OH), 7.76 (d, J = 5.7 Hz, 1H, H-6), 7.33 (d, J = 7.5 Hz, 2H, H-3’,5’) 7.01
m, 2H, H-6’,4’’), 6.89 (m, 1H, H-4’), 6.76 (m, 1H, H-3’’), 6.72 (d, J = 5.7 Hz, 1H,
H-7), 6.58 (t, J = 7.4 Hz, 1H, H-5’’), 6.36 (d, J = 7.4 Hz, 1H, H-6’’), 5.23 (s, 2H,
CH ), 4.05 (m, 4H, piperazine H-2,6), 2.44 (m, 4H, piperazine H-3,5), 2.22 (s, 3H,
2
13
NCH ). C NMR (151 MHz, DMSO-d ) δ 155.85 (C-2’), 154.58 (C-2’’), 150.96 (C-
3
6
4
), 149.14 (C-2), 141.14 (C-7a), 139.56 (C-6), 131.35 (C-3’), 128.35 (C-4’’), 127.44
(C-3a), 127.02 (C-6’’), 126.80 (C-1’), 122.45 (C-1’’), 119.13 (C-4’), 118.80 (C-5’’),
1
4
17.11 (C-5’), 115.90 (C-6’), 114.88 (C-3’’), 97.93 (C-7), 54.59 (piperazine C-3,5),
13
5.64 (NCH ), 45.51 (piperazine C-2,6), 42.83 (CH ). The C NMR chemical shifts
3
2
for this compound were extracted from the corresponding HMBC spectrum. HRMS-
+
ESI(+) (C H N O ) [M+H] m/z calcd 416.2081, found 416.2080.
2
4
26
5
2
4
.1.1.5.
2-(2,5-Dimethoxyphenyl)-4-(4-methylpiperazin-1-yl)-1H-imidazo[4,5-
c]pyridine (10e)
This compound was synthesized using a procedure analogous to that of 10c. The
residue
was
purified
using
silica
gel
column
chromatography
(
4
dichloromethane/methanol: 9/1) and the product was obtained as a yellow solid, yield
o
1
2%, m.p. 180-181 C (n-hexane). H NMR (600 MHz, DMSO-d ) δ 12.34-12.20 (brs,
6
1
H, NH), 7.79 (d, J = 5.5 Hz, 1H, H-6), 7.76 (d, J = 3.2 Hz, 1H, H-6’), 7.15 (d, J =

1
5
ACCEPTED MANUSCRIPT
9
.0 Hz, 1H, H-3’), 7.05 (dd, J = 9.0, 3.2 Hz, 1H, H-4’), 6.97 (d, J = 5.5 Hz, 1H, H-7),
4
.14 (m, 4H, piperazine H-2,6), 3.94 (s, 3H, 2’-OCH ), 3.78 (s, 3H, 5’-OCH ), 2.57
3
3
1
3
(m, 4H, piperazine H-3,5), 2.28 (s, 3H, NCH ). C NMR (151 MHz, DMSO-d ) δ
3
6
1
6
53.23 (C-5’), 151.06 (C-2’), 150.78 (C-4), 145.70 (C-2), 140.61 (C-7a), 139.48 (C-
), 127.80 (C-3a), 118.35 (C-1’), 116.27 (C-4’), 114.57 (C-6’), 113.36 (C-3’), 99.48
(
C-7), 56.18 (2’-OCH ), 55.62 (5’-OCH ), 54.53 (piperazine C-3,5), 45.52 (NCH -
3 3 3
-
piperazine C-2,6). HRMS-ESI(-) (C H N O ) [M-H] m/z calcd 352.1768, found
1
9
22
5
2
3
52.1770.
4
.1.1.6.
4-(4-Methylpiperazin-1-yl)-2-(pyridin-4-yl)-1H-imidazo[4,5-c]pyridine
(
10f)
This compound was synthesized using a procedure analogous to that of 10c. The
residue was purified using silica gel column chromatography
dichloromethane/methanol: 9/2) and upon successive preparative thin layer
chromatography (chloroform/methanol: 9/2) compound 10f was isolated as yellow oil
(
1
(
yield 28%), together with a small amount of 14. Data for 10f: H NMR (600 MHz,
DMSO-d ) δ 13.80-13.10 (brs, 1H, NH), 8.74 (d, J = 5.7 Hz, 2H, H-2’,6’), 8.04 (d, J
6
=
5.7 Hz, 2H, H-3’,5’), 7.84 (d, J = 5.6 Hz, 1H, H-6), 6.90 (d, J = 5.6 Hz, 1H, H-7),
4
.14 (m, 4H, piperazine H-2,6), 2.46 (m, 4H, piperazine H-3,5), 2.23 (s, 3H, NCH3).
1
3
C NMR (151 MHz, DMSO-d ) δ 151.22 (C-4), 150.49 (C-2’,6’), 145.66 (C-2),
6
1
41.09 (C-7a), 140.66 (C-6), 136.62 (C-4’), 128.61 (C-3a), 120.13 (C-3’,5’), 98.50
(
C-7), 54.78 (piperazine C-3,5), 45.90 (NCH ), 45.69 (piperazine C-2,6). HRMS-
3
-
ESI(-) (C H N ) [M-H] m/z calcd 293.1509, found 293.1512
1
6
17
6
Data
for
4-(4-methylpiperazin-1-yl)-1-[(pyridin-4-yl)methyl]-2-(4-pyridyl)-1H-
1
imidazo[4,5-c]pyridine (14): Yellow oil, Yield 4%. H NMR (600 MHz, DMSO-d ) δ
6
8
.70 (d, J = 6.0 Hz, 1H, H-2’,6’), 8.47 (d, J = 5.8 Hz, 1H, H-2’’,6’’), 7.86 (d, J = 5.7
Hz, 1H, H-6), 7.66 (d, J = 6.0 Hz, 1H, H-3’,5’), 6.97 (d, J = 5.8 Hz, 1H, H-3’’,5’’),
.87 (d, J = 5.7 Hz, 1H, H-7), 5.66 (s, 2H, CH ), 4.13 (m, 4H, piperazine H-2,6), 2.45
6
2
1
3
(m, 4H, piperazine H-3,5), 2.22 (s, 3H, NCH ). C NMR (151 MHz, DMSO-d ) δ
3
6
1
1
51.32 (C-4), 150.29 (C-2’,6’), 150.10 (C-2’’,6’’), 147.17 (C-2), 145.36 (C-4’’),
42.59 (C-7a), 141.08 (C-6), 136.74 (C-4’), 127.29 (C-3a), 122.96 (C-3’,5’), 121.16
(
C-3’’,5’’), 97.50 (C-7), 54.80 (piperazine C-3,5), 46.75 (CH ), 45.94 (NCH ), 45.66
2 3
+
(piperazine C-2,6). HRMS-ESI(+) (C H N ) [M+H] m/z calcd 386.2088, found
22 24 7
3
4
86.2089.
.1.1.7. 4-Benzylamino-1,3-dihydroimidazo[4,5-c]pyridin-2-one (11a) and 7-
amino-3-benzyl-3H-imidazo[4,5-b]pyridin-2-one (12a)
These compounds were synthesized using a procedure analogous to that of 10a. The
residue was purified using silica gel column chromatography (ethyl acetate/methanol:
9
/1) and each isomer was isolated using preparative thin layer chromatography
(
dichloromethane/methanol 95:5).
o
1
Data for 11a: Yield 15%, white solid, m.p. 197-198 C (ether), H NMR (600 MHz,
Acetone-d ) δ 9.90-9.82 (brs, 1H, imidazole NH), 9.48-9.40 (brs, 1H, imidazole NH),
6
7
=
.73 (d, J = 5.4 Hz, 1H, H-6), 7.40 (d, J = 7.6 Hz, 2H, benzylamine H-2,6), 7.31 (t, J
7.3 Hz, 2H, benzylamine H-3,5), 7.23 (t, J = 7.3 Hz, 1H, benzylamine H-4), 6.47 (d,
J = 5.4 Hz, 1H, H-7), 5.71 (s, 1H, benzylamine NH), 4.71 (d, J = 5.5 Hz, 2H,
13
benzylamine CH ). C NMR (151 MHz, Acetone-d ) δ 155.12 (C-2), 144.16 (C-4),
2
6
1
1
4
41.62 (benzylamine C-1), 141.20 (C-6), 136.06 (C-7a), 129.18 (benzylamine C-3,5),
28.57 (benzylamine C-2,6), 127.62 (benzylamine C-4), 112.10 (C-3a), 97.72 (C-7),
-
5.55 (benzylamine CH ). HRMS-ESI(-) (C H N O) [M-H] m/z calcd 239.0938,
2
13 11
4
found 239.0945).

1
6
ACCEPTED MANUSCRIPT
o
1
Data for 12a: Yield 35%, brown solid m.p. 256-257 C (ether), H NMR (600 MHz,
DMSO-d ) δ 10.34-10.28 (brs, 1H, NH), 7.56 (d, J = 5.7 Hz, 1H, H-5), 7.29 (m, 4H,
6
benzyl H-2,3,5,6), 7.23 (m, 1H, benzyl H-4), 6.33 (d, J = 5.7 Hz, 1H, H-6), 5.77 (s,
1
3
2
1
3
H, NH ), 4.93 (s, 2H, benzyl CH ). C NMR (151 MHz, DMSO-d ) δ 152.68 (C-2),
2 2 6
43.90 (C-3a), 141.31 (C-5), 137.77 (benzyl C-1), 137.35 (C-7), 128.37 (benzyl C-
,5), 127.39 (benzyl C-2,6), 127.13 (benzyl C-4), 107.33 (C-7a), 104.95 (C-6), 42.12
+
(
benzyl CH ). HRMS-ESI (+) (C H N O) [M+H] m/z calcd 241.1084, found
2
13 13
4
2
4
41.1086).
.1.1.8. 4-Benzylamino-1,3-dihydroimidazo[4,5-c]pyridine-2-thione (11b) and 7-
amino-3-benzyl-3H-imidazo[4,5-b]pyridine-2-thione (12b)
These compounds were synthesized using a procedure analogous to that of 10a. The
residue was purified using silica gel column chromatography (cyclohexane/ethyl
acetate: 1/2) and each isomer was isolated using preparative thin layer
chromatography (dichloromethane/methanol 95:5).
o
1
Data for 11b: Yield 15%, brown solid m.p. 250-251 C (ether), H NMR (600 MHz,
Acetone-d ) δ 11.54-11.40 (brs, 1H, imidazole NH), 11.37-11.20 (brs, 1H, imidazole
6
NH), 7.82 (d, J = 5.6 Hz, 1H, H-6), 7.43 (d, J = 7.6 Hz, 2H, benzylamine H-2,6), 7.32
(t, J = 7.6 Hz, 2H, benzylamine H-3,5), 7.25 (t, J = 7.4 Hz, 1H, benzylamine H-4),
6
.61 (d, J = 5.6 Hz, 1H, H-7), 6.13-6.10 (brs, 1H, benzylamine NH), 4.75 (d, J = 5.2
13
Hz, 2H, benzylamine CH ). C NMR (151 MHz, Acetone-d ) δ 169.64 (C-2), 144.37
2
6
(C-4), 142.52 (C-6), 141.00 (benzylamine C-1), 138.78 (C-7a), 129.24 (benzylamine
C-3,5), 128.62 (benzylamine C-2,6), 127.77 (benzylamine C-4), 115.96 (C-3a), 97.38
+
(
C-7), 45.55 (benzylamine CH ). HRMS-ESI(+) (C H N S) [M+H] m/z calcd
2 13 13 4
2
57.0855, found 257.0861)
o
1
Data for 12b: Yield 30%, white solid m.p. 262-263 C (ether), H NMR (600 MHz,
DMSO-d ) δ 13.07-11.68 (brs, 1H, NH), 7.74 (d, J = 5.5 Hz, 1H, H-5), 7.37 (d, J =
6
7
.3 Hz, 2H, benzyl H-2,6), 7.29 (t, J = 7.4 Hz, 2H, benzyl H-3,5), 7.23 (t, J = 7.1 Hz,
1
H, benzyl H-4), 6.43 (d, J = 5.5 Hz, 1H, H-6), 6.17 (s, 2H, NH ), 5.37 (s, 2H, benzyl
2
13
CH ). C NMR (151 MHz, DMSO-d ) δ 166.36 (C-2), 145.87 (C-3a), 143.74 (C-5),
2
6
1
1
38.55 (C-7), 136.85 (benzyl C-1), 128.25 (benzyl C-3,5), 127.74 (benzyl C-2,6),
27.23 (benzyl C-4), 110.89 (C-7a), 104.73 (C-6), 44.92 (benzyl CH ). HRMS-ESI(+)
2
+
(C H N S) [M+H] m/z calcd 257.0855, found 257.0848).
1
3
13
4
4
3
.1.1.9. N-Benzyl-2-(4-nitrophenyl)-1H-imidazo[4,5-c]pyridin-4-amine (11c) and
-benzyl-2-(4-nitrophenyl)-3H-imidazo[4,5-b]pyridin-7-amine (12c)
These compounds were synthesized using a procedure analogous to that of 10c. The
residue was purified using silica gel column chromatography (ethyl acetate) and each
isomer was isolated using preparative thin layer chromatography (ethyl acetate).
o
1
Data for 11c: Yield 25%, yellow solid m.p.: 255-256 C (ether), H NMR (600 MHz,
Acetone-d ) δ 12.23-11.28 (brs, 1H, imidazole NH), 8.40 (d, J = 8.9 Hz, 2H,
6
nitrophenyl H-3,5), 8.36 (d, J = 8.9 Hz, 2H, nitrophenyl H-2,6), 7.80 (d, J = 5.8Hz,
1
H, H-6), 7.46 (d, J = 7.5 Hz, 2H, benzylamine H-2,6), 7.30 (t, J = 7.6 Hz, 2H,
benzylamine H-3,5), 7.21 (t, J = 7.4 Hz, 1H, benzylamine H-4), 6.81 (d, J = 5.8 Hz,
13
1
H, H-7), 6.78-6.71 (brs, 1H, benzylamine NH), 4.90 (s, 2H, benzylamine CH ). C
2
NMR (151 MHz, Acetone-d ) δ 152.30 (C-4), 149.07 (nitrophenyl C-1), 148.14
6
(nitrophenyl C-4), 142.21 (C-3a), 141.98 (C-6), 137.01 (C-2), 129.09 (benzylamine
C-3,5), 128.42 (benzylamine C-2,6), 127.96 (nitrophenyl C-3,5), 127.47 (benzylamine
C-4), 125.02 (nitrophenyl C-2,6), 98.62 (C-7), 45.00 (benzylamine CH ). HRMS-
2
-
ESI(-) (C H N O ) [M-H] m/z calcd 344.1153, found 344.1141).
1
9
14
5
2
o
1
Data for 12c: Yield 35%, yellow solid m.p.: 198-199 C (methanol), H NMR (600
MHz, DMSO-d ) δ 8.32 (d, J = 8.8 Hz, 2H, nitrophenyl H-3,5), 8.01 (d, J = 8.8 Hz,
6

1
7
ACCEPTED MANUSCRIPT
2
H, nitrophenyl H-2,6), 7.90 (d, J = 5.5 Hz, 1H, H-5), 7.24 (t, J = 7.3 Hz, 2H, benzyl
H-3,5), 7.19 (t, J = 7.2 Hz, 1H, benzyl H-4), 6.98 (d, J = 7.3 Hz, 2H, benzyl H-2,6),
13
6
.61 (s, 2H, NH ), 6.46 (d, J = 5.5 Hz, 1H, H-6), 5.62 (s, 2H, benzyl CH ). C NMR
2 2
(
151 MHz, DMSO-d ) δ 149.19 (C-2), 147.64 (nitrophenyl C-4), 147.17 (C-3a),
6
1
(
2
47.00 (C-7), 145.74 (C-5), 137.18 (benzyl C-1), 136.58 (nitrophenyl C-1), 129.70
nitrophenyl C-2,6), 128.64 (benzyl C-3,5), 127.35 (benzyl C-4), 126.29 (benzyl C-
,6), 123.80 (nitrophenyl C-3,5), 122.95 (C-7a), 102.84 (C-6), 45.94 (benzyl CH ).
2
+
HRMS-ESI(+) (C H N O ) [M+H] m/z calcd 346.1299, found 346.1288).
1
9
16
5
2
4
.1.1.10. N-Benzyl-2-(2-hydroxyphenyl)-1H-imidazo[4,5-c]pyridin-4-amine (11d)
and 3-benzyl-2-(2-hydroxyphenyl)-3H-imidazo[4,5-b]pyridin-7-amine (12d)
These compounds were synthesized using a procedure analogous to that of 10c. The
residue
was
purified
using
silica
gel
column
chromatography
(
dichloromethane/methanol: 9/1) to provide compounds 11d and 12d.
o
1
Data for 11d: Yield 15%, brown solid m.p.: 118-119 C (ether/n-pentane), H NMR
(
600 MHz, Acetone-d ) δ 12.54-12.36 (brs, 1H, imidazole NH), 12.32-12.14 (brs, 1H,
6
OH), 7.94 (dd, J = 7.8, 1.3 Hz, 1H, hydroxyphenyl H-3), 7.83 (d, J = 5.6 Hz, 1H, H-
), 7.45 (d, J = 7.5 Hz, 2H, benzylamine H-3,5), 7.35 (td, J = 7.8, 1.3 Hz, 1H,
hydroxyphenyl H-5), 7.29 (t, J = 7.6 Hz, 2H, benzylamine 2,6), 7.20 (t, J = 7.4 Hz,
H, benzylamine H-4), 7.17-7.11 (brs, 1H, benzylamine NH), 7.01 (dd, J = 7.8, 0.9
Hz, 1H, hydroxyphenyl H-6), 6.96 (td, J = 7.8, 0.9 Hz, 1H, hydroxyphenyl H-4), 6.79
6
1
1
3
(
d, J = 5.6 Hz, 1H, H-7), 4.88 (d, J = 5.2 Hz, 2H, benzylamine CH ). C NMR (151
2
MHz, Acetone-d ) δ 159.26 (C-2), 151.50 (C-4), 150.60 (hydroxyphenyl C-1), 142.37
6
(C-6), 142.15 (benzylamine C-1), 138.56 (C-7a), 132.23 (hydroxyphenyl C-5), 129.03
(benzylamine C-2,6), 128.40 (benzylamine C-3,5), 127.37 (benzylamine C-4), 126.38
(hydroxyphenyl C-3), 119.88 (hydroxyphenyl C-4), 118.28 (hydroxyphenyl C-6),
1
13.56 (hydroxyphenyl C-2), 97.97 (C-7), 44.83 (benzylamine CH ). HRMS-ESI(+)
2
+
(C H N O) [M+H] m/z calcd 317.1397, found 317.1407.
19 17 4
o
1
Data for 12d: Yield 35%, white solid m.p.: 253-254 C (ether), H NMR (600 MHz,
DMSO-d ) δ 11.43 (s, 1H, OH), 7.85 (d, J = 5.5 Hz, 1H, H-5), 7.41 (dd, J = 7.7, 1.4
6
Hz, 1H, hydroxyphenyl H-6), 7.32 (td, J = 7.7, 1.4 Hz, 1H, hydroxyphenyl H-4), 7.22
(
7
t , J = 7.3 Hz, 2H, benzyl H-3,5), 7.18 (t, J = 7.2 Hz, 1H, benzyl H-4), 7.02 (dd, J =
.7, 0.8 Hz, 1H, hydroxyphenyl H-3), 6.97 (d, J = 7.3 Hz, 2H, benzyl H-2,6), 6.84 (td,
J = 7.7, 0.8 Hz, 1H, hydroxyphenyl H-5), 6.54 (s, 2H, NH ), 6.43 (d, J = 5.5 Hz, 1H,
2
1
3
H-6), 5.54 (s, 2H, benzyl CH2). C NMR (151 MHz, DMSO-d ) δ 156.61
6
(hydroxyphenyl C-2), 148.26 (C-2), 147.99 (C-7), 146.25 (C-3a), 144.93 (C-5),
1
1
1
1
3
4
37.36 (benzyl C-1), 131.15 (hydroxyphenyl C-4), 129.41 (hydroxyphenyl C-6),
28.48 (benzyl C-3,5), 127.18 (benzyl C-4), 126.40 (benzyl C-2,6), 121.32 (C-7a),
18.88 (hydroxyphenyl C-5), 116.64 (hydroxyphenyl C-3), 115.72 (hydroxyphenyl C-
+
), 102.79 (C-6), 46.06 (benzyl CH ). HRMS-ESI(+) (C H N O) [M+H] m/z calcd
2
19 17
4
17.1397, found 317.1392.
.1.1.11. N-Benzyl-2-(2,5-dimethoxyphenyl)-1H-imidazo[4,5-c]pyridin-4-amine
(11e) and 3-benzyl-2-(2,5-dimethoxyphenyl)-3H-imidazo[4,5-b]pyridin-7-amine
(12e)
These compounds were synthesized using a procedure analogous to that of 10c. The
residue
was
purified
using
silica
gel
column
chromatography
(
dichloromethane/methanol: 19/1) to provide compounds 11e and 12e.
o
1
Data for 11e: Yield 20%, brown solid m.p.: 190-191 C (ether/n-pentane), H NMR
(
600 MHz, Acetone-d ) δ 11.82-11.76 (brs, 1H, imidazole NH), 8.02 (d, J = 3.1 Hz,
6
1
H, dimethoxyphenyl H-6), 7.74 (d, J = 5.5 Hz, 1H, H-6), 7.45 (d, J = 7.5 Hz, 2H,
benzylamine H-2,6), 7.30 (t, J = 7.5 Hz, 2H, benzylamine H-3,5), 7.20 (t, J = 7.3 Hz,

1
8
ACCEPTED MANUSCRIPT
1
9
1
H, benzylamine H-4), 7.15 (d, J = 9.0 Hz, 1H, dimethoxyphenyl H-3), 7.01 (dd, J =
.0, 3.1 Hz, 1H, dimethoxyphenyl H-4), 6.75 (d, J = 5.5 Hz, 1H, H-7), 6.53-6.49 (brs,
H, benzylamine NH), 4.88 (d, J = 5.8 Hz, 2H, benzylamine CH ), 4.00 (s, 3H,
2
1
3
dimethoxyphenyl CH O-5), 3.81 (s, 3H, dimethoxyphenyl CH O-2). C NMR (151
3
3
MHz, Acetone-d ) δ 154.88 (dimethoxyphenyl C-2), 152.16 (dimethoxyphenyl C-5),
6
1
1
52.09 (C-4), 148.02 (C-2), 142.33 (benzylamine C-1), 141.14 (C-6), 139.30 (C-7a),
29.05 (benzylamine C-3,5), 128.42 (benzylamine C-2,6), 128.09 (C-3a), 127.37
(
1
(
benzylamine C-4), 119.61 (dimethoxyphenyl C-1), 117.45 (dimethoxyphenyl C-4),
14.76 (dimethoxyphenyl C-6), 113.94 (dimethoxyphenyl C-3), 98.72 (C-7), 56.57
dimethoxyphenyl CH O-5), 56.06 (dimethoxyphenyl CH O-2), 44.97 (benzylamine
3
3
-
CH ). HRMS-ESI(-) (C H N O ) [M-H] m/z calcd 359.1514, found 359.1498.
Data for 12e: Yield 45%, orange solid m.p.: 90-91 C (ether/n-pentane), H NMR (600
2
21 19
4
2
o
1
MHz, DMSO-d ) δ 7.83 (d, J = 5.5 Hz, 1H, H-5), 7.16 (m, 3H, dimethoxyphenyl H-4,
6
benzyl H-3,5), 7.07 (m, 2H, dimethoxyphenyl H-3, benzyl H-4), 6.88 (d, J = 6.9 Hz,
2
H, benzyl H-2,6), 6.85 (d, J = 2.6 Hz, 1H, dimethoxyphenyl H-6), 6.40 (d, J = 5.5
Hz, 1H, H-6), 6.36 (s, 2H, NH ), 5.21 (s, 2H, benzyl CH ), 3.66 (s, 3H,
2
2
1
3
dimethoxyphenyl CH O-2), 3.62 (s, 3H, dimethoxyphenyl CH O-5). C NMR (151
3
3
MHz, DMSO-d ) δ 152.80 (dimethoxyphenyl C-2), 151.17 (dimethoxyphenyl C-5),
6
1
(
48.26 (C-2), 147.29 (dimethoxyphenyl C-1), 146.53 (C-3a), 144.59 (C-5), 137.44
benzyl C-1), 128.14 (dimethoxyphenyl C-4), 126.99 (benzyl C-3,5), 126.79 (benzyl
C-2,6), 122.68 (C-7), 120.27 (C-7a), 116.98 (dimethoxyphenyl C-6), 116.65
dimethoxyphenyl C-3), 112.62 (benzyl C-4), 102.32 (C-6), 55.67 (dimethoxyphenyl
CH O-2), 55.55 (dimethoxyphenyl CH O-5), 45.55 (benzyl CH ). HRMS-ESI(+)
(
3
3
2
+
(C H N O ) [M+H] m/z calcd 361.1659, found 361.1655.
21 21 4 2
4
3
.1.1.12. N-Benzyl-2-(pyridin-4-yl)-1H-imidazo[4,5-c]pyridin-4-amine (11f) and
-benzyl-2-(pyridin-4-yl)imidazo[4,5-b]pyridin-7-amine (12f)
These compounds were synthesized using a procedure analogous to that of 6c. The
residue
was
purified
using
silica
gel
column
chromatography
(
dichloromethane/methanol: 9/1) to provide compounds 11f and 12f.
o
1
Data for 11f: Yield 25%, brown solid m.p.: 150-151 C (ether/n-pentane), H NMR
(
600 MHz, Acetone-d ) δ 12.78-11.75 (brs, 1H, NH imidazole), 8.71 (dd, J = 4.5, 1,5
6
Hz, 2H, pyridinyl H-3,5), 8.04 (dd, J = 4.5, 1.5 Hz, 2H, pyridinyl H-2,6), 7.81 (d, J =
5
2
.5 Hz, 1H, H-6), 7.45 (d, J = 7.7 Hz, 2H, benzylamine H-2,6), 7.30 (t, J = 7.6 Hz,
H, benzylamine H-3,5), 7.21 (t, J = 7.4 Hz, 1H, benzylamine H-4), 6.79 (d, J = 5.5
Hz, 1H, H-7), 6.70-6.65 (brs, 1H, benzylamine NH), 4.89 (d, J = 4.9 Hz, 2H,
1
3
benzylamine CH2). C NMR (151 MHz, Acetone-d ) δ 152.40 (C-4), 151.45
6
(
7
(
7
pyridinyl C-3,5), 147.78 (C-2), 142.31 (C-6), 142.03 (benzylamine C-1), 140.24 (C-
a), 138.00 (pyridinyl C-1), 129.12 (C-3a), 129.08 (benzylamine C-3,5), 128.41
benzylamine C-2,6), 127.45 (benzylamine C-4), 120.90 (pyridinyl C-2,6), 98.38 (C-
-
), 44.96 (benzylamine CH ). HRMS-ESI(-) (C H N ) [M-H] m/z calcd 300.1255,
2 18 14 5
found 300.1243.
o
1
Data for 12f: Yield 40%, brown solid m.p.: 222-223 C (ether), H NMR (600 MHz,
DMSO-d ) δ 8.67 (d, J = 5.8 Hz, 2H, pyridinyl H-3,5), 7.90 (d, J = 5.5 Hz, 1H, H-5),
6
7
.69 (d, J = 5.8 Hz, 2H, pyridinyl H-2,6), 7.25 (t, J = 7.4 Hz, 2H, benzyl H-3,5), 7.20
(
t, J = 7.2 Hz, 1H, benzyl H-4), 6.98 (d, J = 7.5 Hz, 2H, benzyl H-2,6), 6.59 (s, 2H,
1
3
NH ), 6.46 (d, J = 5.5Hz, 1H, H-6), 5.62 (s, 2H, benzyl CH ). C NMR (151 MHz,
2
2
DMSO-d ) δ 150.10 (pyridinyl C-3,5), 149.16 (C-2), 147.18 (C-3a), 146.56 (C-7),
6
1
1
45.77 (C-5), 137.68 (pyridinyl C-1), 137.25 (benzyl C-1), 128.66 (benzyl C-3,5),
27.35 (benzyl C-4), 126.20 (benzyl C-2,6), 122.80 (C-7a), 122.57 (pyridinyl C-2,6),

1
9
ACCEPTED MANUSCRIPT
+
1
3
4
02.81 (C-6), 45.83 (benzyl CH ). HRMS-ESI(+) (C H N ) [M+H] m/z calcd
2 18 16 5
02.1400, found 302.1408.
.2 Aspergillus manipulations. Standard complete and minimal media (MM) for A.
nidulans were used. Media and supplemented auxotrophies were at the concentrations
given in http://www.fgsc.net. 10 mM, NaNO was used as a nitrogen source.
3
Inhibitors are added in MM dissolved in DMSO at 500 µΜ. Transformations were
2
0,23
performed as described previously. Strains have been described before.
In brief,
the three strains used each expressed either FcyB, or AzgA or UapA, in an otherwise
isogenic genetic background that lacked, due to total deletions of the corresponding
genes, any other nucleobase-nucleoside-allantoin transporter (uapA∆ uapC∆ azgA∆
furD∆ furA∆ fcyb∆ cntA). The transporter to be studied in each case was introduced
2
4
by standard genetic transformation with a plasmid carrying the corresponding gene
expressed by its native promoter.
4
.3 Transport assays. Transport assays for measuring the activity of purine
transporters, such as FcyB, AzgA or UapA, is carried out in germinating
conidiospores, as recently described in detail by Krypotou and Diallinas.[22] For
3
transport competition assays, 0.5 µΜ of H-radiolabelled substrate (adenine or
xanthine) is added in a mix with 1000-fold excess analogues (500 µM). Assays are
terminated by freezing, immediate centrifugation and washing of cells. K values are
i
estimated from IC50 measurements using the Cheng and Prussof equation [Ki =
IC50/1+ [S]/Km, where [S] is the fixed concentration of radiolabeled substrate used]
and analyzed by the GraphPad Prism software. All experiments are carried out at
three times, with each assays performed in triplicate. Standard deviation in all cases is
3
3
less than 30%. Radiolabeled purines used are: [2,8- H]-adenine or [8- H]-xanthine
2
2.8 Ci/mmol, all from Moravek Biochemicals.
4
.4 Homology Modeling. The 3D Model of Fcyb was constructed based on the
crystal structure of bacterial benzyl-hydantoin Mhp1 transporter (PDB entry
D1D)[13] and we utilized the alignment already published by our group.[7] The final
4
model was built using Prime software (Schrodinger Release 2017-4: Prime,
Schrödinger, LLC, New York, NY, 2017) and the resulted model was prepared using
the Protein Preparation Wizard as implemented on Maestro 10 (Schrödinger Release
2
017-4: Schrödinger Suite 2017-4 Protein Preparation Wizard; Epik, Schrödinger,
LLC, New York, NY, 2017; Impact, Schrödinger, LLC, New York, NY, 2017; Prime,
Schrödinger, LLC, New York, NY, 2017).
4
4
.5 Docking calculations.
.5.1. Ligand and Protein Preparation. All analogues designed were prepared using
the ligand preparation wizard as implemented in Maestro 10 (Schrödinger Release
2
017-4: LigPrep, Schrödinger, LLC, New York, NY, 2017).
4
.5.2. Induced Fit Docking. Schrödinger developed and validated an Induced Fit
Docking protocol based on Glide and the Refinement module in Prime for accurate
prediction of ligand binding modes and concomitant structural changes in the
receptor.( Schrödinger Release 2017-4: Schrödinger Suite 2017-4 Induced Fit
Docking protocol; Glide, Schrödinger, LLC, New York, NY, 2016; Prime,
Schrödinger, LLC, New York, NY, 2017).[25] The IFD protocol models induced fit
docking of ligands using the following steps:
1
. An optional constrained minimization of the receptor (protein preparation,
refinement only) with an RMSD cutoff of 0.18 Å. Normally this is done when
preparing the protein with the Protein Preparation Wizard.

2
0
ACCEPTED MANUSCRIPT
2
. Initial Glide docking of each ligand using a softened potential (van der Waals radii
scaling), and optional removal of side chains and application of constraints. By
default, a maximum 20 poses per ligand are retained, and by default poses to be
retained must have a Coulomb-vdW score less than 100 and an H-bond score less than
–
3
0.05.
. Prime side-chain prediction for each protein-ligand complex, on residues within a
given distance of any ligand pose (default 5 Å), with optional inclusion or exclusion
of other residues, and an optional implicit membrane model.
4
. Prime minimization of the same set of residues and the ligand for each
protein/ligand complex pose. The receptor structure in each pose now reflects an
induced fit to the ligand structure and conformation.
5
. Glide redocking of each protein-ligand complex structure within a specified energy
of the lowest-energy structure (default 30 kcal/mol). The ligand is now rigorously
docked, using default Glide settings, into the induced-fit receptor structure.
6
. Estimation of the binding energy (IFDScore) for each output pose.
4
.6. Molecular Dynamics Simulations. Molecular Dynamic simulations were
conducted with Desmond v.3 software (Schrödinger Release 2017-3: Desmond
Molecular Dynamics System, D. E. Shaw Research, New York, NY, 2017. Maestro-
Desmond Interoperability Tools, Schrödinger, New York, NY, 2017.).[26] First the
system was prepared by embedding the protein in a POPC lipid bilayer, solvating the
membrane by TIP3P explicit water, neutralizing with counter ions and adding
1
50ꢀmM salt. Initially a stepwise equilibration protocol was utilized, developed by
Desmond for membrane proteins. A 90 ns simulation was performed in the NPT
ensemble with Langevin thermostat and barostat and semi isotropic pressure restraints
for the substrate studied (compound 10a). All figures were created with Maestro v10
(Schrödinger Release 2017-4: Maestro, Schrödinger, LLC, New York, NY, 2017).
ACKNOWLEDGEMENTS
This research was co-supported by the European Union (European Social Fund) and
Greek national funds through the Operational Program “Education and Lifelong
Learning” of the National Strategic Reference Framework-Research Funding
Program: THALES, investing in knowledge society through the European Social
Fund. G. Papadaki was supported by a “Stavros Niarchos Foundation” fellowship
within the framework of a research grant to the National and Kapodistrian University
of Athens.
SUPPLEMENTARY DATA
The Supplementary Data contains the 1H-NMR and 13C-NMR spectra of synthesized
compounds, and Supplementary Figure SF1 related to the MD simulations.
REFERENCES
[
1] Darlington, A. J., and Scazzocchio, C. (1967) Use of Analogues and the Substrate-
Sensitivity of Mutants in Analysis of Purine Uptake and Breakdown in Aspergillus
nidulans, Journal of Bacteriology 93, 937-940.
[
2] Diallinas, G., and Scazzocchio, C. (1989) A gene coding for the uric acid-xanthine
permease of Aspergillus nidulans: inactivational cloning, characterization, and
sequence of a cis-acting mutation, Genetics 122, 341.

2
1
ACCEPTED MANUSCRIPT
[
3] Pantazopoulou, A., and Diallinas, G. (2007) Fungal nucleobase transporters,
FEMS Microbiology Reviews 31, 657-675.
4] Diallinas, G., and Gournas, C. (2008) Structure-function relationships in the
[
nucleobase-ascorbate transporter (NAT) family: lessons from model microbial genetic
systems, Channels (Austin, Tex.) 2, 363-372.
[
5] Gournas, C., Papageorgiou, I., and Diallinas, G. (2008) The nucleobase–ascorbate
transporter (NAT) family: genomics, evolution, structure–function relationships and
physiological role, Molecular BioSystems 4, 404.
[
(
6] Krypotou, E., Lambrinidis, G., Evangelidis, T., Mikros, E., and Diallinas, G.
2014) Modelling, substrate docking and mutational analysis identify residues
essential for function and specificity of the major fungal purine transporter AzgA,
Molecular Microbiology 93, 129-145.
[
7] Krypotou, E., Kosti, V., Amillis, S., Myrianthopoulos, V., Mikros, E., and
Diallinas, G. (2012) Modeling, Substrate Docking, and Mutational Analysis Identify
Residues Essential for the Function and Specificity of a Eukaryotic Purine-Cytosine
NCS1 Transporter, Journal of Biological Chemistry 287, 36792-36803.
[
8] Sioupouli, G., Lambrinidis, G., Mikros, E., Amillis, S., and Diallinas, G. (2017)
Cryptic purine transporters in <i>Aspergillus nidulans</i> reveal the role of specific
residues in the evolution of specificity in the NCS1 family: Cryptic purine
transporters in <i>A. nidulans</i>, Molecular Microbiology 103, 319-332.
[
9] Hamari, Z., Amillis, S., Drevet, C., Apostolaki, A., Vágvölgyi, C., Diallinas, G.,
and Scazzocchio, C. (2009) Convergent evolution and orphan genes in the Fur4p-like
family and characterization of a general nucleoside transporter in <i>Aspergillus
nidulans</i>, Molecular Microbiology 73, 43-57.
[
10] Krypotou, E., Evangelidis, T., Bobonis, J., Pittis, A. A., Gabaldón, T.,
Scazzocchio, C., Mikros, E., and Diallinas, G. (2015) Origin, diversification and
substrate specificity in the family of NCS1/FUR transporters: Origin and specificity
of Fur transporters, Molecular Microbiology 96, 927-950.
[
11] Lu, F., Li, S., Jiang, Y., Jiang, J., Fan, H., Lu, G., Deng, D., Dang, S., Zhang, X.,
Wang, J., and Yan, N. (2011) Structure and mechanism of the uracil transporter UraA,
Nature 472, 243-246.
[
12] Shimamura, T., Weyand, S., Beckstein, O., Rutherford, N. G., Hadden, J. M.,
Sharples, D., Sansom, M. S. P., Iwata, S., Henderson, P. J. F., and Cameron, A. D.
2010) Molecular Basis of Alternating Access Membrane Transport by the Sodium-
Hydantoin Transporter Mhp1, Science 328, 470-473.
13] Simmons, K. J., Jackson, S. M., Brueckner, F., Patching, S. G., Beckstein, O.,
(
[
Ivanova, E., Geng, T., Weyand, S., Drew, D., Lanigan, J., Sharples, D. J., Sansom, M.
S., Iwata, S., Fishwick, C. W., Johnson, A. P., Cameron, A. D., and Henderson, P. J.
[
2014) Molecular mechanism of ligand recognition by membrane transport protein,
Mhp1, The EMBO Journal 33, 1831-1844.
14] Weyand, S., Shimamura, T., Yajima, S., Suzuki, S., Mirza, O., Krusong, K.,
[
Carpenter, E. P., Rutherford, N. G., Hadden, J. M., O'Reilly, J., Ma, P., Saidijam, M.,
Patching, S. G., Hope, R. J., Norbertczak, H. T., Roach, P. C. J., Iwata, S., Henderson,
P. J. F., and Cameron, A. D. (2008) Structure and Molecular Mechanism of a
Nucleobase-Cation-Symport-1 Family Transporter, Science 322, 709-713.
[
15] Yu, X., Yang, G., Yan, C., Baylon, J. L., Jiang, J., Fan, H., Lu, G., Hasegawa, K.,
Okumura, H., Wang, T., Tajkhorshid, E., Li, S., and Yan, N. (2017) Dimeric structure
of the uracil:proton symporter UraA provides mechanistic insights into the
SLC4/23/26 transporters, Cell Research 27, 1020-1033.

2
2
ACCEPTED MANUSCRIPT
[
16] Alguel, Y., Amillis, S., Leung, J., Lambrinidis, G., Capaldi, S., Scull, N. J.,
Craven, G., Iwata, S., Armstrong, A., Mikros, E., Diallinas, G., Cameron, A. D., and
Byrne, B. (2016) Structure of eukaryotic purine/H+ symporter UapA suggests a role
for homodimerization in transport activity, Nature Communications 7, 11336.
[
17] Diallinas, G. (2014) Understanding transporter specificity and the discrete
appearance of channel-like gating domains in transporters, Frontiers in Pharmacology
5
.
[
18] Diallinas, G. (2016) Dissection of Transporter Function: From Genetics to
Structure, Trends in Genetics 32, 576-590.
19] Alguel, Y., Cameron, A. D., Diallinas, G., and Byrne, B. (2016) Transporter
oligomerization: form and function, Biochemical Society Transactions 44, 1737-1744.
20] Lougiakis, N., Gavriil, E.-S., Kairis, M., Sioupouli, G., Lambrinidis, G., Benaki,
[
[
D., Krypotou, E., Mikros, E., Marakos, P., Pouli, N., and Diallinas, G. (2016) Design
and synthesis of purine analogues as highly specific ligands for FcyB, a ubiquitous
fungal nucleobase transporter, Bioorganic & Medicinal Chemistry 24, 5941-5952.
[
21] Yin, X., and Schneller, S. W. (2004) L-deaza-5'-noraisteromycin, Nucleosides,
Nucleotides & Nucleic Acids 23, 67-76.
22] Krypotou, E., and Diallinas, G. (2014) Transport assays in filamentous fungi:
[
kinetic characterization of the UapC purine transporter of Aspergillus nidulans,
Fungal Genet Biol 63, 1-8.
[
(
23] Shivakumar, D., Williams, J., Wu, Y., Damm, W., Shelley, J., and Sherman, W.
2010) Prediction of Absolute Solvation Free Energies using Molecular Dynamics
Free Energy Perturbation and the OPLS Force Field, Journal of Chemical Theory and
Computation 6, 1509-1519.
[
24] Koukaki, M., Giannoutsou, E., Karagouni, A., and Diallinas, G. (2003) A novel
improved method for Aspergillus nidulans transformation, Journal of Microbiological
Methods 55, 687-695.
[
25] Sherman, W., Day, T., Jacobson, M. P., Friesner, R. A., and Farid, R. (2006)
Novel procedure for modeling ligand/receptor induced fit effects, J Med Chem 49,
34-553.
5
[
(
26] Shivakumar, D., Williams, J., Wu, Y., Damm, W., Shelley, J., and Sherman, W.
2010) Prediction of Absolute Solvation Free Energies using Molecular Dynamics
Free Energy Perturbation and the OPLS Force Field, Journal of Chemical Theory and
Computation 6, 1509-1519.

ACCEPTED MANUSCRIPT
Structure-activity relationships in fungal nucleobases
transporters as dissected by the inhibitory effects of novel
purine analogues
a
a
b
Efthymios-Spyridon Gavriil, Spyridon Dimitrakis, Georgia Papadaki,
b
a
a
Sophia Balaska, George Lambrinidis, Nikolaos Lougiakis, Panagiotis
a
,b
a
,a
Marakos, George Diallinas* , Nicole Pouli, Emmanuel Mikros*
Highlights
•
Rationally designed 3-deazapurine analogues are synthesized and used in studying
transporter-ligand interactions
•
Novel 3-deazaxanthine analogues are recognized by a single (FcyB) or multiple (FcyB
and AzgA) fungal transporters
•
•
Active compounds stabilize the outward transporter conformation.
Fungal purine transporters can be used as specific gateways for the design of highly
targeted novel antifungals or drug cocktails