Structure-activity relationship study of prion inhibition by 2-aminopyridine-3,5-dicarbonitrile-based compounds: Parallel synthesis, bioactivity, and in vitro pharmacokinetics

Article abstract of DOI:10.1021/jm061045z

2-Aminopyridine-3,5-dicarbonitrile compounds were previously identified as mimetics of dominant-negative prion protein mutants and inhibit prion replication in cultured cells. Here, we report findings from a comprehensive structure-activity relationship study of the 6-aminopyridine-3,5-dicarbonitrile scaffold. We identify compounds with significantly improved bioactivity (approximately 40-fold) against replication of the infectious prion isoform (PrP^Sc) and suitable pharmacokinetic profiles to warrant evaluation in animal models of prion disease.

Full text of DOI:10.1021/jm061045z

J. Med. Chem. 2007, 50, 65-73
65
Structure-Activity Relationship Study of Prion Inhibition by
2-Aminopyridine-3,5-dicarbonitrile-Based Compounds: Parallel Synthesis, Bioactivity, and in
Vitro Pharmacokinetics
Barnaby C. H. May,*,†,‡ Julie A. Zorn,§ Juanita Witkop,† John Sherrill,§ Andrew C. Wallace,§ Giuseppe Legname,†,‡
Stanley B. Prusiner,†,‡,# and Fred E. Cohen†,§
Institute for NeurodegeneratiVe Diseases and Departments of Neurology, Cellular and Molecular Pharmacology, and Biochemistry and
Biophysics, UniVersity of CaliforniasSan Francisco, San Francisco, California 94158
ReceiVed August 31, 2006
2-Aminopyridine-3,5-dicarbonitrile compounds were previously identified as mimetics of dominant-negative
prion protein mutants and inhibit prion replication in cultured cells. Here, we report findings from a
comprehensive structure-activity relationship study of the 6-aminopyridine-3,5-dicarbonitrile scaffold. We
identify compounds with significantly improved bioactivity (approximately 40-fold) against replication of
the infectious prion isoform (PrPSc) and suitable pharmacokinetic profiles to warrant evaluation in animal
models of prion disease.
Introduction
Misfolding of the cellular prion protein, PrPC, to a â-rich
conformation, denoted PrPSc, is the underlying molecular event
that gives rise to the prion diseases.1 Subsequent deposition of
oligomeric PrPSc in the central nervous system leads to neuronal
loss and rapid death in animals and humans.2 The conversion
of PrPC to PrPSc is thought to proceed via formation of a
complex between the PrP isoforms and an as-yet unidentified
molecular chaperone (Figure 1A, denoted “X”).3 The proposed
PrP-X binding may involve an epitope that maps to dominant-
negative PrP mutants, involving residues Q168 and Q172 of
helix B and residues T215 and Q219 of helix C (Figure 1B,
human PrP numbering).4 Dominant-negative PrP mutants protect
from disease, as demonstrated by familial polymorphisms in
the human and ovine prion protein gene (PRNP),5,6 transgenic
animal models,7 and transfection studies in scrapie-infected
cells.4
In an effort to identify inhibitors to PrPSc formation using a
structure-based paradigm, computational chemistry was used
to derive pharmacophore models based on the conformation and
electronic space enciphered by dominant-negative mutant PrPs.8
The pharmacophore models were used to query a virtual
compound library to identify compounds that, given their
mimicry of the dominant-negative epitope, might bind to the
molecular chaperone “X”. In binding the molecular chaperone,
these ligands would occlude binding of PrPs and hence inhibit
conversion of PrPC. Screening the candidate dominant-negative
mimetic compounds in scrapie-infected neuroblastoma (ScN2a)
cells identified a lead 2-aminopyridine-3,5-dicarbonitrile com-
pound, 1 (Figure 1C).
Figure 1. (A) Schematic diagram illustrating the conformational
conversion of the cellular prion protein (PrPC) to the infectious isoform
PrPSc, which is thought to involve an unknown molecular chaperone,
denoted ”X”. The representation of PrPSc is based on the â-helical model
of Govaerts et al.35 (B) Dominant-negative epitope on the surface of
PrPC, involving residues 168, 172, 215, and 219 (space-filled, based
on numbering of the human PrP sequence). (C) Lead 2-aminopyridine-
3,5-dicarbonitrile compound 1.
Here, we present biological and in vitro pharmacokinetic data
on a potent subset of 2-aminopyridine-3,5-dicarbonitrile com-
pounds. These new lead compounds are characterized by a
halobenzene and basic alkyl substituent at C-4 and C-6,
respectively, of the pyridine heterocycle. Additionally, we
identified a related potent scaffold, the thienopyridine. Certain
newly identified analogues are approximately 40-fold more
potent than the previous lead compounds in this class and
possess suitable physicochemical properties for adsorption,
distribution, metabolism, excretion, and toxicity (ADMET) and
efficacy studies in animals.
As part of a structure-activity relationship (SAR) of this class
of compound, we have identified 2-aminopyridine-3,5-dicar-
bonitrile compounds that are active at low micromolar concen-
trations in a scrapie-infected cell model of prion replication.
* To whom correspondence should be addressed. Phone: 415-514-4243.
Fax: 415-476-6515. E-mail: bmay@ind.ucsf.edu.
† Institute for Neurodegenerative Diseases.
‡ Department of Neurology.
§ Department of Cellular and Molecular Pharmacology.
# Department of Biochemistry and Biophysics.
10.1021/jm061045z CCC: $37.00 © 2007 American Chemical Society
Published on Web 12/13/2006

66 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1
May et al.
Scheme 1a
as LD50, the compound concentration at which 50% of the cells
were viable. Bioactivity against PrPSc accumulation was ex-
pressed as EC50, the compound concentration at which 50% of
PrPSc had been removed from the culture on exposure to
compound.
Biological Activity. Through iterative screening and synthesis
we have identified 2-aminopyridine-3,5-dicarbonitrile com-
pounds that are bioactive at low micromolar concentrations
against PrPSc accumulation and nontoxic up to 25 µM. The
placement of halobenzene substituents at C-4 of the scaffold
provided improved activity to the 2-aminopyridine-3,5-dicar-
bonitrile compounds (Table 1, e.g., 7-14). The meta-substituted
monohalobenzene substituents were more potent than equivalent
para-substituted analogues (Table 1; compare 7 and 8). Par-
ticularly potent and nontoxic analogues were identified bearing
basic alkyl substituents at C-6 of the pyridine heterocycle (Table
1, e.g., 17-23). The bioactivity of such analogues was depend-
ent on the size of the distal amine substituent (Table 1; compare
19 and 25) and the length of the alkyl linker that separates the
distal tertiary amine from the pyridine heterocycle (Table 1;
compare 22 and 23). While tertiary amines were bioactive at
low micromolar concentrations, free amines and carboxamides
(Table 1, 15 and 16, respectively) were not effective even at
concentrations up to 25 µM. Rigidifying the C-6 substituent by
cyclizing to a bicyclic thienopyridine heterocycle gave analogues
that were bioactive at low micromolar concentrations (Table 1,
28 and 29).
a Reagents and conditions: (a) 2-cyanothioacetamide (1 equiv), piperidine
(1.5 equiv), EtOH, reflux, 8 h; (b) 10% KOH, DMF, room temp, 1 min;
(c) aryl or alkyl halide, room temp, 5 h.
Scheme 2a
a Reagents and conditions: (a) 10% KOH, DMF, room temp, 5 h.
Results
Library Synthesis. The synthesis proceeded via formation
of 2-(arylidene)malononitriles (2) via â-alanine catalyzed Kno-
evenagel condensation of malononitrile and commercially
available aldehydes (Scheme 1).9 Precursor 2-(arylidene)-
malononitriles (2) were isolated by filtration and used without
further purification. The target pyridine compounds (4) were
synthesized in a one-pot, two-step reaction, whereby 2-(arylidene)-
malononitriles (2) were first reacted with 2-cyanothioacetamide
to yield the key intermediate (3).10 Second, solvent was removed,
and crude intermediates (3) were reacted with a chemset of alkyl
halides and aryl halides in the presence of 10% aqueous KOH
to furnish the C-6 substituent. The final products (4) were
purified by parallel HPLC, and the library components were
characterized by 1H NMR and LCMS. In addition, a thienopy-
ridine scaffold (Scheme 2, 6) was also targeted to introduce
rigidity to the C-6 substituent of the pyridine scaffold. Ap-
propriately substituted pyridine compounds (5) were cyclized
in the presence of 10% aqueous KOH to yield the corresponding
substituted thienopyridines (6).10 The original lead compound,
1, was prepared as previously reported.11 In total, 152 library
components were synthesized.
Duplex Cellular Screening Assay. Scrapie-infected neuro-
blastoma cells (ScN2a)12 were plated onto 96-well fluorescence
compatible cell culture plates (approximately 40 000 cells/well)
and incubated with compound for 5 days (approximately
150 000 cells/well). Compounds were initially screened at a
single concentration point (20 µM), and certain compounds were
subsequently screened over a concentration range, up to 25 µM.
Following the viability assay, the calcein reagent was removed
and intact ScN2a cells were treated with lysis buffer and
Benzonase to effect both cell lysis and concomitant poly-nucleic-
acid digestion. The cell lysates were then digested with
proteinase K (PK) to yield the protease-resistant PrP fragment,
PrP 27-30.13 The digested lysates were treated with sodium
phosphotungstic acid (PTA) and centrifuged to precipitate
selectively PrP 27-30.14 The pelleted fraction was denatured
and coated onto Immunlon II ELISA plates. PrP 27-30 was
detected using the anti-PrP antibody D1815 and goat anti-human
IgG Fab alkaline phosphatase conjugate. Untreated ScN2a cells,
no cells, and the antiprion compound quinacrine16 were used
as controls for the duplex assay. Dose-response curves over a
concentration range were determined in triplicate from three
independent experiments. Compound toxicity, as determined
from the cell viability component of the assay, was expressed
In Vitro Pharmacokinetics. A duplex solubility-perme-
ability assay was adapted from the procedure of Wexler et al.17
Compound solubility was categorized as follows: 0 µM <
“poor” < 200 µM; 200 µM < “medium” < 400 µM; “excellent”
> 400 µM (Table 1). Membrane permeability was determined
via a parallel artificial membrane permeability assay (PAMPA),18
using a dioleoylphosphocholine (DOPC) lipid membrane, and
log Pe was calculated using the equation of Wohnsland et al.19
(Table 1). Estradiol, furosemide, chlorpromazine, quanabenz,
methotrexate, carbamazepine, and famotidine were used as
controls for the solubility and permeability assays and gave the
expected relative pharmacokinetic profiles (Table 1).19
The duplex in vitro pharmacokinetic assay demonstrated that
halobenzene-substituted analogues (Table 1, e.g., 11 and 12),
had limited solubility (<400 µM). However, the relatively high
PAMPA permeability rates for compounds in this series
suggested that suitable serum concentrations may still be
attainable. Compounds substituted with tertiary amines at C-6
of the pyridine scaffold demonstrated the potential to partition
effectively into serum based on their solubility-permeability
results. Compounds 18, 19, 22, and 23 had solubilities in
aqueous buffer greater than 400 µM and PAMPA permeability
rates equivalent to or better than those of the controls (Table
1). Thienopyridine compounds 28 and 29 had low solubility
and readily precipitated in aqueous buffer and thus require
further medicinal chemistry to optimize the physicochemical
properties of this class.
Modeling Studies. The dominant-negative epitope on the
surface of PrP involving residues Q168, Q172, T215, and Q219
was modeled and represented as a pharmacophore model as
described previously (Figure 2A).8 Multiple conformers of
compounds 1 and 12 and the novel thienopyridine 28 were
generated with the programs CORINA,20 followed by OMEGA,21
using the MMFF94 force field, a maximum of 400 conforma-
tions with a minimum distance of 0.6 Å between conformations,
and an energy window of 20. The program GENX8 was used
to match the molecular conformers with the PrP pharmacophore

Prion Inhibition by 3,5-Dicarbonitrile-Based Compounds
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1 67
Table 1. Bioactivity and in Vitro Pharmacokinetic Properties of 2-Aminopyridine-3,5-dicarbonitrile Compounds
a A complete list of 2-aminopyridine-3,5-dicarbonitrile compounds synthesized and for which full dose-response curves were obtained is available as
Supporting Information online. b Compound concentration required to reduce PrPSc to 50% relative to untreated ScN2a cells. c Compound concentration
required to reduce the number of viable cells to 50% relative to untreated control ScN2a cells. d Compound solubility at 500 µM in aqueous 50 mM potassium
phosphate (monobasic), pH 6.5 buffer. Compound solubility was classified as follows: 0 µM < “poor” < 200 µM; 200 µM < “medium” < 400 µM;
“excellent” > 400 µM. e PAMPA permeability across a DOPC lipid membrane. f NA, no activity up to 25 µM. ND, not determined. Standard error determined
from three independent experiments.
model; a rms distance of less than 2 Å was classified as a match.
The resulting matched molecular conformers were visualized
with MOE22 to determine which conformers best satisfied the
electronic and steric constraints of the PrP pharmacophore model
(Figure 2B-D).
and cellular PrPSc, using the same compound-treated cell culture.
Quantification of toxicity was beneficial. First, toxic compounds
gave false positives in the quantification of PrPSc, and second,
toxicity data were helpful in selecting nontoxic compounds for
further study. Previously, we have shown that compound toxicity
toward ScN2a cells correlates well with toxicity toward human
kidney HEK92 and liver HEPG2 cells.25 The duplex screening
assay developed as part of this current work quantified cell
viability using the fluorescent probe calcein-AM26 and subse-
quently quantified PrP 27-30 by direct ELISA. Quantification
of compound toxicity and efficacy against PrPSc using the duplex
assay gave Z factors of 0.7-0.8 and 0.5-0.6, respectively, and
thus provided the necessary dynamic range and low deviation
for screening purposes.27 The ability to quantify both compound
toxicity and efficacy from the same cell culture and the
Discussion
Deriving a SAR for 2-aminopyridine-3,5-dicarbonitrile com-
pounds required the development of a screening assay that would
provide the necessary throughput and high-quality data needed
to identify trends over the compound library. Existing ap-
proaches to quantify compound bioactivity against cellular PrPSc
suffer from poor quantitation and/or an absence of a microtiter
plate format.16,23,24 We envisaged a duplex microtitre plate-based
screening assay that would quantify both compound toxicity

68 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1
May et al.
pounds (6) were more potent than the corresponding pyridine
compounds (5) (see Supporting Information). The improved
bioactivity of the thienopyridine scaffold may be attributed to
the additional rigidity provided by the bicyclic scaffold that went
some way toward preorganizing the C-6 substituent into a
relevant receptor-bound conformation.
To complement compound toxicity and efficacy data and to
improve the criteria by which bioactive 2-aminopyridine-3,5-
dicarbonitrile of compounds are selected for detailed in vivo
studies, we implemented secondary assays for compound
solubility and membrane permeability. These pharmacokinetic
parameters are important determinants of the effective free-
serum concentration of a compound following oral administra-
tion.28 In developing in vitro pharmacokinetic screens, we did
not aim to predict accurately partitioning of test compounds
but rather to guide the selection of candidate compounds for
further in vivo studies by providing relevant pharmacokinetic
properties. Ideally, a compound would have an in vitro aqueous
solubility greater than 400 µM and permeability constant greater
than -7.0, as determined by the duplex solubility-permeability
assay. Several new 2-aminopyridine-3,5-dicarbonitrile leads
satisfied these criteria (Table 1), and subsequent in vivo
pharmacokinetic studies are planned to validate the pharmaco-
kinetic profiles of these compounds.
Figure 2. (A) Dominant-negative epitope on the surface of PrPC,
involving residues 168, 172, 215, and 219. Also shown are the overlay
of the dominant-negative epitope (gray) and (B) the original 2-ami-
nopyridine-3,5-dicarbonitrile lead 1, (C) compound 12, and (D)
thienopyridine compound 28.
amenability to high-throughput screening are two unique benefits
of the current duplex screening assay.
Identifying relevant cellular targets for the 2-aminopyridine-
3,5-dicarbonitrile compounds would obviously aid the develop-
ment of this class as therapeutics for the treatment of prion
disease. Additionally, target identification and mechanistic
insights would be important to understanding the cellular
replication of PrPSc. The original lead 2-aminopyridine-3,5-
dicarbonitrile compound 1 was identified computationally as a
dominant-negative PrP mimetic. However, it remains to be
established whether compound 1 and related bioactive com-
pounds in the class inhibit prion replication in ScN2a cells by
binding to an unknown replication auxiliary “X” (Figure 1). A
retrospective study was used to determine if 2-aminopyridine-
3,5-dicarbonitrile compounds identified in the current work still
satisfied the original pharmacophore used to identify dominant-
negative mimetics. Computational modeling studies confirmed
that newly identified 2-aminopyridine-3,5-dicarbonitrile com-
pound 12 and the thienopyridine compound 28 spatially mimic
the dominant-negative epitope encoded on the surface of PrP
(Figure 2). However, additional studies are required to validate
experimentally the binding of certain 2-aminopyridine-3,5-
dicarbonitrile compounds to a macromolecule that participates
in prion replication and hence validate the proposed mechanism
of action of this class.
The original lead compound 1 (Table 1) was screened in the
duplex assay and was weakly bioactive, EC50 ≈ 80-100 µM.
The discrepancy between our findings and those originally
published for compound 1 (EC50 ) 20 µM)8 can be attributed
to the alternative cellular assays used to quantify bioactivity.
Whereas the current duplex assay quantifies changes in total
accumulated PrPSc, the previous screening assay quantified
changes in newly formed PrPSc. Screening against newly formed
PrPSc requires transient transfection of ScN2a cells, typically
with a mouse-hamster chimeric (MHM2) PrP.8 In this way,
newly formed PrPSc can be distinguished from pre-existing PrPSc
using antibodies directed against the hamster sequence of the
chimera. We chose to forego this approach, as transient
transfection of ScN2a cells suffers from variable efficiency and
reduces the overall robustness of the cellular screening assay.
Chemistries were identified and optimized that allowed for
the parallel solution synthesis of the 2-aminopyridine-3,5-
dicarbonitrile scaffold with diverse substituents at C-4 and C-6
of the pyridine heterocycle. In the first instance, aldehyde, aryl
halide, and alkyl halide chemsets were selected in such a way
that the resultant compound library encoded diverse chemical
space with C-4 and C-6 substituents. We did not computationally
filter target library components with the original pharmacophore
used to identify the 2-aminopyridine-3,5-dicarbonitrile scaffold.
Instead, traditional SAR was used to guide the focus of
subsequent smaller compound libraries such that promising
bioactive structural motifs at C-4 and C-6 could be explored in
greater detail. Generally, a variety of substituents were tolerated
at both the C-4 and C-6 positions of the pyridine heterocycle
(see Supporting Information). Compound 10, bearing a 4-bro-
mobenzene substituent at C-4, and compound 19, bearing a N,N-
diethylamine substituent at C-6, were identified as promising
leads from a larger diverse compound library. The SAR of
compounds 10 and 19 was further elaborated with smaller
focused compound libraries (Table 1) and identified compounds
active at low micromolar concentrations. Compounds in these
series would be candidates for subsequent medicinal chemistry
optimizations. Cyclization of certain pyridine compounds
yielded a novel class of bioactive thienopyridine compounds
(Scheme 2). In some instances, certain thienopyridine com-
The possibility exists that alternative mechanisms may be
responsible for the observed bioactivity of 2-aminopyridine-
3,5-dicarbonitrile compounds in a cellular context. Reddy et al.
have proposed that certain 2-aminopyridine-3,5-dicarbonitrile
compounds bind PrPC to exert their effect in scrapie-infected
cells.29 Reddy et al. docked a virtual library of 2-aminopyridine-
3,5-dicarbonitrile compounds against a putative binding pocket
on the surface of PrPC and demonstrated that several hits
interacted weakly with PrPC. However, the lead 2-aminopyri-
dine-3,5-dicarbonitrile compounds from that study were gener-
ally inactive in scrapie-infected cells.29 In comparison, we have
used bioactivity in ScN2a cells to optimize our lead 2-ami-
nopyridine-3,5-dicarbonitrile compounds for activity against
PrPSc. None of the lead 2-aminopyridine-3,5-dicarbonitrile
compounds described by Reddy et al. overlap with those of our
own work. To compare the relative bioactivity of known PrPC-
binding 2-aminopyridine-3,5-dicarbonitrile compounds with our

Prion Inhibition by 3,5-Dicarbonitrile-Based Compounds
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1 69
Table 2. Bioactivity of 2-Aminopyridine-3,5-dicarbonitrile Compound 9
against Human Adenosine Receptor Subtypes hA1, hA2a, and hA3
antagonists were inactive in ScN2a cells at 25 µM. It is noted
that human and mouse adenosine receptors share significant
sequence homology in the ligand-binding epitopes. These results
suggest that the antiprion activity of 2-aminopyridine-3,5-
dicarbonitrile compounds may not be attributed to adenosine
receptor binding. Additional studies are needed to elucidate the
mechanism of action of this class, potentially via suitably
functionalized chemical genetics probes based on the 2-amino-
pyridine-3,5-dicarbonitrile scaffold.
Conclusion
In conclusion, potent 2-aminopyridine-3,5-dicarbonitrile com-
pounds have been identified that inhibit accumulation of PrPSc
in a cell model of prion replication at low micromolar
concentrations. Current lead compounds in this series represent
a marked improvement in bioactivity over the original lead
compound. A detailed SAR for bioactive 2-aminopyridine-3,5-
dicarbonitrile compounds has been defined, which includes a
tertiary alkylamine of limited steric bulk at position C-6 and a
mono- or dihalobenzene substituent at position C-4 of the
pyridine scaffold. A novel bicyclic thienopyridine scaffold was
identified as being effective against prion replication in culture.
A qualitative assessment of lead compounds from this series
has identified candidate compounds for further in vivo phar-
macokinetic and efficacy evaluation in animal models of prion
disease. Additionally, these current leads are of suitable potency
to form the basis of chemical genetics probes with which to
identify targets that may be implicated in prion replication.
a Inhibition of human adenosine receptor subtypes hA1, hA2a, and hA3
at a ligand concentration of 5 µM, relative to the known selective inhibitors
also assayed at 5 µM: DPCPX (8-cyclopentyl-1,3-dipropylxanthine, 100%
inhibition hA1), NECA (adenosine-5′-N-ethylcarboxamide, 100% inhibition
hA2a), or IB-MECA (N-6-(3-iodobenzyl)adenosine-5′-N-methyluronamide,
100% inhibition hA3). b Bioactivity in ScN2a cells. Standard error deter-
mined from three independent experiments. c Inhibitor dissociation constant.
ND, not determined.
Experimental Section
Chemistry and Biology. General Methods. Malononitrile,
â-alanine, and 2-cyanothioacetamide were purchased from Sigma-
Aldrich Chemical Co. All aldehydes, aryl halides, and alkyl halides
were purchased from either Sigma-Aldrich Chemical Co. or
Maybridge Chemical Co. All solvents were obtained commercially
in the highest purity (Sigma-Aldrich Chemical Co.) and used
without further purification. Compound libraries were synthesized
in parallel using a Bohdan 48-well, temperature-controlled miniblock
fitted with a water-cooled condenser. Solvent was removed from
individual reaction vessels using a Genevac HT-4 speedvac. Crude
reactions were purified using a Parallex Flex (Biotage) parallel
preparative HPLC instrument fitted with Xterra C-18 columns
(Waters) and automated fraction collector. Solvent was removed
own lead compounds, we independently synthesized the sole
bioactive 2-aminopyridine-3,5-dicarbonitrile compound identi-
fied by Reddy et al., 27. We determined that compound 27 was
inactive up to 25 µM in our duplex ScN2a cell assay. These
results suggest that while certain 2-aminopyridine-3,5-dicarbo-
nitrile compounds may interact weakly with PrPC via a suitable
binding pocket, alternative mechanisms leading to potentially
more potent inhibition of PrPSc replication may be operating in
cellular models of prion replication.
1
Previously, compounds bearing the 2-aminopyridine-3,5-
dicarbonitrile scaffold have been shown to be potent antagonists
of adenosine receptors.30,31 For example, compounds 30 and
31 (Table 2) bind human adenosine receptor subtypes with
inhibitory constants in the nanomolar concentration range (for
30, Ki(hA1) ) 2.4 ( 1.0 nM, Ki(hA2a) ) 28 ( 4 nM, Ki(hA3)
) 171 ( 109 nM; for 31, Ki(hA1) ) 7.0 ( 0.8 nM, Ki(hA2a) )
214 ( 37 nM, Ki(hA3) ) 24 ( 7.6 nM).30 To investigate
whether 2-aminopyridine-3,5-dicarbonitrile compounds elicit
their antiprion effects in ScN2a cells via adenosine receptors,
we determined whether the representative compound 9 was an
adenosine receptor antagonist. Compound 9 was active at 5 µM
against human hA1, hA2a, and hA3 receptors and had an
inhibitory constant of Ki ) 590 nM against hA1 (Table 2).
However, compound 9 was approximately 100- to 300-fold less
potent against the adenosine receptor subtypes than the two
known 2-aminopyridine-3,5-dicarbonitrile compounds 30 and
31 (Table 2).30 Additionally, we synthesized compounds 30 and
31 and assayed these 2-aminopyridine-3,5-dicarbonitrile com-
pounds, along with the structurally unrelated and potent ad-
enosine antagonists 8-cyclopentyl-1,3-dipropylxanthine (DPCPX)32
and N-6-(3-iodobenzyl)-adenosine-5′-N-methyluronamide (IB-
MECA),33 in ScN2a cells. These potent adenosine receptor
from fractions using a custom-built Genevac megaevaporator. H
and 13C NMR data were recorded on a Varian 400 MHz spectrom-
eter in either DMSO-d6 or CD3OD/acetic acid-d4. LCMS was
conducted on a Waters Micromass ZQ in ESI+ mode, equipped
with a Waters 2487 dual wavelength absorbance detector and
Waters Alliance 2695 separations module, with the sample eluting
through an analytical Xterra C-18 column. Compound purity was
determined by LCMS using either method A (Waters Xterra C-18
column (2.1 mm × 50 mm, 3.5 µm) eluting water (0.05%
trifluroacetic acid)/methanol (0.05% trifluroacetic acid), flow rate
0.2 mLmin-1, run time of 30 min) or method B (Waters Xterra
phenyl column (2.1 mm × 50 mm, 3.5 µm) eluting water (0.05%
trifluroacetic acid)/acetonitrile (0.05% trifluroacetic acid), flow rate
0.3 mLmin-1, run time of 20 min). Purity was determined from
integrating peak areas of the liquid chromatogram, monitored at
254 nm. High-resolution mass spectrometry (HRMS) data were
obtained by the Vincent Coates Foundation Mass Spectrometry
Laboratory, Stanford University Mass Spectrometry. Compound
structures, associated spectral data, and biological activity were
stored as a chemical database using ISIS/Base (MDL, San Ramon).
Sodium phosphotungstic acid (PTA), guanidine thiocyanate, di-
ethanolamine, phenylmethylsulfonyl fluoride (PMSF), p-nitrophenyl
phosphate tablets, and bovine serum albumin (BSA) were obtained
from Sigma-Aldrich Chemical Co. Minimal essential media (MEM)
with Earles salts, phosphate-buffered saline (PBS) without Ca2+

70 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1
May et al.
and Mg2+, trypsin-EDTA, and GlutaMax were purchased from
Gibco. Fetal bovine sera (FBS) were obtained from HyClone, and
PK was from Invitrogen. Benzonase nuclease was obtained from
Novagen. Anti-PrP Fab antibody D18 was prepared in-house, and
goat antihuman IgG Fab alkaline phosphatase conjugate was
purchased from Fisher Scientific. Fluorescence intensity was
determined using a Tecan Genios fluorescence microplate reader
running XFluor4 software. Absorbance was determined using a
SpectraMax Plus microplate reader running SoftMaxPro. ELISA
plates were washed and aspirated using a Bio-Tek Elx405 plate
washer. Plate-to-plate transfers and liquid dispensing were done
using a Labctye S4 liquid handler fitted with a fixed 96-well head.
Adenosine receptor binding studies were conducted by Cerep
(Poitiers, France) using established methods.34
J ) 1.2, 7.2 Hz, 1H), 7.39 (dd, J ) 1.6, 7.6 Hz, 1H), 3.60 (m, 2H),
3.48 (m, 2H), 3.34 (q, J ) 7.2 Hz, 4H), 1.35 (t, J ) 7.2 Hz, 6H);
LCMS (ESI) m/z 386 (MH+); HRMS m/z calculated for C19H21N5-
ClS (MH+) 386.1206, found 386.1209; purity method A, 95%,
method B, 95%.
2-Amino-4-(4-bromophenyl)-6-(2-(diethylamino)ethylthio)py-
1
ridine-3,5-dicarbonitrile (10). H NMR (CD3OD/CD3CO2D) δ
7.74 (d, J ) 8.8 Hz, 2H), 7.44 (d, J ) 8.4 Hz, 2H), 3.60 (m, 2H),
3.48 (m, 2H), 3.34 (q, J ) 7.2 Hz, 4H), 1.35 (t, J ) 7.2 Hz, 6 H);
LCMS (ESI) m/z 432 (MH+); HRMS m/z calculated for C19H21N5-
SBr (MH+) 430.0701, found 430.0702; purity method A, 97%,
method B, 95%.
2-Amino-4-(3-bromophenyl)-6-(2-(diethylamino)ethylthio)py-
1
ridine-3,5-dicarbonitrile (11). H NMR (CD3OD/CD3CO2D) δ
General Procedure A. Synthesis of 2-Aminopyridine-3,5-
dicarbonitrile Compounds. Starting 2-(arylidene)malononitriles
(2) were prepared from commercially available arylaldehydes using
the procedure of Gazit et al. Briefly, a solution of the arylaldehyde
(1 mmol), malononitrile (3 mmol), and a catalytic amount of
â-alanine (approximately 0.4 mmol) in ethanol (30 mL) was stirred
at room temperature for 72 h. The mixture was cooled on ice, and
H2O was added (10 mL) to precipitate the product. In cases for
which the product remained soluble, solvent volume was reduced
to half under vacuum, and the product precipitated on cooling. The
product was removed by filtration, washed with ice-cold H2O (5
mL), ice-cold ethanol (5 mL), and hexanes (5 mL) and dried under
high vacuum. 2-(Arylidene)malononitriles were obtained in 60-
95% yields, with purity of >90% as determined by LCMS, and
were used without further purification.
To each well of a Bohdan miniblock, 2-(arylidene)malononitrile
(2, 0.5 mmol), piperidine (0.75 mmol), and 2-cyanothioacetamide
(0.5 mmol) in ethanol (3 mL) were added. The mixture was heated
at 80 °C for 5 h with stirring. Solvent and piperidine were removed
under vacuum centrifugation to yield the crude thione intermediates
(3). The crude intermediates were resuspended in DMF (0.2 mL)
and stirred with 10% KOH (1 equiv, 0.28 mL) at room temperature
for 1 min. Aryl or alkyl halides (0.5 mmol) were added, and the
mixture was stirred at room temperature for 5 h. Reaction vessels
were again transferred to a Genevac HT4 speedvac, and solvent
was removed. The crude products were resuspended in DMSO and
purified using a Parallex Flex (Biotage) parallel preparative HPLC
instrument fitted with Xterra C-18 columns (Waters). A solvent
gradient of 10-100% methanol/20 mM ammonium acetate and a
flow rate of 20 mL min-1 were employed. Injections were
monitored using a UV absorbance at 254 nm, and fractions were
collected using an automated fraction collector. Solvent was
removed from all fractions using a Genevac megaevaporator.
Aliquots of each fraction were analyzed by LCMS to determine
purity and compound identity. Fractions were combined on this
basis and solvent removed under reduced pressure to yield the target
compounds as solids. All final compounds were characterized by
1H NMR and LCMS. Yields for the library ranged from ap-
proximately 10% to 70%. Compounds were dissolved in DMSO
at a stock concentration of 10 mM and stored at -20 °C.
2-Amino-4-(4-chlorophenyl)-6-(2-(diethylamino)ethylthio)py-
ridine-3,5-dicarbonitrile (7). 1H NMR (CD3OD/CD3CO2D) δ 7.57
(d, J ) 8.4 Hz, 2H), 7.51 (d, J ) 8.8 Hz, 2H), 3.60 (m, 2H), 3.48
(m, 2H), 3.34 (q, J ) 7.2 Hz, 4H), 1.35 (t, J ) 7.2 Hz, 6H); LCMS
(ESI) m/z 386 (MH+); HRMS m/z calculated for C19H21N5ClS
(MH+) 386.1206, found 386.1202; purity method A, 98%, method
B, 95%.
7.75 (td, J ) 2.0 Hz, 4.8 Hz, 1H), 7.69 (d, J ) 2.8 Hz, 1H), 7.49
(d, J ) 5.2 Hz, 2H), 3.60 (m, 2H), 3.48 (m, 2H), 3.34 (q, J ) 7.2
Hz, 4H), 1.35 (t, J ) 7.2 Hz, 6H); LCMS (ESI) m/z 432 (MH+);
HRMS m/z calculated for C19H21N5SBr (MH+) 430.0701, found
430.0697; purity method A, 97%, method B, 95%.
2-(2-(Diethylamino)ethylthio)-6-amino-4-(3,5-dichlorophenyl)-
pyridine-3,5-dicarbonitrile (12). 1H NMR (CD3OD/CD3CO2D) δ
7.69 (t, J ) 2.0 Hz, 1H), 7.52 (d, J ) 2.0 Hz, 2H), 3.60 (m, 2H),
3.48 (m, 2 H), 3.34 (q, J ) 7.2 Hz, 4H), 1.35 (t, J ) 7.2 Hz, 6H);
LCMS (ESI) m/z 421 (MH+); HRMS m/z calculated for C19H20N5-
Cl2S (MH+) 420.0816, found 420.0808; purity method A, 95%,
method B, 95%.
2-(2-(Diethylamino)ethylthio)-6-amino-4-(2,3-dichlorophenyl)-
pyridine-3,5-dicarbonitrile (13). 1H NMR (CD3OD/CD3CO2D) δ
7.76 (dd, J ) 1.6, 8.0 Hz, 1H), 7.51 (t, J ) 8.0 Hz, 1H), 7.36 (dd,
J ) 1.2, 7.6 Hz, 1H), 3.60 (m, 2H), 3.48 (m, 2H), 3.34 (q, J ) 7.2
Hz, 4H), 1.35 (t, J ) 7.2 Hz, 6H); LCMS (ESI) m/z 421 (MH+);
HRMS m/z calculated for C19H20N5Cl2S (MH+) 420.0816, found
420.0807; purity method A, 93%, method B, 95%.
2-(2-(Diethylamino)ethylthio)-6-amino-4-(3,4-dichlorophenyl)-
pyridine-3,5-dicarbonitrile (14). 1H NMR (CD3OD/CD3CO2D) δ
7.74 (d, J ) 6.0 Hz, 1H), 7.73 (s, 1H), 7.45 (dd, J ) 2.4, 8.4 Hz,
1H), 3.60 (m, 2H), 3.48 (m, 2H), 3.34 (q, J ) 7.2 Hz, 4H), 1.35 (t,
J ) 7.2 Hz, 6H); LCMS (ESI) m/z 421 (MH+); HRMS m/z
calculated for C19H20N5Cl2S (MH+) 420.0816, found 420.0809;
purity method A, 88%, method B, 100%.
2-(6-Amino-3,5-dicyano-4-(furan-2-yl)pyridine-2-ylthio)aceta-
1
mide (15). H NMR (DMSO-d6) δ 8.01 (s, 1H), 7.48 (bs, 2H),
7.39 (d, J ) 3.2 Hz, 1H), 7.23 (bs, 2H), 6.83 (s, 1H), 3.85 (s, 2H);
LCMS (ESI) m/z 300 (MH+); HRMS m/z calculated for C13H9N5O2-
NaSS (MNa+) 322.0375, found 322.0379; purity method A, 94%,
method B, 100%.
2-Amino-6-(3-aminopropylsulfanyl)-4-furan-2-ylpyridine-3,5-
dicarbonitrile (16). 1H NMR (CD3OD/CD3CO2D) δ 7.86 (d, J )
1.2 Hz, 1H), 7.47 (d, J ) 3.2 Hz, 1H), 6.74 (dd, J ) 1.6, 3.6 Hz,
1H), 3.34 (t, J ) 7.2 Hz, 2H), 3.07 (t, J ) 7.6 Hz, 2H), 2.09 (quintet,
J ) 7.6 Hz, 2H); 13C NMR (DMSO-d6) 167.8, 160.1, 146.4, 145.1,
143.7, 116.2, 115.8, 115.7, 112.8, 89.7, 81.3, 31.6, 31.5, 27.3;
LCMS (ESI) m/z 300 (MH+); HRMS m/z calculated for C14H13N5-
CONaS (MNa+) 322.0739, found 322.0733; purity method A, 69%,
method B, 70%.
2-(2-(Dimethylamino)ethylthio)-6-amino-4-(furan-2-yl)pyri-
dine-3,5-dicarbonitrile (17). 1H NMR (CD3OD/CD3CO2D) δ 7.88
(d, J ) 1.6 Hz, 1H), 7.51 (d, J ) 3.6 Hz, 1H), 6.75 (dd, J ) 1.6,
3.6 Hz, 1H), 3.61 (m, 2H), 3.51 (m, 2H), 2.98 (s, 6H); LCMS (ESI)
m/z 314 (MH+); HRMS m/z calculated for C15H15N5ONaS (MNa+)
336.0895, found 336.0901; purity method A, 97%, method B, 95%.
2-(3-(Dimethylamino)propylthio)-6-amino-4-(furan-2-yl)py-
2-Amino-4-(3-chlorophenyl)-6-(2-(diethylamino)ethylthio)py-
ridine-3,5-dicarbonitrile (8). 1H NMR (CD3OD/CD3CO2D) δ 7.58
(d, J ) 6.4 Hz, 1H), 7.55 (m, 2H), 7.44 (dd, J ) 1.6, 7.2 Hz, 1H),
3.60 (m, 2H), 3.48 (m, 2H), 3.34 (q, J ) 7.2 Hz, 4H), 1.35 (t, J )
7.2 Hz, 6H); LCMS (ESI) m/z 386 (MH+); HRMS m/z calculated
for C19H21N5ClS (MH+) 386.1206, found 386.1199; purity method
A, 93%, method B, 100%.
1
ridine-3,5-dicarbonitrile (18). H NMR (DMSO-d6) δ 8.10 (d, J
) 1.6 Hz, 1H), 7.39 (d, J ) 3.6 Hz, 1H), 6.84 (dd, J ) 2.0, 3.6
Hz, 1H), 3.22 (t, J ) 6.8 Hz, 2H), 3.14 (m, 2H), 3.09 (m, 2H),
2.76 (s, 6H); LCMS (ESI) m/z 328 (MH+); HRMS m/z calculated
for C16H17N5ONaS (MNa+) 350.1052, found 350.1045; purity
method A, 90%, method B, 100%.
2-Amino-4-(2-chlorophenyl)-6-(2-(diethylamino)ethylthio)py-
ridine-3,5-dicarbonitrile (9). 1H NMR (CD3OD/CD3CO2D) δ 7.62
(dd, J ) 1.2, 8.0 Hz, 1H), 7.56 (td, J ) 1.6, 8.0 Hz, 1H), 7.50 (td,
2-Amino-6-(2-diethylaminoethylsulfanyl)-4-furan-2-ylpyridine-
3,5-dicarbonitrile (19). H NMR (CD3OD/CD3CO2D) δ 7.88 (d,
J ) 1.2 Hz, 1H), 7.51 (d, J ) 3.6 Hz, 1H), 6.75 (dd, J ) 1.6, 3.6
1

Prion Inhibition by 3,5-Dicarbonitrile-Based Compounds
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1 71
Hz, 1H), 3.60 (m, 2H), 3.48 (m, 2H), 3.34 (q, J ) 7.2 Hz, 4H),
1.35 (t, J ) 7.2 Hz, 6H); LCMS (ESI) m/z 342 (MH+); HRMS m/z
calculated for C17H20N5OS (MH+) 342.1389, found 342.1375; purity
method A, 99%, method B, 100%.
2-Amino-6-(2-(diisopropylamino)ethylthio)-4-furan-2-ylpyri-
dine-3,5-dicarbonitrile (20). 1H NMR (CD3OD/CD3CO2D) δ 7.88
(d, J ) 1.2 Hz, 1H), 7.51 (d, J ) 3.6 Hz, 1H), 6.75 (dd, J ) 2.0,
3.6 Hz, 1 H), 3.83 (quintet, J ) 6.8 Hz, 2H), 3.61 (m, 2H), 3.44
(m, 2H), 1.42 (d, J ) 6.8 Hz, 12H); LCMS (ESI) m/z 370 (MH+);
HRMS m/z calculated for C19H24N5OS (MH+) 370.1702, found
370.1701; purity method A, 95%, method B, 100%.
(bs, 1H), 7.18 (d, J ) 3.2 Hz, 1H), 6.96 (d, J ) 8.4 Hz, 1H), 6.86
(dd, J ) 1.6, 3.6 Hz, 1H), 4.30 (d, J ) 4.8 Hz, 4H); LCMS (ESI)
m/z 419 (MH+); HRMS m/z calculated for C21H14N4O4NaS (MNa+)
441.0633, found 441.0631; purity method A, 74%, method B, 95%.
Duplex Cellular Toxicity and Antiprion Assay. ScN2a cells
were maintained in filter-sterilized (0.2 mm) MEM, supplemented
with FBS and GlutaMax. On day 1, media was aspirated from a
confluent 100 mm plate of ScN2a cells and cells were detached by
addition of 0.05% trypsin-EDTA (1 mL). MEM media was added
(approximately 10 mL), and the cell concentration was determined
using Packed Cell Volume tubes (TPP). The ScN2a cell concentra-
tion was adjusted to 4 × 105 cells mL-1 with MEM media. A 96-
well, tissue-culture-treated, clear-bottom, black plate (Greiner Bio-
One) wetted with MEM media (90 µL) was incubated at 37 °C
prior to use. To this plate, 100 µL of the ScN2a cell suspension
was transferred, and the cells were allowed to settle for 2 h prior
to addition of the test compound. Compound libraries were prepared
in 100% DMSO at the required concentrations in a 96-well format
and then diluted 1/20 with sterile PBS prior to use. Compounds (10
µL) were transferred to the 96-well cell culture plate, and the plates
were incubated at 37 °C. Final DMSO concentrations did not exceed
0.25% (v/v). Media were aspirated on day 5, and cells were washed
twice with PBS (200 µL). Calcein-AM (100 µL, 2.5 µg mL-1
solution in PBS) was added, and the plates were incubated at
37 °C for 25 min. Fluorescent emission intensity was quantified
using a fluorescence plate reader, ex/em ) 492 nm/525 nm.
The calcein-AM solution was aspirated, 50 µL of lysis buffer
(10 mM Tris-HCl, pH 8.0; 150 mM NaCl; 0.5% sodium deoxy-
cholate; 0.5% Nonidet P-40) was added, and then plates were
shaken for 5 min at room temperature. Benzonase nuclease (25
µL, 7.5 U/mL in 50 mM Tris-HCl, pH 8.0; 20 mM NaCl; 2 mM
MgCl2) was added, and plates were incubated at 37 °C for 15 min
or until the DNA pellet was no longer visible. Proteinase K (25
µL, 25 µg mL-1 solution in lysis buffer) was added, and the plates
were incubated at 37 °C for 1 h. Proteolysis was inhibited by the
addition of PMSF (10 µL, 20 mM solution in ethanol), with a 10
min incubation at room temperature. PK-digested PrPSc was
precipitated by the addition of PTA (110 µL, 7.5% solution in 64.75
mM MgCl2, pH 7.4). Plates were incubated overnight at 37 °C and
then centrifuged at 2200 rpm for 60 min using a Beckman-Coulter
GS-6R centrifuge. The supernatant was aspirated. The PTA-
precipitated protein pellet was denatured with 6 M guanidine
thiocyanate (55 µL) in coating buffer (55 µL, 0.1 M sodium
bicarbonate, pH 8.6) for 5 min at room temperature with shaking.
The suspension (100 µL) was transferred to Immunlon II ELISA
plates (Fisher Scientific), sealed, and incubated overnight at 4 °C.
ELISA plates were washed twice with TBST (20 mM Tris-HCl,
137 mM NaCl, 0.05% Tween-20, pH 7.5), blocked with 200 µL of
3% BSA as a solution in TBS (20 mM TrisHCl, 137 mM NaCl,
pH 7.5), sealed, and incubated at 37 °C. After 1 h, the ELISA plates
were washed twice with TBST and incubated with 100 µL of anti-
PrP antibody D18 (1 µg mL-1) as a solution in 1% BSA/TBS.
ELISA plates were sealed and incubated at 37 °C for 2 h, then
washed seven times with TBST. Goat antihuman IgG Fab AP
conjugate (100 µL) diluted 1:2500 with 1% BSA/TBS was added
to the plates, sealed, and incubated at 37 °C for 1 h. Plates were
washed seven times with TBST. Plates were developed with
p-nitrophenyl phosphate (100 µL, 1 mg mL-1) as a solution in
alkaline phosphatase buffer (1 M diethanolamine, 0.5 mM
MgCl2‚6H20, pH 9.8). Absorbance at 405 nm was measured using
a microplate reader. Dose-response curves and EC50 values were
computed using SigmaPlot.
2-(2-(Pyrrolidin-1-yl)ethylthio)-6-amino-4-(furan-2-yl)pyridine-
1
3,5-dicarbonitrile (21). H NMR (CD3OD/CD3CO2D) δ 7.88 (s,
1H), 7.50 (d, J ) 3.6 Hz, 1H), 6.75 (dd, J ) 1.6, 3.6 Hz, 1H), 3.58
(m, 6H), 3.48 (m, 2H), 2.12 (m, 4H); LCMS (ESI) m/z 340 (MH+);
HRMS m/z calculated for C17H18N5OS (MH+) 340.1232, found
340.1244; purity method A, 97%, method B, 100%.
2-(2-(Piperidin-1-yl)ethylthio)-6-amino-4-(furan-2-yl)pyridine-
1
3,5-dicarbonitrile (22). H NMR (CD3OD/CD3CO2D) δ 7.88 (d,
J ) 2.0 Hz, 1 H), 7.50 (d, J ) 3.6 Hz, 1H), 6.75 (dd, J ) 1.6, 3.6
Hz, 1H), 3.65 (m, 2H), 3.59 (m, 2H), 3.45 (m, 2H), 3.04 (m, 2H),
1.95 (m, 2H), 1.83 (m, 3H), 1.55 (m, 1H); LCMS (ESI) m/z 354
(MH+); HRMS m/z calculated for C18H20N5OS (MH+) 354.1389,
found 354.1381; purity method A, 95%, method B, 100%.
2-(3-(Piperidin-1-yl)propylthio)-6-amino-4-(furan-2-yl)pyri-
dine-3,5-dicarbonitrile (23). 1H NMR (DMSO-d6) δ 8.10 (d, J )
1.2 Hz, 1H), 7.39 (d, J ) 3.6 Hz, 1H), 6.83 (dd, J ) 1.6, 3.6 Hz,
1H), 3.43 (m, 2H), 3.23 (t, J ) 7.2 Hz, 2H), 3.14 (m, 2H), 2.85
(m, 2H), 2.04 (m, 2H), 1.80 (m, 2H), 1.62 (m, 3H), 1.35 (m, 1H);
LCMS (ESI) m/z 368 (MH+); HRMS m/z calculated for C19H22N5-
OS (MH+) 368.1545, found 368.1537; purity method A, 93%,
method B, 100%.
2-(2-Morpholinoethylthio)-6-amino-4-(furan-2-yl)pyridine-
1
3,5-dicarbonitrile (24). H NMR (CD3OD/CD3CO2D) δ 7.88 (d,
J ) 1.2 Hz, 1H), 7.50 (d, J ) 3.2 Hz, 1H), 6.75 (dd, J ) 2.0, 3.6
Hz, 1H), 3.92 (m, 4H), 3.59 (m, 2H), 3.46 (m, 2H), 3.35 (m, 4H);
LCMS (ESI) m/z 356 (MH+); HRMS m/z calculated for C17H17N5O2-
NaS (MNa+) 378.1001, found 378.1002; purity method A, 87%,
method B, 100%.
2-(2-(Dibutylamino)ethylthio)-6-amino-4-(furan-2-yl)pyridine-
1
3,5-dicarbonitrile (25). H NMR (CD3OD/CD3CO2D) δ 7.88 (s,
1H), 7.50 (d, J ) 3.6 Hz, 1H), 6.75 (dd, J ) 1.6, 3.6 Hz, 1H), 3.61
(m, 2H), 3.52 (m, 2H), 3.25 (m, 4H), 1.72 (m, 4H), 1.45 (m, 4H),
1.01 (t, J ) 7.2 Hz, 6H); LCMS (ESI) m/z 398 (MH+); HRMS m/z
calculated for C21H28N5OS (MH+) 398.2015, found 398.2011; purity
method A, 95%, method B, 100%.
2-(2-(N-Benzyl-N-methylamino)ethylthio)-6-amino-4-(furan-
2-yl)pyridine-3,5-dicarbonitrile (26). 1H NMR (CD3OD/CD3-
CO2D) δ 7.89 (d, J ) 1.6 Hz, 1H), 7.50 (d, J ) 3.6 Hz, 1H), 7.46
(m, 5H), 6.76 (dd, J ) 1.6, 3.6 Hz, 1H), 4.41 (s, 2H), 3.60 (m,
2H), 3.51 (m, 2H), 2.97 (s, 3H); LCMS (ESI) m/z 390 (MH+);
HRMS m/z calculated for C21H20N5OS (MH+) 390.1389, found
390.1385; purity method A, 89%, method B, 100%.
General Procedure B. Synthesis of Thienopyridine Com-
pounds. Thienopyridines 6 were prepared according to the proce-
dure of Dyachenko et al. Briefly, pyridine compounds 5 were treated
with 10% aqueous KOH (1 equiv) in DMF (0.4 mL) for 5 h at the
room temperature. The product was precipitated by the addition of
H2O (0.4 mL) and collected by filtration, washed with ice-cold H2O
(5 mL), ice-cold ethanol (5 mL), and hexanes (5 mL) and dried
under high vacuum. Compounds were obtained in 60-80% yields.
3,6-Diamino-4-furan-2-yl-2-(2-nitrobenzoyl)thieno[2,3-b]py-
1
ridine-5-carbonitrile (28). H NMR (DMSO-d6) δ 8.17 (d, J )
Duplex Solubility and PAMPA Assay. Compound stocks at
10 mM were serially diluted over a concentration range of 500-
3.13 µM in DMSO (final volume 200 µL) in a UV-compatible,
96-well plate (Greiner Bio-One), in triplicate. Absorbance at 320
nm was quantified using a SpectraMax microplate reader and a
standard curve was derived such that r2 > 0.85. Standard curves
for control compounds were determined at an appropriate absorption
maxima. A membrane-embedded solubility plate (Millipore) was
wetted with previously filtered donor buffer (275 µL, 50 mM
8.0 Hz, 1H), 8.12 (bs, 1H), 7.90 (t, J ) 7.2 Hz, 1H), 7.78 (d, J )
8.0 Hz, 1H), 7.75 (bs, 1H), 7.70 (d, J ) 7.2 Hz, 1H), 7.22 (bs,1H),
6.88 (dd, J ) 1.6, 3.6 Hz, 1H); LCMS (ESI) m/z 406 (MH+);
HRMS m/z calculated for C19H12N5O4S (MH+) 406.0610, found
406.0602; purity method A, 69%, method B, 95%.
3,6-Diamino-4-furan-2-yl-2-(2,3-dihydrobenzo[b][1,4]dioxan-
2-oyl)thieno[2,3-b]pyridine-5-carbonitrile (29). 1H NMR (DMSO-
d6) δ 8.10 (s, 1H), 7.66 (bs, 1H), 7.24 (d, J ) 8.4 Hz, 1H), 7.19

72 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1
May et al.
(7) Perrier, V.; Kaneko, K.; Safar, J.; Vergara, J.; Tremblay, P.;
DeArmond, S. J.; Cohen, F. E.; Prusiner, S. B.; Wallace, A. C.
Dominant-negative inhibition of prion replication in transgenic mice.
Proc. Natl. Acad. Sci. U.S.A. 2002, 99 (20), 13079-13084.
(8) Perrier, V.; Wallace, A. C.; Kaneko, K.; Safar, J.; Prusiner, S. B.;
Cohen, F. E. Mimicking dominant negative inhibition of prion
replication through structure-based drug design. Proc. Natl. Acad.
Sci. U.S.A. 2000, 97 (11), 6073-6078.
(9) Gazit, A.; Osherov, N.; Posner, I.; Yaish, P.; Poradosu, E.; Gilon,
C.; Levitzki, A. Tyrphostins. 2. Heterocyclic and alpha-substituted
benzylidenemalononitrile tyrphostins as potent inhibitors of EGF
receptor and ErbB2/neu tyrosine kinases. J. Med. Chem. 1991, 34
(6), 1896-1907.
(10) Dyachenko, V. D.; Litvinov, V. P. Synthesis and recyclization of
4-aryl-2,6-diamino-3,5-dicyano-4H-thiopyrans. Russ. J. Org. Chem.
1998, 34 (4), 557-563.
(11) Kambe, S.; Saito, K.; Sakurai, A.; Midorikawa, H. Synthetic studies
using alpha,beta-unsaturated nitriles: facile synthesis of pyridine
derivatives. Synthesis 1981, 531-532.
(12) Butler, D. A.; Scott, M. R.; Bockman, J. M.; Borchelt, D. R.;
Taraboulos, A.; Hsiao, K. K.; Kingsbury, D. T.; Prusiner, S. B.
Scrapie-infected murine neuroblastoma cells produce protease-
resistant prion proteins. J. Virol. 1988, 62 (5), 1558-1564.
(13) Oesch, B.; Westaway, D.; Walchli, M.; McKinley, M. P.; Kent, S.
B.; Aebersold, R.; Barry, R. A.; Tempst, P.; Teplow, D. B.; Hood,
L. E.; Prusiner, S. B.; Weissmann, C. A cellular gene encodes scrapie
PrP 27-30 protein. Cell 1985, 40 (4), 735-746.
potassium phosphate (monobasic), pH 6.5). To the solubility plate,
15 µL of test compound at 10 mM in DMSO was added, and the
plate was shaken at room temperature for 90 min. Vacuum was
applied to the underside of the membrane plate using a vacuum
manifold (Millipore), and the filtrate was collected in a 96-well
receiver plate. Filtrate (200 µL) was transferred to a UV-compatible,
96-well plate (Greiner Bio-One). Absorbance at 320 nm was
measured {Amax(filtrate)}, and the concentration was derived from
the linear slope of the standard curve. The filtrate was retained for
the secondary permeability screen. Compound solubility was
categorized as follows: 0 µM < “poor” < 200 µM; 200 µM <
“medium” < 400 µM; “excellent” > 400 µM.
For the permeability screen, a PAMPA sandwich was prepared
by transferring filtered (0.2 mm) acceptor buffer (300 µL, 50 mM
potassium phosphate (monobasic), 2% DMSO, pH 7.4) to a Teflon
acceptor plate (Millipore). A PAMPA donor plate (Millipore) was
coated with 4 µL of DOPC lipid, as a solution in dodecane, with
1% BHT additive (Avanti Polar Lipids). Filtrate from the solubility
assay (150 µL) was carefully added to the donor plate, and the
PAMPA sandwich was assembled. The sandwich was incubated
at room temperature in the dark for 15 h and then disassembled.
Acceptor buffer (200 µL) was transferred from the Teflon acceptor
block to a UV-compatible, 96-well plate (Greiner Bio-One), and
absorbance was measured at 320 nm {Amax(acceptor)}. Permeability
(log Pe) was calculated on the basis of the formula of Wohnsland
et al.:
(14) Lee, I. S.; Long, J. R.; Prusiner, S. B.; Safar, J. G. Selective
precipitation of prions by polyoxometalate complexes. J. Am. Chem.
Soc. 2005, 127 (40), 13802-13803.
(15) Williamson, R. A.; Peretz, D.; Pinilla, C.; Ball, H.; Bastidas, R. B.;
Rozenshteyn, R.; Houghten, R. A.; Prusiner, S. B.; Burton, D. R.
Mapping the prion protein using recombinant antibodies. J. Virol.
1998, 72 (11), 9413-9418.
VDVA
C )
(area)(time)
(16) Korth, C.; May, B. C.; Cohen, F. E.; Prusiner, S. B. Acridine and
phenothiazine derivatives as pharmacotherapeutics for prion disease.
Proc. Natl. Acad. Sci. U.S.A. 2001, 98 (17), 9836-9841.
(17) Wexler, D. S.; Gao, L.; Anderson, F.; Ow, A.; Nadasdi, L.;
McAlorum, A.; Urfer, R.; Huang, S. G. Linking solubility and
permeability assays for maximum throughput and reproducibility. J.
Biomol. Screening 2005, 10 (4), 383-390.
(18) Kansy, M.; Senner, F.; Gubernator, K. Physicochemical high
throughput screening: parallel artificial membrane permeation assay
in the description of passive absorption processes. J. Med. Chem.
1998, 41 (7), 1007-1010.
(19) Wohnsland, F.; Faller, B. High-throughput permeability pH profile
and high-throughput alkane/water log P with artificial membranes.
J. Med. Chem. 2001, 44 (6), 923-930.
(20) Aires-de-Sousa, J.; Gasteiger, J.; Gutman, I.; Vidovic, D. Chirality
codes and molecular structure. J. Chem. Inf. Comput. Sci. 2004, 44
(3), 831-836.
A
max(acceptor)
log Pe ) log C -ln 1 -
{
(
) }
]
[
Amax(filtrate)
for which VD ) 0.15 cm3 (150 µL), VA ) 0.30 cm3 (300 µL), area
) 0.048 cm2, time ) 54 000 s (15 h).
Acknowledgment. This work was supported by the National
Institutes of Health (Grants AG02132, AG10770, and AG021601)
and by a gift from the G. Harold and Leila Y. Mathers Charitable
Foundation. The authors thank Hang Nguyen, Cedric Covaerts,
and Jay Choi (UCSF) for assistance in preparation of this
manuscript. B.C.H.M., S.B.P. and G.L have financial interest
in InPro Biotechnology, Inc. (California).
(21) Bostrom, J.; Greenwood, J. R.; Gottfries, J. Assessing the performance
of OMEGA with respect to retrieving bioactive conformations. J.
Mol. Graphics Modell. 2003, 21 (5), 449-462.
(22) Wolber, G. Fast neutron beams for boron neutron capture therapy?
Z. Med. Phys. 2004, 14 (1), 55-63.
Supporting Information Available: NMR, MS, and and
bioactivity data. This material is available free of charge via the
Internet at http://pubs.acs.org.
(23) Rudyk, H.; Vasiljevic, S.; Hennion, R. M.; Birkett, C. R.; Hope, J.;
Gilbert, I. H. Screening Congo Red and its analogues for their ability
to prevent the formation of PrP-res in scrapie-infected cells. J. Gen.
Virol. 2000, 81 (4), 1155-1164.
(24) Kocisko, D. A.; Baron, G. S.; Rubenstein, R.; Chen, J.; Kuizon, S.;
Caughey, B. New inhibitors of scrapie-associated prion protein
formation in a library of 2000 drugs and natural products. J. Virol.
2003, 77 (19), 10288-10294.
(25) May, B. C.; Witkop, J.; Sherrill, J.; Anderson, M. O.; Madrid, P. B.;
Zorn, J. A.; Prusiner, S. B.; Cohen, F. E.; Guy, R. K. Structure-
activity relationship study of 9-aminoacridine compounds in scrapie-
infected neuroblastoma cells. Bioorg. Med. Chem. Lett. 2006, 16 (18),
4913-4916.
(26) Lichtenfels, R.; Biddison, W. E.; Schulz, H.; Vogt, A. B.; Martin,
R. CARE-LASS (calcein-release-assay), an improved fluorescence-
based test system to measure cytotoxic T lymphocyte activity. J.
Immunol. Methods 1994, 172 (2), 227-239.
(27) Zhang, J. H.; Chung, T. D.; Oldenburg, K. R. A simple statistical
parameter for use in evaluation and validation of high throughput
screening assays. J. Biomol. Screening 1999, 4 (2), 67-73.
(28) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J.
Experimental and computational approaches to estimate solubility
and permeability in drug discovery and development settings. AdV.
Drug DeliVery ReV. 2001, 46 (1-3), 3-26.
References
(1) Prusiner, S. B. Prions. Proc. Natl. Acad. Sci. U.S.A. 1998, 95 (23),
13363-13383.
(2) DeArmond, S. J.; Prusiner, S. B. Transgenetics and neuropathology
of prion diseases. Curr. Top. Microbiol. Immunol. 1996, 207, 125-
146.
(3) Telling, G. C.; Scott, M.; Mastrianni, J.; Gabizon, R.; Torchia, M.;
Cohen, F. E.; DeArmond, S. J.; Prusiner, S. B. Prion propagation in
mice expressing human and chimeric PrP transgenes implicates the
interaction of cellular PrP with another protein. Cell 1995, 83 (1),
79-90.
(4) Kaneko, K.; Zulianello, L.; Scott, M.; Cooper, C. M.; Wallace, A.
C.; James, T. L.; Cohen, F. E.; Prusiner, S. B. Evidence for protein
X binding to a discontinuous epitope on the cellular prion protein
during scrapie prion propagation. Proc. Natl. Acad. Sci. U.S.A. 1997,
94 (19), 10069-10074.
(5) Kitamoto, T.; Tateishi, J. Human prion diseases with variant prion
protein. Philos. Trans. R. Soc. London, Ser. B 1994, 343 (1306),
391-398.
(6) Westaway, D.; Zuliani, V.; Cooper, C. M.; Da Costa, M.; Neuman,
S.; Jenny, A. L.; Detwiler, L.; Prusiner, S. B. Homozygosity for prion
protein alleles encoding glutamine-171 renders sheep susceptible to
natural scrapie. Genes DeV. 1994, 8 (8), 959-969.

Prion Inhibition by 3,5-Dicarbonitrile-Based Compounds
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1 73
(29) Reddy, T. R.; Mutter, R.; Heal, W.; Guo, K.; Gillet, V. J.; Pratt, S.;
Chen, B. Library design, synthesis, and screening: pyridine dicar-
bonitriles as potential prion disease therapeutics. J. Med. Chem. 2006,
49 (2), 607-615.
(30) Beukers, M. W.; Chang, L. C.; von Frijtag Drabbe Kunzel, J. K.;
Mulder-Krieger, T.; Spanjersberg, R. F.; Brussee, J.; IJzerman, A.
P. New, non-adenosine, high-potency agonists for the human
adenosine A2B receptor with an improved selectivity profile compared
to the reference agonist N-ethylcarboxamidoadenosine. J. Med. Chem.
2004, 47 (15), 3707-3709.
selective A1 adenosine receptor antagonist. Life Sci. 1987, 40 (6),
555-561.
(33) Gallo-Rodriguez, C.; Ji, X. D.; Melman, N.; Siegman, B. D.; Sanders,
L. H.; Orlina, J.; Fischer, B.; Pu, Q.; Olah, M. E.; van Galen, P. J.;
et al. Structure-activity relationships of N6-benzyladenosine-5′-
uronamides as A3-selective adenosine agonists. J. Med. Chem. 1994,
37 (5), 636-646.
(34) Townsend-Nicholson, A.; Schofield, P. R. A threonine residue in
the seventh transmembrane domain of the human A1 adenosine
receptor mediates specific agonist binding. J. Biol. Chem. 1994, 269
(4), 2373-2376.
(35) Govaerts, C.; Wille, H.; Prusiner, S. B.; Cohen, F. E. Evidence for
assembly of prions with left-handed beta-helices into trimers. Proc.
Natl. Acad. Sci. U.S.A. 2004, 101 (22), 8342-8347.
(31) Chang, L. C.; von Frijtag Drabbe Kunzel, J. K.; Mulder-Krieger, T.;
Spanjersberg, R. F.; Roerink, S. F.; van den Hout, G.; Beukers, M.
W.; Brussee, J.; IJzerman, A. P. A series of ligands displaying a
remarkable agonistic-antagonistic profile at the adenosine A1
receptor. J. Med. Chem. 2005, 48 (6), 2045-2053.
(32) Haleen, S. J.; Steffen, R. P.; Hamilton, H. W. PD 116,948, a highly
JM061045Z