Reactivity of 9-aminoacridine drug quinacrine with glutathione limits its antiprion activity

Article abstract of DOI:10.1111/cbdd.12918

Quinacrine—the drug based on 9-aminoacridine—failed in clinical trials for prion diseases, whereas it was active in in vitro studies. We hypothesize that aromatic nucleophilic substitution at C9 could be contributing factor responsible for this failure be

Full text of DOI:10.1111/cbdd.12918

Received Date : 05-Aug-2016
Revised Date : 19-Oct-2016
Accepted Date : 23-Nov-2016
Article type : Research Article
Reactivity of 9-aminoacridine drug quinacrine with glutathione limits its
anti-prion activity
Short running title: “Glutathione reduces quinacrine effects on prions”
Martin Šafařík,a,# Tibor Moško,b,# Zbigniew Zawada,a Eva Šafaříková,b Martin Dračínský,a Karel
Holada,b Jaroslav Šebestíka,*
aInstitute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic,
v.v.i., Flemingovo nám. 2, 166 10 Prague 6, Czech Republic;
bInstitute of Immunology and Microbiology, First Faculty of Medicine, Charles University in
Prague, Studničkova 7, 128 00 Prague 2, Czech Republic
Corresponding Author
*E-mail: jsebestik@seznam.cz. Phone: (+420)220-183-445. Fax: (+420)224-310-090.
This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process, which may lead to
differences between this version and the Version of Record. Please cite this article as doi:
10.1111/cbdd.12918
This article is protected by copyright. All rights reserved.

Author Contributions
#M.Š. and T.M. contributed equally. Z.Z. provided reference sample of GlutQui and data for
previous way of preparation. E.Š. provided recombinant prion protein for fluorescence assays.
M.D. measured NMR spectra. K.H. and J.Š. designed the experiments and wrote the manuscript.
KEYWORDS: quinacrine, failure in clinical trials, nucleophilic displacement, prion-protein
binding, anti-prion activity
ABSTRACT: Quinacrine – the drug based on 9-aminoacridine – failed in clinical trials for prion
diseases, whereas it was active in in-vitro studies. We hypothesize that aromatic nucleophilic
substitution at C9 could be contributing factor responsible for this failure because of the transfer
of acridine moiety from quinacrine to abundant glutathione. Here, we described the semi-large-
scale synthesis of the acridinylated glutathione and the consequences of its formation on
biological and biophysical activities. The acridinylated glutathione is one order of magnitude
weaker prion protein binder than the parent quinacrine. Moreover, according to log DpH 7.4 the
glutathione conjugate is two orders of magnitude more hydrophilic than quinacrine. Its higher
hydrophilicity and higher dsDNA binding potency will significantly decrease its bioavailability
in membrane-like environment. The glutathione deactivates quinacrine not only directly but also
decreases its bioavailability. Furthermore, the conjugate can spontaneously decompose to
practically insoluble acridone, which is precipitated out from the living systems.
This article is protected by copyright. All rights reserved.

1. Introduction
9-Aminoacridines are important compounds used for the treatment of cancer, viruses and
neurodegenerative diseases (1-3). Quinacrine (1) is perhaps one of the most important drugs
from the family of 9-aminoacridines. It is used for the treatment of various diseases such as
rheumatoid arthritis, lupus erythrematosus, chloroquine-resistant malaria, tapeworm infections
(Taenia saginata), Chagas disease, and epilepsy (refractory petit mal) (4-8). In in-vitro
experiments quinacrine diminished the propagation of prions in cell culture, suggesting that the
drug can be used for treatment of patients with Creutzfeldt-Jakob disease or the new variant of
this disease (5). Bis-acridinylated compounds were tested not only for their anti-prion activity (9)
but also for their anti-Alzheimer's activity (10). Although quinacrine demonstrated excellent in-
vitro results (5), it failed in clinical trials (11,12) and did not prolong survival in a murine model
of Creutzfeldt-Jakob disease (13) or in naturally scrapie infected sheep (14). Firstly, the failure of
quinacrine in-vivo was explained by selection of quinacrine resistant prions during the treatment
(15). In game model, the application of quinacrine even seemed to accelerate conformational
change and spreading of chronic wasting disease prions (16). Secondly, the failure was correlated
with clearance of quinacrine from the brain (17). Despite quinacrine reached a brain in
concentration corresponding to 10-fold of EC50 of in-vitro test (18), quinacrine was rapidly
cleared by P-glycoprotein efflux transporter (17). MDR0/0 mice, deficient in P-glycoprotein
multi-drug resistance (MDR) transporter, can accumulate up to 100 μM of quinacrine in the
brain. Such a high concentration of quinacrine leads to only temporal decrease of abnormal prion
protein (PrPSc) in brains of infected mice and gradually increases after several days despite the
presence of the drug (15). Thirdly, the lack of quinacrine efficiency in-vivo may have
pharmacodynamic origin as quinacrine is capable to limit formation of new molecules of PrPSc,
but does not affect already existing PrPSc (19).
This article is protected by copyright. All rights reserved.

As we have previously shown, quinacrine reacts with thiol-containing peptides including
glutathione (GSH) with which it forms S-acridinylated product – GlutQui (20) (Figure 1). We
hypothesize that this could be another factor responsible for quinacrine failure in clinical trials.
The failure can be then explained by different levels of GSH in neuronal cell cultures (5) and in
prion infected brains with increased amount of astrocytes (21). Especially, astrocytes not only
contain higher levels of GSH than neurons, but also release significant amounts of GSH into the
extracellular space (22). A better understanding of the quinacrine anti-prion limits can facilitate
drug design of in-vivo active compounds based on acridine structure.
In order to corroborate our hypothesis, we studied the influence of acridine moiety transfer on
various biological and biophysical parameters. The prion protein binding affinity, distribution
between aqueous and organic phase, dsDNA binding activity and in-vitro anti-prion activity were
evaluated.
2. Methods and Materials
2.1. General Procedures
All reagents were analytical grade and were purchased from Sigma-Aldrich, Czech Republic,
unless stated otherwise. During the syntheses, the molecular weights of the peptide conjugates
were determined using ESI-MS (Bruker Daltonics Reflex IV and Waters Q-Tof micro
instruments). Agilent 1200 instrument (Santa Clara, CA, USA) with a quaternary pump,
thermostat, diode array detector and a reversed-phase C18 column (PoroShell 120 SB-C18 2.7 µm,
3x50 mm, Agilent Technologies, Santa Clara, CA, USA) were used for HPLC using acetonitrile
(ACN) gradient from 1-1-13-19-57-100% within 0-1-2-8-13-15 min. The column was
This article is protected by copyright. All rights reserved.

o
termostated to 40 C. Unless stated otherwise, water for HPLC was modified with 0.1%
trifluoroacetic acid (TFA). UV-Vis spectra were acquired using a Varian Carry 5000 instrument
(Palo Alto, CA, USA) with a quartz cell (1 cm). NMR spectra of dimethylsulfoxide-d6 (DMSO-
d6) solutions were recorded with a Bruker Avance III spectrometer equipped with a 5-mm
1
13
diameter cryoprobe operating at 600.1 MHz for H and 150.9 MHz for C. The signals were
1
referenced to residual solvent signals (δ = 2.50 and 37.9 ppm in H and 13C, respectively).
Resonance assignment of all signals was obtained by a combination of 1D and 2D (H,H-COSY,
H,C-HSQC, H,C-HMBC) experiments.
2.2. Syntheses of S-(6-Chloro-2-methoxyacridin-9-yl)glutathione acetate (3a) and
S-(6-Chloro-2-methoxyacridin-9-yl)glutathione hydrochloride (3b)
The compounds were prepared similarly to procedure published previously, but the purification
steps were significantly improved (20). GSH (1 g, 3.25 mmol) and quinacrine dihydrochloride
(1.54 g, 3.25 mmol) were dissolved in 45 mL of water. The pH was adjusted by 1M NaOH to
o
7.5. The solution was stirred and heated at 37 C for 6 h. The conversion was approximately
80%; however, the prolonged reaction time did not provide significant improvement of
conversion. Small sample (150 mg) of the reaction mixture was purified by RP-HPLC using
gradient from 5%-100% ACN within 40 min. The TFA salt was converted to acetate by repeated
dissolution in 5% acetic acid (AcOH) and freeze drying. The yield was 30 mg (1%) of 3a.
According to mixed inject method, the product was identical with the previously published one.
HPLC RT 7.8 min. ESI HRMS (m/z): for [M+H]+ C24H26O7N4ClS calcd. 549.12052; found
549.12058 (0.10695 ppm). ESI HRMS neg. (m/z): for [M-H]+ C24H24O7N4ClS calcd.
547.10597; found 547.10486 (-2.03310 ppm). 1H-NMR is almost identical to that of 3b reported
This article is protected by copyright. All rights reserved.

below with one extra signal at 1.91 ppm belonging to acetate. One half of the crude reaction
mixture was loaded on weak anion exchanger (Zerolit G, SRA97, 100 mL) in hydroxide cycle.
The unreacted quinacrine was removed by washing with water (ca 300 mL) till the eluent is
colorless. Most of the quinacrine was eluted with water; however, a small amount of quinacrine
remained bound to the ionex even after 2 days elution as checked in another experiment. Then
free unreacted GSH was recovered by fractional elution with 5% acetic acid (ca 400 mL).
According to analytical HPLC, it was contaminated with desired product 3a. When the free GSH
was removed, the product 3b was eluted with 0.5M HCl (ca 150 mL). After evaporation and
o
drying in vacuum at 40 C over night, it provides orange solid 3b with yield 300 mg (14%) of
HPLC pure compound. According to mixed inject method, the product was identical with
previously published one. HPLC RT 7.8 min. For C24H25N4O7SCl (548.11) found ESI-MS, m/z:
549.1 (M+H)+. The Zerolite G was regenerated with 1M HCl and 1M NaOH and second half of
the reaction mixture was purified. Yield 205 mg (9%). The total preparative yield of 3b was
23%. For [H-γ-Glu-Cys(Qui)-Gly-OH]4 . 9 HCl . 9 H2O (C96H127N16O37S4Cl13) calc. 42.92%
C, 4.77% H, 8.34% N, 4.77% S, 17.16% Cl; found: 43.17% C, 4.42% H, 8.04% N, 4.80% S,
17.49% Cl. ESI HRMS neg. (m/z): for [M-H]+ C24H24O7N4ClS calcd. 547.10597; found
547.10486 (-2.03310 ppm). 13C NMR: 171.1 (Glu-δ), 171.0 (Gly-Cʹ), 171.0 (Glu-Cʹ), 170.0
(Cys-Cʹ), 158.2 (C2), 146.3 (C4b), 146.1 (C4a), 139.5 (C9), 134.1 (C6), 131.9 (C4), 130.0 (C8b),
18.7 (C8), 128.2 (C5), 128.0 (C7), 127.3 (C8a), 126.3 (C3), 102.4 (C1), 55.9 (OCH3), 53.2 (Cys-
1
α), 51.8 (Glu-α), 40.8 (Gly-α), 38.8 (Cys-β), 30.8 (Glu-γ), 26.0 (Glu-β). H NMR: 8.67 (d, 1H,
J8,7 = 9.3, H-8), 8.41 (d, 1H, JNH,α = 8.3, Cys-NH), 8.37 (t, 1H, JNH,α = 5.9, Gly-NH), 8.23 (bs,
+
3H, NH3 ), 8.23 (d, 1H, J5,7 = 2.2, H-5), 8.12 (d, 1H, J4,3 = 9.4, H-4), 7.90 (d, 1H, J1,3 = 2.8, H-
1), 7.69 (dd, 1H, J7,8 = 9.3, J7,5 = 2.2, H-7), 7.61 (dd, 1H, J3,4 = 9.4, J3,1 = 2.8, H-3), 4.41 (ddd,
This article is protected by copyright. All rights reserved.

1H, Jα,NH = 8.3, Jα,β = 9.4 and 4.6, Cys-α), 4.02 (s, 3H, OCH3), 3.90 (m, 1H, Glu-α), 3.59 (d, 2H,
Jα,NH = 5.9, Gly-α), 3.34 (dd, 1H, Jgem = 13.4, Jβ,α = 4.8, Cys-β), 3.19 (dd, 1H, Jgem = 13.4, Jβ,α
=
9.4, Cys-β), 2.18–2.25 (m, 2H, Glu-γ), 1.91–2.00 (m, 2H, Glu-β). 1H-NMR and 13C-NMR agreed
with previously published one (20). For copies of spectra see Supplementary Information
(Figures S3 and S4 for NMR, and Figures S6 and S7 for HRMS spectra). We have observed
concentration dependent decomposition of 3b in DMSO-d6 with traces of water.
2.3. Reactivity of 9-aminoacridines with GSH
9-Aminoacridine (230 mg, 1 mmol) or tacrine (234 mg, 1 mmol) and GSH (307 mg, 1 mmol)
o
were stirred in phosphate buffer (50 mM, pH 7.5, 100 mL) at 37 C under Ar for 24 h. The
progress of reaction was monitored by analytical HPLC in gradient from 5%-100% ACN within
20 min. Neither 9-aminoacridine nor tacrine were consumed by chemical reaction.
2.4. Stability of GlutQui in buffer and DMSO
Corresponding amount of GlutQui (3b, 0.5-50 mg) was dissolved in 1 mL of medium (either
phosphate buffer 0.1 M, pH 7.5 or neat DMSO) and maintained under Ar at 37 oC for 7-10 days.
Concentrations of GlutQui and decomposition products were determined using HPLC in gradient
5-5-100% CH3OH within 0-1-11 min with flow 1 ml/min, thermostated Poroshell 120 EC-C18
2.7 μm, 3.0x50 mm column at 40 °C. The decomposition products were also characterized by
NMR spectroscopy. Compounds 2 and 4 were identified in the decomposition mixture in 1:1
13
ratio. GSH (2) C NMR: 171.4 (Glu-δ), 171.1 (Gly-Cʹ), 170.9 (Glu-Cʹ), 170.5 (Cys-Cʹ), 51.9
(Cys-α), 51.8 (Glu-α), 41.0 (Gly-α), 39.9 (Cys-β), 30.9 (Glu-γ), 26.1 (Glu-β). 1H NMR: 8.44 (m,
+
3H, NH3 ), 8.41 (d, 1H, JNH,α = 8.5, Cys-NH), 8.37 (t, JNH,α = 5.9, Gly-NH), 4.58 (ddd, 1H, Jα,NH
This article is protected by copyright. All rights reserved.

= 8.5, Jα,β = 9.9 and 4.2, Cys-α), 3.91 (m, 1H, Glu-α), 3.75 d, 2H, Jα,NH = 5.9, Gly-α), 3.13 (dd,
1H, Jgem = 13.5, Jβ,α = 4.2, Cys-β), 2.87 (dd, 1H, Jgem = 13.5, Jβ,α = 9.9, Cys-β), 2.43 (m, 1H, Glu-
γ), 2.33 (m, 1H, Glu-γ), 1.98–2.09 (m, 2H, Glu-β). 6-Chloro-2-methoxyacridin-9(10H)-one (4)
13C NMR: 175.8 (C-8), 154.5 (C-2), 141.4 (C-4b), 137.8 (C-6), 135.9 (C-4a), 128.5 (C-8), 124.7
(C-3), 121.4 (C-8b), 121.3 (C-7), 119.6 (C-4), 118.4 (C-8a), 116,5 (C-5), 105.2 (C-1), 55.6
(OCH3). 1H NMR: 12.23 (s, 1H, H-10), 8.21 (d, 1H, J8,7 = 8.7, H-8), 7.63 (d, 1H, J5,7 = 2.0, H-5),
7.60 (d, 1H, J1,3 = 3.0, H-1), 7.58 (d, 1H, J4,3 = 8.9, H-4), 7.42 (dd, 1H, J3,4 = 9.0, J3,1 = 3.0, H-3),
7.22 (dd, 1H, J7,8 = 8.7, J7,5 = 2.0, H-7), 3.85 (s, 3H, OCH3). Copies of NMR spectra after
decomposition are depicted in Figure S5.
2.5. Solubility of 9-aminoacridines in water
To a sample of 9-aminoacridines (50 mg) was consecutively added 50 µL of solvent. The
suspension was sonicated for 5 min and evaluated by naked-eye with magnifying glass for
transparency. When undissolved particles remained, another portion of solvent was added. When
the sample was solubilized in volume lower than 100 µL of solvent, it was evaporated to dryness
and after the first 50 µL the process was repeated with 5 µL step. For sparingly soluble
compounds, only 1 mg was used for the assay.
2.6. Acridine-prion binding assay
The recombinant full length mouse prion protein with N-terminal His tag (His-mPrP23-231) was
prepared and purified as described previously (20,23). The assay based on quenching of prion-
protein intrinsic fluorescence was adapted according to previous work (24). The intrinsic
fluorescence of His-mPrP23-231 (6 µM) in phosphate buffer (10 mM, pH 8) was quenched by
This article is protected by copyright. All rights reserved.

increasing concentration of acridine compound using a Jasco FP-6600 spectrofluorimeter with
excitation at 295 3 nm (wavelength in which the filter effect caused by investigated acridines
was negligible) and emission at 337 2 nm at 18 oC. The fluorescence data were acquired 5 min
after the ligand addition in order to achieve ligand-protein equilibrium. Fluorescence of the same
system without prion protein was subtracted from the corresponding experiment. Non-linear
regression analysis was used for suppression of experimental errors using program xmgrace
(http://plasma-gate.weizmann.ac.il/Grace/) and fitting with exponential function:
where Fmax was the fluorescence intensity of the native His-mPrP23-231 in the absence of each
compound, F∞ was a theoretical value for complete saturation of all binding sites, and QC50 was
the concentration of acridine species causing 50% quenching of prion-protein fluorescence. This
function and equation 2 were used for generation of 1000 data points with inverse fractional
saturation (1/ΔF) as a dependent value and ligand concentration as an independent value.
The constructed data were fitted with the rectangular hyperbolic equation:
where KD was dissociation constant.
2.7. Apparent distribution coefficients of acridines between n-octanol-buffer phase
The apparent distribution coefficients were obtained by shake-flask method (25,26). Briefly,
aqueous phosphate buffer (10 mM, pH 7.4, 250 mL) and n-octanol (250 mL) were well shaken
and let stand overnight at 6 oC to separate as saturated phases. Prior to distribution experiment 5
mL of each of saturated phases were intensively mixed with 16 µmol acridine compound
o
overnight at 6 C. The low temperature was used in order to prevent hydrolysis of acridines to
This article is protected by copyright. All rights reserved.

corresponding acridones (20,27). The phases were separated by centrifugation and examined by
o
UV-Vis spectroscopy at 18 C. The absorbances at 400, 402, and 420 nm were used for
calibration curves and quantification of acridines in the both saturated phases. The average of
concentrations determined in each wavelength was used. When the absorbance was above 1, the
aliquot of the phase was diluted and reexamined.
2.8. Acridine-dsDNA interaction assay by displacement of ethidium bromide
An ethidium-DNA complex was formed by mixing 1.27 µM solution of ethidium bromide and
0.696 µM (in base pairs) of calf thymus-DNA (CT-DNA). Measurements were carried out in 3
o
mL of phosphate buffer (50 mM, pH 7.4, 18 C). Small volumes of a concentrated solution of
inhibitor were then titrated while the fluorescence of the solution was monitored with excitation
at 546 3 nm and emission at 600 2 nm. Fluorescences were read at least 3-times and average
values were used. Fluorescence of the same system without ethidium was subtracted from the
corresponding experiment. Nonlinear curve fitting was used to determine IC50 values by fitting
the raw displacement data to a one-phase exponential decay curve using xmgrace software
utilizing modified equation 1. QC50 value was substituted with IC50 – the concentration of
inhibitor causing 50% displacement of ethidium bromide.
The conversion (28,29) of IC50 values to corresponding Ki (apparent inhibition constant) is based
on Eqs. (4-6),
where [NA] is the concentration of ‘‘free’’ (unbound) nucleic acid, [Obs] is the concentration of
the ‘‘free’’ observable species (ethidium), [NA–Obs] is the concentration of the NA–Obs
complex, [I] is the concentration of ‘‘free’’ inhibitor, and [NA–I] is the concentration of inhibitor-
bound nucleic acid. This calculation assumes that one equivalent of inhibitor displaces one
This article is protected by copyright. All rights reserved.

equivalent of ethidium and thus the term ‘‘apparent’’ affinity is used. According to the ethidium–
CT-DNA association curve, the KD = 0.88 µM and the binding stoichiometry is 0.46 ethidium
binding sites per base pair (n). Upon adding 0.696 µM of CT-DNA to 1.27 µM of ethidium, the
concentration of ethidium–DNA [NA–Obs] = 0.18 µM, free ethidium [Obs] = 1.09 µM, and free
DNA [NA] = 0.14 µM (in available binding sites). Upon titration of the inhibitor to its IC50
value, the concentration of the complex [NA–Obs] = 0.09 µM, free ethidium [Obs] = 1.18 µM,
free DNA [NA] = 0.14 µM (in available binding sites), [NA–I] = 0.09 µM, and the concentration
of free inhibitor [I] = {(IC50 value) - [NA–I]}. These values were used to solve Eq. (6).
2.9. Cell line and prion strain
As a model of prion infection, we used prion sensitive CAD5 cells (Cath.a-differentiated cell
line) (30) provided by Charles Weissmann (Department of Infectology, Scripps Research
Institute, FL, USA). CAD5 cells were chronically infected with Rocky Mountain Laboratory
(RML) strain of mouse-adapted scrapie prions as described previously (31). Cells were
maintained in Opti-MEM medium supplemented with 10% bovine growth serum and
penicillin/streptomycin (Invitrogen).
2.10. Toxicity and cell culture manipulation
Stock solution of quinacrine (5 mM) was prepared in phosphate-buffer saline (PBS) pH 7.4 and
GlutQui (3b) was dissolved in DMSO (85 mM solution). In the first round of experiments, the
stock solution was used during the course of the experiment, in the second round of experiments,
the solutions were prepared fresh each day to eliminate the effect of above mentioned
decomposition of GlutQui. Prior to testing anti-prion activity, the toxicities of quinacrine and
This article is protected by copyright. All rights reserved.

GlutQui towards CAD5 cells were evaluated and the concentrations which did not prevent the
growth of the cells were determined. Based on the toxicity testing the concentrations 0.3 - 4 µM
and 10 -200 µM were used for quinacrine and GlutQui, respectively. Chronically RML infected
or non-infected CAD5 cells (70,000 cells) were seeded per tissue culture dish (10 cm in
diameter) to ensure enough space for cells to growth for at least 7 days. Medium with quinacrine
or GlutQui was exchanged every day for continuous treatment or the acridine was applied only
once at first day for “one dose” treatment (acridine free medium was then changed every day). At
7th day, the cells were quantitatively harvested and counted. For toxicity measurement, the LC50
values were determined using equation analogous to Eq. (1). However, the number of cells
equivalent to “F∞” parameter was fixed to be zero. From every tissue culture dish, 110,000 cells
were used for standard scrapie cell assay (SSCA) and the rests of cells were washed in PBS and
pellets were stored at -80°C for western blot analysis.
2.11. Standard Scrapie Cell Assay
The samples were investigated by SSCA (32). All steps were carried out at room temperature
unless stated otherwise. The ELISPOT plates (Multiscreen 96 well filter plates with high protein
binding Immobilon-P membrane, 0.45 µm, Millipore) were activated with 60 µL of 70% ethanol
per well. The ethanol was removed by vacuum. Wells were washed twice by 160 µL PBS and 50
µL of PBS was added to each well to prevent drying out. Cells were washed and diluted in PBS
to 100,000 and 10,000 cells per ml and 100 µL aliquots plated to the wells of ELISPOT plate in
eightplicates (10,000 or 1,000 cells per well). Cells were immobilized to the membrane by
vacuum (Supelco) and drying the ELISPOT plates at 50 °C for 1 hour. Plates were stored at 4 °C
for up to two weeks before further processing.
This article is protected by copyright. All rights reserved.

For proteinase K (PK) treatment, 60 µL of PK (2 µg/mL) were added to each well and incubated
at 37 °C for 30 minutes. After 30 minutes, cells were washed with 160 µL of PBS and the PK
was inhibited by 2 mM phenylmethanesulfonyl fluoride (PMSF) in PBS for 10 minutes.
Subsequently, cells were washed with 160 µL of PBS and denatured by treatment with 150 µL of
3 M guanidine thiocyanate in 10 mM tris(hydroxymethyl)aminomethane (TRIS)-HCl, pH 8 for
10 minutes. After denaturation, the cells were washed 4 times with 160 µL of PBS and the
membrane was blocked using Superblock solution (Pierce) for 1 hour. Blocking reagent was
removed and primary antibody (60 µL per well, 6D11 Covance) in TRIS buffered saline pH 7.4
with 0.05% Tween 20 (TBS-T) and 1% non-fat milk (Biorad) was added and incubated for 2
hours. Cells were washed 4 times with TBS-T and incubated with 60 µL of secondary antibody
(Alkaline phosphatase conjugated affinity purified F(ab')2 fragment DAM IgG, Jackson) for 1
hour. The cells were washed 5 times with TBS-T and developed with AP Conjugate Substrate Kit
(Bio-Rad). Spots on ELISPOT plates were quantified using the Nikon system equipped with
analytical software (NIS-Elements, ver. 4.20). The SSCA data were fitted by equation analogous
to Eq. (1). However, the number of spots equivalent to “F∞” parameter was fixed to be zero.
2.12. Western blot
Cell pellets were lysed using protein lysis buffer (50 mM TRIS pH 7.4, 150 mM NaCl, 0.5%
Triton X-100, and 0.5% Na deoxycholate) on ice for 30 minutes with occasional vortexing. Cell
nuclei were not excluded from lysates and samples were treated with benzonase (Novagen).
Protein concentrations were determined using bicinchoninic acid assay (Pierce) according to
manufacturer's instructions. For samples not treated with PK, 30 µg of total protein was loaded
on gel per line. For samples treated with PK, 300 µg of total proteins was incubated with 50
This article is protected by copyright. All rights reserved.

µg/ml of PK for 30 minutes at 37 °C. The PK treatment was stopped by adding PMSF to final
concentration 2 mM for 10 minutes. Proteins were precipitated (salting-out in presence of 10%
NaCl and 0.05% N-lauroyl sarcosine Na) on ice (33), washed 2 times in 25 mM TRIS-HCl pH
8.8 and dissolved in Laemmli sample buffer. Gel electrophoresis and blotting was performed as
described previously (34). The membranes were developed after probing with a mix of mAbs
AH6 (epitope PrP159–174, Roslin Institute, Midlothian, UK) and DC2 (epitope PrP39–46,
provided by Vladka Curin-Serbec, Blood Transfusion Centre of Slovenia, Ljubljana).
3. Results
3.1. Semi-large-scale synthesis of quinacrine—GSH conjugate
In order to avoid expensive HPLC separation, ionex chromatography with Zerolit G weak anion
exchanger was used for separation of reaction mixture after acridinylation of GSH (20). The
unconsumed quinacrine (base) is almost not retained by the ionex. The removal of unreacted
GSH from GlutQui was done by various strength of acid used for elution. AcOH was sufficient
for elution of mixture of the GSH contaminated with small amount of its conjugate 3a. Stronger
acid (HCl) was used for final release of the conjugate 3b from the resin in more than 95% purity.
The new synthesis provides hundreds of milligrams of the peptide conjugate, which allows most
of biological and physicochemical assays.
3.2. Resistance of primary 9-aminoacridines towards thiols
In contrast to quinacrine, both 9-aminoacridine and tacrine remained stable during incubation
with GSH within 24 h at 37 oC under Ar.
This article is protected by copyright. All rights reserved.

3.3. Stability of GlutQui
In order to investigate biological properties of GlutQui (3b), it was important to know whether
GlutQui is stable under physiological conditions, as well as, in stock solution. Within the 7 days,
the concentration of GlutQui did not change more than 1% in buffer and more than 3% in
DMSO, when the starting concentration was lower than 1 mg/mL. However, the situation
became different when the initial concentration of GlutQui was increased. In DMSO, almost
complete decomposition was achieved within few days with starting concentration higher than
15 mg/mL (Figure 2).
According to NMR study, the GlutQui (3b) decomposes to GSH (2) and corresponding acridone
(4). Interestingly, measurement of NMR spectra of pure compounds requires the utilization of
low sample concentration ca 1.5 mg/mL and employment of CryoProbe. Otherwise partial
decomposition takes place (Figure S1 in Supplementary Information).
3.4. Prion-binding of the acridines
Binding of His-mPrP23-231 with various acridines was investigated by assay of Villa et al. (24).
This assay revealed that quinacrine is more than 20-times stronger binder than the GlutQui
(Figure 3 and Table 1). The conjugation with GSH led to 95% deactivation of quinacrine.
Consequently, the GlutQui conjugate 3a is more than 3-times weaker prion protein binder than
unsubstituted 9-aminoacridine.
This article is protected by copyright. All rights reserved.

3.5. Bioavailability of the acridines
Two factors influencing bioavailability of 9-aminoacridines were evaluated i.e. DNA binding and
distribution between n-octanol/buffer system. When GSH accepts acridine moiety from
quinacrine, the DNA affinity increases ca 5-folds (Table 2). The GlutQui conjugate 3a is ca 21-
times tighter DNA binder than 9-aminoacridine.
Reaction of GSH with quinacrine increased water solubility (with exception of dihydrochloride
salt 3b); however, decreased membrane permeability (Table 3, Table 4). Quinacrine easily
penetrates to organic phase, whereas GlutQui conjugate 3a is mostly located in aqueous one, i.e.
the GlutQui prefers to stay in aqueous phase two orders of magnitude more than the original
drug.
3.6. Toxicity and anti-prion activity of the acridines
The assay in dividing prion permissive CAD5 cells was established. The toxic effects such as
mainly observed inhibition of cell divisions and almost unseen detachment of dead cells in
culture were noticeable at 0.3 µM concentration of quinacrine and 30 µM concentration of
GlutQui (Figure 4). Calculated values of LC50 under the culture condition setup were 1.3 µM and
≥97 µM for quinacrine and GlutQui (3b), respectively.
The efficacy of prion clearance from the cells was investigated by standard scrapie cell assay
(SSCA) (32) (Figure 5). The concentrations of quinacrine and of GlutQui (3b) responsible for
50% decrease of infected cell numbers were 0.68 µM and 15 µM, respectively.
We have realized that the prion healing curves for GlutQui (3b) are almost independent of the
use of freshly prepared stock solution or aged stock solution. This opened a question whether the
continuous treatment of infected cells in culture is necessary or only the first dose is responsible
for the anti-prion effect. We have investigated it in more detail for FDA approved drug
This article is protected by copyright. All rights reserved.

quinacrine (Figure 6). Our results suggest that one dose of quinacrine (3 µM) can lead to
complete disappearance of prion infected cells form culture; however, the continuous treatment
was more effective in lowering the proportion of infected cells at lower quinacrine
concentrations.
The healing of infected cells from prion infection with quinacrine and GlutQui (3b) including
one dose mechanisms was confirmed by western blot (Figure 7). Complete loss of signal of PrPSc
during continuous treatment was observed at 2 µM or 30 µM concentration of quinacrine or
GlutQui (3b), respectively. For one dose treatment, the loss of signal of PrPSc was achieved at 3
µM or 50 µM concentration of quinacrine or GlutQui (3b), respectively. This shift of healing
capabilities correlated well with the 20-fold lower affinity of GlutQui (3a) toward prion protein
compared to quinacrine. The decrease of efficacy during one dose treatment was also observed
in SSCA assay.
4. Discussions
The ability of 9-aminoacridines to react with thiol nucleophiles is well recognized (35,36). We
and others previously shown (20,35,37) that quinacrine reacts also with GSH – quite abundant
biological thiol nucleophile present in human body. Depletion of cellular GSH was observed
during the treatment of cancer cell lines with 9-aminoacridine-thiazolidine derivatives (38). The
reactivity of 9-aminoacridine derivatives with GSH could be directly responsible for GSH
depletion. In our hands quinacrine reacted easily with GSH producing GlutQui, but unsubstituted
9-aminoacridine was stable toward GSH. The stability of unsubstituted 9-amino group towards
nucleophilic substitution is in accordance with the stability of 9-aminoacridines towards nitrogen
nucleophiles (27).
This article is protected by copyright. All rights reserved.

In order to test biological and physicochemical properties of GlutQui, we have developed
a semi-large scale synthesis procedure. In contrast to previously published small-scale synthesis
protocol (20) requiring HPLC purification of a reaction mixture the new separation strategy
using ionex chromatography provides hundreds of milligrams of the peptide conjugate allowing
to carry out biological and physicochemical assays. By working with higher amount of GlutQui,
we also have realized that the conjugate can decompose to acridone and that this decomposition
is concentration dependent.
Regarding to anti-prion activity, we can assume that quinacrine, to be effective, have to
interact with prion protein. We have previously observed discrepancy (20,27) between weak
binding of quinacrine to truncated recombinant prion protein hPrP121-230 (KD 4.6 mM) (39) and
strong anti-prion activity of quinacrine in-vitro (EC50 0.3 µM) (5,9). However, when prion
protein spanning from 90 to 231 amino acids is used (hPrP90-231), more potent binding
appeared (KD 0.49 µM) (24). Removal of ca 30 amino acids containing cluster of charged
residues and part of the hydrophobic core of the protein led to totally different KD values. The
later reported KD correlated well with the described anti-prion activity. Therefore, we applied the
described assay for assessment of binding affinity of quinacrine and GlutQui to more accurate
model – full-length His-mPrP23-231. In this assay, the conjugation with GSH led to 95%
deactivation of quinacrine. Consequently we showed that the GlutQui conjugate is more than 3-
times weaker prion protein binder than unsubstituted 9-aminoacridine. Whereas 9-aminoacridine
is weaker binder than quinacrine, it is stable towards GSH. Thus, after quinacrine deactivation,
9-aminoacridine is 3-fold more potent prion protein binder than GSH deactivated quinacrine.
This article is protected by copyright. All rights reserved.

9-Aminoacridines are quite strong DNA binders. Tight binding to DNA severely limits
the extravascular distributive properties of drugs, because diffusion is driven by the free drug
concentration (2). For these compounds, the effective diffusion coefficient (Deff) that rules the
rate of extravascular diffusion, is significantly lowered by DNA binding and correlates with
dissociation constant. It is given by
, where D stands for the diffusion coefficient,
KD is the DNA dissociation constant and Sf corresponds to the concentration of available DNA-
binding sites (2). For a total DNA-binding site concentration of about 1 mM and non-specific
binder (about 10% of the sites in chromatin), the lower limit of KD to maintain the diffusion time
useful on the pharmacological timescale is about 100 µM (2). Thus, even quite moderate DNA
binding limits diffusion of would-be drug. For 9-aminoacridines, the dissociation constants are in
10 µM levels and their bioavailability is quite low. In some cases, the conjugation of 9-
aminoacridines with peptides can increase acridine bioavailability by suppression of their DNA
binding (40). However, when GSH accepts acridine moiety from quinacrine, the DNA affinity
increases ca 5-folds (Table 2). We showed that the GlutQui conjugate is ca 21-times tighter
binder than 9-aminoacridine. Thus, GSH not only suppresses prion binding of quinacrine but
also decreases its bioavailability by enhanced DNA binding.
Another factor influencing bioavailability of drugs in human body is distribution
coefficient log P between n-octanol—water (25), where n-octanol serves as a model of biological
membranes. For polyelectrolytes such as peptides, the better measure is log DpH 7.4, where instead
of pure water, the aqueous physiological buffer is used with pH 7.4 (26). The compounds with
log DpH 7.4 in range -0.5 – 3 belong to families of most used drugs (41-44), i.e. they have well
balanced membrane permeability and aqueous solubility in biological systems. Conjugation of
GSH with quinacrine made the drug more water soluble and suppressed its membrane
This article is protected by copyright. All rights reserved.

permeability. Since prions in infected cells are located both on the cytoplasmic membrane and
intracellularly (45), the hydrophilization of acridine moiety decreased approximately 100-fold
local concentration of the active drug in prion active sites.
Toxicity of tested drugs also has to be taken in to the account. Toxic effect, presented as
inhibition of cell division, of quinacrine or GlutQui occurred in concentration dependent manner.
According to LC50 values in CAD5 cells, quinacrine is ca 80 times more toxic than GlutQui.
Effectivity of treatment tested on RML infected CAD5 cell by SSCA have shown 25-fold
decrease of GlutQui anti-prion activity compared to quinacrine. This observation corresponds to
lower binding affinity of GlutQui to His-mPrP23-231. The activities of fresh and aging GlutQui
stock solutions in SSCA cannot be explained by higher efficacy of formed acridone, because
acridone 4 in saturated solution has only minor effect on number of prion infected cells in SSCA
assay (Figure S2), which is in range of standard error of the assay (32). To simulate the drug
regimen of aging GlutQui using quinacrine we treated infected cells only ones. We used only
one dose at the first day of the assay. The similar effect of “only the first dose of drug heals the
cells” appeared also for quinacrine. One dose of quinacrine (3 µM) can lead to complete
disappearance of prion infected cells from culture. Healing of prion infection by application of
one dose only can indicate that acridines may (I.) either actively accumulate in the cells in
concentration that is sufficient for prion removal even after its dilution by cell divisions; (II.) or
can somehow reset the susceptibility of the cells to prion infection for at least several cell
divisions enough for disappearance of PrPSc from the culture. This is supported by fact that the
culture is growing exponentially after application and the medium is replaced every day with
acridine-free system. It suggests that the cell divisions in combination with the initial impact of
This article is protected by copyright. All rights reserved.

acridines rather than permanent presence of acridines in media are necessary for healing of cells
from the prion infection. Moreover, it could be also an explanation why the quinacrine anti-
prion activity is effective only in replicating cell culture (15).
5. Conclusions
The exposure of quinacrine drug towards biogenic thiol – GSH – led to formation of acridine-
peptide conjugate, which has 20-fold lower affinity toward recombinant prion protein. This
biophysical parameter correlates well with 15-times lower anti-prion activity of GlutQui
confirmed by western-blots and 25-fold decrease of anti-prion activity in standard scrapie cell
assay. Moreover, GlutQui is 100-times more hydrophilic and 5-times more powerful dsDNA
binder than the parent drug. Taking into account these findings, by conversion of quinacrine to
the peptide-conjugate, we can estimate ca 2,000 times (20 x 100) lowering of activity in
experiments in-vivo than that in-vitro. Moreover, the conjugate slowly decomposes to acridone,
which is practically insoluble in water. However, the findings of our study should be interpreted
cautiously as in vitro models utilized may imperfectly replicate complex in vivo situation.
Interestingly, it seems that acridines can heal prion infection of cultured cells by application of
one dose only. Since from that time the culture is growing and the medium exchanged, most
probably, the acridines have to either influence prion propagation by attacking the replication
mechanism at the beginning of the treatment, or accumulate into the cells in concentration much
higher than that required for treatment with subsequent dilution to lower but still effective
concentration during cell replication.
This article is protected by copyright. All rights reserved.

Biological activity of 9-aminoacridines, which react with GSH, can be altered. It seems that
acridines having 9-amino group without other substituents (primary amine) are resistant towards
reaction with GSH. Thus, we hypothesize that new drugs can be designed as 9-aminoacridines
with introduction of aminoalkyl chains in different positions than C-9. By this, they will be
supposedly inert toward GSH.
Acknowledgments
We thank Prof. Charles Weissmann (SCRIPPS, Florida) and Prof. Vladka Curin-Serbec (Blood
Transfusion Centre of Slovenia, Ljubljana) for the donation of materials. This work was
supported by grants of the Czech Science Foundation (GA CR) No. 14-00431S; 15-09072S; 16-
00270S; P303/12/1791; and Research Project RVO 61388963. TM, ED and KH were supported
by projects of Charles University in Prague: PRVOUK-P24/LF1/3, UNCE 204022 and SVV
260026.
Conflict of interest: Authors declare that they have no conflict of interests.
References
1. Demeunynck, M., Charmantray, F., Martelli, A. (2001) Interest of acridine derivatives in the
anticancer chemotherapy. Curr Pharm Des 7:1703-1724.
2. Denny, W. A. Acridine-4-carboxamides and the concept of minimal DNA intercalators, in:
Demeunynck, M., Bailly, C., Wilson, W.D. editors. Small molecule DNA and RNA binders,
Wiley-VCH Verlag GmbH & Co, Weinheim, 2003; p 482-502.
3. Šebestík, J., Hlaváček, J., Stibor, I. (2007) A role of the 9-aminoacridines and their conjugates
in a life science. Curr Prot Pept Sci 8:471-483.
This article is protected by copyright. All rights reserved.

4. Wallace, D. J. (1989) The use of Quinacrine (Atabrine) in rheumatic diseases - a
reexamination. Seminars in Arthritis and Rheumatism 18:282-297.
5. Korth, C., May, B. C. H., Cohen, F. E., Prusiner, S. B. (2001) Acridine and phenothiazine
derivatives as pharmacotherapeutics for prion disease, Proc Natl Acad Sci USA 98:9836-9841.
6. Krauth-Siegel, R. L., Bauer, H., Schirmer, H. (2005) Dithiol proteins as guardians of the
intracellular redox milieu in parasites: Old and new drug targets in Trypanosomes and Malaria-
causing Plasmodia. Angew Chem Int Ed 44:690-715.
7. Saravanamuthu, A., Vickers, T. J., Bond, C. S., Peterson, M. R., Hunter, W. N., Fairlamb,
A. H. (2004) Two interacting binding sites for Quinacrine derivatives in the active site of
Trypanothione reductase. A template for drug design. J Biol Chem 279:29493-29500.
8. Burnett, J. C., Schmidt, J. J., Stafford, R. G., Panchal, R. G., Nguyen, T. L., Hermone, A. R.,
Vennerstrom, J. L., McGrath, C. F., Lane, D. J., Sausville, E. A., Zaharevitz, D. W., Gussio, R.,
Bavari, S. (2003) Novel small molecule inhibitors of Botulinum neurotoxin A metalloprotease
activity. Biochem Biophys Res Commun 310:84-93.
9. May, B. C. H., Fafarman, A. T., Hong, S. B., Rogers, M., Deady, L. W., Prusiner, S. B.,
Cohen, F. E. (2003) Potent inhibition of Scrapie prion replication in cultured cells by bis-
acridines, Proc Natl Acad Sci USA 100:3416-3421.
10. Bolognesi, M. L., Cavalli, A., Valgimigli, L., Bartolini, M., Rosini, M., Andrisano, V.,
Recanatini, M., Melchiorre, C. (2007) Multi-target-directed drug design strategy: From a dual
binding site Acetylcholinesterase inhibitor to a trifunctional compound against Alzheimer’s
Disease. J Med Chem 50:6446-6449.
This article is protected by copyright. All rights reserved.

11. Collinge, J., Gorham, M., Hudson, F., Kennedy, A., Keogh, G., Pal, S., Rossor, M., Rudge, P.,
Siddique, D., Spyer, M., Thomas, D., Walker, S., Webb, T., Wroe, S., Darbyshire, J. (2009)
Safety and efficacy of Quinacrine in human prion disease (PRION-1 study): A patient-preference
trial. Lancet Neurol 8:334-344.
12. Geschwind, M. D., Kuo, A. L., Wong, K. S., Haman, A., Devereux, G., Raudabaugh, B. J.,
Johnson, D. Y., Torres-Chae, C. C., Finley, R., Garcia, P., Thai, J. N., Cheng, H. Q., Neuhaus,
J. M., Forner, S. A., Duncan, J. L., Possin, K. L., Dearmond, S. J., Prusiner, S. B., Miller, B. L.
(2013) Quinacrine treatment trial for sporadic Creutzfeldt-Jakob disease. Neurology 81:2015-
2023.
13. Collins, S. J., Lewis, V., Brazier, M., Hill, A. F., Fletcher, A., Masters, C. L. (2002)
Quinacrine does not prolong survival in a murine Creutzfeldt-Jakob Disease model, Ann Neurol
52:503-506.
14. Gayrard, V., Picard-Hagen, N., Viguié, C., Laroute, V., Andréoletti, O., Toutain, P. L. (2005)
A possible pharmacological explanation for quinacrine failure to treat prion diseases:
pharmacokinetic investigations in a ovine model of scrapie. British J Pharmacol 144:386-393.
15. Ghaemmaghami, M. S., Ahn, M., Lessard, P., Giles, K., Legname, G., DeArmond, S. J.,
Prusiner, S. B. (2009) Continuous Quinacrine treatment results in the formation of drug-resistant
prions. PLoS Pathogens 5: art. no. e1000673.
16. Bian, J., Kang, H.-E., Telling, G. C. (2014) Quinacrine promotes replication and
conformational mutation of chronic wasting disease prions. Proc Natl Acad Sci USA 111:6028–
6033.
This article is protected by copyright. All rights reserved.

17. Ahn, M., Ghaemmaghami, S., Huang, Y., Phuan, P.-W., May, B. C. H., Giles, K.,
DeArmond, S. J., Prusiner, S. B. (2012) Pharmacokinetics of Quinacrine Efflux from Mouse
Brain via the P-glycoprotein Efflux Transporter. PLoS One 7: art. no. e39112.
18. Yung, L., Huang, P. Y., Lessard, P., Legname, G., Lin, E. T., Baldwin, M., Prusiner, S. B.,
Ryou, C., Guglielmo, B. J. (2004) Pharmacokinetics of Quinacrine in the treatment of prion
disease. BMC Infectious Diseases 4: art no 53.
19. Barret, A., Tagliavini, F., Forloni, G., Bate, C., Salmona, M., Colombo, L., De Luigi, A.,
Limido, L., Suardi, S., Rossi, G., Auvré, F., Adjou, K. T., Salčs, N., Williams, A., Lasmézas, C.,
Deslys, J. P. (2003) Evaluation of quinacrine treatment for prion diseases. J Virol 77:8462-8469.
20. Zawada, Z., Šafařík, M., Dvořáková, E., Janoušková, O., Březinová, A., Stibor, I., Holada,
K., Bouř, P., Hlaváček, J., Šebestík, J. (2013) Quinacrine reactivity with prion proteins and
prion-derived peptides. Amino Acids 44:1279-1292.
21. Hannaoui, S., Maatouk, L., Privat, N., Levavasseur, E., Faucheux, B. A., Haïk, S. (2013)
Prion Propagation and Toxicity Occur In Vitro with Two-Phase Kinetics Specific to Strain and
Neuronal Type. J Virol 87:2535-2548.
22. Guitart, K., Loers, G., Schachner, M. and Kleene, R. (2015) Prion protein regulates
glutathione metabolism and neural glutamate and cysteine uptake via excitatory amino acid
transporter 3. J Neurochem 133: 558–571.
23. Pavlíček, A., Bednárová, L., Holada, K. (2007) Production, purification and oxidative
folding of the mouse recombinant prion protein. Folia Microbiol (Praha) 52:391-397.
This article is protected by copyright. All rights reserved.

24. Villa, V., Tonelli, M., Thellung, S., Corsaro, A., Tasso, B., Novelli, F., Canu, C., Pino, A.,
Chiovitti, K., Paludi, D., Russo, C., Sparatore, A., Aceto, A., Boido, V., Sparatore, F., Florio, T.
(2011) Efficacy of novel acridine derivatives in the inhibition of hPrP90-231 prion protein
fragment toxicity. Neurotox Res 19: 556-574.
25. Leo, A., Hansch, C., Elkins, D. (1971) Partition coefficients and their uses. Chem Rev
71:525-616.
26. Rubas, W., Cromwell, M. E. M. (1997) The effect of chemical modifications on
octanol/water partition (log D) and permeabilities across Caco-2 monolayers. Adv Drug Deliv
Rev 23:157-162.
27. Šebestík, J., Šafařík, M., Stibor, I., Hlaváček, J. (2006) Acridin-9-yl exchange: A proposal
for the action of some 9-aminoacridine drugs. Biopolymers (Pept Sci) 84:605-614.
28. Šebestík, J., Hlaváček, J., Stibor, I. (2006) Rational design and synthesis of a double-
stranded DNA-binder library. Biopolymers (Pept Sci) 84:400-407.
29. Luedtke, N. W., Liu, Q., Tor, Y. (2003) RNA-Ligand Interactions:ꢀ Affinity and Specificity
of Aminoglycoside Dimers and Acridine Conjugates to the HIV-1 Rev Response Element.
Biochem 42:11391-11403.
30. Qi, Y., Wang, J. K., McMillian, M., Chikaraishi, D. M. (1997) Characterization of a CNS
cell line, CAD, in which morphological differentiation is initiated by serum deprivation. J
Neurosci 17:1217-1225.
31. Janoušková, O., Rakušan, J., Karásková, M., Holada, K. (2012) Photodynamic inactivation
This article is protected by copyright. All rights reserved.

of prions by disulfonated hydroxyaluminium phthalocyanine. J Gen Virol 93:2512-2517.
32. Mahal, S. P., Demczyk, C. A., Smith, E. W. Jr, Klohn, P. C., Weissmann, C. (2008) Assaying
prions in cell culture: the standard scrapie cell assay (SSCA) and the scrapie cell assay in end
point format (SCEPA). Methods in Molecular Biology 459:49-68.
33. Polymenidou, M., Verghese-Nikolakaki, S., Groschup, M., Chaplin, M. J., Stack, M. J.,
Plaitakis, A., Sklaviadis, T. (2002) A short purification process for quantitative isolation of
PrPSc from naturally occurring and experimental transmissible spongiform encephalopathies.
BMC Infectious Diseases 2:art no 23.
34. Panigaj, M., Glier, H., Wildova, M., Holada, K. (2011) Expression of prion protein in mouse
erythroid progenitors and differentiating murine erythroleukemia cells. PLoS One 6: art no
e24599.
35. Wild, F., Young, J. M. (1965) The reaction of Mepacrine with thiols. J Chem Soc 7261-
7274.
36. Weltrowski, M., Ledochowski, A., Sowinski, P. (1982) Research on tumor-inhibiting
compounds 70. Reactions of 1-nitroacridines with ethanethiol. Polish J Chem 56: 77-82.
37. Zawada, Z., Šebestík, J., Šafařík, M., Krejčiříková, A., Březinová, A., Hlaváček, J., Stibor, I.
Holada, K., Bouř, P. (2010) What could be the role of Quinacrine in Creutzfeldt-Jakob Disease
treatment? in: Lebl, M. Meldal, M., Jensen, K., Hoeg-Jensen, T. editors, Peptides 2010. Proc.
31st Eur. Pept. Symp., Eur. Pept. Soc.: Copenhagen DK, p 84-85.
38. Paulíková, H., Vantová, Z., Hunáková, Ľ., Čižeková, L., Čarná, M., Kožurková, M.,
This article is protected by copyright. All rights reserved.

Sabolová, D., Kristian, P., Hamuľaková, S., Imrich, J. (2012) DNA binding acridine-
thiazolidinone agents affecting intracellular glutathione. Bioorg Med Chem 20:7139-7148.
39. Vogtherr, M., Grimme, S., Elshorst, B., Jacobs, D. M., Fiebig, K., Griesinger, C., Zahn, R.
(2003) Antimalarial drug Quinacrine binds to C-terminal helix of cellular prion protein. J Med
Chem 46:3563-3564.
40. Šebestík, J., Stibor, I., Hlaváček, J. (2006) New peptide conjugates with 9-aminoacridine:
synthesis and binding to DNA. Journal of Peptide Science 12:472-480.
41. Zhu, C., Jiang, L., Chen, T.-M., Hwang, K.-K. (2002) A comparative study of artificial
membrane permeability assay for high throughput profiling of drug absorption potential. Eur J
Med Chem 37:399-407.
42. Hill, A. P., Young, R. J. (2010) Getting physical in drug discovery: a contemporary
perspective on solubility and hydrophobicity. Drug Discovery Today 15:648-655.
43. Ungell, A.-L., Nylander, S., Bergstrand, S., Sjöberg, Å., Lennernäs, H. (1998) Membrane
Transport of Drugs in Different Regions of the Intestinal Tract of the Rat. J Pharmaceut Sci 87:
360-366.
44. Kansy, M., Senner, F., Gubernator, K. (1998) Physicochemical High Throughput Screening:
Parallel Artificial Membrane Permeation Assay in the Description of Passive Absorption
Processes. Journal of Medicinal Chemistry 41:1007-1010.
45. Prusiner, S. B. (1998) Proc Natl Acad Sci USA 95:13363-13383.
This article is protected by copyright. All rights reserved.

Supporting Information
Supporting information for this article contain figures: Figure S1 Decomposition of 3b to GSH
(2) and acridone (4); Figure S2 SSCA and western blot of acridone (4); Figure S3 Copy of
NMR spectrum of purified GlutQui (3a); Figure S4 Copy of NMR spectra of purified GlutQui
(3b); Figure S5 Copy of NMR spectra of GlutQui after decomposition to GSH (2) and acridone
(4); Figure S6 Copy of HRMS spectra of GlutQui (3a); Figure S7 Copy of HRMS spectrum of
GlutQui (3b).
Figure legends
Figure 1 Reaction of FDA approved drug quinacrine (1) with GSH (2). GlutQui (3) is formed.
The GlutQui is quite unstable and slowly decompose to starting GSH (2) and corresponding
acridone (4), which is practically insoluble in water.
Figure 2 Normalized 1H-NMR spectra of acridine systems, peptide amides, and ammonium salts
of GlutQui (3b) and decomposition mixture containing acridone (4) and GSH (2) using
CryoProbe. At the top spectrum of freshly prepared solution with low concentration of 3b in
order to prevent decomposition of the sample. The bottom represents acridone (4) and GSH (2)
spectrum obtained after complete decomposition. In the figure, only aromatic H are assigned, for
full assignment see also section 2.4. Spectra showing the progress of decomposition at high
concentration of 3b are depicted in Figure S1 in supplementary materials.
Figure 3 Förster resonance transfer assay for determinantion of acridine-prion protein binding.
Quinacrine (x,---), 9-aminoacridine (*, ---), GlutQui (3a, +, ---).
This article is protected by copyright. All rights reserved.

Figure 4 Toxicity of quinacrine (left) and GlutQui (3b, right) to CAD5 cells in culture. Each
concentration was tested in two independent experiments (black and red). LC50 values were 1.3
and ≥ 97 μM for quinacrine and GlutQui (3b), respectively.
Figure 5 The effect of quinacrine and GlutQui (3b) treatment on prion clearance from infected
CAD5 cells was evaluated by SSCA. Each point represents the mean from 8 values.
Concentrations were tested in two independent experiments (black and red; green and magenta).
In case of GlutQui, black and red represent treatment with aging stock solution i.e. contaminated
with acridone after the GlutQui decomposition. Green and magenta stand for application of
freshly prepared stock solution. Unfortunately, fresh GlutQui at higher concentration prevented
SSCA by formation of yellow spots on ELISPOT plate. EC50 values were 0.68, 15, and 15 μM
for quinacrine, aging GlutQui, and fresh GlutQui, respectively.
Figure 6 The effect of continuous and one dose quinacrine treatment on prion clearance from
infected CAD5 cells detected by SSCA. The green and blue circles represent the continuous
treatment with quinacrine, where the fresh solution of quinacrine was added daily after every
change of media. The brown and orange squares stand for one dose treatment, with quinacrine
added only once at the beginning of the experiment. Each point represents the mean from 8
values. Concentrations were tested in two independent experiments (replicate 1 (green and
orange) and replicate 2 (blue and brown)). From the fitting, the EC50 values for continuous and
one dose treatments were 0.68 and 1.1 μM, respectively.
Figure 7 Elimination of abnormal prion protein from RML infected CAD5 cells detected by
western blot. RML infected CAD5 cells were treated with different concentrations of quinacrine
(A every day, B only one dose) or GlutQui (3b, C every day, D only one dose) for 7 days. Cells
This article is protected by copyright. All rights reserved.