Development of new scaffolds as reversible tissue transglutaminase inhibitors, with improved potency or resistance to glutathione addition

Article abstract of DOI:10.1039/c6md00565a

Previous studies within our group have yielded a class of cinnamoyl-based competitive reversible inhibitors for tissue transglutaminase (TG2), with K_i values as low as 1.0 μM (compound CP4d). However, due to the electrophilic nature of their al

Full text of DOI:10.1039/c6md00565a

View Article Online
View Journal
MedChemComm
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: K. Y. P. Apperley, I.
Roy, V. Saucier, N. Brunet-Filion, S. Piscopo, C. Pardin, É. De Francesco, C. Hao and J. W. Keillor, Med.
Chem. Commun., 2016, DOI: 10.1039/C6MD00565A.
This is an Accepted Manuscript, which has been through the
Volume 7 Number 1 January 2016 Pages 1–204
Royal Society of Chemistry peer review process and has been
accepted for publication.
MedChemComm
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Broadening the field of opportunity for medicinal chemists
www.rsc.org/medchemcomm
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
Themed issue: Antibiotic Resistance
ISSN 2040-2503
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
rsc.li/medchemcomm

Page 1 of 24
MedChemComm
View Article Online
DOI: 10.1039/C6MD00565A
Development of new scaffolds as reversible tissue transglutaminase inhibitors, with
improved potency or resistance to glutathione addition†
Kim Y.P. Apperleya, Isabelle Royb, Vincent Sauciera, Nicholas BrunetꢀFiliona, SaraꢀPier
Piscopoa, Christophe Pardina, Élise De Francescoa, Catherine Haoa, Jeffrey W. Keillora,*
aDepartment of Chemistry and Biomolecular Sciences
University of Ottawa
10 Marie-Curie Pvt.
Ottawa, Ontario K1N 6N5
Canada
bDépartement de chimie
Université de Montréal
C. P. 6128, Succursale Centre-ville
Montréal, Québec H3C 3J7
Canada
*Corresponding author: jkeillor@uottawa.ca
Keywords
tissue transglutaminase; enzyme inhibition; cinnamoyl; triazole; glutathione
† The authors declare no competing interests.
1

MedChemComm
Page 2 of 24
View Article Online
DOI: 10.1039/C6MD00565A
Abstract
Previous studies within our group have yielded a class of cinnamoylꢀbased competitive
reversible inhibitors for tissue transglutaminase (TG2), with Ki values as low as 1.0 ꢁM
(compound CP4d). However, due to the electrophilic nature of their alkene moiety, this class of
inhibitors is susceptible to nucleophilic attack by glutathione, a key element in cellular
metabolism and toxicity response. To address this issue, we made several modifications to the
inhibitor scaffold, ultimately showing that a bis(triazole) scaffold increased resistance to
nucleophilic attack, with compound 27d being the most potent (Ki = 10 ꢁM). In the process of
reducing reactivity, we also prepared a new class of inhibitors, replacing the alkene of CP4d
with an alkyne, leading to a significant increase in potency for compound 22b (Ki = 420 nM).
2

Page 3 of 24
MedChemComm
View Article Online
DOI: 10.1039/C6MD00565A
Introduction
Transglutaminases (E.C. 2.3.2.13) are a family of calciumꢀdependent enzymes that
natively catalyse the crosslinking of proteins, through the formation of Nε(γꢀglutamyl)lysine
bridges between glutamine and lysine residues. In the absence of a lysine (or acylꢀacceptor
amine), transglutaminases can also hydrolyse glutamine side chains to form glutamate. Tissue
transglutaminase (TG2), as its name implies, is ubiquitously expressed in tissues. On the cell
surface and in the extracellular matrix, TG2 adopts an open1, 2 conformation in the presence of
calcium (Ca2+), its coenzyme required for transamidation3. Extracellular TG2 may also be
deactivated4 by disulfide bond formation, even in the presence of Ca2+. In the cytosol, where the
concentration of free calcium is low, and the concentration of guanine nucleotides is high, TG2
is allosterically regulated5 to adopt a closed conformation in which it functions as a Gꢀprotein6.
Human TG2 (hTG2) plays many physiological roles, acting in cellular differentiation7, 8,
apoptosis9, 10, and extracellular matrix assembly11, and it can also function as an adhesion
receptor12. However, the upꢀregulation of TG2ꢀmediated acylꢀtransfer has been implicated in
several physiological disorders13, such as fibrosis, cancer, and celiac disease. In light of these
pathological roles, our group has studied the development of inhibitors against TG214ꢀ16, leading
to the discovery of a family of cinnamoyl derivatives that function as reversible inhibitors. Our
original studies were done using guinea pig liver TG2 (gplTG2), which displays high sequence
homology with the human enzyme and a much greater solubility when expressed in recombinant
form, and was therefore used as a model. The most potent inhibitors of those studies were based
on a nitrocinnamoyl scaffold, such as CP30a15, CP15a15, m-CP30a17 and CP4d16, whose IC50
values are indicated in Figure 1. Following the development of a protocol for the expression and
purification of hTG2 within our group, we began to work with the human form of the enzyme,
3

MedChemComm
Page 4 of 24
View Article Online
DOI: 10.1039/C6MD00565A
and confirmed17 that the kinetic parameters such as KM and Ki values are comparable when
determined under similar reaction conditions. Earlier inhibition results are therefore shown with
gplTG2, while the later ones were obtained with hTG2.
Figure 1. 4ꢀnitrocinnamoylꢀbased reversible inhibitors for TG2
aIC50 value determined with gplTG2
Although these parent compounds are reversible inhibitors of TG2, they do contain an
electrophilic α,βꢀunsaturated carbonyl motif, which may undergo Michael addition with
nucleophiles such as glutathione (GSH), a tripeptide thiol whose intracellular concentration is in
the millimolar range. This reaction may result in a loss of potency in vivo; we therefore sought to
determine the extent of this reactivity, and to develop a family of inhibitors that would not be
susceptible to this potential nucleophilic attack.
In this work, we started from these known scaffolds and made modifications to the
electrophilic moiety to reduce its reactivity. In a first approach, we added sterically hindering
groups or increased electron density at the suspected reactive site; the replacement of a carbonꢀ
carbon double bond by a triple bond resulted in a significant increase in potency, with compound
22b being a competitive inhibitor with a Ki of 420 nM. In a second approach, we incorporated
4

Page 5 of 24
MedChemComm
View Article Online
DOI: 10.1039/C6MD00565A
the electrophilic alkene into a heteroaromatic group, resulting in increased resistance to
nucleophilic attack by glutathione, shown for compound 27d.
Results
GSH susceptibility of lead compounds
To test directly for reactivity with GSH, an incubation experiment was designed, using
m-CP30a18 (Figure 1) as a representative inhibitor for suspected Michael acceptors. Inhibitor
m-CP30a was developed and evaluated17 directly against hTG2, displaying an IC50 of 19.6 ±
2.5 ꢁM. The reaction of m-CP30a with glutathione in DMSO, the standard conditions for
1
measuring such reactivity19, was followed by H NMR by monitoring for changes from the
spectra of the starting materials (see Figure S1 in SI). The mass of the formed adduct (Scheme 1)
was then verified by HRMS, confirming monoꢀaddition. After only ten minutes, the reaction
between m-CP30a and GSH had proceeded to a significant extent, and after one hour, the
reaction was complete, with no further changes after an additional three hours, suggesting
significant GSH susceptibility within this class of inhibitors. A similar reactivity was observed
with CP4d (see Figure S2), where the reaction was more than 50% complete after 10 minutes.
Scheme 1. Suspected adduct formation between m-CP30a and glutathione
5

MedChemComm
Page 6 of 24
View Article Online
DOI: 10.1039/C6MD00565A
Synthesis and evaluation of inhibitor analogues
CβꢀAlkylated analogues of CP4d
Several compounds were therefore designed to mimic known inhibitors, while decreasing
their electrophilicity and susceptibility towards attack by GSH. Our first design featured the
alkylation of the βꢀposition of the cinnamoyl scaffold, while retaining a common benzotriazole
group for comparison with parent compound CP15a15 (Figure 1). Both methyl derivative 3 and
isopropyl derivatives 8 & 9 (Figure 2) were synthesized by a similar approach (see Scheme S1 in
SI), starting with a HornerꢀWadsworthꢀEmmons reaction from, respectively, 4ꢀnitrophenyl
methyl and 4ꢀnitrophenyl isopropyl ketones to afford the ethyl esters. While the diastereomers of
the methyl derivative were not successfully separated, their ratio of 1:1 was determined by
13C NMR spectroscopy by comparison of peak intensities. However, the isopropyl diasteromers
4 and 5 (Scheme S1) were successfully separated by flash column chromatography and identified
by NOESY, based on the interaction between the vinylic proton and the neighbouring protons on
one of the methyl groups. These esters were then saponified to yield the corresponding
carboxylic acids (2, 6, and 7; Scheme S1), according to a literature protocol20. Reaction with
oxalyl chloride led to the corresponding acyl chlorides, which were coupled directly with
benzotriazole to afford products 3, 8 and 9.
Figure 2. CβꢀAlkylated analogues of CP4d
6

Page 7 of 24
MedChemComm
View Article Online
DOI: 10.1039/C6MD00565A
Coumarin analogue of CP15a
Coumarin derivative 11 (Scheme 2) was designed as an alternative analogue in which the
α,βꢀunsaturation was incorporated into an aromatic cycle. Precursor 10 was prepared18 according
to a published procedure21 and reacted directly with thionyl chloride to form the corresponding
acyl chloride, which was coupled with benzotriazole to form 11.
Scheme 2. Synthesis of coumarin derivative 11
Alkyne analogue of CP15a
In a different approach, the cinnamoyl double bond was replaced with a triple bond
(Scheme 3). Sonogashira coupling22, 23 between 1ꢀiodoꢀ4ꢀnitrobenzene and ethyl propynoate led
to ester 12, which was saponified to the corresponding carboxylic acid 13. The acid chloride was
then formed24 in situ prior to coupling with benzotriazole to obtain compound 14, an alkyne
analogue of CP15a. The inhibition results for these benzotriazole compounds are gathered in
Table 1.
7

MedChemComm
Page 8 of 24
View Article Online
DOI: 10.1039/C6MD00565A
O
OEt
O2N
O2N
(PPh3)2PdCl2 (cat)
CuI (cat), K2CO3
O2N
LiOH
OEt
OH
anhydrous THF
5 h, 66°C
I
acetone, H2O
16 h, RT
O
12
13
O
O2N
SOCl2, benzotriazole
N
N
N
DCM
2.5 h, RT
O
14
Scheme 3. Synthesis of alkynyl derivative 14
Table 1. IC50 values for analogues of benzotriazole CP15a
Compound IC50 (µM)
74 ± 15 a
CP15a
n.d. b
3
n.d. b
8
n.d. b
9
145 ± 23
2.6 ± 0.2
11
14
a determined with gplTG2
b n.d. : inhibition not detected at the compound’s solubility limit
8

Page 9 of 24
MedChemComm
View Article Online
DOI: 10.1039/C6MD00565A
Replacement of the enone moiety of m-CP30a
The α,βꢀunsaturated carbonyl motif was also mimicked by direct incorporation of an
epoxide group or a cyclopropyl group in azachalcone m-CP30a. A CoreyꢀChaykovsky
cyclopropanation25 yielded a racemic mixture of the anti cyclopropane26 15, as determined by
NOESY. A CoreyꢀChaykovsky epoxidation was used27 to form epoxide 16, by replacing the
trimethyloxosulfonium iodide used for cyclopropanation with trimethylsulfonium iodide
(Scheme 4).
Scheme 4. Synthesis of α,βꢀunsaturated carbonyl mimics 15 and 16
Furthermore, an isoxazole mimic was prepared (17, Scheme 5), where the heterocycle
was formed regioselectively from the α,βꢀunsaturated ketone mꢀCP30a using Nꢀhydroxyꢀ4ꢀ
toluenesulfonamide, according to a published method28. The inhibition results for each of these
derivatives of mꢀCP30a are shown in Table 2.
Scheme 5. Synthesis of isoxazole derivative 17
9

MedChemComm
Page 10 of 24
View Article Online
DOI: 10.1039/C6MD00565A
Table 2. IC50 values for the analogues of azachalcone m-CP30a
Compound IC50 (µM)
19.6 ± 2.5
147 ± 10
151 ± 21
n.d. a
m-CP30a
15
16
17
a n.d. : inhibition not detected at the compound's solubility limit
Alkyne analogues of CP4d
Two alkynyl mimics of CP4d, namely 22a and 22b (Scheme 6), were also prepared, by
replacing the cinnamoyl double bond with a triple bond. Sonogashira coupling between 1ꢀiodoꢀ
4ꢀnitrobenzene and propargyl alcohol gave alcohol 18, which was oxidized with TEMPO to 19;
this aldehyde was then used in a Grignard addition with ethynylmagnesium bromide to form
alcohol 20. A second oxidation with TEMPO afforded29 ketone 21, which was then used in a
copper(I)ꢀcatalysed azideꢀalkyne cycloaddition (CuAAC), with either benzyl azide (leading to
22a) or 4ꢀnitrobenzyl azide (leading to 22b). These azides were previously prepared16 from the
reaction of the corresponding benzyl bromide with sodium azide in DMSO and used without
further purification. The inhibition results for these alkynyl derivatives are shown in Table 3.
Notably, like its parent inhibitor CP4d, compound 22b was determined via our NMR assay to be
equally, if not more, susceptible to GSH addition (see Figure S3 in SI).
10

Page 11 of 24
MedChemComm
View Article Online
DOI: 10.1039/C6MD00565A
Scheme 6. Synthesis of alkyne analogues 22a and 22b
Table 3. IC50 results for the alkyne analogues of CP4d
Compound IC50 (µM)
2.1 ± 0.3a
CP4d
0.61 ± 0.05
0.79 ± 0.05
22a
22b
a determined with gplTG2
Bis(triazole) derivatives of CP4d
In a final series of analogues, the electrophilic π bonds of inhibitors CP4d, 22a, and 22b, were
replaced with a second triazole ring (Scheme 7, Table 4). A library of derivatives was
constructed by maintaining one 4ꢀnitrobenzyl group and varying the ring substituent of the
opposite triazole. We began with a CuAAC between 4ꢀnitrobenzylazide and propargyl alcohol,
to form alcohol 23, and then used the same three steps as in the synthesis of 22, namely the first
oxidation with TEMPO to form aldehyde 24, Grignard addition with ethynylmagnesium bromide
11

MedChemComm
Page 12 of 24
View Article Online
DOI: 10.1039/C6MD00565A
to prepare alcohol 25, and a second oxidation with TEMPO to form ketone 26. For both
oxidation steps, trichloroisocyanuric acid (TCICA) was used30 as a radicalꢀregenerating reactant
instead of the previouslyꢀused (diacetoxyiodo)benzene, due to the low yields obtained with the
latter. Another CuAAC was used to form the second triazole, with various substituents, mostly
para substituted aromatics (Table 4). Given the low IC50 obtained with compound 27e, we also
prepared symmetrical compound 27m, using a similar approach as for the rest of the library.
Again, the substituted azides were prepared16 as described previously. The inhibition data for this
bis(triazole) library of compounds are shown in Table 4. Compound 27d was also analysed for
resistance to GSH addition; gratifyingly, we observed no reaction by NMR over 72 hours (see
Figure S4 in SI).
Scheme 7. Synthesis of the library of bis(triazole) compounds
12

Page 13 of 24
MedChemComm
View Article Online
DOI: 10.1039/C6MD00565A
Table 4. IC50 values for bis(triazole) compounds
Compound R group
R' group
N/A
IC50 (µM)
2.1 ± 0.3
n.d. a
N/A
ꢀNO2
ꢀNO2
ꢀNO2
ꢀNO2
ꢀNO2
ꢀNO2
ꢀNO2
ꢀNO2
ꢀNO2
ꢀNO2
ꢀNO2
ꢀNO2
ꢀCF3
CP4d
27a
27b
27c
27d
27e
27f
ꢀPh
ꢀ(2ꢀNO2)Ph 64.1 ± 4.6
> 100 b
ꢀ(3ꢀNO2)Ph
ꢀ(4ꢀNO2)Ph 14.1 ± 1.7
ꢀ(4ꢀCF3)Ph 25.9 ± 1.8
ꢀ(4ꢀOCF3)Ph
ꢀ(4ꢀCN)Ph
ꢀ(4ꢀCH3)Ph
ꢀ(4ꢀOCH3)Ph
ꢀ(4ꢀF)Ph
> 100 b
> 100 b
n.d. a
27g
27h
27i
n.d. a
n.d. a
27j
ꢀ(4ꢀBr)Ph
n.d. a
27k
27l
ꢀcyclohexyl
ꢀ(4ꢀCF3)Ph
n.d. a
n.d. a
27m
a n.d. : inhibition not detected at the compound's solubility limit
b >100 denotes weak but detectable inhibition at the compound’s solubility limit
13

MedChemComm
Page 14 of 24
View Article Online
DOI: 10.1039/C6MD00565A
Discussion
In our first series of mimics, the C=C double bond of the cinnamoyl moiety of CP15a
was modified by βꢀalkylation (Figure 2). We hypothesized that the added alkyl group at the
βꢀposition would reduce reactivity through steric hindrance and through its electronꢀdonating
effect. However, evaluation of these alkylated inhibitors against hTG2 (as described in the
Experimental Section) showed that alkylation led to a complete loss of potency (Table 1). To
verify that the absence of inhibition was not due to a combination of one isomer being much
more potent than the other, we tested compounds 8 (isopropyl, Eꢀisomer) and 9 (isopropyl, Zꢀ
isomer) separately, but noted no detectable inhibition for either one. Given their lack of
inhibitory potency, we did not evaluate these compounds for resistance against glutathione.
In a different approach to reduce electrophilicity of the α,βꢀunsaturated carbonyl motif,
we introduced two threeꢀmembered rings: a cyclopropane (15) to mimic the C=C double bond,
and an epoxide (16) to mimic the carbonyl. With analogue 15, a trans cyclopropyl group was
used to replace the trans double bond, while retaining the carbonyl. This modified compound
showed a significant loss of potency in comparison to parent m-CP30a, as did analogue 16. The
latter loss in inhibitory efficiency may be due to the significant change in geometry around the
oxygen. Furthermore, despite the replacement of the carbonyl with an epoxide, 16 is arguably
equally as susceptible to nucleophilic attack as m-CP30a; addition to 16 could occur either by
direct nucleophilic attack on the epoxide, or by an SN2' mechanism, where nucleophilic attack on
the β carbon of the alkene moiety occurs prior to the opening of the epoxide. Considering how
weakly these analogues inhibit hTG2 (Table 2), we did not evaluate them for glutathione
resistance.
14

Page 15 of 24
MedChemComm
View Article Online
DOI: 10.1039/C6MD00565A
In our next attempt to design potent inhibitors with diminished susceptibility to
glutathione addition, the α,βꢀdouble bond of the parent inhibitor was incorporated into a
heteroaromatic ring. A coumarin group was used as a mimic of the cinnamoyl group in CP15a,
yielding compound 11. Given its 10ꢀfold loss of potency compared to the parent compound, we
did not test it for glutathione resistance. We also incorporated an isoxazole ring (17) in parent m-
CP30a, where the heterocycle has increased stability as conferred by aromaticity. Furthermore,
unlike analogue 16, derivative 17 maintains the sp2 hybridisation state of the carbonylꢀ
mimicking oxygen, which helps maintain a similar compound geometry.
However, this compound showed no inhibition of hTG2, possibly due to the geometric
constraints associated with the introduction of the isoxazole. Again, given the loss in inhibitory
potency (Table 1 and Table 2), we did not evaluate these compounds for resistance to
glutathione.
In an effort to remain closer in structure to parent compound CP15a, we replaced its
double bond with a triple bond, leading to compound 14. We hypothesized that the greater
electronic density surrounding the spꢀhybridised carbons may decrease their electrophilic
reactivity with glutathione. This modified inhibitor was the first to display increased potency
(Table 1) when compared to its parent CP15a. Encouraged by these initial results, we also
prepared an alkyne analogue of CP4d, 22b, and an analogue without the nitro group on the
triazoleꢀside benzyl group, 22a. Similar substitutions were done in the original study16 and,
since nitro groups are generally disfavoured31 in drug development due to the significant
secondary effects related to their metabolism, we wanted to evaluate the importance of the
second nitro group on this scaffold. Just like ynone 14, both these compounds showed high
potency against hTG2 (Table 3). We also confirmed that, like its parent CP4d, 22b is also a
15

MedChemComm
Page 16 of 24
View Article Online
DOI: 10.1039/C6MD00565A
1
competitive inhibitor (Table 5). While these alkynyl compounds were observed by H NMR to
be as susceptible to GSH addition as the parent CP4d, they remain the most potent inhibitors we
have prepared to date.
Finally, starting from CP4d and 22b, we replaced the electrophilic double bond with a
second triazole group, resulting in the bis(triazole) 27d. We reasoned that the aromaticity of the
heterocycle would diminish the double bond’s electrophilicity and confer resistance against GSH
addition. In accordance with that hypothesis, we noted no adduct formation over 72 hours in our
1
GSH H NMR assay (Figure S4 in SI). Following this positive result, we prepared a library of
analogues. Considering that CP15a and CP4d have a paraꢀnitro substitution, but that compound
m-CP30a has a metaꢀnitro group, we first determined the optimal position for substitution on
this scaffold. While all three of 27b, 27c, and 27d showed inhibition towards hTG2, unlike the
monoꢀnitro parent 27a, the results show that, for this scaffold as well, substitution at the para
position is optimal for inhibitory potency. Based on the encouraging results with 27e, we also
prepared the symmetrical compound 27m. However, the replacement of the second nitro group
with a trifluoromethyl group resulted in complete loss of inhibition.
In comparing each of the most potent scaffolds (alkene, alkyne and triazole), we
confirmed that the competitive inhibition observed with parent CP4d was maintained with each
of the dinitro analogues 22b and 27d,. Apparent MichaelisꢀMenten constants (KMapp) were
determined by varying substrate (AL5) concentrations for a series of inhibitor concentrations
app
(Figure 2A for 22b, Figure S5 in SI for 27d). Plotting both KM and Vmax against inhibitor
concentration (Figure 2B, Figure S6 in SI for 27d) confirmed the competitive mode of inhibition,
and allowed us to determine Ki. A similar approach was used with 27d, leading to the data shown
in Table 5.
16

Page 17 of 24
MedChemComm
View Article Online
DOI: 10.1039/C6MD00565A
Figure 3. Determination of the inhibition constant for 22b a) MichaelisꢀMenten curves at various
concentrations of inhibitor and b) Determination of the mode of inhibition and value of Ki
17

MedChemComm
Page 18 of 24
View Article Online
DOI: 10.1039/C6MD00565A
Table 5. Inhibition constants (Ki) for selected inhibitors towards human TG2
Compound Ki (µM)
1.0 ± 0.117
CP4d
0.42 ± 0.05
9.9 ± 1.6
22b
27d
While 27d showed a reduction in affinity of one order of magnitude compared to parent
CP4d, this decrease is accompanied by an increased resistance to GSH addition, which is a
significant concern in the targeting of intracellular hTG2. In comparison, inhibitor 22b is our
most potent reversible hTG2 inhibitor to date, with the first subꢀmicromolar Ki value, at 420 nM.
This compound would be particularly appropriate for the inhibition of extracellular hTG2, where
the concentration of GSH is much lower and less likely to react with our compound.
Conclusion
1
In summary, we have developed a rapid HꢀNMRꢀbased assay to measure the reactivity
of our reversible hTG2 inhibitors with GSH. Given the significant extent of reactivity observed
with the parent inhibitors, their scaffolds were modified to reduce their electrophilicity. In this
study, two notable new scaffolds were developed: alkynes and bis(triazoles). The former
compounds, including 22b with a Ki of 420 nM, show great promise as competitive inhibitors of
extracellular hTG2, given that the extracellular concentration of GSH is one to three orders of
magnitude lower than inside cells, reaching values below 5 ꢁM in plasma32. The latter scaffolds
show potential for targeting intracellular hTG2 by virtue of their increased resistance to GSH
addition. Future work will involve development of these scaffolds and in cellulo evaluation.
18

Page 19 of 24
MedChemComm
View Article Online
DOI: 10.1039/C6MD00565A
Acknowledgements
This work was supported financially by the Natural Sciences and Engineering Research Council
of Canada (NSERC). K.Y.P.A. is grateful to the University of Ottawa for an Admission
(Excellence) scholarship and to the Fonds de Recherche Québécois: Nature et Technologies
(FRQNT) for a doctoral scholarship. I.R. is grateful to NSERC and the Fonds Québecois de
Recherche sur la Nature et les Technologies (FQRNT) for postgraduate scholarships. V.S., N.B.ꢀ
F., S.ꢀP.P. and E.D.F. are grateful to NSERC for Undergraduate Summer Research Awards
(USRA).
Supporting Information
Synthetic methods, spectra for the NMR assay and graphs for the determination of the Ki value
for 27d are included.
Kinetic methods
Evaluation of resistance to glutathione by 1H NMR
1H NMR spectra of a solution of 12 mM inhibitor and 80 mM glutathione in DMSOꢀd6
(1 mL) were recorded on a Bruker AVANCE 300 spectrometer at the indicated time points.
Spectra were collected for each substrate individually, as well as after combining the inhibitor
and GSH, at 10 minutes, one hour, two hours and four hours to monitor any changes different
from an additive spectrum. The product was then analysed by mass spectrometry to confirm the
presence the proposed glutathione adduct.
HRMS (ESI+) for 1: [C24H27N5O9S+Na]+ calc. 584.1422 Da, obt. 584.1500 Da
19

MedChemComm
Page 20 of 24
View Article Online
DOI: 10.1039/C6MD00565A
Determination of IC50 values
Kinetic runs were recorded on a Varian Cary 100 Bio UVꢀvisible spectrophotometer at
405 nm and 25°C, in a buffer composed of 111 mM MOPS (pH 7.0), 3.33 mM CaCl2, and
0.05 mM EDTA. All aqueous solutions were prepared using deionised water. Each kinetic assay
was performed using 900 ꢁL buffer, 25 ꢁL of a DMF stock solution of substrate NꢀCbzꢀGlu(γꢀpꢀ
nitrophenylester)Gly (AL5)33 (1 mg/mL, 2.2 mM), 50 ꢁL of a solution of hTG2 at 0.05 U/mL
(final concentration of 2.5 mU/mL) and 0ꢀ25 ꢁL of a DMF stock solution of inhibitor (contingent
on solubility; the highest concentration tested was generally between 5 and 30 ꢁM). The volume
of DMF was then adjusted so that it represented 5% of the final volume. The reaction was
initiated with the addition of hTG2 (expressed and purified according to a literature protocol17),
and the increase in absorbance was monitored over 10 minutes; initial rates were measured over
the first five minutes of absorbance change. IC50 values were obtained as the negative xꢀintercept
of a Dixon plot (reciprocal of initial rates vs. inhibitor concentration).
Determination of Ki values
Kinetic runs were recorded on a UVꢀvisible spectrophotometer at 405 nm and 25°C, in a
buffer composed of 111 mM MOPS (pH 7.0), 3.33 mM CaCl2, and 0.05 mM EDTA. All aqueous
solutions were prepared using deionised water. Stock solutions of AL5 (1 mg/mL, 2.2 mM) and
inhibitor (0.65 mg/mL, 1.50 mM for compound 27d and 0.04 mg/mL, 0.11 mM for 22b) were
prepared in DMF. For each solution, the volume of DMF was adjusted to represent 5% of the
final volume, and 2.5 mU/mL of hTG2 were used to run the assay. MichaelisꢀMenten curves
were determined using AL5 concentrations of 6.5 ꢁM, 14 ꢁM, 28 ꢁM and 70 ꢁM (representing
0.5, 1, 2 and 5 times KM (14 ꢁM18)), and inhibitor concentrations of 1.5 ꢁM, 3.0 ꢁM, 6.0 ꢁM and
15 ꢁM for compound 27d or 0.11 ꢁM, 0.21 ꢁM, 0.42 ꢁM and 1.1 ꢁM for 22b. The reaction was
20

Page 21 of 24
MedChemComm
View Article Online
DOI: 10.1039/C6MD00565A
initiated with the addition of hTG2 (expressed and purified according to a literature protocol17),
and the increase in absorbance was monitored over 10 minutes; initial rates were measured over
the first five minutes of absorbance change. MichaelisꢀMenten curves were fitted by nonꢀlinear
regression to equation 1, using Graph Pad Prism 6, to obtain KMapp values, while Ki values were
then determined by linear regression of KMapp vs. inhibitor concentration, using equation 2.
ꢀ
ꢁ
AL5
V
max
v =
Eq. 1
Eq. 2
app
M
ꢀ
ꢁ
K
+ AL5
KaMpp = KM ꢂ1+ ꢀinhibitor ꢃ =
ꢁ
K
M ꢀ
ꢁ
inhibitor + KM
K
i
K
i
21

MedChemComm
Page 22 of 24
View Article Online
DOI: 10.1039/C6MD00565A
References
1.
C. M. Clouthier, G. G. Mironov, V. Okhonin, M. V. Berezovski and J. W. Keillor,
Angew. Chem., Int. Ed., 2012, 51, 12464ꢀ12468.
2.
3.
4.
D. M. Pinkas, P. Strop, A. T. Brunger and C. Khosla, PLoS Biol., 2007, 5, 2788ꢀ2796.
L. Fesus and M. Piacentini, Trends Biochem. Sci., 2002, 27, 534ꢀ539.
X. Jin, J. Stamnaes, C. Kloeck, T. R. Di Raimondo, L. M. Sollid and C. Khosla, J. Biol.
Chem., 2011, 286, 37866ꢀ37873.
5.
J. W. Keillor, C. M. Clouthier, K. Y. P. Apperley, A. Akbar and A. Mulani, Bioorg.
Chem., 2014, 57, 186ꢀ197.
6.
7.
M. Griffin, R. Casadio and C. M. Bergamini, Biochem. J., 2002, 368, 377ꢀ396.
J. Tucholski, M. Lesort and G. V. W. Johnson, Neuroscience (Oxford, United Kingdom),
2001, 102, 481ꢀ491.
8.
Z. Balajthy, K. Csomos, G. Vamosi, A. Szanto, M. Lanotte and L. Fesus, Blood, 2006,
108, 2045ꢀ2054.
9.
Z. Sarang, P. Molnar, T. Nemeth, S. Gomba, T. Kardon, G. Melino, S. Cotecchia, L.
Fesus and Z. Szondy, Hepatology (Hoboken, NJ, U. S.), 2005, 42, 578ꢀ587.
L. K. Boroughs, M. A. Antonyak and R. A. Cerione, J. Biol. Chem., 2014, 289, 10115ꢀ
10125.
10.
11.
12.
G. Melino, M. AnnicchiaricoꢀPetruzzelli, L. Piredda, E. Candi, V. Gentile, P. J. A.
Davies and M. Piacentini, Mol. Cell. Biol., 1994, 14, 6584ꢀ6596.
C. A. Gaudry, E. Verderio, D. Aeschlimann, A. Cox, C. Smith and M. Griffin, J. Biol.
Chem., 1999, 274, 30707ꢀ30714.
13.
14.
L. Lorand and R. M. Graham, Nat. Rev. Mol. Cell Biol., 2003, 4, 140ꢀ156.
C. Pardin, S. M. F. G. Gillet and J. W. Keillor, Bioorg. Med. Chem., 2006, 14, 8379ꢀ
8385.
15.
16.
17.
C. Pardin, J. N. Pelletier, W. D. Lubell and J. W. Keillor, J. Org. Chem., 2008, 73, 5766ꢀ
5775.
C. Pardin, I. Roy, W. D. Lubell and J. W. Keillor, Chem. Biol. Drug Des., 2008, 72, 189ꢀ
196.
I. Roy, O. Smith, C. M. Clouthier and J. W. Keillor, Protein Expres. Purif., 2013, 87, 41ꢀ
46.
22

Page 23 of 24
MedChemComm
View Article Online
DOI: 10.1039/C6MD00565A
18.
19.
I. Roy, Ph.D., Université de Montréal, 2014.
B. Budke, J. H. Kalin, M. Pawlowski, A. S. Zelivianskaia, M. Wu, A. P. Kozikowski and
P. P. Connell, J. Med. Chem., 2013, 56, 254ꢀ263.
20.
E. Carceller, J. Salas, M. Merlos, M. Giral, R. Ferrando, I. Escamilla, J. Ramis, J. Garciaꢀ
Rafanell and J. Forn, J. Med. Chem., 2001, 44, 3001ꢀ3013.
21.
22.
23.
24.
25.
26.
27.
28.
29.
A. Song, X. Wang and K. S. Lam, Tetrahedron Lett., 2003, 44, 1755ꢀ1758.
K. Sonogashira, Y. Tohda and N. Hagihara, Tetrahedron Lett., 1975, 4467ꢀ4470.
T. Eckert and J. Ipaktschi, Synth. Commun., 1998, 28, 327ꢀ335.
A. R. Katritzky, Y. Zhang and S. K. Singh, Synthesis, 2003, 2795ꢀ2798.
E. J. Corey and M. Chaykovsky, J. Am. Chem. Soc., 1965, 87, 1353ꢀ1364.
R. J. Paxton and R. J. K. Taylor, Synlett, 2007, 633ꢀ637.
E. J. Corey and M. Chaykovsky, J. Am. Chem. Soc., 1962, 84, 3782ꢀ3783.
S. Tang, J. He, Y. Sun, L. He and X. She, Org. Lett., 2009, 11, 3982ꢀ3985.
D. J. Vugts, L. Veum, K. alꢀMafraji, R. Lemmens, R. F. Schmitz, F. J. J. de Kanter, M. B.
Groen, U. Hanefeld and R. V. A. Orru, Eur. J. Org. Chem., 2006, 1672ꢀ1677.
S. I. Presolski, V. Hong, S.ꢀH. Cho and M. G. Finn, J. Am. Chem. Soc., 2010, 132,
14570ꢀ14576.
30.
31.
C.ꢀH. Hsu, T. Stedeford, E. OkochiꢀTakada, T. Ushijima, H. Noguchi, C. MuroꢀCacho, J.
W. Holder and M. Banasik, J. Environ. Sci. Health, Part C Environ. Carcinog.
Ecotoxicol. Rev., 2007, 25, 155ꢀ184.
32.
33.
B. Halliwell and J. M. C. Gutteridge, Free Radicals in Biology and Medicine, Oxford
University Press, 2015.
A. Leblanc, C. Gravel, J. Labelle and J. W. Keillor, Biochemistry, 2001, 40, 8335ꢀ8342.
23

MedChemComm
Page 24 of 24
View Article Online
DOI: 10.1039/C6MD00565A
Graphical Table of Contents entry
"CP4d"
NO2
O2N
N
N
N
improved
glutathione
resistance
O
improved
potency
NO2
O2N
NO2
O2N
N
N
N
N
N
N
N
N
N
22b
27d
O
O
Starting from known reversible inhibitor CP4d, we have designed two new classes of
inhibitors, improving both potency (22b) and glutathione resistance (27d).