Synthesis and evaluation of peptidic maleimides as transglutaminase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2006.10.061

A series of novel transglutaminase inhibitors was prepared, based on the scaffold of a commonly used peptide substrate and bearing an electrophilic maleimide group. These compounds were evaluated in vitro and shown to lead to irreversible inactivation of tissue transglutaminase. Comparison with inhibitors studied previously provides insight into the steric environment of the enzyme active site.

Full text of DOI:10.1016/j.bmcl.2006.10.061

Bioorganic & Medicinal Chemistry Letters 17 (2007) 305–308
Synthesis and evaluation of peptidic maleimides as
transglutaminase inhibitors
Dany Halim, Karine Caron and Jeffrey W. Keillor*
De´partement de chimie, Universite´ de Montre´al, C.P. 6128, Succursale centre-ville, Montre´al, Que., Canada H3C 3J7
Received 6 September 2006; revised 11 October 2006; accepted 23 October 2006
Available online 25 October 2006
Abstract—A series of novel transglutaminase inhibitors was prepared, based on the scaffold of a commonly used peptide substrate
and bearing an electrophilic maleimide group. These compounds were evaluated in vitro and shown to lead to irreversible inacti-
vation of tissue transglutaminase. Comparison with inhibitors studied previously provides insight into the steric environment of
the enzyme active site.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Transglutaminases (TGases, EC 2.3.2.13) are Ca2+-de-
pendent enzymes that catalyze the formation of isopep-
tide cross-links between the c-carboxamide group of a
protein- or peptide-bound glutamine residue and a pri-
mary amino group, such as the e-amino group of pro-
tein- or peptide-bound lysine residues (Scheme 1).1–3
Tissue TGase (tTG) has been identified as a contributor
to the formation of cataracts and to Celiac disease, and
a growing body of evidence suggests that it may be in-
volved in atherosclerosis, inflammation, fibrosis, diabe-
tes, cancer metastases, autoimmune diseases, lamellar
ichthyosis, and psoriasis (for a review, see 4). TGase
has also been implicated in neurodegenerative diseases
associated with an increase in polyglutamine-containing
peptides in the brain such as Huntington’s disease,
Alzheimer disease, Parkinson disease, and supranuclear
palsy.5–7
based on the same peptidic scaffold, and their in vitro
evaluation as irreversible tTG inhibitors (Scheme 2).
The starting points of our syntheses of the peptidic
maleimides were the Cbz-protected diamino acids 2a–
4a, numbered according to side-chain length. Com-
pounds 3a and 4a were commercially available,14
whereas compound 2a was obtained from the Hoffmann
rearrangement of the corresponding amide, as previous-
ly reported.10 Ensuing protection of the pendant amino
group,15 esterification16 or peptide coupling17 and
deprotection of the side-chain amine18 were accom-
plished by straightforward synthetic routes, affording
amines 2c–4c and 2d–4d. Hydrophobic ester groups
were incorporated into the inhibitor design both for syn-
thetic simplicity and in consideration of their effect on
enzyme affinity as observed previously.8,11,12 These
amines were then transformed into the final maleimide
inhibitors by the typical two-step condensation reaction
with maleic anhydride.18,19 Reaction conditions and
yields were not optimized, but ample quantities of the fi-
nal products were obtained for subsequent kinetic
analyses.
A number of potential TGase inactivators have been
developed in order to regulate excess TGase activity.
These include dihydroisoxazole derivatives,8 gluten pep-
tide analogs,9 and dipeptide-bound a,b-unsaturated
amides, epoxides, and 1,2,4-thiadiazoles.10–12 The design
of the latter inhibitors was based on the structure of car-
bobenzyloxy-L-glutaminylglycine (Cbz-Gln-Gly),
a
Recombinant guinea pig liver TGase was expressed in
Escherichia coli and subsequently purified according to
a procedure developed in our laboratories.20,21 Guinea
pig liver tTG was chosen for this study because it can
be obtained easily in excellent yield and solubility, and
shows high homology with human tTG,22 thereby vali-
dating its use as a model for the evaluation of inhibitors
of potential therapeutic utility.
commonly used dipeptide acyl-donor substrate.13 Here-
in we report the synthesis of a series of novel maleimides
Keywords: Transglutaminase; Maleimide; Inhibition kinetics.
*
Corresponding author. Tel.: +1 514 343 6219; fax: +1 514 343
7586; e-mail: jw.keillor@umontreal.ca
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.10.061

306
D. Halim et al. / Bioorg. Med. Chem. Lett. 17 (2007) 305–308
Scheme 1. Cross-linking reaction catalyzed by TGase.
Scheme 2. Synthesis of inhibitors studied herein. Reagents and conditions: (A) 7 (Boc)2O/Et3N/MeOH/24 h, rt; (B) Bn–Br/Et3N/DMF/24 h, rt; (C)
Gly-OMe/DIEA/TBTU/DMF/24 h, rt; (D) i—TFA/1 h, rt; ii—maleic anhydride/CHCl3/24 h, rt; iii—acetic anhydride/NaOAc/24 h, 95 ꢁC.
All peptidic maleimides demonstrated time-dependent
inhibition. Kinetic parameters were determined by incu-
bating TGase with inhibitor and measuring residual
TGase activity in the absence of excess inhibitor (upon
40-fold dilution) as a function of time.23 First-order rate
constants of inactivation (kobs) were thus measured with
respect to inhibitor concentration. However, saturating
concentrations could not be attained, owing to the rela-
tively low solubility of the inhibitors with respect to their
apparent affinity constants. Therefore, the apparent
second-order rate constants for inactivation (kinact/KI,
Table 1) were determined from the initial slopes of the
plots of kobs versus inhibitor concentration.23
Despite the lack of saturation kinetics, two lines of evi-
dence suggest that inactivation is taking place through
reaction at the active site. First, although homology
modeling24 of guinea pig tTG suggests that Cys229
may be solvent exposed, and therefore theoretically
capable of reacting with a maleimide in a second-order
fashion, its distance from the substrate binding site
˚
(>15 A) makes its putative reaction with even the lon-
gest inhibitor unlikely to inactivate the enzyme. Second,
incubation with inhibitor in the presence of substrate
afforded temporary protection, slowing inactivation
appreciably. For example, in the presence of 12 mM
Cbz-Gln-Gly (four times its Km value) inactivation by
100 lM 2e was roughly 2-fold slower than in the absence
of substrate.
Table 1. Kinetic parameters determined for inhibition of tTG
Compound kinact/KI (mMÀ1 minÀ1 a
)
kinact (minÀ1
)
KI (lM)
Comparison of the relative efficiencies of benzyl esters
2e–4e and dipeptides 2f–4f reveals the latter series to
be ꢀ2- to 20-fold less efficient. Previous docking studies
from our group25 have suggested the peptide backbone
of acyl-donor substrates are bound in a shallow, rather
hydrophobic groove on the surface of the enzyme. The
specificity of tTG with respect to the peptide sequence
of its Gln substrates is consistent with the nature of this
putative binding site, as noted previously.26 The greater
efficiency of the benzyl esters, compared to the glycine
methyl esters, may reflect higher affinity for, or better
positioning in, this binding site.
2e
3e
4e
2f
3f
4f
0.67 0.08
6.26 1.69
17.08 0.12
0.28 0.03
0.43 0.05
0.83 0.01
—
—
—
—
—
—
—
—
—
—
—
—
2gb
3gb
4gb
1180
890
0.60
0.75
0.49
0.51
0.85
0.23
2200
a Std. error from fitting of data.
b Kinetic parameters given as ‘kinact’, ‘kcat’ and KI, respectively, in Ref.
11.

D. Halim et al. / Bioorg. Med. Chem. Lett. 17 (2007) 305–308
307
Acknowledgments
The authors acknowledge the financial support of the
Natural Sciences and Engineering Research Council of
Canada (NSERC). K.C. is also grateful to NSERC for
a Undergraduate Student Research Award (USRA).
Figure 1. Acrylamide inhibitors studied previously.10,11
Supplementary data
Chromatography procedures and spectral data of all
synthetic intermediates are available. Supplementary
data associated with this article can be found, in the on-
line version, at doi:10.1016/j.bmcl.2006.10.061.
Within each series it can be noted that efficiency increas-
es with chain length. This trend has also been observed
for three other series of inhibitors based on the same
peptide scaffold.10,12 Since TGases are designed to ex-
clude asparagine residues as acyl-donor substrates,
reacting only with glutamine residues, they are capable
of discriminating against substrate analogs having short
side chains. However, tTG cannot exclude longer acyl-
donor substrates (or irreversible inhibitors) having the
conformational flexibility necessary to properly position
their pendant reactive groups near the active site thiol.
References and notes
1. Folk, J. E.; Cole, P. W. Biochim. Biophys. Acta 1996, 22,
244.
2. Achyuthan, K. E.; Slaughter, T. F.; Santiago, M. A.;
Enghild, J. J.; Greenberg, C. A. J. Biol. Chem. 1993, 268,
21284.
3. Greenberg, C. S.; Birkbichler, P. J.; Rice, R. H. FASEB J.
1991, 5, 3071.
4. Kim, S.-Y.; Jeitner, T. M.; Steinert, P. M. Neurochem. Int.
2002, 40, 85.
5. Cooper, A. J. L.; Jeitner, T. M.; Gentile, V.; Blass, J. P.
Neurochem. Int. 2002, 40, 53.
6. Singer, S. M.; Zainelli, G. M.; Norlund, M. A.; Lee, J. M.;
Muma, N. A. Neurochem. Int. 2002, 40, 17.
7. Karpuj, M. V.; Becker, M. W.; Steinman, L. Neurochem.
Int. 2002, 40, 31.
8. Choi, K.; Siegel, M.; Piper, J. L.; Yuan, L.; Cho, E.;
Strnad, P.; Omary, B.; Rich, K. M.; Khosla, C. Chem.
Biol. 2005, 12, 469.
9. Hausch, F.; Halttunen, T.; Ma¨ki, M.; Khosla, C. Chem.
Biol. 2003, 10, 225.
It is also instructive to draw a comparison with a series
of acrylamide inhibitors studied previously. As shown in
Figure 1, acrylamides 2g–4g also comprise a dipeptide
scaffold having a comparable C-terminal ester, and their
pharmacophore differs from maleimide by only one car-
bonyl completing the heterocycle. Compounds 2g–4g
were determined11 to have KI values in the range of
0.23–0.85 lM, and their efficiency constants are shown
in Table 1. From this comparison it is evident that the
maleimides are ꢀ103- to 104-fold less efficient inhibitors.
Although direct comparison of reactivity and affinity
constants is not possible, one may presume that the
maleimides possess inherently greater reactivity (larger
kinact values), given the activation of the double bond to-
ward nucleophilic addition by two conjugated carbonyl
groups. By this reasoning, it would appear that the affin-
ity constants of the maleimides are well above those
measured for the acrylamides (even larger KI values).
This is consistent with the lack of complete saturation
that was observed for the maleimides, for concentrations
up to 750 lM. Our modeling of the glutamine substrate
binding tunnel of tTG25 suggests that it is a sterically
constrained environment that may not easily accommo-
date the greater volume of the cyclic maleimide group.
Furthermore, this hypothesis is supported by our obser-
vation that any substitution for example, by methyl or
phenyl, on the c-carboxamide nitrogen of glutamine re-
sults in the complete loss of donor substrate activity.27
´
10. de Macedo, P.; Marrano, C.; Keillor, J. W. Bioorg. Med.
Chem. 2002, 10, 355.
´
11. Marrano, C.; de Macedo, P.; Keillor, J. W. Bioorg. Med.
Chem. 2001, 9, 1923.
12. Marrano, C.; de Macedo, P.; Gagnon, P.; Lapierre, D.;
´
Gravel, C.; Keillor, J. W. Bioorg. Med. Chem. 2001, 9,
3231.
13. Folk, J. E.; Chung, S. I. Methods Enzymol. 1985, 113, 358.
14. Starting materials were obtained from Sigma–Aldrich.
15. Procedure A: Boc protection. Compounds 2a, 3a, and 4a
(1.75 mmol) were dissolved in 10 mL of methanol. After
the addition of 1.45 mL of triethylamine (10.5 mmol),
460 mg of (Boc)2O (2.10 mmol) and a few drops of 1 M
NaOH were added. The mixture was stirred overnight at
room temperature and then evaporated under reduced
pressure. The residue was dissolved in 20 mL of 1 N
NaOH. The aqueous phase was washed with 3· 20 mL
CH2Cl2 and acidified to pH ꢀ1.5 by the addition of 6 N
HCl. The product was extracted with 3· 25 mL EtOAc
and the organic layer was dried over MgSO4, filtered, and
evaporated under reduced pressure to give the final
product (36–81% yield).
In summary, the series of novel maleimide inhibitors
presented herein confirm the validity of the peptidic
scaffold, bearing an electrophilic ‘warhead’ on a long
side-chain, for the design of small molecule inhibitors
that target the active site of tTG. The maleimide group
itself, although well known for inactivation of thiol-de-
pendent enzymes, appears to be just large enough to de-
crease the ease of its insertion and productive
orientation in the narrow donor substrate binding tun-
nel of tTG.
16. Procedure B: Benzyl ester formation. In the dark, com-
pounds 2b, 3b, and 4b (1.15 mmol) were dissolved in
10 mL DMF. After the addition of 0.17 mL of benzyl
bromide (1.38 mmol), 0.24 mL of triethylamine
(1.73 mmol) was added. The mixture was stirred overnight
at room temperature. The reaction mixture was then

308
D. Halim et al. / Bioorg. Med. Chem. Lett. 17 (2007) 305–308
evaporated under reduced pressure and the residue was
purified by flash chromatography (66–92% yield).
established in our group.20 One unit of TGase was defined
as the amount of enzyme that catalyzes the formation of
1.0 lmol of hydroxamate per min (based on the hydroxa-
mate activity assay, where Cbz-Gln-Gly is used as a
c-glutamyl donor substrate and hydroxylamine is used as
an acyl acceptor substrate20). All materials were of reagent
grade purity and obtained from Sigma–Aldrich Chemical
Company. Water was purified using a Millipore BioCell
water purification system.
17. Procedure C: Peptide coupling. Under nitrogen atmo-
sphere, compounds 2b, 3b, and 4b (3 mmol) were dissolved
in 10 mL of anhydrous acetonitrile. After the addition of
2.59 mL DIEA (15 mmol), 744 mg of Gly-OMe (6 mmol)
was added. Finally, the reaction was initiated by the
addition of 2.39 g TBTU (7.5 mmol). The mixture was
stirred overnight at room temperature. The reaction
mixture was then evaporated under reduced pressure and
the residue was purified by flash chromatography (59–70%
yield).
22. Leblanc, A.; Gravel, C.; Labelle, J.; Keillor, J. W.
Biochemistry 2001, 40, 8335.
23. Assay procedure. The progress of the inhibition reactions
studied was determined by following the loss of activity as
a function of time of incubation with inhibitor. Incubation
solutions were composed of 10 lL inhibitor (50–750 lM
final concentration), 200 lL TGase (150 lg/mL), 12 lL of
100 mM CaCl2, 168 lL of 200 mM Mops (pH 7), and
10 lL of 20 mM EDTA (total volume 400 lL). At various
incubation times a 25-lL aliquot was removed and added
to a solution composed of 25 lL stock substrate solution22
(2.18 mM N-Cbz-Glu(c-p-nitrophenyl ester)Gly in DMF),
50 lL water, and 900 lL of buffer (100 mM Mops (pH 7),
3 mM CaCl2, and 0.05 mM EDTA final concentrations).
Residual TGase activity was measured at 37 ꢁC as an
increase of absorbance at 410 nm. The rate constant for
the loss of residual activity as a function of time (kobs) was
determined from fitting the data to a single exponential
equation, for each concentration of inhibitor. Plots of kobs
against inhibitor concentration did not show saturation
kinetics and were subsequently fitted to a linear equation,
the slopes giving the ratios (kinact/KI).
18. Procedure D: BOC deprotection and maleimide formation.
Compounds 2c, 3c, 4c, 2d, 3d, and 4d (2 mmol) were
dissolved in 8 mL TFA. After 1 h of reaction, acetone and
cyclohexane were added in 1:5 proportion to make an
azeotrope solution. This solution was then evaporated
three or four times under reduced pressure. The residue
containing the free amine was finally dried under vacuum
overnight. Then, under nitrogen atmosphere, it was
dissolved in 10 mL of chloroform. Finally, 217 mg of
maleic anhydride (2.2 mmol) was added and the reaction
mixture was stirred for 20 h at room temperature. The
solution was then evaporated under reduced pressure. At
this point, the product maleamide can be isolated by flash
chromatography, or used without further purification in
the ring closure step: Still under nitrogen atmosphere,
10 mL of acetic anhydride and 0.6 mmol of sodium acetate
were added to the residue. The mixture was stirred at
95 ꢁC overnight. The final product was purified by flash
chromatography (11–46% yield for three steps).
19. Girouard, S.; Houle, M.-H.; Grandbois, A.; Keillor, J. W.;
Michnick, S. W. J. Am. Chem. Soc. 2005, 127, 559.
20. Gillet, S. M. F. G.; Chica, R. A.; Keillor, J. W.; Pelletier,
J. N. Protein Expr. Purif. 2004, 33, 256.
21. Enzyme preparation. Recombinant guinea pig liver tTG
was expressed and purified according to the protocol
24. Chica, R.A .; Keillor, J. W.; Pelletier, J. N. unpublished
observations.
25. Chica, R. A.; Gagnon, P.; Keillor, J. W.; Pelletier, J. N.
Protein Sci. 2004, 13, 979.
26. Keillor, J. W. Chem. Biol. 2005, 12, 410.
´ ´
27. Day, N. M.Sc. Thesis, Universite de Montreal, 1999.