In situ selection of lead compounds by click chemistry: Target-guided optimization of acetylcholinesterase inhibitors

Article abstract of DOI:10.1021/ja043031t

The target-guided, in situ click chemistry approach to lead discovery has been successfully employed for discovering acetylcholinesterase (AChE) inhibitors by incubating a selected enzyme/tacrine azide combination with a variety of acetylene reagents that were not previously known to interact with the enzyme's peripheral binding site. The triazole products, formed by the enzyme, were identified by HPLC-mass spectrometry analysis of the crude reaction mixtures. The target-guided lead discovery search was also successful when performed with reagent mixtures containing up to 10 components. From 23 acetylene reagents, the enzyme selected two phenyltetrahydroisoquinoline (PIQ) building blocks that combined with the tacrine azide within the active center gorge to form multivalent inhibitors that simultaneously associate with the active and peripheral binding sites. These new inhibitors are up to 3 times as potent as our previous phenylphenanthridinium-derived compounds, and with dissociation constants as low as 33 femtomolar, they are the most potent noncovalent AChE inhibitors known. In addition, the new compounds lack a permanent positive charge and aniline groups and possess fewer fused aromatic rings. Remarkably, despite the high binding affinity, the enzyme displayed a surprisingly low preference for one PIQ enantiomer over the other.

Full text of DOI:10.1021/ja043031t

Published on Web 04/19/2005
In Situ Selection of Lead Compounds by Click Chemistry:
Target-Guided Optimization of Acetylcholinesterase Inhibitors
Antoni Krasin´ski,† Zoran Radic´,‡ Roman Manetsch,† Jessica Raushel,†
Palmer Taylor,‡ K. Barry Sharpless,†,§ and Hartmuth C. Kolb*,†,§
Contribution from the Department of Chemistry and the Skaggs Institute for Chemical Biology,
The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037, and
the Department of Pharmacology, #0636, UniVersity of California, San Diego,
9500 Gilman DriVe, La Jolla, California 92093
Received November 18, 2004; E-mail: hckolb@scripps.edu
Abstract: The target-guided, in situ click chemistry approach to lead discovery has been successfully
employed for discovering acetylcholinesterase (AChE) inhibitors by incubating a selected enzyme/tacrine
azide combination with a variety of acetylene reagents that were not previously known to interact with the
enzyme’s peripheral binding site. The triazole products, formed by the enzyme, were identified by HPLC-
mass spectrometry analysis of the crude reaction mixtures. The target-guided lead discovery search was
also successful when performed with reagent mixtures containing up to 10 components. From 23 acetylene
reagents, the enzyme selected two phenyltetrahydroisoquinoline (PIQ) building blocks that combined with
the tacrine azide within the active center gorge to form multivalent inhibitors that simultaneously associate
with the active and peripheral binding sites. These new inhibitors are up to 3 times as potent as our previous
phenylphenanthridinium-derived compounds, and with dissociation constants as low as 33 femtomolar,
they are the most potent noncovalent AChE inhibitors known. In addition, the new compounds lack a
permanent positive charge and aniline groups and possess fewer fused aromatic rings. Remarkably, despite
the high binding affinity, the enzyme displayed a surprisingly low preference for one PIQ enantiomer over
the other.
guided synthesis,15,16 and kinetically controlled target-guided
synthesis.17-25 The latter approach uses irreversible reactions
Introduction
By employing the biological targets themselves for as-
sembling inhibitors within the confines of their binding sites,
target-guided synthesis (TGS) promises to revolutionize lead
discovery. The newly formed inhibitors usually display much
higher binding affinities for their biological targets than the
individual components, since they simultaneously engage in
multiple binding interactions.1,2 In principle, lead discovery by
TGS is independent of the function of the target, since it relies
solely on its ability to hold the reagents in close proximity until
they become connected via the “arranged” chemical reaction.
As long as 20 years ago, Rideout et al. reported a marked
synergism between the cytotoxic effects of decanal and N-
amino-guanidines, which they suggested to result from the self-
assembly of cytotoxic hydrazones inside cells.3,4 Since then,
several approaches to target-guided synthesis have been devel-
oped: dynamic combinatorial chemistry,5-14 stepwise target-
to unite reagents inside the protein’s binding pockets. Most
(6) Ramstro¨m, O.; Lehn, J.-M. ChemBioChem 2000, 1, 41-48.
(7) Lehn, J.-M.; Eliseev, A. V. Science 2001, 291, 2331-2332.
(8) Bunyapaiboonsri, T.; Ramstro¨m, O.; Lohmann, S.; Lehn, J.-M.; Peng, L.;
Goeldner, M. ChemBioChem 2001, 2, 438-444.
(9) Eliseev, A. V. Pharm. News 2002, 9, 207-215.
(10) Ramstro¨m, O.; Lehn, J.-M. Nat. ReV. Drug DiscoVery 2002, 1, 26-36.
(11) Otto, S. Curr. Opin. Drug DiscoVery DeV. 2003, 6, 509-520.
(12) Erlanson, D. A.; Braisted, A. C.; Raphael, D. R.; Randal, M.; Stroud, R.
M.; Gordon, E. M.; Wells, J. A. Proc. Natl. Acad. Sci. U.S.A. 2000, 97,
9367-9372.
(13) Erlanson, D. A.; Lam, J. W.; Wiesmann, C.; Luong, T. N.; Simmons, R.
L.; DeLano, W. L.; Choong, I. C.; Burdett, M. T.; Flanagan, W. M.; Lee,
D.; Gordon, E. M.; O’Brien, T. Nat. Biotechnol. 2003, 21, 308-314.
(14) Ramstro¨m, O.; Lohmann, S.; Bunyapaiboonsri, T.; Lehn, J.-M. Chem.-
Eur. J. 2004, 10, 1711-1715.
(15) Maly, D. J.; Choong, I. C.; Ellman, J. A. Proc. Natl. Acad. Sci. U.S.A.
2000, 97, 2419-2424.
(16) Kehoe, J. W.; Maly, D. J.; Verdugo, D. E.; Armstrong, J. I.; Cook, B. N.;
Ouyang, Y.-B.; Moore, K. L.; Ellman, J. A.; Bertozzi, C. R. Bioorg. Med.
Chem. Lett. 2002, 12, 329-332.
(17) Lewis, W. G.; Green, L. G.; Grynszpan, F.; Radic´, Z.; Carlier, P. R.; Taylor,
P.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed. 2002, 41, 1053-
1057.
† Department of Chemistry, The Scripps Research Institute.
§ The Skaggs Institute for Chemical Biology, The Scripps Research
Institute.
(18) Bourne, Y.; Kolb, H. C.; Radic´, Z.; Sharpless, K. B.; Taylor, P.; Marchot,
P. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 1449-1454.
(19) Inglese, J.; Benkovic, S. J. Tetrahedron 1991, 47, 2351-2364.
(20) Boger, D. L.; Haynes, N.-E.; Kitos, P. A.; Warren, M. S.; Ramcharan, J.;
Marolewski, A. E.; Benkovic, S. J. Bioorg. Med. Chem. 1997, 5, 1817-
1830.
(21) Greasley, S. E.; Marsilje, T. H.; Cai, H.; Baker, S.; Benkovic, S. J.; Boger,
D. L.; Wilson, I. A. Biochemistry 2001, 40, 13538-13547.
(22) Nguyen, R.; Huc, I. Angew. Chem., Int. Ed. 2001, 40, 1774-1776.
(23) Nicolaou, K. C.; Hughes, R.; Cho, S. Y.; Winssinger, N.; Smethurst, C.;
Labischinski, H.; Endermann, R. Angew. Chem., Int. Ed. 2000, 39, 3823-
3828.
‡ University of California, San Diego.
(1) Jain, A.; Huang, S. G.; Whitesides, G. M. J. Am. Chem. Soc. 1994, 116,
5057-5062.
(2) Mammen, M.; Chio, S.-K.; Whitesides, G. M. Angew. Chem., Int. Ed. 1998,
37, 2755-2794.
(3) Rideout, D. Science 1986, 233, 561-563.
(4) Rideout, D.; Calogeropoulou, T.; Jaworski, J.; McCarthy, M. Biopolymers
1990, 29, 247-262.
(5) Huc, I.; Lehn, J.-M. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 2106-2110.
9
6686
J. AM. CHEM. SOC. 2005, 127, 6686-6692
10.1021/ja043031t CCC: $30.25 © 2005 American Chemical Society

In Situ Selection of Lead Compounds
A R T I C L E S
Figure 1. Four in situ click chemistry hit compounds based on tacrine and phenylphenanthridinium reagents.
approaches to TGS employ highly reactive reagents (strong
electrophiles or nucleophiles, metathesis catalysts, etc.), which
can cause side reactions and even destroy the biological target.
To avoid such complications, we have developed an extremely
reliable approach to kinetically controlled TGS, called in situ
click chemistry,17,26 which employs the completely bio-
orthogonal [1,3]-dipolar cycloaddition reaction27 between azides
and acetylenes. This process is self-contained, hence no external
reagents, catalysts, or byproducts that might interfere, and the
“reactants” themselves are largely “invisible” in a biological
milieu. Most importantly, despite its high driving force (>50
kcal/mol) the uncatalyzed reaction has a surprisingly high
activation barrier of approximately 25 kcal/mol, causing it to
be extremely slow at room temperature and its rate to be highly
dependent on parameters that stabilize the transition state.28 This
was first exploited by Mock et al., who observed a 105-fold
increase of the cycloaddition reaction rate when azide and
acetylene groups are held together in close proximity inside the
synthetic receptor, cucurbituril, leading to irreversible formation
of a triazole.29-31 In previous work we have shown that
acetylcholinesterase (AChE) is able to assemble extremely
potent inhibitors, which simultaneously access the enzyme’s
active and peripheral binding sites,32-35 from azide and acetylene
reagents, each linked to known active and peripheral site
inhibitors, tacrine and phenylphenanthridinium, respectively.17,26
Later we found that carbonic anhydrase is also capable of
assembling its own inhibitors within the confines of its active
center region, suggesting the in situ click chemistry technique
is applicable to a broad range of targets.36 Indeed, the scope of
the method is not limited to proteins, as demonstrated by Dervan
et al., who have used the azide/acetylene cycloaddition to
explore the double-stranded DNA-templated interconnection of
hairpin polyamides in the minor groove to produce tandem
hairpin dimers in site-specific fashion, which are capable of
targeting longer sequences.25
Recent key breakthroughs in our labs were made possible
by an improved method for analyzing the in situ click chemistry
reaction mixtures.26 Instead of using MALDI/DIOS (desorption/
ionization on silicon) mass spectrometry,37,38 as done previ-
ously,17 we now use HPLC with compound detection through
electrospray mass spectrometry in the positive selected ion mode
(LC/MS-SIM). For our purpose this method is more reliable,
enabling identification of the product triazoles by both retention
time and molecular weight. Additionally, chromatographic
removal of the molecules that might otherwise obscure the mass
spectrum of the product allows us to reduce the reaction time
from 6 days to 6 hours and lower the reagent concentrations
considerably. The following results were obtained under these
conditions.
(1) From 52 combinations of azide- and acetylene-bearing
tacrine and phenylphenanthridinium reagents, potentially giving
rise to 104 products, only four were assembled inside mouse
or eel acetylcholinesterase to form, with high selectivity, 1,5-
disubstituted (“syn”) triazoles (Figure 1). These enzyme-
generated compounds are femtomolar inhibitors, whereas the
corresponding 1,4-disubstituted (“anti”) triazole derivatives are
much less active.
(2) The triazole units of all enzyme-generated inhibitors were
two methylene units away from the tacrine moiety, even though
the enzyme had the opportunity to assemble products with
different linker spacings. Recent X-ray crystallographic studies
of complexes of anti- and syn-TZ2PA6 and mouse AChE
revealed that the formed triazole moieties are optimally posi-
tioned to contribute to protein binding through hydrogen bonding
and stacking interactions.18 This implies not only that triazoles
(24) Nicolaou, K. C.; Hughes, R.; Cho, S. Y.; Winssinger, N.; Labischinski,
H.; Endermann, R. Chem.-Eur. J. 2001, 7, 3824-3843.
(25) Poulin-Kerstien, A. T.; Dervan, P. B. J. Am. Chem. Soc. 2003, 125, 15811-
15821.
(26) Manetsch, R.; Krasin´ski, A.; Radic´, Z.; Raushel, J.; Taylor, P.; Sharpless,
K. B.; Kolb, H. C. J. Am. Chem. Soc. 2004, 126, 12809-12818.
(27) Huisgen, R. In 1,3-Dipolar Cycloaddition Chemistry; Padwa, A., Ed.;
Wiley: New York, 1984; Vol. 1, pp 1-176.
(28) Himo, F.; Lovell, T.; Hilgraf, R.; Rostovtsev, V. V.; Noodleman, L.;
Sharpless, K. B.; Fokin, V. V. J. Am. Chem. Soc. 2005, 127, 210-216.
(29) Mock, W. L.; Irra, T. A.; Wepsiec, J. P.; Manimaran, T. L. J. Org. Chem.
1983, 48, 3619-3620.
(30) Mock, W. L.; Irra, T. A.; Wepsiec, J. P.; Adhya, M. J. Org. Chem. 1989,
54, 5302-5308.
(31) Mock, W. L. Top. Curr. Chem. 1995, 175, 1-24.
(32) Du, D.-M.; Carlier, P. R. Curr. Pharm. Des. 2004, 10, 3141-3156.
(33) Han, Y. F.; Li, C. P.-L.; Chow, E.; Wang, H.; Pang, Y.-P.; Carlier, P. R.
Bioorg. Med. Chem. 1999, 7, 2569-2575.
(36) Mocharla, V. P.; Colasson, B.; Lee, L.; Roeper, S.; Sharpless, K. B.; Wong,
C.-H.; Kolb, H. C. Angew. Chem., Int. Ed. 2005, 44, 116-120.
(37) Wei, J.; Buriak, J. M.; Siuzdak, G. Nature 1999, 399, 243-246.
(38) Thomas, J. J.; Shen, Z.; Crowell, J. E.; Finn, M. G.; Siuzdak, G. Proc.
Natl. Acad. Sci. U.S.A. 2001, 98, 4932-4937.
(34) Carlier, P. R.; Han, Y. F.; Chow, E. S.-H.; Li, C. P.-L.; Wang, H.; Lieu,
T. X.; Wong, H. S.; Pang, Y.-P. Bioorg. Med. Chem. 1999, 7, 351-357.
(35) Pang, Y.-P.; Quiram, P.; Jelacic, T.; Hong, F.; Brimijoin, S. J. Biol. Chem.
1996, 271, 23646-23649.
9
J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005 6687

A R T I C L E S
Krasin´ski et al.
Figure 2. In situ click chemistry screening for AChE inhibitors, containing novel peripheral site ligands.
Figure 3. A library of 23 acetylene reagents for in situ click chemistry screening. All chiral compounds are racemic.
are valuable pharmacophoric units but also that the enzyme may
have actively accelerated their formation by lowering the
transition state energies of their formation through favorable
binding interactions.
(3) Due to the chromatographic separation and greater
reliability, the new LC/MS-SIM based analysis method enabled
us to increase screening throughput by using multicomponent
mixtures of building blocks. Thus, medium- or high-throughput
in situ click chemistry screening for lead discovery is now within
reach.
approach is conceptually interesting, as it addresses the question
of whether an enzyme complex of one reaction partner, capable
of triazole genesis (e.g., the tacrine azide TZ2), can find and
select its “best” triazole-forming partner(s) when presented with
mixtures of candidates with unknown binding affinities and so
discover its own potent biligand inhibitors. In the case at hand,
the goal was to replace the phenylphenanthridinium component
of our previous “in situ-made” AChE inhibitors with a moiety
conferring greater pharmacologic potential.
The acetylenic building blocks were readily synthesized by
alkylating commercially available amines with the appropriate
iodoalkynes or by forming hydrazones from 7-heptynal (cf.
Supporting Information). The complete acetylene reagent library
is shown in Figure 3. The heterocycles were chosen to simplify
the structure and eliminate the permanent positive charge, yet
retain many of the features of the phenanthridinium moiety.
Results and Discussion
These results have set the stage for performing the first ever
search for AChE inhibitors through in situ click chemistry based
on building blocks that were not previously known to interact
with the target. All previous experiments had employed known
active site and peripheral binding site ligands. To minimize the
number of variables, we decided to continue to use the tacrine
building block TZ2 as an “anchor molecule” that, in complex
with the enzyme, would recruit and irreversibly link together
novel peripheral site binders to form multivalent AChE inhibi-
tors that simultaneously access multiple binding sites within the
enzyme (Figure 2). A two-methylene spacer between tacrine
and the azide was chosen, since previous experiments had
proven this distance to be optimal.26 On the basis of analogous
considerations, we designed a library of complementary acety-
lene reagents carrying aromatic heterocyclic phenylphenanthri-
dinium mimics with a spacing of five or six methylene units.
To increase the screening throughput, we planned to test
multireagent mixtures containing up to 10 acetylene reagents
at a time. This multicomponent in situ click chemistry screening
For concept validation, initial in situ click chemistry screens
were performed by incubating binary TZ2/acetylene mixtures
with eel or mouse AChE at pH 7.4 for at least 6 h and analyzing
each reaction mixture by LC/MS-SIM.26 Most alkynes gave no
detectable product, except for the phenyltetrahydroisoquinolines
PIQ-A5 and PIQ-A6, which formed significant amounts of
triazoles. Their identity was confirmed by chromatographic
comparison of the in situ click chemistry reaction mixtures with
authentic samples of TZ2PIQ-A5 and TZ2PIQ-A6, which were
synthesized by a thermal cycloaddition reaction. Thus, two new
in situ hits have been found, composed of the tacrine active
site ligand and the phenyltetrahydroisoquinoline peripheral site
ligands. The latter were not previously known to bind to the
peripheral binding site of AChE, and they may have better
9
6688 J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005

In Situ Selection of Lead Compounds
A R T I C L E S
Table 1. Inhibition Constants of Acetylene Reagents for Eel and
Mouse AChEa,b
Kd (µM)
entry
compound
mouse AChE
eel AChE
1
IQN-A5
IQN-A6
PIQ-A5
PIQ-A6
IIQ-A5
IIQ-A6
C-A5
77
210
60
74
2
3
34
18
4
21
7.8
100
67
>400
>400
>400
>400
n.d.c
n.d.c
84
5
>400
>400
>400
>400
>400
>400
n.d.c
n.d.c
42
6
7
8
C-A6
9
PO-A5
PO-A6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
PHN-A5
PHN-A6
DPA-A5
DPA-A6
HIQ-A5
HIQ-A6
DMB-A5
DMB-A6
PIP-A5
PIP-A6
PQH-A4
QH-A4
BOH-A4
PA6
83
99
190
14
14
>400
>400
38
110
46
43
20
20
320
170
5.8
>400
>400
0.50
n.d.
4.1
0.36
a The measurements were performed in 100 mM phosphate buffer pH
7.0 (+0.01% BSA to stabilize the enzyme) at 22 °C. Constants were
determined in duplicate experiments using Hunter and Downs plots.40 b For
comparison, the dissociation constant Kd of TZ2 for mouse AChE is 0.023
µM. c Not determined due to low solubility of the compound.
Figure 4. Multicomponent in situ click chemistry screen. Extracted ion
LC/MS-SIM chromatograms for the multicomponent in situ reaction (traces
A-1, B-1, C-1) and for the background reaction containing all reagents, but
no enzyme (traces A-2, B-2, C-2). Traces A-1 and -2: Extracted ion
chromatograms for TZ2PIQ-A5; note the presence of product in the enzyme
reaction, while no product is formed in the absence of AChE. Traces B-1
and -2: Extracted ion chromatograms for TZ2PIQ-A6. Again, product is
present only in the enzyme reaction. Traces C-1 and -2: Extracted ion
chromatograms for TZ2IIQ-A6. No product is formed in either the enzyme
or the control reactions. The extracted ion traces for the remaining building
blocks are similar to this trace, demonstrating that the corresponding triazoles
are below the detection limit in the in situ reaction mixture.
in situ click chemistry products are formed even when compet-
ing reagents (here PIQ-A5 and PIQ-A6) are present.
Determination of Dissociation Constants for Acetylene
Reagents. The inhibition of AChE by the acetylene reagents
was measured at 22 °C and pH 7.0 (Table 1). With inhibition
constants in the micromolar range (8-34 µM), the phenyltet-
rahydroisoquinoline derivatives PIQ-A5 and PIQ-A6 are about
1 to 2 orders of magnitude less potent than the phenylphenan-
thridinium acetylenes used previously (Kd of PA6 for mouse
AChE: 0.36 µM), demonstrating that even this relatively low
level of affinity is sufficient for the target-templated reaction
to take place. While the PIQ compounds are among the higher
affinity in the set, there are five other compounds (entries 13,
17, 18, 21, and 23) with an approximately equal or higher
potency. Surprisingly, these compounds did not form in situ
products, despite their high binding affinity to the protein and
the presence of identical linker moieties, revealing a lack of
correlation between a reagent’s binding affinity and its ability
to undergo target-templated cycloaddition. It is possible that
these compounds are not peripheral site ligands, but rather active
site binders, thereby preventing the in situ reaction from
occurring, or that the spatial orientation of the acetylene group
of the enzyme-bound reagent is suboptimal. The fact that these
compounds do not show substrate-competitive inhibition ex-
pected of association with the active center appears to rule out
the former possibility, and stereoelectronic factors seem to be
decisive.
pharmacological properties than the previous phenylphenan-
thridinium-derived inhibitors, due to the lack of a positive
charge, the absence of aniline groups, and the presence of fewer
fused aromatic rings. The new hits were validated by demon-
strating that no triazole was formed in the absence of AChE, or
when the enzyme was replaced by bovine serum albumin (BSA)
(cf. Figure 4 for results from a multicomponent screen).
After successful completion of in situ click chemistry
experiments with binary TZ2/acetylene mixtures, we turned our
attention to multicomponent screens. Incubation of a mixture
of 10 structurally related alkynes (16 compounds, if all
enantiomers are counted, cf. Figure 4) with TZ2 and the enzyme
gave only the expected triazole products TZ2PIQ-A5 and
TZ2PIQ-A6; that is, none of the other acetylenes that were
present in the mixture were converted into triazoles. Thus, the
TZ2/enzyme complex (over 99% active site saturation by TZ2
under the reaction conditions39) was able to recognize subtle
differences in alkyne structure (compare IQN, PIQ, and IIQ)
and form triazole products exclusively with its preferred
reagents, PIQ-A5 and PIQ-A6. These results demonstrate that
highly efficient multicomponent screens are practical and that
Enantioselectivity. Initially, all reagents were either achiral
or racemic, prompting the question whether AChE would prefer
one enantiomer over the other for the in situ reaction in the
case of PIQ-A5 and PIQ-A6. To study the relative in situ click
chemistry rates, we prepared each reagent in its pure enantio-
meric forms through the synthesis and optical resolution of the
(39) AChE concentration: 1µM; TZ2 concentration: 4-5 µM; dissociation
constant of TZ2 for mouse AChE: Kd ) 23 nM.
9
J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005 6689

A R T I C L E S
Krasin´ski et al.
Scheme 1. Synthesis of Enantiomerically Pure PIQ Reagents
Scheme 2. Synthesis of TZ2-Derived syn- and anti-Triazoles from PIQ-A5 and PIQ-A6a
a Reagents and conditions: (a) EtMgCl (1.0 equiv), THF, rt to 60 °C, 30 min; (b) TZ2 (0.5 equiv), rt to 60 °C, 4 h, 65-85% yield over 2 steps; (c) TZ2
(1.0 equiv), CuI (5 mol %), MeCN, rt, 12 h, 80-90% yield.
precursor rac-6,7-dimethoxy-1-phenyl-1,2,3,4-tetrahydroiso-
quinoline,41,42 followed by alkylation with 7-iodohept-1-yne (1)
or 8-iodooct-1-yne (2) (cf. Scheme 1). The enantiopure alkynes
were also needed for kinetic and structural studies of enzyme-
inhibitor complexes (vide infra).
Interestingly, the in situ click chemistry reaction rates for the
enantiomers of each acetylene component are quite similar (cf.
Supporting Information). In case of the mouse enzyme, the R
isomers react slightly faster than the S enantiomers, whereas
the opposite is true for the eel enzyme. We are in the process
of investigating the molecular origin of the lack of selectivity
through X-ray crystallography.
Determination of the Regioselectivity. We synthesized all
regioisomers and enantiomers of the TZ2-derived PIQ-A5 and
PIQ-A6 triazoles to elucidate the regioselectivity of the enzyme
reaction and to develop a deeper understanding of the structure-
activity relationship (Scheme 2). For the preparation of anti-
triazoles, the recently discovered copper(I)-catalyzed process
was employed,43,44 whereas syn-isomers were synthesized by
way of magnesium acetylides.45-48
similarity to the tacrine/phenylphenanthridinium system in that
all in situ reaction products are 1,5-disubstituted (syn) triazoles
(Figure 5). The syn selectivity was independent of the linker
length, the source of the enzyme (eel or mouse AChE), and the
absolute configuration of the PIQ component (cf. Supporting
Information for LC/MS-SIM traces).
Determination of AChE-Inhibitor Association and Dis-
sociation Rate Constants. All kinetic parameters of inhibitor
binding to and dissociation from eel and mouse AChE were
measured as described previously,26 except for one modification.
For determining the first-order dissociation rate constants by
measuring the return of AChE activity upon 5000-fold dilution
of 50-100 nM concentrations of AChE‚inhibitor complex, we
employed purified inactive mouse AChE mutant Ser203Ala
(20-70 nM), instead of DNA, for sequestering the inhibitors
upon their release from the complex with the active wild-type
AChE to prevent their reassociation at concentrations approach-
ing or higher than their Kd. This modification was necessary,
since the current set of inhibitors did not intercalate with DNA
as efficiently as the phenylphenanthridinium derivatives used
previously. The equilibrium dissociation constants for the syn-
and anti-isomers, calculated as the ratios of their first-order
The comparison of LC/MS-SIM traces of enzyme-produced
triazoles and the reference compounds revealed a striking
(40) Hunter, A.; Downs, C. D. J. Biol. Chem. 1945, 87, 427-446.
(41) Kumar, P.; Dhawan, K. N.; Kishor, K.; Bhargava, K. P.; Satsangi, R. K. J.
Heterocycl. Chem. 1982, 19, 677-679.
(45) Krasin´ski, A.; Fokin, V. V.; Sharpless, K. B. Org. Lett. 2004, 6, 1237-
1240.
(46) Akimova, G. S.; Chistokletov, V. N.; Petrov, A. A. Zh. Org. Khim. 1967,
(42) Lee, D. U.; Wiegrebe, W. Bull. Korean Chem. Soc. 1991, 12, 373-376.
(43) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew.
Chem., Int. Ed. 2002, 41, 2596-2599.
3, 968-974.
(47) Akimova, G. S.; Chistokletov, V. N.; Petrov, A. A. Zh. Org. Khim. 1967,
3, 2241-2247.
(44) Tornøe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002, 67, 3057-
3064.
(48) Akimova, G. S.; Chistokletov, V. N.; Petrov, A. A. Zh. Org. Khim. 1968,
4, 389-394.
9
6690 J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005

In Situ Selection of Lead Compounds
A R T I C L E S
Figure 5. Regioisomer (syn/anti) determination for mouse AChE-derived in situ hits. The in situ product, (R)-TZ2PIQ-A5, was compared by LC/MS-SIM
to authentic samples from the Cu(I)-catalyzed and Mg-mediated reactions. (A) anti-(R)-TZ2PIQ-A5 prepared by the Cu(I)-catalyzed reaction; (B) co-
injection of (R)-TZ2PIQ-A5, formed by AChE, and anti-(R)-TZ2PIQ-A5, prepared by the Cu(I)-catalyzed reaction; (C) in situ click reaction; (D) co-
injection of the in situ click reaction and syn-(R)-TZ2PIQ-A5 prepared by the Mg-mediated reaction; (E) syn-(R)-TZ2PIQ-A5 prepared by the Mg-mediated
reaction.
Table 2. Kinetic Parameters and Dissociation Constants for in
Situ-Generated Inhibitors and Related Compounds
A5, which do not carry a permanent positive charge and aniline
groups as did the previous champion, TZ2PA6, while being 3
kon
koff
(min-
Kd
AChE
times as potent in the case of eel AChE. In fact, with a
dissociation constant of less than 40 fM (eel AChE), these
compounds are the most potent noncovalent AChE inhibitors
known to date. As a general observation, all in situ-generated
compounds tested here show a greater affinity for the eel enzyme
(33-360 fM) than for mouse AChE (100-1700 fM).
Despite the tight binding of the TZ2PIQ triazoles, there is
no clear enantiomeric preference. This observation is in line
with the lack of selectivity for either one of the two enantiomeric
reactants in the in situ click chemistry reaction (vide supra).
Thus, in the case of TZ2PIQ-A5, having a 5-methylene linker
between the triazole ring and the tetrahydroisoquinoline moiety,
the R isomer is bound about 5 times more tightly than the S
enantiomer in the case of mouse AChE, while there is no
difference between the enantiomers in case of eel enzyme. The
reverse trend is observed in the case of the TZ2PIQ-A6
triazoles, having a 6-methylene linker, where the S isomer has
a higher binding affinity for either enzyme.
As before, the in situ-generated syn-triazoles are several orders
of magnitude more potent inhibitors than the corresponding anti-
isomers, not formed by the enzyme. However, the extent of the
difference is more than an order of magnitude larger than for
phenylphenanthridinium triazoles, ranging from 600- to 5600-
fold preference for the syn-triazoles (the average difference in
the free energy of binding is about 4 kcal/mol, cf. Supporting
Information for additional free energy data). In the case of the
phenylphenanthridinium triazoles, the syn-preference was only
14- to 420-fold. In general, our data suggest that compounds
that are not formed by the enzyme, e.g., anti-triazoles and
TZ2HIQ-A6, bind more weakly than the in situ-generated
triazoles.
1
1
inhibitor
(1010 M-1 min-
)
)
(fM)
source
syn-
anti-
syn-
anti-
syn-
anti-
syn-
anti-
syn-
anti-
syn-
anti-
0.70
0.77
0.52
0.72
0.67
0.84
0.40
0.58
0.90
0.97
0.41
0.67
0.67
0.73
0.50
0.64
1.0
0.00023
0.0038
1.10
33 eel
500 mouse
210 000 eel
870 000 mouse
36 eel
100 mouse
190 000 eel
180 000 mouse
96 eel
1100 mouse
200 000 eel
420 000 mouse
360 eel
1700 mouse
170 000 eel
220 000 mouse
1200 eel
5400 mouse
2 100 000 eel
1 900 000 mouse
99 eel
(S)-TZ2PIQ-A5
6.2
0.00024
0.00088
0.77
(R)-TZ2PIQ-A5
(S)-TZ2PIQ-A6
(R)-TZ2PIQ-A6
TZ2HIQ-A6
TZ2PA6
1.0
0.00087
0.0110
0.83
2.8
0.0024
0.012
0.82
1.4
0.012
0.081
10
11
0.0015
0.0071
0.25
1.5
0.47
0.58
1.5
1.7
1.8
410 mouse
14 000 eel
8900 mouse
2.5
0.22
dissociation and second-order association rate constants, are
listed in Table 2, along with the dissociation constants for the
TZ2PA6 isomers reported previously.26
Interestingly, despite their higher binding affinities, the PIQ-
derived inhibitors display 2 to 3 times lower on-rates than the
original PA-derived inhibitors. Nevertheless, the on-rates kon
are all very large, close to diffusion controlled, and the observed
binding affinity trends are due mainly to variations in off-rates
koff, which are extremely slow for the tightest binding com-
pounds.
Experimental Section
The best inhibitors proved to be the enzyme-generated
phenyltetrahydroisoquinoline derivatives (S)- and (R)-TZ2PIQ-
CAUTION! All of the compounds described here (and especially
the most potent polyvalent inhibitors) are potentially neurotoxic. They
9
J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005 6691

A R T I C L E S
Krasin´ski et al.
must be handled with extreme care by trained personnel. Azide-
containing compounds, particularly those lower in saturated carbon and
oxygen content, are potentially explosive and must be handled with
care.
General Procedures for in Situ Click Chemistry Experiments.
Determination of Acetylcholinesterase and Stock Concentrations.
The enzyme concentrations were determined by quantitative measure-
ment of AChE activity using the Ellman assay as described previ-
ously.17,26 All in situ click chemistry reactions were performed at an
active site concentration of 1 µM. The stock concentrations for all
triazole compounds were determined in duplicate by titration of the
inhibitor solutions with two different AChE preparations of known
concentration.
to serve as an “anchor molecule”. In the present study, the
anchor is the azide TZ2, which is present at a concentration
sufficient to saturate the enzyme active site. We have discovered
two new potent in situ hit compounds through screening of
reagent mixtures using the reliable LC/MS-SIM method for
analyzing in situ click chemistry mixtures. The hit compounds
TZ2PIQ-A5 and TZ2PIQ-A6 were made by both eel and
mouse AChE with high regioselectivity for the syn-triazole
product. With low-femtomolar dissociation constants, these
compounds are the most potent noncovalent AChE inhibitors
known. They lack a permanent positive charge and aniline
groups and possess fewer fused aromatic rings than the original
inhibitors. The corresponding anti-isomers, not made by the
enzymes, were 2-4 orders of magnitude less potent. The
affinities of the offered building block components do not always
correlate well with the propensity for formation of hit com-
pounds. The unique synergism found for formation of these hit
compounds within the enzyme likely results from the proximity
of the reactant moieties and their ability to adopt the appropriate
orientation. In addition, the enzyme-templated cycloaddition is
most likely promoted by an enthalpic stabilization of the
transition state, probably through compensation of the strong
dipole moment that is developed during triazole genesis.
Previous crystallographic studies have revealed the unusual
positioning of a tryptophan and a tyrosine at the peripheral
binding site pointing toward a potential ability of the reactive
building blocks to induce or stabilize a unique enzyme
conformation that allows the reaction to take place. Currently,
we are focusing our research on the mechanism of the in situ
click chemistry involving a combination of kinetic studies and
structural biology.
In Situ Click Chemistry Screening Procedure for Binary Reagent
Mixures. The tacrine azide TZ2 was dissolved in MeOH and added
to ∼1 µM solutions of eel AChE (Type V-S, Sigma) or mouse
AChE49,50 in buffer (2 mM ammonium citrate, 100 mM NaCl, pH )
7.3-7.5) followed immediately by one of the acetylene components
and mixed. The final concentrations were as follows: eel or mouse
AChE, 1 µM; tacrine azide (TZ2), 4.6 µM, acetylene component, 24
µM; MeOH, 1.5%. Each reaction mixture was incubated at 37 °C for
at least 6 h. Samples of the reactions were injected directly (15 µL)
into the LC/MSD instrument to perform LC/MS-SIM analysis (Zorbax
SB-C8 reverse-phase column, preceded by a Phenomenex C18 guard
column, electrospray ionization, and mass spectroscopic detection in
the positive selected ion mode, tuned to the expected molecular weight
of the product). The cycloaddition products were identified by their
retention times and molecular weights. Control experiments in the
absence of enzyme or in the presence of bovine serum albumin (BSA,
3 mg/mL) instead of enzyme failed to produce product signals. For
these control experiments, methanol (1:1 dilution) was added to the
reaction mixtures prior to LC/MS-SIM analysis, to prevent possible
precipitation of the expected triazole product.
In Situ Click Chemistry Screening Procedure for Multicompo-
nent Incubations. A methanolic solution (1.0 µL) of acetylene building
blocks (10 reagents at 2 mM concentration each) was added to a solution
of TZ2 and mouse AChE (99 µL of 1 µM AChE, 4.2 µM TZ2, 2 mM
ammonium citrate, 100 mM NaCl, pH ) 7.3-7.5). The final concentra-
tions were as follows: mouse AChE, 1 µM; tacrine component (TZ2),
4.2 µM; acetylene component, 20 µM; MeOH, 1%. Each reaction
mixture was incubated at 37 °C for at least 24 h. Samples of the
reactions were injected directly (15 µL) into the LC/MSD instrument
to perform LC/MS-SIM analysis, as described previously.26 The
chromatograms were analyzed for the presence of in situ reaction
products by extracting single ion traces for all expected molecular
weights (cf. Figure 4).
Acknowledgment. We are grateful to Prof. G. Siuzdak and
Mr. J. Apon for MALDI mass spectrometry support. We also
thank Prof. Pascale Marchot (Laboratoire de Biochimie, Institut
Fe´de´ratif de Recherche Jean Roche, Universite´ de la Me´diter-
rane´e, Marseille, France) for providing samples of purified eel
AChE. This work was supported by the Swiss National Science
Foundation (R.M.); the Novartis Research Foundation (R.M.);
the Skaggs Institute for Research (A.K., J.R., H.C.K.); the
National Institute of General Medical Sciences, National
Institutes of Health, GM28384 (K.B.S.) and GM R37-18360
(P.T.); DAMDC17C-02-2-0025 (P.T.); and the W. M. Keck
Foundation (K.B.S.).
Conclusions
This study reveals that the in situ click chemistry approach
has great potential for lead discovery and optimization. As
established here, suitable reagents for the generation of potent
inhibitors within the enzyme’s binding sites can be found
without prior knowledge of their affinities for the protein,
provided that one of the two components has sufficient affinity
Supporting Information Available: LC/MS-SIM traces for
in situ click chemistry and control experiments, experimental
details and LC/MS-SIM traces for regioisomer determination,
comparison of reaction rates between the PIQ acetylene
enantiomers, experimental procedures for reagent and triazole
synthesis, and their characterization, and tables with free energy
increments for structural modifications. This material is available
free of charge via the Internet at http://pubs.acs.org.
(49) Radic´, Z.; Pickering, N. A.; Vellom, D. C.; Camp, S.; Taylor, P.
Biochemistry 1993, 32, 12074-12084.
(50) Marchot, P.; Ravelli, R. B. G.; Raves, M. L.; Bourne, Y.; Vellom, D. C.;
Kanter, J.; Camp, S.; Sussman, J. L.; Taylor, P. Protein Sci. 1996, 5, 672-
679.
JA043031T
9
6692 J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005