Synthesis of fluorescent probes directed to the active site gorge of acetylcholinesterase

Article abstract of DOI:10.1016/S0960-894X(00)00275-4

Six organophosphorus compounds linked to fluorophore groups were prepared in an effort to selectively modify the active site of acetylcholinesterase and deliver probes to the gorge region. Two compounds that vary by the length of a methylene (CH₂) group, pyrene-SO₂NH(CH₂)(n)NHC(O)CH₂CH₂P(O)(OEt)(F) (where n = 2 or 3) were found to be potent, irreversible inhibitors of recombinant mouse AChE (K(i)?10⁵ M^-1 min^-1). Size exclusion chromatography afforded a fluorescently-labeled cholinesterase conjugate. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00275-4

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1523±1526
Synthesis of Fluorescent Probes Directed to the Active
Site Gorge of Acetylcholinesterase
Jennifer R. Saltmarsh,a Aileen E. Boyd,b Oscar P. Rodriguez,a Zoran Radic,b
Palmer Taylorb and Charles M. Thompsona,c,*
aDepartment of Chemistry, The University of Montana, Missoula, MT 59812, USA
bDepartment of Pharmacology 0636, University of California, San Diego, La Jolla, CA 92093-0636, USA
cDepartment of Pharmaceutical Sciences, The University of Montana, Missoula, MT 59812, USA
Received 6 March 2000; accepted 25 April 2000
AbstractÐSix organophosphorus compounds linked to ¯uorophore groups were prepared in an e€ort to selectively modify the
active site of acetylcholinesterase and deliver probes to the gorge region. Two compounds that vary by the length of a methylene
(CH2) group, pyrene-SO2NH(CH2)nNHC(O)CH2CH2P(O)(OEt)(F) (where n=2 or 3) were found to be potent, irreversible inhibi-
1
tors of recombinant mouse AChE (Ki105 M min 1). Size exclusion chromatography a€orded a ¯uorescently-labeled cholines-
terase conjugate. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
Design
The crystal structure of acetylcholinesterase (AChE),1
the enzyme responsible for the hydrolysis of the neuro-
transmitter acetylcholine (ACh),2 has provided many
insights into key structural elements and motifs central to
its catalytic mechanism and mode of ACh processing. One
of the more interesting structural features of AChE
revealed from the crystal structure and predicted in part by
¯uorescence experiments3 is the presence of a long, narrow
hydrophobic gorge approximately 20 A deep leading to
the active site. Key to understanding the remarkably
high turnover rate and speci®c activity of AChE is the
manner in which substrates and inhibitors traverse this
passageway to reach the catalytic center. Selective place-
ment of a reporter group into the gorge region would
help elucidate the properties and dynamics of this unique
portal. The aim of this work was to design, synthesize,
and evaluate the anticholinesterase activity of pyrene- or
dansyl-containing organophosphates3 that covalently
attach to the active site serine and place ¯uorescent repor-
ter groups within the gorge region. As such, the resultant
¯uorescence data would permit examination of solvent
exposure and segmental motion of the reporter group.
The primary objective was to develop and evaluate
AChE inhibitor molecules (Fig. 1) capable of anchoring
a ¯uorophore at predictable depths within the protein.
To seat the ¯uorescent probe, an organophosphate (OP)
group was selected to covalently modify the active site
serine hydroxyl of AChE4 thereby placing the probe in
the gorge region but removed from the immediate vici-
nity of the catalytic serine. Spacer groups were used to
position the reactive phosphorus atom 16±19 A from the
further edge of the ¯uorophore moiety. Other consider-
ations used in the design of the target molecules (Fig. 1)
include the need for: (a) small phosphoester alkyl groups
(R=Me, Et) and leaving groups to reduce steric inter-
actions at the active site, (b) spacer variations in length
and chemical composition, (c) a non-hydrolyzable link-
age tethering the phosphorus group to the ¯uorophore
to obviate post-inhibitory reactions such as aging or
Figure 1. Proposed design for OP-based inhibitors tethered to a
reporter group.
*Corresponding author. Tel.: +1-406-243-4643; fax: +1-406-243-4227;
e-mail: cmthomp@selway.umt.edu
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00275-4

1524
J. R. Saltmarsh et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1523±1526
Scheme 1. Synthesis of OP-based probes (4, 5, 6 and 7) using lysineamide as the linker.
dealkylation and (d) ¯uorophores which show emission
properties characteristic of their immediate chemical
environment and segmental motion. Additional bonds
between the ¯uorophore, spacer and OP moieties were
planned as sulfonamide/amide to ensure stability and
chemical similarity to protein structure.
molecules 4 and 5, respectively. In an identical synthetic
sequence, O,S-dimethyl phosphonoacetic acid 2b7 was
used to prepare probe 6 in which the SMe moiety was
expected to function as the leaving group upon reaction
with AChE.8 Compound 5 was dealkylated at the P-OCH3
with sodium iodide (NaI, acetone, re¯ux) and then
converted to the phosphono¯uoridate 710 with cyanuric
¯uoride (C3N3F3)9 in DMSO. The conversion of 5 to 7
using cyanuric ¯uoride was near quantitative and was
easily monitored by 31P NMR, namely, the 5 ppm
down®eld shift (relative to the phosphorus monoacid) and
formation of the characteristic P-F coupling constant
(JP F=1055 Hz).
Synthesis5
The ®rst synthesis utilized commercially available
N(e)-tBoc-l-lysineamide 1 as the spacer group since
attachment of OP and ¯uorophore to the amine groups
would a€ord probe-containing molecules of suitable length
and chemical composition. The syntheses of compounds
4±7 are outlined in Scheme 1.
A second synthetic approach was undertaken in an e€ort
to provide additional options in the target molecule
length and concurrently circumvent problems in the
phosphorylation of AChE by probes 4±7 (vide infra).
Scheme 2 shows the synthetic route to probes 11a and 11b
using simple diamine spacer groups linked to pyrene and
3-(diethylphosphono)propanoic acid 9 as the reactive OP
precursor.
Dimethyl phosphonoacetic acid 2a6 was coupled to the
amino group of N(e)-tBoc-l-lysineamide hydrochloride
using dicyclohexylcarbodiimide (DCC) and triethylamine
t
(TEA). Removal of the Boc group with tri¯uoroacetic
acid (TFA) and coupling to dansyl chloride or pyrene-sul-
fonyl chloride in the presence of TEA a€orded probe
Scheme 2. Synthesis of OP-based pyrene probes (11a and 11b) using diamines as the linkers.

J. R. Saltmarsh et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1523±1526
1525
Pyrenesulfonyl chloride was reacted with excess ethylene
diamine or 1,3-diaminopropane to a€ord pyrenesulfon-
amido intermediates 8a and 8b, respectively. These inter-
mediates were coupled to 3-(diethylphosphono)propanoic
acid 9 (prepared from acrylic acid and triethylphosphite)11
using DCC to a€ord the 2-carbon spacer product 10a
and three-carbon spacer product, 10b. The phosphorus
ester groups were changed to ethyl in this series to reduce
the possibility of post-inhibitory hydrolysis reactions.
Hydrolysis of the phosphorus diethyl ester to the mono-
acid was accomplished with LiOH/THF-H2O. Use of
other base combinations led to either no reaction or
hydrolysis of both ester groups. Dealkylation using
sodium iodide failed to provide the monoacid. Following
neutralization with HCl, the two-carbon spacer and three-
carbon spacer monoacids were converted to the phos-
phono¯uoridates 11a (d 37.1; JP-F=1057 Hz) and 11b (d
37.3; JP-F=1057 Hz) using cyanuric ¯uoride in DMSO.
All phosphono¯uoridates were generated in DMSO and
used directly in inhibition studies without further pur-
i®cation.
atom to ¯uorophore edge are 17.8±18.1 and 19.3±19.4
A, respectively. Since the diameter of the pyrene group
is approximately 7.0 A, inhibitors 11a and 11b center
the ¯uorophore group at average distances of 14.4 and
15.9 A from the active site and 4 to 5.5 A internal to the
outer rim of the gorge. Despite excellent inhibitory
activity against WT mAChE, compounds 11a and 11b
were poor inhibitors of EEAChE. We are currently
exploring this species-dependent di€erence.
To demonstrate that these new inhibitors became cova-
lently conjugated to the enzyme and resulted in an
extrinsically ¯uorescent enzyme, as designed, an excess
of 11b was reacted with recombinant mAChE until
>95% inhibition was achieved. Free inhibitor was sepa-
rated from phosphonylated enzyme by means of size
exclusion chromatography (Sephadex G-25), and the
fractions containing protein were pooled and analyzed
spectroscopically. Fractions containing protein showed
an absorbance maxima at 352 nm in a ratio to the pro-
tein aromatic side-chain absorption suggesting near
stoichiometric conjugation. Furthermore, ¯uorescence
emission spectra upon excitation at 352 nm yielded the
characteristic emission pro®le of pyrene compounds.
Inhibition of AChE by Probes 4±7, 11a and 11b
To demonstrate that the target compounds 4±7, 11a and
11b placed the ¯uorophore portion of the inhibitor into
the active site gorge it was necessary to ®rst show that
they were covalent inhibitors of AChE.12 Compounds 5,
6 and 7 bear identical spacers and pyrene ¯uorophores but
vary in the phosphorus leaving group. As a result, di€er-
ences were expected in the Ki (inhibition rate constant)
despite the fact that identical inhibitor-AChE conjugates
would result. While modest anticholinesterase inhibition
was achieved against recombinant mouse acetylcholin-
esterase (mAChE)13 by compounds 5±7 in the 5±40 mM
range, they were virtually ine€ectual against electric eel
acetylcholinesterase (EEAChE). Inhibition of mAChE
by compounds 6 and 7 led to apparent reactivation
(spontaneous scission of the serine-phosphate bond)14
that may be due to noncovalent interactions between these
inhibitors and mAChE or alternatively, spontaneous
hydrolysis. The dansyl analogue 4 was also a weak
inhibitor indicating that the ¯uorophore size was not a
major in¯uence in the poor inhibition pro®les of 5±7.
Conclusion
Novel inhibitor structures that selectively position a
¯uorophore group into the gorge region of AChE have
been synthesized. Studies are currently underway using
inhibitors 11a and 11b to probe the environment and
dynamics of the gorge region of AChE using ¯uorescence
techniques.
Acknowledgements
C.M.T. thanks the University of Montana Sponsored
Research Programs Award, the National Science Foun-
dation (MCB-9808372 and CHE-9807469) and National
Institutes of Health (NS38248) for ®nancial support. The
NMR used in this study was purchased with funding
from the NSF (CHE9302468). P.T. thanks the United
States Public Health Service (GM 18360) and the
Department of Army Medical Defense (17-1-8014) for
®nancial support. A.E.B. was supported by predoctoral
fellowships from the Lucille P. Markey Charitable Trust
and ASERT.
Branching at the spacer group (lysine chiral center) of
compounds 4±7, which could diminish entry and access
to the active site, was seen as one potential problem to
overcome. Additionally, the phosphonoacetate group
contains an enolizeable methylene that could both
reduce the phosphoryl reactivity and increase reactiva-
tion. Two variations were made in the construction
of 11a and 11b to address these problems. Aliphatic dia-
mine spacer groups replaced lysine to eliminate branching
and 3-phosphonopropanoic acid was used in place of
phosphonoacetic acid to reduce enolization. Com-
References and Notes
1. (a) Sussman, J. L.; Harel, M.; Frolow, F.; Oefner, C.;
Goldman, A.; Toker, L.; Silman, I. Science 1991, 253, 872. (b)
Bourne, Y.; Taylor, P.; Marchot, P. Cell 1995, 83, 503.
2. For reviews of acetylcholinesterase see: (a) Quinn, D.
Chem. Rev. 1987, 87, 955. (b) Taylor, P. In The Pharmacolo-
gical Basis of Therapeutics, 9th Edition; McGraw Hill: New
York, 1996; pp 161±176. (c) Rosenberry, T. L. Adv. Enzymol.
Relat. Areas Mol. Biol. 1975, 43, 103.
pounds 11a and 11b (Ki=1.0Â105
M
1min 1) were
found to be excellent irreversible inhibitors of wild-type
(WT) mAChE. These values compare well with related
phosphonate inhibitors.15 Molecular modeling16 of the
chain-extended forms of these two inhibitor probes indi-
cates that the interatomic distances from phosphorus
3. For examples of OP inhibitors containing ¯uorophore
groups see: (a) Steinberg, N.; van der Drift, A. C. M.; Grun-

1526
J. R. Saltmarsh et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1523±1526
wald, J.; Segall, Y.; Shirin, E.; Haas, E.; Ashani, Y.; Silman, I.
Biochemistry 1989, 28, 1248. (b) Amatai, G.; Ashani, Y.;
Gafni, A.; Silman, I. Biochemistry 1982, 21, 2060. (c) Berman,
H. A.; Taylor, P. Biochemistry 1978, 17, 1704. (d) Epstein, D.
J.; Berman, H. A.; Taylor, P. Biochemistry 1979, 18, 4749.
4. For reviews of acetylcholinesterase inactivation by organo-
phosphates see: (a) Jarv, J. Bioorg. Chem. 1984, 12, 259. (b)
Ecobichon, D. J. In Pesticides and Neurological Diseases; CRC:
Boca Raton, 1982: pp 151±203. (c) Eto, M. Organophosphorus
Pesticides; Organic and Biological Chemistry; CRC: Florida,
1974. (d) Fukuto, T. R. Bull. Wld. Health Org. 1971, 44, 31.
5. All new compounds were fully characterized by spectral
and elemental or mass spectral analyses.
S.; Jackson, J. A.; Quinn, D. A.; Larsen, A. In Rev. Pestic.
Toxicol.; Toxicology Communications: Raleigh, 1993; Vol. 2,
pp 131±146.
9. (a) Olah, G. A.; Nojima, M.; Kerkes, I. Synthesis 1973, 487.
(b) Kokotos, G.; Noula, C. J. Org. Chem. 1996, 61, 6994. (c)
Carpino, L. A.; Sadat-Aalaee, D.; Chao, H. G.; DeSelms, R.
H. J. Am. Chem. Soc. 1990, 112, 9651.
10. NOTE: Phosphono¯uoridates are likely to be neurotoxic
and should only be handled by quali®ed personnel in a well-
ventilated hood. The anticholinesterase activity of the orga-
nophosphates described in this study can be destroyed by
stirring the compounds with 1 M NaOH. For a general review
of phosphono¯uoridate nerve gas agents see: Benshop, H. P.;
De Jong, L. P. A. Acc. Chem. Res. 1988, 21, 368.
11. Dixon, H. B. F.; Sparkes, M. J. Biochem. J. 1974, 141,
715.
6. Dimethyl phosphonoacetic acid was prepared by acid
hydrolysis (CF3CO2H, rt, 4±6 h) of tert-butyl dimethyl phos-
phonoacetate (Aldrich).
7. O,S-Dimethyl phosphonoacetic acid was prepared from
tert-butyl dimethyl phosphonoacetate using the following
sequence in 40% overall yield: (1) thionation with Lawesson's
reagent (toluene, re¯ux, 24 h), (2) EtOC(S)S K+/MeOH
re¯ux then CH3I, (3) CF3CO2H, rt, 4±6 h. For representative
experimentals see: Thompson, C. M.; Frick, J. A.; Natke, B.
E.; Hansen, L. K. Chem. Res. Toxicol. 1989, 2, 386.
8. For examples of thiolates as leaving groups see: (a)
Thompson, C. M.; Fukuto, T. R. J. Agric. Food Chem. 1982,
30, 282. (b) Thompson, C. M. In Organophosphorus Insecti-
cides: Chemistry, Fate and E€ects; Academic: San Diego,
1992, pp 19±46. (c) Thompson, C. M.; Berkman, C. E.; Ryu,
12. Ellman, G. L.; Courtney, K. D.; Andres, V.; Featherston,
R. M. Biochem. Pharmacol. 1961, 7, 88.
13. Bourne, Y.; Taylor, P.; Bougis, P. E.; Marchot, P. J. Biol.
Chem. 1999, 274, 2963.
14. Ecobichon, D. J. In Pesticides and Neurological Diseases;
CRC: Boca Raton, 1982; pp 151±203.
15. Hosea, N.; Berman, H. A.; Taylor, P. Biochemistry 1995,
34, 11528.
16. Spartan Pro, Wavefunction Inc., Irvine, CA. Modeling of
active site probes was conducted with SP-enantiomers which
directs the smaller substituent to the acyl pocket and the
phosphoryl oxygen to the oxyanion hole.