Enzyme-controlled molecular recognition: Selective targeting of trypsin with a substrate-inhibitor supramolecular complex

Article abstract of DOI:10.1021/ja980810z

We have synthesized and studied the host compound (1) that forms a noncovalent complex (K = 480 ± 30 M^-1) with 1,4-diamidinobenzene (2) at pD 8.0 in D₂O. Guest 2 has been shown to inhibit the proteases specific for cationic side chains (K(i) values for thrombin, plasmin, and trypsin being 1.0 ± 0.4, 0.6 ±0.1, and 0.88 ± 0.07 mM, respectively). Host 1 and the 1- 2 complex are shown to undergo the trypsin-catalyzed cleavage that results in the increased concentration of the free form of 2 and, therefore, induces the enzyme inhibition. Plasmin and thrombin are shown to cleave 1 at a significantly lower rate than trypsin. Thus, the complex of 1 with 2 represents the first example of a selective enzyme-sensitive supramolecular system capable of enzyme targeting through a 'trojan horse'-type mechanism in which a bioactive compound acts on the same molecule that provides its release from a delivery system.

Full text of DOI:10.1021/ja980810z

VOLUME 120, NUMBER 45
NOVEMBER 18, 1998
© Copyright 1998 by the
American Chemical Society
Enzyme-Controlled Molecular Recognition: Selective Targeting of
Trypsin with a Substrate-Inhibitor Supramolecular Complex
Sonali Rudra and Alexey V. Eliseev*
Contribution from the Department of Medicinal Chemistry, School of Pharmacy, State UniVersity of
New York at Buffalo, Buffalo, New York 14260
ReceiVed March 11, 1998. ReVised Manuscript ReceiVed July 13, 1998
Abstract: We have synthesized and studied the host compound (1) that forms a noncovalent complex (K )
480 ( 30 M-1) with 1,4-diamidinobenzene (2) at pD 8.0 in D2O. Guest 2 has been shown to inhibit the
proteases specific for cationic side chains (Ki values for thrombin, plasmin, and trypsin being 1.0 ( 0.4, 0.6
( 0.1, and 0.88 ( 0.07 mM, respectively). Host 1 and the 1‚2 complex are shown to undergo the trypsin-
catalyzed cleavage that results in the increased concentration of the free form of 2 and, therefore, induces the
enzyme inhibition. Plasmin and thrombin are shown to cleave 1 at a significantly lower rate than trypsin.
Thus, the complex of 1 with 2 represents the first example of a selective enzyme-sensitive supramolecular
system capable of enzyme targeting through a “trojan horse”-type mechanism in which a bioactive compound
acts on the same molecule that provides its release from a delivery system.
Among multiple branches of supramolecular chemistry,1 one
of the most interesting directions is focused on the creation of
new synthetic host-guest systems engaged in molecular rec-
ognition of biological compounds, such as peptides and proteins,
in their native environment, i.e., aqueous solutions.2 When the
recognition targets are enzymes, the synthetic molecular as-
semblies can be tailored not only to bind to the target, but also
to serve as enzyme substrates. As a variation of such an
approach, enzyme-induced decapsulation3 has been previously
used as a selective controlled release system.
Further development of synthetic enzyme-controlled molec-
ular recognition systems may have important implications for
medicinal chemistry. Indeed, various methods of drug laten-
tiation have been used to transform known bioactive compounds
to their enzyme-labile derivatives thereby improving their
pharmacokinetic profiles.4 A similar approach has been used
to build enzyme-sensitive delivery systems in which a drug
molecule bound to a carrier is released upon specific enzymatic
attack.5
The design of the system described below was based on the
idea that an enzyme-targeted delivery system would work best
if the enzyme catalyzing the release would also be the target
for the released drug (Scheme 1). The catalyzed cleavage of
the host carrier (in its free or guest-complexed form) would
(1) Lehn, J.-M. Supramolecular Chemistry. Concepts and PerspectiVes;
VCH: Weinheim, 1995.
(2) (a) Still, W. C. Acc. Chem. Res. 1996, 29, 155-163. (b) Schneider,
H.-J. Angew. Chem., Int. Ed. Engl. 1993, 32, 848. (c) Torneiro, M.; Still,
W. C. Tetrahedron 1997, 53, 8739-8750. (d) Albert, J. S.; Goodman, M.
S.; Hamilton, A. D. J. Am. Chem. Soc. 1995, 117, 1143-1144. (e) Peczuh,
M. W.; Hamilton, A. D.; Sanchezquesada, J.; de Mendoza, J.; Haack, T.;
Giralt, E. J. Am. Chem. Soc. 1997, 119, 9327-9328. (f) Hamuro, Y.;
Calama, M. C.; Park, H. S.; Hamilton, A. D. Angew. Chem., Int. Ed. Engl.
1997, 36, 2680-2683. (g) Hinzen, B.; Seiler, P.; Diederich, F. HelV. Chim.
Acta 1996, 79, 942-960.
(4) (a) Korolkovas, A. Essentials of Medicinal Chemistry, 2nd ed.; Wiley-
Interscience: New York, 1988. (b) Wermuth, C. G.; Gaignault, J.-C.;
Marchandeau, C. Designing Prodrugs and Bioprecursors. 1. Carrier Prodrugs.
In: The Practice of Medicinal Chemistry; Wermuth, C. G., Ed.; Academic
Press: London, 1996.
(5) Schacht, E. H.; Vansteenkiste, S.; Seymour, L. Macromolecular
Carriers for Drug Targeting. In: The Practice of Medicinal Chemistry;
Wermuth, C. G., Ed.; Academic Press: London, 1996.
(3) Menger, F. M.; Johnston, D. E. J. Am. Chem. Soc. 1991, 113, 5467-
5468.
10.1021/ja980810z CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/27/1998

11544 J. Am. Chem. Soc., Vol. 120, No. 45, 1998
Rudra and EliseeV
Scheme 1
Scheme 2
thus increase concentration of the free form of the enzyme
inhibitor and thereby close the negative feedback loop, as shown
in Scheme 1. Such a Trojan horse-type complex may be
potentially used as a delivery system that discharges the inhibitor
at the immediate site of its action providing precision of targeting
at the molecular level.
We report here a demonstration of the above general concept,
implemented in a model synthetic host carrier that is capable
of complexing with an inhibitor of several serine proteases and
at the same time serves as a selective substrate for one of the
proteases.
Design and Synthesis of the Host-Inhibitor System. The
enzymes chosen as potential targets were trypsin, thrombin, and
plasmin, proteases that cleave peptide bonds specifically at the
sites containing positively charged residues, Arg and Lys.6 The
host-inhibitor combination was designed to meet the require-
ments of Scheme 1, as described below.
1,4-Diamidinobenzene 2 has been previously shown to inhibit
trypsin via binding to its cation-recognition site.7 Our prelimi-
nary studies have shown that 2 exhibits similar inhibitory
properties toward thrombin and plasmin. The Ki values of 2
for thrombin (from bovine plasma), plasmin (from human
plasma), and trypsin (type-IX from porcine plasma) were found
to be 1.0 ( 0.4, 0.6 ( 0.1, and 0.88 ( 0.07 mM, respectively.
Our measured Ki for trypsin differed from the previously
reported value (5 mM,7 perhaps, because of somewhat different
experimental conditions).
Host compound 1 has been designed to contain two “arms”
that would be capable of binding to 2 via the formation of up
to four salt bridges between the phosphate and amidinium groups
(Scheme 2).8 Formation of the guanidinium-phosphate salt
bridges has been shown to be an effective recognition mecha-
nism in natural and synthetic complexes.9 Molecular modeling
of the 1‚2 complex (Sybyl/Tripos software) has shown that the
phosphate groups (d(P-P) ) 14.0 Å in the energy-minimized
conformation) are located favorably for the formation of two
sets of the ion pairs with the dication of 2 (d(C(amidine)-
C(amidine)) ) 7.0 Å), as shown in Figure 1. The two
phosphate-containing halves of 1 have been linked together by
peptide bonds through a lysine residue that can serve as an
anchor rendering the host compound a potential substrate for
proteases specific to cationic side chains. The enzymatic
cleavage of 1 should thus result in the increased concentration
of the free form of 2 that can subsequently inhibit the protease.
Host 1 was synthesized via amide formation and phospho-
rylation procedures (Scheme 3) and isolated as its tetraammo-
Figure 1. Energy-minimized molecular model of the 1‚2 complex.
nium salt. Its binding with 2 was studied by the 31P NMR
titration in aqueous solution at pD 8.0, close to the optimum
for the protease action.6 Considerable complexation-induced
31P chemical shifts were indicative of the expected ionic binding
mechanism. The titration curves (Figure 2) were satisfactory
fitted by a 1:1 binding equation10 and yielded the binding
constant value of 480 ( 30 M-1 (∆G ) 15.3 kJ/mol). The
observed host-inhibitor affinity is slightly lower than one should
expect from the formation of four salt bridges in the aqueous
solution.11 Such a difference may be attributed to the formation
of an internal ion pair between the lysine ammonium and one
of the phosphate groups in 1 which effectively decreases the
host negative charge. The stability of the host-inhibitor
(6) Barman, T. E. The Enzyme Handbook; Springer-Verlag: New York,
1969; Vol. 2.
(7) Geratz, J. D. Arch. Biochem. Biophys. 1967, 118, 90.
(8) As suggested by the reviewer, this binding mode may be alternatively
described as a combination of two (2+1) salt bridges with possible formation
of up to four ionic hydrogen bonds.
(9) (a) Perreault, D. M.; Cabell, L. A.; Anslyn, E. V. Bioorg. Med. Chem.
1997, 5, 1209-1220. (b) Schrader, T. Chem. Eur. J. 1997, 3, 1537-1541.
(c) Schrader, T. H. Tetrahedron Lett. 1998, 39, 7, 517-520.
(10) K was determined from the nonlinear regression fitting of the titration
data to the equation ∆δ ) ∆δ∞K[I]/(1 + K[I]) that corresponds to the excess
of the inhibitor over the complex maintained in the experiment.
(11) (a) Schneider, H.-J.; Theis, I. Angew. Chem., Int. Ed. Engl. 1989,
28, 753. (b) Schneider, H.-J. Angew. Chem., Int. Ed. Engl. 1991, 30, 1417.

Enzyme-Controlled Molecular Recognition
J. Am. Chem. Soc., Vol. 120, No. 45, 1998 11545
Scheme 3
performed at pH 8.5 and 37 °C in 0.1 M tris-HCl buffer were
1
monitored by HPLC and by H and 31P NMR.
Incubation of 1 under the above conditions in the absence of
enzymes showed no detectable cleavage within 54 h. Incubation
of 10 mM 1 with thrombin (0.64 mM) for 24 h showed no
changes in the substrate concentration and spectra, within
experimental error. Similarly, an experiment with plasmin (0.06
mM)13 showed <20% cleavage after 54 h. Such inactivity of
the enzymes is not surprising given that the negatively charged
side chains of 1 may strongly increase its Km values. However,
trypsin (0.81 mM) did catalyze the hydrolysis of 1, as followed
by HPLC (Figure 3). The progress of the hydrolysis reaction
1
was also verified by the appearance of new signals in the H
NMR spectrum which were assigned to the cleavage products
8 and 9 by comparison with the independently synthesized
standards (see the Experimental Section).
To test the trypsin effect on the inhibitor-host system, a
similar cleavage experiment of 1 was performed in the presence
of an equimolar concentration of 2. The results shown in Figure
3 demonstrate that the cleavage was also observed in these
conditions, although at a lower rate, as expected from the
inhibitory action of 2 on the enzyme.
It should be noted that in the latter reaction, host 1 plays a
dual role, that of the scavenger for the inhibitor and of the
substrate for the enzyme. Thus the enzyme activity, affected
by the released inhibitor, is probed by the rate of the host
Figure 2. 31P NMR titration of 1 (4.71 mM) with 2 at pD 8.1 and 20
°C. Changes in chemical shifts (∆δ) are shown for two phosphate
groups in 1.
complex is similar to that commonly observed, for example,
for drug-cyclodextrin complexes.12
Proteolytic Cleavage of the Host. Host 1 was then tested
as a potential substrate for the proteases. The cleavage reactions
(13) Plasmin was tested at a lower concentration than other proteases
for solubility reasons and also because plasmin is known to be ca. 50 times
more active than trypsin (per mM of enzyme) in the hydrolysis of their
specific amide substrates (see Supporting Information).
(12) (a) Szejtli, J. Cyclodextrins and Their Inclusion Complexes; Aka-
de´miai Kiado´: Budapest, 1982. (b) Ducheˆne, D. New Trends in Cyclodex-
trins and Their DeriVatiVes; Editions de Sante´: Paris, 1991.

11546 J. Am. Chem. Soc., Vol. 120, No. 45, 1998
Rudra and EliseeV
Experimental Section
Materials and Methods. All reagents were purchased from Aldrich
unless otherwise noted. N-ꢀ-CBZ-L-lysine methyl ester, specific
substrates for Ki determination, trypsin-type IX (15 900 units/mg of
protein) from porcine pancreas, thrombin (98 units/mg of protein) from
bovine plasma, and plasmin (3.9 units/mg of protein) from human
plasma were purchased from Sigma. 1H, 13C, and 31P NMR were
recorded on Varian Unity Inova-400, Varian Unity Inova-300, and
Varian Gemini (300 MHz) spectrometers. UV spectra were recorded
with a Cary 3E UV-vis spectrophotometer. HPLC analyses were
performed on a Beckman System Gold chromatograph equipped with
a photodiode array UV detector.
Synthesis. (a) N-r-(4-Hydroxybenzoyl)-N-E-CBZ-lysine Methyl
Ester (4). N-ꢀ-CBZ-L-lysine methyl ester hydrochloride was converted
in its free base form 3 by ether extraction from aqueous solution at pH
10. DCC (2.20 g, 10.66 mmol) was added to a stirred solution of 3
(2.408 g, 8.19 mmol) and 4-hydroxybenzoic acid (1.13 g, 8.19 mmol)
in anhydrous THF (20 mL) at 5 ïC, and the temperature was raised to
ambient. Upon completion of the reaction after 4 h, the solvent was
evaporated in Vacuo and the residue was dissolved in EtOAc (150 mL),
extracted with 10% citric acid (2 × 75 mL), saturated NaHCO3 solution
(2 × 75 mL), brine (1 × 75 mL), and water (1 × 75 mL), and dried
over anhydrous MgSO4. Evaporation of the solvent yielded the oily
product 4 (2.3291 g, 69%, Rf ) 0.45, silica, 1:1 CHCl3/EtOAc).
Compound 4 was used in subsequent steps without further purification;
1H NMR (300 Mz, DMSO) δ 8.38 (d, J ) 7.3 Hz, 1H, -R-CH-NH-
), 7.76 (d, J ) 7.8 Hz, 2H, -CO-Ph-OH), 7.33 (m, 5H, -Ph), 7.22
(t, 1H, -ꢀ-NH-COO-), 6.80 (d, J ) 7.8 Hz, 2H, -CO-Ph-OH),
5.00 (s, 2H, -CH2-Ph), 4.46-4.26 (m, 1H, -R-CH-), 3.63 (s, 3H,
-COOCH3), 2.98 (m, 2H, -CH2-ꢀ-NH-), 1.94-0.92 (m).
Figure 3. Time courses for the trypsin-catalyzed hydrolysis of 1 (10
mM) in the absence (a) and in the presence (b) of 10.6 mM 2 (each
curve is an average of two independent runs, see text for details).
cleavage. The complex kinetics of this process makes it difficult
to analyze the precise mechanism of the reaction, namely
whether the host itself or its complex with the inhibitor acts as
the enzyme substrate. However, regardless of the mechanism,
the trypsin-catalyzed host cleavage leads to the increasing
concentration of the inhibitor’s free form in solution. Given
the measured 1‚2 stability constant, it can be estimated that under
the experimental conditions (Figure 3), 64% of the inhibitor
initially exists in the host-complexed form and therefore is
effectively removed from the interaction with the enzyme. The
subsequent host hydrolysis increases the fraction of the free
inhibitor. The overall process of trypsin interaction with the
1‚2 complex can thus be considered as a product-inhibited
reaction. Apparently, the inhibitory effect of 2 results in
deactivation of trypsin with respect to not only 1 but also any
other potential substrate, thereby serving the general purpose
of targeting.
(b) N-r-(4-Hydroxybenzoyl)-N-E-CBZ-lysine (5). To a stirred
solution of 4 (2.568 g, 6.19 mmol) in 3:1 methanol-water (60 mL)
were added NaOH pellets (1.24 g, 31.0 mmol) at 5 °C. After 2.5 h,
when the TLC (silica, 1:1 CHCl3/EtOAc) showed disappearance of all
starting material, the solvent was evaporated and the residue was
dissolved in water (50 mL) and acidified to pH 3.5 with a dilute solution
of HCl. Extraction with EtOAc (2 × 125 mL), drying the organic
layer over anhydrous MgSO4, followed by the solvent evaporation and
drying yielded the carboxylic acid derivative 5 (2.142 g, 86%) as an
oily residue. The product 5 was used in subsequent steps without
1
further purification; H NMR (300 MHz, DMSO) δ 9.95 (br s, 1H,
-Ph-OH), 8.23 (d, 1H, -R-NH-), 7.76 (d, 2H, -Ph-OH), 7.34 (m,
5H, -Ph), 7.22 (t, 1H, -NH-COO-), 6.81 (d, 2H, -Ph-OH), 4.99
(s, 2H, -CH2-Ph), 4.31 (m, 1H, -R-CH-), 2.99 (m, 2H, -CH2-ꢀ-
NH-), 1.88-1.68 (m, 2H, -R-CH-CH2-), 1.55-1.22 (m, 4H,
-(CH2)2-CH2-ꢀ-NH-).
Conclusion
We have described a host compound that scavenges the
protease inhibitor via noncovalent complex formation and may
release it upon proteolytic cleavage. An essential feature of
this system is that the host increases the selectivity of the
inhibitor action. Indeed, the Ki values of 2 are similar for all
three tested enzymes, but only trypsin is capable of cleaving
host 1 and the 2‚1 complex. This is no surprise given that the
substrate specificity is determined by the overall catalytic
mechanism of the enzyme, while the inhibitor specificity reflects
only the binding event.
(c) N-r-(4-Hydroxybenzoyl)-N-E-CBZ-lysine 4-Hydroxyanilide
(6). Carboxylic acid 5 (1.601 g, 3.99 mmol) and 4-aminophenol (0.436
g, 3.99 mmol) were dissolved in anhydrous THF (65 mL) in an argon
atmosphere. The mixture was then cooled to 5 °C and to it was added
DCC (1.07 g, 5.18 mmol). The reaction mixture was allowed to warm
to room temperature and the progress of the reaction was monitored
by TLC (silica, 7.5% MeOH in CHCl3). On completion of the reaction
after 4 h, the solvent was removed in Vacuo and the residue was
dissolved in EtOAc (150 mL) and extracted with saturated NaHCO3
(2 × 75 mL), 10% citric acid (2 × 75 mL), brine water (75 mL), and
water (75 mL). The EtOAc layer was then dried over anhydrous
MgSO4, evaporated, and dried under Vacuo to yield product 5 (1.687
g, 86%). This crude product was used in the next step without further
purification. A small portion of the product was purified by column
chromatography (eluant: 7.5% MeOH in CHCl3) for NMR spectros-
copy; 1H NMR (300 MHz, DMSO) δ 9.96 (s, 1H, -OH), 9.76 (s, 1H,
-OH), 9.15 (s, 1H, -NH-Ph-OH), 8.18 (d, 1H, -NH-R-CH-), 7.80
(d, 2H, -CO-Ph-OH), 7.34 (m, 7H, -Ph + -NH-Ph-OH), 7.24
(t, 1H, -ꢀ-NH-COO-), 6.81 (d, 2H, -CO-Ph-OH), 6.68 (d, 2H,
-NH-Ph-OH), 5.00 (s, 2H, -CH2-PH), 4.49 (m, 1H, -R-CH-),
3.01 (m, 2H, -CH2-ꢀ-NH-), 2.88-2.68 (m, 2H, -R-CH-CH2-),
1.59-1.20 (m, 4H, -(CH2)2-CH2-ꢀ-NH-). 1H-1H COSY (300
MHz) indicated coupling between protons at δ 7.80 and 6.81 ppm and
Our current effort is focused on the development of systems
that use reversible coValent bonds between the host and the
inhibitor and, therefore, form stronger complexes that may lead
to new therapeutic applications. The ultimate trojan horse-type
system should act in a way similarly to the suicide enzyme
inactivators14 with two major differences: (1) while the product
of the suicide reaction usually irreversibly links to the enzyme,
the “trojan horse” product leaves the enzyme in its native state;
and (2) the “trojan horse” approach requires no special inhibitor
design, because there is a known inhibitor that can be used in
the complex.
(14) (a) Sjoerdsma, A. Clin. Pharm. Therap. 1981, 30, 3-22. (b) Walsh
C. T. Annu. ReV. Biochem. 1984, 53, 493-535.

Enzyme-Controlled Molecular Recognition
J. Am. Chem. Soc., Vol. 120, No. 45, 1998 11547
between those at δ 7.34 and 6.68 ppm. FABMS m/z 492.3 (M + H)+;
calcd for C27H30N3O6 492.5.
MS (MeOH/water 1:1 v/v) showed a major singly charged molecular
peak at m/z 516.1 corresponding to one negative charge on each
phosphate group of 1 and one positive charge on the lysine ammonium.
The positive mode ESI MS showed a singly charged peak at m/z 518.0
corresponding to fully protonated phosphate groups and positively
charged ammonium.
Two byproducts of the reaction recovered from the fractions
containing 0.5-0.6 and 0.6-0.7 M salt were identified as cleavage
products 8 and 9, respectively. 8: 1H NMR (400 MHz, D2O, pD ) 8)
δ 7.61 (d, J ) 8.0 Hz, 2H, Ph), 7.10 (d, J ) 8.0 Hz, 2H, Ph), 4.20 (m,
1H, R-CH), 2.81 (t, J ) 7.2 Hz, 2H, -CH2-ꢀ-NH3+), 1.8-1.6 (m,
2H, -R-CH-CH2-), 1.6-1.5 (m, 2H, -CH2-CH2-ꢀ-NH3+), 1.30 (t,
J ) 7.6 Hz, 2H, -R-CH-CH2-CH2-); 31P NMR (162 MHz, D2O,
pD ) 8) δ -3.05. 9: 1H NMR (400 MHz, D2O, pD ) 8) δ 7.25, 7.02
(dd); 31P NMR (162 MHz, D2O, pD ) 8) δ -3.71. NMR data for 8
and 9 were later used to confirm assignments of the enzymatic cleavage
products.
(d) Tetrabenzyl [N-r-(4-Benzoyl)-N-E-CBZ-lys 4-anilide]diphos-
phate (7). To a stirred solution of 6 (0.2 g, 0.41 mmol) in anhydrous
THF (14 mL) was added sodium hydride (dry 95%, 0.021 g, 0.83 mmol)
under argon. After the suspension was stirred for 5 min a solution of
tetrabenzylpyrophosphate (Fluka, 0.57 g, 1.06 mmol) in anhydrous THF
(7 mL) was added dropwise. Within 15 min a white viscous substance
precipitated out and TLC (eluant: 5% MeOH in CHCl3) showed no
presence of starting material 6. After removing the solvent and drying
in Vacuo, the crude product was purified by silica chromatography (5%
MeOH in CHCl3) to yield compound 7 (0.257 g, 62%, Rf ) 0.54); 1H
NMR (400 MHz, DMSO) δ 10.15 (s, 1H, -NH-Ph-), 8.59 (d, 1H,
-NH-R-CH-), 7.91 (d, J ) 8.8 Hz, 2H, -Ph-), 7.57 (d, J ) 8.8
Hz, 2H, -Ph-), 7.33 (m, 27 H, -CH2-Ph + -Ph-), 7.24 (t, 1H,
-ꢀ-NH-), 7.10 (d, J ) 8.4 Hz, 2H, -Ph-), 5.12 (m, 6 H, -CH2-
Ph), 4.99 (m, 4H, -CH2-Ph), 4.49 (m, 1H, -R-CH-), 2.98 (m, 2H,
-CH2-ꢀ-NH-), 1.77 (m, 2H, -R-CH-CH2-), 1.57-1.00 (m, 4H,
-R-CH-CH2-(CH2)2-); 31P NMR (162 MHz, DMSO) δ -5.75,
-6.35; FABMS m/z 1012.6 (M + H)+; calcd for C55H55N3O12P2 1012.3.
(e) Tetraammonium [N-r-(4-Benzoyl)-lysine-4-anilide]diphos-
phate (1). Trimethylsilyl bromide (0.053 mL, 0.401 mmol) was added
to a solution of 5 (0.074 g, 0.073 mmol) in anhydrous THF (2 mL)
(f) 1,4-Diamidinobenzene Dihydrochloride (2). 2 was synthesized
by a modified procedure of Garigipati.15 1,4-Dicyanobenzene (0.99
g, 7.72 mmol) was added as a solid to a 0.97 M stock solution of
methylchloroaluminum amide in benzene,16 and the resulting mixture
was refluxed for 48 h under argon. The reaction mixture was then
cooled and poured into a slurry of silica gel (20 g) in chloroform. The
slurry was stirred for 5 min, filtered, resuspended in water, filtered,
and washed with water (200 mL). The water and the chloroform
filtrates were evaporated and the residues combined. The crude product
was crystallized from water/EtOH to yield an off-white crystalline solid
ï
under argon at -40 C. The reaction mixture was allowed to stir for
1 h while adjusting to ambient temperature. The progress of the reaction
was followed by 31P NMR (DMSO). Further trimethylsilyl bromide
(0.053 mL, 0.401 mmol) was added four times at intervals of 0.5 h.
The solvent from the reaction mixture was then removed and the residue
was dissolved in MeOH (2 mL) and stirred for 2 days. The solvent
was then removed, and the residue was dissolved in water (2 mL), pH
adjusted to 7-8, and purified by ion-exchange chromatography
(Sephadex DEAE 25, NH4HCO3 gradient of 0 to 1 M). Lyophilization
of the eluant from the fractions containing 0.7-0.8 M salt yielded
1
of 2 (1.14 g, 63%); H NMR (300 MHz, D2O + DMSO) δ 8.08 (s,
-Ph-). Anal. Calcd for C8H10N4‚2HCl: C, 40.87; H, 5.14; N, 23.83;
Cl, 30.16. Found: C, 40.97; H, 5.11; N, 23.71; Cl, 30.08.
Acknowledgment. The authors thank Mr. J. Zweigenbaum
and Dr. J. Henion at Cornell University for performing mass
spectrometry analysis of compound 1.
1
compound 1 (0.011 g, 29%, as determined by H NMR) as a white
1
fluffy solid; H NMR (400 MHz, D2O, pD )8.1) δ 7.60 (d, J ) 6.4
Hz, 2H, -CO-Ph-), 7.11 (m, 4H, -CO-Ph- + -NH-Ph-), 7.00
(d, J ) 8.0 Hz, 2H, -NH-Ph-), 4.40 (m, 1H, -R-CH-), 2.82 (t, J
) 7.6 Hz, 2H, -CH2-ꢀ-NH3+), 1.90-1.65 (m, 2H, -R-CH-CH2-),
1.63-1.50 (m, 2H, -CH2-CH2-ꢀ-NH3+), 1.50-1.30 (m, 2H, -R-
CH-CH2-CH2-); 1H-1H COSY (300 MHz) indicated coupling within
the following pairs of signals: δ 7.6 and 7.11 ppm, δ 7.11 and 7.00
ppm, δ 2.82 and 1.63-1.50 ppm, and δ 1.90-1.65 and 1.50-1.30
ppm. 31P NMR (162 MHz, D2O, pD ) 8.1) δ 0.343, 0.019. HPLC
analysis of 1 (see Supporting Information) showed >95% purity. The
structure of 1 was confirmed by the electrospray ionization (ESI) MS
and MS/MS (see Supporting Information). The negative mode ESI
Supporting Information Available: Ki determination pro-
tocols, procedures for HPLC analysis of the cleavage reactions,
and ESI MS and MS/MS spectra of 1 (6 pages, print/PDF).
See any current masthead page for ordering information and
Web access instructions.
JA980810Z
(15) Garigipati, R. S. Tetrahedron Lett. 1990, 31, 1969.
(16) Levin, J. I.; Turos, E.; Weinreb, S. M. Synth. Commun. 1982, 12,
989.