Synthesis and bioactivities of nitronyl nitroxide and RGD containing pseudopeptides

Article abstract of DOI:10.1016/j.bmcl.2003.08.041

1-(1′,3′-dioxyl-4′,4′,5′,5′- tetramethyldihydroimidazol-2′-yl)-phenyl-4-yloxylacetic acid (3), and 1-(1′,3′-dioxyl-4′,4′,5′,5′- tetramethyldihydroimidazol-2′-yl)-phenyl-4-yloxylacetyl-RGDS (13), -RGDV (14), -RGDF (15) were synthesized. The ESR measurement gave the same spectroscopy for 3 and 13-15. The NO scavenging tests in vitro, anti-platelet aggregation tests in vitro and the anti-thrombosis assay in vivo indicated that introducing 3 into the N-terminal of RGDS, RGDV and RGDF the corresponding bioactivities for both of 3 and RGD peptides can be remained completely. The present combinations provided a beneficial strategy for simultaneous scavenging NO and anti-thrombosis, and for the use of spin label of RGD peptides in the conformational researches.

Full text of DOI:10.1016/j.bmcl.2003.08.041

Bioorganic & Medicinal Chemistry Letters 13 (2003) 4065–4069
Synthesis and Bioactivities of Nitronyl Nitroxide and RGD
Containing Pseudopeptides
Junling Liu, Ming Zhao, Chao Wang and Shiqi Peng*
College of Pharmaceutical Sciences, Peking University, Beijing 100083, PR China
Received 26 May 2003; accepted 18 August 2003
Abstract—1-(10,30-Dioxyl-40,40,50,50-tetramethyldihydroimidazol-20-yl)-phenyl-4-yloxylacetic acid (3), and 1-(10,30-dioxyl-40,40,50,50-
tetramethyldihydroimidazol-20-yl)-phenyl-4-yloxylacetyl-RGDS (13), -RGDV (14), -RGDF (15) were synthesized. The ESR
measurement gave the same spectroscopy for 3 and 13–15. The NO scavenging tests in vitro, anti-platelet aggregation tests in vitro
and the anti-thrombosis assay in vivo indicated that introducing 3 into the N-terminal of RGDS, RGDV and RGDF the corre-
sponding bioactivities for both of 3 and RGD peptides can be remained completely. The present combinations provided a beneficial
strategy for simultaneous scavenging NO and anti-thrombosis, and for the use of spin label of RGD peptides in the conformational
researches.
# 2003 Elsevier Ltd. All rights reserved.
Introduction
oligopeptides with anti-thrombotic and/or thrombolytic
activity.11ꢀ13 The results obtained indicated that the
contribution of RGD sequence to the specific interac-
tion of RGD containing peptides and pseudopeptides
may resultin ht e desirable lead compounds. Based on
the abilities of nitronyl nitroxide as NO scavenging or
spin labeling agent and of RGD peptide as anti-throm-
botic agent a number of combinations of them were
considered. As part of a program to develop thrombus
related therapeutic psedopeptides in the present paper 1-
(10,30-dioxy-40,40,50,50-tetramethyldi-hydroimidazol-20-yl)-
phenyl-4-yloxyacetic acid (3) was introduced into the N-
terminal of RGDS, RGDV and RGDF and their ESR
measurement, scavenging NO in vitro, anti-platelet
aggregation in vitro, anti-platelet aggregation in vitro
and anti-thrombotic in vivo were observed.
The data resulted from a number of investigations sug-
gest that the impact of NO on ischemic brain injury
depend on the stage of evolution of the tissue damage
and glutamate–NO pathway may occur in ischemic
penumbra, by which large amounts of NO are produced
by nNOS and might contribute to metabolic deteriora-
tion of the penumbra leading to lager infarction.1ꢀ3 The
therapeutic strategies have attempted to reduce the
damage either by intervening in the formation process
of NO or by scavenging NO already formed. For the
strategy of scavenging NO already formed and spin
labeling of peptides nitronyl nitroxide attracts extensive
attention.4ꢀ6
Platelet aggregation plays an essential role in normal
homeostasis and is dependent on the interaction of
membrane glycoprotein GPIIb/IIIa complex with
plasma adhesive glycoproteins.7,8 In addition to mono-
clonal antibodies the binding function of GPIIb/IIIa
can be blocked by RGD containing peptides which can
also prevent platelet-dependent thrombus formation in
experimental models of coronary artery thrombosis.9,10
Chemistry
1-(10,30-dioxyl-40,40,50,50-tetramethyldihydromidazol-20-
yl)-phenyl-4-yloxylacetic acid (3) was synthesized
according to Scheme 1. Treating 40-hydroxy-2-phenyl-
1,3-dioxyl-4,4,5,5-tetra-methyldihydroimida-zol
(1),
which was prepared by use of the same procedure as
that in the literature14 from 2,3-dihydroxyamino-2,3-
dimethylbutane and p-hydroxylbenzaldehyde with
In our previous papers RGDS, RGDV and RGDF were
used as the building block in the modification of the
BrCH2COOC2H5
gave
1-(10,30-dioxyl-40,40,50,50-
tetramethyldihydroimidazol-20-yl)-phenyl-4-yloxylacetic
ethyl ester (2) as an intense blue powder in 90% yield.
In the presence of NaOH (2 mol/L), compound 2 was
*Corresponding author. Tel.: +86-10-8280-2482; fax: +86-10-8280-
2311; e-mail: sqpeng@mail.bjmu.edu.cn
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.08.041

4066
J. Liu et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4065–4069
Scheme 1. Preparation of 1-(10,30-dioxyl-40,40,50,50-tetramethyldihydroimidazol-20-yl)-phenyl-4-yloxylacetic acid 3: (a) Br2, NaOH (6 mol/l); (b) Zn,
NH4Cl; (c) p-HO–C6H4CHO; (d) PbO2; (e) BrCH2COOC2H5, NaOC2H5 THF; (f) NaOH (2 mol/L).
converted into the corresponding acid 3 in theoretical
yield (Scheme 1).
chromatography (Sephadex G10) 1-(10,30-dioxyl-
40,40,50,50-tetramethyldihydro-imidazol-20 -yl)-phenyl-4-
yloxylacetyl-RGDS (13),-RGDV (14), and-RGDF (15)
were obtained in 73%, 70%, and 80% yield,15 respec-
tively (Scheme 2).
With the same procedure as that in the literature10,12 the
protective tetrapeptides Boc-Arg(Tos)-Gly-Asp- (OcHex)-
Ser(Bzl)-OBzl (4), Boc-Arg(Tos)-Gly-Asp-(OcHex)-Val-
OBzl (5), Boc-Arg(Tos)-Gly-Asp-(OcHex)-Phe-OBzl (6)
were prepared. After removal of their Boc protecting
groups the protective peptides with N-terminal free were
coupled with compound 3 to give the nitronyl nitroxide
containing protective peptides (10–12) in 87, 90, 90%
yield, respectively. The deprotection of the protective
groups in the side chain of the amino acid residue with
anhydrous HF and the deposition of the reaction resi-
due in air or ether for 2 h the crude product, a colorless
powder, may turn into blue. After purification by
ESR Measurement
Center field: 3480 Gauss, Sweep Width: 100 Gauss,
Sweep Time: 100 s, Modulation Amplitude: 1.0ꢁ10ꢀ1
G, Time Constant 1.6ꢁ10ꢀ1 S, Modulation Frequency
100 KHZ, Microwave Frequency 9.72 GHZ, and
Microwave Power: 10 MW were used for ESR
measurementand no any difference between compound
3 and 13–15 was observed. In phosphate buffer (pH 7.4)
the ESR spectra of compound 3 and 13–15 may be
ꢀ7
monitored at 10ꢀ7 mol/L. At10
mol/L–10ꢀ5 mol/L
no any difference between the ESR spectra of com-
pound 3 and 13–15 was observed. The typical spectro-
scopy is given in Scheme 3.
Scavenging NO In Vitro16
Immediately after decapitation, rat aortic strips were
prepared and put in a perfusion bath with 5 mL of
Scheme 2. Preparation of 1-(10,30-dioxyl-40,40,50,50-tetramethyl-dihy-
droimidazol-20-yl)-phenyl-4-yloxylacetyl-RGDS (13), -RGDV (14),
and -RGDF (15): (a) DCC and NMM; (b) HF.
Scheme 3. The ESR spectra of compound 3 (a) and 15 (b) in phos-
phate buffer (pH 7.4, final concentration 10ꢀ5 mol/L).

J. Liu et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4065–4069
4067
warmed (37 ꢂC), oxygenated (95% O2, 5% CO2) Krebs
solution (pH 7.4). The aortic strips were connected to
the tension transducers and their relaxation–contraction
curves were recorded. The noradrenaline (NE) solution
(final concentration 10ꢀ9 mol/L) was added for inducing
the contraction of the strips, and when the hypertonic
contraction reached to the maximal level NE was
washed1 and the vessel strips were stabilized for 30 min.
After renewal of the solution NE (final concentration
10ꢀ9 mol/L) was added. When the hypertonic contrac-
tion value of aortic strips reached the peak 15 mL of
water (vehicle), or the solution of 3, 13–15 in 15 mL of
water (final concentration 10ꢀ6 mol/L) was added
respectively. After stabilization, 1.5 mL of Ach (final
concentration 10ꢀ6 mol/L) were added and the NO
scavenging activities of the compounds are expressed
with the inhibition percentage of Ach-induced vasor-
elaxation and listed in Table 1.
mol/L) and adenosine diphosphate (ADP, final concen-
tration 10ꢀ5 mol/L). The effects of compounds 3, 13–15
on PAF or ADP induced platelet aggregation were
observed. The maximal rate of platelet aggregation
(Am%) was represented by the peak height of aggrega-
tion curve. The data are listed in Tables 2 and 3 and the
statistical analysis of the date is carried out by use of
ANOVA test, p<0.05 is considered significant.
Anti-thrombotic In Vivo13
Male Wistar rats weighing 250–300 g (purchased from
Animal Center of Peking University) were used. The
tested compounds were dissolved in NS just before use
and kept in an ice bath. The rats were anesthetized with
pentobarbital sodium (80.0 mg/kg, ip), and the right
carotid artery and left jugular vein were separated. A 6
cm thread with exact weight was put into the middle of
the polyethylene tube. The polyethylene tube was full
with heparin sodium (50 IU/mL of NS) and one end
was inserted into the left jugular vein. From the other
end of the polyethylene tube, heparin sodium was
injected as anticoagulant, then the tested compounds
were injected, which this end was inserted into the right
carotid artery. In this case the tube was full of NS or
tested compound containing NS. The blood was flowed
Anti-platelet Aggregation In Vitro13
Platelet-rich plasma was prepared by centrifugation of
normal rabbit blood anticoagulation with sodium
citrate at a final concentration of 3.8%. The platelet
counts were adjusted to 2ꢁ105/mL by addition of auto-
logous plasma. Platelet aggregation studies were con-
ducted in an aggregometer using the standard
turbidimetric technique. The agonists used were plate-
let-activating factor (PAF, final concentration 10ꢀ7
Table 3. Effect of the compounds on PAF induced platelet aggrega-
tion
ꢀ
Compd
Am % (XꢃSD), atthe dose of
Table 1. Inhibition percentage of Ach-induced vasorelaxation
10ꢀ7 mol/L
10ꢀ6 mol/L
10ꢀ5 mol/L
ꢀ
Compd
%, XꢃSD
NS
Aspirin
3
RGDS
RGDV
RGDF
13
57.82ꢃ4.55
53.42ꢃ3.44
54.25ꢃ3.16
54.21ꢃ3.19
52.66ꢃ4.91
51.88ꢃ2.65a
52.10ꢃ3.25c
50.88ꢃ4.10b
50.02ꢃ3.98b
36.98ꢃ2.95a
52.98ꢃ4.39
37.10ꢃ3.25a
35.11ꢃ2.70a
25.18ꢃ2.28a
35.11ꢃ2.48a
33.22ꢃ3.09a
24.26ꢃ2.42a
29.99ꢃ2.95a
52.92ꢃ4.19
22.66ꢃ2.36a
20.50ꢃ2.01a
10.22ꢃ2.15a
22.05ꢃ2.33a
20.45ꢃ2.26a
9.98ꢃ2.27a
Water
3
13
14
15
RGDS
RGDV
RGDF
1.6ꢃ1.4
99.2ꢃ1.4a
88.9ꢃ1.2a,b
91.2ꢃ0.5a,b
88.9ꢃ5.3a,b
3.2ꢃ2.8
2.4ꢃ1.9
14
15
2.6ꢃ2.5
n=6; water=vehicle.
aCompare to water, p<0.001.
bCompare to 3, p>0.05.
n=8; NS=vehicle.
aCompare to NS, p<0.001.
bCompare to NS, p<0.01.
cCompare to NS, p<0.05.
Table 2. Effect of the compounds on ADP induced platelet aggrega-
tion
Table 4. Effect of the compounds on the thrombus weight
ꢀ
Compd
Am % (XꢃSD), atthe dose of
ꢀ
ꢀ
Compd
Wetthrombus ( XꢃSD mg) Dry thrombus (XꢃSD mg)
10ꢀ7 mol/L
10ꢀ6 mol/L
10ꢀ5 mol/L
NS (3 mL/kg)
Aspirin
3
RGDS
RGDV
RGDF
13
40.19ꢃ4.13
28.11ꢃ2.99c
38.10ꢃ3.18
36.96ꢃ3.78
32.62ꢃ3.20a
25.61ꢃ3.14c
32.09ꢃ3.42a
29.49ꢃ3.30c
24.30ꢃ2.43c
6.95ꢃ2.01
3.99ꢃ1.030b
6.17ꢃ0.70
6.68ꢃ1.54
5.13ꢃ1.06a
3.94ꢃ0.95b
5.08ꢃ1.21a
4.07ꢃ1.00b
3.16ꢃ0.45c
NS
Aspirin
3
RGDS
RGDV
RGDF
13
55.97ꢃ5.01
54.89ꢃ2.88
54.68ꢃ2.40
53.55ꢃ2.46
42.58ꢃ2.99a
51.96ꢃ4.18
40.14ꢃ3.11a
30.01ꢃ2.66a
51.60ꢃ3.99
17.02ꢃ2.69a
12.07ꢃ1.89a
10.13ꢃ1.02a
16.91ꢃ2.55a
11.85ꢃ1.65a
10.21ꢃ1.95a
50.06ꢃ3.29b
33.76ꢃ2.44a
51.46ꢃ3.24c
49.98ꢃ3.81b
35.25ꢃ2.96a
28.90ꢃ2.89a
24.15ꢃ2.96a
40.05ꢃ3.21a
24.99ꢃ2.54a
20.66ꢃ2.96a
14
15
14
15
n=8; dosage: 170 mmol/kg for aspirin, 5 mmol/kg for others; NS,
n=8; NS, vehicle.
vehicle.
aCompare to NS, p<0.001.
bCompare to NS, p<0.01.
cCompare to NS, p<0.0.5
aCompare to NS, p<0.05.
bCompare to NS, p<0.01.
cCompare to NS, p<0.001.

4068
J. Liu et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4065–4069
from the right carotid artery to the left jugular vein via
the polyethylene tube for 15 min. The thread was taken
out and weighed and the weight of the wet thrombus
was recorded. The thread was kept in a desiccator for 2
weeks and the weight of the dry thrombus was recorded.
The data are listed in Table 4. The statistical analysis of
the date is also carried out by use of ANOVA test,
p<0.05 is considered significant.
treatment potency for 13, 14 or 15 was again sub-
stantially same as that for RGDS, RGDV or RGDF
(Table 4). The results demonstrated that when 3 was
introduced into the N-terminal of RGDS, RGDV or
RGDF the formed linker as the the ligand may be still
suitable for its binding to GP IIb/IIIa receptor and the
anti-thrombolitic activity of RGDS, RGDV or RGDF
was consequently remained.
Scavenging NO is an important strategy in the preven-
tion of ischemic brain injury. The combination of
nitronyl nitroxide and RGD peptide provides a possi-
bility to scavenge NO and prevent thrombus formation
simultaneously. On the other hand the label of RGD
peptides with nitronyl nitroxide may provide new path
for further biological investigation as well.
Discussion
As the NO scavenging agent block 3 can be smoothly
introduced into the N-terminal of RGDS, RGDV and
RGDF. In the ESR measurements 3, 13–15 give the
same spectroscopy means that as the NO scavenging
block 3 was stable enough to the deprotection of the
protective groups on the side chain of the amino acid
residue in 3 containing peptides with CF3SO3H/TFA
(1:4, v/v) or anhydrous HF as the deprotection system.
The special stability of 3 to the conditions of peptide
synthesis indicates that as peptide spin label block 3
may have potential use.
Acknowledgements
The author Shiqi Peng thanks the National Natural
Science Foundation (20232020) and Key Research
Project(G 1998051111) of China for financial support.
Even though RGDS, RGDV and RGDF exhibited no
any NO scavenging activity the inhibition percentage of
Ach-induced vasorelaxation for 3, 13, 14, and 15 was
99.2ꢃ1.4, 88.9ꢃ1.2, 91.2ꢃ0.5, and 88.9ꢃ5.3, respec-
tively, which was significantly different from that
(1.6ꢃ1.4) for water (Table 1). The approximate equal
ability of NO scavenging for 3, 13, 14, and 15 demon-
strated that when RGDS, RGDV or RGDF was intro-
duced into 3 the formed linker as the substrate may be
still suitable for NO scavenging reaction and NO
scavenging activity of 3 was consequently remained.
References and Notes
1. Iadecola, C. TINS 1997, 20, 132.
2. Zhang, F.; Iodecola, C. J. Cereb. Blood Flow Metab. 1994,
14, 574.
3. Hobbs, A. J.; Higgs, A.; Moncada, S. Ann. Rev. Pharmacol.
Toxicol. 1999, 39, 191.
4. Weinkam, R. J.; Jorgensen, E. J. Am. Chem. Soc. 1971, 93,
7033.
At the concentrations of 10ꢀ7, 10ꢀ6, and 10ꢀ5 mol/L 3
exhibited no any effect on ADP and PAF induced
platelet aggregation but at the concentration of 10ꢀ6
and 10ꢀ5 mol/L 13, 14 and 15 obviously inhibited pla-
telet aggregation. When ADP activated platelets were
treated with 10ꢀ6 mol/L, and 10ꢀ5 mol/L of 13, 14 or 15
the platelet aggregation rate (%) was lowered from
55.97ꢃ5.01 (for NS, control) to 40.05ꢃ3.21,
24.99ꢃ2.54, 20.66ꢃ2.96, and 16.91ꢃ2.55, 11.85ꢃ1.65,
10.21ꢃ1.95, respectively. The treatment potency for 13,
14 or 15 was substantially same as that for RGDS,
RGDV or RGDF (Table 2). Approximately same
effects of 13, 14, 15, RGDS, RGDV or RGDF on PAF
induced platelet aggregation was also observed (Table
3). The results indicated that when 3 was introduced
into the N-terminal of RGDS, RGDV or RGDF the
formed linker as the ligand may be still suitable for its
binding to GP IIb/IIIa receptor and the anti-platelet
aggregation activity of RGDS, RGDV or RGDF was
consequently remained.
5. Swartz, H. M. Pure Appl. Chem. 1990, 62, 235.
6. Maeda, H.; Akaike, T.; Yoshida, M.; Suga, M. J. Leukicyte
Biology 1994, 56, 588.
7. Gralnick, H. R.; Coller, S. B. Blood 1984, 64, 797.
8. Timmons, S.; Hawiger, J. Trans. Assoc. Am. Phys. 1986, 99,
226.
9. Strony, J. T.; Folts, J. D.; Adelman, B. Clin. Res. 1989, 54,
117.
10. Rapoport, R. M. Eur. J. Pharmacol. 1985, 115, 219.
11. Zhao, M.; Peng, S. Q. J. Prakt. Chem. 1999, 341, 668.
12. Zhao, M.; Wang, C.; Yang, J.; Peng, S. Q. Prep. Biochem.
Biotechnol. 2000, 30, 247.
13. Lin, N.; Zhao, M.; Wang, C.; Peng, S. Q. Bioorg. Med.
Chem. Lett. 2002, 12, 585.
14. Ullman, F. E.; Osiecki, J. H.; Boocock, D. G. B.; Darcy,
R. J. Am. Chem. Soc. 1972, 94, 7049.
15. Wu, Y.F.; Zhao, M.; Wang, C.; Peng, S.Q. Bioorg. Med.
Chem. Lett. 2002, 12, 2331. The physical data: 1-(10,30-dioxyl-
40,40,50,50 -tetramethyldihydroimidazol-20 -yl)-phenyl-4-yloxyl-
acetic acid (3), Mp 155–157 ꢂC, EI–MS (m/e): 307 [M]+; Boc-
Arg(Tos)-Gly-Asp(OcHex)-Ser(Bzl)-OBzl (4), Mp 73–75 ꢂC,
[a]2D0=ꢀ34 (c 0.4, CHCl3), FAB-MS (m/e): 950 [M+H]+;
BocArg-(Tos)-Gly-Asp(OcHex)-Val-OBzl (5), Mp 79–81 ꢂC
[a]2D0=ꢀ24 (c 0.4, CHCl3), FAB-MS (m/e): 872 [M+H]+;
Boc-Arg(Tos)-Gly-Asp(OcHex)-Phe-OBzl (6), mp 165–170 ꢂC,
[a]2D0=ꢀ25 (c 0.4, CHCl3), FAB-MS (m/e): 920 [M+H]+;
RGDS, mp 110–111 ꢂC, [a]2D0=ꢀ20 (c 0.2, 6 N HCl), FAB-MS
(m/e): 434 [M+H]+; RGDV, mp 120–121 ꢂC, [a]2D0=ꢀ20 (c 0.3,
6 N HCl), FAB-MS (m/e): 446 [M+H]+; RGDF, mp 125–
The anti-thrombolitic in vivo of 3 and 13–15 indicated
that 3 exhibited no any effect on the thrombus weight
butatthe dose of 5 mmol/kg 13, 14, or 15 did exhibited
anti-thrombolitic effect. The thrombus weight after the
treatment with 13, 14, or 15 was 32.09ꢃ3.42,
29.49ꢃ3.30, and 24.30ꢃ2.43 mg, respectively. The

J. Liu et al. / Bioorg. Med. Chem. Lett. 13 (2003) 4065–4069
4069
126 ꢂC, [a]2D0=ꢀ40 (c 0.2, 6 N HCl), FAB-MS (m/e): 494
[M+H]+; 1-(10,30-dioxyl-40,40,50,50-tetramethyldihydroimidazol-
20 - yl) - phenyl - 4 - yloxylacetylArg(Tos) - Gly - Asp(OcHex)-
Ser(Bzl)-OBzl, mp 76–78 ꢂC, [a]D20=+50 (c 1, CH3OH), FAB-
MS (m/e): 1109 [M+H]+; 1-(10,30-dioxyl-40,40,50,50-tetra-
methyl-dihydroimidazol-20ꢂ-yl)-phenyl-4-yloxylacetylArg-Gly-
Asp-Ser-OH (13), mp 154 C (decomp.), [a]2D0=40 (c 1, H2O),
FAB-MS (m/e): 723 [M+H]+; 1-(10,30-dioxyl-40,40,50,50-tetra-
methyl-dihydroimidazol-20-yl)-phenyl-4-yloxylacetylArg-(Tos)-
Gly-Asp(OcHex)-Val-OBzl, mp 92–94 ꢂC [a]D20=+55 (c 1,
CH3OH), FAB-MS (m/e): 1031 [M+H]+; 1-(10,30-dioxyl-
40,40,50,50 -tetramethyl-dihydroimidazol-20 -yl)phenyl-4-yloxyl-
acetylArg-Gly-Asp-Val-OH (14), mp 160 ꢂC (decomp.),
[a]2D0=+17 (c 1, H2O), FAB-MS (m/e): 735 [M+H]+; 1-(10,30-
dioxyl-40,40,50,50-tetramethyl-dihydroimidazol-20 -yl)phenyl-4-
yloxylacetylArg-(Tos)-Gly-Asp(OcHex)-Phe-OBzl, mp 108–
110 ꢂC, [a]D20=ꢀ38 (c 1, CH3OH), FAB-MS (m/e): 1079
[M+H]+; 1-(10,30-dioxyl-40,40,50,50-tetramethyl - dihydroimi-
dazol-2-yl)phe-nyl-4-yloxylacetylArg-Gly-Asp-Phe-OH (15), mp
158 ꢂC (decomp.), [a]D20=+20 (c 1, H2O), FAB-MS
783[M+H]+.
16. Cowart, M.; Kowaluk, E. A.; Daanen, J. F.; Kohlhaas,
K. L.; Alexander, K. M.; Wagenaar, F. L.; Kerwin, J. F. J.
Med. Chem. 1998, 41, 2636.