Synthesis and biological evaluation of anthranilamide-based non-peptide mimetics of ω-conotoxin GVIA

Article abstract of DOI:10.1016/j.tet.2006.05.041

Non-peptide mimetics based on an anthranilamide 'scaffold' possessing fragments that mimic Lys2, Tyr13 and Arg17 in ω-conotoxin GVIA have been prepared. Compounds were assayed for binding to the voltage-gated calcium channels Ca_v2.2 ('N-type')

Full text of DOI:10.1016/j.tet.2006.05.041

Tetrahedron 62 (2006) 7284–7292
Synthesis and biological evaluation of anthranilamide-based
non-peptide mimetics of u-conotoxin GVIA
Jonathan B. Baell,a,b, Peter J. Duggan,c,d Stewart A. Forsyth,a Richard J. Lewis,e
*
Y. Phei Lok,c,d Christina I. Schroedere and Nicholas E. Shepherdc
aBiomolecular Research Institute, 343 Royal Parade, Parkville, Vic 3052, Australia
bStructural Biology—Chemistry Group, The Walter and Eliza Hall Institute of Medical Research, Biotechnology Centre,
4 Research Avenue, La Trobe R & D Park, Bundoora, Vic 3086, Australia
cSchool of Chemistry, Monash University, Clayton, Vic 3800, Australia
dCSIRO Molecular and Health Technologies, Bag 10, Clayton South, Vic 3169, Australia
eInstitute for Molecular Bioscience, The University of Queensland, St. Lucia, Qld 4072, Australia
Received 8 February 2006; revised 5 May 2006; accepted 18 May 2006
Available online 12 June 2006
Abstract—Non-peptide mimetics based on an anthranilamide ‘scaffold’ possessing fragments that mimic Lys2, Tyr13 and Arg17 in u-cono-
toxin GVIA have been prepared. Compounds were assayed for binding to the voltage-gated calcium channels Cav2.2 (‘N-type’) and Cav2.1
(‘P/Q-type’) in rat brain. The primary synthetic target, 2-(6-amino-hexanoylamino)-5-(3-guanidino-propoxy)-N-[4-(4-hydroxyphenoxy)-
phenyl]-benzamide (2a), exhibited low mM binding to Cav2.2 and was more than 30-fold selective for Cav2.2 over Cav2.1.
Crown Copyright Ó 2006 Published by Elsevier Ltd. All rights reserved.
1. Introduction
Pain management is one of the largest pharmaceutical mar-
ketsintheworldandisexpected4 toincreaseatacompounded
annual growth rate (CAGR) of 10% to reach $29.8 billion
(US) in 2008. It is therefore not surprising that intensive
screening events have focused on the discovery of small
molecule Cav2.2 inhibitors that might be orally active. This
has culminated in NeuroMed’s NMED-160, which is an
orally available blocker of Cav2.2 channels that is now in
Phase II clinical trials for a variety of pain conditions.5
u-Conotoxins are pharmacologically active toxins derived
from the venom of cone snails.1 u-Conotoxin GVIA (GVIA)
is produced by Conus geographus, while u-conotoxin
MVIIA (MVIIA) is produced by Conus magus. Both of these
polypeptides potently block the neuronal voltage-gated N-
type calcium channel (Cav2.2), which in humans is a target
for the relief of neuropathic pain. Indeed, u-conotoxin
MVIIA (PrialtÒ, ZiconotideÒ, SNX-111) was approved late
in 2004 for the treatment of severe chronic pain and shows
efficacy in cases where morphine-based analgesics are less
effective.2
Our interest is in the rational design of small molecule mi-
metics of peptide and protein binding epitopes. GVIA and
MVIIA are structurally defined and their binding epitopes
have been extensively mapped, making them attractive tar-
gets for mimetic design.1,6
Such therapeutic promise has driven the search for other
inhibitors of Cav2.2. Another u-conopeptide, u-conotoxin
CVID (AM336), has recently3 been taken into phase II clin-
ical trials by AMRAD and reportedly has a better therapeutic
index than Elan’s MVIIA. However, being polypeptides,
these agents suffer drawbacks in that they are relatively ex-
pensive to manufacture and possess poor pharmacokinetic
properties. Indeed, both MVIIA and CVID of necessity are
administered intrathecally.
We recently reported7 the synthesis and biological activity of
the benzothiazole derivative 1a (Fig. 1), which was designed
to mimic the side chains of K2, Y13 and R17 in GVIA and
bound Cav2.2 channels with a Ki of 1.8 mM. This simple 3-
residue mimetic was also selective for Cav2.2 over Cav2.1
(‘P/Q-type calcium channels’) and this was seen as therapeu-
tically desirable in order to minimise off-target side effects.8
The key to designing these mimetics is the application of
interactive de novo design since no commercially available
software could adequately address the challenge posed by
the large, discontinuous and disparate u-conotoxin binding
epitope.
Keywords: u-Conotoxin; Mimetic; Cav2.2 (‘N-type’) calcium channel
blocker.
* Corresponding author. Tel.: +61 3 9345 2108; fax: +61 3 9345 2211;
e-mail: jbaell@wehi.edu.au
0040–4020/$ - see front matter Crown Copyright Ó 2006 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.05.041

J. B. Baell et al. / Tetrahedron 62 (2006) 7284–7292
7285
with N-(3-bromopropyl)phthalimide and potassium carbon-
ate in DMF at 65 ꢀC for 4 h to give 4a in 74% yield. The nitro
group of 4a was then reduced, after dissolution in hot 60%
ethanolic THF, by portionwise addition of sodium dithionite
and water at 80 ꢀC, to give aniline 4b in 93% yield. This
aniline was then acylated with Cbz-protected 6-aminohexa-
noic acid using a typical HOBT/HBTU coupling protocol,
to give 5a in 90% yield. The phthalimide group was cleaved
using sodium borohydride in aqueous isopropanol10 fol-
lowed by treatment with acetic acid at 60 ꢀC for 48 h to
give the free amine 5b in 55% yield. In turn, this amine
was converted into a protected guanidyl group by treatment
with N,N0-bis(benzyloxycarbonyl)-1H-pyrazole-1-carbox-
amidine and triethylamine in methanolic DCM, to give 5c
in 44% yield. Finally, this was fully deprotected by catalytic
hydrogenation in trifluoroethanol to afford 2a as a colourless
solid in 64% yield.
Figure 1. The structure of the type-III mimetic of u-conotoxin GVIA based
on an N0-benzylated benzamidobenzothiazole core (1a), along with the
structure of an analogue (1b), both recently reported by us.7 Annotations
indicate the GVIA residue mimicked, and emboldened bonds are those
that mimic the corresponding a–b bond vectors in GVIA.
Analogue 2b was obtained from a sample of precursor 5b
that was deprotected by catalytic hydrogenation in metha-
nolic ethanol, to give 2b in 31% yield.
Since the side chain termini targeted for the mimicry of
GVIA are not spatially well defined in solution, our approach
is to concentrate on the design of scaffolds that mimic the tar-
geted a–b bond vectors as these are conformationally better
resolved. In order to investigate whether other scaffolds
could be developed besides those based on a benzothiazole
template, we have designed the anthranilamide derivative
2a (Fig. 2). A solid-state structure of the anthranilamide scaf-
fold confirmed that it could serve as a suitable K2–Y13–R17
mimetic of GVIA. We have previously reported this crystal
structure along with preliminary synthetic and functional
activity details for this mimetic.9 Herein, we characterise
the full synthesis of 2a and its analogue 2b, using a different
and more efficient synthetic route to that which we have
previously outlined, and report binding data for these
compounds to Cav2.2 and Cav2.1.
Intermediate 3 was not our initial choice of mimetic precur-
sor. Instead, the aryl fluoride 8 in Scheme 2 was targeted on
the basis that previous preliminary studies showed that the
labile fluorine atom could be displaced by N-Boc-protected
3-aminopropanol as its alkoxide in DME. The aryl fluoride
8 was readily made by coupling our previously reported7
aniline 6 with 5-fluoro-2-nitrobenzoic acid (7). In an initial
side project on R17 guanidine isosteres, we investigated the
use of pyridin-2-ylethanol and sodium hydride in DME to
see ifwe could displace the labile fluoride togiveacompound
such as 9. However, only occasionally did 9 result, and even
then only in small yields, and the predominant product in-
stead was repeatedly 3. With large amounts of 3 at our dis-
posal, we realised that alkylation of the phenolic group in 3
using readily available 3-aminopropylbromide, N-protected
as the phthalimide, should provide a facile route to our other
target mimetics 2a and 2b. As described above for Scheme 1,
this indeed proved to be the case.
Although labile aryl fluoride atoms are reported to readily
furnish phenols through alkaline hydrolysis,11,12 model reac-
tions led us to suggest that a mechanism involving base-cat-
alysed beta-elimination of initially formed intended product
9 is involved in the production of 3, facilitated by the acidic
nature of the pyrid-2yl-methylene proton and the excellent
leaving group ability of the para-nitrophenoxide anion, as
shown in Scheme 3. The model reactions involved addition
of 2,4-dinitrofluorobenzene to pyridin-2-ylethanol/NaH in
DME, whereupon the only pyridine-containing species pro-
duced was 2-vinylpyridine. On the other hand, addition of
the alkoxide to the model aryl fluoride, conditions under
which excess strong base would not exist, gave the model
product as intended. Our proposed mechanism is also sup-
ported by the published use of S-pyrid-2-ylethyl groups as
surrogates for thiols, which are unmasked through a base-
catalysed beta-elimination mechanism analogous to that
postulated here.13
Figure 2. The structure of the type-III mimetic of u-conotoxin GVIA based
on an N0-aryl-N00-acyl anthranilamide core (2a), along with the structure of
an analogue (2b), reported herein. Annotations indicate the GVIA residue
mimicked, and emboldened bonds are those that mimic the corresponding
a–b bond vectors in GVIA.
2. Results
2.1. Mimetic synthesis
Other unexpected behaviour was encountered for this scaf-
fold when we synthesised mimetics with a greater degree
of orthogonal protection than that exhibited in 5c. Thus,
Mimetics were synthesised from key intermediate 3, as
shown in Scheme 1. Firstly, the phenolic group was alkylated

7286
J. B. Baell et al. / Tetrahedron 62 (2006) 7284–7292
O
O
O
NO2
O
R
a
BnO
N
H
BnO
N
H
4a R = NO2 (74%)
4b R = NH2 (93%)
b
3
OH
O(CH2)3NPhth
(CH2)5NHCbz
N O
O
H
O
c
BnO
N
H
O(CH2)3R
5a R = NPhth (90%)
5b R = NH2 (55%)
d
e
f
2b (31%)
g
2a (64%)
5c R = N=C(NHCbz)2 (44%)
Scheme 1. Synthesis of mimetic 2a and its analogue 2b from key precursor 3. Reagents and conditions: (a) N0-(3-bromopropyl)phthalimide, K2CO3, DMF,
65 ꢀC; (b) Na2S2O4, EtOH, H2O, reflux; (c) CbzN(CH2)5CO2H, HBTU, HOBt, Et3N, DMF, rt; (d) (i) NaBH4, i-PrOH, H2O, DCM, rt; (ii) CH3CO2H, reflux;
(e) N,N0-bis(benzyloxycarbonyl)-1H-pyrazole-1-carboxamidine, DCM, rt; (f) H2, Pd/C, MeOH/EtOH, rt; (g) H2, Pd/C, trifluoroethanol, rt.
O
O
NO2
O
BnO
N
H
9
(minor, 19%)
O
O
a
b
O
NO2
F
+
N
BnO
N
H
BnO
NH2
6
8 (68%)
3
(predominant, 80%)
Scheme 2. Synthesis of key precursor 3 as the predominant by-product instead of the intended target 9. Reagents and conditions: (a) 5-fluoro-2-nitrobenzoic acid
(7), HBTU, Et3N, DMF, rt; (b) 2-(2-hydroxyethyl)pyridine, NaH, DME, rt.
O
O
O
NO2
O
NO2
O–
N
+
BnO
N
H
BnO
N
H
3
O
9
N
2-vinylpyridine
H
B
Scheme 3. Proposed base-induced (B) beta-elimination in 9 to account for the observed production of 3 under anhydrous conditions (see Scheme 2) where the
intended product was 9.
5b was readily converted to 10 by reaction with N,N0-
bis(tert-butyloxycarbonyl)-1H-pyrazole-1-carboxamidine in
DCM as shown in Scheme 4. However, when we attempted
simultaneous bis-BOC and benzyl ether deprotection, using
the TFA/thioanisole method of Kiso et al.,14 only the cleaved
compound 11 was isolated after work up and purification.
2.2. Mimetic binding affinity for Cav2.2 and Cav2.1
Mimetics 2a and 2b were assayed for binding to Cav2.2 using
125I-GVIA as a ligand. The K2–Y13–R17 mimetic 2a bound
with a Ki of 3.5 mM while the K2–Y13–R17K mimetic 2b
bound with a Ki of 13 mM. These same mimetics were
(CH2)5NH2
O
(CH2)5NHCbz
H
O
H
a
N
O
N
O
b
5b
HO2C
BnO
N
H
11 (58%)
O(CH2)3N=C(NH2)2
10 (39%)
O(CH2)3N=C(NHBoc)2
Scheme 4. Synthesis of orthogonally protected mimetic 10 and its subsequent decomposition to 11 during attempted deprotection. Reagents and conditions:
(a) N,N0-bis(tert-butyloxycarbonyl)-1H-pyrazole-1-carboxamidine, DCM, rt; (b) TFA/thioanisole, DCM, rt.

J. B. Baell et al. / Tetrahedron 62 (2006) 7284–7292
7287
assayed for binding to Cav2.1 using 125I-MVIIC as a ligand.
The K2–Y13–R17 mimetic 2a bound with a Ki of 111 mM
while the K2–Y13–R17K mimetic 2b bound with a Ki of
176 mM (Table 1).
solution conformation of GVIA, which is represented by
its peptide backbone conformation as a yellow tube. All
residues in GVIA have been removed apart from the side
chains of the targeted residues. This makes it quite clear how
the anthranilamide scaffold can mimic the a–b bond vectors
of the targeted residues, K2, Y13 and R17, in GVIA and
how, if suitably functionalised, it would be a structural
K2–Y13–R17 GVIA mimetic.
Table 1. Binding potencies (mM) for mimetics 2a and 2b to Cav2.2 (N-type)
and Cav2.1 (P/Q-type) calcium channels (95% confidence intervals shown
in parentheses)
We have previously reported7 another K2–Y13–R17 mi-
metic using a benzothiazole-based scaffold (1a) and for
this system, undertook more extensive testing of truncated
analogues to show that each of the three mimetic side chains
contributed to the low micromolar affinity for Cav2.2. For
the current anthranilamide system, we have only investi-
gated the variant 2b, which contains a lysine-like side chain
instead of an arginine-like side chain as the R17 side chain
mimetic. However, the drop in potency that we see parallels
the drop in potency that accompanies the analogous change
in the benzothiazole-based mimetic system (1a to 1b), and
we propose that the mimetic side chains in 2a all interact
favourably with the ion channel as they do for 1a and
moreover, that both compounds are true GVIA mimetics.
Compound
Ki (Cav2.2)
Ki (Cav2.1)
2a
2b
3.5 (2.5–4.8)
13.1 (9.5–18.0)
111 (70–180)
176 (140–220)
3. Discussion
The relatively potent, low micromolar binding affinity for
the K2–Y13–R17 mimetic 2a to Cav2.2 (N-type calcium)
channels is impressive and is made more remarkable by
the selectivity of this compound for these channels over
Cav2.1 (P/Q-type calcium) channels. This compound ap-
pears to be highly optimisable since the K2 and R17 mimetic
side chains contain significant flexibility and are suited to
conformational constraint. It is not proven that 2a mimics
the three targeted residues in GVIA exactly in binding to
Cav2.2 as designed, but this is a reasonable assumption since
molecular modelling has shown9 that this scaffold is a good
structural mimic of K2, Y13 and R17 in GVIA. This is more
clearly shown in Figure 3, where the solid-state conforma-
tion of the scaffold9 is superimposed on the NMR-derived
Interactive de novo design can be the most efficient way to
develop type-III mimetics of binding epitopes.1,6,7,9,15,16
Here, synthetic and conformational knowledge is applied
by the medicinal chemist interactively with a modelling
package to design a suitable mimetic scaffold that is syn-
thetically tractable, conformationally appropriate, not too
Figure 3. Three-dimensional picture of the anthranilamide scaffold ‘core’ superimposed on the NMR solution structure of u-conotoxin GVIA. The peptide
backbone is shown as a mustard yellow translucent tube. Only residues Lys2, Tyr13 and Arg17 of GVIA are shown, these in cyan with their respective
Ca–Cb bonds being coloured purple. The scaffold is coloured by atom type with only the NH hydrogens being displayed for the sake of clarity. The mimetic
Ca–Cb bonds are coloured yellow. It can be seen that the purple GVIA bonds match closely with their yellow counterparts in the mimetic. The conformation of
the scaffold used is that determined by X-ray crystallographic analysis, as previously described.9

7288
J. B. Baell et al. / Tetrahedron 62 (2006) 7284–7292
drug-unlike and highly analogable. Occasional structural re-
laxation with reference to known crystallographic substruc-
ture conformation is used in the process. Although this
procedure is inherently suited to automation and computa-
tional approaches have been developed,17 in our hands we
find our approach to be the most successful and efficient.
Binding epitopes targeted by interactive de novo design
may vary from continuous and contiguous, to discontinuous
and discontiguous1,6,7,9,15,16 and so although we have tar-
geted mimetics of pharmacologically active toxins, our
approach may be suitable for other polypeptide–protein or
protein–protein interactions. We are currently applying this
technique with considerable success to the design of mi-
metics of the alpha-helical binding domain of pro-apoptotic
BH3-only proteins in their complex with anti-apoptotic
Bcl-2 homologues.18 In the current work, we target a pharma-
cophore that is both discontinuous and discontiguous. In a
sense, the disparate nature of the pharmacophore allows for
more choice in the scaffold construction, and we have now
shown how two, quite distinct scaffolds can act as GVIA mi-
metics in targeting the same three residues. This gives some
additional versatility to the process where one scaffold may
be interchanged for another with better ADMET properties.
Resolution MS was recorded on a Micromass Platform II
mass spectrometer. The compounds analysed were dissolved
in organic solvent and ionised using an electrospray ionisa-
tion source. MS data are recorded as positive electrospray
ions unless indicated by (ESI-) where negative ions are
reported. Elemental analyses were conducted by CMAS
Chemical and Micro Analytical Services Pty. Ltd. (P.O.
Box 248, Belmont, Victoria 3216, Australia).
5.1. 2-(6-Aminohexanoylamino)-5-(3-guanidinoprop-
oxy)-N-[4-(4-hydroxyphenoxy)-phenyl]-benzamide (2a)
The fully protected compound (5c) (17.5 mg, 0.017 mmol)
was dissolved in trifluoroethanol (6 mL) and 10% Pd/C
was added. The reaction mixture was stirred under H2 at at-
mospheric pressure for 4 h. The catalyst was then removed
by filtration and washed with MeOH (2ꢁ3 mL) and the com-
bined filtrates and washings concentrated to dryness under
reduced pressure. The solid was then triturated with DCM
(2ꢁ3 mL), dissolved in water and centrifuged. Freeze-drying
of the supernatant afforded 2a as a white solid (5.8 mg, 64%).
Mp 134–136 ꢀC; 1H NMR (300 MHz, methanol-d4) d ppm:
1.44 (m, 4H), 1.70 (p, J¼7.2 Hz, 2H), 2.10 (p, J¼7.2 Hz,
2H), 2.40 (t, J¼6.0 Hz, 2H), 2.60 (t, J¼6.9 Hz, 2H), 3.44
(t, J¼6.8 Hz, 2H), 4.16 (t, J¼5.9 Hz, 2H), 6.76–6.94 (m,
6H), 7.14 (dd, J¼3.0, 9.0 Hz, 1H), 7.32 (d, J¼2.7 Hz, 1H),
7.58 (dd, J¼2.1, 9.0 Hz, 2H), 7.89 (d, J¼8.7 Hz, 1H); 13C
NMR-APT (75 MHz, methanol-d4; 1 unassigned ArC)
d ppm: 27.5 (CH2), 28.3 (CH2), 30.5 (CH2), 34.3 (CH2),
38.9 (CH2), 40.4 (CH2), 43.2 (CH2), 67.4 (CH2), 116.0
(CH), 118.6 (CH), 119.1 (CH), 119.3 (CH), 122.7 (CH),
124.7 (CH), 127.2 (CH), 130.4 (C), 132.1 (C), 134.7 (C),
150.5 (C), 157.7 (C), 158.3 (C), 159.6 (C), 169.5 (C), 175.5
(C); ATR (neat) cmꢂ1: 3278 (O–H), 2932, 2847 (aliphatic
C–H), 1652, 1600 (C]O), 1497 (C]N); HRMS calcd for
C29H36N6O5 (M+H+) 549.28254, found 549.2828.
One important factor in the success of our approach is the
choice of neither more nor less than three residues to target.
Less than three is unlikely to register activity, more than
three is likely to render initial design and synthesis overly
complex. In this regard, we found it quite surprising that mi-
metic 2a is as potent as it is. Rudimentary thermodynamic
considerations suggest that the side chains in this type of
mimetic might contribute significantly more to the binding
process than they do in GVIA itself.1,7,15
4. Conclusion
Non-peptide mimetics of u-conotoxin GVIA have been pre-
pared based on an anthranilamide scaffold that projects side
chain mimetics of lysine, tyrosine and arginine in a similar
respective manner to the projection of K2, Y13 and R17 in
GVIA. In so doing, mimetic 2a exhibits a Ki to Cav2.2 of
3.5 mM and is over 30-fold more selective for this channel
than Cav2.1 channels. We have selected 2a for further opti-
misation and assessment of functional antagonism of the
Cav2.2 channel.
5.2. 2-(6-Aminohexanoylamino)-5-(3-aminopropoxy)-
N-[4-(4-hydroxyphenoxy)-phenyl]-benzamide (2b)
The protected compound (5b) (26 mg, 0.036 mmol) was dis-
solved in absolute EtOH (5 mL) and MeOH (2 mL) and
stirred under H2 with 10% Pd/C for 3 h. The mixture was fil-
tered and the filtrate evaporated to dryness. The residue was
triturated with hexanes (4ꢁ8 mL) and ether (3ꢁ8 mL) and
then dissolved in water and centrifuged. The supernatant
was removed and freeze-dried to give 2b as a green solid
(4.8 mg, 31%). Mp 146–148 ꢀC; 1H NMR-COSY
(300 MHz, methanol-d4) d ppm: 1.46 (p, J¼7.5 Hz, 2H),
1.71 (m, J¼8.2, 7.2 Hz, 4H), 2.20 (p, J¼6.6 Hz, 2H), 2.43
(t, J¼7.4 Hz, 2H), 2.91 (t, J¼7.4 Hz, 2H), 3.19 (t,
J¼7.2 Hz, 2H), 4.22 (t, J¼5.9 Hz, 2H), 6.81–6.96 (m, 6H),
7.14–7.18 (dd, J¼3.0, 9.0 Hz, 1H), 7.37 (d, J¼2.7 Hz, 1H),
7.52 (d, J¼9.0 Hz, 2H), 7.93 (d, J¼9.0 Hz, 1H); 13C NMR
(75 MHz, methanol-d4) d ppm: 26.9 (CH2), 27.7 (CH2),
29.1 (CH2), 29.4 (CH2), 38.6 (CH2), 39.5 (CH2), 41.4
(CH2), 67.7 (CH2), 112.5 (CH), 116.0 (CH), 116.1 (2CH),
117.9 (2CH), 118.2 (CH), 119.3 (2CH), 119.5 (2CH),
130.0 (C), 132.4 (C), 134.9 (C), 151.5 (C), 155.8 (C), 157.4
5. Experimental
All commercially obtained chemicals and reagents were
used as received. Melting points were recorded on a Reichert
hot stage melting point apparatus. 1H and 13C nuclear mag-
netic resonance (NMR) spectra were recorded on a Varian
Mercury 300 MHz spectrometer using the solvents speci-
fied; exchangeable NH and OH protons are only assigned
1
where specified. 19F-decoupled H and 13C NMR spectra
were run on Bruker DPX 300 MHz spectrometer FTIR
were run on Perkin–Elmer 1600 Series FTIR. ATR IR spec-
tra were run on a Bruker IFS 55 FTIR Specac single reflec-
tion ATR system fitted with a single bounce diamond top
plate. HRMS of compounds were recorded on a Bruker Bio-
Apex 47e Fourier Transform mass spectrometer. Low
(C), 158.1 (C), 169.6 (C), 175.0 (C); ATR (neat) cmꢂ1
3237 (O–H), 2927, 2849 (aliphatic C–H), 1647, 1596
(weak), 1496 (C]O); MS (ESI) (M+H+) 507.3; HRMS calcd
for C28H34N4O5 (M+H+) 507.26075, found 507.2606.
:

J. B. Baell et al. / Tetrahedron 62 (2006) 7284–7292
7289
5.3. N-[4-(4-Benzyloxyphenoxy)-phenyl]-5-hydroxy-2-
nitrobenzamide (3)
(NO2); HRMS calcd for C37H29N3O8 (M+Na+) 666.1852,
found 666.1846.
2-(2-Hydroxyethyl)pyridine (0.1 mL, 0.89 mmol) was added
to a slurry of NaH (60% in mineral oil, 21.6 mg, 0.54 mmol)
in anhydrous DME (0.25 mL) under N2. After 20 min, a solu-
tion of 8 (36 mg, 0.079 mmol) dissolved in anhydrous DMF
(3 mL) was added dropwise over 25 min at room tempera-
ture. The reaction mixture changed from an orange-yellow
colour to a brown colour overnight (24 h). DME was re-
moved under reduced pressure and the resulting residue dis-
solved in EtOAc (30 mL). The EtOAc solution was washed
with 2 M HCl (2ꢁ40 mL). Washings with 2 M NaOH
(2ꢁ40 mL) were carried out until the aqueous solution be-
came colourless. The organic layer was then washed with
saturated brine (2ꢁ30 mL), dried (MgSO4) and filtered.
The organic solvent was removed under reduced pressure.
Purification by radial chromatography using 100% EtOAc
afforded 3 (28.7 mg, 80%) as a yellow solid after removal
of solvent (a small amount of 9 (8.6 mg, 19%) was also fur-
5.5. 2-Amino-N-[4-(4-benzyloxyphenoxy)-phenyl]-5-[3-
(1,3-dioxo-1,3-dihydro-isoindol-2-yl)-propoxy]-benz-
amide (4b)
Nitrobenzene (4a) (0.288 g, 0.448 mmol) was dissolved in
95% EtOH (20 mL) and THF (15 mL) and heated to 80 ꢀC.
Upon dissolution of the solid, Na2S2O4 (0.312 g, 1.8 mmol)
was added, followed by H2O (4 mL). The temperature was
maintained at 80 ꢀC for a further 2 h. The solvent was then
removed under reduced pressure. EtOAc (40 mL) and 2 M
HCl (40 mL) were added to the residue and the aqueous layer
extracted with EtOAc (2ꢁ30 mL). The combined organic ex-
tracts were then washed with saturated NaHCO3 (2ꢁ40 mL),
saturated brine (2ꢁ30 mL) and dried (MgSO4), filtered and
evaporated to dryness to give 4b as a light yellow solid
(0.255 g, 93%). Mp 140–143 ꢀC; 1H NMR (300 MHz,
CDCl3) d ppm: 2.03 (p, J¼6.4 Hz, 2H), 3.87 (t, J¼6.9 Hz,
2H), 3.94 (t, J¼5.7 Hz, 2H), 4.98 (s, 2H), 6.57–6.92, 7.08–
7.09, 7.26–7.38, 7.52–7.55 (m, 16H), 7.60–7.76 (m, 4H);
13C NMR-APT (75 MHz, CDCl3; 1 ArCH and 11 ArC unas-
signed) d ppm: 28.1 (CH2), 35.3 (CH2), 67.7 (CH2), 70.5
(CH2), 115.4 (CH), 115.8 (2CH), 118.4 (2CH), 119.0 (CH),
120.2 (2CH), 122.3 (2CH), 123.2 (2CH), 127.4 (2CH),
127.9 (CH), 128.5 (2CH), 131.9 (C), 132.9 (C), 133.9
(2CH); ATR (neat) cmꢂ1: 3461, 3372 (N–H), 3271 (aromatic
C–H), 2951, 2862 (aliphatic C–H), 1811 (C]O), 1639
(phthalimide C]O); HRMS calcd for C37H31N3O6
(M+H+) 614.2291, found 614.2291.
1
nished as a brown amorphous solid). H NMR (300 MHz,
acetone-d6) d ppm: 5.08 (s, 2H), 6.88–7.05 (m, 8H), 7.35–
7.47 (m, 5), 7.62 (d, J¼9.3 Hz, 2H), 8.09 (d, J¼9.3 Hz,
1H), 10.50 (s, 1H, NH); 13C NMR (75 MHz, acetone-d6;
1 ArCH remains unassigned) d ppm: 70.30 (CH2), 115.52
(CH), 116.13 (2CH), 118.40 (2CH), 120.31 (2CH), 121.50
(2CH), 127.41 (CH), 127.76 (2CH), 127.94 (CH), 128.59
(2CH), 134.48 (C), 136.49 (C), 137.68 (C), 138.21 (C),
151.05 (C), 154.60 (C), 155.20 (C), 163.08 (C), 164.43 (C);
ATR (neat) cmꢂ1: 3217 s, 3047 m, 2875 w, 1580 m, 1494 s,
1444 m, 1386 w, 1323 s, 1263 m, 1209 s, 1102 w, 1063 m,
878 m, 819 m, 742 m, 695 m cmꢂ1. MS (ESI) m/z 457.3
[M+H]+. HRMS: Found 456.1303 (requires 456.1321 for
C26H20N2O6). Microanalysis: Found (%) C 68.51 H 4.38 N
6.19 (requires (%) C 68.42 H 4.42 N 6.14 for C26H20N2O6).
5.6. (5-{2-[4-(4-Benzyloxyphenoxy)-phenyl-carbomoyl]-
4-[3-(1,3-dioxo-1,3-dihydro-isoindol-2-yl)-propoxy]-phe-
nylcarbomyl}-pentyl)carbamic acid benzyl ester (5a)
5.4. N-[4-(4-Benzyloxyphenoxy)-phenyl]-5-[3-(1,3-di-
oxo-1,3-dihydro-isoindol-2-yl)-propoxy]-2-nitrobenz-
amide (4a)
Aniline (4b) (0.180 g, 0.29 mmol) and 6-benzyloxy-carbonyl
aminohexanoic acid (0.156 g, 0.58 mmol) were dissolved
in DMF (10 mL). Triethylamine (0.16 mL, 0.11 mol) was
added, followed by HOBt (0.080 g, 0.5 mmol) and HBTU
(0.223 g, 0.5 mmol). The reaction mixture was stirred at
room temperature overnight then poured into 2 M HCl
(50 mL). This aqueous solution was extracted with DCM
(3ꢁ30 mL). The combined DCM extracts were washed
with 2 M HCl (30 mL), then with saturated NaHCO3
(2ꢁ30 mL) followed by saturated brine (2ꢁ30 mL). After
drying (MgSO4) and filtering, the DCM was removed under
reduced pressure to give the crude product as a white precip-
itate. This precipitate was re-crystallised from EtOAc to af-
ford 5a as a white solid (0.228 g, 90%). Mp 157–159 ꢀC;
1H NMR (300 MHz, CDCl3) d ppm: 1.39 (p, J¼6.6 Hz,
2H), 1.52 (p, J¼7.5 Hz, 2H), 1.71 (p, J¼7.5 Hz, 2H), 2.11
(p, J¼6.2 Hz, 2H), 2.35 (t, J¼7.4 Hz, 2H), 3.17 (q,
J¼6.4 Hz, 2H), 3.92 (t, J¼6.8 Hz, 2H), 4.05 (t, J¼5.6 Hz,
2H), 4.84 (s, 2H, 2NH), 5.06 (s, 2H), 5.07 (s, 2H), 6.90–
7.04 (m, 8H), 7.15–7.16 (m, 2H), 7.28–7.49 (m, 9H), 7.56–
7.59 (m, 2 H), 7.67–7.81 (m, 4H); 13C NMR-APT
(75 MHz, CDCl3) d ppm: 25.1 (CH2), 26.3 (CH2), 28.1
(CH2), 29.7 (CH2), 35.3 (CH2), 37.7 (CH2), 40.9 (CH2),
66.6 (CH2), 70.5 (CH2), 77.0 (CH2), 114.9 (CH), 115.9
(2CH), 118.2 (2CH), 119.1 (CH), 120.5 (2CH), 122.4 (2C),
122.5 (2CH), 123.2 (2CH), 123.4 (CH), 127.3 (4CH),
127.4 (CH), 127.9 (CH), 128.4 (2CH), 128.5 (2CH), 131.9
Phenol (3) (0.223 g, 0.489 mmol) and N-(3-bromopropyl)-
phthalimide (0.131 g, 0.489 mmol) were dissolved in DMF
(10 mL) with K2CO3 (0.27 g, 0.19 mmol) and the reaction
mixture heated under nitrogen at 65 ꢀC for 4 h. After allow-
ing to cool to room temperature, the mixture was poured into
iced 2 M HCl (60 mL) with stirring. The beige flocculant was
filtered off and the residue washed with water. This pre-
cipitate was then dissolved in 40:60 EtOAc/THF, dried
(MgSO4), filtered and the solvent removed under reduced
pressure. The crude residue was re-crystallised from EtOAc
to give 4a as a beige coloured powder (0.232 g, 74%). Mp
1
186–188 ꢀC; H NMR (300 MHz, acetone-d6) d ppm: 2.28
(p, J¼6.2 Hz, 2H), 3.94 (t, J¼6.5 Hz, 2H), 4.35 (t, J¼
5.9 Hz, 2H), 5.16 (s, 2H), 6.99–7.15 (m, 9H), 7.34–7.57(m,
5H), 7.75–7.91 (m, 5H+NH), 8.12–8.16 (m, 1H); 13C
NMR-APT (75 MHz, acetone-d6) d ppm: 27.9 (CH2), 35.2
(CH2), 66.7 (CH2), 70.5 (CH2), 114.1 (CH), 115.1 (CH),
115.8 (2CH), 116.2 (2CH), 120.1 (2CH), 121.9 (2CH),
123.2 (2CH), 126.9 (CH), 127.8 (CH), 127.3 (2CH), 128.4
(2CH), 131.7 (C), 132.6 (C), 134.1 (2CH), 135.4 (C), 136.7
(C), 138.6 (2C), 150.5 (C), 154.7 (C), 154.8 (C), 162.6
(2C), 164.5 (C), 168.4 (C); ATR (neat) cmꢂ1: 3274 (aromatic
C–H), 2940, 2875 (aliphatic C–H), 1705 (C]O) 1510, 1208

7290
J. B. Baell et al. / Tetrahedron 62 (2006) 7284–7292
(C), 132.0 (2C), 133.1 (C), 134.0 (2CH), 136.5 (C), 136.8
(C), 150.3 (C), 153.6 (C), 155.0 (C), 155.4 (C), 166.8 (C),
168.4 (2C), 171.4 (C); ATR (neat) cmꢂ1: 3270, 3058 (amide
N–H), 2936, 2869 (aliphatic C–H), 1711, 1502 (C]O);
HRMS calcd for C51H48N4O9 (M+Na+) 883.3319, found
883.3306.
25.0 (CH2), 26.2 (CH2), 28.4 (CH2), 29.6 (CH2), 37.9
(CH2), 39.4 (CH2), 40.8 (CH2), 66.5 (CH2), 67.0 (CH2),
67.3 (CH2), 38.2 (CH2), 70.5 (CH2), 114.1 (CH), 115.9
(2CH), 117.2 (CH), 118.2 (2CH), 120.5 (2CH), 122.8 (2CH),
123.3 (CH), 127.5 (4CH), 127.9 (2CH), 127.9 (2CH), 128.0
(2CH), 128.1 (2CH), 128.3 (2CH), 128.4 (2CH), 128.5 (CH),
128.6 (CH), 128.7 (CH), 128.8 (CH), 132.0 (C), 132.7 (C),
134.3 (C), 136.5 (C), 136.7 (C), 136.9 (2C), 150.3 (C),
153.7 (C), 153.9 (C), 155.1 (C), 155.6 (C), 155.8 (C), 156.4
(C), 163.4 (C), 167.1 (C), 171.6 (C); ATR (neat) cmꢂ1: 3228
(N–H), 3078, 3036 (aromatic C–H), 3743, 3871 (aliphatic
C–H), 1729, 1679 (C]O), 1500 (C]N); HRMS calcd for
C60H60N6O11 (M+Na+) 1063.4218, found 1063.4220.
5.7. (5-{4-(3-Aminopropoxy)-[4-(4-benzyloxyphenoxy)-
phenylcarbonoyl]-phenylcarbamoyl}-pentyl)-carbamic
acid benzyl ester (5b)
The phthalimide protected precursor (5a) (85 mg, 99 mmol)
was dissolved in a solution of 6:1 i-PrOH/H2O (10 mL),
and DCM was added until a homogeneous mixture was ob-
tained. To this was added NaBH4 (0.019 g, 0.5 mmol) was
added and the reaction mixture stirred at room temperature
for 18 h. Acetic acid was added to adjust the pH to 2 and
the reaction heated at 60 ꢀC for 48 h. The solvent was then
evaporated from the reaction mixture. Saturated NaHCO3
(20 mL) was added and the resulting mixture was extracted
with EtOAc (3ꢁ20 mL). The combined EtOAc extracts
were washed with water (20 mL), saturated brine
(2ꢁ20 mL), dried (MgSO4) and filtered. After the removal
of solvent from the filtrate, the residue was purified by radial
chromatography using a 2:1 CHCl3/MeOH eluent, or by first
flushing through an 80:20 EtOAc/hexanes eluent, then elut-
5.9. 5-Fluoro-2-nitrobenzoic acid (7)
2-Fluorobenzoic acid (1.807 g, 12.9 mmol) was dissolved in
concentrated H2SO4 (100 mL) after which P2O5 (w8 g) was
added. The mixturewas then cooled to 0 ꢀC and concentrated
HNO3 (15 mL) was added. After stirring for 3 h at 0 ꢀC, the
reaction mixturewas poured into iced water (500 mL) and fil-
tered. The precipitatewas dissolved in DCM (30 mL) and the
filtrate was extracted with DCM (3ꢁ30 mL). The combined
DCM extracts were washed with water (2ꢁ30 mL) and satu-
rated brine (2ꢁ40 mL). After drying (MgSO4) and filtering,
DCM was removed under reduced pressure to give 7 as a
white solid (2.280 g, 95%). Mp 136–138 ꢀC (lit.19 138–
139 ꢀC); 1H NMR (300 MHz, CDCl3) d ppm: 7.37 (m, 1H),
7.53 (dd, J¼2.7, 7.8 Hz, 1H), 8.00 (dd, J¼4.5, 9.0 Hz, 1H);
1H (19F-decoupled) NMR (300 MHz, CDCl3) d ppm: 7.37
(dd, J¼1.8, 8.9 Hz, 1H), 7.52 (d, J¼2.7 Hz, 1H), 8.00 (d,
J¼9.0 Hz, 1H); 13C NMR (75 MHz, CDCl3) d ppm: 117.5
(d, J¼25.4 Hz), 119.2 (d, J¼23.1 Hz), 126.7 (d, J¼9.4 Hz),
129.0(d, J¼8.6 Hz), 144.3(s), 164(d, J¼256.7 Hz), 168.4(s).
1
ing the product 5b (40 mg, 55%) with MeOH. H NMR
(300 MHz, methanol-d4) d ppm: 1.39 (p, J¼6.6 Hz, 2H),
1.50 (p, J¼6.6 Hz, 2H), 1.69 (p, J¼6.6 Hz, 2H), 2.19 (p,
J¼6.6 Hz, 2H), 2.39 (t, J¼7.2 Hz, 2H), 3.07 (t, J¼6.9 Hz,
2H), 3.18 (t, J¼7.2 Hz, 2H), 4.21 (t, J¼5.9 Hz, 2H), 5.06
(s, 2H), 5.09 (s, 2H), 6.94–7.05 (m, 6H), 7.13–7.6 (m,
15H); ATR (neat) cmꢂ1: 3463, 3367 (N–H), 3272, 3049
(aromatic C–H), 2946, 2877 (aliphatic C–H), 1710 (C]O);
HRMS calcd for C43H46N4O7 (M+H+) 731.3445, found
731.3444.
5.10. N-[4-(4-Benzyloxyphenoxy)-phenyl]-5-fluoro-2-
nitrobenzamide (8)
5.8. (5-{4-(3-[ N,N0-Bis(benzyloxycarbonyl)-guanidino-
propoxy])-[4-(4-benzyloxyphenoxy)-phenylcarbonoyl]-
phenylcarbamoyl}-pentyl)-carbamic acid benzyl ester
(5c)
4-(4-Benzyloxyphenoxy)-phenylamine
(6)
(0.889 g,
3 mmol) and 5-fluoro-2-nitrobenzoic acid (7) (0.558 g,
3 mmol) were dissolved in DMF (15 mL). Triethylamine
(1.2 mL, 12 mmol) was added followed by HBTU (1.14 g,
3 mmol). The reaction mixture was stirred for 5 h at room
temperature then poured into 2 M HCl (40 mL). The acidic
mixture was then extracted with EtOAc (3ꢁ30 mL). The
combined organic extracts were washed with water
(2ꢁ25 mL), saturated brine (2ꢁ20 mL), dried (MgSO4)
and filtered. The EtOAc solution was then concentrated to
one-third volume and cooled in ice to afford the coupled
product as a beige coloured precipitate which was filtered
Crude amine (5b) (0.106 g, 0.12 mmol) was dissolved in
MeOH (4 mL) with DCM (2 mL) added to aid dissolution.
Triethylamine was then added (0.05 mL, 0.51 mmol),
followed by N,N0-bis(benzyloxycarbonyl)-1H-pyrazole-1-
caboxamidine (38.3 mg, 0.10 mmol). The reaction mixture
was stirred at room temperature for 8 h during which a white
precipitate formed. The precipitatewas filtered off and the fil-
trate concentrated to afford more precipitate. The combined
precipitate was dissolved in DCM (15 mL) and washed with
2 M HCl (3ꢁ10 mL) followed by water (20 mL) and satu-
rated brine (2ꢁ10 mL). The product was purified by radial
chromatography using a 1:4 EtOAc/DCM solvent system
to give 5c as a white amorphous solid (56 mg, 44%) which
solidified only after an extended period under vacuum
1
off (0.940 g, 68%). Mp 177–179 ꢀC; H (19F-decoupled)
NMR (300 MHz, CDCl3) d ppm: 5.07 (s, 2H), 6.98 (s, 4H),
7.00 (s, 1H), 7.28–7.53 (m, 10H), 8.22 (d, J¼9.0 Hz, 1H);
13C NMR (75 MHz, DMSO-d6) d ppm: 70.52 (CH2), 116.8
(2CH), 117.5 (d, J¼25.7 Hz, CH), 118.5 (d, J¼23.3 Hz,
CH), 119.0 (2CH), 120.8 (2CH), 122.2 (2CH), 128.4 (CH),
128.5 (2CH), 128.6 (CH), 129.2 (2CH), 134.5 (C), 136.4
(C), 137.9 (C), 143.5 (d, J¼3.1 Hz, C), 151.0 (C), 154.6 (C),
155.2 (C), 163.2 (C), 164.9 (d, J¼253.6 Hz, C); ATR (neat)
cmꢂ1: 3250 m, 3070 m, 2940 w, 2870 w, 1650 s, 1550 s,
1490 s, 1340 s, 1210 s, 1010 m, 820 m, 740 w; MS (ESI):
m/z 459.2 [M+H]+. HRMS: Found 481.1177 (requires
481.1176 for [C26H19FN2O5]Na+). Microanalysis: Found
1
(25 ꢀC, 0.1 mmHg). Mp 114–116 ꢀC; H NMR (300 MHz,
CDCl3) d ppm: 1.36 (p, J¼8.4 Hz, 2H), 1.51 (p, J¼7.5 Hz,
2H), 1.70 (p, J¼7.5 Hz, 2H), 2.05 (p, J¼6.3 Hz, 2H), 2.35
(p, J¼7.5 Hz, 2H), 3.16 (q, J¼6.4 Hz, 2H), 3.66 (q, J¼
6.0 Hz, 2H), 4.06 (t, J¼5.6 Hz, 2H), 5.05 (s, 2H), 5.06 (s,
2H), 5.06 (s, 2H), 5.09 (s, 2H), 6.90–6.93 (m, 6H), 7.20–
7.45 (m, 25H); 13C NMR-APT (75 MHz, CDCl3) d ppm:

J. B. Baell et al. / Tetrahedron 62 (2006) 7284–7292
7291
(%) C 68.07 H 4.12 N 6.08 (requires (%) C 68.12 H 4.18 N
6.11 for C26H19FN2O5).
reduced pressure and the residue triturated with hexanes
(6ꢁ5 mL) followed by ether (4ꢁ5 mL) to remove any resi-
dual thioanisole. The crude compound was dissolved in
DCM and purified by flushing through a silica plug with a
50:50 EtOAc/hexanes eluant. The most polar band was
eluted with MeOH to give 11 (5.6 mg, 58%) as a brown
5.11. N-[4-(4-Benzyloxyphenoxy)-phenyl]-2-nitro-5-
(2-pyridin-2-yl-ethoxy)-benzamide (9)
1
This was obtained only in low yields from the reaction of 2-
(2-hydroxyethyl)pyridine with 8, as described earlier for the
synthesis of 3. Mp 99–102 ꢀC; 1H NMR (300 MHz, CDCl3)
d ppm: 3.31 (t, J¼6. 5 Hz, 2H), 4.52 (t, J¼6. 5 Hz, 2H), 5.05
(s, 2H), 6.95–7.05 (m, 9H), 7.19–7.52 (m, 8H), 7.67 (dt, J¼
1.8, 7.8 Hz, 1H), 8.14 (d, J¼8.7 Hz, 1H), 8.54 (dd, J¼0.9,
5.7 Hz, 1H); 13C NMR-APT (75 MHz, CDCl3) d ppm: 37.4
(CH2), 68.0 (CH2), 70.5 (CH2), 114.2 (CH), 115.3 (CH),
115.8 (CH), 118.2 (CH), 120.2 (CH), 121.9 (CH), 122.2
(CH), 123.8 (CH), 127.0 (CH), 127.4 (CH), 127.9 (CH),
128.4 (CH), 132.1 (C), 135.1 (C), 136.6 (CH), 136.8 (C),
138.3 (C), 149.1 (CH), 150.4 (C), 154.8 (C), 155.1 (C),
157.1 (C), 162.8 (C), 164.4 (C); ATR (neat) cmꢂ1: 3283,
3065 (aromatic C–H), 2997, 2941 (aliphatic C–H), 1620
(C]O), 1582 (C]N); MS (ESI) (M+H+) 562.5; HRMS
calcd for C33H27N3O6 (M+Na+) 548.17975, found 584.1793.
solid. Mp 99–102 ꢀC; H NMR (300 MHz, methanol-d4)
d ppm: 1.51 (p, J¼6.3 Hz, 2H), 1.78 (m, 4H), 2.09 (q,
J¼6.3 Hz, 2H), 2.47 (t, J¼7.4 Hz, 2H), 2.97 (t, J¼7.4 Hz,
2H), 3.44 (t, J¼6.8 Hz, 2H), 4.11 (t, J¼5.7 Hz, 2H), 7.09
(d, J¼9.0 Hz, 1H), 7.64 (s, 1H), 8.43 (d, J¼9.0 Hz, 1H);
13C NMR (75 MHz, methanol-d4; 2 ArC remain unassigned)
d ppm: 26.9 (CH2), 27.69 (CH2), 29.09 (CH2), 30.59 (CH2),
39.59 (CH2), 40.59 (CH2), 41.39 (CH2), 67.29 (CH2), 118.3
(CH), 119.7 (C), 120.3 (CH), 123.6 (CH), 131.8 (C), 156.2
(C), 159.7 (C); ATR (neat) cmꢂ1: 3440 (O–H), 3166, 3063
(aromatic C–H), 2952, 2923, 2868 (aliphatic C–H), 1660,
1589 (C]O), 1449 (C]N); HRMS calcd for C17H27N5O4
(M+H+) 366.21423, found 366.2153.
6. Biological methods
5.12. (5-{4-(3-[N,N0-Bis(tert-butoxycarbonyl)-guanidino-
propoxy])-[4-(4-benzyloxyphenoxy)-phenylcarbonoyl]-
phenylcarbamoyl}-pentyl)-carbamic acid benzyl ester
(10)
Peptide synthesis, radiolabelling of the peptides with 125I and
rat brain preparation were conducted following previously
described procedures.20–22 Radioligand binding assays
were run in triplicate in 96-well plates at room temperature.
Each well of the 96-well plate (Polystyrene, Round bottom,
NuncÔ, Denmark) contained compound 2a or 2b (first dilu-
tion 0.6 mM of compound total of seven dilutions, 1:10),
5–10 fmol of radiolabelled peptide (125I-GVIA) and 8 mg of
crude rat membrane (added last). All dilutions were made
in assay buffer (20 mM HEPES, 75 mM NaCl, 0.2 mM
EDTA, 0.2 mM EGTA, 2 mM Leupeptin, 2 mL apoprotinin
(to 30 mL assay buffer), 0.1% BSA, pH 7.4) and the final vol-
ume in each well was 150 mL. The plate was left on a shaker
for 1 h at room temperature before being filtered. Incubation
was terminated by washing the plate with wash buffer
(20 mM HEPES, 125 mM NaCl, pH 7.4) and filtered under
vacuum (Tomtec). The glass fibre filter used (90ꢁ120 mm,
double thickness, Wallac, Finland) was soaked in 0.6% poly-
ethyleneimine immediately prior to filtering to reduce non-
specific binding. The filter was put in a Sample Bag (Wallac,
Finland) containing 8 mL BetaPlate Scint (Wallac, Finland)
and the radioactivity bound to the filter was counted using
a 1450 MicroBeta Wallac Jet (Wallac, Finland). The data
were analysed using GraphPad Prism 2.0 (GraphPad Soft-
ware, Inc, San Diego, USA).
The protected anthranilamide derivative (5b) (55 mg,
0.064 mmol) was stirred with 1-H-pyrazole-1-(N,N0-bis-
(tert-butoxycarbonyl)]-caboxamidine (20 mg, 0.064 mmol)
in DCM (5 mL) overnight. The DCM was then removed
under reduced pressure. Excess guanidylating agent was
removed by triturating with hexanes (3ꢁ3 mL). After purifi-
cation of the residue by radial chromatography using a 50:50
EtOAc/hexanes solvent system, 10 was obtained as a white
1
amorphous solid (24 mg, 39%). Mp 56–59 ꢀC; H NMR
(300 MHz, CDCl3) d ppm: 1.43, 1.45 (s, 18H), 1.34 (p,
J¼7.2 Hz, 2H), 1.49 (p, J¼7.2 Hz, 2H), 1.65 (p, J¼7.2 Hz,
2H), 2.02 (p, J¼7.2 Hz, 2H), 2.31 (t, J¼7.5 Hz, 2H), 3.14
(t, J¼6.4 Hz, 2H), 3.58 (t, J¼6.0 Hz, 2H), 4.03 (t, J¼
5.4 Hz, 2H), 5.29 (s, 2H), 5.31 (s, 2H), 6.94–6.97 (m, 6H),
7.10–7.54 (m, 15H); 13C NMR (75 MHz, CDCl3) d ppm:
25.1 (CH2), 26.3 (CH2), 28.1 (3CH3), 28.3 (3CH3), 28.7
(CH2), 29.7 (CH2), 37.9 (CH2), 39.1 (CH2), 40.9 (CH2),
66.5 (CH2), 67.0 (CH2), 70.5 (CH2), 79.4 (C), 83.1 (C), 113.7
(CH), 115.8 (2CH), 117.3 (CH), 118.2 (2CH), 120.4 (2CH),
122.6 (2CH), 122.9 (CH), 123.2 (CH), 127.4 (2CH), 127.7
(2CH), 128.4 (2CH), 128.5 (3CH), 132.1 (C), 132.3 (C),
136.5 (C), 136.8 (C), 150.3 (C), 152.9 (C), 153.9 (2C),
154.9 (C), 155.3 (C), 156.0 (C), 156.2 (C), 163.3 (C), 167.0
(C), 171.4 (C); ATR (neat) cmꢂ1: 3314 (N–H), 2925, 2856
(aliphatic C–H), 1718, 1607 (C]O), 1494 (C]N); HRMS
calcd for C54H64N6O11 (M+H+) 973.4711, found 973.4713.
Acknowledgements
Financial support of this work was provided by the Bio-
molecular Research Institute, the National Health and Medi-
cal Research Council of Australia and Monash University.
Y.P.L. is a recipient of a Monash Graduate Scholarship and
an International Postgraduate Research Scholarship.
5.13. 2-(6-Aminohexanoylamino)-5-(3-guanidino-
propoxy)-benzoic acid (11)
References and notes
To the bis-Boc protected compound (10) (25.6 mg, 26 mmol)
was added thioanisole (0.15 mL, 1.3 mmol) and TFA
(1 mL). The clear, colourless solution turned red upon addi-
tion of TFA. The resulting mixture was stirred at room tem-
perature for 8 h. The solvent was then removed under
1. Norton, R. S.; Baell, J. B.; Angus, J. A. Calcium Channel
Pharmacology; McDonough, S. I., Ed.; Kluwer/Academic/
Plenum: New York, NY, 2003; pp 143–181.

7292
J. B. Baell et al. / Tetrahedron 62 (2006) 7284–7292
2. Staats, P. S.; Yearwood, T.; Charapata, S. G.; Presley, R. W.;
Wallace, M. S.; Byas-Smith, M.; Fisher, R.; Bryce, D. A.;
Mangieri, E. A.; Luther, R. R.; Mayo, M.; McGuire, D.;
Ellis, D. J. Am. Med. Assoc. 2004, 291, 63–70.
11. Farmer, J. D.; Rudnicki, S. M.; Suggs, J. W. Tetrahedron Lett.
1988, 29, 5105–5108.
12. Anderson, J. S.; Brown, K. C. Synth. Commun. 1983, 13, 233–
236.
3. Smith, M. T.; Cabot, P. J.; Ross, F. B.; Robertson, A. D.; Lewis,
R. J. Pain 2002, 96, 119–127.
4. Taken from a Frontline Strategic Consulting Report. See: (a)
http://66.102.7.104/search?q¼cache:ooJ21LCiKkgJ:www.
frontlinesmc.com/SMR/PainManagement/painPR.pdf+Pain+
and+compounded+annual+growth+rate+(CAGR)+of+%2510+
13. Itoh, T.; Mase, T. Org. Lett. 2004, 6, 4587–4590.
14. Kiso, Y.; Ukawa, K.; Nakamura, S.; Ito, K.; Akita, T. Chem.
Pharm. Bull. 1980, 28, 673–676.
15. Baell, J. B.; Harvey, A. J.; Norton, R. S. J. Comput.-Aided
Mol. Des. 2002, 16, 245–262.
16. Harvey, A. J.; Gable, R. W.; Baell, J. B. Bioorg. Med. Chem.
Lett. 2005, 15, 3193–3196.
17. Lauri, G.; Bartlett, P. A. J. Comput.-Aided Mol. Des. 1994, 8,
51–66.
to+reach+%2429.8+in+2008&hl¼en;
(b)
http://www.
newsrx.com/newsletters/Pharma-Business-Week/2003-04-28/
0428200333327PB.html; accessed 5th May 2006.
5. McGivern, J. G. Drug Discov. Today 2006, 11, 245–253.
6. Baell, J. B.; Duggan, P. J.; Lok, Y. P. Aust. J. Chem. 2004, 57,
179–185.
18. Baell, J. B.; Lessene, G. PCT/AU2005/000968, 2005.
19. Kissener, W.; Kuprat, J.; Emde, H; Paetz, K. C. U.S. Patent
5,756,831, 1998.
7. Baell, J. B.; Duggan, P. G.; Forsyth, S. A.; Lewis, R. J.;
Lok, Y. P.; Schroeder, C. I. Bioorg. Med. Chem. 2004, 12,
4025–4037.
20. Kristipati, R.; Nadasdi, L.; Tarczy-Hornoch, K.; Lau, K.;
Miljanich, G. P.; Ramachandran, J.; Bell, J. R. Mol. Cell
Neurosci. 1994, 5, 219–228.
8. Cohen, C. J.; Kraus, R. L. Calcium Channel Pharmacology;
McDonough, S. I., Ed.; Kluwer/Academic/Plenum: New
York, NY, 2003; pp 73–93.
9. Baell, J. B.; Forsyth, S. A.; Gable, R. W.; Norton, R. S.;
Mulder, R. J. J. Comput.-Aided Mol. Des. 2001, 15, 1119–
1136.
21. Lewis, R. J.; Nielsen, K. J.; Craik, D. J.; Loughnan, M. L.;
Adams, D. A.; Sharpe, I. A.; Luchian, T.; Adams, D. J.;
Bond, T.; Thomas, L.; Jones, A.; Matheson, J. L.;
Drinkwater, R.; Andrews, P. R.; Alewood, P. F. J. Biol.
Chem. 2000, 275, 35335–35344.
22. Nadasdi, L.; Yamashiro, D.; Chung, D.; Tarczy-Hornoch, K.;
Adriaenssens, P.; Ramachandran, J. Biochemistry 1995, 34,
8076–8081.
10. Osby, J. O.; Martin, M. G.; Ganem, B. Tetrahedron Lett. 1984,
25, 2093–2096.