Synthesis and nicotinic binding of novel phenyl derivatives of UB-165. Identifying factors associated with α7 selectivity

Article abstract of DOI:10.1016/S0960-894X(03)00594-8

Four racemic phenyl-substituted analogues 3-6 of the potent nicotinic agonist UB-165 1 have been synthesised and evaluated against the α ₄β₂, α₃β₄, and α₇ neuronal nicotinic receptors. The 2′-phenyl derivative 3 shows no activity at these major receptor subtypes, while the 4′-phenyl analogue 4 shows an enhanced level of α₇ selectivity as compared to UB-165 and deschloro UB-165 2. These results are discussed within the context of recent pharmacophore models.

Full text of DOI:10.1016/S0960-894X(03)00594-8

Bioorganic & Medicinal Chemistry Letters 13 (2003) 2825–2828
Synthesis and Nicotinic Binding of Novel Phenyl Derivatives of
UB-165. Identifying Factors Associated with ꢀ7Selectivity
Gunter Karig,a Jonathan M. Large,a Christopher G. V. Sharples,b
Andrew Sutherland,a Timothy Gallaghera,* and Susan Wonnacottb,*
aSchool of Chemistry, University of Bristol, Bristol BS8 1TS, UK
bDepartment of Biology and Biochemistry, University of Bath, Bath BA2 7AY, UK
Received 14 March 2003; revised 5 May 2003; accepted 14 June 2003
Abstract—Four racemic phenyl-substituted analogues 3–6 of the potent nicotinic agonist UB-165 1 have been synthesised and
evaluated against the a4b2, a3b4, and a7 neuronal nicotinic receptors. The 20-phenyl derivative 3 shows no activity at these major
receptor subtypes, while the 40-phenyl analogue 4 shows an enhanced level of a7 selectivity as compared to UB-165 and deschloro
UB-165 2. These results are discussed within the context of recent pharmacophore models.
# 2003 Elsevier Ltd. All rights reserved.
Neuronal nicotinic acetylcholine receptors (nAChRs)
are intimately involved in a diverse range of important
CNS functions and are also linked to a variety of clinically
significant disease states.1ꢀ6 A role for a4-containing
nAChRs in the control of pain has recently provided
further incentive for the characterisation of individual
receptors via subtype selective ligands.7 Such molecular
entities would facilitate elucidation of the relationship
between receptor subtypes and function and, in turn,
generate useful leads for drug discovery.8,9 UB-165 1, a
hybrid ligand based on combining structural elements
derived from anatoxin-a and epibatidine, is a potent
nicotinic agonist but shows (like anatoxin-a and epi-
batidine) limited discrimination between a4b2, a3b4 and
a7 receptor subtypes.10,11
achieve enhanced subtype selectivity. In this paper, we
disclose the synthesis of a series of racemic phenyl-sub-
stituted 30-pyridyl derivatives 3–6 of UB-165, and the
ability of these novel ligands to bind to a4b2, a3b4 and
a7 nAChR subtypes is defined. The results obtained
reinforce some recent suggestions relating to nicotinic
pharmacophores and point towards structural factors
that may contribute to selectivity for the a7 subtype.
Recently, UB-165 1 and the deschloro variant 2 have
been the subject of two independent structure–activity
relationship and pharmacophore studies, with the 20-
and 40-pyridyl analogues, together with a series of diazine
variants, having been prepared and evaluated.12,13 It is
clear that the 30-pyridyl moiety (as in 1 and 2), either alone
or embedded within a diazine unit, is critical for binding
and agonist activity. Further substitution of the 30-pyridyl
unit of 1 or 2 has not, however, been examined, but
additional interactions associated with this region of the
ligand would provide an important opportunity to
The 20-phenyl, 40-phenyl and 60-phenyl analogues 3, 4,
and 6, respectively, were prepared by means of a key
Negishi coupling reaction using the versatile enol triflate
*Corresponding author. Tel.:+44-117-9288260; fax:+44-117-
9298611; e-mail: t.gallagher@bristol.ac.uk
0960-894X/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00594-8

2826
G. Karig et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2825–2828
Scheme 1. Reagents and conditions: (a) (i) 2-bromopyridine, LDA, THF, ꢀ78 ꢁC, 30 min; (ii) ZnCl2, ꢀ78 ꢁC to rt; (iii) Pd(PPh3)4, 7, reflux;
(b) PhB(OH)2, Pd(PPh3)4, Na2CO3, PhMe, EtOH, reflux; (c) (i) concd HCl, 1,4-dioxane, water, reflux; (ii) Boc2O, Et3N, THF, water; (iii) 2 M HCl,
1,4-dioxane; (d) (i) 4-bromopyridine, LDA, THF, ꢀ78 ꢁC, 30 min; (ii) ZnCl2, ꢀ78 ꢁC to rt; (iii) Pd(PPh3)4, 7, reflux; (e) (i) 2-chloro-5-iodopyridine,
BuLi, THF, ꢀ78 ꢁC; (ii) ZnCl2, ꢀ78 ꢁC to rt; (iii) Pd(PPh3)4, 7, reflux; (f) PhB(OH)2, Pd(PPh3)4, DME, K2CO3, reflux.
7 as outlined in Scheme 1.10 The basic strategy followed
is illustrated for the preparation of the 20-phenyl ana-
logue 3. Halide-directed lithiation14 of 2-bromopyridine
followed by transmetallation with ZnCl2 and Negishi
cross coupling with enol triflate 7 gave the 20-bromo-
pyridyl derivative 8 in 93% yield.15 Reaction of 8 with
phenylboronic acid under Suzuki conditions gave the
20-phenylpyridyl derivative 9 in 81% yield, and removal
of the vinyloxycarbonyl (Voc) protecting group was
then carried out using a three step deprotection strat-
egy10,12 (via the Boc intermediate) to give the target
compound 3 as the hydrochloride salt, in 91% yield.
by incorporation of the phenyl moiety in the first step to
give 15 in 60% yield. Conversion of 15 to the corres-
ponding organozinc species followed by reaction with
enol triflate 7 under Negishi conditions gave the desired
adduct 16 in 66% yield. Deprotection of 16 (as descri-
bed above) gave the 50-phenylpyridyl analogue 5 in 56%
yield over three steps.
The four phenyl-substituted UB-165 analogues 3–6 (as
their hydrochloride salts) were evaluated for binding to
three major nAChR subtypes present in the CNS and
the peripheral nervous system. These comprised the rat
brain a4b2 and a7 subtypes which were labelled by
[3H]nicotine and [3H]MLA, respectively, and the rat
a3b4 nAChR expressed in a stably transfected cell line
and labelled by [3H]epibatidine.12,17 The results of these
binding studies along with those for UB-165 1 and the
deschloro variant 2 are shown in Table 1.
Application of the directed deprotonation-trans-
metallation strategy16 to 4-bromopyridine, Negishi
coupling with 7, subsequent Suzuki coupling of 10 and
deprotection of adduct 11 gave the 40-phenylpyridyl
analogue 4. The synthesis of 6 involved iodine-lithium
exchange of 2-chloro-5-iodopyridine, followed by
transmetallation with ZnCl2 and Negishi coupling to
give the 60-chloro-30-pyridyl adduct 12 in 59% yield. A
modified Suzuki coupling with 12 to phenylboronic acid
gave the N-Voc derivative 13 which was deprotected (as
above) to give the 60-phenylpyridyl analogue 6. This
chemistry all proceeds efficiently and the multistep
deprotection protocol (N-Voc to N-Boc to N-H) was
used because we have found it most convenient to purify
at the N-Boc stage. In our hands, N-Boc cleavage is
cleaner and more efficient than N-Voc cleavage.
It should also be appreciated that loss of the chlorine
has little effect on the affinity of UB-165 for these
receptors; UB-165 1 and the deschloro analogue 2
We had intended to prepare the 5-phenyl analogue 5 via
monometallation and Negishi coupling of 3,5-dibromo-
pyridine with enol triflate 7, then Suzuki coupling to
incorporate the phenyl moiety. However, bromine–
lithium exchange of 3,5-dibromopyridine followed by
transmetallation with ZnCl2 and coupling with the enol
triflate 7 gave only the desbromo coupled product 14 in
46% yield (Scheme 2). This problem was circumvented
Scheme 2. Reagents and conditions: (a) (i) BuLi, THF, ꢀ78ꢁC, 30 min;
(ii) ZnCl2, ꢀ78 ꢁC to rt; (iii) Pd(PPh3)4, 7, reflux; (b) PhB(OH)2,
Pd(PPh3)4, Na2CO3, PhMe, EtOH, reflux; (c) (i) concd HCl, 1,4-dioxane,
water, reflux; (ii) Boc2O, Et3N, THF, water; (iii) 2 M HCl, 1,4-dioxane.

G. Karig et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2825–2828
2827
Table 1. Binding affinities for UB-165 and analogues (note com-
pounds 1–6 are racemic)a
Compd
nAChR subtype
Ki (nM)
a4b2
a3b4
a7
UB-165 1
Deschloro UB-165 2
20-Ph 3
0.27
0.43
>50,000
234
4.8
6.5
22.5
>50,000
1990
468
2760
89
>50,000
927
15,980
1306
Figure 1. Schematic representation of the a4b2* nAChR pharmaco-
40-Ph 4
phore. The structure in the box is a view down the C(2)-C(30) bond.
50-Ph 5
60-Ph 6
21.4
56.3
arrangement might be expected to inhibit the interaction
of this key pharmacophore component with its asso-
ciated receptor binding site and this would explain the
lack of nicotinic activity associated with 3. Phenyl sub-
stitution at the 50- and 60-positions does not have a
major impact on subtype selectivity although as noted
above, some moderation of binding potency is asso-
ciated with substitution of these sites (see Table 1). The
40-phenyl derivative 4 is well able to attain the con-
formation associated with the Tønder pharmacophore
model and the substituent does not impact on the ability
of the N(9) ammonium center to interact with the
receptor. The retention of affinity at a7 (but not a4b2
and a3b4) that is observed in this case may indicate that
the 40-phenyl substituent of 4 occupies a region of space
relative to the other key components of the ligand
structure that could usefully be exploited for enhancing
selectivity for this nAChR subtype, and studies to test
this hypothesis are underway.24
aValues are the means from at least 3 independent assays.
(which is more closely related to ligands 3–6) have
similar profiles. Four significant observations can be
made regarding the data for the phenyl-substituted
ligands shown in Table 1. Firstly, incorporation of a
phenyl substituent at the 20-position (as in 3) results in a
complete loss of nicotinic potency against all three
receptors subtypes. Secondly, steric bulk associated with
the phenyl substituent does reduce potency compared
with 2, but clearly also makes an impact on selectivity.
Thirdly, phenyl substitution at the 50- and 60-positions
(5 and 6, respectively) results in profiles that are quali-
tatively similar to deschloro UB-165 2: potency at
a4b2>a3b4 > a7. However, it should be pointed out
that 6 in particular is less capable than deschloro
UB-165 2 of discriminating between these three receptor
subtypes.
The final and, in our view, most interesting observation
relates to the 40-phenyl analogue 4. This ligand, while
substantially (2 orders of magnitude) less potent than
UB-165 1 and deschloro UB-165 2 at a4b2 and a3b4
subtypes, retains a comparatively high potency at the a7
receptor. Analogue 4 is only 8.5 times more potent at
a4b2 than at a3b4, whereas deschloro UB-165 is 52 times
more potent, respectively. However, at a7 4 is only four
times less potent than at a4b2, whereas deschloro UB-
165 is 200 times less potent at a7. Interestingly, UB-165
1 shows a binding differential >104 times between these
two receptor subtypes.
In conclusion, we have synthesised the four possible
phenyl-substituted variants of UB-165. Results of bind-
ing studies are consistent with the recent suggestion that
the conformation represented in Figure 1 is important
for an effective nicotinic ligand and this is further sup-
ported by the observation that the 20-phenyl isomer 3
shows the most dramatic loss in potency. The 40-phenyl
derivative 4 shows enhanced a7 selectivity and this may
provide a means of refining and extending current mod-
els to derive a useful template specific for a7 ligands.
Acknowledgements
Nicotinic pharmacophores continue to attract sig-
nificant interest, with important contributions being
made from a number of groups.6,18ꢀ22 Given our earlier
studies,12 it is pertinent to consider these biological
results alongside the pharmacophore model recently
described by Tønder et al.,18,19 the key conformational
features of which are shown in schematic form in Figure
1.
The authors thank Dr. Chris Kruse and Dr. Axel Stoit
(Solvay Pharmaceuticals) for helpful discussions, and
the BBSRC (Biomolecular Sciences Project Grants 86/
B11785 and 7/MOLO4724) for financial support.
References and Notes
In this model, which was developed for the a4b2*
nAChR, the bioactive conformation of deschoro
UB-165 2 has the C(1)–C(2)–C(30)–C(20) in a cisoid
arrangement with the pyridyl nitrogen (N(10)) above the
C(1)–C(2)–C(3) plane; the C(1)–C(2)–C(30)–C(20) dihe-
dral angle is approximately 55ꢁ (cf. Fig. 1).23 In this
conformation, the 20-phenyl substitutent in 3 (a ligand
which completely lacks nicotinic activity) is located in
very close proximity to the N(9) ammonium center. This
1. Role, L. W.; Berg, D. K. Neuron. 1996, 16, 1077.
2. Lindstrom, J. Mol. Neurobiol. 1997, 15, 193.
3. Arneric, S. P., Brioni, J. D., Eds. Neuronal Nicotinic
Receptors. John Wiley and Sons: New York, 1999.
4. Jones, S.; Sudweeks, S.; Yakel, J. L. Trends Neurosci. 1999,
22, 555.
5. Paterson, D.; Nordberg, A. Prog. Neurobiol. 2000, 61, 75.
6. Curtis, L.; Chiodini, F.; Spang, J. E.; Bertrand, S.; Patt, J. T.;
Westera, G.; Bertrand, D. Eur. J. Pharmacol. 2000, 393, 155.

2828
G. Karig et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2825–2828
7. Bannon, A. W.; Decker, M. W.; Holladay, M. W.; Curzon,
P.; Donnelly-Roberts, D.; Puttfarcken, P. S.; Bitner, R. S.;
Diaz, A.; Dickenson, A. H.; Porsolt, R. D.; Williams, M.;
Arneric, S. P. Science 1998, 279, 77.
8. Holladay, M. W.; Dart, M. J.; Lynch, J. K. J. Med. Chem.
1997, 40, 4169.
18. Tønder, J. E.; Olesen, P. H.; Hansen, J. B.; Begtrup, M.;
Pettersson, I J. Comput.-Aided Mol. Des. 2001, 15, 247.
19. Tønder, J. E.; Olesen, P. H. Curr. Med. Chem. 2001, 8, 651.
20. Dukat, M.; Damaj, M. I.; Glassco, W.; Dumas, D.; May,
E. L.; Martin, B. R.; Glennon, R. A. Med. Chem. Res. 1993, 4,
131.
9. Lloyd, G. K.; Williams, M. J. Pharmacol. Exp. Ther. 2000,
292, 461.
21. Glennon, R. A.; Herndon, J. A.; Dukat, M. Med. Chem.
Res. 1994, 6, 465.
10. Wright, E.; Gallagher, T.; Sharples, C. G. V.; Wonnacott,
S. Bioorg. Med. Chem. Lett. 1997, 7, 2867.
11. Sharples, C. G. V.; Kaiser, S.; Soliakov, L.; Marks, M. J.;
Collins, A. C.; Washburn, M.; Wright, E.; Spencer, J. A.;
Gallagher, T.; Whiteaker, P.; Wonnacott, S. J. Neurosci. 2000,
20, 2783.
12. Sharples, C. G. V.; Karig, G.; Simpson, G. L.; Spencer,
J. A.; Wright, E.; Millar, N. S.; Wonnacott, S.; Gallagher, T.
J. Med. Chem. 2002, 45, 3235.
13. Gohlke, H.; Gundisch, D.; Schwarz, S.; Seitz, G.; Tilotta,
M. C.; Wegge, T. J. Med. Chem. 2002, 45, 1064.
14. Gribble, G. W.; Saulnier, M. G. Tetrahedron Lett. 1980,
21, 4137.
22. Meyer, M. D.; Decker, M. W.; Rueter, L. E.; Anderson,
D. J.; Dart, M. J.; Kim, K. H.; Sullivan, J. P.; Williams, M.
Eur. J. Pharmacol. 2000, 393, 171.
23. The conformations associated with ligands 3–6 are based
on the computational methods and results obtained earlier.12
24. The quinoline analogue of 2, which was prepared by
essentially the same strategy as shown in Scheme 1, shows
binding affinity values (nM) of 140, 2535 and >50,000 for
a4b2, a3b4, and a7, respectively. This would suggest that too
much rigidity associated with substituents in the 50 and
60-region blocks a7 recognition. Glennon25 has reported binding
data for 6-phenyl nicotine and the corresponding homo-
azanicotine analogue (cf. ligand 6), both of which show low
affinity for the a4b2 nAChR. The corresponding quinolines
show somewhat increased levels of potency.
15. Karig, G.; Spencer, J. A.; Gallagher, T. Org. Lett. 2001, 3,
835.
16. Karig, G.; Thasana, N.; Gallagher, T. Synlett. 2002, 808.
17. Lewis, T. M.; Harkness, P. C.; Sivilotti, L. G.; Colqu-
houn, D.; Millar, N. S. J. Physiol. 1997, 505, 299.
25. Ferretti, G.; Dukat, M.; Giannella, M.; Piergentili, A.;
Pigini, M.; Quaglia, W.; Damaj, M. I.; Martin, B. R.; Glen-
non, R. A. Bioorg. Med. Chem. Lett. 2003, 13, 733.