J. D. Speake et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1183–1186
1185
completely destroyed agonist activity at both the a1B
and a1D subtypes, while still retaining activity at the a1A
receptor. Also, as size increased from methyl to propyl,
there was a more rapid falloff in potency and efficacy at
a1A than that observed for analogous substitutions in
the triazole series. Methyl compound 9k reduced
potency by an order of magnitude, while retaining effi-
cacy. Ethyl substitution, 9l, further reduced potency
almost another order of magnitude, with some effect on
efficacy, and propyl substitution, 9m, continued the
trend. Unlike the triazole series, isopropyl substitution
was not tolerated, as demonstrated by the loss of all
agonist activity of compound 9n.
reports, functional activity did not correlate well with
affinity.15 For instance, 9j had a pIC50 of 7.6, 6.8, and
7.4 at a1A, a1B, and a1D, respectively, while 9k possessed
a pIC50 of 6.7, 6.4, and 6.3 at the three receptors, a
rather small difference considering the dramatic change
in functional activity.16 This incongruence has been
attributed to differences in intrinsic activity of the
ligands at the three receptors, and these types of ligands
have been referred to as efficacy driven or efficacy
dominant agonists.17
In conclusion, 20-heteroaryl-2-(anilinomethyl)imidazo-
lines have been identified that are potent, fully effica-
cious, and selective for the cloned human a1A adrenergic
receptor. The size and shape of the 20-heteroaryl group
were found to have a profound impact on functional
activity at all three a1 receptors. Compound 9j was less
than 10-fold selective for the a1A receptor, while com-
pound 9k shows that an appropriately placed methyl
group can confer >10,000-fold selectivity for activation
of the a1A receptor.
In the C3-linked pyrazoles, substitution at R2 with
methyl and ethyl, compounds 9p and 9q, respectively,
showed the same trend, but were slightly less dramatic
in their impact on potency and efficacy than the analo-
gous C5-linked pyrazoles 9k and 9l. Methyl substitution
was not enough to completely remove efficacy at the a1B
and a1D subtypes and only reduced potency at the a1A
subtype by ꢀ3-fold. Upon ethyl substitution, the ago-
nist activity at the a1B and a1D subtypes was removed
with, again, some reduction of potency and efficacy at
the a1A subtype.
References and Notes
1. (a) Schwinn, D. A.; Dwatra, M. M. Adv. Pharmacol. 1998,
42, 390. (b) Michel, M. C.; Taguchi, K.; Schafers, R. S.; Wil-
liams, T. J.; Clarke, D. E.; Ford, A. P. D. W. Adv. Pharmacol.
1998, 42, 394. (c) Ruffolo, R. R., Jr.; Hieble, J. P. Eur. Urol.
1999, 36, 17.
2. (a) The pharmacologically-defined native a1-adrenoceptors
are identified as a1A, a1B and a1D. The corresponding subtypes
characterized by molecular cloning techniques are designated
as a1a, a1b and a1d. For a brief review of a1-adrenoceptor
molecular pharmacology and a recent discussion of adreno-
ceptor classification, see: Zhong, H.; Minneman, K. P. Eur. J.
Pharmacol. 1999, 375, 261. (b) Guarino, R. D.; Perez, D. M.;
Piascik, M. T. Cell. Signal. 1996, 8, 323. (c) Hieble, J. P.
Pharm. Acta Helv. 2000, 74, 163. (d) Alexander, S.; Peters, J.;
Mead, A. Trends Pharmacol. Sci. 1998 (Suppl.), 1.
Substitution of the C4-linked pyrazoles, compounds 9r–
9t, had an even more pronounced effect on agonism at
all three a1 receptor subtypes. Methyl substitution was
enough to almost completely remove efficacy at all of
the receptors, with only the a1A subtype maintaining
some weak agonism. Ethyl compound 9t had no mea-
surable agonist activity.
One exception to the observed trends occurs in the pyr-
role series, compounds 9u–9w. In this case ethyl sub-
stitution, 9v, is actually somewhat more potent at the
a1A subtype than methyl, 9u. However, at a1D the pre-
viously observed trends held, in that 9u still had some
efficacy, while 9v did not. Neither compound had any
efficacy at the a1B subtype. Additional substitution, 9w,
further reduced potency and efficacy at a1A as expected.
3. Hieble, J. P.; Ruffolo, R. R., Jr. Drugs Pharm. Sci. 1998,
89, 231.
4. (a) Taniguchi, N.; Hamada, K.; Ogasawara, T.; Ukai, Y.;
Yoshikuni, Y.; Kimura, K. Eur. J. Pharmacol. 1996, 318, 117.
(b) Taniguchi, N.; Ukai, Y.; Tanaka, T.; Yano, J.; Kimura,
K.; Moriyama, N.; Kawabe, K. Naunyn-Schmiedeberg’s Arch.
Pharmacol. 1997, 355, 412. (c) Alberts, P.; Bergstrom, P. A. C.;
Fredrickson, M. G. Eur. J. Pharmacol. 1999, 371, 31. (d)
There have been several literature reports that a fourth a1-
adrenoceptor subtype, the a1L-subtype, is involved in urethral
smooth muscle contraction. The a1L-subtype has not been
cloned, but has been pharmacologically defined: Flavahan,
N. A.; Vanhouette, P. M. Trends Pharmacol. Sci. 1986, 7, 347.
(e) The a1L receptor may not be a distinct subtype, but a dif-
ferent affinity state of the a1A-subtype. For a discussion, see:
Ford, A. P. D. W.; Daniels, D. V.; Chang, D. J.; Gever, J. R.;
Jasper, J. R.; Lesnick, J. D.; Clarke, D. E. Br. J. Pharmacol.
1997, 121, 1127.
5. (a) Na, Y. J.; Guo, Y. L.; Gu, F. J. Med. 1998, 29, 289. (b)
Lee, E.; Lee, C. Br. J. Urol. 1997, 80, 606.
6. Brown, R. E. US Patent 3,754,002, 1971; Chem. Abstr. 79,
92219.
7. Human a1A (clone #137-12), a1B (clone #37-11), and a1D
(clone #16-7) a-adrenoceptors were expressed in Rat 1 fibro-
blast cells. Receptor activation was determined via calcium
mobilization through the Gq coupled PLC pathway using
calcium-sensitive fluorescent dyes (Calcium Green–Molecular
A possible rationale for the observed SAR arises from
consideration of the differing likely shapes of the mole-
cules, determined at least in part, by whether or not they
are able to form an intramolecular hydrogen bond to
the aniline N–H. Those heterocycles that can readily
make this H-bond, such as the unsubstituted triazole 9a,
and pyrazole 9j, would likely be relatively planar with
regard to the dihedral angle between the two aryl rings.
However, compounds such as pyrazole 9k, where the H-
bond acceptor has been blocked by methyl substitution,
may tend to be somewhat less planar. Further, com-
pounds that have progressively larger substituents,
either adjacent to the H-bond acceptor, 9p and 9q for
instance, or adjacent to the biaryl bond, 9b–g, for
example, may also have increasing difficulty adopting a
planar conformation. It appears that the a1A receptor
may be better able to accommodate this non-planarity
and still achieve the appropriate conformation for
receptor activation than either the a1B or a1D receptors.
It should be noted that, consistent with previous