Ergoline derivatives as highly potent and selective antagonists at the somatostatin sst1 receptor

Article abstract of DOI:10.1016/j.bmcl.2007.12.030

Non-peptidic compounds containing the octahydro-indolo[4,3-fg]quinoline (ergoline) structural element have been optimized into derivatives with high affinity (pK_d r sst₁ > 9) and selectivity (>1000-fold for h sst₁ over h sst₂-h sst₅) for the somatostatin sst₁ receptor. In functional assays, these ergolines act as antagonists at human recombinant sst₁ receptors. Pharmacokinetic studies in rodents reveal good oral bioavailability and brain penetration for some of these compounds.

Full text of DOI:10.1016/j.bmcl.2007.12.030

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 979–982
Ergoline derivatives as highly potent and selective antagonists
at the somatostatin sst1 receptor
Thomas Troxler,* Albert Enz, Daniel Hoyer, Daniel Langenegger, Peter Neumann,
Paul Pfa¨ffli, Philippe Schoeffter and Konstanze Hurth*
Neuroscience Chemistry, Novartis Institutes for BioMedical Research, WSJ-088.3.06, CH-4002 Basel, Switzerland
Received 23 October 2007; revised 12 December 2007; accepted 12 December 2007
Available online 26 December 2007
Abstract—Non-peptidic compounds containing the octahydro-indolo[4,3-fg]quinoline (ergoline) structural element have been opti-
mized into derivatives with high affinity (pKd r sst1 > 9) and selectivity (>1000-fold for h sst1 over h sst2–h sst5) for the somatostatin
sst1 receptor. In functional assays, these ergolines act as antagonists at human recombinant sst1 receptors. Pharmacokinetic studies
in rodents reveal good oral bioavailability and brain penetration for some of these compounds.
Ó 2007 Elsevier Ltd. All rights reserved.
Somatostatin (somatotropin-release-inhibiting factor,
SRIF) is a widely distributed peptide hormone/neuro-
transmitter1 that occurs in two biologically active forms,
a tetradecapeptide SRIF14 and a 28-amino acid peptide
SRIF28. To date, five somatostatin receptor subtypes
(sst1 to sst5) have been cloned and characterized, all
belonging to the G-protein-coupled receptor superfam-
ily.2–4 In order to elucidate the function of the different
SRIF receptor subtypes and to evaluate the potential of
somatostatin receptor ligands as therapeutic agents,
there is a need for non-peptidic, metabolically stable,
potent and subtype selective SRIF receptor agonists
and antagonists.5 In two recent publications6,7 we have
described the identification of potent and selective
non-peptidic somatostatin sst1 receptor antagonists
based on the octahydrobenzo[g]quinoline (obeline) core
structure (e.g., 1). Herein, we present the optimization of
a second class of sst1 receptor antagonists which are
based on the octahydro-indolo[4,3-fg]quinoline (ergo-
line)8 structural moiety (e.g., 2). The main goal of this
effort was to identify structurally diverse sst1 antagonists
which retain the excellent potency, selectivity and PK
properties of the obelines, but are synthetically more
easily accessible. Ergolines are bioisosteres of obelines,9
and their core moiety is known to be readily available
from the natural products, lysergic acid or paspalic acid.
A set of compounds from our corporate compound col-
lection containing the ergoline substructure was
screened in a radioligand binding assay for the rat sst1
and sst2 receptors. Indeed, several of these ergolines
showed appreciable affinity for the sst1 receptor, the
most potent being the pyridin-2-yl-piperazine derivative
2 with a pKd of 7.85 for r sst1 and good selectivity over r
sst2 (pKd = 4.75).
O
O
O
H
H
H
H
HN
N
N
N
N
N
N
N
6
2
NO2
2
1
In this paper, we present the structure–activity relation-
ship that was explored for lead compound 2, focusing on
positions 2 and 6 of the ergoline ring system, as well as
the aryl piperazine moiety. Affinities to the sst1 and sst2
receptors were determined in a radioligand binding as-
say performed in rat cortex membranes using
[
125I]SRIF-14 in the presence of 120 mM NaCl (sst1)10
or [125I][Tyr3]octreotide (sst2).11
For derivatives with halogen substituents in position 2,
sst1 affinity was retained (3 and 5, Table 1) or even in-
creased in the case of bromine (4), while no major effect
on selectivity over sst2 was observed. Of four other vari-
ations in this position (6–9), only the thiomethyl moiety
of 7 led to an improvement in sst1 affinity. Due to con-
Keywords: Somatostatin; GPCR; Somatostatin sst1 receptor; Selective
sst1 receptor antagonists; Ergolines; Lysergic acid derivatives.
*
Corresponding authors. Tel.: +41 61 3246604; fax: +41 61 3246760
(T.T.); e-mail: thomas.troxler@novartis.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.12.030

980
T. Troxler et al. / Bioorg. Med. Chem. Lett. 18 (2008) 979–982
Table 1. Binding affinities of ergoline derivatives to rat sst1 and sst2
receptors (variations at position 2)
of the 4-nitrophenyl moiety which led to derivatives with
very high affinities in the obeline class, but much less so
in the ergoline series (25–27). Based on their impressive
rat receptor binding profile, as well as structural consid-
erations (maximally diverse aryl piperazine moiety), the
two 2-bromo-ergoline derivatives 19 and 30 were chosen
for further profiling.
O
H
HN
N
N
N
N
H
R
a
a
R
Compound
pKd r sst1
pKd r sst2
All ergoline derivatives described herein were prepared
starting from biotechnologically available, enantiome-
rically pure paspalic acid 34. The syntheses of somato-
statin ligands 19 and 30 are outlined in Scheme 1 as
representative examples. Hydrogenation of 34, fol-
lowed by esterification, regioselective bromination in
position 2 and saponification of the ester group affor-
ded acid 36. Amide formation with the two respective
aryl piperazines7 was affected using either carbonyl-
diimidazole or propyl-phosphonic anhydride to afford
the final products in overall yields of 44% and 43%,
respectively.
–H
–Cl
2
3
4
5
6
7
8
9
7.85 0.08
7.89 0.12
8.35 0.18
7.78 0.08
6.62 0.05
8.62 0.28
5.58 0.07
6.51 0.06
4.75 0.16
n.d.
–Br
–I
4.78 0.06
5.25 0.02
5.07 0.06
5.07 0.21
5.03 0.07
5.11 0.07
–OH
–SMe
–SOMe
–SO2Me
a Mean SEM. Number of experiments: n = 3–6.
cerns with the metabolic stability of 7, the 2-bromo-
substituted ergoline core of 4 was chosen for further
optimization.
The enantiomers of 19 and 30 were prepared starting
from the methyl ester of natural (ꢀ)-lysergic acid 37
(Scheme 2). Racemization of 37 to acid rac-38 was
achieved by heating with hydrazine,12,13 followed by
hydrolysis of the intermediate racemic hydrazide.
rac-38 was converted to rac-19 and rac-30, respectively,
in a sequence analogous to the one described for 19
and 30. Separation by HPLC on chiral stationary phase
afforded the desired (+)-enantiomers ent-19 and
ent-30, respectively, along with the corresponding
(ꢀ)-enantiomers.
Removal of the N-methyl substituent of the 2-bromo-
ergoline core (10), as well as its replacement with longer
or functionalized alkyl groups (11–14), led to a dramatic
loss of sst1 affinity (Table 2). Also, an ethyl group did
not confer an improvement regarding potency and selec-
tivity, as demonstrated with the pair 15/16. The methyl
group as it is found in ergot natural products was there-
fore retained in this position for all further derivatives.
Investigation of the structure–sst1 affinity relationship
for the aryl piperazine moiety revealed that independent
of the nature of the substituent in position 2 (R = H, Cl
or Br), the 1-methyl-1H-pyridin-2-one (17–19),
benzo[1,2,5]oxadiazole (29 and 30) and [1,2,5]thiadiazolo-
[3,4-b]pyridine (31–33) moieties conferred the most
promising sst1 affinity and sst2 selectivity values (Table
3). These findings are well in line with the SAR data gen-
erated in the obeline series,6,7 with the notable exception
Compounds 19 and 30 were tested for their binding
affinities to the human recombinant somatostatin recep-
tors h sst1–h sst5.14 Both compounds displayed affinities
to the h sst1 receptor in the low nanomolar range
(pKd = 8.76 and 8.91, respectively) and excellent selec-
tivities (>1000-fold) over the other four somatostatin
receptors (Table 4). In extensive radioligand binding
studies, 19 and 30 proved to be selective over a range
of 40 different neurotransmitter receptors and ligand-
gated ion channels, with highest affinities found for the
h dopamine D2 receptor (pKd = 7.25 and 6.90, respec-
tively), the h dopamine D4 receptor (pKd = 6.91 and
Table 2. Binding affinities of ergoline derivatives to rat sst1 and sst2
receptors (variations at position 6)
O
H
7.11, respectively) and the
h
5HT1A receptor
HN
Br
N
(pKd = 7.30 and 7.55, respectively).15
N
N
Aryl
H
R'
Binding studies with the enantiomers of 19 and 30
showed that only derivatives with the natural (ꢀ)-con-
figuration show appreciable sst1 affinity: 19 is >1000-
fold more potent at native rat or recombinant human
sst1 receptors than its (+)-antipode ent-19 (Table 4).
Similarly, 30 shows >400-fold higher affinity than ent-
30 at rat and human sst1 receptors.
a
Compound pKd r sst1 pKd r sst2
a
R0
Aryl
–H
–Me
–nBu
2-Pyridyl
2-Pyridyl
2-Pyridyl
10
2
6.77 0.07 5.10 0.07
8.35 0.18 4.78 0.06
5.94 0.05 n.d.
11
2-Pyridyl
2-Pyridyl
2-Pyridyl
12
13
14
5.76 0.05 n.d.
5.83 0.04 n.d.
5.63 0.07 n.d.
In functional assays, sst1 ligands 19 and 30 act as antag-
onists at human recombinant somatostatin sst1 recep-
tors negatively coupled to cAMP accumulation in
CCL39 cells.16 They cause a concentration-dependent,
surmountable antagonism of SRIF-14 induced inhibi-
tion of forskolin-stimulated adenylate cyclase activity
with pKb values of 7.85 0.30 (n = 5) and 7.62 0.23
O
OH
–Me
–Et
3,4-Di-F-phenyl 15
3,4-Di-F-phenyl 16
9.31 0.06 5.09 0.05
9.01 0.05 4.88 0.09
a Mean SEM. Number of experiments: n = 3–6.

T. Troxler et al. / Bioorg. Med. Chem. Lett. 18 (2008) 979–982
981
Table 3. Binding affinities of ergoline derivatives to rat sst1 and sst2 receptors (variations of piperazine substituent)
O
H
HN
N
N
N
Aryl
H
R
Aryl
R = H
R = Cl
R = Br
a
a
a
a
a
a
Compound
pKd r sst1
pKd r sst2
Compound
pKd r sst1
pKd r sst2
Compound
pKd r sst1
pKd r sst2
N
2
7.85 0.08
9.12 0.08
4.75 0.16
5.11 0.06
3
7.89 0.12
9.41 0.07
9.42 0.03
8.38 0.01
8.90 0.04
9.22 0.06
9.41 0.05
9.62 0.08
n.d.
4
8.35 0.18
9.69 0.05
9.55 0.02
8.26 0.06
8.87 0.07
9.31 0.06
9.43 0.10
9.43 0.09
4.78 0.06
4.71 0.04
4.58 0.03
5.70 0.02
5.14 0.06
5.09 0.05
5.19 0.10
5.34 0.13
N
N
O
O
17
18
20
23
26
28
29
32
4.31 0.21
4.92 0.14
5.23 0.05
5.05 0.13
4.89 0.04
5.29 0.20
5.41 0.05
19
21
24
27
15
30
33
22
25
7.71 0.07
8.41 0.02
5.66 0.02
5.50 0.05
Cl
NO2
F
F
N
O
N
N
N
S
N
31
8.99 0.03
5.52 0.05
a Mean SEM. Number of experiments: n = 3–6.
O
O
O
H
H
H
H
H
a, b
c, d
e or f
HN
Br
HN
HN
OH
OH
O
19 or 30
N
N
N
H
34
35
36
Scheme 1. Synthesis of ergoline sst1 receptor antagonists 19 and 30. Reagents and yields: (a) H2, Pd/C (>90%); (b) MeOH/H2SO4 (95%); (c) NBS,
CH2Cl2 (77%); (d) NaOH, H2O/MeOH/THF (91%); (e) 1-methyl-6-piperazin-1-yl-1H-pyridin-2-one, carbonyl-diimidazole, DMF (74%); (f) 5-
piperazin-1-yl-benzo[1,2,5]oxadiazole, propyl-phosphonic anhydride, pyridine/DMF (72%).
O
O
g, h
i to n
o
HN
HN
O
OH
( )
ent-19 or ent-30
(and 19 or 30)
rac-19 or rac-30
N
N
H
H
37
rac-38
Scheme 2. Synthesis of unnatural (+)-enantiomers ent-19 and ent-30. Reagents and yields: (g) hydrazine, H2O, 20 min, 140° (50%); (h) KOH/H2O,
2 h, 130° (100%); (i) H2, Pd/C (79%); (j) MeOH/H2SO4, 2 h, 70° (75%); (k) tris-2-pyrrolidone-perbromide, THF, 15 h, rt (50%); (l) KOH, Dioxane,
15 h, rt (63%); (m) 1-methyl-6-piperazin-1-yl-1H-pyridin-2-one, propyl-phosphonic anhydride, pyridine/DMF, 15 h, rt (61%); (n) 5-piperazin-1-yl-
benzo[1,2,5]oxadiazole, propyl-phosphonic anhydride, pyridine/DMF, 15 h, rt (82%); (o) HPLC on chiral stationary phase.
(n = 12), respectively. Both compounds also act as
antagonists at human recombinant sst1 receptor driven
luciferase activity17; they inhibit SRIF-28 induced lucif-
erase activity with a pKb value of 8.85 0.19 (n = 3) for
19 and 9.42 0.15 (n = 3) for 30, and are devoid of
intrinsic activity.
Pharmacokinetic studies in rats showed that both com-
pounds are well absorbed after oral administration
(absolute bioavailability of 87% and 48% for 19 and
30, respectively). While concentrations of 19 in rat brain
after oral or intravenous administration were below the
limit of detection, most probably due to an efflux system

982
T. Troxler et al. / Bioorg. Med. Chem. Lett. 18 (2008) 979–982
Table 4. Compounds 19, ent-19, 30 and ent-30: physicochemical parameters and affinities for different somatostatin receptor subtypes
O
O
O
O
H
H
H
H
H
H
H
H
HN
Br
HN
Br
HN
Br
HN
Br
N
N
N
N
N
N
N
N
N
N
N
N
O
O
N
N
N
N
N
N
O
O
19
Mpb
30
ent-19
ent-30
20
½aꢁD
c
a
Compound
pKd
r sst1
r sst2
h sst1
h sst2
h sst3
h sst4
h sst5
19
275° (decomp.)
250° (decomp.)
266° (decomp.)
260° (decomp.)
ꢀ96.6°
+98.6°
ꢀ82.7°
+84.6°
9.69 0.05
6.38 0.02
9.43 0.10
6.74 0.03
4.71 0.04
4.69 0.02
5.19 0.10
4.90 0.04
8.76 0.01
5.71 0.02
8.91 0.04
6.30 0.01
4.91 0.05
5.03 0.03
5.17 0.05
5.41 0.01
5.53 0.01
5.27 0.04
5.63 0.04
5.14 0.05
5.31 0.05
5.09 0.01
5.44 0.06
5.25 0.03
4.25 0.10
4.16 0.09
5.19 0.01
4.63 0.02
ent-19
30
ent-30
a Mean SEM. Number of experiments: n = 3–6.
b Free base.
c Free base (DMF, c = 0.5).
7. Hurth, K.; Enz, A.; Floersheim, P.; Gentsch, C.; Hoyer,
D.; Langenegger, D.; Neumann, P.; Pfa¨ffli, P.; Sorg, D.;
Swoboda, R.; Vassout, A.; Troxler, T. Bioorg. Med. Chem.
Lett. 2007, 17, 3988.
8. (a) Pfaeffli, P.; Neumann, P.; Swoboda, R.; Stutz, P.
¨
PCT Int. Appl. WO 9854183, 1998; (b) Pfaeffli, P.;
at the blood–brain barrier, 30 readily entered the brain.
Oral administration of 30 lmol/kg 30 to rats led to
brain/plasma ratios ranging from 0.7 after 10 min to
1.7 after 24 h. Details as well as in vivo pharmacology
data will be published elsewhere in due course.
Neumann, P.; Swoboda, R.; Stutz, P. Chem. Abstr.
1998, 130, 25220.
9. Nordmann, R.; Petcher, T. J. J. Med. Chem. 1985, 28, 367.
10. Hoyer, D.; Perez, J.; Schoeffter, P.; Langenegger, D.;
¨
In conclusion, we have identified sst1 antagonists of a
novel ergoline-type structural class with sub-nanomolar
affinities to somatostatin sst1 receptors and >1000-fold
selectivity over other somatostatin receptor subtypes.
They behave as full antagonists in functional assays
and show promising PK properties in rodents.
Schupbach, E.; Kaupmann, K.; Lubbert, H.; Bruns, C.;
¨
¨
Reubi, J. C. Eur. J. Pharmacol. 1995, 289, 151.
11. Schoeffter, P.; Perez, J.; Langenegger, D.; Schupbach, E.;
¨
Lubbert, H.; Bruns, C.; Hoyer, D. Eur. J. Pharmacol.
¨
1995, 289, 163.
12. Stoll, A.; Hofmann, A. Hoppe-Seyler’s Z. Physiol. Chem.
1937, 250, 7.
References and notes
13. Stoll, A.; Hofmann, A. Helv. Chim. Acta 1943, 26, 922.
14. Siehler, S.; Seuwen, K.; Hoyer, D. Naunyn-Schmiedeberg’s
Arch. Pharmacol. 1999, 360, 488.
15. Hoyer, D.; Nunn, C.; Hannon, J.; Schoeffter, P.; Feuer-
bach, D.; Schuepbach, E.; Langenegger, D.; Bouhelal, R.;
Hurth, K.; Neumann, P.; Troxler, T.; Pfaeffli, P. Neurosci
Lett. 2004, 361, 132.
16. Nunn, C.; Schoeffter, P.; Langenegger, D.; Hoyer,
D. Naunyn-Schmiedeberg’s Arch. Pharmacol. 2003,
367, 1.
17. Nunn, C.; Rueping, M.; Langenegger, D.; Schuepbach, E.;
Kimmerlin, T.; Mich, P.; Hurth, K.; Seebach, D.; Hoyer,
D. Naunyn-Schmiedeberg’s Arch. Pharmacol. 2003, 367,
95.
1. Brazeau, P.; Vale, W.; Burgus, R.; Ling, N.; Butcher, M.;
Rivier, J.; Guillemin, R. Science 1973, 179, 77.
2. Patel, Y. C. Front. Neuroendocrinol. 1999, 20, 157.
3. Hoyer, D.; Bell, G. I.; Berelowitz, M.; Epelbaum, J.; Feniuk,
W.; Humphrey, P. P. A.; O’Carroll, A.-M.; Patel, Y. C.;
Schonbrunn, A. Trends Pharmacol. Sci. 1995, 16, 86.
4. Bell, G. I.; Reisine, T. Trends Neurosci. 1993, 16, 34.
5. Hannon, J. P.; Nunn, C.; Stolz, B.; Bruns, C.; Weckbec-
ker, G.; Lewis, I.; Troxler, T.; Hurth, K.; Hoyer, D. J.
Mol. Neurosci. 2002, 18, 15.
6. Troxler, T.; Hoyer, D.; Langenegger, D.; Neumann, P.;
Pfa¨ffli, P.; Schoeffter, P.; Sorg, D.; Swoboda, R.; Hurth,
K. Bioorg. Med. Chem. Lett. 2007, 17, 3983.