Parallel solution- and solid-phase synthesis of spiropyrrolo-pyrroles as novel neurokinin receptor ligands

Article abstract of DOI:10.1016/S0960-894X(02)00659-5

The combination of a 3,5-bis(trifluoromethyl)phenyl needle with the spiropyrrolo-pyrrole motive as a privileged GPCR scaffold was the basis for designing a focused combinatorial library targeted towards the neurokinin-1 receptor. A solution- and solid-phase method is described and binding affinities of representative compounds are presented.

Full text of DOI:10.1016/S0960-894X(02)00659-5

Bioorganic & Medicinal Chemistry Letters 12 (2002) 3073–3076
Parallel Solution- and Solid-Phase Synthesis of Spiropyrrolo-
Pyrroles as Novel Neurokinin Receptor Ligands
Konrad H. Bleicher,* Yves Wuthrich, Geo Adam, Torsten Hoffmann
and Andrew J. Sleight
F. Hoffmann-La Roche AG, Pharma Research, CH-4070 Basel, Switzerland
Received 17 May 2002; revised 24 July 2002; accepted 5 August 2002
Abstract—The combination of a 3,5-bis(trifluoromethyl)phenyl needle with the spiropyrrolo-pyrrole motive as a privileged GPCR
scaffold was the basis for designing a focused combinatorial library targeted towards the neurokinin-1 receptor. A solution- and
solid-phase method is described and bindingaffinities of representative compounds are presented.
# 2002 Elsevier Science Ltd. All rights reserved.
Currently three subtypes of neurokinin receptors are
described as being7-transmembrane G-protein coupled,
namely the NK-1, NK-2 and the NK-3 receptors. All
three belongto the peptide receptor subgroup of GPCR
family 1.1 Their endogenous ligands are the tachykinins,
a group of well known peptide amides that all share the
common C-terminal amino acid sequence Phe-X-Gly-
Leu-Met-NH2 where X is either Phe or Val. ‘Substance P’
(or SP) is an undecapeptide (where X=Phe) which shows
highest binding affinity for the NK-1 receptor. NKA and
NKB (X=Val) are both decapeptides that bind pre-
ferentially to the NK-2 and NK-3 receptor, respectively.2
the NK-1 receptor we started with the classical route of
high-throughput screening (HTS) as a first step for the
identification and validation of hit series.
As a complementary approach to the HTS campaign we
envisaged an NK-1 focused library design based on the
combination of two concepts—the ‘privileged structure’
and the ‘needle’ approach as already stated in an earlier
communication.4
Since spiropiperidines are considered as pharmaco-
logically relevant molecules for various targets within
the area of G-protein coupled receptors, this class of
chemical motives is often referred to as ‘privileged
GPCR-structures’.5,6
All three receptors have been described as beingof
pharmacological relevance in various areas. NK-1 and
NK-3 receptors are discussed as potential CNS disease
targets since both receptors are mainly expressed in the
brain. Antagonists of these receptors are currently
investigated as potential treatments for diseases such as
anxiety and depression (NK-1 antagonists) or chemo-
therapy-induced emesis (NK-3 antagonists). The NK-2
receptor is mainly expressed in the periphery and descri-
bed as a pharmacologically relevant target for respiratory
diseases such as asthma as well as urinary dysfunction.3
Complementary to the ‘privileged structure’ motive the
terminus ‘needle’ was described in the literature as the
fragment of an active molecule showing very specific
interactions with one particular biological target.7
Within the neurokinin family one such example is the
3,5-bis(trifluoromethyl)phenyl moiety which has been
described as an essential fragment for several NK-1
receptor ligands.3
Due to the fact that neurokinin receptors are G-protein
coupled receptors, biostructural information is very
limited. Thus for the initiation of a research program on
Spiropyrrolo-pyrroles were already described by us as
ligands for the orphanin FQ receptor (OFQ).8 This
protein was characterized as beinga member of the G-
protein coupled receptors. Similar to neurokinin recep-
tors the OFQ receptor belongs to the GPCR subfamily
1 where a 17 amino acid neuropeptide was identified as
the endogenous ligand.9 Since these OFQ ligands also
*Correspondingauthor. Tel.: +41-6168-70387; fax: +41-6168-86459;
e-mail: konrad.bleicher@roche.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00659-5

3074
K. H. Bleicher et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3073–3076
Table 1. NK-1 receptor affinities of representatives from library A
Scheme 1. Targeted library design for NK-1 receptor ligands by com-
bination of a GPCR ‘privileged structure’ motive with the 3,5-bis(tri-
fluoromethyl)phenyl ‘needle’ (Ph(CF3)2).
Compd
R
R0
pKi (hNK-1)
9a
Phe
7.52
contain the piperidine architecture we envisaged this
template as a ‘privileged structure’ for G-protein cou-
pled receptors.
9b
9c
Phe
Phe
7.51
7.12
Followingthe approach of combiningsuch promiscuous
motives with the ‘needle’ concept, a focused library
synthesis was initiated as depicted in Scheme 1. Three sets
of sublibraries were prepared in a parallel fashion using
either solution- or solid-phase methodologies.
A compound collection of type A was generated via
parallel solution phase synthesis. The core scaffold was
thereby obtained from a (3+2) cycloaddition reaction
startingfrom N-phenylmaleimide 4 as a dienophile and
the correspondingdiene generated in situ from 1-Boc-
piperidone and N-benzylglycine via base catalysis.10
Under the reaction conditions described in Scheme 2 a
spontaneous decarboxylation takes place leadingto the
9d
9e
Phe
Phe
6.52
5.97
spiropyrrolo-pyrrole intermediate 5. Subsequent deben-
zylation usingstandard Pd/C hydrogenation conditions
gave deprotected intermediate 6. The nucleophilicity of
the resulting a,a-disubstituted pyrrolidine nitrogen
turned out to be very low due to steric hindrance of the
neighboring spiropiperidine ring. Reactivity enhance-
ment was achieved by silylation of the pyrrolidine
nitrogen prior to the acylation step.
Therefore, intermediate
6 was treated with bis(-
trimethylsilyl)acetamide (BSA) at room temperature
and stirred for 1 h. Subsequently the reaction mixture
was cooled down to 0 ꢀC. The dropwise addition of 3,5-
bis(trifluoromethyl)benzoyl chloride and subsequent
warmingto room temperature agve the benzoylated
intermediate 7 in mediocre yield only. Kieselgel chro-
matography and final treatment with trifluoroacetic acid
resulted in the TFA salt of precursor 8 ready for further
modification. To generate a compound collection of
type A spiropyrrolo-pyrrole 8 was treated under DIC
couplingconditions with a set of carboxylic acids to
form the correspondingamides. Sulfonylchlorides were
coupled to give sulfonamides, isocyanates yielded the
correspondingureas and finally reductive alkylation reac-
tions were performed to introduce a basic tertiary nitro-
gen into the molecular framework. The compounds were
generated using standard in-house developed parallel
synthesis equipment. All products were worked up by a
liquid/liquid extraction step and finally purified using
preparative HPLC. For quality control and compound
Scheme 2. (a) 1 equiv Boc-piperidone, 1 equiv N-benzylglycine, 2.5
equiv DIPEA, toluene, reflux, 16 h; (b) Pd/C, H2, 2 atm, MeOH/
CH2Cl2 (4:1), rt, 16 h; (c) 1.2 equiv 3,5-bis(trifluoromethyl)benzoyl
chloride, 1.2 equiv BSA, CH2Cl2, 0 ꢀC–rt, 72 h; (d) TFA/DCM (1:1),
rt, 2 h; (e) 1.1 equiv RCOOH, 1.1 equiv DIC, CH2Cl2, rt, 16 h; (f) 1.1
equiv RSO2Cl, 1.1 equiv TEA, CH2Cl2, rt, 16 h; (g) 1.1 equiv RNCO,
CH2Cl2, rt, 16 h; (h) 1.1 equiv RCHO, 1.3 equiv NaBH3CN, DMF/
MeOH/CH3COOH (87:10:3), rt, 16 h.

K. H. Bleicher et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3073–3076
3075
Due to low reactivity of the pyrrolidine nitrogen
observed duringthe 3,5-bis(trifluoromethyl)benzoyl-
ation reaction we considered a solid-phase protocol for
this particular modification step to generate compounds
of type B and C. This would allow the use of a large
excess of reagents to drive the reaction to completion.
Unconsumed material can be washed away and product
isolation is obtained via simple filtration. For the ease of
parallel processingwe considered an acid labile polymer
support as most convenient. The allyloxycarbonyl-
(Alloc) group was considered as an orthogonal protect-
inggroup that can be cleaved under very mild condi-
tions. The solution phase synthesis of the corresponding
precursors 14 is shown in Scheme 3. Analogous to
Scheme 2, the spiropyrrolo-pyrrole is obtained via a
thermal (3+2) cycloaddition reaction startingfrom
either commercially available N-substituted maleimides
11 or from intermediates prepared via maleic acid
anhydride 10 and primary amines. Debenzylation of
intermediates 12 and subsequent acylation with allyl
chloroformate resulted in intermediates 13. Boc clea-
vage by TFA treatment gave scaffolds 14.
Table 2. NK-1 receptor affinities of representatives from library B
Compd
R0
R
pKi (hNK-1)
19a
6.58
19b
H
H
6.30
19c
19d
5.64
5.63
For the immobilization of precursors 14 the compounds
were added to trityl chloride resin resultingin solid
supported intermediate 15. Palladium catalyzed Alloc
cleavage gave resin 16. Compound subset B was gener-
ated via modification at the pyrrolidine nitrogen by
couplingof carboxylic acids, isocyanates, sulfonyl
chlorides and aldehydes resultingin resins 18. After
TFA cleavage the compounds were either purified or
further modified by treatingwith acetic anhydride or
mesyl chloride for cappingof the piperidine nitrogen.
Excess of reagent was removed by simple evaporation
19e
H
5.37
characterization online-coupled LCMS was applied to
check the purity of generated material.
Five representative compounds (9a–e) showingreason-
able activities for the NK-1 receptor are depicted in
Table 1.
Scheme 4. (a) 1 equiv 14, 3 equiv DIPEA, CH2Cl2, rt, 16 h; (b) 0.05
equiv Pd(PPh3)4, 10 equiv morpholine, CH2Cl2, rt, 16 h; (c) 3 equiv
RCOOH, 3 equiv DIC, DMF, rt, 16 h; (d) 3 equiv RNCO, CH2Cl2, rt,
16 h; (e) 3 equiv RSO2Cl, 3 equiv TEA, CH2Cl2, rt, 16 h; (f) 10 equiv
RCHO, 5 equiv NaBH3CN, DMF/MeOH/CH3COOH (87:10:3), rt,
16 h; (g) 3 equiv bromoacetic acid, 3 equiv DIC, DMF, rt, 16 h; (h) 3
equiv amine, DMF, rt, 4 h; (i) TFA, CH2Cl2 (1:1), rt, 2 h; (j) 2 equiv
Ac2O, 2 equiv TEA, CH2Cl2, rt, 16 h; (k) 2 equiv MeSO2Cl, 2 equiv
Scheme 3. (a) 1 equiv primary amine, CH2Cl2, rt, 1 h, evaporation,
1.25 equiv sodium acetate, acetic anhydride, 90 ꢀC, 16 h; (b) 1 equiv
Boc-piperidone, 1 equiv N-benzylglycine, 2.5 equiv DIPEA, toluene,
reflux, 16 h; (c) Pd/C, H2, 2 atm, MeOH/CH2Cl2 (4:1), rt, 2 h; (d) 1.2
equiv Alloc-chloride, 1.2 equiv BSA, CH2Cl2, 0 ꢀC–rt, 72 h; (e) TFA/
DCM (1:1), rt, 2 h.
TEA, CH2Cl2, rt, 16 h; for library C: (l) 2 equiv 3,5-bis(tri-
fluoromethyl)benzoyl chloride, 2 equiv DIPEA, CH2Cl2, rt, 16 h.

3076
K. H. Bleicher et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3073–3076
Table 3. NK-1 receptor affinities of representatives from library C
material. To guarantee solid biological data all com-
pounds were further purified with preparative HPLC
and characterized as already described.
Table 3 shows five representative compounds generated
for library C (19f–j) with reasonable to high NK-1
receptor affinities.
In summary, we described the parallel solution- and
solid-phase synthesis of a focused compound library
targeted towards the neurokinin-1 receptor. The basis
for the library design was a combination of the spiro-
pyrrolo-pyrrole template envisaged as a ‘privileged
structure’ for G-protein coupled receptors with the 3,5-
bis(trifluoromethyl)phenyl motive as an NK-1 specific
‘needle’. As a result novel neurokinin receptor ligands
with low nanomolar affinities were identified.
Compd
R
R0
pKi (hNK-1)
19f
Phe
8.29
19g
19h
19i
19j
Phe
Me
Me
Me
8.08
7.41
6.29
6.10
Acknowledgements
The authors would like to thank Dr. Mark Rogers-
Evans for critical readingof the manuscript and Alain
Rudler for biological testing.11
References and Notes
1. Beck-Sickinger, A. Drug Discov. Today 1996, 1, 502.
2. Leroy, V.; Mauser, P.; Gao, Z.; Peet, N. P. Exp. Opin.
Invest. Drugs 1999, 9, 735.
after quenchingthe reaction mixture with methanol. All
compounds were purified by preparative HPLC.
3. Seward, E. M.; Swain, C. Exp. Opin. Ther. Pat. 1999, 9,
571.
4. Bleicher, K. H.; Wuthrich, Y.; De Boni, M.; Kolczewski,
S.; Hoffmann, T.; Sleight, A. J. Bioorg. Med. Chem. Lett.
2002, 12, 2519.
5. Mason, J. S.; Morize, I.; Menard, P. R.; Cheney, D. L.;
Hulme, C.; Labaudiniere, R. F. J. Med. Chem. 1999, 42, 3251.
6. Patchett, A. A.; Nargund, R. P. Annu. Rep. Med. Chem.
2000, 35, 289.
7. Boehm, H.-J.; Boehringer, M.; Bur, D.; Gmuender, H.;
Huber, W.; Klaus, W.; Kostrewa, D.; Kuehne, H.; Luebbers,
T.; Meunier-Keller, N.; Mueller, F. J. Med. Chem. 2000, 43,
2664.
Table 2 shows five representative compounds of library
B (19a–e) with moderate NK-1 receptor activities.
Compounds generated for library C were designed to
have the NK-1 specific 3,5-bis(trifluoromethyl)phenyl
needle at the piperidine nitrogen. Since this was foreseen
as the attachment point to the resin as described in
Scheme 4, a solid supported synthesis approach is not
obvious. We anyway decided to proceed via a solid-
phase synthesis route due to low reactivity of the pyr-
rolidine nitrogen allowing the use of an excess of
reagents for this particular coupling step. Additionally a
two step modification procedure was envisaged by first
DIC couplingof resin 16 with bromoacetic acid result-
ingin resin 17 and subsequent nucleophilic displace-
ment of the bromide with a set of available secondary
amines. The compounds were cleaved from the trityl
resin under TFA treatment to give the free piperidine
8. Adam, G.; Dautzenberg, F.; Kolczewski, S.; Roever, S.;
Wichmann, J. Eur. Pat. Appl. 963987 1999; Chem. Abstr.
1999, 132, 35624.
9. Reinscheid, R. K.; Nothacker, H. P.; Bourson, A.; Ardati,
A.; Henningsen, R. A.; Bunzow, J. R.; Grandy, D. K.; Lan-
gen, H.; Monsma, F. J., Jr.; Civelli, O. Science 1995, 270, 792.
10. Tsuge, O.; Kanemasa, S.; Ohe, M.; Yorozu, K.; Take-
naka, S.; Ueno, K. Bull. Chem. Soc. Jpn. 1987, 60, 4067.
11. Affinities of the compounds for the human NK1 receptor
were evaluated in CHO cells infected with the human NK1
receptor (usingthe Semliki virus expression system) and
radiolabelled with [3H] substance P. Displacement curves
were determined with at least seven concentrations of the
compound.
analoges.
Subsequent
coupling
of
3,5-bis(tri-
fluoromethyl)benzoyl chloride was applied to crude
material after evaporation of the cleavage reagent pro-
vidingcompounds of type C. The excess of unconsumed
carboxylic acid chloride was scavenged with amino-
methylated polystyrene resin resultingin hihgly pure