Spirocyclic NK1 antagonists I: [4.5] and [5.5]-Spiroketals

Article abstract of DOI:10.1016/S0960-894X(02)00506-1

A series of novel spiroketal-based NK₁ antagonists is described. The effect of modifications to the spiroether ring and aromatic substituents are discussed, leading to the identification of compounds with high affinity and excellent CNS penetration.

Full text of DOI:10.1016/S0960-894X(02)00506-1

Bioorganic & Medicinal Chemistry Letters 12 (2002) 2515–2518
Spirocyclic NK1 Antagonists I: [4.5] and [5.5]-Spiroketals
Eileen M. Seward,a,* Emma Carlson,b Timothy Harrison,a
Karen E. Haworth,a Richard Herbert,a Fintan J. Kelleher,a
Marc M. Kurtz,c Jonathan Moseley,a Simon N. Owen,a
Andrew P. Owens,a Sharon J. Sadowski,c
Christopher J. Swaina and Brian J. Williamsa
aDepartment of Medicinal Chemistry, Merck Sharp and Dohme Research Laboratories,
The Neuroscience Research Centre, Terlings Park, Eastwick Road, Harlow, Essex, UK
bDepartment of Pharmacology, Merck Sharp and Dohme Research Laboratories,
The Neuroscience Research Centre, Terlings Park, Eastwick Road, Harlow, Essex, UK
cDepartment of Biochemistry, Merck Research Laboratories, Rahway, NJ 07065, USA
Received 2 May 2002; revised 18 June 2002; accepted 28 June 2002
Abstract—A series of novel spiroketal-based NK1 antagonists is described. The effect of modifications to the spiroether ring and
aromatic substituents are discussed, leading to the identification of compounds with high affinity and excellent CNS penetration.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
face p–p interaction between two aromatic rings which
We have previously described our ongoing studies to
define the NK1 receptor antagonist pharmacophore.1
This work culminated in the identification of MK-869
(1),2 the first in a new class of anti-emetics, allowing
superior and sustained protection from acute and
delayed chemotherapy-induced nausea and vomiting.3
MK-869 has also shown efficacy in major depression;
this seminal discovery represents the first novel
mechanism of action to treat depression in 40 years.4
may be important in stabilizing the bioactive conforma-
tion. In our back-up program we were interested in con-
formationally restricted ligands in which the disposition of
these two key aromatic rings would be fixed by virtue of
the structural framework. One system we considered was
based on the spirocyclic ketal 1,7-dioxaspiro[5.5]undecane
(2). Stereoelectronic effects have long been recognized that
influence the conformation of acetals and indeed it has
been shown that spiro system 2 exists in conformation 2a
only (below). This conformer benefits from two reinfor-
cing anomeric effects: both C–O bonds being anti-
periplanar to an oxygen lone pair in the spiro ring. These
effects are additive and it has been estimated that 2a is 9.5
kJ/mol lower in energy than 2b, possessing only a single
anomeric effect, and 19 kJ/mol lower than 2c. Thus 2 is
regarded as a conformationally rigid system.5
A common feature of our high affinity NK1 antagonists,
based on NMR and crystallographic evidence,1 is the
presence of an intramolecular edge-to-face or face-to-
Molecular modelling indicated that the optimal dis-
position of the aryl rings for binding to the receptor
would be achieved by the 5S,6S,9S-stereoisomer (3).
*Corresponding author. E-mail: eileen_seward@merck.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00506-1

2516
E. M. Seward et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2515–2518
Furthermore, a key H-bond between the receptor and
the antagonist would be possible from O7.1
Table 1. Effect of stereochemistry on hNK1 binding affinitya
We herein report the synthesis and biological profile of
a series of [5.5] and related [4.5]-spiroketals—novel NK1
receptor antagonists.
10a R=Bn
11a R=H (hNK1 44% @ 1 mM)
10b R=Bn
11b R=H (hNK1 28Æ10 nM)
Synthetic Chemistry
The required spiroketals were prepared by reaction of
N-protected 3-(S)-phenylmorpholinone (4)1 with a sui-
tably substituted acetylene anion (Scheme 1). The
required acetylene for the synthesis of the [5.5] spiro-
ketals was obtained using Seebach6 chemistry: this
involved the regiospecific attack of a TMS-protected
titanium acetylide complex at the benzylic position of
an aromatic epoxide 5 to afford the homopropargylic
alcohol 6. Clean TMS-deprotection was achieved by
treatment of 6 with silver nitrate followed by potassium
cyanide.7 Following protection of the alcohol, acetylene
7 was reacted with morpholinone 4 to give hemiacetal 8.
Saturation and deprotection afforded the key inter-
mediate diol 9. We anticipated that acetalisation of 9
under acid catalysis might be problematic due to the
proximity of the basic nitrogen; indeed, no cyclisation
10c R=Bn
11c R=H (hNK1 47% @ 1 mM)
10d R=Bn
11d R=H (hNK1 400Æ108 nM)
aDisplacement of [125I]-labelled substance P from the cloned human
receptor expressed in CHO cells.12 Values are displacements or
IC50sÆS.D. and are means of three experiments.
was observed upon treatment with a variety of mild acid
catalysts. However, cyclisation was achieved by heating
at reflux in strong mineral acid. This gave rise to a
mixture of the four possible diastereoisomers (10a–d)
(a:b:c:d=6:3:3:1 Table 1) which were separated by
chromatography or crystallization. It was not possible
to re-equilibrate 10a under these conditions.
The relative stereochemistries and solution conforma-
1
tions (Table 1) were determined by H NMR.8 In all
isomers the pyran and morpholine oxygens were axial
underlining the stabilizing influence of the anomeric
effect in this system. Upfield shifts for the tri-
fluoromethyl-substituted aromatic ring in isomer 10b
(5S,6S,9S-stereoisomer) were consistent with a con-
formation in which the two aromatic rings have an
edge-to-face p–p interaction.
The N-benzyl group was removed by hydrogenolysis to
afford the parent morpholines (11a–d Table 1) which
were further derivatised to the corresponding triazo-
lones 12 or triazoles 13 by alkylation and cyclisation
(Scheme 1).
The same general strategy (not shown) was used to pre-
pare the [4.5] spiroketals. The requisite acetylenes in this
case were commercially available as the propargylic
alcohols. A more flexible approach to the [4.5] system
was also developed which allowed a rapid exploration
of aromatic substitution (Scheme 2).
Scheme 1. Reagents and conditions: (i) nBuLi, THF, À78 ꢀC; (ii)
(OiPr)3TiCl, À50 ꢀC; (iii) (5), THF, À78 ꢀC; (iv) AgNO3, EtOH, 0 ꢀC;
(v) KCN, H2O; (vi) 2,6-lutidine, TBS triflate, CH2Cl2, 0 ꢀC; (vii)
nBuLi, THF, À78 ꢀC; (viii) (4), THF, À78 ꢀC; (ix) H2, 30 psi, PtO2,
EtOAc; (x) MeOH/H+; (xi) 6 NHCl, MeOH, reflux; (xii) H 2, 40 psi,
Pd/C, MeOH/H+; (xiii) chloromethylamidrazone, K2CO3, DMF,
60 ꢀC; (xiv) xylene, 140 ꢀC; (xv) ClCH2CꢁCCH2Cl, K2CO3, DMF;
(xvi) NaN3, DMSO; (xvii) NHMe2, dioxane, 80 ꢀC.
Addition of allylmagnesium chloride to the morpholi-
none 4 proceeded smoothly to afford the alkylated lac-
tol 14. Dihydroxylation of the double bond gave the
triol 15 as a mixture of isomers. Cyclization in strong
mineral acid at reflux generated two diastereomeric

E. M. Seward et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2515–2518
2517
Discussion
A comparison of the four possible epimers of the [5.5]
spiroketals at the 5- and 9-positions confirmed that the
5S,6S,9S-stereochemistry was preferred (Table 1), con-
sistent with our pharmacophore model. The parent
unsubstituted [5.5]-spiroketal had modest NK1 binding
affinity (25 IC50 170 nM, Table 2), but it is well docu-
mented that appropriate aromatic and N-substitution
can have a dramatic effect on NK1 affinity.1,11 Accord-
ingly, a 10-fold increase in affinity was observed upon
N-substitution with a polar heterocycle (26 IC50 19 nM);
alternatively, substitution at the ortho position of the
9-phenyl ring was also beneficial (11b IC50 28 nM).
In contrast, substitution in the meta or para positions
(27, 28) resulted in a considerable reduction in affi-
nity. Combining ortho substitution in the 9-phenyl
ring with N-substitution, and indeed, with fluorine
substitution at the 4-position of the 5-phenyl ring
afforded compounds with very high NK1 binding
affinity (29, 30).
Compounds 29 and 30 were selected for further evalua-
tion in vivo. The CNS penetration of selected com-
pounds was assessed by their ability in gerbils to block
the foot-tapping induced by central infusion of the
NK1-selective agonist GR73632.13 Both compounds
showed good CNS penetration (29 ID50 1.1 mg/kg iv; 30
ID50 0.6 mg/kg iv) when given immediately prior to
agonist challenge.
Scheme 2. Reagents and conditions: (i) allylmagnesium chloride,
THF, À78 ꢀC; (ii) OsO4, NMO, THF, tBuOH, H2O; (iii) 6 NHCl,
MeOH, 65 ꢀC; (iv) (COCl)2, DMSO, CH2Cl2, À78 ꢀC, then Et3N;
(v) NaHMDS, THF, À78 ꢀC; (vi) (19); (vii) (20), Pd(PPh3)4, LiCl,
Na2CO3, DME, H2O, 80 ꢀC; (viii) Pd/C, HCO2NH4, MeOH, reflux;
(ix) Me3SnSnMe3, Pd(PPh3)4, LiCl, Li2CO3, THF, H2O, 80 ꢀC; (x)
(23), Pd(PPh3)4, LiCl, DMF, 80 ꢀC; (xi) nBuLi, THF, À78 ꢀC; (xii)
B(OMe)3; (xiii) HCl, 0 ꢀC.
We also explored the corresponding [4.5] series (Table
3). The NK1 affinity of the parent unsubstituted [4.5]-
spiroketal (31 IC50 99 nM) compares favorably with the
analogous [5.5]-spirocycle (25 IC50 170 nM). Substitu-
tion at both the ortho and meta positions of the 9-aryl
ring appears to be well tolerated (32–35), although para
pairs of alcohols 16 which were oxidized to afford the
ketones 17. These were separated by chromatography
and ketone 17a was enolised regioselectively using
NaHMDS; quenching of the resulting enolate with tri-
flimide 199 gave reproducible yields of the enol triflate
18. The triflate was then cleanly displaced under Suzuki
conditions using appropriately substituted aromatic
boronic acids 20 to give the dihydrofuran intermediates
21. The triflate could also be converted to the tri-
methylstannane derivative 22 which was cross-coupled
with a variety of aryl bromides 23 under Stille condi-
tions. The requisite aryl bromides 23 and boronic acids
20 were either commercially available or prepared by
standard literature procedures.10
Table 2. hNK1 Binding affinity of [5.5]-spiroketal NK1 antagonists
Compd
R
H
X
H
R1
H
hNK1 IC50, nMa
25
170Æ17
26
H
H
19Æ3
11b
27
28
2-CF3
3-CF3
4-CF3
H
H
H
F
F
F
28Æ10
513Æ45
The dihydrofuran intermediates 21 were reduced to the
product tetrahydrofurans by hydrogenation to give the
desired 3S stereoisomers 24; the 10-aryl substituent
apparently blocked one face of the double bond from
the hydrogenation catalyst. This stereochemistry was
55% @1 mM
29
30
2-CF3
F
1.1Æ0.8
1.4Æ0.7
1
2-OCF3
F
confirmed by H NMR nOe studies. N-Debenzylation
occurred concomitantly under the hydrogenation con-
ditions and the resulting amines were N-substituted, if
required, using the same conditions described in
Scheme 1.
aDisplacement of [125I]-labelled substance P from the cloned human
receptor expressed in CHO cells.12 Values are displacements or
IC50sÆS.D. and are means of three experiments.

2518
E. M. Seward et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2515–2518
Table 3. hNK1 Binding affinity of [4.5]-spiroketal NK1 antagonists
Acknowledgements
We thank Steve Thomas for invaluable analytical and
NMR support.
(3S, 5S, 10S)
References and Notes
Compd
R
R’
X
NK1 IC50 (nM)a
1. Swain, C. J.; Seward, E. M.; Cascieri, M. A.; Fong, T. M.;
Herbert, R.; MacIntyre, D. E.; Merchant, K. J.; Owen, S. N.;
Owens, A. P.; Sabin, V.; Teall, M.; VanNiel, M. B.; Williams,
B. J.; Sadowski, S.; Strader, C.; Ball, R. G.; Baker, R. J. Med.
Chem. 1995, 38, 4793.
2. Hale, J. J.; Mills, S. G.; MacCoss, M.; Finke, P. E.; Cascieri,
M. A.; Sadowski, S.; Ber, E.; Chicchi, G. G.; Kurtz, M.; Metz-
ger, J.; Eiermann, G.; Tsou, N. N.; Tattersall, F. D.; Rupniak,
N. M. J.; Williams, A. R.; Rycroft, Wayne; Hargreaves, R.;
MacIntyre, D. E. J. Med. Chem. 1998, 41, 4607.
31
32
33
34
35
36
H
H
H
H
H
H
H
H
H
H
H
H
H
99Æ18
24Æ19
2-CF3
3-CF3
2-OMe
3-OMe
4-OMe
30Æ10
13Æ6
183Æ42
27%@1 mM
37
2-OMe
H
0.56Æ0.21
3. Navari, R. M.; Reinhardt, R. R.; Gralla, R. J.; Kris, M. G.;
Hesketh, P. J.; Khojasteh, A.; Kindler, H.; Grote, T. H.; Pen-
dergrass, K.; Grunberg, S. M.; Carides, A. D.; Gertz, B.
J. NewEngl. J. Med. 1999, 340, 190.
38
39
40
41
2-OMe
2-OMe
2-OMe
2-OMe
5-OMe
5-F
5-iPr
H
H
H
H
16Æ2
10Æ1
2.8Æ1
1.3Æ0.4
5-CF3
4. Kramer, M. S.; Cutler, N.; Feighner, J.; Shrivastava, R.;
Carman, J.; Sramek, J. J.; Reines, S. A.; Liu, G.; Snavely, D.;
Wyatt-Knowles, E.; Hale, J. J.; Mills, S. G.; MacCoss, M.;
Swain, C. J.; Harrison, T.; Hill, R. G.; Hefti, F.; Scolnick,
E. M.; Cascieri, M. A.; Chicchi, G. G.; Sadowski, S.; Wil-
liams, A. R.; Hewson, L.; Smith, D.; Carlson, E. J.; Har-
greaves, R. J.; Rupniak, N. M. J. Science 1998, 281, 1640.
5. Deslongchamps, P.; Rowan, D. D.; Pothier, N.; Sauve, T.;
Saunders, J. K. Can. J. Chem. 1981, 59, 1105.
42
43
2-OMe
2-OMe
H
0.1Æ0.06
0.1Æ0.05
aDisplacement of [125I]-labelled substance P from the cloned human
receptor expressed in CHO cells.12 Values are displacements or
IC50sÆS.D. and are means of three experiments.
6. Seebach, D. Chem. Ber. 1988, 121, 1315.
7. Schmidt, H. M.; Arens, J. F. Recl. Trav. Chim. Pays-Bas
1967, 86, 1138.
8. The stereochemistry at C5 was fixed as (5S) and the relative
3
stereochemistry at C9 was deduced from the values of J for
substitution (36) is detrimental to binding. As in the
[5.5] series, introduction of a polar heterocycle onto the
morpholine nitrogen provided a twenty-fold increase in
affinity (37), thus affording the first compound in this
series with sub-nanomolar NK1 binding affinity. Alter-
natively, introduction of a second substituent onto the
3S aryl ring, most favorably at the 5-position, afforded
a series of very high affinity antagonists (38–42) and
obviated the need for an N-substituent. Re-introduction
of the polar heterocycle was well tolerated (43) but
offered no increase in affinity.
H9. The relative stereochemistry at the spiro centre was
deduced using NOESY data; the orientation of the 5-phenyl
group was equatorial for the (6S) isomers and axial for the (6R)
isomers. Coupling constants were consistent with normal chair
conformations for both rings in each of the isomers (10a–d).
9. Comins, D. L.; Dehghani, A. Tetrahedron Lett. 1992, 33,
6299.
10. Baker, R.; Harrison, T.; Swain, C. J.; Williams, B. J. WO
9719084.
11. Ladduwahetty, T.; Baker, R.; Cascieri, M. A.; Chambers, M.;
Haworth, K.; Keown, L. E.; MacIntyre, D. E.; Metzger, J. M.;
Owen, S. N .; Rycroft, W.; Sadowski, S.; Seward, E. M.; Shep-
heard, S. L.; Swain, C. J.; Tattersall, F. D.; Watt, A. P.; William-
son, D. W.; Hargreaves, R. J. J. Med. Chem. 1996, 39, 2907.
12. Cascieri, M. A.; Ber, E.; Fong, T. M.; Sadowski, S.; Ban-
sal, A.; Swain, C. J.; Seward, E. M.; Frances, B.; Burns, D.;
Strader, C. D. Mol. Pharmocol 1992, 42, 458.
13. Rupniak, N. M. J.; Tattersall, F. D.; Williams, A. R.;
Rycroft, W.; Carlson, E.; Cascieri, M. A.; Sadowski, S.; Ber,
E.; Hale, J. J.; Mills, S. G.; MacCoss, M.; Seward, E.;
Huscroft, I.; Owen, S.; Swain, C. J.; Hill, R. G.; Hargreaves,
R. J. Eur. J. Pharmacol. 1997, 326, 201.
The 2-methoxy-5-(5-trifluoromethyl)tetrazolyl14 deriva-
tive (42, IC50 0.1 nM), was evaluated in our in vivo CNS
assays. The compound showed excellent CNS penetra-
tion as measured by blockade of agonist-induced foot
tapping in the gerbil (ID50 0.1 mg/kg iv) when given
immediately prior to agonist challenge.
In conclusion, we describe a novel series of NK1 recep-
tor antagonists based upon a rigid spiroketal framework
which have high affinity and excellent CNS penetration.
Further in vivo evaluations and additional SAR studies
will be reported in subsequent publications.
14. Armour, D. R.; Chung, K. M. L.; Congreve, M.; Evans,
B.; Guntrip, S.; Hubbard, T.; Kay, C.; Middlemiss, D.; Mor-
daunt, J. E.; Pegg, N. A.; Vinader, M. V.; Ward, P.; Watson,
S. P. Bioorg. Med. Chem. Lett. 1996, 6, 1015.