Parallel solid-phase synthesis of disubstituted (5-biphenyltetrazolyl) hydantoins and thiohydantoins targeting the growth hormone secretagogue receptor

Article abstract of DOI:10.1016/j.bmcl.2003.11.012

An efficient solid-phase protocol for the synthesis of substituted (5-biphenyltetrazolyl)-hydantoins and -thiohydantoins has been developed. Suzuki cross-coupling between resin-bound 2-(tetrazol-5-yl)-phenylborinane and 4-bromobenzaldehyde gave the corresponding tetrazolylbiphenyl aldehyde. Subsequent reductive amination using amino acid esters gave the pivotal resin bound amino acid esters which were transformed to hydantoins or thiohydantoins via two routes: (i) treatment with isocyanates or isothiocyanates or (ii) successive treatment with triphosgene and primary amines. Using molecular modeling, we were able to jump from L-692,429, a well known non-peptidyl growth hormone secretagogue (GHS), to biphenyltetrazolyl hydantoins, obtaining compounds with IC₅₀ values below 600 nM after two iterative cycles only.

Full text of DOI:10.1016/j.bmcl.2003.11.012

Bioorganic & Medicinal Chemistry Letters 14 (2004) 317–320
Parallel solid-phase synthesis of disubstituted
(5-biphenyltetrazolyl) hydantoins and thiohydantoins targeting
the growth hormone secretagogue receptor
Rune Severinsen,a Jesper F. Lau,a Kent Bondensgaard,b Birgit S. Hansen,a
Mikael Begtrupc and Michael Ankersena,*
aMedicinal Chemistry, Novo Nordisk A/S, Novo-Nordisk Park 2760 Maaloev, Denmark
bProtein Engineering, Novo Nordisk A/S, Novo-Nordisk Park 2760 Maaloev, Denmark
cThe Danish University for Pharmaceutical Sciences, Universitetsparken 2, 2100 Copenhagen O, Denmark
Received 5August 2003; revised 22 October 2003; accepted 10 November 2003
Abstract—An efficient solid-phase protocol for the synthesis of substituted (5-biphenyltetrazolyl)-hydantoins and -thiohydantoins
has been developed. Suzuki cross-coupling between resin-bound 2-(tetrazol-5-yl)-phenylborinane and 4-bromobenzaldehyde gave
the corresponding tetrazolylbiphenyl aldehyde. Subsequent reductive amination using amino acid esters gave the pivotal resin
bound amino acid esters which were transformed to hydantoins or thiohydantoins via two routes: (i) treatment with isocyanates or
isothiocyanates or (ii) successive treatment with triphosgene and primary amines. Using molecular modeling, we were able to jump
from L-692,429, a well known non-peptidyl growth hormone secretagogue (GHS), to biphenyltetrazolyl hydantoins, obtaining
compounds with IC50 values below 600 nM after two iterative cycles only.
# 2003 Elsevier Ltd. All rights reserved.
1. Introduction
Growth hormone (GH) is synthesized and released from
somatotrophs (i.e., GH releasing cells) in the anterior
pituitary.1 Both GH synthesis and GH release from
somatotrophs are under tight control by two hypothal-
amic hormones; growth hormone releasing hormone
(GHRH)2 which stimulates the synthesis and release of
GH, and somatostatin3 which inhibits GH release. In
addition to the effect of GHRH, a new class of com-
pounds, called growth hormone secretagogues (GHS), is
able to induce GH release from the pituitary.4 An
endogenous ligand of the GHS pathway is Ghrelin,5 a
28 amino acid O-n-octanoylated peptide, which was
identified more than 20 years later than the first syn-
thetic compounds. Ghrelin has been shown to be
released from the stomach and is able to release GH in
vitro as well as in vivo. It is widely believed that ghrelin
and the synthetic GHSs elicit their effect at both the
hypothalamic and pituitary level and work synergisti-
Figure 1. L-629,429 and general library scaffold.
cally with GHRH.6 Several non-peptidyl mimics of
Ghrelin have been reported, notably the compound L-
692,429 from Merck (Fig. 1).7 Here we report a new
series of nonpeptidyl compounds containing the biphe-
nyltetrazole moiety of L-629,429 combined with dis-
ubstituted hydantoins and thiohydantoins in the general
structure 1. All compounds were prepared on solid
support using a parallel synthesis strategy.
2. Modeling
Keywords: Biphenyltetrazole; Hydantoins; Cross-coupling reactions;
Growth hormone.
Previously, we have reported a model of L-692,429
docked in a receptor model of the GHS receptor.8 L-
* Corresponding author. Tel.: +45-4443-2181; fax: +45-4443-4547;
e-mail: miak@novonordisk.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.11.012

318
R. Severinsen et al. / Bioorg. Med. Chem. Lett. 14 (2004) 317–320
692,429 and the general scaffold 1 share a biphenyl-
tetrazole fragment. Using this common fragment, the
core structure of hydantoin 1 was docked into a GHS
receptor model such that position of the biphenylte-
trazole fragment in the L-692,429–GHS receptor com-
plex was preserved. In both cases, the biphenyltetrazole
fragment is situated in a pocket between TM 5and TM
6 spanned by Phe204(TM 5), Phe249(TM 6) and
Trp253(TM 6) with the tetrazole moiety in the vicinity
of Lys199(TM 5). The docked core of scaffold 1 con-
strains the accessible orientations of the R1 and R2
substitutents. R1 is directed towards a pocket formed by
a cluster of aromatic residues W104(TM 2), F119 (TM
3), F309(TM 7), F312(TM 7) and Y313(TM 7) whereas
R2 points towards Glu124(TM 3), which acts as a
counterion for the amino group of L-692,429.9 Due to
the nature of these pockets, substrates with aromatic
and amino group substituents were tested. Distances
from the core of 1 to the complementary groups in the
GHS receptor were estimated using the receptor model,
and R1- and R2-groups were chosen accordingly. A
similar arrangement of pharmacophoric elements can be
achieved by swapping R1 and R2, and rotating the
hydantoin system 180 degrees relative to the biphe-
nyltetrazole group. However, the directionality of the
aromatic and the amino substituents is somewhat dif-
ferent, and seems less favorable as judged by the recep-
tor model. Likewise was the S stereoisomer of 1
preferred over the R stereoisomer.
tetrazolyl-2-phenyl)-5,5-dimethyl-[1,3,2]-dioxa-borinane
3 in 91% yield (Scheme 1). The tetrazole 3 was readily
immobilized on a 2-chloro trityl resin in the presence of
DIPEA. Subsequent treatment with 4-bromobenzaldehyde
under standard Suzuki cross-coupling conditions10 yiel-
ded the biphenyl aldehyde 5. Amino acid esters were
introduced via reductive amination using NaBH3CN in
a mixture of DMF–MeOH–trimethyl orthoformate
containing 1% acetic acid to give the resin bound amino
acid ester 6. Subsequent reaction with isothiocyanates
or isocyanates in the presence of DIPEA followed by
cleavage with 5% TFA in DCM yielded the corre-
sponding thiohydantoins 9 or hydantoins 10 in high
purity (>90%) and moderate yields (40–70%).
To extend the scope of the scaffold, another strategy
was investigated. Resin bound amino acid ester 6 was
treated with triphosgene11 in the presence of DIPEA
followed by reaction with primary amines. Upon clea-
vage using 5% TFA in DCM, the hydantoins 10 was
isolated in high purities ꢀ90% and in moderate to good
yields (40–80%).13,14 In the case of the sterically hindered
4-amino-1-Boc-piperidine, cleavage from the resin yiel-
ded the uncyclicized derivative, which was treated with
10% acetic acid/THF at 50 ꢁC to give 10a in 45% yield.
The scope of the strategy outlined in Scheme 2 was fur-
ther extended by the usage of different aromatic bromo
aldehydes in the Suzuki reaction. Thiophenes and fluoro
substituted benzenes reacted with high purity and good
yields, while furans yielded poor results in the Suzuki
reaction. Pyridines worked well in the Suzuki reaction
and in the reductive amination reaction, but did not
cyclicize to form hydantoins.
3. Combinatorial solid-phase chemistry
In order to find an efficient method for the introduction of
variable groups in the hydantoin/thiohydantoin scaffold 1,
a robust solid-phase strategy leading to 1 was developed.
The compounds synthesized following the strategy out-
lined in Scheme 2, were tested for displacement of 125I-
Ghrelin to the human GHS-receptor expressed in BHK-
cells.12 Selected results are shown in Table 1. In the
initial library R1 and R2 were varied, taking the phar-
macophore model described earlier into consideration.
When comparing the IC50 values of thiohydantoins 9
and hydantoins 10, the results suggest that hydantoins
fits the binding-pocket better than thiohydantoins (9e
and 10d). The results further suggested that the distance
between the aromate in R1 and the hydantoin ring had
significant effect on the binding. When the aromatic ring
was attached directly to the hydantoin ring, or when
separated by one carbon atom, the binding was sig-
nificantly lower than when separated by two carbon
atoms (9a, 9b and 9e).
Phenyltetrazole was treated with n-BuLi and TMEDA
followed by addition of triisopropyl borate. The resulting
mixture of borinanes, boronic acids and boroxines were
esterified with neopentyl glycol, to give the necessary (5-
Scheme 1. (a) 2 in TMEDA, THF at À78 ꢁC, then BuLi in hexanes
and stirred 1 h to À10 ꢁC, then B(OiPr)3 at À78 ꢁC; (b) neopentylgly-
col, ethyl acetate, 4 h.
Scheme 2. Reagents and conditions: (a) 2-chlorotrityl resin, 3 equiv DIPEA, DCM, rt 2 h; (b) 4-bromobenzaldehyde, aqueous NaOH, toluene/
ethanol, Pd(PPh3)4; (c) amino acid ester, (TMOF/MeOH/DMF/AcOH)/(33:33:33:1), 2 h, NaBH3CN, MeOH/DMF, 4 h; (d) R2NCO, DIPEA,
DCM, 4 h, 45 ꢁC; (e) DIPEA, DCM, CO(OCCl3)2, (f) R2NH2, DIPEA, DCM then 45 ꢁC 4 h; (g) 5% TFA/DCM.

R. Severinsen et al. / Bioorg. Med. Chem. Lett. 14 (2004) 317–320
319
Table 1. Selected results from the library synthesized following the route outlined in Scheme 2
Product
Amino acid (S)-isomer
(R1)
Isothiocyanate,
isocyanate or amine (R2)
Aromatic bromo
aldehydes Br-Ar1-CHO
Purity
LC–MS ELS (%)
IC50 GHS-R
(mM)
Yield after
purification
9a
99
97
89
96
99
45359
94
91
97
89
98
98
92
91
3574
3565
>100
4549
8.2
9b
9c
45
57
9d
9e
10a
10b
10c
10d
10e
10f
10g
10h
10i
0
>100
>100
4.0
82
52
76
69
78
85
1.1
0.6
1.6
1572
1.3
70
Anilines, primary amines, secondary amines, amides
and tertiary amines were all tested in the R2 position,
but only the tertiary amines yielded significant binding.
Based on these results a second library was made in
which different R2 groups were combined with the most
potent R1 (homophenylalanine) resulting in several
compounds with IC50 values below 2 mM. The most
potent compounds all contained a tertiary amine, situ-
ated three carbon atoms from the hydantoin ring (10d,
10e and 10f). When the tertiary amine was constrained
in a ring system, the binding was further enhanced. The
R-isomer of 10b, 10d, and 10f was synthesized to
explore the statement about the S-isomer being pre-
ferred over the R-isomer. The R-isomer all had IC50
values above 100 mM, thus supporting the results
obtained by modeling.
The introduction of heteroaromatic or fluorosubstituted
rings, did not have any significant influence on the
binding (10g, 10i and 10f), even though 10h displayed a
substantial lower binding.
In summary R1 in the model should contain an aro-
matic ring separated from the hydantoin-ring by at least
two carbon atoms. R2 should contain a tertiary amine,
separated from the hydantoin ring by three carbon
atoms and preferable the amine nitrogen should be part
of a constraining ring. The SAR obtained is in good

320
R. Severinsen et al. / Bioorg. Med. Chem. Lett. 14 (2004) 317–320
agreement with both the chemical nature and spatial
arrangement of the R-groups of the pharmacophoric
model developed using the receptor model.
12. Hansen, B. S.; Raun, K.; Nielsen, K. K.; Johansen, P. B.;
Hansen, T. K.; Peschke, B.; Lau, J.; Andersen, P. H.;
Ankersen, M. Eur. J. Endocrinol. 1999, 141, 180.
13. General method for preparation of biphenyl tetrazole
hydantoins 10 using RNH2. (5-Tetrazolyl-2-phenyl)-5,5-
dimethyl-[1,3,2]-dioxaborinane 3 was immobilized on a
1% DVB resin, using a 2-chlorotrytil linker. Borinane 3
(26 mg, 0.1 mmol, 2 equiv) was dissolved in DCM (2 mL)
and DIPEA (68 mL, 0.4 mmol, 8 equiv) and added to the
resin (50 mg, 0.05 mmol, 1 equiv). The mixture was agitated
for 1 h at 25 ꢁC and subsequently washed with 3ÂDCM,
2ÂMeOH and 3ÂDCM yielding resin bound borinane 4.
4-Bromobenzaldehyde (27.6 mg, 0.15mmol, 3 equiv) was
dissolved in toluene (1.8 mL), EtOH (0.2 mL) and 1 N
NaOH(aq) (0.15mL, 0.15mmol, 3 equiv) and the resin
was swelled in the mixture. Subsequently the mixture was
4. Conclusion
In conclusion, an efficient solid-phase protocols for the
synthesis of biphenyltetrazolyl thiohydantoins 9 and
hydantoins 10 from simple building blocks has been
developed. Employing these strategies, libraries target-
ing GHSR were produced yielding compounds with
IC50 values below 1 mM. A SAR described, based on the
results, is in good agreement with the initial pharmaco-
phoric model.
purged with N2 for 5min, where upon Pd(PPh ) (6 mg,
3 4
0.005mmol, 0.1 equiv) was added to the mixture. The reac-
tion vessel was then sealed with a rubber septum, and agi-
tated at 50ꢁC for 8 h. Subsequent washing with 3ÂDMF,
2ÂMeOH, 2ÂDCM and 3ÂDMF gave aldehyde 5.
Amino acid esters (0.15mmol, 3 equiv) were dissolved in
a mixture of (MeOH/DMF/TMOF/AcOH 33:33:33:1), (1
mL) and added to the resin. The mixture was agitated for
1 h before NaBH3CN (25.2 mg, 0.4 mmol, 4 equiv) dis-
solved in DMF:MeOH/1:1, (1 mL) was added, and the
mixture was agitated at 45 ꢁC for further 5h. The resin
was then washed with 3ÂDMF and 3ÂDCM resulting in
the amino acid esters 6.
Acknowledgements
The authors would like to thank Corporate Research
Affairs at Novo Nordisk A/S and The Danish Academy
of Technical Sciences for financial support.
References and notes
1. Reichlin, S.; Saperstein, R.; Jackson, I. M. D.; Boyd,
A. E.; Patel, Y. Ann. Rev. Physiol. 1976, 38, 389.
2. Rivier, J.; Spiess, J.; Thorner, M.; Vale, W. Nature 1982,
300, 276.
3. Krulich, L.; Dhariwal, A. P.; McCain, S. M. Endocrinol-
ogy 1968, 83, 783.
Amino acid esters 6 was then dissolved in DCM and
DIPEA (1 mL DCM and 0.1 mL DIPEA) and agitated 10
min. Triphosgene (45mg, 0.15mmol, 3 equiv) in DCM (1
mL) were added and the mixture was agitated for addi-
tionally 2 h, and washed with 3ÂDCM. Primary amines
dissolved in DCM (2mL) and DIPEA (25 mL, 0.15mmol,
3 equiv) was added to the resin, agitated for 3 h and sub-
sequently washed with 3ÂDCM, 2ÂDMF and 5ÂDCM
yielding hydantoins 8. Cleavage from the resin was affec-
ted using TFA/DCM 5:95, containing 1% TIS, and the
products were purified by reverse phase HPLC. The pro-
ducts were characterized by LC–MS and NMR. Data for
4. Bercu, B. B.; Walker, R. F. Growth Hormone Secretago-
gues; Springer: New York, 1996.
5. Kojima, M.; Hosoda, H.; Date, Y.; Nakazato, M.; Mat-
suo, H.; Kangawa, K. Nature 1999, 402, 656.
6. Bowers, C. Y.; Reynolds, G. A.; Durham, D.; Barrera,
C. M.; Pezzoli, S. S.; Thorner, M. O. J. Clin. Endocrinol.
Metab. 1990, 70, 975.
1
compound 10f: H NMR (400 MHz; CDCl3): d 1.36 (m,
7. Schoen, W. R.; Pisano, J. M.; Prendergast, T. K.; Wyv-
ratt, M. J.; Fisher, M. H.; Cheng, K.; Chan, W. W. S.;
Butler, B.; Smith, R. G.; Ball, R. G. J. Med. Chem. 1994,
37, 897.
8. Bondensgaard, K.; Thøgersen, H.; Ankersen, M.; Han-
sen, B. S.; Wulff, B. S.; Bywater, R. P. J. Med. Chem., in
press.
9. Feighner, S. D.; Howard, A. D.; Prendergast, K.; Palyha,
O. C.; Hreniuk, D. L.; Nargund, R.; Underwood, D.;
Tata, J. R.; Dean, D. C.; Tan, C. P.; McKee, K. K.;
Woods, J. W.; Patchett, A. A.; Smith, R. G.; Van der
Ploeg, L. H. T. Mol. Endocrinol. 1998, 12, 137.
10. Blettner, C. G.; Konig, W. A.; Ruhter, G.; Stenzel, W.;
Schotten, T. Synlett 1999, 307.
1H), 1.80 (m, 3H), 1.87 (m, 2H), 2.09 (m, 3H) 2.29, (m,
1H), 2.58 (m, 1H), 2.65 (m, 3H), 2.97 (t, J=12 Hz, 1H),
3.07 (t, J=12 Hz, 1H), 3.48 (m, 1H), 3.58 (m, 3H), 3.81
(d, J=16 Hz, 1H), 4.02 (dd, J1=4 Hz, J2=8 Hz, 1H),
4.88 (d, J=16 Hz, 1H), 6.97 (d, J=8 Hz, 2H), 7.03 (d,
J=8 Hz, 2H), 7.14 (d, J=8 Hz, 2H), 7.19 (t, J=8 Hz,
1H), 7.27 (t, J=8 Hz, 3H), 7.36 (dd, J1=2 Hz, J2=8 Hz,
1H), 7.46 (dt, J1=2 Hz, J2=8 Hz, 1H), 7.53 (dd, J1=2
.
Hz, J2=8 Hz, 1H). Anal. calcd for C30H33N7OS CF3-
.
CO2H H2O: C 60.42, H 5.79, N 14.09. Found: C 60.18, H
5.58, N 14.16.
14. General method for preparation of biphenyltetrazolyl
hydantoins usingRNCS or RNCO . Same procedure as
above, except that triphosgene treatment was omitted and
isocyanates or isothiocyanates were added in DCM/
DIPEA.
11. Cotarca, L.; Delogu, P.; Nardelli, A.; Sunjic, V. Synthesis
Stuttgart 1996, 5, 553.