Utilizing the intramolecular Fukuyama-Mitsunobu reaction for a flexible synthesis of novel heterocyclic scaffolds for peptidomimetic drug design

Article abstract of DOI:10.1016/j.bmcl.2005.06.035

We report the synthesis of the novel scaffolds pyrazino[1,2-b]isoquinoline and pyrrolo[1,2-a]pyrazine displaying the somatostatin pharmacophores. Both classes of compounds contain a pyrazine heterocycle, which can be prepared in a straightforward manner utilizing an intramolecular Fukuyama-Mitsunobu reaction. As both the families derive from amino acids, they can be accessed in high optical purity.

Full text of DOI:10.1016/j.bmcl.2005.06.035

Bioorganic & Medicinal Chemistry Letters 15 (2005) 4033–4036
Utilizing the intramolecular Fukuyama–Mitsunobu reaction for
a flexible synthesis of novel heterocyclic scaffolds for
peptidomimetic drug design
Christoph W. Zapf,* Juan R. Del Valle and Murray Goodmanꢀ
Department of Chemistry and Biochemistry, University of California San Diego, La Jolla, CA 92093-0343, USA
Received 23 May 2005; revised 6 June 2005; accepted 6 June 2005
Available online 5 July 2005
Dedicated to the memory of Prof. Murray Goodman for a lifetime of scientific contributions and mentorship
Abstract—We report the synthesis of the novel scaffolds pyrazino[1,2-b]isoquinoline and pyrrolo[1,2-a]pyrazine displaying the
somatostatin pharmacophores. Both classes of compounds contain a pyrazine heterocycle, which can be prepared in a straightfor-
ward manner utilizing an intramolecular Fukuyama–Mitsunobu reaction. As both the families derive from amino acids, they can be
accessed in high optical purity.
Ó 2005 Elsevier Ltd. All rights reserved.
As part of our efforts to design and synthesize com-
pounds displaying an increased selectivity to the
somatostatin subtype receptors and enhanced in vivo
stability,1–5 we were interested in a heterocyclic scaf-
fold-based approach to somatostatin analogs.
Although the native peptide hormone itself shows strong
binding in the nanomolar-range to the five receptors
subtypes, it is essentially lacking any selectivity for the
respective subtypes.6 Much progress has been made in
the design of small molecules that bind effectively and
selectively to one of the five known subtypes of the
somatostatin receptor families.7 Some of the small mol-
ecule somatostatin analogs reported in the literature
(Fig. 1) have been shown to bind in vitro to a specific
receptor subtype with selectivities of up to five orders
of magnitude over the other four receptors.8,9 It could
be argued that the molecular topology of these com-
pounds has little in common with the structural features
of the Phe7-Trp8-Lys9-Thr10 unit of somatostatin which
has been shown to be the pharmacophore region of the
peptide hormone.10,11
Figure 1. Compounds which have been shown to selectively bind with
high affinity to one of the five human somatostatin receptor subtypes.
Inspired by the promising results of these small molecule
somatostatin analogs, we intended to extend this work
by a scaffold-based approach to nonpeptide somatostat-
in analogs. Recently, we were successful in designing and
preparing scaffold-based compounds which selectively
bound to the l- and d-opioid receptors.12 We were mind-
ful to design a scaffold that could incorporate a variety of
functionalities in an appropriate three-dimensional
arrangement and would allow late-stage diversification.
These concepts are of paramount interest to facilitate
the study of structure–activity relationships to optimize
the potency and selectivity of target molecules.
Keywords: Scaffolds; Pyrazino[1,2-b]isoquinoline; Pyrrolo[1,2-a]pyra-
zine; Fukuyama–Mitsunobu; Peptidomimetics.
*
Corresponding author at present address: Department of Chemistry,
Princeton University, Princeton, NJ 08544, USA. Tel.: +1 609 258
8162; fax: +1 609 258 1980; e-mail: czapf@Princeton.edu
Deceased June 1st 2004.
ꢀ
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.06.035

4034
C. W. Zapf et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4033–4036
Our design focused on two structurally related scaffolds,
the pyrazino[1,2-b]isoquinoline 1 and the pyrrolo[1,2-
a]pyrazine 2 scaffold decorated with the pharmaco-
phores of somatostatin analogs that relate to the native
peptide: the indole nucleus, the aminoalkyl chain, and
the commonly occurring aromatic moiety (Scheme 1).
The attachment of the aminoalkyl chain to the second-
ary amine was to occur among the last synthetic steps.
This strategy provided an advanced intermediate which
could be utilized for structure–activity studies of the al-
kyl chain. Ring closure to produce the piperazine core
was to be accomplished via alcohol through a Mitsun-
obu-related reaction. Alcohol 3 was to derive from an
amide-bond formation utilizing the corresponding ami-
no alcohol 4 and tryptophan 5.
The key step in this reaction scheme is the cyclization
reaction furnishing the protected cyclic secondary amine
on the piperazine ring. It was our intention to take
advantage of a protecting group which allowed us to
form this carbon–nitrogen bond. Derivatization of
amines as the corresponding 2-nitro-benzenesulfon-
amide (nosyl-amide) allows the formation of C–N bonds
under Mitsunobu conditions utilizing alcohols.13
Previously, related scaffolds were synthesized by a simi-
lar route albeit for application toward solid-phase
chemistry.14–16
Scheme 2. Reagents: (a) SOCl2, MeOH; (b) 2-nitro-benzenesulfonyl
chloride, Et3N, DCM, 80%; (c) LiOH, MeOH, H2O, quant.; (d)
BF3ÆOEt2, BH3ÆSMe2, THF, 78%; (e) EDC, HOBt, Et3N, DMF, 77%;
(f) DIAD, PPh3, THF; (g) Boc2O, DMAP, CH3CN; and (h) PhSH,
DBU, DMF, 41%, three steps.
Amide 9 was formed in good yield (77%) from acid D-6
and amine 8 utilizing 1-[3-(dimethylamino)propyl]-3-
ethylcarbodiimide (EDC) and 1-hydroxybenzotriazole
(HOBt) as the coupling reagents in DMF. As deter-
mined by 1H NMR, a single diastereomer was produced
for amide 9 which confirms the enantiopurity of amino
alcohol 8.
Synthesis of target compound 1 commenced with
D-tryptophan 5 (Scheme 2) which was converted into
its methyl ester and subsequently allowed to react with
2-nitro-benzenesulfonyl chloride in the presence of
triethylamine yielding the protected tryptophan. Sapon-
ification of the methyl ester provided acid D-6 in 80%
yield over the three steps.
Amide 9 set the stage for the intramolecular Fukuyama–
Mitsunobu reaction to form the protected piperazinone
structure. Diisopropyl azodicarboxylate (DIAD) togeth-
er with triphenylphosphine was utilized to accomplish
the cyclization reaction. Following the intramolecular
cyclization reaction, the indole ring was Boc-protected19
yielding the indolylmethyl substituted tricyclic structure.
Amine 8 was obtained from the corresponding enantio-
merically pure (S)-tetrahydroisoquinoline carboxylic
acid (TIC, 7) which was reduced utilizing borane meth-
ylsulfide complex in the presence of boron trifluoride
diethyl etherate in THF.17 After recrystallization, the
amino alcohol 8 was obtained in 78% yield. The optical
rotation was in agreement with the value reported in
literature.18
Among the various methods for the removal the nosyl
group reported in the literature, the use of thiophenol
in the presence of DBU was found to be the method
of choice for our purposes. This reagent combination
allowed the removal of the nosyl-protecting group pro-
ducing the cyclic secondary amine 10 in a 41% yield over
the three steps.
Amine 10 was acylated utilizing the amine-protected
acid 11 to produce the fully protected target structure
(Scheme 3). To minimize the steric bulk of the activated
carboxylic acid, PyBroP was employed as the coupling
reagent providing the desired amide in 80% yield. Subse-
quently, the protecting groups were removed from the
fully protected structure with 4 N HCl in dioxane pro-
ducing target compound 12.
Alkyl bromide 13, prepared from acid 11,20 was utilized
to alkylate the secondary amine 10 in the presence of
Scheme 1. Retrosynthetic analysis of target structures 1 and 2.

C. W. Zapf et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4033–4036
4035
Scheme 5. Reagents: (a) 16, PyBroP, Et3N, DMF, 27%; (b) 4 N HCl,
dioxane, quant.
synthesized. We recently reported on the asymmetric
synthesis of protected 4-alkylprolinols and prolines.21,22
Our newly developed method allowed for the effective
and diastereoselective incorporation of a wide range of
substituents onto the pyrrolidine ring.
Scheme 3. Reagents: (a) 11, PyBroP,Et3N, DMF, 80%; (b) 4 N HCl,
dioxane, quant.; and (c) 13, DIEA, DMF, 31%.
With enantiomerically pure (2S,4R)-4-benzylprolinol 18
available, the amine was coupled to previously synthe-
sized nosyl-protected tryptophan L-6 utilizing EDC
and HOBt as the coupling reagents in the presence of tri-
ethylamine (Scheme 6). Amide 19 was obtained in 62%
diisopropylethyl amine at elevated temperature produc-
ing the tertiary amine. The removal of the protecting
groups from this compound was accomplished quantita-
tively with 4 N HCl in dioxane producing target com-
pound 14.
1
yield and as a single diastereomer as determined by H
NMR.
Utilizing the same synthetic procedure leading to
compound 14 the corresponding diastereomer epi-12
(Scheme 4) was obtained starting from commercially
available L-tryptophan methyl ester 15.
Subsequently, the amide 19 was cyclized under Fukuy-
ama–Mitsunobu conditions furnishing the bicyclic core.
Boc-protection of the indole ring followed by removal of
the nosyl-protecting group produced the secondary
amine 20.
Intermediate epi-10, which is diastereomeric to second-
ary amine 10, lends itself for introduction of an element
of diversity as well. Coupling of lysine derivative 16,
which is acylated as a pivaloyl amide at the a-amine
and Boc-protected at the e-amine, gave rise to the corre-
sponding amide in a moderate yield. Incorporating the
sterically demanding pivaloyl amide side chain should
reduce the overall flexibility of the final target structure
17, which was obtained after deprotection with 4 N HCl
in dioxane (Scheme 5).
As indicated above, we contemplated a second class of
scaffolds which could serve as the basis for novel non-
peptidic somatostatin analogs. This scaffold is based
on a 1,4-diaza-2-keto-bicyclo[4.3.0]-nonane moiety 2
containing three asymmetric centers. As indicated in
Scheme 1, substituted prolinol 4 and tryptophan 5 serve
as starting materials from which the scaffold was to be
Scheme 6. Reagents: (a) EDC, HOBt, Et3N, DMF, 62%; (b) DIAD,
PPh3, THF; (c) Boc2O, DMAP, CH3CN; (d) PhSH, DBU, DMF, 33%,
three steps; (e) 11, PyBroP, Et3N, DMF, 80%; and (f) 4 N HCl,
dioxane, quant.
Scheme 4. The synthesis of epi-12 was carried out using L-tryptophan.

4036
C. W. Zapf et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4033–4036
As before, the aminoalkyl side chain was installed via an
amide bond utilizing protected aminohexanoic acid 11.
The amide-bond formation was carried out with PyBroP
as the coupling reagent thus providing the fully protect-
ed target structure in 80% yield. The final molecule 21
was obtained in quantitative yield by treating this amide
with 4 N HCl in dioxane.
nonpeptidic analogs of other biologically relevant
structures.
References and notes
1. Falb, E.; Salitra, Y.; Yechezkel, T.; Bracha, M.; Litman,
P.; Olender, R.; Rosenfeld, R.; Senderowitz, H.; Jiang, S.;
Goodman, M. Bioorg. Med. Chem. 2001, 9, 3255.
2. Mattern, R. H.; Moore, S. B.; Tran, T. A.; Rueter, J. K.;
Goodman, M. Tetrahedron 2000, 56, 9819.
Figure 2 summarizes the five target structures synthe-
sized. On the basis of their elements of diversity, we
hope to be able to answer relevant questions with regard
to the utility of our designed scaffold. The main issues
which can be addressed with this series of somatostatin
analogs revolve around regiochemistry, stereochemistry,
steric constraints, and charge.
3. Mattern, R.-H.; Tran, T.-A.; Goodman, M. J. Med.
Chem. 1998, 41, 2686.
4. Tran, T.-A.; Mattern, R.-H.; Afargan, M.; Amitay, O.;
Ziv, O.; Morgan, B. A.; Taylor, J. E.; Hoyer, D.;
Goodman, M. J. Med. Chem. 1998, 41, 2679.
5. Melacini, G.; Zhu, Q.; Osapay, G.; Goodman, M. J. Med.
Chem. 1997, 40, 2252.
Valuable information about the stereochemical require-
ments for the tryptophan moiety will be obtained by
comparing somatostatin analog 12 with epi-12. Com-
pound 14 contains a tertiary amine which will be pro-
tonated under physiological conditions. This structural
feature will enable us to draw conclusions about the ef-
fects of a positive charge in the somatostatin ligand.
Analog 17, with the additional pivaloyl amide feature,
follows the established trend of incorporating steric hin-
drance to the ligand therefore reducing the overall flex-
ibility of the compound and thus increasing its binding
affinity. Bicyclic structure 21 with the substituted prolin-
ol building block represents a member of a potentially
large family of somatostatin analogs based on the 1,4-
diaza-2-keto-bicyclo[4.3.0]-nonane scaffold.
6. Janecka, A.; Zubrzycka, M.; Janecki, T. J. Pept. Res.
2001, 58, 91.
7. Rohrer, S. P.; Birzin, E. T.; Mosley, R. T.; Berk, S. C.;
Hutchins, S. M.; Shen, D.-M.; Xiong, Y.; Hayes, E. C.;
Parmar, R. M.; Foor, F.; Mitra, S. W.; Degrado, S. J.;
Shu, M.; Klopp, J. M.; Cai, S.-J.; Blake, A.; Chan, W. W.
S.; Pasternak, A.; Yang, L.; Patchett, A. A.; Smith, R. G.;
Chapman, K. T.; Schaeffer, J. M. Science 1998, 282,
737.
8. Yang, L.; Guo, L.; Pasternak, A.; Mosley, R. T.; Rohrer,
S. P.; Birzin, E. T.; Foor, F.; Cheng, K.; Schaeffer, J. M.;
Patchett, A. A. J. Med. Chem. 1998, 41, 2175.
9. Yang, L.; Berk, S. C.; Rohrer, S. P.; Mosley, R. T.; Guo,
L.; Underwood, D. J.; Arison, B. H.; Birzin, E. T.; Hayes,
E. C.; Mitra, S. W.; Parmar, R. M.; Cheng, K.; Wu, T.-J.;
Butler, B. S.; Foor, F.; Pasternak, A.; Pan, Y.; Silva, M.;
Freidinger, R. M.; Smith, R. G.; Chapman, K. T.;
Schaeffer, J. M.; Patchett, A. A. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 10836.
10. Vale, W.; Brazeau, P.; Rivier, C.; Brown, M.; Boss, B.;
Rivier, J.; Burgus, R.; Ling, N.; Guillemin, R. Recent
Progr. Horm. Res. 1975, 31, 365.
11. Nutt, R. F.; Veber, D. F.; Curley, P.; Saperstein, R.;
Hirschmann, R. Int. J. Pept. Protein Res. 1983, 21, 66.
12. Creighton, C. J.; Zapf, C. W.; Bu, J. H.; Goodman, M.
Org. Lett. 1999, 1407.
The design and successful synthesis of these two novel
scaffolds are not limited to the area of somatostatin.
As was shown previously, we have full control over
the stereochemical properties of these scaffolds as well
as the substituents which serve as pharmacophores.
We believe that the facile and efficient syntheses of these
scaffolds will lend themselves to the design of related
13. Fukuyama, T.; Jow, C.-K.; Cheung, M. Tetrahedron Lett.
1995, 36, 6373.
14. Swayze, E. E. Tetrahedron Lett. 1997, 38, 8643.
15. Kung, P. P.; Swayze, E. Tetrahedron Lett. 1999, 40, 5651.
16. Arya, P.; Wei, C. Q.; Barnes, M. L.; Daroszewska, M.
J. Comb. Chem. 2004, 6, 65.
17. Aggarwal, V. K.; Humphries, P. S.; Fenwick, A. J. Chem.
Soc., Perkin Trans. 1999, 1, 2883.
18. Yamaguchi, R.; Hamasaki, T.; Sasaki, T.; Ohta, T.;
Utimoto, K.; Kozima, S.; Takaya, H. J. Org. Chem. 1993,
58, 1136.
19. Nakamura, K.; Baker, T. J.; Goodman, M. Org. Lett.
2000, 2, 2967.
20. Bromide 13 was prepared from acid 11 via reduction of
acid 11 using borane methyl sulfide complex (78%)
followed by bromination utilizing triphenylphosphine,
bromine, and imidazole (78%).
21. Del Valle, J. R.; Goodman, M. J. Org. Chem. 2003, 68,
3923.
Figure 2. Series of compounds prepared by means of the intramolec-
ular Fukuyama–Mitsunobu reaction.
22. Del Valle, J. R.; Goodman, M. Angew. Chem. Int. Ed.
Engl. 2002, 41, 1600.