New utilizations of optically active homoallylamines: Highly stereoselective synthesis of cyclic guanidine and thiourea

Article abstract of DOI:10.3987/COM-05-S(K)43

Halocyclizations of optically active homoallylguanidine and homoallylthiourea were examined. These reactions proceeded stereoselectively to give six membered cyclic guanidines and thiourea. High 1,3-asymmetric induction by the homoallylic substituents is

Full text of DOI:10.3987/COM-05-S(K)43

HETEROCYCLES, Vol. 66, 2005
101
HETEROCYCLES, Vol. 66, 2005, pp. 101 – 106. © The Japan Institute of Heterocyclic Chemistry
Received, 31st August, 2005, Accepted, 11th October, 2005, Published online, 14th October, 2005. COM-05-S(K)43
NEW UTILIZATIONS OF OPTICALLY ACTIVE HOMOALLYL-
AMINES: HIGHLY STEREOSELECTIVE SYNTHESIS OF CYCLIC
GUANIDINE AND THIOUREA
Reiko Yanada,a Akira Kaieda,a Kazuo Yanada,b and Yoshiji Takemoto*a
aGraduate School of Pharmaceutical Sciences, Kyoto University, Yoshida,
Sakyo-ku, Kyoto 606-8501, Japan. e-mail: takemoto@pharm.kyoto-u.ac.jp
bFaculty of Pharmaceutical Sciences, Setsunan University, Nagaotoge-cho,
Hirakata, Osaka 573-0101, Japan.
Abstract – Halocyclizations of optically active homoallylguanidine and
homoallylthiourea were examined. These reactions proceeded stereoselectively to
give six membered cyclic guanidines and thiourea. High 1,3-asymmetric
induction by the homoallylic substituents is observed in these halocyclizations.
The allylation of imines, providing the corresponding homoallylamines, is an important synthetic
transformation.1 Therefore asymmetric allylation reactions of allylic metal compounds to carbon-nitrogen
double bond of optically active imines have broadly been studied.2 Recently we also reported
diastereoselective samarium- or indium-mediated allylations of optically active aromatic imines
containing a β-alkoxyamino group or a β-amino hydroxy group as a chiral auxiliary unit on the nitrogen
4
of imines.3, So far, optically active homoallylamines have been transformed to β-amino acids,5
β-aminoesters,6 γ-lactams,7 piperidines,8 and nitrogen containing bicycles9 (Scheme 1). As a new
application method of optically active homoallylamine, we planned to build the second asymmetric
position by the cyclization reaction and applied to the construction of polyfunctionalized cyclic
compounds such as guanidine and thiourea derivatives which attract attentions for the synthesis of HIV-1
protease inhibitors,10 bromopyrrole alkaloide Manzacidin A11 and the ability as super bases.12 Here we
report the new synthetic application of optically active homoallylamines. The application contains the
novel stereoselective halocyclization of optically active homoallyl guanidines (4, 9) and
homoallylthiourea (11) as a key step.
† This paper is dedicated to the memory of Professor Kenji Koga (1938–2004).

102
HETEROCYCLES, Vol. 66, 2005
Scheme 1. Synthetic utilities of chiral homoallylamines.
NR4
O
HN
H
NH
HN
H
NH
NHR1
R2
X
X
R
R
R
H
H
R2 = CO2H
CO2R
CN
guanidine derivative
HIV-1 protease
inhibitors
R1HN
R
R2
O
R1N
R3
R
n
chiral homoallylamine
n = 1, 2
R5
SMe
Br
R1N
R
HN
O
N
O
N
N
R2
CO2H
N
X
O
R3
Me
O
Me
H
O
thiourea derivative
Manzacidin A
According to our method,4 we first synthesized optically active homoallylamine (2) from aromatic imine
(1) in 95% yield and >99% de. By removing chiral auxiliary of 2 with HIO4•2H2O and 40% aq. NH3,
optically active homoallylamine derivative (3) was obtained in 95% yield according to our method.3b
Subsequent condensation of homoallylamine (3) with N,N’-bis-Boc-thiourea lead to homoallyl guanidine
(4) in good yield.
Scheme 2. Synthesis of optically active acyclic homoallyl guanidines.
S
BocHN
NHBoc
HIO4·2H2O
40% aq. NH3
Allyl bromide
In
Ph NBoc
i-Pr
Ph i-Pr
Ph
WSC
OH
OH
N
H
NHBoc
Ph
N
N
H
NH2
DMF, rt
95% yield
99% de
MeOH/H2O, rt
95% yield
DMF, rt
79% yield
1
2
3
4
With the requisite substrate in hand, we next turned our attention to the key halocyclization.
Iodocyclization of guanidines bearing allyl, alkynyl or allenyl groups12c,13 has already been reported. We
investigated the cyclization reaction of guanidine bearing homoallyl group under variable conditions
(Table 1). The cyclization of compound (4) was first examined with several electrophiles in dioxane (runs
1–4). The cyclization of 4 with NIS proceeded smoothly to give the desired product (5a) in moderate
yield and stereoselectivity (run 1). Surprisingly when the same reaction was carried out with NCS,
stereoselectivity was quite opposite, resulting in the major formation of trans isomer (run 3). This reason

HETEROCYCLES, Vol. 66, 2005
103
is unclear at present. With pyridinium hydrobromide perbromide in dichloromethane, the rate acceleration
was dramatically improved (run 5). The similar better result was obtained when THF was used as solvent
(run 6). Use of pyridine, DMF and acetonitrile gave lower yields and stereoselectivities (runs 7–9). The
relative cis stereochemistry was established on the basis of NMR spectroscopy. The NOE experiment of
NMR spectrum (14.2%) between C4-hydrogen and C6-hydrogen in major stereoisomer (5b, X=Br) was
observed, showing the cis relationship between the bromomethyl and phenyl groups. We propose the
observed diastereoselectivity can be rationalized by considering a chair-like transition state for attack on
the halonium ion, where the phenyl group and the halonium ion are both oriented in a pseudo-equatorial
position. Similar models have been proposed to explain homoallylic induction for iodocyclization of
thioimidates13e and isothioureas.12c
Table 1. Cyclization reaction of homoallylguanidine.
NBoc
Ph
NBoc
NBoc
NHBoc
H
NHBoc
HN
H
electrophile
solvent
BocN
NH
6
5a X=I
5b X=Br
5c X=Cl
Ph
N
H
4
X
+
Ph
X
H
4
H
H
proposed transition state
run electrophile
X
I
solvent time (h) yield (%) 5 (cis/trans)
1
2
3
4
5
6
7
8
9
NIS
NBS
NCS
Py•HBr3
Py•HBr3
Py•HBr3
Py•HBr3
Py•HBr3
Py•HBr3
Dioxane
Dioxane
Dioxane
Dioxane
CH2Cl2
THF
10
10
48
10
5
68
47
35
93
90
87
52
69
80
5a (6 : 1)
5b (3 : 1)
5c (1 : 3)
6 : 1
Br
Cl
Br
Br
Br
Br
Br
Br
6 : 1
6 : 1
5
Pyridine
DMF
MeCN
10
10
10
2 : 1
5 : 1
3 : 1
Then we applied the bromoguanidination method described above to the synthesis of multifunctionalized
bicyclic compounds (10) and (12) (Scheme 3). Condensation of glyoxylic acid ethyl ester with
(S)-2-amino-2-isopropylethanol or (S)-2-amino-2-phenylethanol gave oxazolidine (6a) or (6b) as a
mixture of diatereomers, respectively. These diastereoisomers could be separated by column
chromatography on basic silica gel to give (2S)-6a and (2R)-6a or (2S)-6b and (2R)-6b, then reacted
respectively with methallylindium bromide. Surprisingly, the reaction gave only one diastereoisomer (7a)
or (7b) in sharp contrast to 2-fluoroalkyl-1,3-oxazolidines bearing an electronwithdrawing
trifluoromethyl group on the imine carbon atom, which also could be separated by column
chromatography, to provide a mixture of the diastereomers via without formation of an imine
intermediate.14 Our results indicate that the allylation reaction of (2S)- and (2R)-oxazolidines (6) with In
and methallyl bromide proceeded via the same imine intermediate (6’a) or (6’b). Then we transformed

104
HETEROCYCLES, Vol. 66, 2005
compound (7) to morpholinone (8) in the presence of a catalytic amount of acetic acid in toluene.
Guanidination of compound (8a) with N-Boc-thiourea gave compound (9). Fortunatelly, the
bromoguanidination according to our method with Py•HBr3 or NBS in CH2Cl2 gave only one
diastereomer (10)15 in excellent yield. The cis stereostructure between C10-hydrogen and C8-methyl
group was also determined by NOE experiment of NMR spectroscopy (37%). On the other hand, the
treatment of morpholinone derivative (8b) with benzyloxycarbonyl isothiocyanate (ZNCS) gave thiourea
(11). Then treatment of 11 with NBS also proceeded stereoselectively to furnish bicyclic compound (12)16
in excellent yield. The stereostructure of compound (12) was also elucidated as 1,3-cis by NOE
measurement. The bromocyclization of compounds (9) and (11) could proceed in stereoselective manners
to give bicyclic guanidine (10) and bicyclic thiourea (12), respectively. They are interested in versatile
synthetic intermediates of several biologically active compounds.
Scheme 3. Synthesis of multifunctionalized bicyclic compounds.
H
N
R
R
R
(2S)-6a: R=i-Pr
(2S)-6b: R=Ph
H
OHCCO2Et
H
N
EtO2C
EtO2C
R
N
+
R
MS4A
O
H
N
separation
In, methallyl bromide
EtOH, rt
OH
EtO2C
EtO2C
In
PhH
rt
O
OH
(2R)-6a: R=i-Pr
(2R)-6b: R=Ph
H2N
6'a
6'b
O
R=i-Pr, Ph
S
BocHN
NBoc
NBoc
BocHN
NHBoc
N
O
Py•HBr3 or NBS
BocN
N
10
WSC
8a
8b
8
Br
O
CH2Cl2, rt, 90–93%
H
MeCN, rt, 70%
O
R
N
R
Me
H
O
cat.AcOH
9
10
toluene
reflux
3 steps
EtO2C
N
H
O
O
ZHN
S
Ph
S
N
HO
7a
7b
8a
8b
Ph
N
ZNCS
ZN
55–65%
NBS
Br
O
CH2Cl2, 0 °C, 78%
CH2Cl2, rt, 91%
O
O
Me
12
H
O
11
In summary, we accomplished the novel utilization as chiral source of optically active homoallylamines.
This method contained the stereoselective cyclization reaction of optically active homoallylguanidine and
thiourea with some electrophiles. In this reaction, the complete 1,3-cis-stereoselectivity was observed. In
this manner, 1,3-cis six-membered monocyclic compounds and multifunctionalized bicyclic compounds
were synthesized easily. The construction of polyfunctionalized cyclic compounds was established using
highly stereoselective allylations with indium and halocyclizations. These compounds are useful as
diverse precursors for pharmaceutically active compounds, super bases and the intermediate for
Manzacidin A.

HETEROCYCLES, Vol. 66, 2005
105
ACKNOWLEDGEMENTS
This work was partially supported by Grant-in-Aid for Scientific Research (B) and (C) from the Ministry
of Education, Culture, Sports, Science, and Technology of Japan, and by grant from 21st Century COE
Program “Knowledge Information Infrastructure for Genome Science”.
REFERENCES AND NOTES
1. S. Kobayashi, M. Sugiura, H. Kitagawa and W. Lam, Chem. Rev., 2002, 102, 2227.
2. Y. Yamamoto and N. Asao, Chem. Rev., 1993, 93, 2207.
3. a) R. Yanada and T. Ibuka, J. Synth. Org. Chem. Jpn., 2000, 58, 597; b) R. Yanada, M. Okaniwa, A.
Kaieda, T. Ibuka, and Y. Takemoto, J. Org. Chem., 2001, 66, 1283.
4. R. Yanada, A. Kaieda, and Y. Takemoto, J. Org. Chem., 2001, 66, 7516.
5. Y. Sugimoto, H. Imamura, H. Sakoh, K. Yamada, and H. Morishima, Synlett, 2001, 1747.
6. B. Choucair, H. Léon, M.-A. Miré, C. Lebreton, and P. Mosset, Org. Lett., 2000, 2, 1851.
7. V. Nyzam, C. Belaud, F. Zammattio, and J. Villiéras, Tetrahedron: Asymmetry, 1996, 7, 1835.
8. M. Billelt, P. Klotz, and A. Mann, Tetrahedron Lett., 2001, 42, 631.
9. A. Ferry, T. Billard, and B. R. Langlois, Synlett, 2005, 1027.
10. For example, nonpeptide HIV protease inhibitors a) P. Gayathri, V. Pande, R. Sivakumar, and S. P.
Gupta, Bioorg. Med. Chem., 2001, 9, 3059; b) C. N. Anil, J. Philippa, W. Xia, M. Stanislav, and J.
W. William, J. Med. Chem., 2002, 45, 973; c) A. R. Katritzky, A. Oliferenko, A. Lomaka, and M.
Karelson, Bioorg. Med. Chem. Lett., 2002, 12, 3453; d) V. Lather and A. K. Madan, J. Mol. Graph.
Modell., 2005, 23, 339.
11. a) K. Namba, T. Shinada, T. Teramoto, and Y. Ohfune, J. Am. Chem. Soc., 2000, 122, 10708; b) P.
M. Wehn and J. D. Bois, J. Am. Chem. Soc., 2002, 124, 12950; c) J. C. S. Woo and D. B. MacKay,
Tetrahedron Lett., 2003, 44, 2881; d) C. Drouin, J. C. S. Woo, D. B. MacKay, and R. M. A. Lavigne,
Tetrahedron Lett., 2004, 45, 7197.
12. a) E. J. Corey and M. J. Grogan, Org. Lett., 1999, 1, 157; b) T. Isobe, K. Fukuda, and T. Ishhikawa,
J. Org. Chem., 2000, 65, 7770; c) T. Isobe, K. Fukuda, T. Tokunaga, H. Seki, K. Yamaguchi, and T.
Ishikawa, J. Org. Chem., 2000, 65, 7774; d) T. Isobe, K. Fukuda, Y. Araki, and T. Ishikawa, Chem.
Commun., 2001, 243; e) B. Kovacevic, Z. B. Maksic, R. Vianello, and M. Primorac, New J. Chem.,
2002, 26, 1329; f) B. Kovacevic and Z. B. Maksic, Chem. Eur. J., 2002, 8, 1694.
13. Iodocylization of guanidines: a) M. Watanabe, H. Okada, T. Teshima, M. Noguchi, and A. Kakehi,
Tetrahedron, 1996, 52, 2827; b) M. Noguchi, H. Okada, M. Watanabe, K. Okuda, and O. Nakamura,
Tetrahedron, 1996, 52, 6581; Iodocylization of amidines: c) P. A. Hunt, C. May, and C. J. Moody,
Tetrahedron Lett., 1988, 29, 3001; d) J.-W. Chern, C. -Y. Shiau, and G.-Y. Lu, Bioorg. Med. Chem.

106
HETEROCYCLES, Vol. 66, 2005
Lett., 1991, 1, 571. Iodocylization of thioimidates: e) H. Takahata, T. Takamatsu, Y.-S. Chen, N.
Ohkubo, T. Yamazaki, T. Momose, and T. Date, J. Org. Chem., 1990, 55, 3792.
14. a) A. Ishii, K. Higashiyama, and K. Mikami, Synlett, 1997, 1381; b) J. Legros, F. Meyer, M.
Coliboeuf, B. Crousse, D. B.-Delpon, and J.-P. Bégué, J. Org. Chem., 2003, 68, 6444.
15. Compound (10): 1H-NMR (CDC13, 500 MHz) δ 1.07 (d, J = 6.7 Hz, 6H), 1.46 (s, 9H), l.46 (s, 9H),
1.80 (s, 3H), 1.93 (m, 1H), 2.12 (dd, J = 12.5, 13.1 Hz, 1H), 2.51 (dd, J = 1.5, 13.1 Hz, 1H), 3.67 (d,
J = 10.4 Hz, 1H), 4.17 (d, J = 10.4 Hz, 1H), 4.30 (dd, J = 1.5, 12.5 Hz, 1H), 4.41 (dd, J = 2.4, 12.2
13
Hz, 1H), 4.57 (dd, J = 1.5, 12.2 Hz, 1H), 4.58 (m, 1H); C-NMR (CDC13, 126 MHz) δ 19.1, 19.2,
22.6, 28.0, 28.4, 28.3, 40.6, 45.3, 51.6, 54.2, 60.5, 67.9, 79.3, 83.7, 151.3, 151.4, 158.6, 167.0; IR
(CHCl3) ν 3028, 2980, 2934, 1735, 1666, 1578 cm-1; MS (CI) m/z 518 (MH+); HRMS (CI) calcd. for
C22H37N3O6Br (MH+) 518.1866, found 518.1872; [α]31D –90.7° (c 0.69, CHC13).
16. Compound (12): 1H-NMR (CDC13, 500 MHz) δ 1.58 (s, Me, 3H), 2.31 (dd, J = 12.2, 14.0 Hz, 1H),
2.69 (dd, J = 4.0, 14.0 Hz, 1H), 3.52 (d, J = 11.3 Hz, 1H), 3.54 (d, J = 11.3 Hz, 1H), 4.36 (dd, J =
4.0, 12.2 Hz, 1H), 4.70 (dd, J = 3.7, 12.2 Hz, 1H), 4.86 (dd, J = 3.4, 12.2 Hz, 1H), 5.10 (d, J = 12.2
13
Hz, 1H), 5.14 (d, J = 12.2 Hz, 1H), 6.42 (dd, J = 3.4, 3.7 Hz, 1H), 7.27–7.39 (m, 10H); C-NMR
(CDC13, 126 MHz) δ 25.8, 38.9, 42.4, 45.7, 53.9, 54.3, 67.7, 68.6, 126.6, 128.0, 128.2, 128.4, 128.6
129.3, 134.3, 136.2, 160.8, 163.5, 166.4; IR (CHCl3) ν 3031, 3011, 2961, 1749, 1667, 1496 cm-1;
MS (CI) m/z 503 (MH+); HRMS (CI) calcd. for C23H24N2O4BrS (MH+) 503.0641, found 503.0627;
[α]24D +21.1° (c 1.05, CHC13).