Aglairubine - Discrepancies during the course of structure elucidation

Article abstract of DOI:10.1016/S0040-4039(02)00869-9

A short synthesis of the originally proposed structure 1 for the putrescine alkaloid aglairubine is presented as well as for a conceivable structure alternative 11. Due to the ascertained mismatch of spectroscopical data for synthetic and natural compounds, the published aglairubine structure has to be revised.

Full text of DOI:10.1016/S0040-4039(02)00869-9

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 4609–4612
Aglairubine—discrepancies during the course of structure
elucidation
Richard Detterbeck and Manfred Hesse*
Organisch-chemisches Institut der Universita¨t Zu¨rich, Winterthurerstrasse 190, CH-8057 Zu¨rich, Switzerland
Received 2 May 2002; accepted 7 May 2002
Abstract—A short synthesis of the originally proposed structure 1 for the putrescine alkaloid aglairubine is presented as well as
for a conceivable structure alternative 11. Due to the ascertained mismatch of spectroscopical data for synthetic and natural
compounds, the published aglairubine structure has to be revised. © 2002 Elsevier Science Ltd. All rights reserved.
A short communication, published by Saifah and
Suparakchinda in 1998,1 dealt with isolation and struc-
ture elucidation of a formerly unknown putrescine
alkaloid, named aglairubine. This alkaloid was isolated
from the leaves of Aglaia rubigenosa (Hiern) Pannel, a
plant that belongs to the genus Meliaceae. Aglaia spe-
cies—especially widespread in the southeastern part of
Asia—are well known as a rich source of putrescine
alkaloids, a subclass of the common, but rather unsys-
tematically distributed class of polyamine alkaloids.
The published aglairubine structure attracted our atten-
tion due to the fact that one of the amide moieties is
incorporated in an unsaturated five-membered ring.
Contrary to this, all other putrescine bisamides known
SePh
CH3
SePh
CH3
c)
a)
b)
HN
BocN
BocN
HN
CH3
CH3
O
O
O
O
2
3
4
5
O
Br
d)
N
7
O
SePh
CH3
N
O
N
SePh
CH3
e)
N
O
O
H2N
6
8
O
O
f)
Cl
10
O
O
SePh
CH3
g)
N
N
CH3
N
H
N
H
O
O
9
1
Scheme 1. (a) Boc2O, NEt3, DMAP, CH3CN, rt, 10 h, quant. (b) KHMDS, PhSeBr, Et2O, −65°C, 30 min, 89%. (c) CF3COOH,
toluene, rt, 1 h, 83%. (d) KHMDS, 7, DMF, 0°C“rt, 10 h, 60%. (e) NH2-NH2, EtOH, reflux, 2 h, 90%. (f) NEt3, 10, CH2Cl2,
rt, 15 min, 95%. (g) MCPBA, CH2Cl2, −35°C“rt, 2 h, 94%.
* Corresponding author. Fax: +41-1-635-68-12; e-mail: mtbohley@oci.unizh.ch
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00869-9

4610
R. Detterbeck, M. Hesse / Tetrahedron Letters 43 (2002) 4609–4612
so far as natural occurring alkaloids2 show an open
chain structure i.e. the amide groups are not incorpo-
rated in a cyclic framework.3 In the course of our study
of such compounds, we decided to synthesise this alka-
loid in order to compare the spectroscopical data and
to verify this new structure type.
measurements comfortably. Surprisingly for such a
rather uncomplicated structure without any stereogenic
centre, the NMR data published did not fit with our
measured spectra. Of course, care has been taken to use
the same conditions of measurement as in literature. In
order to be completely sure about the structure of our
product, we examined it extensively by two-dimensional
NMR spectroscopy (HSQC, COSY, HMBC and even
INADEQUATE). Not surprisingly, all of our recorded
data perfectly confirmed the structure of 1.
For this purpose, we were watching out for a synthesis
sequence which could bring up the desired structure
unequivocally, so the synthesis itself had to be an
independent source of information of the final structure
to accompany the structure elucidation by spectro-
scopic means.
Due to the significant differences with the published
data by Saifah and Suparakchinda for aglairubine, we
have therefore to conclude that either their structure
elucidation is wrong or there are some problems in the
analytical data set published. The published NMR
spectra for aglairubine are definitely not in accordance
with structure 1.
As a starting material for the synthesis shown in
Scheme 1 served commercially available 3-methyl-2-
pyrrolidone (2). Boc-protection of this cyclic amide to
tert-butyl 3-methyl-2-oxo-1-pyrrolidine-carboxylate (3)
could be performed in excellent yield using a protocol
of Grehn et al.,4 which describes the reaction of Boc2O
with amides in acetonitrile in the presence of catalytic
amounts of DMAP. The double bond needed in the
target structure 1 could be introduced in a latent form,
using the versatile phenylselenyl group, so deprotona-
tion of 3 by KHMDS at low temperature and subse-
quent addition of phenylselenyl bromide resulted
quickly in 4.5 For the following Boc-deprotection, we
found it much more convenient, in view of reaction
time, to use TFA in toluene instead of the more com-
monly used halogenated solvents. The free amide 56
obtained could be successfully alkylated to 67 by N-(4-
bromobutyl)phthalimide (7), using DMF as solvent and
KHMDS again as a base. The putrescine core thus
constructed in 6 now demanded the liberation of the
amine functionality to bring in the cinnamic acid
residue in the next step. Hydrazinolysis of the phthal-
imide group in EtOH worked well and allowed isola-
tion of free amine 88 without problems. With 8 in hand,
it was easy to isolate 99 after acylation with cinnamoyl
chloride (10). Compound 9 represented the whole skele-
ton of structure 1, besides the double bond. We were
pleased to see that a simple oxidation of the phenylse-
lenyl group by MCPBA as an oxidant at lower temper-
atures resulted in the desired 110 by means of the
scheduled elimination reaction.
Unfortunately, we could not examine the natural
product ourselves because of a lack of material. On the
other hand, we took into consideration that only the
position of the methyl group on the five-membered ring
might be wrong in structure 1. We assumed that a
methyl group shifted to the neighbouring position—as
shown in structure 11—could possibly be the correct
structure for natural aglairubine. To survey this pre-
sumption, we decided also to synthesise structure 11,
using the pathway presented in Scheme 2. The reaction
of ethyl (Z)-4-bromo-3-methyl-2-butenoate (12) with
easily obtainable mono-cinnamoyl-putrescine (13)11
allowed isolation of the desired 11 in one single step, so
a highly effective synthesis was at hand. Bromo ester 12
itself could be obtained from commercially available
3-methyl-but-2-enoic acid (14), following a method
described by Mata and Thomas.12 It may be added that
a possible, but somewhat tedious chromatographic sep-
aration of both geometric isomers 12 and 15, obtained
in a ratio of nearly 1:1 by the NBS bromination of ethyl
3-methyl-2-butenoate (16), proved to be dispensable for
a successful synthesis of 11.13 For obvious reasons of
hindered rotation around the double bond, only the
(Z)-isomer 12 can undergo the cyclisation to the five-
membered ring in 11.
Again, we used the same extensive NMR spectroscopi-
cal examination as mentioned above to successfully
authenticate our synthetic product. Disappointingly,
The reaction sequence presented allowed the isolation
of 1 in a few hundred-milligram scale, enabling spectral
HOOC
H3C
EtOOC
H3C
EtOOC
H3C
EtOOC
BrH2C
b)
a)
+
CH3
CH2Br
CH3
CH3
14
16
12
15
CH3
O
O
NH2
c)
N
N
H
+ 12 + 15
N
H
O
13
11
Scheme 2. (a) EtOH, H2SO4, reflux, 10 h, 72%. (b) NBS, AIBN, CCl4, reflux, 3 h, 96%. (c) NEt3, CH2Cl2, reflux, 1.5 h, 30%.14

R. Detterbeck, M. Hesse / Tetrahedron Letters 43 (2002) 4609–4612
4611
the new structure 11 also did not fit the published
spectroscopical data and had therefore to be excluded
as a structure alternative for aglairubine, provided
that the published NMR spectra are correct for iso-
lated aglairubine.
troscopical data with the published spectra by Saifah
and Suparakchinda.1 If there was no mistake
during their publication procedure for these data,
the structure 1 is in fact incorrect. At least, the
already published NMR spectra are definitely wrong
for compound 1. The correct spectra for 1 and 11
are presented in our Tables 1 and 2. A final deci-
sion about the virtual aglairubine structure demands
for a new structure elucidation on authentically mate-
rial.
Conclusion
Neither the originally proposed structure 1 nor the
structural alternative 11 show identity in NMR-spec-
1
Table 1. Comparison of the H NMR data (CD3OD, 500 MHz) of the synthesised compounds 1 and 11 with the published
spectra for aglairubine
CH3
O
O
N
CH3
N
N
H
N
H
Published 1H NMR for aglairubine1
O
1
O
11
Dl in ppm
[ppm]
[ppm]
[ppm]
Dl in ppm
+0.11
+0.08
+0.07
7.65 (2H)
7.60 (1H)
7.45 (3H)
7.54 (2H)
7.52 (1H)
7.38 (2H)
7.36 (1H)
6.59 (1H)
6.33 (1H)
4.23 (2H)
3.32 (2H)
3.28 (2H)
1.83 (3H)
1.60 (4H)
7.56 (2H)
7.52 (1H)
7.36 (3H)
+0.02
0.0
−0.02
0.0
+0.09
+0.61
−0.28
+0.26
+0.14
+0.10
+0.12
+0.04
6.68 (1H)
6.94 (1H)
3.95 (2H)
3.58 (2H)
3.42 (2H)
1.93 (3H)
1.72 (2H)
1.64 (2H)
6.59 (1H)
5.79 (1H)
4.02 (2H)
3.45 (2H)
3.34 (2H)
2.08 (3H)
1.65 (2H)
1.56 (2H)
0.0
−0.54
−0.21
+0.13
−0.06
+0.25
+0.05
−0.04
Table 2. Comparison of the 13C NMR data (CD3OD, 500 MHz) of the synthesised compounds 1 and 11 with the published
spectra for aglairubine
CH3
O
O
N
N
CH3
N
H
N
H
Published 13C NMR for aglairubine1
[ppm]
O
O
11
1
Dl in ppm
[ppm]
[ppm]
Dl in ppm
+2.4
+0.2
+1.0
+0.3
+2.5
+3.5
+0.2
0.0
−0.5
+0.3
−7.2
+3.0
+0.1
+0.1
−0.6
−1.6
174.4
168.8
141.8
136.5
138.4
136.0
130.9
130.1
128.9
122.1
52.3
172.0
168.6
140.8
136.2
135.9
132.5
130.7
130.1 (2C)
129.4 (2C)
121.8
59.5
40.3
174.8
168.8
141.8
136.5
122.7
159.4
130.9
130.1
128.9
122.1
57.2
+2.8
+0.2
+1.0
+0.3
−13.2
+26.9
+0.2
0.0
−0.5
+0.3
−2.3
+2.5
+0.8
+0.0
−0.6
+2.2
43.3
40.3
27.9
27.2
42.8
40.3
27.8
27.2
40.2
27.8 (2C)
11.4
13.0
15.2

4612
R. Detterbeck, M. Hesse / Tetrahedron Letters 43 (2002) 4609–4612
(ddd, 1H), 2.13–2.03 (m, 1H), 1.72–1.60 (m, 2H), 1.58 (s,
Acknowledgements
3H), 1.57–1.45 (m, 2H); 13C NMR (75 MHz, CDCl3): l
174.9 (s), 168.2 (s), 137.7 (d), 133.8 (d), 132.0 (s), 129.0
(d), 128.6 (d), 126.8 (s), 123.1 (d), 49.4 (s), 43.9 (t), 42.4
(t), 37.4 (t), 35.1 (t), 25.9 (t), 24.7 (q), 24.4 (t); CIMS
(NH3): m/z 457 (100, [M+1]+), 299 (56, [M−C6H5Se]+).
8. 1-(4-Aminobutyl)-3-methyl-3-(phenylselanyl)-2-pyrrolidi-
none (8): Oil; IR (CHCl3): w 2990, 2930, 2860, 1675, 1475,
We thank the NMR department of our institute and
especially Mrs. Nadja Walch for recording the INADE-
QUATE spectra. Furthermore, the Swiss National Sci-
ence Foundation is gratefully acknowledged for
generous support.
1
1450, 1435, 1370, 1280, 1100, 1020, 1000 cm−1; H NMR
(300 MHz, CDCl3): l 7.67 (d, J=8.0 Hz, 2H), 7.41–7.26
(m, 3H), 3.27–2.94 (m, 4H), 2.72 (t, J=6.7 Hz, 2H),
2.37–2.30 (ddd, 1H), 2.13–1.96 (m, 1H), 1.80 (br, 2H),
1.59 (s, 3H), 1.58–1.40 (m, 4H); 13C NMR (75 MHz,
CDCl3): l 174.9 (s), 137.7 (d), 129.0 (d), 128.6 (d), 126.8
(s), 49.5 (s), 43.9 (t), 42.7 (t), 41.5 (t), 35.1 (t), 30.4 (t),
24.8 (t), 24.4 (q); CIMS (NH3): m/z 327 (100, [M+1]+),
169 (8, [M−C6H5Se]+).
References
1. Saifah, E.; Suparakchinda, N. Planta Med. 1998, 64, 682.
2. For a review about polyamine chemistry as well as a
complete list of all up to June 2000 published polyamine
alkaloids see: Bienz, S.; Detterbeck, R.; Ensch, C.; Gug-
gisberg, A.; Ha¨usermann, U.; Meisterhans, C.; Wendt,
B.; Werner, C.; Hesse, M. In Putrescine, Spermine, Sper-
midine, and Related Polyamine Alkaloids; Cordell, G. A.,
Ed; Academic Press: New York, 2002; in press.
3. There is indeed a whole class of cyclic polyamine alka-
loids existing, which are derived from putrescine, but they
all possess the special 2-aminopyrrolidine skeleton and
are therefore considered as their own subclass.
9. (E)-N-{4-[3-Methyl-2-oxo-3-(phenylselanyl)-1-pyrroli-
dinyl]butyl}-3-phenyl-2-propenamide (9): Oil IR (CHCl3):
w 3470, 3300, 3060, 2990, 2920, 2860, 1665, 1625, 1575,
1510, 1450, 1435, 1370, 1330, 1280,1105, 985, 975, 910
1
cm−1; H NMR (300 MHz, CDCl3): l 7.68–7.59 (m, 3H),
7.51–7.46 (m, 2H), 7.39–7.25 (m, 6H), 6.72 (t, J=5.3 Hz,
1H), 6.50 (d, J=15.7 Hz, 1H), 3.43–3.37 (m, 2H), 3.29–
2.94 (m, 4H), 2.37–2.30 (m, 1H), 2.16–2.02 (m, 1H), 1.60
(s, 3H), 1.62–1.52 (m, 4H); 13C NMR (75 MHz, CDCl3):
l 175.4 (s), 166.1 (s), 140.3 (d), 137.7 (d), 135.0 (s), 129.3
(d), 129.2 (d), 128.7 (d), 128.6 (d), 127.7 (d), 126.7 (s),
121.1 (d), 49.5 (s), 44.0 (t), 42.5 (t), 39.3 (t), 35.0 (t), 26.0
(t), 24.8 (t), 24.8 (q); CI-MS (NH3): m/z 457 ([M+1]+).
10. (E)-N-[4-(3-Methyl-2-oxo-2,5-dihydro-1H-pyrrol-1-yl)-
butyl]-3-phenyl-2-propenamide (1): Oil IR (CHCl3): w
3450, 3300, 3300, 2970, 2860, 1670, 1645, 1630, 1580,
1515, 1460, 1450, 1415, 1335, 1240, 990, 980, 910, 815
cm−1; CI-MS (NH3): m/z 299 ([M+1]+). For the NMR
data of structure 1 see Tables 1 and 2.
4. Grehn, L.; Gunnarsson, K.; Ragnarsson, U. Acta Chem.
Scand. B 1986, 40, 745–750.
5. tert-Butyl 3-methyl-2-oxo-3-(phenylselanyl)-1-pyrrolidine-
1
carboxylate (4): Oil; H NMR (300 MHz, CDCl3): l 7.65
(d, J=8.0 Hz, 2H), 7.41–7.38 (m, 1H), 7.33–7.27 (m, 2H),
3.59–3.55 (m, 1H), 3.38–3.34 (m, 1H), 2.30–2.20 (m, 1H),
2.12–2.00 (m, 1H), 1.59 (s, 3H), 1.52 (s, 9H). 13C NMR
(75 MHz, CDCl3): l 173.8 (s), 150.2 (s), 137.7 (d), 129.5
(d), 128.8 (d), 126.2 (s), 82.6 (s), 49.6 (s), 42.9 (t), 33.5 (t),
27.9 (q), 24.1 (q).
6. 3-Methyl-3-(phenylselanyl)-2-pyrrolidinone
(CHCl3): w 3439, 3021, 1697, 1622, 1476, 1417, 1376,
1305, 1273, 1213, 1102, 661, 561 cm−1 1H NMR (300
(5):
IR
11. Experimental details for the synthesis of N-mono-cin-
namoyl-putrescine (13) are published elsewhere: Detter-
beck, R.; Hesse, M.; Tetrahedron 2002, in press.
;
MHz, CDCl3): l 7.70 (d, J=6.8 Hz, 2H), 7.42–7.32 (m,
1H), 7.31–7.27 (m, 2H), 6.90 (br, 1H), 3.17–3.11 (m, 1H),
3.07–2.99 (m, 1H), 2.41–2.33 (ddd, 1H), 2.23–2.12 (m,
1H), 1.60 (s, 3H); 13C NMR (75 MHz, CDCl3): l 178.9
(s), 137.6 (d), 129.1 (d), 128.8 (d), 126.8 (s), 48.3 (s), 39.1
(t), 37.1 (t), 24.3 (q); CIMS (NH3): m/z 273 (100,
[M+18]+), 256 (54, [M+1]+).
12. Mata, E. G.; Thomas, E. J. J. Chem. Soc., Perkin Trans.
1 1995, 785–799.
13. (E)-N-[4-(4-Methyl-2-oxo-2,5-dihydro-1H-pyrrol-1-yl)-
butyl]-3-phenyl-2-propenamide (11): Oil IR (CHCl3): w
3440, 3300, 2995, 2930, 2860, 1670, 1630, 1580, 1550,
1510, 1460, 1450, 1410, 1330, 1150, 985, 975, 940, 840
cm−1; CIMS (NH3): m/z 299 ([M+1]+); EI MS: m/z 298
7. 2-{4-[3-Methyl-2-oxo-3-(phenylselanyl)-1-pyrrolidinyl]-
butyl}1H-isoindole-1,3(2H)-dione (6): Oil; IR (CHCl3): w
3040, 2990, 2940, 2860, 1770, 1710, 1680, 1615, 1470,
1435, 1395, 1370, 1360, 1330, 1280, 1265, 1100, 1040,
+
’
+
(71, [M] ), 152 (51, [M−C6H5CHCHCONH] ), 131 (100,
[M−C6H5CHCHCO]+), 103 (63, [C6H5CHCH]+), 77 (32,
[C6H5]+). The NMR spectra of structure 11 are given in
Tables 1 and 2.
1020, 1000, 690 cm−1 1H NMR (300 MHz, CDCl3): l
;
7.86–7.80 (m, 2H), 7.74–7.66 (m, 2H), 7.65–7.63 (m, 2H),
7.38–7.25 (m, 3H), 3.69 (t, J=7.0 Hz, 2H), 3.33–3.23 (m,
1H), 3.18–3.03 (m, 2H), 2.97–2.89 (m, 1H), 2.37–2.29
14. No efforts have been undertaken to optimise the reaction
conditions.