Synthesis and structure elucidation of open-chained putrescine-bisamides from Aglaia species

Article abstract of DOI:10.1016/S0040-4020(02)00751-2

The structures of the eight recently described open-chained putrescine bisamide alkaloids secoodorine (1), aglaithioduline (3), aglaiduline (4), aglaidithioduline (5), grandiamide B (13), grandiamide C (14), pyramidatine (20), and secopiriferine (22), which were isolated from different Aglaia species, have been verified by synthesis. In addition to that, the published structure 2 for hemileptagline had to be revised, and for secoodorine (1) the absolute configuration could be established.

Full text of DOI:10.1016/S0040-4020(02)00751-2

Tetrahedron 58 (2002) 6887–6893
Synthesis and structure elucidation of open-chained
putrescine-bisamides from Aglaia species
*
Richard Detterbeck and Manfred Hesse
¨ ¨ ¨
Organisch-chemisches Institut der Universitat Zurich, Winterthurerstrasse 190, CH-8057 Zurich, Switzerland
Received 22 March 2002; revised 6 June 2002; accepted 27 June 2002
Abstract—The structures of the eight recently described open-chained putrescine bisamide alkaloids secoodorine (1), aglaithioduline (3),
aglaiduline (4), aglaidithioduline (5), grandiamide B (13), grandiamide C (14), pyramidatine (20), and secopiriferine (22), which were
isolated from different Aglaia species, have been verified by synthesis. In addition to that, the published structure 2 for hemileptagline had to
be revised, and for secoodorine (1) the absolute configuration could be established. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
known. Therefore, we decided to synthesize these alkaloids
in order to fill this gap in literature.
The whole group of naturally occurring polyamine alkaloids
can be divided according to the incorporated polyamine part
into the three sub-classes of putrescine, spermidine, and
spermine alkaloids.† For each of these groups numerous
examples exist, which occur either in an open-chained, but
also in a cyclic form. The latter is well known for
spermidine and spermine alkaloids since centuries, whereas
the knowledge about the cyclic form of putrescine alkaloids
is not so widespread. Surely, this comes up with the rather
restricted naturally occurrence of these 2-aminopyrrolidine
compounds, which—to the best of our knowledge—could
up to now only be isolated from plants of the Aglaia genus
(Meliaceae).‡ It seems to be rather likely, that all alkaloids
with such a 2-aminopyrrolidine framework have an open-
chained putrescine-bisamide precursor, which can be
cyclized by a special enzymatic system. The open-chained
putrescine alkaloids are unsystematically distributed
throughout the plant kingdom. However, Aglaia species
are a rich source of such compounds, so all of the nine
substances, we want to deal within this report, were isolated
from different Aglaia plants. During the last few years these
putrescine bisamides were isolated and their structure
proposed mainly by spectroscopical means. Up to now,
there was no final structure proof by synthesis nor was the
absolute configuration of one of them—secoodorine (1)—
2. Results and discussion
In 2000, Greger et al.2 reported the isolation of three open-
chained putrescine bisamides from Aglaia leptantha Miqu.,
namely hemileptagline, leptagline (3), and aglanthine (4).
The proposed structures 2 and 3 both included (E )-3-
(methylthio)propenoic acid (6) as an acyl residue, therefore
suggesting to see these structures in context to the three
already described, also sulphur containing putrescine
bisamides, which were isolated by Saifah et al.3 in 1999
from the leaves of A. edulis A. Gray. These authors reported
about aglaiduline (4), aglaithioduline (3), and aglaidithio-
duline (5). A simple comparison of the structures, proposed
by both authors, revealed that leptagline (3) is identical with
aglaithioduline, whereas aglanthine (4) is identical to
aglaiduline. Attention should be drawn to the fact, that
Saifah et al. published their results in 1999, whereas Greger
et al. in the year 2000, so the compound names from Saifah
et al. take priority. For this reason the latter are preferred in
this report, the other ones are only given to clarify the
situation. In order to avoid the danger of confusion, we
suggest to eliminate the redundant ones from literature. For
the structure 2 of hemileptagline—proposed by Greger
et al.—we had from the very beginning the strong suspicion,
that the structure elucidation of the Austrian authors has
been wrong, because the published NMR-spectroscopic data
for hemileptagline point out for a symmetric structure,
whereas the proposed structure 2 is non-symmetric. Once
more, a simple comparison of the published NMR data for
hemileptagline and aglaidithioduline (5) showed identity of
both data sets. The structure of aglaidithioduline (5) consists
beside the putrescine backbone of two amide linked (E )-3-
(methylthio)propenoic acid residues and fits therefore with
Keywords:
putrescine-bisamides;
Aglaia;
hemileptagline;
aglaidithioduline; aglaithioduline; grandiamide B and C; aglaiduline;
pyramidatine; secopiriferine; secoodorine.
*
Corresponding author. Tel.: þ41-1-635-4281; fax: þ41-1-635-6812;
e-mail: mtbohley@oci.unizh.ch
For a complete and up to date review article on all aspects of polyamine
†
chemistry as well as a list of all up to now described polyamine alkaloids
see Bienz et al.1
A complete list of all known 2-aminopyrrolidine alkaloids together with
‡
the corresponding literature is given in Ref. 1.
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)00751-2

6888
R. Detterbeck, M. Hesse / Tetrahedron 58 (2002) 6887–6893
Scheme 1. (a) SOCl2, CH2Cl2, rt, 20 min. (b) H2N–(CH2)4–NHBoc (9), NEt3, CH2Cl2, RT, 15 min, 38%. (c) NEt3, CH2Cl2, rt, 1 h, 30%. (d) (1) TFA, toluene,
rt, 15 min, (2) aq. 2N NaOH, rt, 5 min, 80%. (e) C6H5CH2COCl (10), NEt3, CH2Cl2, rt, 30 min, 78%.
Scheme 2. (a) NEt3, CH2Cl2, rt, 2 h, 81%.
the measured and reported spectroscopical data of both
research teams. In order to be sure about the conclusions
around this somewhat dazzling facts, we decided to
synthesize the proposed structure 2 as well as 3–5.
The spectroscopical data measured for 2 fitted perfectly with
this non-symmetric structure, but were not at all identical
with the data published by Greger et al.2 Therefore, an
almost impossible coincidence in the spectroscopical data
has been definitely excluded. At this point it can be stated,
that the hemileptagline structure 2 is incorrect.
For this purpose, we first needed access to the commercially
not available (E )-3-(methylthio)propenoic acid (6). Besides
a scarcely user-friendly method, which uses methanthiole
under high pressure conditions,4 we were glad to see that De
Medeiros et al.5 reported a much easier synthesis for 6,
starting from propiolic acid (11) and potassium thiocyanate
as an olfactorily unproblematic sulphur source (Scheme 1).
The obtained acid 6 could be activated to the acid chloride 7
by means of thionyl chloride. The latter was converted
without any purification in modest yield§ to 8, using
N-mono-Boc-putrescine (9)k as an amine. After final
deprotection with TFA and extractive work-up from aq.
base, the desired non-protonated form of 2 was obtained.{
On the other hand, synthesized compound 2 provided us
with a valuable starting material, which enabled the
synthesis of aglaithioduline (3) in one simple step.
Amidation of 2 with phenylacetyl chloride 10 in the
presence of NEt3 yielded 3 directly. The spectroscopic
data for 3 were in good accordance with the published data
for the natural products.2,3 This means, as mentioned above,
that leptagline is identical with aglaithioduline (3).
The last one of the sulphur containing putrescine bisamides,
aglaidithioduline (5), has also been synthesised from 6 by
activating this acid as acid chloride 7 and subsequent
reaction with free putrescine (12). Beside small amount of 2
as by-product, the required aglaidithioduline (5) could be
isolated. Spectroscopic characterization again showed
complete agreement with the published data by Saifah
et al.3 for the natural product. More important, the
congruence of the analytical data of synthesized 5 and the
published data for so-called hemileptagline is the proof, that
Greger et al. had isolated in fact aglaidithioduline (5). To
§
Working conditions were not optimized, because we only needed enough
material to prove our suspicion that the originally proposed hemileptaglin
structure 2 is incorrect.
Obtained according to a literature method by Krapcho and Kuell6. We
found it more convenient to use THF instead of the proposed 1,4-dioxane
k
{ as a solvent.
Care has been taken to isolate 2 definitely as the free amine and not in
protonated form, because the salt would of course show different
chemical shifts in the NMR spectra.

R. Detterbeck, M. Hesse / Tetrahedron 58 (2002) 6887–6893
6889
Scheme 3. (a) NEt3, CH2Cl2, rt, 15 min, 92%. (b) aq. 1N HCl, EtOH, 408, 3 h, 60%. (c) EDC, CH2Cl2, rt, 12 h, 77%. (d) DCC, CH2Cl2, rt, 8 h, 60%. (e) NEt3,
CH2Cl2, rt, 15 min, 86%.
make a long story short: both authors isolated the same three
alkaloids from two different Aglaia species (Scheme 1).
Synthesis of grandiamide C (14) (Scheme 3) demanded
access to 3-hydroxy-2-methylene butanoic acid (19).
Amberg and Seebach8 reported a preparation of this
compound, where they took advantage from a DABCO‡‡
catalyzed Baylis–Hillman reaction between acetaldehyde
and methyl acrylate followed by an ester hydrolysis. DCC
mediated coupling of acid 19 and amine 15 allowed
straightforward isolation of racemic 14. The modest yield
originated from some purification problems because of
coelution of the urea by-product. Again, spectroscopic data
comparison corroborated the structure elucidation by Inada
et al. Attention should be drawn to the fact, that grandiamide
C (14) features a centre of chirality, so one might expect its
occurrence in nature in an enantiopure form. Nevertheless,
the Japanese authors reported about [a ]D¼^0 (CHCl3) and
therefore they claimed, that natural 14 has to be a racemate.
On the other hand, it may be also possible—admittedly
more unlikely—that 14 shows no optical rotation at the
[a ]D-line by coincidence.
To verify the non-sulphur containing aglaiduline structure
(4) by synthesis, we used the simple reaction of phenyl-
acetyl chloride (10) with putrescine (12) in presence of base
(Scheme 2). The product (4) was obtained in good yield and
showed the expected spectroscopical properties, therefore
proving the structure (Scheme 2).
Also in 2000, Inada et al.7 published the isolation and
structure elucidation by spectroscopic means of three
previously unknown putrescine bisamides from the leaves
of A. grandis Korth. and named them as grandiamide A, B
(13), and C (14). Common feature of substances 13 and 14
(as well as the later discussed alkaloids) is a N-mono-
cinnamoyl-putrescine residue, so we found it convenient to
prepare the building block 15 first (Scheme 3) and acylate it
afterwards in an appropriate manner. N-mono-Boc-
putrescine (9, Scheme 1) once more served as a starting
material, which could be easily acylated with cinnamoyl
chloride (16) to obtain 17. Removal of the Boc-protecting
group under acidic conditions yielded in desired 15. For
grandiamide B (13, Scheme 3) we decided to use EDCpp as
an amide-coupling reagent, instead of the more common
DCC,†† in order to profit from the better water solubility of
the inevitable urea by-product of the coupling reaction and
the therefore easier purification process. As an acyl
component served commercially available tiglic acid (18),
so the synthesis of 13 could be performed in one single step.
Comparison of the analytical data of 13 with the published
one of the natural product revealed no difference.
An earlier publication of Saifah et al.9 dealt with the
isolation and structure elucidation of pyramidatine (20)
from the leaves of A. pyramidata Hance. This alkaloid also
contains the N-mono-cinnamoyl-putrescine residue, so with
15 as a building block, its synthesis could be achieved
quickly by reaction with benzoyl chloride (21) according to
Scheme 3. As expected, the synthesized material was found
to be identical to the natural one described in literature
(Scheme 3).
Once more, Greger et al.10 published in 2001 the isolation
and structure elucidation of two previously unknown
putrescine bisamides: secopiriferine (22) and secoodorine
(1) (Scheme 4). In this case, they used a leaf extract of A.
p p
EDC¼1-[3-(dimethylamino)propyl]-3-ethylcarbodiimide-
hydrochloride.
DCC¼1,3-dicyclohexylcarbodiimide.
††
‡‡
DABCO¼1,4-diazabicyclo[2.2.2]octane.

6890
R. Detterbeck, M. Hesse / Tetrahedron 58 (2002) 6887–6893
Scheme 4. (a) NEt3, CH2Cl2, rt, 15 min, 95%. (b) DCC, HOBt, EtOAc, rt, 1 h, 66%.
gracilis A. C. Smith, a plant originated from Fijian islands.
Whereas 22 is an achiral molecule, 1 possess a stereogenic
centre. The Austrian authors reported an optical rotation of
[a ]D¼þ11 (CHCl3) for compound 1, but they failed to
determine the absolute configuration. Therefore, we decided
to synthesize 1 in an enantiopure form. DCC mediated
coupling of commercially available (þ)-(S )-2-methyl
butanoic acid (23) with 15 in the presence of HOBt§§—
which is well established especially in peptide chemistry to
avoid racemization—enabled the synthesis of optically
active 1. Besides the identity of measured spectra for
synthetic and natural alkaloid, the comparison of optical
rotations, [a]D¼þ12.3 (CHCl3) for the synthetic product,
revealed the (S)-configuration of naturally occurring (þ)-
secoodorine (1).
graphy (TLC) on Merck precoated plates Kieselgel 60 F254.
All extracts were dried before evaporation over MgSO4,
unless otherwise stated. Column chromatography (CC):
Kieselgel 60 (230–400 mesh ASTM) from Merck. Melting
points (mp): Mettler FP52. IR spectra were measured in
CHCl3 (Fluka for IR spectroscopy) unless otherwise stated:
Perkin–Elmer 781. Optical rotations [a]D in CHCl3 (Fluka
for IR spectroscopy): Perkin–Elmer 241 polarimeter. NMR
spectra: 1H NMR: Bruker ARX-300 (300 MHz); 13C NMR:
Bruker ARX-300 (75 MHz); chemical shifts d in ppm rel. to
TMS as internal standard. MS: Finnigan SSQ-700 for
Chemical Ionization (CI) with NH3, and Finnigan TSQ-700
for electrospray ionization (ESI); m/z (rel. intensity in %).
4.1.1. tert-Butyl (E )-[4-(3-methylthio-prop-2-enoyl-
amino)-butyl]-carbamate (8). To a suspension of 0.15 g
(1.27 mmol) 6kk in 10 ml CH2Cl2 was added dropwise
0.30 g (2.54 mmol, 0.18 ml) SOCl2 at room temperature.
After complete dissolution, the mixture was stirred for
20 min, before all volatile compounds were removed in
vacuo. The residue was taken up in 10 ml CH2Cl2 and again
completely evaporated to remove the excess of SOCl2. The
crude acid chloride 7 was dissolved in 10 ml CH2Cl2
without any further purification and a mixture of 0.15 g
(1.5 mmol; 0.21 ml) NEt3 and 0.28 g (1.5 mmol) 9{{ was
added. After 10 min stirring at room temperature, the
mixture was poured into aq. NaCl solution and extracted
with CH2Cl2. After removing the solvent in vacuo,
purification of the residue by CC (SiO2; CH2Cl2/MeOH
20:1) yielded 0.14 g (38%) 8 as a yellowish solid. IR n
(cm21): 3450m, 3320m, 3000m, 2980m, 2860w, 1710s,
1610s, 1580s, 1510s, 1390w, 1370m, 1330w, 1250s, 1170s,
1000w, 940m, 840w. 1H NMR (CDCl3, 300 MHz): 7.58 (d,
J¼14.7 Hz, 1H), 6.40 (br., 1H), 5.72 (d, J¼14.7 Hz, 1H),
4.90 (br. 1H), 3.31 (q, J¼6.1 Hz, 2H), 3.11 (q, J¼6.1 Hz,
2H), 2.31 (s, 3H), 1.60–1.48 (m, 4H), 1.44 (s, 9H). 13C
NMR (CDCl3, 75 MHz): 164.7 (s), 156.1 (s), 142.0 (d),
116.1 (d), 79.0 (s), 40.0 (t), 39.0 (t), 28.3 (q), 27.5 (t), 26.5
(t), 14.4 (q). CIMS/NH3: 289 (100, [Mþ1]þ), 233 (85,
[M2CH3S–CHvCHþ18]þ), 189 (62, [M2Bocþ1]þ).
We closed the series of putrescine bisamide synthesis with
above-mentioned secopiriferine (22). The straightforward
reaction of 15 with isobutyryl chloride (24) allowed the
isolation of 22 in almost quantitative yield (Scheme 4).
Spectra comparison also proved its structure (Scheme 4).
3. Conclusion
The structures of nine newly published naturally occurring
putrescine bisamides could be confirmed by synthesis. For
one of them—hemileptagline—the originally proposed
structure 2 has been revised and the identity to this
compound with aglaidithioduline (5) has been proven.
Furthermore, two compound names in this field—leptagline
and aglanthine—were found to be redundant, because of the
identity of these compounds with the earlier described
aglaithioduline (3) and aglaiduline (4), respectively. The
structures of the other compounds have been successfully
verified and the absolute configuration of (þ)-(S)-seco-
odorine (1) could be established.
4. Experimental
4.1.2. (E )-N-(4-Aminobutyl)-3-methylthio-prop-2-
enamide (2). To a solution of 0.10 g (0.347 mmol) 8 in
10 ml toluene was added 1 ml TFA at room temperature.
After 15 min stirring, the solution was poured in aq. 2N
NaOH and extracted exhaustively with CH2Cl2. The organic
phase was dried (Na2SO4) before evaporating to afford
4.1. General remarks
All commercially available reagents were used without
further purification. Solvents were either puriss. p.a. grade
(Fluka) or distilled prior to use. The application of a
nitrogen atmosphere was not necessary for the described
reactions. They were monitored by thin layer chromato-
kk
{{ Prepared according to the published procedure by De Medeiros et al.5
§§
HOBt¼1-hydroxybenzotriazole.
Synthesized from putrescine (12) by a procedure of Krapcho and Kuell.6

R. Detterbeck, M. Hesse / Tetrahedron 58 (2002) 6887–6893
6891
52 mg (80%) of 2 as a colorless oil. IR n (cm21): 3440m,
3300m, 2990m, 2930s, 2860m, 1650s, 1580s, 1510s,
1435m, 1330m, 1250m, 1200m, 1090w, 990w, 940m,
some product, the main charge was obtained by complete
evaporation and subsequent recrystallisation from EtOH/
hexane. Together, 14.9 g (81%) of 4 could be collected as
colorless crystals. Mp: 175.5–176.58. IR n (cm21): 3335w,
3290m, 3060w, 3000w, 2940w, 2880w, 1660s, 1600w,
1520m, 1495w, 1470w, 1455w, 1350w, 1320w, 1270w,
1160w, 1080w, 695w. 1H NMR (CD3OD, 300 MHz): 7.35–
7.20 (m, 10H), 3.47 (s, 4H), 3.15 (m, 4H), 1.48 (m, 4H). 13C
NMR (CD3OD, 75 MHz): 174.2 (s), 137.0 (s), 130.2 (d),
129.7 (d), 128.0 (d), 44.1 (t), 40.3 (t), 27.9 (t). 1H NMR (d 6-
DMSO, 300 MHz): 8.16 (t, J¼5.1 Hz, 2H), 7.43–7.28 (m,
10H), 3.50 (s, 4H), 3.15 (m, 4H), 1.58–1.40 (m, 4H). 13C
NMR (d 6-DMSO, 75 MHz): 169.8 (s), 136.5 (s), 128.8 (d),
128.0 (d), 126.1 (d), 42.3 (t), 38.2 (t), 26.4 (t). CIMS/NH3:
342 (9, [Mþ1]þ), 325 (100, [Mþ1]þ).
1
840m. H NMR (CD3OD, 300 MHz): 7.54 (d, J¼14.7 Hz,
1H), 5.81 (d, J¼4.8 Hz, 1H), 3.25 (t, J¼6.6 Hz, 2H), 2.68 (t,
J¼6.85 Hz, 2H), 2.34 (s, 3H), 1.60–1.45 (m, 4H). 13C NMR
(CD3OD, 75 MHz): 167.5 (s), 143.6 (d), 117.0 (d), 42.2 (t),
40.3 (t), 30.8 (t), 28.0 (t), 14.5 (q). CIMS/NH3: 189
([Mþ1]þ).
4.1.3. (E )-3-Methylthio-N-[4-(3-methylthio-prop-2-
enoylamino)butyl]-prop-2-enamide (5aglaidithioduline,
5). 3-Methylthio-2-propenoic acid chloride (7) was prepared
as described for 8 and then a solution of putrescine (12)
(ratio 7/12 1:2) in CH2Cl2 added at room temperature. After
1 h, the mixture was poured into aq. NaCl solution and
extracted with CH2Cl2. After separation by CC (SiO2;
CH2Cl2/MeOH 19:1) the product 5 was obtained as a
colorless microcrystalline solid (30%). Mp: 177–1798. IR
(KBr) n (cm21): 3317s, 2943m, 2868m, 1626s, 1577s,
1527s, 1473m, 1433w, 1325m, 1258m, 1240m, 1192m,
1033w, 994m, 937m, 851m, 817w, 742w, 704m 634m,
4.1.6. tert-Butyl N-[4-(3-phenyl-prop-2-enoylamino)-
butyl]carbamate (N-Boc-N0-cinnamoyl-putrescine, 17).
A solution of 7.16 g (43 mmol) cinnamoyl chloride (16) in
30 ml CH2Cl2 was dropped into a mixture of 8.1 g
(43 mmol) 9 and 5.06 g (50 mmol; 7.0 ml) NEt3 in 70 ml
CH2Cl2. Cooling was accomplished by an ice-bath. After
2 h, the mixture was poured into diluted aq. NaOH (to
hydrolyse any unreacted 16) and extracted with CH2Cl2.
After evaporation, 13.13 g (92%) of 17 were obtained as a
yellowish solid in sufficient purity. IR n (cm21): 3440m,
3300m, 2980m, 2930m, 2860w, 1700s, 1660s, 1625s,
1580w, 1500s, 1450w, 1390w, 1365m, 1330w, 1225s,
1
524w. H NMR (CD3OD, 300 MHz): 7.54 (d, J¼14.7 Hz,
2H), 5.81 (d, J¼14.8 Hz, 2H), 3.30–3.18 (m, 4H), 2.34 (s,
6H), 1.62–1.50 (m, 4H). 13C NMR (CD3OD, 75 MHz):
167.5 (s), 143.6 (d), 116.9 (d), 40.2 (t), 28.1 (t), 14.5 (q). 1H
NMR (d 6-DMSO, 300 MHz): 7.84 (t, J¼5.6 Hz, 2H), 7.44
(d, J¼14.8 Hz, 2H), 5.88 (d, J¼14.8 Hz, 1H), 3.22–3.12
(m, 4H), 2.37 (s, 6H), 1.54–1.45 (m, 4H). 13C NMR (d 6-
DMSO, 75 MHz): 163.4 (s), 139.4 (d), 117.3 (d), 38.1 (t),
26.6 (t), 13.7 (q). CIMS/NH3: 289 ([Mþ1]þ). Elemental
analysis calcd for C12H20N2O2S2: C, 49.97; H, 6.99; N,
9.71; S, 22.23; found: C, 49.72; H, 6.71; N, 9.66; S, 21.98.
1
1160s, 985w, 975m, 855w. H NMR (CDCl3, 300 MHz):
7.61 (d, J¼15.7 Hz, 1H), 7.47 (m, 2H), 7.33 (m, 3H), 6.45
(d, J¼15.7 Hz, 1H), 6.30 (br., 1H), 4.70 (br., 1H), 3.41 (q,
J¼5.9 Hz, 2H), 3.14 (m, 2H), 1.75–1.50 (m, 4H), 1.44 (s,
9H). 13C NMR (CDCl3, 75 MHz): 166.0 (s), 156.1 (s), 140.5
(d), 134.9 (s), 129.4 (d), 128.7 (d), 127.6 (d), 120.9 (d), 79.2
(s), 40.0 (t), 39.3 (t), 28.3 (q), 27.6 (t), 26.5 (t).
4.1.4. 3-Methylthio-N-(4-phenylacetylaminobutyl)-prop-
2-enamide (5aglaithioduline, 3). Acylation of 2 (0.10 g;
0.53 mmol) in 15 ml CH2Cl2 with 0.123 g (0.8 mmol; 1 ml)
phenylacetyl chloride (10) in the presence of 0.215 g
(2.1 mmol; 0.3 ml) NEt3 as a base, yielded after 30 min
stirring at room temperature in a reaction mixture, which
was taken up in aq. NaCl solution and extracted with
CH2Cl2. After drying and subsequent evaporation of the
organic phase a raw product resulted, which was purified by
CC (SiO2; CH2Cl2/MeOH 19:1) to give 0.127 g (78%) of 3
as a colorless solid. Mp: 128–1308. IR n (cm21): 3440m,
3300m, 2990m, 2920m, 2860w, 1715m, 1660s, 1580s,
1510s, 1435w, 1330w, 1250m, 990w, 940m, 840w. 1H
NMR (CD3OD, 300 MHz): 7.54 (d, J¼14.7 Hz, 1H), 7.32–
7.18 (m, 5H), 5.80 (d, J¼4.7 Hz, 1H), 3.48 (s, 2H), 3.30–
3.15 (m, 4H), 2.34 (s, 3H), 1.45–1.60 (m, 4H). 13C NMR
(CD3OD, 75 MHz): 174.2 (s), 167.5 (s), 143.5 (d), 137.2 (s),
130.2 (d), 129.7 (d), 128.0 (d), 116.9 (d), 44.1 (t), 40.3 (t),
40.1 (t), 28.0 (t), 27.9 (t), 14.5 (q). CIMS/NH3: 307
([Mþ1]þ).
4.1.7. N-(4-Aminobutyl)-3-phenyl-prop-2-enamide (N-
mono-cinnamoyl-putrescine 15). To a solution of 10.0 g
(31.4 mmol) 17 in 20 ml EtOH was added aq. 1N HCl
(100 ml) and the mixture warmed to 408 for 3 h. Afterwards,
the solution was washed with Et2O and the organic phase
discarded. The aq. phase was first saturated with NaCl and
then aq. NaOH was added to get a strong basic solution,
which was extracted with CH2Cl2. After separating the
organic phase, drying with Na2SO4 and evaporating resulted
in 4.1 g (60%)ppp 15 as a viscous yellow oil, which could be
used without any further purification. IR n (cm21): 3420m,
3290m, 3060w, 2980m, 2930s, 2860m, 1660s, 1620s,
1575w, 1510s, 1450m, 1330m, 1300w, 1280w, 1220m,
1090w, 1070w, 975m, 855m, 660w. 1H NMR (CDCl3,
300 MHz): 7.61 (d, J¼15.7 Hz, 1H), 7.47 (m, 2H), 7.40–
7.25 (m, 3H), 6.72 (br.s, 1H), 6.45 (d, J¼15.7 Hz, 1H), 3.38
(q, J¼5.9 Hz, 2H), 2.73 (t, J¼6.5 Hz, 2H), 1.75–1.40 (m,
6H). 13C NMR (CDCl3, 75 MHz): 165.9 (s), 140.4 (d),
134.9 (s), 129.4 (d), 128.7 (d), 127.6 (d), 121.0 (d), 41.6 (t),
39.5 (t), 30.7 (t), 27.0 (t).
4.1.5. 2-Phenyl-N-(4-phenylacetylaminobutyl)acetamide
(5aglaiduline, 4). In a solution of 5.0 g (56.7 mmol)
putrescine (12) and 14.34 g (141 mmol; 19.75 ml) NEt3 in
100 ml CH2Cl2 18.4 g (119 mmol; 15.8 ml) phenylacetyl
chloride (10) was dropwise introduced. Cooling (ice-bath)
was necessary. After 2 h stirring, the mixture was poured
into diluted aq. HCl and extracted with CH2Cl2. Concen-
tration of the organic phase in vacuo led to crystallisation of
4.1.8. (E )-2-Methyl-N-(4-[(E )-3-phenyl-2-propenoyl]-
aminobutyl)-2-butenamide (5grandiamide B, 13). To a
suspension of 0.37 g (1.67 mmol) 15 in 30 ml CH2Cl2 was
added EDC (0.349 g; 1.82 mmol) and afterwards tiglic acid
ppp
Some product was lost during the extraction process.

6892
R. Detterbeck, M. Hesse / Tetrahedron 58 (2002) 6887–6893
(18; 0.167 g; 1.67 mmol) in 10 ml CH2Cl2 was introduced.
Stirring at room temperature was continued for 12 h before
the mixture was taken up in aq. 1N HCl. Extraction with
CH2Cl2 followed by washing with aq. NaCl solution yielded
after evaporation directly in 0.39 g (77%) 13 as a colorless
powder. Mp: 100.5–101.58. IR n (cm21): 3450m, 3320m,
3300s, 3940m, 3860w, 1670s, 1630s, 1580w, 1515s, 1450m,
1380w, 1335m, 1280w, 1170w, 1090w, 990w, 980m, 855w,
(86%) of 20 as a colorless powder. Mp: 166.5–167.58. IR n
(cm21): 3442w, 3310s, 3000w, 2940w, 2870w, 1705w,
1660s, 1630s, 1620s, 1580m, 1520s, 1480m, 1450w, 1325w,
1280w, 985m, 975m. 1H NMR (d 6-DMSO, 300 MHz): 8.50
(t, J¼5.5 Hz, 1H), 8.15 (t, J¼5.4 Hz, 1H), 7.9 (m, 2H),
7.60–7.35 (m, 9H), 6.67 (d, J¼15.8 Hz, 1H), 3.33 (q,
J¼6.4 Hz, 2H), 3.26 (q, J¼6.5 Hz, 2H), 1.70–1.45 (m, 4H).
13C NMR (d 6-DMSO, 75 MHz): 166.0 (s), 164.7 (s), 138.2
(d), 134.8 (s), 134.6 (s), 130.8 (d), 129.2 (d), 128.8 (d),
128.1 (d), 127.3 (d), 127.0 (d), 122.3 (d), 38.8 (t), 38.3 (t),
26.6 (t). CIMS/NH3: 323 ([Mþ1]þ).
1
660w. H NMR (CDCl3, 300 MHz): 7.60 (d, J¼15.7 Hz,
1H), 7.45 (m, 2H), 7.28 (m, 3H), 6.95 (t, J¼5.3 Hz, 1H),
6.53 (d, J¼15.7 Hz, 1H), 6.43 (q, J¼6.9 Hz, 1H), 6.25 (br.t,
1H), 3.48–3.28 (m, 4H), 1.84 (s, 3H), 1.72 (d, J¼6.9 Hz,
3H), 1.68–1.52 (m, 4H). 13C NMR (CDCl3, 75 MHz): 169.7
(s), 166.2 (s), 140.3 (d), 134.9 (s), 131.7 (s), 130.6 (d), 129.4
(d), 128.6 (d), 127.6 (d), 121.2 (d), 39.2 (t),††† 27.1 (t), 26.7
(t), 13.8 (q), 12.3 (q). CIMS/NH3: 301 ([Mþ1]þ). Elemental
analysis calcd for C18H24N2O2: C, 71.97; H, 8.05; N, 9.33;
found: C, 71.72; H, 8.31; N, 9.07.
4.1.12. (1)-(S )-N-[4-(2-Methyl-butanoylamino)butyl]-3-
phenyl-prop-2-enamide (5(1)-(S )-secoodorine, 1). To a
solution of 0.14 g (1.37 mmol) commercially available (þ)-
(S )-2-methylbutanoic acid (23) in 10 ml EtOAc were
subsequently added 0.186 g (1.37 mmol) HOBt and
0.283 g (1.37 mmol) DCC. After 5 min, 0.30 g
(1.37 mmol) 15 in 10 ml EtOAc were introduced at room
temperature. This resulted in a white suspension. After 1 h
stirring, the precipitated colorless solid was filtered off and
discarded after washing with EtOAc. The filtrate was
evaporated, and the residue purified by CC (SiO2; CH2Cl2/
MeOH 15:1): 0.27 g (66%) of 1 as colorless crystals. Mp:
146.5–147.08. [a]D¼þ12.38 (c¼0.33, CHCl3). IR n
(cm21): 3445m, 3320br., 2995m, 2960m, 2930m, 2870m,
1660s, 1625s, 1580w, 1510s, 1450m, 1370w, 1335m,
1300w, 1280w, 1090w, 1070w, 985m, 975m, 910w,
4.1.9. (6)-3-Hydroxy-2-methylen-butanoic acid (19).
Prepared according to Ref. 8. IR n (cm21): 3600–2400br.,
2980w, 1690s, 1625m, 1400m, 1270m, 1170m, 1090s,
1035w, 1015w, 965m, 925w, 870w. 1H NMR (CDCl3,
300 MHz): 7.20 (br., 2H), 6.37 (s, 1H), 5.94 (s, 1H), 3.51 (q,
J¼7.0 Hz, 1H), 1.41 (d, J¼7.0 Hz, 3H). 13C NMR (CDCl3,
75 MHz): 170.9 (s), 142.7 (s), 126.4 (t), 66.8 (d), 21.9 (q).
4.1.10. 2-(1-Hydroxyethyl)-N-[4-(3-phenyl-prop-2-enoyl-
amino)butyl]prop-2-enamide (5grandiamide C, 14). To
a suspension of 0.365 g (1.67 mmol) 15 in 30 ml CH2Cl2
were subsequently added 0.345 g (1.67 mmol) DCC and
0.17 g (1.67 mmol) 19. After a short induction period, the
precipitation of a colorless solid (dicyclohexyl urea) could
be observed. Stirring was continued at room temperature for
8 h. Afterwards, the solid was filtered off and discarded. The
clear filtrate was concentrated in vacuo and the residue
purified by CC (SiO2; CH2Cl2/MeOH 20:1) which yielded
in 0.32 g (60%) 14 as a colorless solid. Mp: 104.5–106.58.
IR n (cm21): 3450m, 3320m, 3000m, 2940m, 2860w,
1665s, 1620s, 1580w, 1520s, 1450m, 1370w, 1330m,
1280w, 1100w, 1020w, 975m. 1H NMR (CDCl3,
300 MHz): 7.57 (d, J¼15.7 Hz, 1H), 7.53 (t, J¼7.9 Hz,
1H), 7.45 (m, 2H), 7.28 (m, 3H), 7.06 (t, J¼5.6 Hz, 1H),
6.53 (d, J¼15.7 Hz, 1H), 5.83 (s, 1H), 5.43 (s, 1H), 4.60 (q,
J¼6.5 Hz, 1H), 3.90 (br. OH, 1H), 3.40–3.20 (m, 4H),
1.65–1.50 (m, 4H), 1.37 (d, J¼6.5 Hz, 3H). 13C NMR
(CDCl3, 75 MHz): 168.1 (s), 166.5 (s), 146.3 (s), 140.6 (d),
134.7 (s), 129.5 (d), 128.7 (d), 127.6 (d), 120.9 (d), 119.3 (t),
68.5 (d), 39.2 (t), 38.9 (t), 26.7 (t), 14.0 (q). CIMS/NH3: 317
([Mþ1]þ). Elemental analysis calcd for C18H24N2O3: C,
68.33; H, 7.65; N, 8.85; found: C, 68.10; H, 7.65; N, 8.79.
1
855w. H NMR (CDCl3, 300 MHz): 7.61 (d, J¼15.7 Hz,
1H), 7.50–7.46 (m, 2H), 7.37–7.31 (m, 3H), 6.66 (br.t, 1H),
6.50 (d, J¼15.7 Hz, 1H), 6.05 (br.t, 1H), 3.42–3.38 (m, 2H),
3.35–3.28 (m, 2H), 2.13 (hex., J¼7.4 Hz, 1H), 1.69–1.55
(m, 1H), 1.61–1.59 (m, 4H), 1.50–1.36 (m, 1H), 1.13 (d,
J¼6.9 Hz, 3H), 0.90 (t, J¼7.4 Hz, 3H). 13C NMR (CDCl3,
75 MHz): 176.8 (s), 166.2 (s), 140.5 (d), 134.9 (s), 129.4 (d),
128.7 (d), 127.6 (d), 121.0 (d), 43.0 (d), 39.2 (t), 38.8 (t),
27.2 (t), 27.1 (t), 26.7 (t), 17.5 (q), 11.8 (q). ESI-MS: 325
([MþNa]þ). Elemental analysis calcd for C18H26N2O2: C,
71.49; H, 8.67; N, 9.26; found: C, 71.52; H, 8.49; N, 9.32.
4.1.13. N-(4-Isobutyrylaminobutyl)-3-phenyl-prop-2-
enoyl-amide (5secopiriferine, 22). Isobutyryl chloride
(24; 0.146 g; 0.145 ml) was added dropwise to a mixture of
0.30 g (1.37 mmol) 15 and 0.21 g (2.1 mmol; 0.29 ml) NEt3
in 15 ml CH2Cl2 at room temperature. After 15 min, the
reaction mixture was poured in diluted aq. HCl and
extracted with CH2Cl2. Removing the solvent in vacuo
and purification of the residue by CC (SiO2; CH2Cl2/MeOH
15:1) yielded in 0.376 g (95%) of 22 as colorless crystals.
Mp: 161.0–161.58. IR n (cm21): 3440m, 3295br., 2995m,
2960m, 2940m, 2870w, 1660s, 1625s, 1580w, 1510s,
1470w, 1450m, 1365w, 1335m, 1300w, 1280w, 1090w,
1
4.1.11. N-[4-(3-Phenyl-prop-2-enoylamino)butyl]benza-
mid (5pyramidatine, 20). To a mixture of 0.50 g
(2.30 mmol) 15 and 0.348 g (3.44 mmol; 0.48 ml) NEt3 in
30 ml EtOAc was added benzoylchloride (0.32 g; 2.3 mmol;
0.27 ml) at room temperature. After 15 min, the mixture
was poured into diluted aq. HCl and extracted with CH2Cl2.
After removing the solvent in vacuo, the residue was
purified by CC (SiO2; CH2Cl2/MeOH 19:1) to yield 0.64 g
985w, 975m, 910w, 855w. H NMR (CDCl3, 300 MHz):
7.62 (d, J¼15.6 Hz, 1H), 7.51–7.48 (m, 2H), 7.39–7.34 (m,
3H), 6.45 (d, J¼15.6 Hz, 1H), 6.33 (br.t, 1H), 5.86 (br.t,
1H), 3.43 (q-like m, 2H), 3.30 (q-like m, 2H), 2.37 (hept.,
J¼6.9 Hz, 1H), 1.63–1.56 (m, 4H), 1.16 (d, J¼6.9 Hz, 6H).
13C NMR (CDCl3, 75 MHz): 177.3 (s), 166.1 (s), 140.7 (d),
134.8 (s), 129.5 (d), 128.67 (d), 127.7 (d), 120.8 (d), 39.2 (t),
38.8 (t), 35.6 (d), 27.1 (t), 26.6 (t), 19.5 (q). ESI-MS: 311
(100, [MþNa]þ), 289 (10, [MþH]þ). Elemental analysis
calcd for C17H24N2O2: C, 70.80; H, 8.39; N, 9.71; found: C,
70.67; H, 8.26; N, 9.64.
†††
The carbon signals of both CH2-groups next to nitrogen atoms show a
signal overlapping by 39.2 ppm.

R. Detterbeck, M. Hesse / Tetrahedron 58 (2002) 6887–6893
6893
3. Saifah, E.; Suttisri, R.; Shamsub, S.; Pengsuparp, T.; Lipipun,
V. Phytochemistry 1999, 52, 1085–1088.
Acknowledgments
4. Ikegami, F.; Sekine, T.; Aburada, M.; Fujii, Y.; Komatsu, Y.;
Murakoshi, I. Chem. Pharm. Bull. 1989, 37, 1932–1933.
5. Medeiros, E. F. D.; Herbert, J. M.; Taylor, R. J. K. J. Chem.
Soc., Perkin Trans. 1 1991, 2725–2730.
We wish to thank Mr N. Agorastos for his help on this work
within the scope of a practica and Mr A. Guggisberg for
helpful discussions. Furthermore, we thank the Swiss
National Science Foundation for generous support.
6. Krapcho, A. P.; Kuell, C. S. Synth. Commun. 1990, 20,
2559–2564.
7. Inada, A.; Shono, K.; Murata, H.; Inatomi, Y.; Darnaedi, D.;
Nakanishi, T. Phytochemistry 2000, 53, 1091–1095.
8. Amberg, W.; Seebach, D. Chem. Ber. 1990, 123, 2429–2438.
9. Saifah, E.; Puripattanavong, J.; Likhitwitayawuid, K.; Cordell,
G. A.; Chai, H.; Pezzuto, J. M. J. Nat. Prod. 1993, 56,
473–477.
References
1. Bienz, S.; Detterbeck, R.; Ensch, C.; Guggisberg, A.;
¨
Hausermann, U.; Meisterhans, C.; Wendt, B.; Werner, C.;
Hesse, M. Putrescine, Spermidine, Spermine and Related
Polyamin Alkaloids. In The Alkaloids; Cordell, G. A., Ed.;
Academic: New York, 2002; in press.
10. Greger, H.; Pacher, T.; Brem, B.; Bacher, M.; Hofer, O.
Phytochemistry 2001, 57, 57–64.
2. Greger, H.; Pacher, T.; Vajrodaya, S.; Bacher, M.; Hofer, O.
J. Nat. Prod. 2000, 63, 616–620.