Cyclopeptide alkaloids from Paliurus ramossisimus

Article abstract of DOI:10.1021/np000136a

Seven zizyphine A-type cyclopeptide alkaloids were isolated from the roots of Paliurus ramossisimus by the combination of centrifugal partition chromatography and conventional separation methods. The novel structures of paliurines A-F (1-6) were characterized and established on the basis of MS and elaborate NMR spectral analyses. Terminal dipeptide stereochemistry was confirmed by correlation with the synthetic dipeptides via comparison of their ¹³C NMR data.

Full text of DOI:10.1021/np000136a

1338
J . Nat. Prod. 2000, 63, 1338-1343
Cyclop ep tid e Alk a loid s fr om P a liu r u s r a m ossisim u s
Hui-yi Lin,† Chung-Hsiung Chen, Bih-J ing You, Karin C. S. Chen Liu, and Shoei-Sheng Lee*
School of Pharmacy, College of Medicine, National Taiwan University, Taipei 100, Taiwan, Republic of China
Received March 29, 2000
Seven zizyphine A-type cyclopeptide alkaloids were isolated from the roots of Paliurus ramossisimus by
the combination of centrifugal partition chromatography and conventional separation methods. The novel
structures of paliurines A-F (1-6) were characterized and established on the basis of MS and elaborate
NMR spectral analyses. Terminal dipeptide stereochemistry was confirmed by correlation with the
synthetic dipeptides via comparison of their 13C NMR data.
Paliurus ramossisimus (Lour.) Poir (Rhamnaceae), a
plant growing along the west coast of Taiwan Island, has
been used as a pain reliever in the treatment of toothache
and stomach ache. Previous investigations of its chemical
constituents yielded the triterpenes ceanothic acid and
betulinic acid as major components1,2 and cyclopeptide
alkaloids as very minor constituents. In the present paper
we report the isolation and structural characterization of
seven cyclopeptide alkaloids and the determination of
stereochemistry for the terminal dipeptide residues.
protons H-1 (ca. δ 5.90, d), H-2 (ca. δ 6.93, dd), and H-3
(ca. δ 8.40, d), appearing as an AMX system,6 and H-12
(ca. δ 6.65, d), H-12′ (ca. δ 6.77, dd) and H-13′ (ca. δ 6.86,
d), appearing as an ABX system; the coupling pattern of
H-8 (ca. δ 4.39, d) and H-9 (ca. δ 5.40, ddd or dt) specific
for the (3S)-â-hydroxyproline unit; and, finally, two methyl
signals (ca. δ 1.00, d; ca. δ 0.90, t) in the isoleucine unit.
The coupling relationships of these protons were identified
by a COSY-45 spectra. These data indicate compounds 1-7
to be zizyphine A-type 13-membered cyclopeptide alka-
loids.7 Further analysis of their 13C NMR spectra revealed
C-11 (s) and C-14 (s) near δ 151, C-1 (d) near δ 107, C-2
(d) near δ 122, and C-9 (d) in the â-hydroxyproline near δ
77.0. The mass spectra of these compounds, except that of
3, revealed characteristic fragment ions at m/z 374,7 which
supported this type of structure having an isoleucine unit
as the ring bond amino acid.
Resu lts a n d Discu ssion
The methanolic extract of the roots was processed in a
manner to facilitate the separation of alkaloids. As the
cyclopeptide alkaloids are weak bases and less polar
relative to alkaloids such as aporphines, the alkaloid-
containing fraction usually mixes with other components
having similar polarity. Thus, a focusing procedure, that
is, passing this fraction through a Sephadex LH-20 column,
was followed, which effectively separated the major alka-
loid fraction. The components in this alkaloid fraction had
very similar polarity and were separated previously by
repeated column chromatography and preparative TLC. We
found that separation could be facilitated by the combina-
tion of partition chromatography with conventional meth-
ods. We observed that the technique of applying acidic
mobile phase would disturb the elution order of these
components, not following their polarity as revealed on TLC
plate. Thus, two components with different polarity to some
extent were usually found to elute in the same fraction,
allowing easy separation simply by column chromatogra-
phy or TLC.
Using this approach, seven 13-membered cyclopeptide
alkaloids (1-7) were isolated. These seven compounds
possess common IR absorptions at 3285-3400 cm-1
(-CONH), 2780-2790 cm-1 (-NCH3), 16800-1690 and
16300-1655 cm-1 (-CONH), 1625 cm-1 for a cis 1,2-
disubstituted olefin, and 12300-1240 cm-1 for aryl alkyl
ether functions. UV showed absorption maxima at 265 and
317 nm, both typical of 13-membered zizyphine A-type
cyclopeptide alkaloids,3,4 which possess a 2,5-dioxystyry-
lamine moiety. Their CD spectra exhibited two negative
Cotton effects around 264 and 324 nm, indicating the same
5S,8S,9S configuration at the 13-membered ring moiety
present in zizyphine A, whose stereochemistry has been
confirmed by total synthesis.5 Their 1H NMR spectra
showed common signals for the 14-OMe (ca. δ 3.77); the
* To whom correspondence should be addressed. Tel./Fax: 886-2-
23916127. E-mail: shoeilee@ha.mc.ntu.edu.tw.
† Current address: Chinese Pharmaceutical Sciences and Department
of Pharmacy, China Medical College, Taichung, Taiwan, Republic of China.
10.1021/np000136a CCC: $19.00
© 2000 American Chemical Society and American Society of Pharmacognosy
Published on Web 08/11/2000

Cyclopeptide Alkaloids from Paliurus
J ournal of Natural Products, 2000, Vol. 63, No. 10 1339
Ta ble 1. 1H NMR Data for Compounds 1-7 in the Terminal Dipeptide Region (δ/ppm, J in Hz)a in CDCl3
position
1b
1(CD3OD)b
Phe
2b
3
4
5
6
7b
intermediate a.a Phe
Phe
Ile
Ile
Ile
2′
5.01 br q
(8.1)
2.93 dd
4.95 dd
(8.0, 6.9)
3.04 dd
5.00 dt
(8.2, 8.6)
2.95 m
4.53 t
(8.5)
1.70 m
4.60 dd
(8.6, 8.2)
1.74 m
4.55 dd
(8.9, 8.5)
1.77 m
3′
(14.4, 8.3)
2.94 dd
(13.6, 6.9)
2.88 dd
(14.4, 6.4)
(13.6, 8.0)
4′
1.09 & 1.38 1.08 & 1.38
0.83 t (7.5) 0.82 t (6.7)
0.78 d (6.7) 0.81 d (6.5)
1.06 & 1.44
0.84 t (7.5)
0.81 d (6.8)
5′(9′)
6′(8′)
7′
7.05 m
7.24 m
7.24 m
7.21 br d (7.1) 7.04 m
7.26 t (7.1)
7.21 m
7.24 m
7.24 m
2R′(NH)
7.05 d (8.8)
7.78 d (8.6)
7.41 d (8.9)
basic end a.a.
2′′
Ile(NMe2)
2.53 d (5.5) 2.66 d (8.8)
Ile(NMe2)
Ile(NHMe) Ile(NMe2)
2.81 d (4.6) 3.07 d (9.7) 4.30 dt
(2.6, 9.7)
Phe(NMe2) Phe(NHMe)
Phe(NMe2)
3.42 dd
(10.3, 3.4)
3.06 dd
Leu(NMe2)
2.89 dd
(8.4, 5.0)
1.40 & 1.53
4.33 ddd
(10.4, 8.8, 1.6)
2.59 m
3′′
1.76 m 1.79 m
1.71 m
1.88 m
2.58 m
(13.0, 10.3)
2.91 dd
(13.0, 3.4)
4′′
1.12 & 1.44 1.07 & 1.57
0.91 t (7.1) 0.88 t (7.3)
0.75 d (7.2) 0.77 (d, 6.7)
1.12 & 1.40 1.10 & 1.58
0.84 t (7.2) 0.87 t (7.3) 7.14 d (5.3) 7.19 br d (6.4)
1.10
5′′(9′′)
6′′(8′′)
7′′
7.04 dd (7.2, 1.2) 0.90 d (6.4)
0.87 d (6.8) 0.71 d (6.5) 7.22 m
7.22 m
7.27 m
7.27 m
2.25 s
7.17 m
7.17 m
2.39 s
0.92 d (6.5)
NMe
2.20 s
2.12 s
2.33 s
2.33 s
2.29 s
2.24 s
a
b
Signals without multiplicity were assigned from COSY-45 or HMQC spectra. Data were assigned from analysis of COSY-45, NOED,
HMQC, and HMBC spectra.
Compound 1 had molecular formula C37H51N5O6 as
deduced from FABMS ([M + H]+ at m/z 662) and analysis
of a three-bond coupling between C-14 and 14-OCH3. For
1
reference, the H and 13C NMR data of 1 measured in CD3-
1
of its 13C NMR data. The H NMR spectrum of 1 (CDCl3)
OD (Tables 1 and 2) were also assigned based on analysis
of COSY-45, NOESY, and HMBC spectra.
(Table 1) revealed signals of two additional methyls from
another isoleucine unit at δ 0.75 (d, J ) 7.2 Hz) and 0.91
(t, J ) 7.1 Hz) and five aryl protons belonging to a
phenylalanine unit between δ 7.05 and δ 7.24 (m). Thus, 1
was composed of four amino acid residues (Phe, Ile, Ile,
Pro) and a styrylamine unit. Of these, the ring amino acids
were Ile and Pro as indicated above. In addition, the 1H
NMR spectrum revealed a six-proton singlet at δ 2.20,
suggesting the terminal amino acid residue to be N,N-
dimethylated. Its mass spectrum displayed the base peak
at m/z 114, corresponding to a fragment ion typically
obtained from â-cleavage of a N,N-dimethylated isoleucine
unit. These data indicated the structure of 1 to be as shown.
Structure 1 was supported by NOED experiments. The
critical NOEs included the common enhancement of N3-H
(δ 8.40, d) upon irradiation of N6-H (δ 7.27, d) and H-12 (δ
6.65, d), respectively. No significant enhancement was
observed between mutual irradiation of H-8 (δ 4.39, d) and
H-9 (δ 5.40, dt). This observation also supported trans-
relationship between H-8 and H-9. Irradiation of the N,N-
dimethyl singlet enhanced a methyl doublet (δ 0.75, Me-
6′′), in addition to the signals at δ 2.53 (d, H-2′′, R-Η) and
δ 1.76 (m, H-3′′). This established N,N-dimethylisoleucine
to be the terminal amino acid residue. An energy-
minimized conformation of 1, which accounts for the above
NOED results, was obtained. In this conformation the
proline plane adopts a perpendicular arrangement with the
benzene ring of styrene, resulting in the closeness of H-9
and H-12. An exocyclic hydrogen bond extending between
N6-H and C-1′ keto group was observed, accounting for the
nearness of H-8 and N6-H.
Compound 2 had the molecular formula C36H49N5O6,
deduced from analysis of its 13C NMR data and FABMS
([M + H]+ at m/z 648, 14 mass units lower than that of 1).
1
Its H NMR spectrum was very similar to that of 1, except
for a three-proton N-Me singlet at δ 2.33 instead of a six-
proton NMe2 at δ 2.20 in 1. Hence, 2 was likely to be the
N-mono-demethylated analogue of 1. The mass spectrum
of 2 exhibited a base peak at m/z 100, which supported this
conlusion. The complete 1H and 13C NMR data assignments
of 2 (Tables 1 and 2) were made by analysis of its COSY-
45, HETCOR, and COLOC spectra. Among these, the
signals of amide carbons, C-4 (δ 167.1), C-7 (δ 169.9), C-1′
(δ 171.2), and C-1′′ (δ 173.3), were distinguished by their
coupling to H-3 (δ 8.40) (C-4), H-8 (δ 4.41) (C-7), H-3′ (δ
2.95) and H-2′ (δ 5.00) (C-1′), and H-2′a (CONH, δ 7.78)
(C-1′′). It is noted that the difference of N-substitution
between 1 and 2 caused the shift change for â-carbon (C-
3′′, δ 34.4 in 1 vs. δ 38.3 in 2) and amide carbon (C-1′′, δ
171.5 in 1 vs δ 173.3 in 2), due to γ-effect from extra
N-methyl group in 1.
Compound 3 had molecular formula C28H42N4O5, based
on analysis of its 13C NMR and FABMS data ([M + H]+ at
m/z 515). One more amino acid residue, in addition to the
two (Ile and Pro) in the 13-membered ring moiety, was
observed as evidenced by the presence of three amido
carbonyl signals (δ 167.3, 170.5, and 170.9) in its 13C NMR
spectrum (Table 2). Besides the similar signals for those
protons in the ring moiety as in 1, its 1H NMR spectrum
also revealed signals for an N,N-dimethyl isoleucine resi-
due, one methyl doublet (δ 0.71, H-6′′), one methyl triplet
(δ 0.87, H-5′′), and a six-proton singlet (δ 2.33, NMe2). The
presence of this terminal amino acid was also supported
by its mass spectrum, which showed the base peak at m/z
114. Based on these data, the structure of compound 3 was
assigned as shown.
Incorporating these 1H NMR assignments (Table 1) from
NOED and COSY-45 analysis, the chemical shifts of those
proton-attached carbons were assigned directly from the
analysis of a HETCOR spectrum, while those of the
quaternary carbons were assigned from an HMBC spectral
analysis (Table 2). The latter technique distinguished the
signals of C-11 (δ 151.2) and C-14 (151.6) via observation
Compound 4 had the molecular formula C37H51N5O6, as
deduced from FABMS and NMR data, the same as that of

1340 J ournal of Natural Products, 2000, Vol. 63, No. 10
Lin et al.
Ta ble 2. 13C NMR Data for Compounds 1-7 in the Terminal Dipeptide Regiona (δ/ppm, mb) in CDCl3
position
1c
1c.CD3OD
2c
3
4
5c
6
7
intermediate a.a
Phe
Phe
Phe
Ile
Ile
Ile
1′
2′
3′
4′
5′(9′)
6′(8′)
7′
171.2 s
51.0 d
39.6 t
135.6 s
129.1 d
128.8 d
127.3 d
172.0 s
53.1 d
38.7 t
137.8 s
130.4 d
129.7 d
128.0 d
171.2 s
51.1 d
39.5 t
135.7 s
129.2 d
128.8 d
127.2 d
171.6 s
53.8 d
37.6 d
24.5 t
10.7 q
15.2 q
171.6 s
53.8 d
37.5 d
24.5 t
10.9 q
15.3 q
172.0 s
53.9 d
37.6 d
24.7 t
10.8 q
15.4 q
basic end a.a.
1′′
2′′
Ile(NMe2)
171.5 s
74.4 d
Ile(NMe2)
173.3 s
74.0 d
Ile(NHMe)
173.3 s
69.9 d
Ile(NMe2)
170.9 s
68.3 d
Phe(NMe2)
172.3 s
71.0 d
Phe(NHMe)
173.7 s
66.1 d
Phe(NMe2)
171.2 s
68.0 d
Leu(NMe2)
173.7 s
67.7 d
3′′
34.4 d
34.8 d
38.3 d
34.7 d
33.1 t
38.9 t
32.2 t
37.2 t
4′′
26.9 t
11.8 q
14.5 q
26.6 t
11.2 q
15.6 q
25.3 t
11.7q
15.8 q
25.3 t
15.1 q
10.7 q
139.7 s
129.1 d
128.3 d
126.1 d
42.3 q
137.3 s
129.0 d
128.7 d
126.9 d
35.4 q
138.2 s
128.9 d
128.6 d
126.5 d
41.9 q
26.0 d
5′′(9′′)
6′′(8′′)
7′′
23.3d
22.2d
q
q
NMe
43.1 q
42.3 q
36.1 q
41.7 q
42.3 q
a
A typical 13C NMR data of the common ring nucleus, for example in 1, are as follows: δ 106.7 (d, C-1), 121.6 (d, C-2), 167.0 (s, C-4),
60.3 (d, C-5), 169.8 (s, C-7), 64.8 (s, C-8), 76.6 (d, C-9), 151.2 (s, C-11), 111.4 (d, C-12), 117.7 (d, C-12′), 124.6 (s, C-13), 114.1 (d, C-13′),
b
151.6 (s, C-14), 35.6 (d, C-15), 24.9 (t, C-16), 11.8 (q, C-17), 16.3 (q, C-18), 32.4 (t, C-19), 46.4 (t, C-20). Multiplicity was obtained from
DEPT experiments. c Data were assigned from analysis of COSY-45, NOED, NOESY, HMQC, and HMBC spectra. Both assignments
d
could be interchanged.
1
1. Analysis of its H and 13C NMR spectra indicated that
it contained the same amino acid residues (Ile × 2, Pro ×
1, Phe × 1) as 1. Because those signals from ring skeletons
were similar for both compounds, it appeared that 4 was a
structural isomer of 1, the difference being in the sequence
of the intermediate and terminal amino acid residues. The
base peak at m/z 148 in the MS established N,N-dimeth-
ylphenylanaline as the terminal amino acid residue. By
elimination, the intermediate amino acid residue was
isoleucine, and the structure of 4 was determined as shown.
displaying a base peak at m/z 114, supported this assign-
ment. Thus, the structure of 7 was determined to be as
shown.
The stereochemistry in the ring nucleus of these seven
compounds was deduced by the CD method as indicated
above. Both intermediate and basic amino acids in other
reported cyclopeptides were determined mostly to be the
L- form by analysis of the products of acid hydrolysis.7 We
thought that NMR data (1H and 13C) could be distinct in
this N-terminal dipeptide portion of different stereochem-
istry, and this could serve as a basis for determining the L
Compound 5 had the molecular formula C36H49N5O6, the
same as 2 but 14 mass units less than 4. The 1H NMR
spectrum of 5 was very similar to that of 4, except for a
three-proton N-Me singlet at δ 2.25 instead of a six-proton
NMe2 at δ 2.29 in 4. Hence, 5 was likely to be the N-mono-
demethylated analogue of 4. The mass spectrum exhibited
a base peak at m/z 134, consistent with fragmentation
typical of the terminal amino acid N-monomethyl phen-
ylalanine. Thus, the structure of 5 was established as
shown.
or
D forms of the amino acid residues. Thus, model
compounds of the N-terminal dipeptides in cyclopeptide
alkaloids of the zizyphine A type were prepared, and their
NMR spectra were analyzed. These model dipeptides
include four sets of diastereoisomers (both L,L and L,D
forms), that is, Phe(OEt)-Leu(NMe2), Ile(OMe)-Leu(NMe2),
Ile(OMe)-Phe(NMe2), and Leu(OMe)-Phe(NMe2); and two
L,L-forms: Ile(OMe)-Ile(NMe2) and Phe(OEt)-Ile(NMe2).
Starting from L-N-tBoc-Ile, L-N-tBoc-Phe, D-Phe (OEt), L-
and D- Leu (N-Boc), these dipeptides were prepared by a
common method, including three major steps, peptide bond
formation via DCC coupling of an N-Boc amino acid and
an amino acid ester,8 followed by N-deprotection (removal
of Boc, trifluoroacetic acid/H2O ) 3:1)9 and N,N-dimeth-
ylation (HCHO/HCO2H, ∆; or HCHO, H2/Pd-C, MeOH).10
Compound 6 had the molecular formula C31H40N4O5. In
addition to signals for the 13-membered nucleus including
styrylamine, isoleucine, and â-hydroxyproline residues, its
1H NMR spectrum displayed signals typical of an N,N-
dimethyl phenylalanine residue: δ 7.04-7.17 (5H), 3.42
(1H, dd, H-2′′), 3.06 (1H, dd) and 2.91 (1H, dd) (Η-3′′), and
2.39 (6H, s, NMe2). The mass spectrum confirmed this
suggestion, with a base peak at m/z 148. Consequently, the
structure of 6 was determined as shown.
1
The H NMR of these compounds showed the signal of
R-H in the amino acid residue of the OR and NMe2 terminal
at δ 4.48-4.90 and 2.49-3.07, respectively, and the N,N-
dimethyl singlet around δ 2.15-2.32. Comparison of the
1H NMR of these diastereoisomers revealed no large
differences among them. Thus 1H NMR data were not
suitable to distinguish them. Most of the 13C NMR data of
the synthetic dipeptides (Table 3) were assigned directly
by analysis of the broad band decoupling and DEPT-135
spectra and also by intercorrelation, and consistency among
these assignments were confirmed by 2D NMR techniques.
For instance, L-Phe(OEt)-L-Leu(NMe2), L-Phe(OEt)-D-Leu-
(NMe2), and L-Phe(OEt)-L-Ile(NMe2) have common C-
terminal residue [Phe(OEt)], by comparison the signals
belonging to Phe(OEt) can be distinguished. The close
carbonyl signals of ester and amide in L-Leu(OMe)-L-Phe-
(NMe2), L-Leu(OMe)-D-Phe(NMe2), L-Ile(OMe)-L-Ile(NMe2),
and L-Phe(OEt)-L-Ile(NMe2) were assigned by COLOC
Compound 7 had the molecular formula C34H53N5O6. In
addition to signals for the 13-membered ring nucleus, the
1H NMR spectrum of 7 (Table 1) revealed signals of four
methyls, three doublets (δ 0.92, 0.90 and 0.81), and one
triplet (δ 0.84) and a six-proton singlet at δ 2.24. These
data indicated that 7 was composed of four amino residues
(Ile × 2, Leu, Pro) and a styrylamine unit. Of these, the
ring amino acids were Ile and Pro as indicated above, while
the intermediate and terminal amino acid residues were
leucine and isoleucine, one of which was N,N-dimethylated.
Upon irradiation of the six-proton singlet at δ 2.24 (NMe2),
the double doublet signal at δ 2.89 (J ) 8.4 and 5.0 Hz)
(R-H, H-2′′) was enhanced, suggesting N,N-dimethylleucine
to be the terminal amino acid residue. The mass spectrum,

Cyclopeptide Alkaloids from Paliurus
J ournal of Natural Products, 2000, Vol. 63, No. 10 1341
Ta ble 3. 13C NMR Data of N,N-Dimethyl Dipeptide Pairs
Phe(OEt)-Leu(NMe2) and Leu(OMe)-Phe(NMe2) in CDCl3 (20
MHz)
mass spectrometer (FABMS: matrix, 4-nitrobenzyl alcohol;
mass correlation, PEG600); Bruker AC-80 and AMX-400 NMR
spectrometers in CDCl3 (δH 7.24, δC 77.0) using Bruker
standard pulse programs: in the HETCOR, HMQC, COLOC,
and HMBC experiments, ∆ ) 1 s and J ) 140, 140, 8, and 8
Hz, respectively, the correlation maps consisted of 512 × 1K
data points per spectrum, each composed of 16 to 64 transients;
Sanki CPC instrument, LLN type with 6 1000E cartridges (410
mL), using the upper and lower layer of the solvent system,
CHCl3-MeOH-6% HOAc (5:5:3), as mobile and stationary
phases, respectively, 2.0 mL/min flow rate, 800 rpm rotation
speed, about 40 kg/cm2 pressure; Si gel TLC analysis [Me2-
CO-toluene (3: 7), saturated with 25% NH4OH].
chemical shift
Phe(OEt)-Leu(NMe2)a
Leu(OMe)-Phe(NMe2)b
∆δ
173.3 s 173.4 s -0.1
c
c
position
LL
LD
∆δd
LL
LD
1
2
3
4
5(9)
6(8)
7
171.8 s 171.8 s
0.0
52.6 d
38.2 t
136.4 s 136.4 s
129.2 d 129.2 d
52.8 d -0.2
50.4 d
41.5 t
24.8 d
22.8 q
21.8 q
50.4 d
41.4 t
24.8 d
22.8 q
21.8 q
0.0
+0.1
0.0
0.0
0.0
38.2 t
0.0
0.0
0.0
128.4 d 128.5 d -0.1
126.9 d 126.9 d 0.0
P la n t Ma ter ia l. The roots of Paliurus ramossisimus for this
study were collected in August 1989, along the west coast of
Ho-Long, Miao Li, Taiwan. A voucher specimen (No. NTUPH-
19890801) has been deposited in the School of Pharmacy,
National Taiwan University.
1′
2′
3′
4′
5′(9′)
6′(8′)
7′
173.4 s 173.2 s +0.2
172.2 s 172.1 s +0.1
67.7 d
37.3 t
25.7 d
23.4 q
22.0 q
67.1 d +0.6e
36.3 t +1.0
71.0 d
33.0 t
70.6 d +0.4
32.3 t +0.7
25.8 d -0.1
23.3 q +0.1
22.1 q -0.1
139.8 s 140.1 s -0.3
129.2 d 129.2 d
128.3 d 128.3 d
126.0 d 126.0 d
0.0
0.0
0.0
Extr a ction a n d Isola tion . The ground, dry roots (36 kg)
were percolated with MeOH (150 L × 3). The MeOH extract
(2.38 kg) was triturated with n-hexane (3 L × 3), and CHCl3
(3 L × 3) to give fractions soluble in n-hexane (113 g) and
CHCl3 (396 g). The residue was suspended in H2O (2 L) and
further partitioned against EtOAc (3 L × 3) and n-BuOH (3 L
× 3) to give fractions soluble in EtOAc (74 g), BuOH (378 g),
and H2O (708 g). The CHCl3-soluble fraction was triturated
with 1% NaOH (0.8 L × 4) to remove triterpenes such as
betulinic acid and ceanothic acid.2 The residue was then
triturated with CHCl3, and the soluble fraction (55 g) was
dissolved in 95% EtOH and then passed through a Sephadex
LH-20 column (1100 mL × 3, 95% EtOH) to give a fraction
(16 g) containing cyclopeptide alkaloids. Part of this fraction
(3 g) was separated on a CPC column using the conditions
described above, monitored by Si gel TLC analysis, to give nine
fractions. Fraction 4 yielded 2 (43 mg) after recrystallization
from MeOH. Fraction 8 (35 mg) was separated by preparative
TLC (0.5 mm) developed thrice with the TLC solvent indicated
above to give 4 (23 mg). Fraction 5 (38 mg) was separated by
preparative TLC (0.5 mm) developed thrice with Me2CO-
toluene (2:8), saturated with 25% NH4OH, to give 2 (12 mg)
and 5 (9 mg). Fraction 6 (356 mg) was chromatographed on a
Si gel column (70 g, 70-230 mesh, 0-6% MeOH in CHCl3) to
give nine subfractions. Subfraction 4 (66 mg) was separated
by preparative TLC (0.5 mm) developed with the TLC solvent
indicated above to give 3 (18 mg) and a 14-membered cyclo-
peptide (39 mg). Subfraction 6 (101 mg) was separated by
preparative TLC (1 mm and 0.5 mm) developed with Me2CO-
toluene (2:3 and 1:4), both saturated with 25% NH4OH,
respectively, to give 1 (34 mg) and 6 (15 mg) from the first
preparative TLC and 7 (22 mg) from second one.
NMe2
42.3 q
42.0 q +0.3
42.4 q
41.9 q +0.5
a
b
δOEt 61.2 (t) and 14.1 (q) in Phe(OEt)-Leu(NMe2). δOMe 52.0
c
(q) in L-Leu(OMe)-Phe(NMe2). L-Phe(OEt)-L-Leu(NMe2) stands
for C-terminal amino acid is L- form and N-terminal one is D-form.
d
∆δ (ppm) ) δLL - δLD
.
technique. For example, in L-Ile(OMe)-L-Ile(NMe2) these
two signals were assigned at δ 172.49 and 171.63, respec-
tively, from the observation of three-bond coupling to ester
OMe (δ 3.68). The ester or amide carbonyl signals in the
other compounds were assigned by correlation. The reso-
nance differences of the corresponding carbons between
these diastereoisomers were observed and are listed in
Table 3. Inspection of these data reveals significant shift
differences between each pair of four sets of diastereoiso-
mers for the chiral R-carbon (C-2′) and those around this
chiral center (C-3′ and NMe2) with different chirality. The
shift difference (∆δLL-LD) of C-2′ ranges from 0.2 to 0.8 ppm,
that of C-3′ ranges from 0.6 to 1.5 ppm. The shift difference
(∆δLL-LD) of N,N-dimethyl carbons is relatively constant (ca.
0.30 ppm). Consequently, this method should be suitable
for determining the stereochemistry of the terminal dipep-
tides in 13- and 14-cyclopeptide alkaloids.
The carbon chemical shifts of the N-terminal amino acid
residue of compounds 1, 2, 4, 5, and 7 were very consistent
with those of the corresponding amino acid residue in the
prepared models of L-L dipeptides, but they were quite
different from those of L-D ones. Hence, the intermediate
and terminal basic amino acids are L forms.
P a liu r in e A (1): amorphous powder; [R]26 -345.0° (c 1.0,
D
MeOH); IR νmax 3390, 3310, 2965, 2940, 2870, 1680, 1640, 1510,
1501, 1220, 1180, 1150, 1081, 879 cm-1; UV λmax (log ꢀ) 216
(4.84), 269 (4.75), 318 (4.56) nm; CD (c 1.51 × 10-5 M) (∆ꢀ)
367 (+3.75), 356 (+4.16), 327 (-8.68), 313 (-10.01), 229
(+5.98); 1H and 13C NMR, see Tables 1 and 2 and Supporting
Information; FABMS (positive) m/z (rel int) [M + H]+ 662 (67),
628 (13), 374 (4, a ), 114 (100, b); HRFABMS (positive) m/z [M
+ H]+ 662.3901 (calcd for C37H52N5O6 662.3918).
Among these seven compounds, 3 was previously isolated
from Zizyphus sativa (sativanine G11). However, we have
now assigned its NMR data, which had not been reported.
The other compounds, to our knowledge, represent the first
natural occurrence of such products. We have, however,
reported the conformation of paliurine B (2) based on the
assumption of an all L-amino acid composition.12 The trivial
name paliurines A-B for compounds 1 and 2, and paliu-
rines C-F for compounds 4-7 were assigned based on
their plant origin. The 14-membered cyclopeptide alkaloids
have been reported to possess sedative effect.13 Our pre-
liminary study on such activity of these 13-membered bases
(1-7), however, was not very consistent. Instead, we have
indication that they possess immunostimulant activity,
which deserves further exploration.
P a liu r in e B (2): colorless amorphous solid, mp 111-112
°C; [R]26 -391.3° (c 0.76, MeOH); IR νmax 3328, 2963, 2934,
D
2876, 1642, 1510, 1455, 1220, 1183, 1089, 1036, 881, 799, 756,
701 cm-1; UV λmax (log ꢀ) 216 (4.81), 268 (4.71), 319 (4.56) nm;
CD (c 1.54 × 10-5 M) (∆ꢀ) 356 (+4.71), 314 (-10.84), 267
(-26.49), 252 (-22.48), 228 (+5.57); 1H and 13C NMR, see
Tables 1 and 2 and Supporting Information; FABMS (positive)
m/z (rel int) [M + H]+ 648 (51), 614 (14), 374 (15, a ), 346 (12),
100 (100, c); HRFABMS (positive) m/z [M + H]+ 648.3754
(calcd for C36H50N5O6 648.3761).
Sa tivn in e G (3): colorless amorphous solid, mp 93-94 °C;
[R]26D -327.0° (c 0.85, MeOH); IR νmax 3334, 2964, 2935, 2876,
1674, 1642, 1510, 1463, 1224, 1183, 1111, 1092, 863, 757, 695
cm-1; UV λmax (log ꢀ) 217 (4.66), 230 (4.51), 269 (4.65), 320
(4.47) nm; CD (c 1.95 × 10-5 M) (∆ꢀ) 356 (+5.75), 340 (0), 323
Exp er im en ta l Section
Gen er al Exper im en tal P r ocedu r es. Perkin-Elmer 1760-X
infrared FT spectrometer (KBr); Hitachi 2000 UV (MeOH);
J ASCO J -710 spectropolarimeter (MeOH); J EOL J MX-HX110

1342 J ournal of Natural Products, 2000, Vol. 63, No. 10
Lin et al.
(-7.93), 315 (-8.15), 267 (-22.92), 250 (-15.31), 237 (0), 232
excess DCC was destroyed by few drops of acetic acid, and the
precipitate (N,N-dicylohexylurea, DCU) was removed by filtra-
tion. After evaporation, the residue was flash chromatographed
over Si gel (17 g, 230-400 mesh) elution with 1 to 5% of Me2-
CO in toluene to give the product L-Ile(OMe)-L-Phe(N-tBoc)
(232 mg, 78% yield): 1H NMR δ (CDCl3) 3.65 (s, OMe), 4.95
(s, NHBOc), 1.39 (s, Boc-Me).
3. P r ep a r a tion of L-Ile(OMe)-L-P h e(NMe2). In a 25-mL
round-bottom flask, l-Ile(OMe)-L-Phe(N-tBoc) was mixed with
trifluoroacetic acid (4.5 mL) and H2O (1.5 mL), and the
suspension was stirred at room temperature for 4 h. After
evaporation under reduced pressure, the residue, which showed
a single spot in TLC analysis without further purification, was
N,N-dimethylated under catalytic conditions (10% Pd/C, 50
mg; H2, 1 atm; EtOH 5 mL, HCHO 0.5 mL, overnight). The
resultant suspension was filtered through a Celite cake, and
the filtrate was condensed. The residue was then partitioned
between CHCl3 (20 mL) and 1% ammonia water (20 mL × 3).
The CHCl3 layer was dried over MgSO4 and evaporated to give
the crude product (222 mg), which was purified via a Si gel
flash column (9 g, 230-400 mesh) eluted with 0 to 0.5% MeOH
in CHCl3 to give the product, L-Ile(OMe)-L-Phe(NMe2) (203 mg,
91.5% yield).
1
(+5.16), 225 (0), 217 (-8.02); H and 13C NMR, see Tables 1
and 2 and Supporting Information; FABMS (positive) m/z (rel
int) [M + H]+ 515 (22), 307 (58), 289 (31), 114 (100, b);
HRFABMS (positive) m/z [M + H]+ 515.3246 (calcd for
C
28H43N4O5 515.3234).
P a liu r in e C (4): colorless amorphous solid; [R]26 -311.0°
D
(c 1.00, MeCN); IR νmax 3395, 2960, 2930, 2850, 1690, 1640,
1500, 1462, 1260, 1220, 1180, 1090, 1037, 880, 770, 700 cm-1
;
UV λmax (log ꢀ) 214 (4.80), 268 (4.62), 318 (4.48) nm; CD (c 1.51
× 10-5 M) (∆ꢀ) 355 (+7.93), 315 (-5.52), 262 (-17.89), 249
(-13.36), 232 (4.12), 224 (0), 219 (-1.25); 1H and 13C NMR,
see Tables 1 and 2 and Supporting Information; FABMS
(positive) m/z (rel int) [M + H]+ 662 (34), 570 (7), 374 (2, a ),
289 (5), 148 (100, d ).
P a liu r in e D (5): [R]26 -164.0° (c 1.00, MeCN); IR νmax
D
3400, 3380, 2960, 2920, 2850, 1690, 1640, 1505, 1456, 1220,
1180, 1035, 890, 770, 700 cm-1; UV λmax (log ꢀ) 216 (4.56), 234
(4.31), 269 (4.34), 318 (4.13) nm; CD (c 1.55 × 10-5 M) (∆ꢀ)
358 (+5.75), 336 (0), 313 (-4.88), 267 (-14.34), 248 (-11.86);
1H and 13C NMR, see Tables 1 and 2 and Supporting Informa-
tion; FABMS (positive) m/z (rel int) [M + H]+ 648 (69), 374
(20, a ), 307 (30), 134 (100, e).
P a liu r in e E (6): [R]26 -382.3° (c 0.94, MeCN); IR νmax
The other dipeptides were prepared from commercially
available (Sigma Co.) N- or O-protected amino acids, L-Phe-
(OEt), D- and L-Phe(NBoc), L- and D-Leu(NBoc), and L-Ile(OMe)
in a similar manner. The physical data of the compounds
follow.
D
3390, 3345, 2960, 2930, 2870, 1670, 1638, 1502, 1460, 1415,
1380, 1290, 1260, 1220, 1180, 1102, 1078, 1055, 1040, 878, 810,
790, 750, 700 cm-1; UV λmax (log ꢀ) 216 (4.82), 269 (4.70), 318
(4.55) nm; CD (c 1.82 × 10-5 M) (∆ꢀ) 356 (+5.69), 267 (-24.04),
240 (0), 231 (+14.11), 214 (+4.27); 1H and 13C NMR, see Tables
1 and 2 and Supporting Information; FABMS (positive) m/z
(rel int) [M + H]+ 549 (100), 374 (2, a ), 307 (29), 154 (88), 148
(99, d ), 136 (60); HRFABMS (positive) m/z [M + H]+ 549.3112
(calcd for C31H41N4O5 549.3077).
L-P h e(OEt)-L-Leu (NMe2): 1H NMR δ (CDCl3, 80 MHz) 7.20
(6H, m, φ-H and amide-H), 4.82 (1H, dt, J ) 6.0, 7.8 Hz, H-2),
3.11 (1H, dd, J ) 11.6, 7.8 Hz, H-3a), 3.03 (1H, J ) 11.6, 6.0
Hz, H-3b), 2.78 (1H, dd, J ) 7.4, 6.1 Hz, H-2′), 2.15 (6H, s,
NMe2), 0.85 (3H, d, J ) 6.0 Hz) and 0.83 (3H, d, J ) 6.0 Hz)
(H-5 and H-6), 4.14 (2H, q, J ) 7.1 Hz, OCH2CH3), 1.35 (3H,
t, J ) 7.1 Hz, OCH2CH3); 13C NMR data, see Table 3; EIMS
m/z (rel int) [M]+ 334 (2), 114 (100).
P a liu r in e F (7): [R]26 -323.0° (c 1.00, MeCN); IR νmax
D
3400, 3330, 2955, 2930, 2860, 1690, 1630, 1505, 1435, 1365,
1262, 1221, 1180, 1040, 880, 820, 770 cm-1; UV λmax (log ꢀ)
217 (4.62), 269 (4.56), 318 (4.38) nm; CD (c 1.59 × 10-5 M)
(∆ꢀ) 355 (+7.42), 268 (-19.62), 262 (-20.27), 249 (-14.37), 231
(0), 219 (-9.13); 1H and 13C NMR (CDCl3), see Tables 1 and 2
and Supporting Information; major NOEs: 14-OMe to H-13′
and H-1, H-3 to H-2, H-5 and H-12, H-5 to H-3, H-15 and H-18,
H-6 to H-8, H-9 to H-12, H-19â and H-20â, H-12 to H-1, H-3
and H-9, H-2R′ (amide-H) to H-2′, H-2′′ and NMe2, NMe2 to
H-2R′, H-2′′ and H-3′′; key HMBC data: H-3 (δ 8.44) and H-5
(δ 4.26) to C-4 (δ 167.1), H-12 (δ 6.68), H-12′ (δ 6.79) and 14-
OMe (δ 3.77) to C-14 (δ 151.5), H-13′ (δ 6.87) to 11 (δ 151.1),
H-6 (δ 7.19) to C-7 (δ 170.3), H-2′ (δ 4.55) to C-1′ (δ 171.8),
H-2′′ (δ 2.89) and H-3′′ (δ 1.53 and 1.40) to C-1′′ (δ 173.1);
FABMS (positive) m/z (rel int) [M + H]+ 628 (40), 374 (2, a ),
255 (4), 165 (2), 114 (100, f).
L-P h e(OEt)-D-Leu (NMe2): 1H NMR δ (CDCl3, 80 MHz)
7.19 (6H, m, φ-H and amide-H), 4.81 (1H, dt, J ) 6.6, 8.2 Hz,
H-2), 3.12 (1H, m, H-3a), 3.04 (1H, m, H-3b), 2.82 (1H, dd, J
) 8.1, 5.2 Hz, H-2′), 2.17 (6H, s, NMe2), 0.89 (3H, d, J ) 5.7
Hz) and 0.86 (3H, d, J ) 6.1 Hz) (H-5 and H-6), 4.14 (2H, q, J
) 7.1 Hz, OCH2CH3), 1.35 (3H, t, J ) 7.1 Hz, OCH2CH3); 13C
NMR data, see Table 3; EIMS m/z (rel int) [M]+ 334 (3), 115
(3), 114 (100).
L-Ile(OMe)-L-Leu (NMe2): 1H NMR δ (CDCl3, 80 MHz) 7.29
(1H, d, J ) 9.0 Hz, amide-H), 4.50 (1H, dd, J ) 9.0, 4.9 Hz,
H-2), 3.68 (3H, s, OCH3), 2.82 (1H, dd, J ) 7.9, 5.6 Hz, H-2′),
2.24 (6H, s, NMe2), 0.88 (9H, d, J ) 6.1 Hz, H-6, H-5′ and H-6′),
0.84 (3H, t, J ) 6.1 Hz, H-5); 13C NMR data, see Supporting
Information; EIMS m/z (rel int) [M]+ 286 (1), 215 (7), 114 (100).
L-Ile(OMe)-D-Leu (NMe2): 1H NMR δ (CDCl3, 80 MHz) 7.37
(1H, d, J ) 9.0 Hz, amide-H), 4.50 (1H, dd, J ) 9.0, 4.8 Hz,
H-2), 3.67 (3H, s, OCH3), 2.90 (1H, dd, J ) 8.0, 5.0, Hz, H-2′),
2.25 (6H, s, NMe2), 0.92-0.83 (12H, m, H-5, H-6, H-5′, and
H-6′); 13C NMR data, see Supporting Information; EIMS m/z
(rel int) [M]+ 286 (2), 114 (100).
Gen er a l Met h od for P r ep a r a t ion of N,N-Dim et h yl
Dip ep tid es. A representative method for preparing L-Ile-
(OMe)-L-Phe(N-Me2) via three-step reaction is described below.
1. P r ep a r a tion of L-Ile (OMe) fr om L-Ile (N-tBoc). To a
methanolic solution (30 mL) of L-Ile (N-tBoc) (3.30 g, 14.3
mmol) in a 250-mL round-bottom flask was added ethereal
diazomethane freshly prepared from diazald (6.13 g). The flask
was then screw-tightened and kept at 4 °C for overnight. The
solution was evaporated to give a residue (3.25 g, 98.5%), which
was essentially pure by TLC. Without further purification, part
of the residue (600 mg, 2.44 mmol) in a 10-mL flask was mixed
with trifluoroacetic acid (1.5 mL) and H2O (0.5 mL), and the
suspension was stirred at room temperature for 4 h. After
evaporation under reduced pressure, the residue was parti-
tioned between CHCl3 (30 mL) and 5% ammonia water (20
mL × 3). The CHCl3 layer was dried over MgSO4 and
evaporated to give a colorless viscous residue, L-Ile(OMe) (277
L-Ile(OMe)-L-P h e(NMe2): 1H NMR δ (CDCl3, 80 MHz) 7.62
(1H, d, J ) 8.8 Hz, amide-H), 7.18 (5H, m, φ-H), 4.45 (1H, dd,
J ) 8.8, 6.0 Hz, H-2), 3.62 (3H, s, OCH3), 3.50-2.77 (3H, m,
H-2′, H-3′a, and 3′b), 2.27 (6H, s, NMe2), 0.86 (3H, t, J ) 5.3
Hz, H-5), 0.82 (3H, d, J ) 6.7 Hz, H-6); 13C NMR data, see
Supporting Information; EIMS m/z (rel int) [M]+ 320 (0.4), 230
(7), 229 (35), 148 (100).
L-Ile(OMe)-D-P h e(NMe2): 1H NMR δ (CDCl3, 200 MHz)
7.23-7.11 (6H, m, φ-H and amide-H), 4.45 (1H, dd, J ) 8.9,
5.0 Hz, H-2), 3.69 (3H, s, OCH3), 3.46 (1H, dd, J ) 7.8, 5.3 Hz,
H-2′), 3.17 (1H, dd, J ) 13.6, 7.8 Hz) and 2.92 (1H, dd, J )
13.6, 5.3 Hz) (H-3′a and 3′b), 2.39 (6H, s, NMe2), 0.83 (3H, t,
J ) 7.2 Hz, H-5), 0.75 (3H, d, J ) 6.9 Hz, H-6); and 13C NMR
data, see Supporting Information; EIMS m/z (rel int) [M]+ 320
(0.5), 230 (6), 229 (33), 147 (7), 148 (100).
1
mg, 78% yield), H NMR (CDCl3) δOMe 3.65.
2. P r ep a r a tion of L-Ile(OMe)-L-P h e(N-tBoc). To the
solution of L-Ile(OMe) (110 mg, 0.76 mmol) and L-Phe(N-tBoc)
(201 mg, 0.76 mmol) in CH2Cl2 (10 mL) and pyridine (five
drops) in a 10-mL round-bottom flask was added dropwise the
solution of N,N′-dicyclohexyl-carbodiimide (DCC) (234 mg, 1.14
mmol) in CH2Cl2 (5 mL) in an ice bath and stirred for
additional 30 min and then at room temperature for 4 h. The
L-Leu (OMe)-L-P h e(NMe2): 1H NMR δ (CDCl3, 80 MHz)
7.20 (6H, m, φ-H and amide-H), 4.52 (1H, dt, J ) 5.6, 8.6 Hz,
H-2), 3.65 (3H, s, OCH3), 3.40-2.75 (3H, m, H-2′, H-3′a, and
H-3′b), 2.29 (6H, s, NMe2), 0.89 (6H, d, J ) 5.0 Hz, H-5 and

Cyclopeptide Alkaloids from Paliurus
J ournal of Natural Products, 2000, Vol. 63, No. 10 1343
H-6); 13C NMR data, see Table 3; EIMS m/z (rel int) [M]+ 320
(0.4), 230 (2), 229 (24), 149 (5), 148 (100).
Ack n ow led gm en t. This work was supported by NSC,
Republic of China, under grants NSC79-0420-B002-144 and
NSC80-0420-B002-154.
L-Leu (OMe)-D-P h e(NMe2): 1H NMR δ (CDCl3, 80 MHz)
7.21 (6H, m, φ-H and amide-H), 4.53 (1H, dt, J ) 5.8, 9.3 Hz,
H-2), 3.68 (3H, s, OCH3), 3.40-2.77 (3H, m, H-2′, H-3′a, and
H-3′b), 2.32 (6H, s, NMe2), 0.85 (6H, d, J ) 5.1 Hz, H-5 and
H-6); 13C NMR see Table 3; EIMS m/z (rel int) [M]+ 320 (0.6),
230 (2), 229 (25), 149 (6), 148 (100).
Su p p or tin g In for m a tion Ava ila ble: HMBC and NOEs of paliu-
rine A, key mass fragments, 1H and 13C NMR data in the ring nucleus,
13C NMR of paliurines D and G versus synthetic dipeptides.
L-Ile(OMe)-L-Ile(NMe2): 1H NMR δ (CDCl3, 400 MHz) 6.93
(1H, d, J ) 8.4 Hz, amide-H), 4.55 (1H, dd, J ) 8.4, 4.8 Hz,
H-2), 3.69 (3H, s, OCH3), 2.55 (1H, d, J ) 5.2 Hz, H-2′), 2.24
(6H, s, NMe2), 0.91-0.86 (12H, m, H-5, H-5′, H-6, and H-6′);
13C NMR δ (CDCl3, 20 MHz) 172.5 (s, C-1), 171.6 (s, C-1′), 74.6
(d, C-2′), 56.1 (d, C-2), 51.7 (q, OCH3), 43.2 (q, NMe2), 37.6 (d,
C-3), 34.5 (d, C-3′), 27.0 (t, C-4′), 25.2 (t, C-4), 15.8 (q, C-6),
14.6 (q, C-6′), 11.9 (q, C-5′), 11.4 (q, C-5); key COLOC data
(100 MHz) C-1 (δ 172.5) to OCH3 (δ 3.69); EIMS m/z (rel int)
[M]+ 286 (28), 275 (6), 230 (28), 229 (94), 115 (94), 114 (100).
L-P h e(OEt)-L-Ile(NMe2): 1H NMR δ (CDCl3, 400 MHz)
7.28-7.21 (5H, m, φ-H), 6.76 (1H, d, J ) 8.6 Hz, amide-H),
4.91 (1H, dt, J ) 5.6, 8.6 Hz, H-2), 3.19 (1H, dd, J ) 14.1, 5.6
Hz, H-3a), 3.01 (1H, dd, J ) 14.1, 8.6 Hz, H-3b), 2.57 (1H, d,
J ) 4.7 Hz, H-2′), 2.21 (6H, s, NMe2), 4.16 (2H, q, J ) 7.1 Hz,
OCH2CH3), 1.23 (3H, t, J ) 7.1 Hz, OCH2CH3), 0.86 (3H, t, J
) 7.3 Hz, H-5′), 0.63 (3H, t, J ) 6.8 Hz, H-6′); 13C NMR δ
(CDCl3, 20 MHz) 171.9 (s, C-1), 171.5 (s, C-1′), 136.4 (s, C-4),
129.1 (d, C-5 and C-9), 128.5 (d, C-6 and C-8), 126.9 (d, C-7),
74.5 (d, C-2′), 61.2 (t, OCH2CH3), 52.6 (d, C-2), 43.1 (q, NMe2),
34.3 (d, C-3′), 26.8 (t, C-4′), 14.2 (q, C-6′), 14.1 (q, OCH2CH3),
11.9 (q, C-5′); key COLOC data (100 MHz) C-1 (δ 171.9) to
H-2 (δ 4.91), H-3 (δ 3.01) and OCH2CH3 (δ 4.16); EIMS m/z
(rel int) [M]+ 334 (0.02), 115 (13), 114 (100).
Refer en ces a n d Notes
(1) Lee, S. S.; Su, W. C.; Liu, K. C. J . Nat. Prod. 1991, 54, 615-618.
(2) Lee, S. S.; Lin, C. J .; Liu, K. C. J . Nat. Prod. 1992, 55, 602-606.
(3) J oullie, M. M.; Nutt, R. F. In Alkaloids: Chemical & Biological
Perspective; Pelletier, S. W., Ed.; Wiley: New York, 1985; Vol. 3, pp
114-168.
(4) Tschesche, R.; Khokhar, I.; Spilles, C. H.; Eckhardt, G.; Cassels, B.
K. Tetrahedron Lett. 1974, 2941-2944.
(5) Schmidt, U.; Lieberknecht, A.; Bo¨kens, H.; Griesser, H. J . Org. Chem.
1983, 48, 2680-2685.
(6) Marchand, J .; Pais, M.; Monseur, X.; J arreau, F. X. Tetrahedron 1969,
25, 937-954.
(7) Tschesche, R.; Kaussmann, E. U. In The Alkaloids; Manske, R. H.,
Ed.; Academic Press: New York, 1975; Vol. 15, pp 165-205, and
references therein.
(8) Klausner, Y. S.; Bodansky, M. Synthesis 1972, 453-463.
(9) Tarbell, D. S.; Yamamoto, Y.; Pope, B. M. Proc. Natl. Acad. Sci. U.S.A.
1972, 69, 730-732.
(10) Emerson, W. S. Org. React. 1948, 4, 174-255.
(11) Shah, A. H.; Pandey, V. B.; Singh, J . P.; Singh, K. N.; Eckhardt, G.
Phytochemistry 1984, 23, 2120-2121.
(12) Yu, C.; Tseng, Y. Y.; Lee, S. S. Biochim. Biophys. Acta 1993, 1156,
334-342.
(13) Han, B. H.; Park, M. H.; Park J . H. Pure Appl. Chem. 1989, 61,
443-448.
NP000136A