Palmatine and Berberine isolation artifacts

Article abstract of DOI:10.1021/np0204996

Some 8-substituted derivatives of the protoberberine alkaloids palmatine (1a) and berberine (1b) have been prepared and investigated by 1D and 2D NMR spectroscopy (H-1, C-13, N-15). Complete NMR data for the 8-hydroxy (2), 8-methoxy (3), 8-ethoxy (4), and

Full text of DOI:10.1021/np0204996

J . Nat. Prod. 2003, 66, 481-486
481
P a lm a tin e a n d Ber ber in e Isola tion Ar tifa cts†
Radek Marek,*,‡ Pavl´ına Secˇka´rˇova´,‡ Dagmar Hulova´,‡ J arom´ır Marek,§ J irˇ´ı Dosta´l, and Vladim´ır Sklena´rˇ‡
National Center for Biomolecular Research, Faculty of Science, Masaryk University, Kotla´rˇska´ 2, CZ-611 37 Brno, Czech
Republic, Laboratory of Functional Genomics and Proteomics, Faculty of Science, Masaryk University, Kotla´rˇska´ 2,
CZ-611 37 Brno, Czech Republic, and Department of Biochemistry, Faculty of Medicine, Masaryk University,
Komenske´ho na´meˇstı´ 2, CZ-662 43 Brno, Czech Republic
Received November 4, 2002
Some 8-substituted derivatives of the protoberberine alkaloids palmatine (1a ) and berberine (1b) have
been prepared and investigated by 1D and 2D NMR spectroscopy (H-1, C-13, N-15). Complete NMR data
for the 8-hydroxy (2), 8-methoxy (3), 8-ethoxy (4), and 8-trichloromethyl (5) 7,8-dihydro derivatives of 1a
and 1b, as well as X-ray data for 8-methoxydihydroberberine (3b), 8-trichloromethyldihydropalmatine
(5a ), and 8-trichloromethyldihydroberberine (5b), are presented. The physicochemical data for all of these
compounds are reviewed and compared with previously published values.
Palmatine (1a ) and berberine (1b) are quaternary pro-
toberberine alkaloids. Their salts (1) are typically yellow
and are widely distributed in many species of the Berberi-
daceae, Fumariaceae, Papaveraceae, and other plant fami-
lies.1,2 Berberine displays a great variety of biological and
pharmacological activities (e.g., antimicrobial,3 antiplas-
modial,4 antidiarrheal,5 cardiovascular6). Both alkaloids
contribute to the chemical defense of plants by complex
actions with several molecular targets.7
The quaternary protoberberines 1 are iminium cations
derived from the 5,6-dihydrodibenzo[a,g]quinolizinium sys-
tem. They are characterized by the sensitivity of the polar
CdN+ bond to nucleophilic attack followed by the formation
of the corresponding 8-adduct.8,9 A number of adducts with,
for example, cyanide,10,11 methoxide,12 alkaline chloro-
form,13-15 and liquid ammonia16 have been described.
Palmatine (1a ) and berberine (1b) are frequently converted
into their free bases (2), called pseudobases, during the
isolation and purification process. In the presence of
common solvents such as MeOH, EtOH, and CHCl3, the
free bases 2 easily produce adducts 3, 4, and 5. In general,
since the pseudobases 2 do not occur in plants, compounds
3-5 must be considered to be isolation artifacts. In this
contribution we present complete 1H and 13C NMR data
the pK of berberine is 15.4, and the pK of palmatine is 15.7
(in MeOH19). Despite this lower stability, we managed to
determine the chemical shifts of both 2a and 2b in CD2Cl2
and C6D6 (Tables 1-4). CDCl3 was not used for measure-
ment since preparing the pseudobases 2a and 2b by
alkalizing the alkaloids with NaOH in CHCl3 solvent could
lead to the formation of dichlorocarbene, followed by its
subsequent reaction with the alkaloid skeleton.
We also detected in the CD2Cl2 solution of berberine free
base 2b minor amounts of bimolecular species, which are
well known to exist in solutions of benzophenanthridine
alkaloids.20-22 Investigation of the formation and structure
relations of these minor components is currently in progress.
The 8-methoxy adduct of palmatine (3a ) and 8-methoxy
(3b) and 8-ethoxy (4b) adducts of berberine were prepared
by reacting the respective quaternary alkaloid with a
solution of sodium methoxide or sodium ethoxide in an
appropriate alcohol. The NMR samples of methoxy adducts
3a and 3b dissolved in CDCl3 were rather unstable. In the
presence of residual H2O these compounds reacted with
CDCl3 to form the 8-trichloromethyl adducts 5a and 5b,
respectively. The reaction mechanism expected for this
conversion is shown in Scheme 1. Protonation of the
for these free bases 2, and for the artifacts 3-5, and
compare them with the previously published values.
Resu lts a n d Discu ssion
The free bases of palmatine (2a ) and berberine (2b) (the
8-hydroxy-7,8-dihydro derivatives), prepared by alkalizing
the quaternary alkaloids (1) with sodium hydroxide (see
Experimental Section), were rather unstable. Over time
(∼12 h), 2a and 2b spontaneously decomposed and formed
several disproportionation products (probably 7,8-dihydro
and 8-oxo-7,8-dihydro derivatives). The 8-hydroxy adducts
2a and 2b are less stable than the analogous hydroxy
derivatives of the benzophenanthridine alkaloids.17 One of
the reasons for this lower stability may be the higher pK
values of protoberberines. For comparison, the pK values
of benzophenanthridines are roughly 7.5-9.0 (in H2O18),
* To whom correspondence should be addressed. Tel: +420-541129634.
Fax: +420-541211214. E-mail: rmarek@chemi.muni.cz.
† Dedicated to Professor J irˇ´ı Slav´ık on the occasion of his 80th birthday.
‡ National Center for Biomolecular Research.
§ Laboratory of Functional Genomics and Proteomics.
Department of Biochemistry.
10.1021/np0204996 CCC: $25.00
© 2003 American Chemical Society and American Society of Pharmacognosy
Published on Web 03/21/2003

482 J ournal of Natural Products, 2003, Vol. 66, No. 4
Marek et al.
Ta ble 1. 1H NMR Chemical Shifts (δ) of Some 8-Substituted Derivatives of Dihydropalmatine in CD2Cl2, C6D6, and CDCl3 at 303 K
2a
3a
5a
2a
3a
5a
3a
5a
H atom
CD2Cl2
CD2Cl2
CD2Cl2
C6D6
C6D6
C6D6
CDCl3
CDCl3
1
4
5
7.19
6.66
7.20
6.67
7.18
6.66
7.29
6.34
7.28
6.36
7.28
6.35
7.21
6.65
7.18
6.63
2.89 m
2.93 m
3.54 m
3.75 m
6.20
2.91 m
2.91 m
3.54 m
3.64 m
6.12
2.74 m
3.39 m
3.72 m
3.84 m
5.64
2.45 m
2.67 m
3.19 m
3.71 m
6.45
2.47 m
2.65 m
3.24 m
3.60 m
6.56
2.34 m
3.14 m
3.39 m
3.59 m
5.92
2.91 m
2.95 m
3.56 m
3.71 m
6.22
2.74 m
3.41 m
3.74 m
3.86 m
5.64
6
8
11
12
13
6.95 d
6.92 d
6.15
6.97 d
6.90 d
6.07
6.99 d
6.88 d
6.17
6.75 d
6.96 d
6.34
6.78 d
6.97 d
6.28
6.73 d
6.89 d
6.38
6.96 d
6.92 d
6.06
6.98 d
6.89 d
6.15
2(OCH3)
3(OCH3)
8(OCH3)
9(OCH3)
10(OCH3)
3.88
3.84
3.88
3.85
3.05
3.89
3.87
3.84
3.51
3.42
3.47
3.43
3.13
4.01
3.46
3.43
3.94
3.90
3.06
3.94
3.93
3.88
3.92
3.86
3.92
3.86
3.98
3.45
3.88
3.38
3.95
3.89
3.87
3.45
3.89
Ta ble 2. 13C NMR Chemical Shifts (δ) of Some 8-Substituted Derivatives of Dihydropalmatine in CD2Cl2, C6D6, and CDCl3 at 303 K
2a
3a
5a
2a
3a
5a
3a
5a
C atom
CD2Cl2
CD2Cl2
CD2Cl2
C6D6
C6D6
C6D6
CDCl3
CDCl3
1
2
3
4
4a
5
108.05
148.52
149.85
111.30
128.38
30.25
108.18
148.54
149.83
111.33
128.28
30.21
107.75
148.49
149.96
111.73
127.89
30.28
109.06
149.18
150.40
111.73
128.40
30.02
109.09
149.26
150.40
111.77
128.15
30.07
108.63
149.15
150.48
112.09
127.65
30.15
107.58
147.88
149.12
110.60
127.70
29.87
107.12
147.85
149.23
111.00
127.34
29.98
6
47.07
47.66
52.21
46.91
47.23
51.93
47.10
51.72
8
79.49
85.42
74.02
79.67
85.10
74.45
84.83
73.63
8a
9
10
11
12
12a
13
122.63
145.48
150.13
114.36
119.68
127.11
94.58
119.53
146.25
150.02
114.46
119.16
129.09
94.30
115.17
147.01
150.12
115.17
118.70
129.98
97.41
123.48
146.26
150.27
114.64
119.54
127.48
94.83
119.82
147.16
150.24
114.89
119.02
130.13
94.57
115.40
147.29
150.03
115.61
118.72
130.46
97.96
118.83
145.87
149.59
114.13
119.00
128.73
94.01
114.86
146.48
149.63
114.70
118.48
129.51
97.01
13a
13b
136.66
124.28
56.44
137.97
124.09
56.43
56.22
52.47
137.97
123.84
56.46
136.99
124.79
55.94
138.53
124.46
55.85
55.59
51.07
138.18
124.23
55.86
137.62
123.76
56.07
55.92
51.80
137.56
123.58
56.11
2(OCH3)
3(OCH3)
8(OCH3)
8(CCl3)
9(OCH3)
10(OCH3)
56.21
56.25
55.54
55.58
55.93
106.22
61.18
56.70
106.51
60.69
56.18
105.60
60.95
56.38
61.45
56.56
61.13
56.54
61.03
56.07
60.81
56.04
60.99
56.26
Ta ble 3. 1H NMR Chemical Shifts (δ) of Some 8-Substituted Derivatives of Dihydroberberine in CD2Cl2, C6D6, and CDCl3 at 303 K
2b
3b
4b
5b
2b
3b
3b
5b
H atom
CD2Cl2
CD2Cl2
CD2Cl2
CD2Cl2
C6D6
C6D6
CDCl3
CDCl3
1
4
5
7.15
6.62
7.17
6.64
7.17
6.63
7.16
6.63
7.23
6.36
7.22
6.37
7.19
6.63
7.16
6.61
2.86 m
2.89 m
3.51 m
3.72 m
6.18
2.88 m
2.88 m
3.52 m
3.64 m
6.11
2.87 m
2.87 m
3.51 m
3.60 m
6.13
2.72 m
3.32 m
3.68 m
3.86 m
5.65
2.28 m
2.50 m
3.05 m
3.57 m
6.36
2.31 m
2.48 m
3.09 m
3.44 m
6.48
2.91 m
2.91 m
3.54 m
3.70 m
6.22
2.72 m
3.34 m
3.70 m
3.87 m
5.64
6
8
11
12
13
6.94 d
6.90 d
6.08
6.96 d
6.89 d
6.01
6.95 d
6.86 d
6.00
6.99 d
6.86 d
6.10
6.73 d
6.87 d
6.11
6.76 d
6.87 d
6.05
6.92 d
6.97 d
6.03
6.97 d
6.86 d
6.10
2(OCH2O)
5.93 d
5.95 d
5.94d
5.96 d
3.03
3.88
3.86
5.94 d
5.95 d
5.93 d
5.94 d
5.34 d
5.35 d
5.32 d
5.34 d
3.03
3.98
3.44
5.94 d
5.96 d
3.06
3.95
3.89
5.94 d
5.95 d
8(OCH3)
9(OCH3)
10(OCH3)
8(OCH2-)
8(-CH3)
3.91
3.85
3.88
3.85
3.28
1.00
3.91
3.86
3.95
3.44
3.94
3.87
methoxy group is probably the first reaction step. We
assume that following the elimination of MeOH, two
compounds coexist in equilibrium: the 8-hydroxy adduct
2 and the hydroxide of the quaternary alkaloid 1. The OH-
anion can extract the proton from chloroform (by eliminat-
C-8 atom of 1 to form the corresponding 8-trichloromethyl
derivative 5. We expect that the palmatine and berberine
pseudobases 2a and 2b, dissolved in CDCl3, behave simi-
larly and spontaneously produce the CCl3 adducts 5a and
5b. To determine the physicochemical data unequivocally,
we prepared the 8-trichloromethyl adducts 5a and 5b
-
ing H2O), and the CCl3 anion released then attacks the

Palmatine and Berberine Isolation Artifacts
J ournal of Natural Products, 2003, Vol. 66, No. 4 483
Ta ble 4. 13C NMR Chemical Shifts (δ) of Some 8-Substituted Derivatives of Dihydroberberine in CD2Cl2, C6D6, and CDCl3 at 303 K
2b
3b
4b
5b
2b
3b
3b
5b
C atom
CD2Cl2
CD2Cl2
CD2Cl2
CD2Cl2
C6D6
C6D6
CDCl3
CDCl3
1
2
3
4
4a
5
104.96
147.29
147.82
108.16
129.73
30.48
104.72
147.25
147.92
108.20
129.75
30.76
104.70
147.22
147.88
108.18
129.73
30.73
104.31
147.25
148.01
108.49
129.31
30.83
104.60
147.24
147.93
108.21
129.85
30.80
105.14
147.85
147.33
108.13
129.60
30.75
104.66
146.76
147.49
107.86
129.16
30.42
104.10
146.71
147.44
108.12
128.69
30.48
6
46.66
47.56
47.53
52.07
46.96
46.96
47.06
51.57
8
79.50
85.32
84.53
73.95
79.39
84.93
84.79
73.47
8a
9
10
11
12
12a
13
123.46
146.17
150.33
114.58
119.72
127.68
95.34
119.55
146.16
150.10
114.38
119.31
128.94
94.78
120.18
146.01
150.06
114.18
119.30
128.85
94.83
115.23
146.95
150.19
115.16
118.83
129.81
97.93
122.70
145.46
150.20
114.34
119.82
127.03
95.06
119.79
147.03
150.29
114.76
119.20
129.96
95.03
118.89
146.42
149.69
114.14
119.17
128.66
94.52
114.83
146.34
149.65
114.53
118.58
129.30
97.46
13a
13b
136.72
126.21
100.99
137.95
125.56
101.68
52.48
137.90
125.60
101.66
137.99
125.43
101.70
136.67
125.78
101.68
138.29
125.85
100.98
50.93
137.64
125.19
101.22
51.86
137.55
125.00
101.03
2(OCH2O)
8(OCH3)
8(CCl3)
9(OCH3)
10(OCH3)
8(OCH2-)
8(-CH3)
106.18
61.17
56.69
105.50
60.95
56.30
61.01
56.07
61.13
56.51
61.12
56.45
60.51
15.79
61.44
56.56
60.77
55.99
61.01
56.28
Sch em e 1
directly by reacting the alkaloid (1) with CHCl3 in the
presence of aqueous ammonia.
solvents are reported here for the first time. The portions
of the 13C NMR spectra of compounds 2a , 3a , and 5a that
indicate unequivocal assignments of the C-8 signals are
shown in Figure 1. The influence of substitution at C-8 is
clearly demonstrated in notable changes of the C-8 and
C-8a chemical shifts. The characteristic 13C NMR shifts of
the C-8 atom agree very well with values of analogous
-OH, -OMe, and -OEt derivatives of benzophenanthri-
dine alkaloids.17
To confirm the structures of 8-methoxydihydroberberine
(3b) and 8-trichloromethylpalmatine (5a ) unambiguously,
the respective crystals were investigated by X-ray diffrac-
tion analysis (see X-ray discussion). The crystal structure
The elucidation of the structures of compounds 2-5 was
based primarily on the NMR data. The structures of
compounds 3b, 5a , and 5b were also obtained by using
X-ray diffraction analysis. The NMR signals were assigned
by using the 2D correlation experiments, GHMBC23,24
(gradient-selected heteronuclear multiple-bond correlation)
and GSQMBC25 (gradient-selected single-quantum multiple-
bond correlation) (see Supporting Information). The NMR
chemical shifts of compounds 2-5 are summarized in
Tables 1-4. The complete assignments of the 13C NMR
signals of the protoberberine derivatives 2-5 in various

484 J ournal of Natural Products, 2003, Vol. 66, No. 4
Marek et al.
Ta ble 5. 15N NMR Chemical Shifts (δ) of Some 8-Substituted
Derivatives of Dihydroberberine at 303 K
compound
15N shifta
solvent
2b
3b
5b
5b
5b
88.5
83.6
64.9
65.0
64.6
CD2Cl2
CD2Cl2
CD2Cl2
CDCl3
DMSO-d6
a
Referenced to 1 M urea in DMSO-d6 (77.0 ppm)39 and liquid
CH3NO2 (381.7 ppm),39 and reported relative to liquid NH3.
at δ 5.64 is consistent with our data for the 8-trichloro-
methyl adducts 5a and 5b.
Systematic comparison of the NMR chemical shifts of the
derivatives obtained in our study with the values reported
previously28 reveals that the data presented by Wafo and
co-workers as belonging to “7,8-dihydro-8-hydroxypalma-
tine” isolated from Enantia chlorantha, correspond unam-
biguously to 8-trichloromethyldihydropalmatine (5a ). The
1H NMR chemical shifts and melting points are also
remarkably consistent with the data for 8-trichlorometh-
yldihydropalmatine (5a ) published earlier.13-15 As de-
scribed, during the procedure28-30 for isolating this com-
pound from plant material, the sample was in touch with
CHCl3 in the presence of aqueous ammonia, and the
8-trichloromethyl adduct (5) was probably formed (cf.
Scheme 1 and Experimental Section). Similarly, on the
basis of our data the compound reported as 2,3,9,10-
dimethylenedioxy-1,8-dihydroxy-7,8-dihydroprotoberber-
ine,31 isolated from Thalictrum delavayi, can be identified
as the 8-trichloromethyl derivative due to the characteristic
chemical shifts of H-8 and C-8. These are approximately δ
6.1-6.2 for H-8 in the 8-O adducts (2, 3, 4) and δ 5.6 for
H-8 in the CCl3 adducts (5) dissolved in CDCl3 or CD2Cl2.
For C-8 we measured chemical shifts of δ 79 (2), 85 (3),
and 74 (5). Previously reported values were δ 5.65 (1H) and
73.6 (13C) for the derivative of palmatine28 and δ 5.38 (1H)
and 74.0 (13C) for the derivative of 2,3,9,10-dimethylene-
dioxy-1-hydroxyprotoberberine.31
F igu r e 1. Portions of the 13C NMR spectra of three palmatine
derivatives in CD2Cl2. Trace A: 8-hydroxydihydropalmatine (2a ). Trace
B: 8-methoxydihydropalmatine (3a ). Trace C: 8-trichloromethyldi-
hydropalmatine (5a ). The influence of substitution at C-8 is clearly
demonstrated in the notable changes of the C-8, C-8a, C-12a, and C-13
chemical shifts.
F igu r e 2. Perspective view of the X-ray structure of compound 5a .
Moreover, in the two cases discussed above,28,31 a crude
alkaloid extract was treated with 2-10% HCl in the early
steps of the extraction procedure. If any 8-hydroxy deriva-
tives (2) had occurred in the plants as natural alkaloids,
this strong acid would have converted them to the corre-
sponding quaternary cations (1). The products were further
processed in CHCl3, even in the presence of aqueous
ammonia.
Similarly, the recent data32 reported for synthetically
prepared 8-hydroxydihydroberberine (2b) do not agree with
our data and probably indicate another derivative of
berberine (cf. NMR data).
15N NMR spectral parameters of compounds 2b, 3b, and
5b, determined at the natural 15N abundance level,33,34 are
summarized in Table 5. A comparison of the influence of
the aromatic and dihydroprotoberberine skeleton on the
15N chemical shifts has recently been published.35,36 The
electron properties of the substitution group are clearly
demonstrated in the 15N chemical shift.37 The nitrogen
resonance of the 8-methoxy derivative 3 is more shielded
(by ∼5 ppm) than the nitrogen of the 8-hydroxy adduct 2.
A substantial upfield shift of ∼20 ppm is further detected
for the 8-trichloromethyl derivative 5. An intense response
was observed between N-7 and H-13 in the 1H-15N
chemical shift multiple bond correlation spectra of the
above-specified dihydroprotoberberines in CD2Cl2 solution.
Other correlations with N-7 were observed for the signals
of H-8 and H-5.
F igu r e 3. Stereodrawing of the X-ray structure of compound 3b. Only
one of the two symmetry-independent molecules is shown.
of 8-trichloromethyldihydroberberine (5b) has already been
solved,26 and it is in good agreement with our new and more
precise data (see Experimental Section). ORTEP plots of
the 3D structures of compounds 5a and 3b are shown in
Figures 2 and 3, respectively. The structure of 7,8-
dimethoxy-2-methyl-3-(4′,5′-methylenedioxy-2′-vinylphenyl)-
1-trichloromethyl-1,2-dihydroisoquinoline has been previ-
ously solved using X-ray analysis by Ishii et al.27 The
chemical shift of the CH-CCl3 proton of this derivative27

Palmatine and Berberine Isolation Artifacts
J ournal of Natural Products, 2003, Vol. 66, No. 4 485
Sin gle-Cr ysta l X-r a y Diffr a ction An a lysis.38 Com-
pounds 3b, 5a , and 5b were also investigated by X-ray
diffraction analysis (Figures 2, 3, and Supporting Informa-
tion). Several structural features of 8-methoxydihydrober-
berine (3b) can be highlighted. The hetero ring C is in a
flattened half-chair conformation. The C8-OMe group
takes a pseudoaxial position, probably being forced to this
by the conformational preference of the polycyclic skeleton
rather than by the anomeric effect. There are two slightly
different conformations of the C-8 methoxy group in the
crystal, one of them with a shorter C8-O3 distance, 1.388-
(8) vs 1.532(8) Å. Ring B is the most puckered (q2 ) 0.510
Å, q3 ) -0.184 Å, φ ) -91.4°). The angle between the two
aromatic rings A and C is 40.61(6)°. In the 8-trichloro-
methyl adducts 5a and 5b, the CCl3 group is equally in
the pseudoaxial position (Supporting Information). The
C8-CCl3 bond in compounds 5 is somewhat longer as
compared with standard C-C length (1.577(3) Å in 5a ;
1.584(3) Å in 5b), probably due to the steric repulsion of
the CCl3 group. The angles between the two aromatic rings
in 5a and 5b are 21.04(7)° and 45.30(6)°, respectively. In
all three compounds, the C-10 methoxy group lies almost
in the plane of the parent aromatic ring, whereas the
neighboring C-9 methoxyl is nearly perpendicular to this
plane. These orientations of the C9-OMe and C10-OMe
groups are characteristic for the structural arrangements
of other protoberberines in the crystal state.11,16,26 The C2-
OMe and C3-OMe groups in the palmatine-chloroform
adduct 5a are more or less in-plane. Each of the compounds
3b, 5a , and 5b contains one stereogenic center (C-8). Owing
to their synthetic origin, these compounds are obviously
racemic.
To conclude, several findings should be summarized. The
derivative of palmatine reported28-30,32 as being the pseudo-
base 2a isolated from Enantia chlorantha corresponds
unequivocally to 8-trichloromethyldihydropalmatine (5a )
and must be classified as an artifact of the isolation.
Similarly, the reported 2,3,9,10-dimethylenedioxy-1,8-di-
hydroxy-7,8-dihydroprotoberberine31 obtained from Thal-
ictrum delavayi is very probably 2,3,9,10-dimethylenedioxy-
1-hydroxy-8-trichloromethyl-7,8-dihydroprotoberberine. The
discrepancies between our data for 8-hydroxydihydrober-
berine (2b) and those reported by Nyasse et al.32 are
likewise substantial. Although the solvent used for the
NMR measurements is not specified in the paper,32 the
reported data probably indicate another derivative of
berberine. In light of the evidence presented in this paper,
as well as in numerous previous contributions, CHCl3
should not be used when isolating protoberberines under
basic conditions. CHCl3 can be replaced advantageously by
other organic solvents (e.g., CH2Cl2), as already recom-
mended by Shamma and Rahimizadeh.14 We believe that
the data summarized in this contribution will clarify the
structure-reactivity relationships of the protoberberine
alkaloids palmatine and berberine.
chemical shifts (δ in ppm) were referenced to the signals of
the solvents [7.16 (1H) and 128.00 (13C) for C6D6; 5.31 (1H) and
53.80 (13C) for CD2Cl2; 7.26 (1H) and 77.00 (13C) for CDCl3].
The 15N chemical shifts were referenced to liquid CH3NO2
(381.7 ppm)39 and are reported relative to liquid NH3. The 2D
NMR experiments GHMBC and GSQMBC were used for
assigning the individual 13C and 1H resonances.40
The diffraction data were collected with a KM4CCD four-
circle area-detector diffractometer (Oxford Diffraction, UK)
equipped with an Oxford Cryostream Cooler (Oxford Cryo-
systems, UK). We performed the ω-scan technique with
different κ and æ offsets in order to cover the entire indepen-
dent part of the reflections. Cell parameters were refined from
all the strong (stronger than 1000 pulses) reflections. Data
reduction was carried out using the program CrysAlis RED.
Direct methods in SHELXS-9741 were used to solve the
structures, and the structures were refined by using SHELXL-
97.42 The tables were prepared for publication by using
SHELXL and PARST,43 and the figures were generated with
ORTEP-III.44
Alk a loid s. Palmatine chloride (1a , X ) Cl) and berberine
chloride (1b, X ) Cl), isolated from J atrorrhiza palmata Miers.
and Berberis vulgaris L., respectively, were kindly donated by
Professor J irˇ´ı Slav´ık (Masaryk University, Brno, Czech Re-
public). The melting points (1a 205-210 °C, 1b 209-210 °C)
and spectral characteristics of these samples agree well with
previously published data.
Sp ectr oscop ic Stu d ies of Com p ou n d s 2a a n d 2b. (i)
Alkaloid 1 (10-30 mg) was dissolved in 1 mL of D2O, and 1
mL of benzene-d6 was added. After alkalization with 20%
NaOH (100 µL) and gentle shaking, the C6D6 phase was
separated and measured. (ii) Alkaloid 1 (10-30 mg) was
suspended in 1 mL of CD2Cl2, and 100 µL of 20% NaOH was
added. The CD2Cl2 phase was separated, dried with anhydrous
Na2CO3, and measured. For 1H and 13C NMR data see Tables
1-4, 15N NMR, see Table 5. 2a : EIMS m/z 369 [M]+ (3), 368
(19), 367 (54), 352 [M - OH]+ (100). 2b: EIMS (30 eV) m/z
353 [M]+ (3), 352 (17), 351 (51), 336 [M - OH]+ (100).
8-Meth oxy-7,8-d ih yd r op a lm a tin e (3a ). Palmatine chlo-
ride (1a , 20 mg) was stirred in 5 mL of a 20% solution of
MeONa in MeOH in an argon atmosphere. After 24 h the
solvent was evaporated and H2O (alkalized with NaOH) was
added. The product was extracted into diethyl ether and dried
with a mixture of NaOH and anhydrous Na2SO4. Evaporation
of the solvent yielded the yellow crystals of 3a (90%), mp 168-
172 °C (dec); for 1H and 13C NMR data, see Tables 1 and 2;
EIMS m/z 383 [M]+ (0.4), 353 (100), 352 [M - OMe]+ (61), 351
(54).
8-Meth oxy-7,8-d ih yd r ober ber in e (3b). This was pre-
pared like 3a , but from berberine chloride (1b). Evaporation
of the solvent yielded yellow crystals of 3b, mp 137-139 °C
(dec); for 1H and 13C NMR data, see Tables 3 and 4; for 15N
NMR data, see Table 5; EIMS m/z 367 [M]+ (0.8), 338 (1.4),
336 [M - OMe]+ (0.4), 57 (100).
Cr ysta l Da ta for 3b. CCDC ref. No. 193920. Crystallized
from diethyl ether, C21H21NO5, Mrel ) 367.39, T ) 120(2) K,
monoclinic, λ ) 0.71073 Å, space group P21/n, a ) 7.4433(4)
Å, b ) 11.0224(6) Å, c ) 21.3225(11) Å, â ) 100.04(3)°, V )
1722.57(16) Å3, Z ) 4, Dcalc ) 1.417 Mg/m3, crystal size 0.55 ×
0.50 × 0.30 mm, R ) 0.052.
8-Eth oxy-7,8-d ih yd r ober ber in e (4b). Berberine chloride
(1b, 15 mg) was stirred in 5 mL of a solution of EtONa in EtOH
in an argon atmosphere. After 24 h the solvent was evaporated
and H2O (alkalized with NaOH) was added. The product was
extracted into diethyl ether and dried with a mixture of NaOH
and anhydrous Na2SO4, and the solvent was evaporated; for
1H and 13C NMR data, see Tables 3 and 4.
Exp er im en ta l Section
Gen er a l Exp er im en ta l P r oced u r es. Melting points were
determined on a Mettler FP 51 apparatus and are uncorrected.
Mass spectra were measured on a Fisons TRIO 1000 quadru-
pole mass spectrometer using electron ionization (EI, 70 eV).
NMR spectra were recorded using a Bruker Avance DRX 500
spectrometer operating at frequencies of 500.13 MHz (1H) and
125.77 MHz (13C), a Bruker Avance 300 spectrometer operating
at frequencies of 300.13 MHz (1H), 75.48 MHz (13C), and 30.41
MHz (15N), and a Bruker Avance 600 spectrometer operating
at frequencies of 600.15 MHz (1H) and 150.92 MHz (13C). All
8-Tr ich lor om eth yl-7,8-d ih yd r op a lm a tin e (5a ). Palma-
tine chloride (1a , 80 mg) was stirred with 10 mL of CHCl3 and
0.5 mL of concentrated aqueous ammonia. After 24 h the
phases were separated and the organic phase was dried with
a mixture of NaOH and anhydrous Na2SO4. Evaporation of
the solvent yielded yellow crystals of 5a quantitatively, mp
1
1
NMR spectra were measured at 303 K. The H and 13C NMR
194-196 °C (dec); for H NMR data, see Table 1; for 13C NMR,

486 J ournal of Natural Products, 2003, Vol. 66, No. 4
Marek et al.
see Table 2; EIMS m/z 471 [M]+ (0.1), 436 (2.4), 435 [M - Cl]+
(0.5), 434 (3.6), 352 [M - CCl3]+ (100).
(14) Shamma, M.; Rahimizadeh, M. J . Nat. Prod. 1986, 49, 398-405.
(15) Slav´ık, J .; Slav´ıkova´, L. Collect. Czech. Chem. Commun. 1989, 54,
2009-2020.
Cr ysta l Da ta for 5a . CCDC ref. No. 193921. Crystallized
from benzene, C22H22Cl3NO4, Mrel ) 470.76, T ) 120(2) K,
triclinic, λ ) 0.71073 Å, space group P1, a ) 7.3910(6) Å, b )
12.2514(10) Å, c ) 12.8568(10) Å, R ) 63.240(8)°, â ) 87.858-
(7)°, γ ) 77.204(7)°, V ) 1011.08(14) Å3, Z ) 2, Dcalc ) 1.546
Mg/m3, crystal size 0.50 × 0.35 × 0.30 mm, R ) 0.035.
8-Tr ich lor om eth yl-7,8-d ih yd r ober ber in e (5b). This was
prepared like 5a , but from berberine chloride (1b). Evaporation
of the solvent yielded yellow crystals of 5b, mp 186-188 °C
(dec); for 1H NMR data, see Table 3, for 13C NMR, see Table
4; for 15N NMR, see Table 5; EIMS m/z 455 [M]+ (0.2), 420
(2.5), 419 [M - Cl]+ (0.4), 418 (3.8), 336 [M - CCl3]+ (100).
Cr ysta l Da ta for 5b. CCDC ref. No. 193922. Crystallized
from diethyl ether, C21H18Cl3NO4, Mrel ) 454.71, T ) 153(2)
K, triclinic, λ ) 0.71073 Å, space group P1, a ) 8.6994(6) Å,
b ) 10.7566(8) Å, c ) 11.0433(8) Å, R ) 69.321(7)°, â )
(16) Man, S.; Dosta´l, J .; Necˇas, M.; Zˇa´k, Z.; Pota´cˇek, M. Heterocycl.
Commun. 2001, 7, 243-248.
(17) Secˇka´rˇova´, P.; Marek, R.; Dosta´l, J .; Dommisse, R.; Esmans, E. L.
Magn. Reson. Chem. 2002, 40, 147-152.
(18) Kova´rˇ, J .; Stejskal, J .; Paulova´, H.; Slav´ık, J . Collect. Czech. Chem.
Commun. 1986, 51, 2626-2634.
(19) Sˇima´nek, V.; Preininger, V.; Lasovsky´, J . Collect. Czech. Chem.
Commun. 1976, 41, 1050-1055.
(20) Marek, R.; Sklena´rˇ, V.; Dosta´l, J .; Slav´ık, J . Tetrahedron Lett. 1996,
37, 1655-1658.
(21) Marek, R.; Tousˇek, J .; Kra´l´ık, L.; Dosta´l, J .; Sklena´rˇ, V. Chem. Lett.
1997, 369-370.
(22) Tousˇek, J .; Dommisse, R.; Dosta´l, J .; Zˇa´k, Z.; Pieters, L.; Marek, R.
J . Mol. Struct. 2002, 613, 103-113.
(23) Bax, A.; Summers, M. F. J . Am. Chem. Soc. 1986, 108, 2093-2094.
(24) Willker, W.; Leibfritz, D.; Kerssebaum, R.; Bermel, W. Magn. Reson.
Chem. 1993, 31, 287-292.
(25) Marek, R.; Kra´l´ık, L.; Sklena´rˇ, V. Tetrahedron Lett. 1997, 38, 665-
668.
(26) Khamidov, I. I.; Tashkhodzhaev, B.; Aripova, S. F.; Telezhenetskaya,
M. V.; Karimov, A. K. Khim. Prir. Soedin. 1996, 889-893.
(27) Ishii, H.; Takeda, S.; Ogata, K.; Hanaoka, M.; Harayama, T. Chem.
Pharm. Bull. 1991, 39, 2712-2714.
87.403(6)°, γ ) 83.486(6)°, V ) 960.55(12) Å3, Z ) 2, Dcalc
)
1.572 Mg/m3, crystal size 0.50 × 0.50 × 0.40 mm, R ) 0.032.
(28) Wafo, P.; Nyasse, B.; Fontaine, C. Phytochemistry 1999, 50, 279-
Ack n ow led gm en t. The authors thank Radovan Karesˇ for
measuring the EI mass spectra. The financial support of this
work by the Ministry of Education of the Czech Republic
(LN00A016) is gratefully acknowledged.
281.
(29) Tan, P. V.; Nyasse, B.; Enow-Orock, G. E.; Wafo, P., Forcha, E. A.
Phytomedicine 2000, 7, 291-296.
(30) Tan, P. V.; Nyasse, B.; Dimo, T.; Wafo, P., Akahkub, B. T. Pharmazie
2002, 57, 409-412.
(31) Li, M.; Chen, X.; Tang, Q.-M.; Zhang, J .-S. Planta Med. 2001, 67,
189-190.
Su p p or tin g In for m a tion Ava ila ble: GSQMBC spectrum of 2b
and perspective views of 3b and 5b. This material is available free of
charge via the Internet at http://pubs.acs.org.
(32) Nyasse, B.; Nkwengoua, E.; Sondengam, B.; Denier, C.; Willson, M.
Pharmazie 2002, 57, 358-361.
(33) Marek, R.; Lycˇka, A. Curr. Org. Chem. 2002, 6, 35-66.
(34) Martin, G. E.; Hadden, C. E. J . Nat. Prod. 2000, 63, 543-585.
(35) Marek, R.; Humpa, O.; Dosta´l, J .; Slav´ık, J .; Sklena´rˇ, V. Magn. Reson.
Chem. 1999, 37, 195-202.
Refer en ces a n d Notes
(1) Bhakuni, D. S.; J ain, S. In The Alkaloids; Brossi, A., Ed.; Academic
Press: New York, 1986; Vol. 28, Chapter 2, pp 95-181.
(2) Preininger, V. In The Alkaloids; Brossi, A., Ed.; Academic Press: New
York, 1986; Vol. 29, Chapter 1, pp 1-98.
(36) Marek, R.; Marek, J ., Dosta´l, J .; Ta´borska´, E.; Slav´ık, J .; Dommisse,
R. Magn. Reson. Chem. 2002, 40, 687-692.
(37) Dosta´l, J .; Marek, R.; Slav´ık, J .; Ta´borska´, E.; Pota´cˇek, M.; Sklena´rˇ,
V. Magn. Reson. Chem. 1998, 36, 869-872.
(3) Hu, J . P.; Takahashi, N.; Yamada, T. Oral Dis. 2000, 6, 297-302.
(4) Wright, C. W.; Marshall, S. J .; Russell, P. F.; Anderson, M. M.;
Phillipson, J . D.; Kirby, G. C.; Warhurst, D. C.; Schiff, P. L., J r. J .
Nat. Prod. 2000, 63, 1638-1640.
(38) Crystallographic data for the structures reported in this paper have
been deposited with the Cambridge Crystallographic Data Center
(CCDC). Copies of the data can be obtained, free of charge, on
application to the Director, CCDC, 12 Union Road, Cambridge CB2
1EZ, UK (fax: +44(0)1223-336033 or e-mail deposit@ccdc.cam.ac.uk).
(39) Wishart, D. S.; Bigam, C. G.; Yao, J .; Abildgaard, F.; Dyson, H. J .;
Oldfield, E.; Markley, J . L.; Sykes, B. D. J . Biomol. NMR 1995, 6,
135-140.
(5) Baird, A. W.; Taylor, C. T.; Brayden D. J . Adv. Drug Delivery Rev.
1997, 23, 111-120.
(6) Lau, C.-W.; Yao, X.-Q.; Chen, Z.-Y.; Ko, W.-H.; Huang, Y. Cardiovasc.
Drug Rev. 2001, 19, 234-244.
(7) Schmeller, T.; Latz-Bru¨ning, B.; Wink, M. Phytochemistry 1997, 44,
257-266.
(40) Marek, R.; Tousˇek, J .; Dosta´l, J .; Slav´ık, J .; Dommisse, R.; Sklena´rˇ,
V. Magn. Reson. Chem. 1999, 37, 781-787.
(8) Shamma, M. The Isoquinoline Alkaloids; Academic Press: New York,
1972; pp 268-314.
(41) Sheldrick, G. M. SHELXS97, Program for the Solution of Crystal
Structures; University of Go¨ttingen: Germany, 1997.
(42) Sheldrick, G. M. SHELXL97, Program for the Refinement of Crystal
Structures; University of Go¨ttingen: Germany, 1997.
(43) Nardelli, M. Comput. Chem. 1983, 7, 95-98.
(44) Burnett, M. N.; J ohnson, C. K. ORTEP-III: Oak Ridge Thermal
Ellipsoid Plot Program for Crystal Structure Illustrations; Oak Ridge
National Laboratory Report ORNL-6895, 1996.
(9) Bentley, K. W. The Isoquinoline Alkaloids; Harwood Academic
Publishers: Amsterdam, 1998; pp 219-263.
(10) Vennerstrom, J . L.; Klayman, D. L. J . Med. Chem. 1988, 31, 1084-
1087.
(11) Man, S.; Pota´cˇek, M.; Necˇas, M.; Zˇa´k, Z.; Dosta´l, J . Molecules 2001,
6, 433-441.
(12) Contreras, M. L.; Rivas, S.; Rozas, R. Heterocycles 1984, 22, 101-
106.
(13) Miana, G. A. Phytochemistry 1973, 12, 1822-1823.
NP0204996