A Short Total Synthesis of rac-Peniolactol

Article abstract of DOI:10.1007/s00706-002-0515-6

A short facile synthesis of the fungal metabolite (±)-3,4-dihydro-3, 6,8-trihydroxy-3-pentadecylisocoumarin (peniolactol) has been achieved. Condensation of hexadecanoyl chloride with 3,5-dimethoxyhomophthalic acid afforded 6,8-dimethoxy-3-pentadecylisoco

Full text of DOI:10.1007/s00706-002-0515-6

Monatshefte f€ur Chemie 134, 457–463 (2003)
DOI 10.1007/s00706-002-0515-6
A Short Total Synthesis of rac-Peniolactol
ꢀ
Aamer Saeed
Department of Chemistry, Quaid-I-Azam University, Islamabad-45320, Pakistan
Received May 23, 2002; accepted May 30, 2002
Published online January 7, 2003 # Springer-Verlag 2003
Summary. A short facile synthesis of the fungal metabolite ( ꢁ )-3,4-dihydro-3,6,8-trihydroxy-3-
pentadecylisocoumarin (peniolactol) has been achieved. Condensation of hexadecanoyl chloride with
3,5-dimethoxyhomophthalic acid afforded 6,8-dimethoxy-3-pentadecylisocoumarin, which on sequen-
tial saponification and demethylation furnished rac-peniolactol in 31% overall yield. The ring-chain
1
tautomerism was studied in solution by H NMR and the mass fragmentation pattern. 3-Pentadecyl-
isocoumarin was also synthesized and saponified to the corresponding keto acid as a model compound
for comparative studies.
Keywords. rac-Peniolactol; Peniophora sanguinea; Isocoumarins; Ring-chain tautomerism.
Introduction
Peniolactol [1] was isolated from the wood attacked by the fungus Peniophora
sanguinea Bres. and was assigned the constitution of 3,4-dihydro-3,6,8-trihy-
droxy-3-pentadecylisocoumarin. It can exist in two tautomeric forms: the open
chain keto acid (1a) and the cyclic hemiacetalic lactol form (1b) (Scheme 1). This
feature is also exhibited by other 3-hydroxy-3-alkyl-3,4-dihydroisocoumarins like
ustic acid [2], sclerotinin A and B [3], 3-hydroxymellein, 3,4,6-trihydroxymellein
[4], 3,6-dihydroxymellein [5], and the lactols of the biogenetic precursors of
cochlioquinones and stemphone [6]. Presumably, owing to this tautomeric inter-
conversion at the prochiral center peniolactol occurs as a racemate unlike other
3-substituted dihydroisocoumarins, the majority of which occur in nature as the
(R)-enantiomers [7–9].
We have already reported on the total synthesis [10] of rac-peniolactol by a
multistep strategy developed during the synthesis of the main dihydroisocouma-
rins [11] of the plant Ononis natrix. Peniolactol is a potential antifungal and anti-
bacterial agent, thus a detailed study of the biological activity required a short and
efficient synthetic route. Condensation of acid chlorides with homophthalic acids
has recently emerged as a standard and authentic route for the construction of the
ꢀ
E-mail: aamersaeed@yahoo.com

458
A. Saeed
Scheme 1
3-substituted isocoumarin skeleton. Herein, a facile three step synthesis of penio-
lactol using this method will be described.
Results and Discussion
The key starting substance for the synthesis of rac-peniolactol, 3,5-dimethoxy-
homophthalic acid (2) was prepared from the corresponding phenylacetic acid
according to the procedure reported by us earlier [12]. Direct reaction of 3,5-
dimethoxyhomophthalic acid with hexadecanoyl chloride at elevated temperature
afforded 6,8-dimethoxy-3-pentadecylisocoumarin (3) in 80% yield [13] (Scheme 2).
This compound showed the characteristic 1H singlet of the isocoumarin moiety at
ꢀ ¼ 6.06 for H-4 in the 1H NMR and signals at ꢀ ¼ 99.71(C-4) and 159.54 (C-3) in
the 13C NMR spectrum. The mass spectrum showed the characteristic isocoumarin
fragment ion at m=z 177, and in its IR spectrum the lactonic carbonyl absorption was
observed at 1734 cmÀ 1. 3-Pentadecylisocoumarin (3a), prepared similarly from
commercial homophthalic acid gave a singlet at ꢀ ¼ 6.24 for H-4 and signals at
ꢀ ¼ 103.91 (C-4) and 158.51 (C-3) in the 1H and 13C NMR spectra. The IR spectrum
showed the lactonic carbonyl absorption at 1728 cmÀ 1
.
Alkaline hydrolysis of the isocoumarin 3 to furnish 4,6-dimethoxy-2-(2-oxo-
heptadecyl) benzoic acid (4a) was accomplished in good yield. The keto acid
Scheme 2

Synthesis of rac-Peniolactol
459
existed partially in equilibrium with its cyclic tautomeric form 6,8-dimethoxy-3-
hydroxy-3-pentadecyl-3,4-dihydroisocoumarin (4b) as evidenced by the diastereo-
topy of prochiral benzylic protons in the 1H NMR spectrum. Thus, in addition to a
2H singlet at ꢀ ¼ 4.05 (H-10 open chain form 4a), each of the protons of ArCH2
showed a double doublet AB pattern at ꢀ ¼ 3.60–3.80, J ¼ 7.1 Hz (H-4 lactol form
1
4b) in the H NMR and at ꢀ ¼ 50.23 (C-10 open chain) in the 13C NMR spectrum.
DEPT 90ꢂ and DEPT 135ꢂ experiments confirmed these assignments. The mass
spectrum showed the characteristic Mþ -H2O peak at m=z 416. The IR spectrum
showed the carboxyl and ketonic carbonyl absorptions at 1683 and 1716 cmÀ 1
,
respectively. The unsubstituted keto acid 4c obtained from the hydrolysis of 3-
pentadecylisocoumarin showed a sharp 2H singlet at ꢀ ¼ 4.10 or a mutiplet at
1
ꢀ ¼ 4.01–4.18 and 77.29 (C10) in the H and 13C NMR spectra.
Selective demethylation of the 4,6-dimethoxy keto acid lactol tautomeric mix-
ture 4ab using BBr3 occurred rapidly under mild conditions ( À 78ꢂC, 10 min) to
furnish 6-hydroxy-4-methoxy-2-(2-oxoheptadecyl) benzoic acid (5a), which pre-
dominantly existed as 3,8-dihydroxy-6-methoxy-3-pentadecyl-3,4-dihydroisocou-
marin (5b) as shown by a quartet at ꢀ ¼ 2.91–3.25, J ¼ 8.54 Hz as compared to a
very feeble singlet at ꢀ ¼ 3.95 for the keto acid form.
The crucial step was the final deprotection to obtain 1. Although BBr3 is the
reagent of choice for complete cleavage of O-methyl ethers of natural products, in
case of a keto acid like 4a, which is highly acid sensitive, the use of a strong Lewis
acid may lead to side reactions, the major competing side reaction being the acid
catalyzed dehydration to the parent dimethoxyisocoumarin 3. If the dehydration
would occur after complete or partial demethylation instead of peniolactol the
corresponding isocoumarins would have resulted. Fortunately, the only major side
product was the parent dimethoxy isocoumarin 3, which could easily be separated
and recycled.
The rac-peniolactol was characterized by the complete absence of MeO-8 and
1
MeO-6 singlets, both in the H NMR and 13C NMR spectra, and by the downfield
shift of the characteristic ArCH2 singlet from 4.05 to 3.94 ppm (open chain 1a) or
the quartet at 2.86–2.65 ppm, J ¼ 7.52 Hz (lactol form 1b) in addition to other
characteristic changes. Thus, the present synthesis constitutes a highly simplified
route to rac-peniolactol (1) involving three linear steps. It proceeds with a 31%
overall yield, which makes 1 available for biological evaluation. Peniolactol is
expected to possess potent antifungal and antibacterial activities as shown by the
related isocoumarins [14]. These studies are in progress.
Some interesting observations regarding the interconversion of keto acid and
lactol forms could be made. Previous studies on such keto acid lactol equilibria
have reported that keto acids with a methoxy group adjacent to the carboxyl exist
mainly in the open chain form, whereas those with a hydroxy substituent at this
position predominantly exist in the lactol form. This is possibly due to internal
1
chelation, both in solution and solid state as shown by their H NMR and X-ray
structures [5, 15]. In the present study it was observed that initially the dimethoxy
keto acid 4a existed only in the keto acid form (2H singlet at ꢀ ¼ 4.05 for CH2-10)
but on standing appreciable conversion to the lactol form 4b occurred as shown by
the clean eight line AB pattern (dddd) of the Ar-CH2 protons at ꢀ ¼ 3.60–3.80,
4
J ¼ 7.1, 6.1 Hz most probably due to J coupling with the aromatic proton. Partial

460
A. Saeed
Scheme 3
demethylation to the hydroxy keto acid 5 also gave an AB pattern at ꢀ ¼ 3.25–2.91,
J ¼ 8.54 Hz but only four lines were visible showing that it mainly existed in the
lactol form 5b. In case of peniolactol again initially a singlet at ꢀ ¼ 3.94 (open
chain 1a) was observed which became a quartet at ꢀ ¼ 2.86–2.65, J ¼ 7.4 Hz (1b).
These observations indicated that although in the solid state the 2-hydroxy keto
acids may exist exclusively in the lactol form, in solution (CDCl3) interconversion
between both forms continues and the dimethoxy acids may also exist as tauto-
meric mixture of two forms in accordance with a recent report on similar keto acids
[16]. Even for the unsubstituted keto acid 4c initially a sharp singlet at ꢀ ¼ 4.10 was
observed which became a mutiplet at ꢀ ¼ 4.01–4.18 on standing indicating that
these are also prone to tautomerization. The mass fragmentation pattern of 1, 4, and
5 also support their existence in the lactol form. Thus, in each case the relative
intensity of cation radicals obtained from the lactol form by elimination of
C14H29OH and C15H31CO2H was much higher as compared to that obtained from
the open chain keto acid form due to elimination of the pentadecanoyl group
(Scheme 3).
Experimental
Commercial ethyl acetate and petroleum ether (60–80ꢂC) were distilled before use, CH2Cl2 was dried
over CaH2 and distilled. 1H NMR and 13C NMR spectra were recorded in CDCl3 on Bruker AM-400
and AM-100 instruments, respectively, and J values are reported in Hz. IR spectra were recorded on a

Synthesis of rac-Peniolactol
461
Bruker Vector 22, the mass Spectra (EI, 70 eV) on a MAT 312 instrument, and the elemental analyses
€
with a CHN-Rapid Heraus. The obtained values agreed favourably with the calculated ones. Flash
Column chromatography (FCC) was carried out on Merck Kieselgel 60 (230–400mesh).
6,8-Dimethoxy-3-pentadecylisocoumarin (3, C26H40O4)
A stirred mixture of 0.5 g 2 (2.08 mmol) and 2.29g hexadecanoyl chloride (8.33 mmol) was heated on
an oil bath at 200ꢂC for 3 h. FCC of the residue (petroleum ether:ethyl acetate ¼ 8:2) afforded a white
solid which on recrystallization from methanol gave 0.69 g isocoumarinþ3 (80%) as colourless prisms.
Mp 101–103ꢂC (Ref. [10] 102 –103ꢂC); MS (70eV): m=z (%) ¼ 416 (M ); IR (film): ꢁ ¼ 2913, 2849,
1734, 1694, 1598, 1572, 1471, 1151, 832 cmÀ 1; 1H NMR (CDCl3, 400MHz): ꢀ ¼ 0.86 (t, J ¼ 7.12 Hz,
3H, H-150), 1.23 (br s, 24H, H30–H140), 1.67 (p, J ¼ 8.4Hz, 2H, H20), 2.44 (t, J ¼ 7.0 Hz, 2H, H-10),
3.87 (s, MeO-8), 3.93 (s, MeO-6), 6.06 (s, H-4), 6.30 (d, J ¼ 2.12 Hz, H-5), 6.40 (d, J ¼ 2.12 Hz, H-7)
ppm; 13C NMR (CDCl3,100MHz): ꢀ ¼ 165.8 (C1), 163.6 (C8), 159.5 (C6), 142.9 (4a), 103.1 (C4),
98.4 (C7), 99.7 (C5), 56.5 (MeO-8), 55.8 (MeO-6), 33.6 (C10), 32.1 (C130), 29.88, 29.86, 29.77, 29.7,
29.6, 29.5, 29.4, 29.3, 29.2, 29.1 (C40–C120), 27.0 (C20), 23.7 (C30), 22.9 (C140), 14.3 (C150) ppm.
3-Pentadecylisocoumarin (3a, C24H36O2)
Prepared in a manner similar to 3 from commercial homophthalic acid and the product flash chromato-
graphed from pure pet ether to afford 3a as colourless needles in 82% yield. Mp 58–59ꢂC; MS (70eV):
m=z (%) ¼ 356 [Mþ ] (97.7), 336 (34.2), 314 (63.7), 160 (100), 118 (97.7); IR (film): ꢁ ¼ 2918, 2849,
1
1728, 1712, 1656, 1604, 1160, 642 cmÀ 1; H NMR (CDCl3, 400 MHz): ꢀ ¼ 0.87 (t, J ¼ 6.28 Hz,
3H, H-150), 1.28 (br s, 24H, H30–H140), 1.70 (p, J ¼ 8.4 Hz, 2H, H20), 2.52 (t, J ¼ 7.08 Hz, 2H, H-
10), 6.24 (s, H-4), 7.34 (d, J ¼ 8.16 Hz, H-5), 7.49 (td, J ¼ 0.88, 7.28 Hz, H-7), 7.65 (m, H-6), 8.25 (d,
J ¼ 8.16Hz, H-8) ppm; 13C NMR (CDCl3, 100MHz): ꢀ ¼ 163.2 (C1), 158.5 (C3), 137.8 (C8a), 134.8
(C6), 129.6 (C8), 127.6 (C7), 125.1 (C5), 120.3 (C4a), 103.9 (C4), 33.7 (C10), 32.0 (C130), 29.88, 29.86,
29.82, 29.80, 29.7, 29.70, 29.6, 29.5 (C30–C120), 27.0 (C20), 22.8 (C140), 14.2 (C150) ppm.
2,4-Dimethoxy-6-(2-oxo-heptadecyl)-benzoic acid (4a) or 6,8-Dimethoxy-
3-hydroxy-3-pentadecyl-3,4-dihydroisocoumarin (4b) (C26H42O5)
A stirred solution of 0.5g 3 (1.20 mmol) in ethanol 20 cm3 was treated with 40cm3 5%KOH and the
mixture refluxed for 4 h. After cooling the reaction mixture, most of the ethanol was rotary evaporated.
Cold water (20 cm3) was added and the mixture acidified with dil. HCl and extracted with 2ꢃ30cm3
dichloromethane. The organic phase was dried (MgSO4), and the solvent evaporated under vacuum to
leave 4ab as a white solid. Recrystallized from MeOH as colourless scales, 0.36 g (70%). Mp 109–111ꢂC
(Ref. [10] 112–115ꢂC); MS: m=z (%) ¼ 434 [Mþ ] (24), 416 (42), 417 (37), 391 (11), 220 (13), 197 (17),
178 (100); IR (film): ꢁ ¼ 2915, 2849, 1713, 1694, 1601, 1202, 1162cmÀ 1; 1H NMR (CDCl3, 400MHz):
ꢀ ¼ 0.87 (t, J ¼ 6.88 Hz, 3H, H-170), 1.25 (m, 24H, H30–H140), 1.54–1.67 (p, 2H, H40), 2.58–2.62 (t,
J ¼ 6.67, 2H, H-30), 3.60–3.80 (dddd, 2H, J ¼ 7.16, 6.9 Hz H4 lactol form 4b), 3.85 (s, MeO-4) 4.0 (s,
MeO-6), 4.05 (s, 2H, H10 keto acid form 4a), 6.39 (s, H-3), 6.48 (s, H-5) ppm; 13C NMR (CDCl3,
100 MHz): ꢀ ¼ 207.83 (C¼ O), 163.07 (COOH), 107.20 (C3), 160.36 (C4), 160.36 (C6), 111.78 (C1),
98.14 (C5), 55.70 (MeO-6), 57.03 (MeO-8), 50.21 (C10), 42.98 (C30), 32.06 (C150), 29.82, 29.80,
29.77, 29.63 (C50–C140), 22.82 (C160), 14.25 (C170) ppm (C=H numbering according to keto acid
form 4a).
2-(2-Oxoheptadecyl) benzoic acid (4c, C24H38O3)
Prepared in a manner similar to 4 fþrom 3-pentadecylisocoumarin in 75% yield as colorless needles. Mp
79–80ꢂC. MS: m=z (%) ¼ 374 [M ] (11), 356 (37), 221 (19), 160 (79), 135 (24), 118 (31); IR (film):
ꢁ ¼ 2915, 2849, 1713, 1694, 1601, 1202, 1162cmÀ 1
;
1H NMR (CDCl3, 400 MHz): ꢀ ¼ 0.87 (t,
J ¼ 6.64 Hz, 3H, H-170), 1.27 (m, 24H, H30–H160), 1.67 (m 2H, H40), 2.51 (brs, 2H, H-10), 4.1 (s,

462
A. Saeed
2H, H10 keto acid) or 4.05–4.18 (p, J ¼ 7.16Hz, 2H, H-4 lactol), 7.20 (d, J ¼ 5.76Hz, H-3), 7.39 (td,
J ¼ 0.88, 8.04 Hz, H-4), 7.54 (dt, J ¼ 7.16 Hz, H-5), 8.2 (d, J ¼ 8.04Hz, H-6) ppm; 13C NMR (CDCl3,
100 MHz): ꢀ ¼ 203.1 (C20), 197.9 (COOH), 132.7 (C3), 131.8 (C4), 127.4 (C5), 77.2 (C10), 125.1 (C6),
33.7 (C150) 32.0 (C30), 29.84, 29.80, 29.77, 29.72, 29.5, 29.4, 29.3, 29.2, 29.1 (C50–C140), 24.6 (C50),
22.8 (C160), 14.2 (C170) ppm (C=H numbering according to keto acid form 4a).
6-Hydroxy-4-methoxy-2-(2-oxoheptadecyl) benzoic acid (5a) or 3,4-Dihydro-3,8-
dihydroxy-6-methoxy-3-pentadecylisocoumarin (5b) (C25H40O5)
A 1 M solution of BBr3 in 1 cm3 CH2Cl2 (1.1mmol) was added dropwise to a stirred solution of 0.16 g
4ab (0.32 mmol) at À 78ꢂC in 4 cm3 dry CH2Cl2 under Ar. After stirring for 10min the reaction
mixture was poured into 50 cm3 ice-water. The layers separated and the aqueous layer extracted with
2ꢃ30 cm3 CH2Cl2 and then 30 cm3 EtOAc. The combined organic phases were dried (MgSO4) and
concentrated. Flash chromatography (petrol ether:ethyl acetate ¼ 7:3) afforded 0.1 g 5ab as colourless
prisms (75%) and some recovered isocoumarin 3. Mp119–120ꢂC; MS: m=z (%) ¼ 421 [Mþ , 1] (11.3),
420 [Mþ ] (24), 402 (22), 377 (50), 240 (32), 164 (100), 137 (22); IR (film): ꢁ ¼ 3410, 1716, 1631,
1
1583, 1157 cmÀ 1; H NMR (CDCl3, 400 MHz): ꢀ ¼ 0.87 (t, J ¼ 6.40Hz, 3H, H-150), 1.20 (brs, 24H,
H30–H140), 1.54–1.67 (m, 2H, H20), 1.85–1.97 (t, J ¼ 8.0 Hz, 2H, H-10), 2.91–3.25 (q, 2H, H-4,
J ¼ 8.54Hz, lactol form 5b), 3.85 (s, MeO-6) 3.95 (s, 2H, H10 open chain 5a), 6.29 (d, J ¼ 2.62 Hz,
H-5), 6.37 (d, J ¼ 2.0 Hz, H-7), 11.80 (s, 1H, OH) ppm (for numbering see lactol form 5b).
rac-4,6-Dihydroxy-2-(2-oxoheptadecyl) benzoic acid (1a) or 3,6,8-Trihydroxy-
3,4-dihydro-3-pentadecylisocoumarin (peniolactol) (1b) (C24H38O5)
To a stirred solution of 0.2 g 4ab (46 mmol) in 6 cm3 dry CH2Cl2 at À 78ꢂC was added dropwise
2.76cm3 1 M of a solution of BBr3 (2.76 mmol) in dry CH2Cl2 under Ar. After stirring for 1 h at
À 78ꢂC, the reaction mixture was warmed to room temperature and stirred for 24h. Then it was poured
into 50 cm3 ice–H2O. The layers were separated and the aqueous layer extracted with 2ꢃ50cm3
CH2Cl2 and then 50 cm3 EtOAc. The combined organic phases were dried (MgSO4) and concentrated.
Flash chromatography (petrol ether:ethyl acetate ¼ 6:4) afforded 0.102g 1ab as colourless prisms
(55%) and some recovered isocoumarin 3. Mp 147–149ꢂC (Ref. [1] 150ꢂC); MS: m=z (%) ¼ 406
[Mþ ] (24), 388 (71), 195 (32), 178 (76), 167 (89), 150 (100); IR (film): ꢁ ¼ 3709, 3598, 2922, 2853,
1710, 1653, 1459, 1260, 1081 cmÀ 1; 1H NMR (CDCl3, 400 MHz): ꢀ ¼ 0.87 (t, J ¼ 7.2 Hz, 3H, H-150),
1.24 (brs, 24H, H30–H140), 1.57 (m, 2H, H20), 2.35 (t, J ¼ 7.52 Hz, 2H, H-10), 2.86–2.65 (q, 2H,
J ¼ 7.6 Hz H-4 lactol form 1b), 3.94 (s, 2H, H10 keto acid form 1a), 6.73 (s, H-5), 7.09 (s, H-7), 10.50
(br, 1H, 6-OH), 11.22 (br, OH) ppm; 13C NMR (CDCl3, 100 MHz): ꢀ ¼ 170.2 (C1), 110.2 (C3), 37.29
(C4), 144.5 (C4a), 107.0 (C5), 166.1 (C6), 99.5 (C7), 164.5 (C8), 101.3 (8a), 35.3 (C10), 23.4 (C20),
29.82, 29.8, 29.7, 29.67, 29.6, 29.5, 29.4, 29.3, 29.2, 29.1 (C30–C120), 31.9 (C130), 22.7 (C140), 14.2
(C150) ppm (C=H numbering according to lactol form 1b).
Acknowledgements
€
The author is indebted to Prof. H. Hopf, Institut fur Organische Chemie, TU Braunschweig, Germany,
for his kind support and valuable discussions. Author also gratefully acknowledges a Georg Forster
research fellowship from the Alexander von Humboldt-Foundation Germany.
References
[1] Gripenberg J (1974) Acta Chem Scand Ser B 28: 505
[2] Raistrick H, Stickings CE (1951) Biochem J 48: 53; Vora VC (1954) J Sci Ind Res (India)
13B: 842

Synthesis of rac-Peniolactol
463
[3] Sassa T, Aoki H, Munakata K (1968) Tetrahedron Lett, 5703; Sassa T, Aoki H, Namiki M,
Munakata K (1968) Agric Biol Chem 32: 1432; Tanaka K, Kobayashi A, Yamashita K (1973)
Agric Biol Chem 37: 669
[4] Grove GF, Pople M (1979) J Chem Soc Perkin Trans I, 337; Money T, Comer FW, Webster GRB,
Wright IG, Scott AI (1967) Tetrahedron 23: 3435; Steyn PS, Holzapfel CW, Ferreira NP (1970)
Phytochemistry 9: 1977
[5] Kameda K, Aoki H, Tanaka H, Namiki M (1973) Agric Biol Chem 37: 2137
[6] Colombo L, Gennari C, Santandrea M, Narisano E, Scolastico C (1980) J Chem Soc Perkin
Trans I, 136
[7] Speranza G, Manitto P, Cassara P, Monti D (1993) Phytochemistry 33: 175
[8] Ramacciotti A, Fiasch R, Napolitano E (1996) J Org Chem 61: 5371
[9] Feliciano AS, Miguel del Corral JM, Canedo LM, Medarde M (1990) Phytochemistry 29: 945
[10] Rama NH, Saeed A, Bird CW (1995) Liebigs Ann Chem 4: 711
[11] Rama NH, Saeed A, Bird CW (1993) Liebigs Ann Chem 12: 1331
[12] Saeed A, Rama NH (1993) J Chem Soc Pak 15: 140
[13] Nozawa K, Yamada M, Tsuda Y, Kawai KI, Nakajima S (1981) Chem Pham Bull 29: 2491, 3486
[14] Rama NH, Iqbal R, Zamani Kh, Saeed A, Iqbal MZ, Chaudary MI (1998) Indian J Chem 7B: 365
[15] Kawai KI, Ito H, Nagase H, Yamaguchi R, Nakajima S (1985) Acta Cryst C41: 415
€
[16] Krohen K, Bahramsari R, Florke U, Ludewig K, Nakajima C, Klichespory A, Aust Michel HJ,
Raeger S, Schulz B, Antus S (1997) Phytochemistry 45: 313