Isomeric kielcorins and dihydroxyxanthones: Synthesis, structure elucidation, and inhibitory activities of growth of human cancer cell lines and on the proliferation of human lymphocytes in vitro

Article abstract of DOI:10.1002/1522-2675(200209)85:9<2862::AID-HLCA2862>3.0.CO;2-R

The synthesis, structure elucidation, and biological activities of five isomeric xanthonolignoids, (±)-transkielcorin C, (±)-cis-kielcorin C, (±)-trans-kielcorin D, (±)-trans-isokielcorin D, and (±)-trans-kielcorin E, are reported. The synthetic approach is based on the oxidative coupling of coniferyl alcohol with an appropriate xanthone. The influence of different oxidizing agents was studied, and the best results were obtained with potassium hexacyanoferrate(III). The structure elucidation was achieved by 2D-NMR techniques such as COSY, HETCOR, HSQC, and HMBC. Long-range C,H connectivities were used to establish the orientation of the substituents on the 1,4-dioxine rings, while NOE experiments were used to determine the configurations of these rings. These xanthonolignoids, as well as (±) -trans-kielcorin, (±)-trans-kielcorin B, (±)-trans-isokielcorin B, and the xanthonic building blocks 3,4-, 1,2-, and 2,3-dihydroxy-9H-xanthen-9-ones, and 2,3-dihydroxy-4-methoxy-9H-xanthen-9-one, were evaluated for their in vitro effect on the growth of three human cancer cell lines, MCF-7 (breast), TK-10 (renal), and UACC-62 (melanoma), and on the proliferation of human lymphocytes.

Full text of DOI:10.1002/1522-2675(200209)85:9<2862::AID-HLCA2862>3.0.CO;2-R

2
862
Helvetica Chimica Acta ± Vol. 85 (2002)
Isomeric Kielcorins and Dihydroxyxanthones: Synthesis, Structure
Elucidation, and Inhibitory Activities of Growth of Human Cancer Cell Lines
and on the Proliferation of Human Lymphocytes In Vitro
a
b
c
a
a
by EmÌlia P. Sousa ) ), Artur M. S. Silva ), Madalena M. M. Pinto* ), Madalena M. Pedro ), F a¬ tima A. M.
a
a
Cerqueira ), and Maria S. J. Nascimento )
a) Centro de Estudos de QuÌmica Org a√ nica, FitoquÌmica e Farmacologia da Universidade do Porto, Faculdade
de Farm a¬ cia, Rua AnÌbal Cunha, 164, 4050-047, Porto, Portugal (fax: ‡351-222057358; e-mail:
madalena@ff.up.pt)
b) Instituto Superior Ci e√ ncias da Sa u¬ de-Norte, Rua Central da Gandra, 1317, 4580 Gandra, Portugal
c
)
Departamento de QuÌmica, Universidade de Aveiro, 3810-193 Aveiro, Portugal
The synthesis, structure elucidation, and biological activities of five isomeric xanthonolignoids, (Æ)-trans-
kielcorin C, (Æ)-cis-kielcorin C, (Æ)-trans-kielcorin D, (Æ)-trans-isokielcorin D, and (Æ)-trans-kielcorin E, are
reported. The synthetic approach is based on the oxidative coupling of coniferyl alcohol with an appropriate
xanthone. The influence of different oxidizing agents was studied, and the best results were obtained with
potassium hexacyanoferrate(III). The structure elucidation was achieved by 2D-NMR techniques such as
COSY, HETCOR, HSQC, and HMBC. Long-range C,H connectivities were used to establish the orientation of
the substituents on the 1,4-dioxine rings, while NOE experiments were used to determine the configurations of
these rings. These xanthonolignoids, as well as (Æ)-trans-kielcorin, (Æ)-trans-kielcorin B, (Æ)-trans-isokielcorin
B, and the xanthonic building blocks 3,4-, 1,2-, and 2,3-dihydroxy-9H-xanthen-9-ones, and 2,3-dihydroxy-4-
methoxy-9H-xanthen-9-one, were evaluated for their in vitro effect on the growth of three human cancer cell
lines, MCF-7 (breast), TK-10 (renal), and UACC-62 (melanoma), and on the proliferation of human
lymphocytes.
1
. Introduction. ± Natural kielcorins are xanthonolignoids isolated from several
plants of Clusiaceae [1]. The biomimetic synthesis of kielcorin (1) [2] and kielcorin B
2) [3], as well as isokielcorin B (3) [3] (Fig. 1), by oxidative coupling of 3,4-
(
dihydroxy-2-methoxy-9H-xanthen-9-one (4) and 2,3-dihydroxy-4-methoxy-9H-xanth-
en-9-one (5), respectively, with coniferyl alcohol (ˆ4-(3-hydroxyprop-1-enyl)-2-
methoxyphenol; 6) was previously achieved (Fig. 2). These kielcorins, as well as their
xanthonic building blocks, were assayed for their biological activities and exhibited
interesting hepatoprotective activity against (tert-butyl)hydroperoxide-induced toxic-
ity in isolated rat hepatocytes [4].
To obtain related bioactive compounds with a kielcorin framework, five constitu-
tional isomers were designed. In this paper, the synthesis of trans- and cis-isomers of
(
[
3
Æ)-2,3-dihydro-3-(4-hydroxy-3-methoxyphenyl)-2-(hydroxymethyl)-7H-1,4-dioxino-
2,3-c]xanthen-7-one, 7 and 8, respectively, and the (Æ)-trans-2,3-dihydro-3-(4-hydroxy-
-methoxyphenyl)-2-(hydroxymethyl)-12H-1,4-dioxino[2,3-a]xanthen-12-one (9), as
well as (Æ)-trans-2,3-dihydro-2-(4-hydroxy-3-methoxyphenyl)-3-(hydroxymethyl)-
1
2H-1,4-dioxino[2,3-a]xanthen-12-one (10), and (Æ)-trans-2,3-dihydro-3-(4-hydroxy-
3-methoxyphenyl)-2-(hydroxymethyl)-11H-1,4-dioxino[2,3-b]xanthen-11-one (11) are
reported (Fig. 1). The synthetic approach, based on a biomimetic pathway as a model,

Helvetica Chimica Acta ± Vol. 85 (2002)
2863
Fig. 1. Structures of kielcorins 1 ± 3 and 7 ± 11
was by oxidative coupling of coniferyl alcohol (6) with an appropriate xanthone, 3,4-
dihydroxy- (12), 1,2-dihydroxy- (13), and 2,3-dihydroxy-9H-xanthen-9-one (14), for
the C, D, and E series of kielcorins, respectively (Scheme 1). Different oxidizing agents
were used (e.g., A gO, Ag CO , and K [Fe(CN) ]) to investigate the oxidative coupling
2
2
3
3
6
reaction [5]. The total synthesis of demethoxykielcorin 7 has already been accom-
plished in low yield by a multi-step pathway [6]. Herein, the cis-isomer 8 was efficiently
converted to the trans-isomer 7 [7].

2
864
Helvetica Chimica Acta ± Vol. 85 (2002)
Fig. 2. Structures of xanthones 4, 5 and 12 ± 14, coniferyl alcohol (6), and the dimeric products 21 and 22
Few data are available for biological activities of kielcorins [4][8]. In this work,
results of the effect of kielcorin derivatives 1 ± 3 and 7 ± 11, and xanthones 5 and 12 ± 14
on the growth of three human cancer cell lines MCF-7 (breast), TK-10 (renal), and
UACC-62 (melanoma), as well as on the proliferation of human lymphocytes are
described.
2
. Results and Discussion. ± 2.1. Synthesis of Kielcorin Derivatives with Several
Oxidizing Agents. In a series of studies on the synthesis of bioactive lignans,
flavonolignoids, and coumarinolignoids, several authors have investigated oxidative
coupling reactions [5][9]. For simpler compounds, it was shown that the substituents on
the catechol ring had a striking influence on the regioselectivity of the reaction; it was
highly regioselective in cases where the catechol skeleton possessed electron-donating
substituents and much less regioselective with catechols bearing electron-withdrawing
substituents [9]. Afree-radical coupling mechanism, with the first step being the
intermolecular O-b coupling of two phenoxyl radicals, has been suggested [10]. Further
studies on the synthesis of model compounds bearing a 3-aryl- or 2-aryl-1,4-
benzodioxin skeleton achieved high regioselectivity by oxidative coupling in the
presence of an Ag salt (oxide, carbonate) or K [Fe(CN) ], respectively [5a]. Based on
3
6
these studies, an investigation was initiated to explore the regio- and stereoselectivity of
the reaction with xanthones as one of the building blocks of benzodioxine rings that
bear a CˆO function as an electrophilic group and a heteroatom with electron-
donating properties. It appeared reasonable to investigate the oxidative coupling in the
presence of the same oxidizing agents (Ag O, Ag CO , and K [Fe(CN) ]).
2
2
3
3
6
The synthesis of 1,2- and 2,3-dihydroxy-9H-xanthen-9-one, 13 and 14, respectively
Scheme 2), involved as starting materials methyl 2-bromobenzoate (15) and 3,4-
(
dimethoxyphenol (16); from the resulting methyl 2-(3,4-dimethoxyphenoxy)benzoate
17) [11], after hydrolysis and subsequent cyclization of 2-(3,4-dimethoxyphenoxy)-
benzoic acid (18) with lithium diisopropylamide (LDA) [12] or with AcCl [13],
(

Helvetica Chimica Acta ± Vol. 85 (2002)
2865
Scheme 1
compounds 19 and 20, respectively, were obtained and demethylated to afford 13 and
4, respectively. 3,4-Dihydroxy-9H-xanthen-9-one (12) was synthesized by a method
1
described in [14]. For compounds 12 ± 14, and 20 are provided, for the first time,
spectral data from INEPT, HETCOR, COSY, or HMBC experiments to clarify the
structural elucidation of these compounds.
The oxidative coupling of the appropriate dihydroxyxanthones 12 ± 14 and an
equimolar amount of coniferyl alcohol (6) were carried out at room temperature, with
1
.5 mol equiv. of Ag O or A gCO , in toluene/Me CO, for 2 ± 3 days. In the case of the
2 2 3 2
other oxidizing agent, equimolar amounts of 12 ± 14 and 6 were treated with 20 mol-
equiv. of K [Fe(CN) ], at room temperature, in a mixture of Me CO/H O and an
3
6
2
2
aqueous 4% AcONa solution (Scheme 1).

2
866
Helvetica Chimica Acta ± Vol. 85 (2002)
Scheme 2
The xanthonolignoid distribution in the crude products of the oxidative coupling
reactions under the conditions indicated above is summarized in Table 1. The isomeric
nature of the kielcorin derivatives obtained did not differ with the oxidizing agents: for
the C series, (Æ)-trans-kielcorin C (7) and (Æ)-cis-kielcorin C (8) were obtained for all
the oxidizing agents; the D series gave two isomeric xanthonolignoids, (Æ)-trans-
kielcorin D (9) and (Æ)-trans-isokielcorin D (10); and the E series afforded (Æ)-trans-
kielcorin E (11). The regioselectivity achieved with Ag salts or with K [Fe(CN) ] for
3
6
simpler compounds was not verified for compounds 7 ± 11. In the C series, none of the
oxidizing agents used were stereoselective, although, when Ag O and Ag CO was
2
2
3
used, a lower amount of the cis-isomer 8 was obtained. Two dimeric compounds,
phenylcoumaran 21 and dehydrodiconiferyl alcohol 22, were also found as by-products
(
Fig. 2) in all reactions [3]. By carrying out the oxidative coupling with K [Fe(CN) ],
3 6
the quantities of 21 and 22 were markedly reduced, and a higher yield of the kielcorins
was obtained.
The most efficient method involved the use of K [Fe(CN) ] as oxidizing agent in the
3
6
oxidative coupling reaction (Table 1). Therefore, in the C series, the mixture of isomer
7
(trans) and 8 (cis) thus obtained was treated with anhydrous K CO [7] to yield
2 3
isomer 7, exclusively (Scheme 3).
The position of the OH substituents had a striking effect on the nature of the
kielcorin isomers obtained: when 3,4-dihydroxy-9H-xanthen-9-one (12) was used as a
building block, a mixture of isomers 7 (trans) and 8 (cis) was formed; with 1,2-
dihydroxy-9H-xanthen-9-one (13), the trans-isomers 9 and 10 were obtained, whereas
2,3-dihydroxy-9H-xanthen-9-one (14) gave only the trans-isomer 11. The electron-

Helvetica Chimica Acta ± Vol. 85 (2002)
2867
Table 1. Distribution of Xanthonolignoids 7 ± 11 in the Crude Reaction Products
Compound
Yields [%]
Ag
20
3.5
22
0.7
3.0
2
O
A
2
CO
3
g
3 6
K [Fe(CN) ]
7a
)
)
17
3.0
20
0.4
2.0
29
11
27
0.8
4.0
a
8
9
b
)
b
1
0 )
c
1
1 )
a)b)c) HPLC conditions: C18-Nucleosil column, 258, 1 ml/min. a) MeOH/1% AcOH in H
O 6 : 4, 239 nm.
2
b
O 5 : 5, 239 nm. c) MeOH/1% AcOH in H
)
MeCN/1% AcOH in H
2
2
O 7 : 3, 241 nm.
Scheme 3
withdrawing and electron-donating effects of the CˆO group, and the heteroatom of
these xanthones 12 ± 14 could influence the electronic character of the radical formed
on positions 3,4, 1,2, and 2,3 of the xanthone in different ways. The involvement of the
olefinic CˆC bound in the formation of an Ag complex can also be an explanation for
this kind of oxidative coupling reaction and deserves further investigation.
1
2
.2. Structure Elucidation of Kielcorins. The analysis of the H- and COSY-NMR
spectra of xanthonolignoids 7 ± 11 allowed the assignment of the ABCD spin system
corresponding to the resonances of HÀC(8), HÀC(9), HÀC(10), and HÀC(11) (in
7
9
± 10) and HÀC(7), HÀC(8), HÀC(9), and HÀC(10) (in 11). HÀC(8) (HÀC(11) in
and 10 and HÀC(10) in 11) (d 8.04 ± 8.20 ppm) is the more deshielded of these
protons, due to the mesomeric and anisotropic deshielding effect of the CˆO group,
followed by HÀC(10) in 7 and 8, HÀC(9) in 9 and 10, and HÀC(8) in 11 (d 7.80 ±
7
.87 ppm), which only feels the mesomeric effect. The assignment of the corresponding
1
3
C resonances were made from the correlations found in the HETCOR or HSQC
spectra, whereas those of the quaternary C-atoms of the unsubstituted ring were based
on the connectivities found in the HMBC and on the one-dimensional selective INEPT
experiments [15] of these compounds (HÀC(8) ! C(11a), C(7), C(10); HÀC(11) and
HÀC(9) ! C(7a)) (a and b; Fig. 3).
Due to the presence of electron-donating substituents on the 1,3,4-trisubstituted
1
phenyl rings, originated from the coniferyl alcohol, their H resonances were the more
shielded ones (HÀC(2'): 7.04 ± 7.10 ppm; HÀC(5'): 6.80 ± 6.83 ppm; HÀC(6'): 6.89 ±
13
6
.90 ppm). The corresponding protonated C resonances were assigned based on the
correlations found in the HETCOR or HSQC spectra, whereas those of the quaternary
C-atoms were based on the connectivities found in their HMBC spectra (HÀC(5') and
HÀC(3) ! C(1'); HÀC(5') ! C(3'); HÀC(2') ! C(4') and C(3) (a and b; Fig. 3).

2
868
Helvetica Chimica Acta ± Vol. 85 (2002)
Fig. 3. Main connectivities found in the HMBC and in the one-dimensional selective INEPT experiments of
xanthonolignoids 7, 9, and 11
The major differences among the NMR spectra of the xanthonolignoids synthesized
1
13
are due to the H and C resonances of the ring bearing the −benzodioxane× moiety. In
the case of (Æ)-trans-kielcorin C (7), HÀC(5) and HÀC(6) appear as two doublets, and
the connectivities found in the HMBC allowed the assignment of the quaternary C-
atoms C(4a), C(12a), C(12b), and C(6a) as depicted in Fig. 3 (a and b). However, the

Helvetica Chimica Acta ± Vol. 85 (2002)
2869
unequivocal identification of the structure of compound 7 as (Æ)-trans-kielcorin C was
accomplished from the one-dimensional selective INEPT [15]. Upon HÀC(3)
irradiation, with a long-range C,H coupling constant optimized for 1 Hz [15b],
enhancements on the C(4a) resonance was observed, allowing the establishment of the
connection of the 1,4-dioxane moiety to the xanthone ring. The assignments of all of the
1
13
H and C resonances of the substituted ring of the compounds 8 ± 11, and the
connection of their 1,4-dioxane moiety were deduced by a similar procedure (c ± f,
Fig. 3).
In the case of (Æ)-trans-isokielcorin D (10), the one-dimensional selective INEPT
did not allow the establishment of any connectivity between HÀC(2) or HÀC(3), and
1
13
the C-atoms of the xanthone ring. However, a comparison of the H and C resonances
on 1,4-dioxane ring with those of (Æ)-trans-kielcorin D (9) shows the downfield shifts of
C(2) of 10 relative to C(3) of 9. Comparable downfield shifts of C(2) of 9 relative to
C(3) of 10 are also observed (Fig. 4). In both cases, the downfield shifts are clearly
caused by the strong deshielded mesomeric effect exerted by the CˆO group, through
the O-substituent at C(12b) of both structures [3].
1
13
Fig. 4. Comparison of the H and C resonances of the 1,4-dioxane of (Æ)-trans-kielcorin D (9) and (Æ)-trans-
isokielcorin D (10)
3
Taking into consideration the coupling constant values J(2,3) ˆ 7.6 ± 8.1 Hz in the
1
H-NMR spectra of xanthonolignoids 7, 9 ± 11, and our previous report [3] allowed the
establishment of a trans-configuration for these two H-atoms. The absence of a NOE
effect between these two H-atoms supported this configurational assignment.
In the course of the synthesis of trans-kielcorin C (7), another compound was
isolated with similar structural features. The main difference is the coupling constant
3
value of J(2,3) ˆ 2.8 Hz, which suggests a cis-configuration for these two H-atoms. This
configuration was confirmed by the observed NOE enhancement (6%) in the HÀC(3)
resonance on irradiation of HÀC(2), supporting their close proximity. This compound
is named cis-isokielcorin C (8).
2
.3. Biological Activities. Table 2 summarizes the in vitro effects of (Æ)-trans-
kielcorin (1), (Æ)-trans-kielcorin B (2), (Æ)-trans-isokielcorin B (3), (Æ)-trans-
kielcorin C (7), (Æ)-cis-kielcorin C (8), (Æ)-trans-kielcorin D (9), (Æ)-trans-
isokielcorin D (10), (Æ)-trans-kielcorin E (11) and of the four xanthonic building
blocks 2,3-dihydroxy-4-methoxy-9H-xanthen-9-one (5), 3,4-, 1,2-, and 2,3-dihydroxy-

2
870
Helvetica Chimica Acta ± Vol. 85 (2002)
Table 2. Effects of Xanthone derivatives 1 ± 3, 5, and 7 ± 14 on the Growth of Human Cancer Cell Lines and
Proliferation of Human Lymphocytes
a
b
a c
Compound
GI50 [mm] ) )
IC50 [mm] ) )
Human
MCF-7
TK-10
(Renal)
UACC-62
(Melanoma)
lymphocytes
(
Breast)
1
5
2
3
2
7
8
3
9
0
4
1
53.5 Æ 8.0
72.1 Æ 5.6
76.6 Æ 7.0
88.5 Æ 8.6
30.9 Æ 5.0
36.6 Æ 2.4
19.8 Æ 1.6
80.8 Æ 11.6
25.4 Æ 1.3
21.6 Æ 2.6
26.2 Æ 1.3
33.3 Æ 2.1
14.0 Æ 0.3
> 100
> 100
d
37.2 )
17.4 Æ 2.7
62.5 Æ 13.3
16.2 Æ 3.4
40.5 Æ 1.5
18.3 Æ 0.7
32.0 Æ 2.4
38.4 Æ 2.7
> 100
82.8 Æ 4.3
40.6 Æ 1.3
21.2 Æ 8.2
> 100
41.9 Æ 1.6
12.2 Æ 1.3
36.2 Æ 7.4
48.6 Æ 6.1
73.3 Æ 2.2
> 100
d
1
59.2 )
25.9 Æ 2.7
45.2 Æ 7.8
65.8 Æ 5.1
97.5 Æ 10.5
> 100
1
1
1
1
> 100
> 100
61.4 Æ 4.3
31.7 Æ 3.7
31.2 Æ 1.5
23.1 Æ 0.5
21.9 Æ 3.9
> 100
a) Results are expressed as GI50 (concentrations of compounds that cause 50% inhibition of cancer cell growth)
or IC50 (concentrations that cause 50% inhibition of lymphocytes proliferation), and show means Æ SEM of 3 ± 6
b
independent observations performed in duplicate. ) Doxorubicin was used as positive control in cancer cell
lines growth (GI50 (MCF-7) ˆ 42.8 Æ 8.2 nm; GI50 (TK-10) ˆ 548.0 Æ 60.0 nm; GI50 (UACC-62)ˆ 94.0 Æ
9
)
.4 nm). c) Cyclosporin Awas used as positive control in lymphocyte proliferation ( IC50 ˆ 0.34 Æ 0.04 mm).
d
Data based on two independently run duplicate experiments.
9
H-xanthen-9-ones 12 ± 14 on the growth of the human cancer cell lines MCF-7, TK-10,
UACC-62, and on the proliferation of human lymphocytes.
.3.1. Effect on the Growth of Human Cancer Cell Lines. An inhibition of the growth
2
of the human cancer cell lines was observed with almost all kielcorins and oxygenated
xanthones studied. The growth inhibitory effect was, in general, moderate but was
shown to be dose-dependent and due to growth arrest and not to cell death, as inferred
from the sulforhodamine B (SRB) assay.
Considering the selectivity, the results in Table 2 showed that 2,3-dihydroxy-4-
methoxy-9H-xanthen-9-one (5), and 3,4- and 1,2-dihydroxy-9H-xanthen-9-ones (12
and 13, resp.) are significantly (P < 0.05) more active against UACC-62 than to MCF-7
and TK-10 cell lines.
The comparison between the effects of 3,4- and 2,3-dihydroxy-9H-xanthen-9-ones
(
12 and 14, resp.) with those of the (Æ)-trans-kielcorin derivatives 7 and 11,
respectively, revealed a significant (P < 0.05) increase of activity in MCF-7 and TK-
0 cell lines. No significant difference was observed for UACC-62 (P > 0.05). On the
1
contrary, compounds 2 and 9 were associated with a decrease and loss of growth-
inhibitory activity, respectively, in all of the cancer cell lines when compared with their
xanthonic building blocks 5 and 13.
The isomerism of the tested kielcorins 7 ± 11 seemed to influence the growth-
inhibitory activity of the human cancer cell lines: the kielcorins of the C (7 and 8) and E
(11) series were significantly more active (P < 0.005) than their D isomers (9 and 10).
In the cis/trans pair, it is interesting to note that (Æ)-trans-kielcorin C (7) was shown to
be a more potent inhibitor than its cis-isomer 8 (P < 0.05). Also, (Æ)-trans-isokielcorin

Helvetica Chimica Acta ± Vol. 85 (2002)
2871
B (3) was associated with a significantly (P < 0.05) stronger effect compared with its
constitutional isomer (Æ)-trans-kielcorin B (2).
It is also interesting to note that the absence of the MeO substituent in the
xanthonic framework was associated with a significant (P < 0.05) increase of growth
inhibitory activity as revealed by kielcorins 7 and 11, relative to those of 1 and 2,
respectively.
2
.3.2. Effect of Kielcorins on Human Lymphocytes Proliferation. Inhibition of
human lymphocyte proliferation induced by phytohemagglutinin (PHA) was detected
with all of the oxygenated xanthones 5 and 12 ± 14, and with (Æ)-trans-isokielcorin B
(
3) and the kielcorins in C series, 7 and 8. The antiproliferative activity detected was
dose-dependent and could not be attributed to a toxic effect on lymphocytes as
demonstrated by the lymphocytoxicity assay (viability of exposed lymphocytes >70%).
(
Æ)-trans-Kielcorin (1) and compounds in D and E series, 9 ± 11, showed no inhibitory
effect even when tested at 100 mm.
Comparing the antiproliferative activity of the dihydroxy-9H-xanthen-9-ones, it
was concluded that the position of the ortho-dihydroxy groups influence the suppresser
effect of these compounds (Table 1). Positions 3,4- were associated with a significantly
(
P < 0.01) stronger antiproliferative effect when compared with the 1,2- and 2,3-
positions. Kielcorins 3, 7, and 8 exhibited a significantly (P < 0.05) less-potent
suppresser effect than that of their xanthonic building blocks 5 and 12.
3
. Conclusions. ± In summary, the synthesis of five new structures based on models
of natural kielcorins is reported. For this kind of compound, the most efficient method
in the oxidative coupling reaction involves the use of K [Fe(CN) ] as oxidizing-
3
6
agent. Long-range C,H connectivities and NOE experiments permit unambiguous
establishment of the structures of the different isomers. The compounds of the C and E
series are the most active for the inhibition of the growth activity. For the inhibition of
human lymphocyte proliferation, generally, the kielcorins are less potent than their
xanthonic building blocks. The differences in the isomerism and in the nature of the
substituents in the xanthonolignoid framework will allow us to evaluate the structure
features, and further modifications are underway to improve the activity of these
compounds.
The authors thank FCT (I & D No. 226/94 and No. 62/94), POCTI (QCAIII), and FEDER for financial
support, and to PRAXIS XXI and FCT for the Ph.D. grants to M. M. P. (SFRH/BD/1456/2000), F. A. M. C.
(
PRAXIS XXI/BD/21801/99), and to E. P. S. (PRAXIS XXI/BD/15663/98). We are indebted to the National
Cancer Institute, Bethesda, MD, USA, for the generous provision of the cancer cell lines and to Dr. G. Eaton,
Department of Chemistry, University of Leicester, UK, for HR-MS measurements.
Experimental Part
1
. General. Column chromatography (CC): Merck silica gel 60 (0.50 ± 0.20 mm). Flash chromatography
FC): Merck silica gel 60 (0.040 ± 0.063 mm). Anal. and prep. TLC: Merck silica gel 60 (GF254). The reactions
were monitored by TLC (CHCl /MeOH) and interrupted, when one of the reagents was no longer detectable by
TLC. Ag CO , AcONa, and K [Fe(CN) ] were supplied by Merck (D-Darmstadt). Coniferyl alcohol and Ag
were obtained from Sigma (D-Steunheim). Lithium diisopropylamide (LDA) soln. was purchased from Fluka
D-Neu-Ulm). Structure elucidations of phenylcoumaran (21) and dehydrodiconiferyl alcohol (22; Fig. 2) were
(
3
2
3
3
6
2
O
(
accomplished by comparison with authentic samples. Reversed-phase (RP) HPLC: Jasco 880-PU, equipped
with a Jasco 875-UV detector, and a CSW 1.7 integrator; Nucleosil-C18 column (250 Â 4.6 mm, 5 mm; Macherey-

2
872
Helvetica Chimica Acta ± Vol. 85 (2002)
Nagel, D-D¸ren) equipped with a guard pre-column (Macherey-Nagel); linear gradient for compounds 7 and 8
with MeOH/1% AcOH in H O 6 :4, for compounds 9 and 10 with MeCN/1% AcOH in H O 5 :5, and for
compound 11 with MeOH/1% AcOH in H O 7:3; at 258; flow rate of 1 ml/min; sample injection volume 20 ml;
detection at 239 nm for kielcorins 7 ± 10 and 241 nm for kielcorin 11; t in min. M.p. Kofler microscope;
2
2
2
R
uncorrected. The optical rotation was zero for compounds 7 ± 11 and was obtained on a Polartronic Universal
À1
polarimeter. IR Spectra: Perkin-Elmer 257 spectrophotometer; KBr microplates; in cm . UV Spectra: Varian
1
13
CARY 1E spectrophotometer; lmax in nm. H- and C-NMR Spectra: Bruker AC-200, DRX-300, or DRX-500
spectrometer; at r.t., CDCl
3
or (D
6
)DMSO soln.; d in ppm relative to Me
4
Si as an internal reference; coupling
1
13
constants J in Hz. H-NMR assignments were made by means of 2D COSY experiments, while C-NMR
assignments were made by 1D selective INEPTand 2D HETCOR, HSQC, and HMBC experiments (long-range
C,H coupling constants were optimized to 7 and 1 Hz). NOE Experiments were conducted by means of NOE
difference technique, with an irradiation time of 2 s and a relaxation delay of 4 s. EI-MS Spectra: Hitachi
Perkin-Elmer RMU-6M spectrometer, in m/z. HR-MS Spectra: VG AutoSpec-Q spectromer; recorded as EI
(electronic impact) mode.
Fetal bovine serum and RPMI-1640 medium were obtained from Gibco BRL. Cyclosporin A, DMF, 3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), DMSO, doxorubicin hydrochloride, gentami-
cin, l-glutamine, Histopaque-1077, phytohemagglutinin (PHA), sodium lauryl sulfate (SDS), and sulforhod-
amine B (SRB) were purchased from Sigma. National Cancer Institute (Bethesda, MD, USA) kindly provided
the cancer cell lines. Stock solns. of compounds were prepared in DMSO and stored at À208 providing uniform
samples for retesting. These frozen concentrates were then diluted to the desired final concentrations with the
appropriate diluents immediately prior the biological assays. Final concentrations of DMSO showed no
interference with the biological activities tested. Unpaired Student×s t-tests were used. Differences with P values
below 0.01 or 0.05 were considered statistically significant. Whenever GI50 > 100 mm, values of 100 mm were
considered to evaluate the statistical significance of differences.
2
. Synthesis of Kielcorins 7 ± 11. General Procedure for the Oxidative Coupling Reaction of Dihydroxy-9H-
xanthen-9-ones 12 ± 14 and Coniferyl Alcohol (6). Method A: with a) Ag O and b) Ag CO . The appropriate
dihydroxyxanthones 12 ± 14 (615 mg, 2.7 mmol) and 6 (500 mg, 2.7 mmol) were taken up in toluene/Me CO 1:1
50 ml) and stirred in the presence of 3.1 mmol of the oxidizing agent (a) 815 mg of Ag O or b) 860 mg of
Ag CO ), at r.t., in the dark for 2 ± 3 days. The suspension was filtered, the filtrate evaporated, and the crude
product was analyzed by HPLC. The crude reaction products were purified by CC with a mixture of hexane/
CHCl /MeOH (several proportions). The isolation of the components was then carried out by prep. TLC
CHCl /MeOH 95 :5). HPLC: t (% yield Ag O, Ag CO ): 17.2 (20, 17) 7, 20.2 (3.5, 3.0) 8, 4.9 (22, 20) 9, 8.4 (0.7,
.4) 10, 12.8 (3.0, 2.0) 11.
Method B: with K [Fe(CN)
.4 mmol) and 6 (250 mg, 1.4 mmol) in Me
then K [Fe(CN) ] soln. (1.0 g, 30 mmol in 25 ml of H
slightly acidified with 10% HCl and diluted with H O. The product was extracted with CH
washed with H O, dried, and the solvent was evaporated to dryness. The crude product was analyzed by HPLC.
The isolation of the components, was then carried out by prep. TLC (CHCl /MeOH 95 :5). HPLC: t (% yield):
2
2
3
2
(
2
2
3
3
(
3
R
2
2
3
0
3
6
]. To a stirred soln. of the appropriate dihydroxyxanthones 12 ± 14 (315 mg,
CO/H O 1:1 (60 ml), AcONa (0.9 g, 11 mmol, in 25 ml of H O) and
O) were added at r.t. After the reaction, the mixture was
Cl ; the extract was
1
2
2
2
3
6
2
2
2
2
2
3
R
1
7.2 (29) 7, 20.2 (11) 8, 4.9 (27) 9, 8.4 (0.8) 10, 12.8 (4.0) 11.
Conversion of (Æ)-cis-Kielcorin C (8) to (Æ)-trans-Kielcorin C (7). To the mixture of 7 and 8 obtained with
[Fe(CN) ] as described above (409.6 mg), H O (200 ml) was added, and the suspension was alkalinized with
CO . The suspension was stirred at r.t. for 1 h, until conversion of cis-isomer 8 to the trans-isomer 7 was
K
K
3
2
6
2
3
2 2 4
complete. The mixture was extracted with AcOEt, washed with H O, dried (Na SO ), filtered, and concentrated
in vacuum to reveal a white solid (407.2 mg, 99%).
(
Æ)-(2R*,3R*)- and (Æ)-(2S*,3S*)-2,3-Dihydro-3-(4-hydroxy-3-methoxyphenyl)-2-(hydroxymethyl)-7H-
,4-dioxino[2,3-c]xanthen-7-one ((Æ)-trans-kielcorin C; 7). M.p. 243 ± 2478 (Me CO). UV (EtOH): 309, 283,
39, 204. IR (KBr): 3398, 1642, 1605, 1562, 1449, 1338, 1264, 1025, 861. H-NMR (300.13 MHz, (D )DMSO):
.25 (s, HOÀC(4')); 8.20 (dd, J ˆ 7.7, 1.6, HÀC(8)); 7.87 (ddd, J ˆ 8.3,7.4, 1.6, HÀC(10)); 7.70 (d, J ˆ 8.8,
1
2
9
2
1
6
HÀC(6)); 7.69 (d, J ˆ 8.3, HÀC(11)); 7.49 (ddd, J ˆ 7.7, 7.4, 0.7, HÀC(9)); 7.07 (d, J ˆ 8.8, HÀC(5)); 7.07 (d, J ˆ
.7, HÀC(2')); 6.92 (dd, J ˆ 8.1, 1.7, HÀC(6')); 6.83 (d, J ˆ 8.1, HÀC(5')); 5.15 (d, J ˆ 7.9, HÀC(3)); 5.13 (t, J ˆ
.1, CH OH); 3.40 ± 3.50 (m, CH OH).
OH); 4.36 ± 4.40 (m, HÀC(2)); 3.79 (s, MeO); 3.71 ± 3.76 (m, CH
C-NMR (75.47 MHz, (D )DMSO): 175.1 (C(7)); 155.4 (C(11a)); 148.8 (C(4a)); 147.7 (C(3')); 147.2 (C(4'));
45.9 (C(12a)); 135.2 (C(10)); 131.8 (C(12b)); 126.7 (C(1')); 126.0 (C(8)); 124.5 (C(9)); 121.0 (C(7a)); 120.7
1
6
2
2
2
1
3
6
1
(
(
C(6')); 118.1 (C(11)); 117.5 (C(6)); 115.7 (C(6a)); 115.4 (C(5')); 113.9 (C(5)); 111.8 (C(2')); 78.0 (C(2)); 76.5
‡
C(3)); 59.9 (CH
2
OH); 55.7 (MeO). EI-MS: 406 (6, M ), 356 (26), 341 (22), 327 (17), 306 (18), 292 (12), 281

Helvetica Chimica Acta ± Vol. 85 (2002)
2873
(
5
100), 261 (38), 253 (5), 228 (13), 215 (8), 180 (10), 165 (8), 141 (10), 115 (18), 105 (9), 91 (17), 77 (18), 63 (10),
5 (13). Anal. calc. for C23 : 407.1131; found: 407.1130.
Æ)-(2S*,3R*)- and (Æ)-(2R*,3S*)-2,3-Dihydro-3-(4-hydroxy-3-methoxyphenyl)-2-(hydroxymethyl)-7H-
,4-dioxino[2,3-c]xanthen-7-one ((Æ)-cis-Kielcorin C; 8). M.p. 200 ± 2028 (Me CO). UV (EtOH): 308, 283,
39, 204. IR (KBr): 3420, 1644, 1607, 1511, 1454, 1379, 1267, 1100, 860. H-NMR (300.13 MHz, (D )DMSO): 9.09
s, HOÀC(4')); 8.21 (dd, J ˆ 7.6, 1.6, HÀC(8)); 7.83 (ddd, J ˆ 8.0, 7.5, 1.6, HÀC(10)); 7.75 (d, J ˆ 8.7, HÀC(6));
.67 (d, J ˆ 8.0, HÀC(11)); 7.46 (dd, J ˆ 7.6, 7.5, HÀC(9)); 7.05 (d, J ˆ 8.8, HÀC(5)); 6.99 (br. s, HÀC(2'));
.95 ± 6.74 (m, HÀC(5'), and HÀC(6')); 5.46 (d, J ˆ 2.8, HÀC(3)); 4.98 (t, J ˆ 5.4, CH OH); 4.63 ± 4.68
)DMSO): 175.5 (C(7)); 155.9
19 7
H O
(
1
2
(
7
2
1
6
6
2
1
3
(
(
(
m, HÀC(2)); 3.83 (s, MeO); 3.42 ± 3.61 (m, CH
2 6
OH). C-NMR (75.47 MHz, (D
C(11a)); 148.7 (C(4a)); 147.8 (C(3')); 147.1 (C(4')); 146.7 (C(12a)); 135.2 (C(10)); 131.1 (C(12b)); 126.7
C(1')); 126.4 (C(8)); 124.6 (C(9)); 121.5 (C(7a)); 119.5 (C(6')); 118.4 (C(11)); 118.2 (C(6)); 116.4 (C(6a));
1
15.9 (C(5')); 114.1 (C(5)); 111.4 (C(2')); 77.8 (C(2)); 76.4 (C(3)); 58.6 (CH
2
OH); 56.0 (MeO). EI-MS: 406 (19,
‡
M ), 281 (1.8), 246 (24), 184 (14), 153 (100), 149 (8), 137 (17), 111 (1.3), 93 (60), 83 (22), 72 (8), 65 (38), 59
(
14), 57 (41). Anal. calc. for C23
Æ)-(2R*,3R*)- and (Æ)-(2S*,3S*)-2,3-Dihydro-3-(4-hydroxy-3-methoxyphenyl)-2-(hydroxyphenyl)-12H-
,4-dioxino[2,3-a]xanthen-12-one ((Æ)-trans-Kielcorin D; 9). M.p. 258 ± 2608 (Me CO). UV (EtOH): 372, 278,
59, 239, 237, 212. IR (KBr): 3498, 1635, 1602, 1309, 1270, 1228, 1047, 762. H-NMR (300.13 MHz, (D )DMSO):
19 7
H O : 407.1131; found: 407.1131.
(
1
2
9
2
1
6
.20 (s, HOÀC(4')); 8.04 (dd, J ˆ 7.9, 1.6, HÀC(11)); 7.80 (ddd, J ˆ 8.1, 7.8, 1.6, HÀC(9)); 7.56 (d, J ˆ 8.1,
HÀC(8)); 7.45 (d, J ˆ 9.1, HÀC(5)); 7.40 (dd, J ˆ 7.9, 7.8, HÀC(10)); 7.14 (d, J ˆ 9.1, HÀC(6)); 7.10 (d, J ˆ 1.7,
HÀC(2')); 6.92 (dd, J ˆ 8.1, 1.7, HÀC(6')); 6.83 (d, J ˆ 8.1, HÀC(5')); 5.05 (d, J ˆ 7.6, HÀC(3)); 5.05 (t, J ˆ 7.6,
CH
2
OH); 4.20 ± 4.23 (m, HÀC(2)); 3.78 (s, MeO); 3.57 ± 3.62 (m, CH
2 2 2
OH); 3.41 (under H O signal, CH OH).
6
)DMSO): 174.9 (C(12)); 154.6 (C(7a)); 150.9 (C(6a)); 147.6 (C(3')); 147.0 (C(4'));
1
3
C-NMR (75.47 MHz, (D
1
43.1 (C(12b)); 138.9 (C(4a)); 134.8 (C(9)); 127.1 (C(1')); 125.9 (C(11)); 124.0 (C(10)); 123.7 (C(5)); 121.7
(
(
(
C(11a)); 120.5 (C(6')); 117.6 (C(8)); 115.3 (C(5')); 111.92, 111.86 (C(12a), C(2')); 109.5 (C(6)); 77.2 (C(2)); 76.1
‡
C(3)); 60.0 (CH
2
OH); 55.7 (MeO). EI-MS: 406 (6, M ), 369 (2), 327 (7), 285 (6), 228 (14), 184 (22), 153
100), 137 (10), 125 (13), 110 (9), 93 (61), 81 (9), 65 (35). Anal. calc. for C23
H
19
O
7
: 407.1131; found: 407.1130.
Æ)-(2R*,3R*)- and (Æ)-(2S*,35S)-2,3-Dihydro-2-(4-hydroxy-3-methoxyphenyl)-3-(hydroxymethyl)-12H-
CO). UV (EtOH): 374,
80, 260, 240, 209. IR (KBr): 3436, 1626, 1520, 1459, 1383, 1319, 1269, 1110, 1060, 1028, 809. 1H-NMR
(
1
2
,4-dioxino[2,3-a]xanthen-12-one ((Æ)-trans-Isokielcorin D; 10). M.p. 136 ± 1398 (Me
2
(
500.13 MHz, (D
6
)DMSO): 9.20 (s, HOÀC(4')); 8.12 (dd, J ˆ 7.8, 1.4, HÀC(11)); 7.81 (ddd, J ˆ 8.1, 7.4, 1.4,
HÀC(9)); 7.57 (d, J ˆ 8.1, HÀC(8)); 7.43 (d, J ˆ 9.1, HÀC(5)); 7.43 (dd, J ˆ 7.8, 7.4, HÀC(10)); 7.12 (d, J ˆ 9.1,
HÀC(6)); 7.04 (d, J ˆ 1.4, HÀC(2')); 6.89 (d, J ˆ 8.1, 1.4, HÀC(6')); 6.80 (d, J ˆ 8.1, HÀC(5')); 5.00 (d, J ˆ 7.8,
HÀC(2)); 4.99 (t, J ˆ 7.6, CH
2
1
OH); 4.31 ± 4.34 (m, HÀC(3)); 3.77 (s, MeO); 3.71 ± 3.73 (m, CH
2
OH); 3.41
3
(
under H
2
O signal, CH
2 6
OH). C-NMR (75.47 MHz, (D )DMSO): 174.6 (C(12)); 154.5 (C(7a)); 150.0 (C(6a));
1
47.7 (C(3')); 147.3 (C(4')); 143.2 (C(12b)); 139.2 (C(4a)); 135.0 (C(9)); 127.3 (C(1')); 126.2 (C(11)); 123.8
(
(
(
C(5)); 123.7 (C(10)); 122.3 (C(11a)); 120.8 (C(6')); 117.7 (C(8)); 115.1 (C(5')); 111.7 (C(12a), C(2')); 107.9
‡
C(6)); 78.0 (C(3)); 75.2 (C(2)); 59.4 (CH
2
OH); 55.4 (MeO). EI-MS: 406 (50, M ), 239 (7), 228 (100), 199
25), 180 (73), 170 (11), 153 (18), 137 (64), 124 (44), 103 (5), 91 (16), 77 (14), 65 (10). Anal. calc. for C23
07.1131; found: 407.1131.
Æ)-(2R*,3S*)- and (Æ)-(2S*,3R)-2,3-Dihydro-3-(4-hydroxy-3-methoxyphenyl)-2-(hydroxymethyl)-11H-
,4-dioxino[2,3-b]xanthen-11-one ((Æ)-trans-Kielcorin E; 11). M.p. 247 ± 2498 (Me CO). UV (EtOH): 353,
07, 277, 241. IR (KBr): 3498, 1635, 1602, 1309, 1270, 1228, 1047, 762. H-NMR (300.13 MHz, (D )DMSO): 9.25
s, HOÀC(4')); 8.17 (dd, J ˆ 7.7, 1.2, HÀC(10)); 7.84 (ddd, J ˆ 8.0, 7.6, 1.2, HÀC(8)); 7.62 (d, J ˆ 8.0, HÀC(7));
.61 (s, HÀC(12)); 7.46 (ddd, J ˆ 7.7, 7.5, HÀC(9)); 7.23 (s, HÀC(5)); 7.06 (d, J ˆ 1.3, HÀC(2')); 6.90 (dd, J ˆ 8.1,
.3, HÀC(6')); 6.82 (d, J ˆ 8.1, HÀC(5')); 5.13 (d, J ˆ 8.1, HÀC(3)); 5.06 (br. s, CH OH); 4.33 ± 4.35
m, HÀC(2)); 3.78 (s, MeO); 3.58 (br. d, J ˆ 11.4, CH OH): 3.38 (under H O signal, CH
75.47 MHz, (D )DMSO): 174.8 (C(11)); 155.7 (C(6a)); 151.1 (C(4a)); 150.6 (C(5a)); 147.7 (C(3')); 147.3
C(4')); 141.3 (C(12a)); 135.0 (C(8)); 126.5 (C(1')); 125.8 (C(10)); 124.1 (C(9)); 120.8 (C(6')); 120.6 (C(10a));
19 7
H O :
4
(
1
3
(
7
2
1
6
1
2
1
3
(
(
(
2
2
2
OH). C-NMR
6
1
18.0 (C(7)); 115.4 (C(5'), C(11a)); 111.9 (C(2')); 111.2 (C(12)); 104.8 (C(5)); 77.8 (C(2)); 77.0 (C(3)); 60.0
‡
(CH
2
OH); 55.7 (MeO). EI-MS: 406 (0.02, M ), 369 (0.8), 327 (6), 310 (0.8), 285 (6), 267 (0.6), 260 (1.4), 246
(
1), 215 (2), 208 (0.6), 184 (21), 153 (100), 137 (3), 125 (12), 110 (9), 93 (62), 81 (8), 65 (35). Anal. calc. for
: 407.1131; found: 407.1131.
. Synthesis of Xanthones 12 ± 14. The following compounds were obtained according to the procedures
cited below.
,4-Dihydroxy-9H-xanthen-9-one (12) [14]. M.p. >3308 CHCl
23 19 7
C H O
3
3
3
([16]: 240 ± 2418 (EtOH/H
2
O); [17]: 238 ±
1
2
408 (MeOH)). UV, IR, and EI-MS: in accord with those in [17]. The analysis of the H, 13C, COSY, HETCOR,

2
874
Helvetica Chimica Acta ± Vol. 85 (2002)
1
and HMBC spectra of 12 permits us to correct the literature [17] assignment of the H resonances of HÀC(6)
and HÀC(5) to 7.81 (ddd, J ˆ 8.6, 6.9, 1.7, HÀC(6)) and 7.62 (dd, J ˆ 8.6, and 0.9, HÀC(5)).
2 2
1,2-Dihydroxy-9H-xanthen-9-one (13) [14][18]. M.p. 163 ± 1658 (CH Cl /hexane); [19]: 166 ± 1678 (EtOH/
H
2
O)). UV (EtOH): 261, 243, 232, 203 (33300), 28800, 14200, 11000); UV (EtOH ‡ NaOH): 311, 236, 222
(
1
76000, 23900, 17000). UV (EtOH ‡ AlCl
3
): 305, 283, 266, 204 (33000, 22900, 8700). IR (KBr): 3467, 1606, 1475,
)DMSO): [20] 12.45 (s, HOÀC(1)); 9.42 (s, HOÀC(2)); 8.16
1
365, 1288, 1045, 748. H-NMR (300.13 MHz, (D
6
(
dd, J ˆ 7.9, 1.6, HÀC(8)); 7.81 (ddd, J ˆ 8.4, 7.0, 1.6, HÀC(6)); 7.60 (dd, J ˆ 8.4, 0.8, HÀC(5)); 7.46 (ddd, J ˆ 7.9,
13
7
1
1
1
.0, 0.8, HÀC(7)); 7.32 (d, J ˆ 9.0, HÀC(4)); 6.96 (d, J ˆ 9.0, HÀC(3)). C-NMR (75.47 MHz, (D
6
)DMSO):
82.3 (C(9)); 155.9 (C(4b)); 148.3 (C(4a)); 147.6 (C(1)); 140.2 (C(2)); 136.4 (C(6)); 125.4 (C(8)); 124.5 (C(3));
‡
24.2 (C(7)); 119.3 (C(8a)); 118.0 (C(5)); 108.8 (C(9a)); 106.2 (C(4)). EI-MS: 230 (1, [M ‡ 2] ), 229 (1, [M ‡
‡
]), 228 (70, M ), 199 (13), 126 (5), 115 (4), 78 (95), 63 (100).
2
,3-Dihydroxy-9H-xanthen-9-one (14) [14][17][21]. M.p. >3308 (Me
17]: 293 ± 2958 (CHCl /MeOH)). UV, IR and EI-MS: in accord with those in [17]. The unequivocal assignments
of C resonances, mainly on the basis of 1D selective INEPT and 2D HETCOR experiments, allowed us to
2
CO/hexane) ([14]: >3508 (AcOEt);
[
3
1
3
1
13
1
correct some H and C resonances of 14 [17], namely the assignment of the H resonances of HÀC(6) to 7.78
1
3
(
ddd, J ˆ 8.4, 6.8, 1.7) and HÀC(5) to 7.58 (d, J ˆ 8.4) and the C resonances of C(2) (144.1), C(3) (151.3), and
C(4a) (154.2).
Methyl 2-Bromobenzoate (15) [22], Methyl 2-(3,4-Dimethoxyphenoxy)benzoate (17) [11 ± 13], 2-(3,4-
Dimethoxyphenoxy)benzoic Acid (18) [11][12], and 1,2-Dimethoxy-9H-xanthen-9-one (19) [12][24]: Data
identical to those reported in [12][25].
2
,3-Dimethoxy-9H-xanthen-9-one (20) [13][25]. M.p. 170 ± 1718 (CH
2
Cl
2
/pentane) ([26]: 164 ± 1658
/petroleum
(
Me
2
CO); [17]: 157 ± 1588 (CHCl ); [27]: 155 ± 1598 (CH Cl /heptane); [28]: 165 ± 1678 (CHCl
3
2
2
3
1
3
ether)). IR and EI-MS: in accord with those in [17]. The unequivocal assignments of C resonances, mainly on
the basis of 1D selective INEPT and 2D HETCOR experiments, allowed us to correct some H and
resonances of 20 [29]. H-NMR (200.13 MHz, (D
1
13
C
1
6
)DMSO): 8.16 (dd, J ˆ 7.9, 1.5, HÀC(8)); 7.82 (ddd, J ˆ 8.4,
7
(
1
1
.4, 1.5, HÀC(6)); 7.60 (d, J ˆ 8.4, HÀC(5)); 7.49 (s, HÀC(1)); 7.44 (ddd, J ˆ 7.9, 7.4, 0.9, HÀC(7)); 7.20
13
s, HÀC(4)); 3.94 (s, MeOÀC(3)); 3.87 (s, MeOÀC(2)). C-NMR (50.03 MHz, (D
6
)DMSO): 174.8 (C(9));
55.7 (C(4b)); 155.6 (C(4a)); 152.0 (C(3)); 146.7 (C(2)); 134.8 (C(6)); 125.9 (C(8)); 124.3 (C(7)); 120.9 (C(8a));
18.0 (C(5)); 114.0 (C(9a)); 104.7 (C(1)); 100.5 (C(4)); 56.6 (MeOÀC(3)); 55.9 (MeOÀC(2)).
. Cell-Growth Assay. The effects of compounds on the growth of human cancer cell lines were evaluated
4
according to the procedure adopted in the NCI×s in vitro anticancer drug screening that uses the SRB assay to
assess growth inhibition [30]. This colorimetric assay estimates cell number indirectly by staining cellular
protein with the protein-binding dye SRB. Three human cancer cell lines were used, MCF-7 (breast cancer),
TK-10 (renal cancer), UACC-62 (melanoma). Cells were routinely maintained as adherent cell cultures in
RPMI-1640 medium containing 5% heat-inactivated fetal bovine serum, 2 mm glutamine, and 50 mg/ml
gentamicin at 378 in a humidified air incubator containing 5% CO
2
. For the SRB assay, each cell line was plated
5
5
at an appropriate density (1.5 Â 10 cells/ml to MCF-7 and TK-10, 1.0 Â 10 cells/ml to UACC-62) in 96-well
plates and allowed to attach overnight. Cells were then exposed for 48 h to serial concentrations of compounds
and to the positive control doxorubicin. Following this incubation period, the adherent cells were fixed in situ,
and washed and dyed with SRB. The bound stain was solubilized, and the absorbance was measured at 492 nm in
a microplate reader. Growth inhibition of 50% (GI50) as well as toxicity of compounds was determined as
described in [31]. Toxicity of xanthone derivatives was inferred from the SRB assay by comparing, after 48 h, the
absorbance of the wells with xanthone derivative-treated cells with the absorbance of the wells containing
untreated cells that were fixed at time zero (time at which compounds were added). Lower absorbances after
4
8 h of treatment indicate occurrence of cell death instead of growth arrest.
. Lymphocyte Assays. The effects of compounds on the mitogenic response of human lymphocytes to
5
PHA(10 mg/ml) were evaluated with a modified colorimetric MTT assay [32]. This assay was previously
described by us [33]. Mononuclear cells were isolated from heparinized peripheral blood of healthy volunteers
by Histopaque-1077 density centrifugation. The toxicity of compounds against human lymphocytes was
evaluated by an assay based on the ability of viable cells to reduce the colourless tetrazolium salt MTT to a
6
colored formazan product. Briefly, in flat-bottom 96-well plates, 2 ± 3 Â 10 mononuclear cells/ml in RPMI-1640
medium containing 10% heat-inactivated foetal bovine serum, 2 mm glutamine, and 50 mg/ml of gentamicin
were exposed for 24 h to the various concentrations of each compound. Following this incubation period,
the MTT soln. (1 mg/ml) was added. After incubation for 4 h, the MTT formazan products were solubilized
with SDS/DMF soln. (20% SDS in a 50% soln. of DMF, pH 4.7) overnight at 378. Absorbance (OD 550 nm) of
the colored soln. was measured with a plate reader. Lymphocytoxicity, determined in terms of the percentage of

Helvetica Chimica Acta ± Vol. 85 (2002)
2875
viable cells, was present when the viability of the exposed cells was less than 70% of the nonexposed control
cells.
REFERENCES
[
[
[
1] O. R. Gottlieb, A. A. L. Mesquita, E. M. Silva, M. T. Melo, Phytochemistry 1969, 8, 665.
2] M. M. M. Pinto, A. A. L. Mesquita, O. R. Gottlieb, Phytochemistry 1987, 26, 2045.
3] E. G. R. Fernandes, M. M. M. Pinto, A. M. S. Silva, J. A. S. Cavaleiro, O. R. Gottlieb, Heterocycles 1999, 51,
8
21.
4] E. R. Fernandes, F. D. Carvalho, F. G. Remi aƒ o, M. L. Bastos, M. M. Pinto, O. R. Gottlieb, Pharm. Res. 1995,
2, 1756.
5] a) S. Antus, E. Baitz-G a¬ es, R. Bauer, A. Gottsegen, O. Seligmann, H. Wagner, Liebigs Ann. Chem. 1989,
147; b) S. Maeda, H. Masuda, T. Tokoroyama, Chem. Pharm. Bull. 1995, 43, 84; c) M. Massacret, C. Goux,
[
[
1
1
P. Lhoste, D. Sinou, Tetrahedron Lett. 1994, 5, 6093.
[
[
6] R. A. Vishwakarma, R. S. Kapil, S. P. Popli, Indian J. Chem., Sect. B 1986, 25, 1021.
7] a) X. She, X. Jing, X. Pan, A. S. C. Chan, T. Yang, Tetrahedron Lett. 1999, 40, 4570; b) M. F. Hibert, M. W.
Gittos, D. N. Middlemiss, A. K. Mir, J. R. Fozard, J. Med. Chem. 1988, 31, 1087.
[8] a) M. Abou-Shoer, A. Habib, C. Chang, J. M. Cassady, Phytochemistry 1989, 28, 2483; b) H. Tosa, M.
Iinuma, T. Tanaka, H. Nozaki, S. Ikeda, K. Tsutsui, K. Tsutsui, M. Yamada, S. Fujimori, Chem. Pharm. Bull.
1997, 45, 418.
[9] L. Merlini, A. Zanarotti, A. Pelter, M. P. Rochefort, R. H‰nsel, J. Chem. Soc., Perkin Trans. 1 1980, 3, 775.
[
[
10] L. Merlini, A. Zanarotti, Tetrahedron Lett. 1975, 3621.
11] W. T. Jackson, R. J. Boyd, L. L. Froelich, D. M. Gapinski, B. E. Mallett, J. S. Sawyer, J. Med. Chem. 1993,
36, 1726.
[
[
[
[
12] L. Gales, M. E. Sousa, M. M. M. Pinto, A. Kijjoa, A. M. Damas, Acta Crystallogr., Sect. C 2001, 57, 1319.
13] R. A. Finnegan, P. L. Bachman, J. Pharm. Sci. 1965, 54, 633.
14] O. R. Gottlieb, A. A. L. Mesquita, G. G. Oliveira, M. T. Melo, Phytochemistry 1970, 9, 2537.
15] a) A. Bax, J. Magn. Reson. 1984, 57, 314; b) G. Venkatraman, L. J. Harrison, K. Y. Sim, Tetrahedron Lett.
1996, 37, 2643; c) L.-J. Lin, G. A. Cordell, J. Chem. Res. (M) 1988, 3052; d) G. A. Cordell, A. D. Kinghorn,
Tetrahedron 1988, 47, 3521.
[16] P. K. Grover, G. D. Shah, R. C. Shah, J. Chem. Soc. 1955, 3982.
[17] C. N. Lin, S. S. Liou, F. N. Ko, C. M. Teng, J. Pharm. Sci. 1993, 82, 11.
[18] B. Vermes, O. Seligmann, H. Wagner, Helv. Chim. Acta 1985, 68, 2359.
[19] J. S. H. Davies, F. Lamb, H. Suschitzky, J. Chem. Soc. 1958, 1790.
[
20] a) D. Barraglough, H. D. Locksley, F. Scheinmann, M. T. Magalh aƒ es, J. Chem. Soc. B 1970, 603; b) P.
Arends, P. Helboe, Acta Chem. Scand. 1972, 26, 4181.
[
[
21] O. R. Gottlieb, A. A. L. Mesquita, G. G. Oliveira, M. T. Melo, Anais da Academia Brasileira de Ci e√ ncias
1970, 42, 117.
22] a) A. Ault, −Techniques and Experiments for Organic Chemistry×, 5th edn., Allyn and Bacon, Inc., Newton,
Massachusetts, 1987, p. 392; b) B. S. Furniss, A. J. Hannaford, P. W. G. Smith, A. R. Tatchell, −Vogel×s
Textbook of Practical Organic Chemistry×, Longman Scientific and Technical, Harlow, England, 1989,
p. 1077.
[
[
23] A. Moroz, M. S. Shvartsberg, Russ. Chem. Rev. 1974, 43, 679.
24] a) O. B. Familoni, I. Ionica, J. F. Bower, V. Snieckus, Synlett 1997, 1081; b) B. Bennetau, J. Mortier, J.
Moyround, J. Guesnet, J. Chem. Soc., Perkin Trans. 1 1995, 1265.
[
25] C. Morel, D. S e¬ raphin, J.-M. Oger, M. Litaudon, T. S e¬ venet, P. Richomme, J. Bruneton, Phytochemistry
2000, 53, 1471.
[
[
26] C. H. Hassal, J. R. Lewis, J. Chem. Soc. 1961, 2, 2312.
27] F. D. Monache, M. M. Mac-Quhae, G. D. Monache, G. B. M. Betollo, R. A. Lima, Phytochemistry 1983, 22,
227.
[28] A. A. L. Gunatilaka, A. M. Y. J. Silva, S. Sotheeswaran, Phytochemistry 1982, 21, 1751.
[29] A. J. Quillinan, F. Scheinmann, J. Chem. Soc., Perkin Trans. 1 1973, 2, 1329.
[30] P. Skehan, R. Storeng, D. Scudiero, A. Monks, J. McMahon, D. Vistica, J. T. Warren, H. Bokesch, S. Kenney,
M. R. Boyd, J. Natl. Cancer Inst. 1990, 82, 1107.

2
876
Helvetica Chimica Acta ± Vol. 85 (2002)
[
31] A. Monks, D. Scudiero, P. Skehan, R. Shoemaker, K. Paull, D. Vistica, C. Hose, J. Langley, P. Cronise, A.
Vaigro-Wolff, M. Gray-Goodrich, H. Campbell, J. Mayo, M. Boyd, J. Natl. Cancer Inst. 1991, 83, 757.
32] T. Mosmann, J. Immunol. Methods 1983, 65, 55.
[
[
33] M. J. Gonzalez, M. S. J. Nascimento, H. M. Cidade, M. M. M. Pinto, A. Kijjoa, C. Anantachoke, A. M. S.
Silva, W. Herz, Planta Med. 1999, 65, 368.
Received June 22, 2002