Synthesis and cytotoxic activities of analogues of thuriferic acid

Article abstract of DOI:10.1016/S0968-0896(01)00280-2

Several analogues of thuriferic acid and derivatives, with the 3,4-methylenedioxyphenyl ring replaced by naphthalene, furan, thiophene and carbazole ring systems, have been prepared. The synthetic strategy is based on a conjugate addition-alkylation metho

Full text of DOI:10.1016/S0968-0896(01)00280-2

Bioorganic & Medicinal Chemistry 10 (2002) 303–312
Synthesis and Cytotoxic Activities of Analogues of Thuriferic Acid
Blanca Madrigal,a Pilar Puebla,a Angel Ramos,a Rafael Pelaez,a Dolores Gravalos,b
Esther Caballeroa and Manuel Medardea,*
a
´
Laboratorio de Quımica Organica y Farmaceutica, Facultad de Farmacia. Campus Miguel de Unamuno, 37007-Salamanca, Spain
´
´
bInstituto BioMar, La Calera 3, 28760-Tres Cantos (Madrid), Spain
Received 1June 2001; accepted 31July 2001
Abstract—Several analogues of thuriferic acid and derivatives, with the 3,4-methylenedioxyphenyl ring replaced by naphthalene,
furan, thiophene and carbazole ring systems, have been prepared. The synthetic strategy is based on a conjugate addition–alkyla-
tion methodology followed by cationic cyclization in order to obtain the isopodophyllone analogues, which are transformed in the
thuriferic acids. Their cytotoxic activities against several tumour cells lines are also described. # 2001Elsevier Science Ltd. All
rights reserved.
Thuriferic acid is a natural cyclolignan, isolated from
Juniperus thurifera leaves,1 whose structure (Fig. 1) and
stereochemistry have been the subject of some debate
and finally confirmed by means of an array of spectro-
scopic techniques and semisynthesis from podophyllo-
toxin.2 The unambiguous establishment of the
configuration of thuriferic acid followed its treatment
with dry HCl and conversion of the terminal methylene
group into a chloromethyl substituent. The carboxylic
group and the a,b-unsaturated ketone, altogether with
the trans relationship between the C70 aryl group and
the carboxylic group at C80 are the most noticeable fea-
tures of a fairly rigid compound. Despite the fact that
the aforementioned structural characteristics are in
overall disagreement with those usually cited in the
structure–activity relationships for cytotoxic lignans:3
presence of a trans lactone ring and a cis C70–C80 rela-
tionship (see podophyllotoxin in Figure 1as a repre-
sentative example), thuriferic acid displays a general
cytotoxic effect with medium-low activity against the 51
cell lines of the NCI.2
Following our studies on the synthesis4 and cytotoxic
activities of non-natural lignans, such as heterolignans5
(lignan analogues in which one or several carbon atoms
of the basic lignanic skeleton have been replaced by het-
eroatoms), in this paper we have prepared and evaluated
a series of thuriferic acid analogues. The benzodioxole
system of the parent compound (A–B rings) has been
substituted by different heterocyclic moieties, such as
the furan, thiophene and carbazole ring systems, as well
as the less polar naphthalene moiety, which has proven
to be a good surrogate for the polyoxygenated phenyl
rings in systems similar to the one now been considered.6
Figure 1. Structure of model lignans and synthesised compounds.
*Corresponding author. Tel.:+34-923-294528; fax:+34-923-294515; e-mail: medarde@usal.es
0968-0896/02/$ - see front matter # 2001Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00280-2

304
B. Madrigal et al. / Bioorg. Med. Chem. 10 (2002) 303–312
The effects of these substitutions on the conformational
preferences of the resulting compounds, as well as on
their cytotoxic activities are discussed. The results show
a totally unexplored structure–activity relationship
(SAR) for the thuriferic acid derivatives, which might be
of interest in the development of new antitumor agents.
of thuriferic acid, which will be described in more detail.
The corresponding furan, thiophene, methylfuran, and
methylthiophene analogues were synthesised in a similar
way, as previously described.10 The cytotoxic activities
for all of them are here described for the first time.
The initial step implies a ‘one pot’ conjugate addition–
alkylation sequence leading to the desired 2,3-bisaryl
lactones.7 The lithium anions of 2-substituted-1,3-
dithianes (obtained by Lewis acid treatment of the cor-
responding aldehydes in the presence of 1,3-propane-
dithiol11), generated by means of n-BuLi in THF at
À78 ꢀC, are added to 5H-furan-2-one over a 3–4 h per-
iod, generating enolates which are alkylated with 3,4,5-
trimethoxybenzaldehyde. The conjugate additions pro-
ceed with good yields,12 but the alkylation steps require
careful control of the temperature and time forcing the
reaction to a maximum conversion (see Table 1). The
reaction leads to racemic non-alkylated 1 and alkylated
trans-lactones as mixtures of the two epimeric alcohols
at C70 (2 and 3). The 2:3 ratio depends on the nature of
the 1,3-dithiane 2-substituent, the electrophile, and the
actual reaction conditions in a non-straightforward way
(see Table 1). The relative stereochemistry at C70 was
Chemistry
Thuriferic acid itself can be prepared by basic treatment
of either podophyllone or picropodophyllone.2 The
synthetic methodology we used for the preparation of
the aryltetralones is based on the cationic cyclization of
the conjugate addition–alkylation products of 5H-
furan-2-one.7 Such a general methodology has led us to
the preparation of podophyllotoxin8 as well as its thio-
phene, furan and naphthalene analogues.4,9,10 In this
paper we have adapted the synthetic sequence for pre-
paring the desired thuriferic acid analogues from iso-
podo- and picropodophyllotoxone analogues (Fig. 2).
The synthetic methodology will be exemplified with the
route to carbazole (Fig. 3) and naphthalene analogues
Figure 2. General synthesis of thuriferic acid analogues and their derivatives. Key: (i) (a) BuLi 1.6 M, THF, À78 ꢀC, 45 min; (b) 5H-furan-2-one,
THF, -78 ꢀC, 3 h; (c) Ar–CHO, THF, TMEDA, À50 ꢀC, 12 h. (ii) TFA, CH2Cl2, 24 h. (iii) HgO, BF3.Et2O, THF–H2O (85:15), 0 ꢀC–> rt, 24 h; (iv)
EtOH/HOAc, reflux, 48 h. (v) 1% KOH/MeOH, rt, 1.5 h; then HCl 2 N. (vi) etheral CH2N2. (vii) p-TsOH, C6H6, reflux, 7 h. (viii) HCl dry stream,
CH2Cl2, 2 h.

B. Madrigal et al. / Bioorg. Med. Chem. 10 (2002) 303–312
305
established by comparison with podorhizol and epipo-
dorhizol,13 based on the 1H NMR chemical shifts of the
C70H (downfield shifted by the carboxyl group in the
threo isomer 2 of podorhizol series) and the C70H–C80H
J-coupling (larger in the epipodorhizol series 3). The
lack of stereoselectivity is not a problem, as both epi-
meric alcohols lead to the same compounds in sub-
sequent steps and can be used without separation on the
way to aryltetralin lactones.
the crude (both are mixtures of the podophyllotoxin 5f
and isopodophyllotoxin 4f relative stereochemistries,
see below). For the naphthalene series, the cyclisa-
tion step proceeds exclusively on the more activated
C1position 4e,5e (analogous to the carbazole ‘bent’ B-
series).
The relative stereochemistries of the cyclised products
were easily established based upon the C70H–C80H and
the C80H–C8H coupling constants. All the lactones
were C8–C80trans; some series having exclusively a trans
C70–C80 arrangement, some others being C70–C8’ cis/
trans mixtures. The product ratios were determined by
The mixtures of alcohols, upon treatment with tri-
fluoroacetic acid or tin tetrachloride (SnCl4) in CH2Cl2
are transformed into cyclised products. For the furan,
thiophene, methylfuran, and methylthiophene analo-
gues only cyclisation products of the rings at C3 are
observed.4,10 For the carbazole series (Fig. 3), starting
from the 2-(9-ethylcarbazol-3-yl)-1,3-dithiane, the cycli-
sation occurs mainly at position C4 of the carbazole
(further referred to as the ‘bent’ B-series), being the
regioisomeric products of cyclisation at the C2 position
(further referred to as the ‘linear’ L-series) detected and
identified as minor products. ‘Bent’ and ‘linear’ carba-
zole analogues, respectively, represent 41and 13% of
1
integration of the H NMR signals of the well resolved
C70 protons (the aromatic resonances for the trimethoxy-
phenyl ring are in some cases unusually broadened,
making them unsuitable for integration). The all-trans
relative stereochemistry 4 (isopodophyllotoxin series) is
preferentially obtained over the trans–cis stereo-
chemistry of podophyllotoxin series 5. The 4:5 ratio for
the carbazole and naphthalene analogues is 3:1to 2:1.
For the thiophene, methylthiophene, furan and methyl-
furan series no podophyllotoxin-like 5 products were
Table 1.
R
R0
T alk/TMEDA
(equiv)
Overall yield %
(erythro 3/threo 2)
R
R0
T alk/TMEDA
(equiv)
Overall yield %
(erythro 3/threo 2)
À2)0 ꢀC/157 (3/1
(eÀry2t0)ꢀC/171
À20 ꢀC / 2
À30 ꢀC / 2
À50 ꢀC /2
À50 ꢀC / 2
À50 ꢀC / 2
À50 ꢀC /2
À50 ꢀC/2
35 (5/1)
60 (5/1)
77 (5/1)
90 (2/1)
77 (2/1)
37 (2/1)
66 (2/1)
41(2/1)
À2)0 ꢀC/162 (3/1
À20 ꢀC /154 (eryt)
À2)0 ꢀC/167 (2/1
À20 ꢀC/130 (eryt)
À60 ꢀC/128 (eryt)
À20 ꢀC/2
42 (eryt)
34 (eryt)
À50 ꢀC /2
À40 ꢀC /2

306
B. Madrigal et al. / Bioorg. Med. Chem. 10 (2002) 303–312
detected.4,10 Such a stereochemical outcome is the main
problem in the synthesis of podophyllotoxin analogues
by this methodology.14 However, on the way to thur-
iferic acid analogues both stereochemistries are con-
verted to the same end product (see below) with the
desired relative configurations, making this strategy
much more efficient for this class of compounds.
(up to 50% depending upon the actual series and the
reaction time). In order to improve the yield of the
thuriferic acid stereochemistry, such epimerisation is
further promoted by means of an acidic treatment with
acetic acid, leading to total conversion. In the ‘bent’
carbazoles (Fig. 3), under the reaction conditions, the
cis-ketolactone is partially converted to the ‘bent-carba-
zolethuriferic’ acid ethyl ester B-10f (R=Et).
The masked carbonyl group at position C7 is depro-
.
tected by means of HgO/BF3 Et2O catalyzed hydrolysis
of the 1,3-dithiane ring producing trans-ketolactones
6.15 Under the reaction conditions the lactone ring is
partially epimerised at position C8 to cis-ketolactones 7
Treatment with 1% methanolic KOH is needed to
transform the trans-ketolactones 6 into the desired
thuriferic acid analogues 8 in good yields (in some cases
9 is also produced by conjugate addition of methanol
Table 2. Cytotoxic potencies against several representative cell lines (lg IC50M)
PÀ388
AÀ549
À6.5
HTÀ29
À6.5
MELÀ28
À6.2
a
À6.5
À6.2
À6.6
À6.6
À6.6
b
À6.2
À6.6
À5.9
À6.6
À6.2
À6.6
À5.9
À6.6
À6.2
À6.6
À5.9
À5.9
c
d
e
10
f
À5.6
À6.6
À5.6
À6.6
À5.6
À6.6
À5.6
À6.6
a
c
À6.5
À6.5
À6.7
À6.2
À6.2
À6.7
À6.2
À5.9
À6.7
À6.2
À5.9
À6.7
12
e
e
À5.7
À5.7
À5.7
À5.7
11

B. Madrigal et al. / Bioorg. Med. Chem. 10 (2002) 303–312
307
under the basic conditions). Careful diazomethane
Discussion
treatments in dry ethyl ether render the methyl esters 10
and 11. If the reaction is not carefully controlled and
allowed to stand for long periods of time, 10 undergoes
polar cycloaddition reactions of diazomethane onto the
double bond to generate pyrazolines. For the naphtha-
lene series, additional acidic treatment (p-TsOH in
refluxing benzene) is necessary to regenerate the thur-
iferic acid analogue 10 from the product 11. In order to
ascertain the relative stereochemistry of the synthesised
analogues, the methyl esters were treated with hydro-
chloric acid in dry CH2Cl2, rendering the corresponding
chlorides 12.
In previous papers, we have set the strategy for the
synthesis of natural lignans and heteroanalogues based
on a conjugate addition–alkylation reaction to assemble
the basic lignan skeleton, followed by cyclisation
towards the aryltetralin structures and epimerisation
reactions to achieve the desired relative stereo-
chemistries.5,8 The replacement of the methylenedioxy-
phenyl ring of podophyllotoxin alters the synthetic
intermediates and the final products, by modifying their
stereochemical preferences and electronic properties and
making them likely to have modified reactivity and/or
biological properties.4,5,9,10
From a chemical point of view, the effects of the thio-
phene and furan substitutions with respect to the podo-
phyllotoxin series can be summarised as a reduction of
the solvent accessible surface of the A–B ring system
and a more strained C–D ring system, probably due to
the different bond angles imposed by the heteroaromatic
moiety, which leads to a greater preference for the less
strained picropodophyllin series 7.4,10
Biological Results
The cytotoxic activities against several representative
tumor cell lines (P-388, HT-29, MEL-28 and A-549)
were tested following a previously described protocol.16
The results are summarised in Table 2. The IC50 for
every compound tested against each cell line are in the
micromolar range. The methyl thuriferates (10) display
a potency similar to the chloro derivatives (12), as pre-
viously observed for thuriferic acid methyl ester and its
chloro derivative. However, not every addition product
to the double bond leads to active compounds, as the
methoxy (11) and pyrazoline (not shown) derivatives of
the naphthalene series are tenfold less potent. As a gen-
eral trend, the furan and thiophene series are less potent
than the naphthalene analogues (which in turn are
almost equipotent to the natural series). The carbazoles
are less potent than the naphthalenes.
For the naphthalene and carbazole series, the cyclisa-
tions onto the more reactive positions yield compounds
in which the A–B ring systems form an angle with the
newly formed C–D ring system (‘bent’ series). As a
result, ring A comes closer to the pendant trimethoxy-
phenyl ring (ring E) than in the methylenedioxy-, furan-,
and thiophene-‘linear’ series. This proximity reduces the
conformational flexibility of the resulting molecules, as
evidenced by comparison of their NMR spectral data
Figure 3. Synthesis of carbazole analogues. Key: (i) TFA, CH2Cl2, 24 h. (ii) HgO, BF3.Et2O, THF–H2O (85:15), 0 ꢀC–> rt, 24 h; (iii) EtOH/HOAc,
reflux, 48 h. (iv) 1% KOH/MeOH, rt, 1.5 h; then HCl 2 N. (v) etheral CH2N2.

308
B. Madrigal et al. / Bioorg. Med. Chem. 10 (2002) 303–312
The cytotoxic activities (Table 2) displayed by the syn-
thesised compounds show that there are no stringent
requirements for the nature of rings A and B of thur-
iferic acid and its derivatives. Replacement of the
methylenedioxyphenyl ring by thiophene, furan, naph-
thalene or carbazole ring systems lead to compounds
with comparable potencies, considering the substantial
changes introduced in terms of stereochemical pre-
ferences (as previously discussed), hydrophobic acces-
sible surface, steric and electronic properties. These
results are contradicting with the described SAR for
other lignan analogues, which usually are not tolerant
to substitutions of the methylenedioxy ring.3
The cytotoxicity results, combined with the stereo-
chemical preferences discussed above imply that the
compounds are able to place their trimethoxyphenyl
rings in a similar position, despite their intrinsical pre-
ferences. The higher potency of the naphthalene deriva-
tives, combined with the fact that the equatorial
arrangement for the E ring in such a system is dis-
favoured suggest a closer to axial disposition of the ring
when bound to its target.
Figure 4. Conformational equilibrium of thiophene and furan, naph-
thalene and carbazole analogues of thuriferic acid.
with the ‘linear’ ones (i.e., the resonances for aromatic
1
The furan and thiophene derivatives, with a preference
for the equatorial disposition display a lower potency
(the alternate conformation can be accessed at the
expense of some energy loss, combined to the lower
hydrophobic contact area). The carbazole ring, how-
ever, introduces a much bulkier substituent over the E
ring, probably making unfavourable contacts or pre-
venting the E ring from making favourable interactions
with the target.
protons of the trimethoxyphenyl ring in the H NMR
spectra appear line-broadened due to a slower motion
of the otherwise freely rotating moiety). This closer
spatial proximity of rings A and E must also be effective
during the cyclisation reaction and affects its stereo-
chemical outcome, leading to a higher proportion of
compounds with a podophyllotoxin stereochemistry 5.
Such compounds are nearly unobserved in the ‘linear’
series (furan, thiophene, methylenedioxyphenyl, etc.).10
When compared to the stringent requirements for the A
and B rings of podophyllotoxin analogues, including
heteroanalogues such as those described in this paper,
the tolerance described suggest a different mechanism of
action or a different interaction with its target.3 Such an
observation is in agreement with the removal of the
lactone ring, which is considered critical for the activity
of most lignan analogues. Further experiments are
under course in order to establish the action mechanism
of these compounds.
The removal of the 1,3-dithiane protecting group pro-
ceeds similarly in every studied series, with moderate to
good yields. Longer reaction times lead to the thermo-
dynamically more stable epimerised products at C8 (7).
This isomerisation, which has previously been used by
us on the way to podophyllotoxin analogues,8 is not
essential in this synthetic route, as the stereochemical
information is lost in the final products, carrying a
double bond at that position.
The lactone opening reactions proceed in good yields to
the desired products 8. Their stereochemistries have
been unambiguously established spectroscopically in the
natural series and the thiophene and furan analogues,
considering the 1H NMR coupling constants of the
thuriferic acids and the corresponding chloro-deriva-
tives.2,10 The differences in both series of compounds
have been attributed to a higher contribution of an
equatorial disposition of the pendant trimethoxyphenyl
ring to the NMR data of the thuriferic acid analogues
for the five-membered heteroaromatic rings (Fig. 4).10
The more sterically demanding (A–B system larger and
angled) naphthalene and carbazole rings, on the other
hand impose and axial disposition to the aforemen-
tioned E ring, rendering the C70 equatorially placed
protons singlets in the 1H NMR spectra. The whole
series thus sample the conformational space, exhibiting
distinct preferential conformations in solution.
Conclusions
Several new heterolignan analogues of thuriferic acid
methyl ester have been synthesised and their cytotoxic
activities against several tumor cell lines tested. The effects
of the substitutions on the synthetic methodology are
deep and also affect the structures of the compounds
accessed. The unusual structural tolerance seen in the bio-
logical activity testing, as compared to other compounds
of these series, could be explained by a different mechan-
ism of action. These compounds could then be used as
parent compounds for further defining and characterising
the target. New compounds in which the trimethoxy-
phenyl ring (deemed essential for the usual mechanisms
of the lignanic analogues) is replaced by other moieties
in order to explore the diversity tolerated on ring E by
such a target site, are currently being designed.

B. Madrigal et al. / Bioorg. Med. Chem. 10 (2002) 303–312
309
Experimental
49.3 (CH), 55.7 (C), 56.3 (OCH3), 56.3 (OCH3), 61.5
(OCH3), 70.5 (CH2), 74.0 (CH), 102.6 (CH), 102.6 (CH),
108.6 (CH), 109.0 (CH), 109.0 (CH), 120.3 (CH), 121.5
(CH), 122.5 (C), 122.7 (C), 125.6 (C), 126.1 (CH), 128.8
(CH), 135.7 (C), 136.5 (C), 138.0 (C), 140.9 (C), 153.1 (C),
153.1 (C), 178.7 (C¼O); EIMS m/z (%): 593 (M+, 20).
General
Reagents were used as purchased without further pur-
ification. Solvents (THF) were dried and freshly distilled
before use according to literature procedures. Chroma-
tographic separations were performed on silica gel col-
umns by flash (Kieselgel 40, 0.040–0.063; Merck)
chromatography. TLC was performed on precoated
silica gel polyester plates (0.25 mm thickness) with
fluorescent indicator UV 254 (Polychrom SI F254).
Melting points were determined on a Buchi 510 appa-
ratus and are uncorrected.1H NMR and 13C NMR
spectra were recorded on a Bruker WP 200-SY spectro-
meter at 200 MHz or on a Bruker SY spectrometer at
400 MHz. Chemical shifts (d) are given in ppm down-
field from tetramethylsilane as internal standard and
coupling constants (J values) are in hertz. For FABMS
analyses, a VG-TS250 apparatus (70 eV) was used.
(ꢁ)(3R,4S)-4-[2-(9-ethyl-3-carbazolyl)-1,3-dithian-2-yl)-
3 - [1S - (3,4,5- trimethoxyphenyl)hydroxymethyl]tetrahy-
1
drofuran-2-one (3f). H NMR (200 MHz, CDCl3) d 1.46
(t, 3H, J=7.3), 1.92 (m, 2H), 2.63–2.81 (m, 4H), 3.01
(m, 1H), 3.28 (m, 1H), 3.55 (s, 6H), 3.71 (t, 1H, J=9.9),
3.76 (s, 3H), 4.38 (c, 2H, J=7.3), 4.63 (d, 1H, J=5.1),
4.72 (dd, 1H, J=2.9, 9.9), 6.28 (s, 2H), 7.25 (t, 1H,
J=7.7), 7.34 (d, 1H, J=8.7), 7.39 (d, 1H, J=7.7), 7.48
(t, 1H, J=7.7), 7.88 (dd, 1H, J=1.8, 8.7), 8.13 (d, 1H,
J=7.7), 8.65 (d, 1H, J=1.8); 13C NMR (50.13 MHz,
CDCl3) d 13.8 (CH3), 24.6 (CH2), 26.8 (CH2), 27.1
(CH2), 37.6 (CH2), 49.5 (CH), 51.2 (CH), 55.7 (OCH3),
55.7 (OCH3), 60.3 (OCH3), 63.8 (C), 68.4 (CH2), 73.9
(CH), 102.8 (CH), 102.8 (CH), 108.6 (CH), 108.9 (CH),
119.2 (CH), 120.4 (CH), 121.6 (CH), 122.5 (C), 123.2
(C), 126.3 (CH), 126.7 (CH), 128.9 (C), 135.4 (C), 137.3
(C), 139.0 (C), 140.4 (C), 152.9 (C), 152.9 (C), 176.3
(C¼O); EIMS m/z (%): 593 (M+, 20).
N-Ethylcarbazole series
Conjugate addition–alkylation reaction:. To a solution of
2-(9-ethyl-3-carbazolyl)-1,3-dithiane (1.5 g, 4.8 mmols)
in dry THF (48 mL) at À78 ꢀC under argon was added
n-BuLi (1.6 M in hexane) (3.3 mL, 5.3 mmols). After
45 min,
a
solution of 5H-furan-2-one (0.3 mL,
Cyclisation reaction
4.8 mmols) in dry THF (4.8 mL) was added dropwise.
The reaction mixture was stirred at À78 ꢀC for 3 h and
then allowed to warm to À50 ꢀC. A solution of 3,4,5-
trimethoxybenzaldehyde (1.4 g, 7.1 mmols) in dry THF
(7.2 mL) and TMEDA (1.5 mL, 10 mmols) were succes-
sively added. The mixture was stirred for 12 h at À50oC
and then quenched by addition of NH4Cl (concd solu-
tion), extracted with ethyl acetate, washed with brine,
dried over MgSO4 and evaporated in vacuo. Flash
chromatography (hexane/ethyl acetate 70:30) afforded 1
(0.5 g, 26%), 2f (0.5 g, 17%) and 3f (0.7 g, 25%).
To a solution of 2f and 3f (1.2 g, 2.0 mmols) in CH2Cl2
(10.1 mL) at room temperature TFA (10.1 ml) was
added dropwise. The reaction mixture was stirred for
24 h and then quenched with NaHCO3 (concd solution),
extracted with ethyl acetate, washed with brine, dried
over Na2SO4 and concentrated in vacuo. Crystallization
of the crude product gave a 2:1mixture (480 mg, 41%)
of B-4f and B-5f as white crystals. Pure B-4f was iso-
lated by crystallization. A 3:1 mixture (150 mg, 13%) of
L-4f and L-5f was isolated from the mother liquors.
4-[2-(9-ethyl-3-carbazolyl)-1,3-dithian-2-yl]tetrahydro-
1
furan-2-one (1). H NMR (200 MHz, CDCl3) d 1.47 (t,
3H, J=7.3), 1.90 (m, 2H), 2.48 (dd, 2H, J=8.8, 17.5),
2.70–3.01 (m, 4H), 3.20 (m, 1H), 4.20 (t, 1H, J=8.2),
4.44 (c, 2H, J=7.3), 4.50 (t, 1H, J=8.1), 7.45 (m, 4H),
8.06 (dd, 1H, J=1.8, 8.1), 8.14 (d, 1H, J=8.0), 8.68 (s,
1H); 13C NMR (50.13 MHz, CDCl3) d 14.5 (CH3), 25.5
(CH2), 27.9 (CH2), 27.9 (CH2), 30.9 (CH2), 38.3 (CH2),
(ꢁ)(3aS,12R,12aS)-7-ethyl-4,4-trimethylenedithio-12-
(3,4,5-trimethoxy phenyl)-3,3a,4,7,12,12a-hexahydroiso-
benzofuran[5,6-c]carbazol-1-one (B-4f). Mp 278–280 ꢀC
1
(hexane/ethyl acetate); H NMR (200 MHz, CDCl3) d
1.46 (t, 3H, J=7.3), 2.17–2.27 (m, 2H), 2.79–3.13 (m,
4H), 3.35 (m, 1H), 3.51 (m, 1H), 3.64 (s, 6H), 3.70 (s,
3H), 4.35 (c, 2H, J=7.3), 4.53 (dd, 1H, J=7.3, 10.6),
4.72 (t, 1H, J=7.3), 5.23 (d, 1H, J=8.8), 6.49 (s, 2H),
7.03 (m, 1H), 7.34 (m, 2H), 7.42 (d, 1H, J=8.8), 8.03 (d,
1H, J=7.8), 8.43 (d, 1H, J=8.8); 13C NMR
(50.13 MHz, CDCl3) d 14.4 (CH3), 24.4 (CH2), 30.2
(CH2), 31.0 (CH2), 38.2 (CH2), 43.9 (CH), 47.5 (CH),
54.0 (C), 56.6 (OCH3), 56.6 (OCH3), 57.3 (CH), 61.3
(OCH3), 69.7 (CH2), 106.4 (CH), 106.4 (CH), 108.3
(CH), 108.8 (CH), 119.5 (CH), 122.6 (C), 122.6 (C),
124.9 (C), 125.9 (CH), 127.1 (CH), 132.5 (C), 134.5 (C),
136.9 (C), 140.7 (C), 140.8 (C), 141.3 (C), 153.3 (C),
153.3 (C), 177.0 (C¼O); EIMS m/z (%): 575 (M+, 100).
49.3 (CH), 62.1(C), 69.3 (CH ), 109.3 (C), 109.4 (CH),
2
119.8 (CH), 121.1 (CH), 121.5 (CH), 122.0 (CH), 123.0
(C), 124.0 (C), 126.7 (CH), 127.2 (CH), 130.0 (C), 139.8
(C), 141.1 (C), 176.6 (C¼O); EIMS m/z (%) 397 (M+,
16), 312 (100%).
(ꢁ)(3R,4S)-4-[2-(9-ethyl-3-carbazolyl)-1,3-dithian-2-yl]-
3 - [1R - (3,4,5- trimethoxyphenyl)hydroxymethyl]tetrahy-
1
drofuran-2-one (2f). H NMR (200 MHz, CDCl3) d 1.87
(m, 2H), 2.55–2.58 (m, 2H), 2.70–2.81(m, 2H), 1.46 (t,
3H, J=7.3), 2.92–2.99 (m, 1H), 3.09–3.13 (m,1H), 3.45
(brs, 6H), 3.78 (s, 3H), 4.34–4.46 (m, 2H), 4.40 (c, 2H,
J=7.3), 4.96 (s, 1H), 6.14 (s, 2H), 7.10 (d, 1H, J=8.2),
7.25 (m, 2H), 7.50 (m, 2H), 8.08 (d, 1H, J=8.0), 8.65 (s,
1H); 13C NMR (50.13 MHz, CDCl3) d 14.6 (CH3), 25.3
(CH2), 27.5 (CH2), 27.8 (CH2), 38.3 (CH2), 48.5 (CH),
(ꢁ)(3aR,12R,12aR)-7-ethyl-4,4-trimethylenedithio-12-
(3,4,5-trimethoxy phenyl)-3,3a,4,7,12,12a-hexahydroiso-
benzofuran[5,6 - c]carbazol - 1 - one (B - 5f). 1H NMR

310
B. Madrigal et al. / Bioorg. Med. Chem. 10 (2002) 303–312
(200 MHz, CDCl3) d 1.45 (t, 3H, J=7.3), 2.65–3.05 (m,
6H), 3.20 (m, 1H), 3.45 (m, 1H), 3.67 (s, 6H), 3.74 (s,
3H), 4.10–4.40 (m, 2H), 4.40 (c, 2H, J=7.3), 5.26 (d,
1H, J=4.4), 6.43 (s, 2H), 7.01(m, 1H), 7.25–7.39 (m,
2H), 7.50 (d, 1H, J=8.8), 7.76 (d, 1H, J=8.4), 8.27 (d,
1H, J=8.8); 13C NMR (50.13 MHz, CDCl3) d 13.7
(CH3), 24.1(CH ), 27.1(CH ), 29.2 (CH2), 37.5 (CH2),
(C), 140.5 (C), 143.7 (C), 152.8 (C), 152.8 (C), 174.1
(C¼O), 194.0 (C¼O); EIMS m/z (%): 485 (M+, 60).
(ꢁ)(3aR,12R,12aS)-7-ethyl-12-(3,4,5-trimethoxyphenyl)-
3a,7,12,12a-tetrahydro-3H-isobenzofuran[5,6-c]carbazol-
1
1,4-dione (B-7f). H NMR (200 MHz, CDCl3) d 1.46 (t,
3H, J=7.7), 3.36 (dd, 1H, J=6.2, 8.0), 3.53 (d, 1H,
J=8.0), 3.69 (s, 6H), 3.74 (s, 3H), 4.41(dd, 1H, J=6.2,
9.0), 4.85 (d, 1H, J=9.0), 5.67 (s, 1H), 6.41 (s, 2H), 7.21
(m, 1H), 7.46 (m, 3H), 8.11 (d, 1H, J=8.0), 8.31(d, 1H,
J=8.8); 13C NMR (50.13 MHz, CDCl3) d 13.9 (CH3),
37.8 (CH2), 40.8 (CH), 43.7 (CH), 47.9 (CH), 56.3
(OCH3), 56.3 (OCH3), 60.8 (OCH3), 71.0 (CH2), 105.0
(CH), 105.0 (CH), 109.0 (CH), 109.3 (CH), 120.7 (C),
120.7 (C), 122.6 (C), 123.4 (CH), 124.8 (C), 125.8 (CH),
126.3 (CH), 137.2 (C), 137.2 (C), 138.5 (C), 140.7 (C),
143.9 (C), 153.8 (C), 153.8 (C), 175.9 (C¼O), 194.9
(C¼O); FABMS m/z (%): 486 (M+, 5).
2
2
44.5 (CH), 45.2 (CH), 45.8 (CH), 56.1(OCH ),), 56.1
3
(OCH3), 60.8 (OCH3), 69.3 (CH2), 69.9 (C), 105.4 (CH),
105.4 (CH), 108.0 (CH), 108.6 (CH), 119.3 (CH), 120.2
(CH), 120.8 (CH), 122.7 (C), 122.7 (C), 126.2 (CH),
131.0 (C), 134.0 (C), 139.8 (C), 140.8 (C), 141.7 (C),
141.8 (C), 152.8 (C), 152.8 (C), 178.0 (C¼O); FABMS
m/z (%): 575 (M+, 75).
(ꢁ)(3aS,12R,12aS)-10-ethyl-4,4-trimethylenedithio-12-
(3,4,5-trimethoxy phenyl)-1,3,3a,4,12,12a-hexahydroiso-
benzofuran - [5,6 - b]carbazol - 3 - one (L - 4f). 1H NMR
(200 MHz, CDCl3) d 1.46 (t, 3H, J=7.3), 2.00–3.27 (m,
6H), 3.31(m, 1H), 3.75 (s, 6H), 3.84 (s, 3H), 3.58–3.84
(m, 1H), 4.37 (c, 2H, J=7.3), 4.37 (m, 1H), 4.56 (d, 1H,
J=6.9), 4.75 (dd, 1H, J=4.0, 10.6), 6.52 (s, 2H), 7.24
(m, 1H), 7.34 (m, 1H), 7.45 (m, 1H), 8.12 (d, 1H,
J=7.7), 8.16 (d, 1H, J=7.7), 8.79 (s, 1H).
Acidic epimerisation
To a solution of B-6f (250 mg, 0.5 mmols) in ethanol
(45 mL) was added glacial acetic acid (35 mL,
0.6 mmols). The reaction mixture was stirred at reflux
for 48 h and then allowed to cool to room temperature
and quenched with a concd solution of NaHCO3,
extracted with ethyl acetate, washed with brine, dried
over Na2SO4 and the organic solvent evaporated in
vacuo. Flash chromatography of the crude product
afforded B-7f (92 mg, 38%) and B-10f (R=Et)(120 mg,
47%).
(ꢁ)(3aR,12R,12aR)-10-ethyl-4,4-trimethylenedithio-12-
(3,4,5-trimethoxyphenyl)-1,3,3a,4,12,12a-hexahydroiso-
benzofuran - [5,6 - b] carbazol - 3 - one (L - 5f). H NMR
1
(200 MHz, CDCl3) d 1.38 (m, 3H), 2.00–3.27 (m, 6H),
3.40–3.87 (m, 2H), 3.64 (s, 3H), 3.75 (s, 6H), 3.90–4.13
(m, 2H), 4.12 (m, 2H), 5.26 (d, 1H, J=4.4), 6.92 (s, 2H),
7.20–7.48 (m, 1H), 7.34 (s, 1H), 7.48 (m, 1H), 8.01 (d,
1H, J=8.8), 8.26 (d, 1H, J=8.8), 8.67 (s, 1H).
1
B-10f (R=Et). H NMR (200 MHz, CDCl3) d 0.98 (t,
3H, J=7.1), 1.45 (t, 3H, J=7.3), 3.69 (s, 3H), 3.75 (s,
3H), 3.96 (c, 2H, J=7.1), 4.04 (s, 1H), 4.40 (c, 2H,
J=7.3), 5.33 (s, 1H), 5.73 (s, 1H), 6.37 (s, 3H), 7.20 (m,
1H), 7.24–7.50 (m, 2H), 7.46 (d, 1H, J=8.8), 8.09 (d,
1H, J=8.0), 8.40 (d, 1H, J=8.8); 13C NMR (50.13 MHz,
CDCl3) 13.8 (CH3), 13.8 (CH3), 37.8 (CH2), 46.1(CH),
56.1(OCH 3), 56.1(OCH 3), 56.9 (CH), 60.7 (OCH3), 61.3
(CH2), 105.7 (CH), 105.7 (CH), 108.5 (CH), 108.9 (CH),
120.2 (C), 120.3 (C), 122.6 (CH), 125.3 (C), 125.3 (C),
125.7 (CH2), 125.9 (CH), 126.0 (CH), 136.1 (C), 136.9
(C), 138.3 (C), 138.5 (C), 140.6 (CH), 143.4 (C), 153.3
(C), 153.3 (C), 171.8 (CO), 185.6 (C¼O). FABMS m/z
(%) 514 (M+, 15), 150 (100).
Deprotection reaction
A solution of B-4f and B-5f (0.48 g, 0.83 mmols) in
25 mL of THF/H2O (85:15) was added to a suspension
of HgO (390 mg, 1.9 mmols) in THF/H2O (85:15) at
0 ꢀC BF.3OEt2 (0.23 mL, 1.9 mmols). The reaction mix-
ture was stirred for 24 h at room temperature, then
CH2Cl2 (20 mL) was added and the precipitate filtered.
The organic layer was washed with brine, dried over
Na2SO4 and the solvent evaporated in vacuo. The crude
product was purified by crystallization in hexane/ethyl
acetate affording a mixture (2:1) of B-6f and B-7f (379 g,
79%), which were then separated by chromatography
(hexane/ethyl acetate 8:2).
Lactone opening reaction
(ꢁ)(3aS,12R,12aS)-7-ethyl-12-(3,4,5-trimethoxyphenyl)-
3a,7,12,12a-tetrahydro-3H-isobenzofuran[5,6-c]carbazol-
1,4-dione (B-6f). Mp 242–244 ꢀC (hexane/ethyl acetate);
1H NMR (200 MHz, CDCl3) d 1.45 (t, 3H, J=6.9), 3.12
(dd, 1H, J=9.8, 16.4), 3.45 (m, 1H), 3.63 (s, 6H), 3.70
(s, 3H), 4.40 (c, 2H, J=6.9), 4.50 (t, 1H, J=8.8), 4.62
(m, 1H), 5.32 (d, 1H, J=9.8), 6.49 (brs, 2H), 7.08 (m,
1H), 7.39 (m, 2H), 7.48 ( d, 1H, J=9.0), 7.92 (d, 1H,
J=9.6), 8.21(d, 1H, J=9.0); 13C NMR (50.13 MHz,
CDCl3) d 13.7 (CH3), 37.8 (CH2), 44.8 (CH), 48.3 (CH),
48.7 (CH), 56.1(OCH 3), 56.1(OCH 3), 60.7 (OCH3),
66.3 (CH2), 106.6 (CH), 106.6 (CH), 108.2 (CH), 108.7
(CH), 120.0 (C), 120.0 (C), 121.9 (C), 124.5 (CH), 124.6
(CH), 126.2 (CH), 126.8 (C), 138.3 (C), 138.3 (C), 140.5
Compound B-7f (250 mg, 0.5 mmols) was treated with a
solution of KOH in MeOH (1%) for 1.5 h and then
quenched by addition of HCl (2 N), extracted with ethyl
acetate, washed with brine, dried over MgSO4 and eva-
porated in vacuo affording B-8f (225 mg, 93%).
(ꢁ)(1R,2S)-7-ethyl-3-methylen-1-(3,4,5-trimethoxyphe-
nyl)-4-oxo-1,2,3,4-tetrahydrobenzo[c]carbazole-2-car-
1
boxylic acid (B-8f). H NMR (200 MHz, CDCl3) d 1.44
(t, 3H, J=6.9), 3.65 (s, 6H), 3.72 (s, 3H), 3.98 (m, 1H),
4.36 (c, 2H, J=6.9), 5.29 (s, 1H), 5.65 (s, 1H), 6.31 (s,
2H), 6.34 (s, 1H), 7.12–7.21 (m, 1H), 7.41–7.48 (m, 2H),
7.47 (d, 1H, J=8.0), 7.98 (d, 1H, J=8.0), 8.38 (d,0 1H,
J=8.8);13C NMR (50.13 MHz, CDCl3) ꢀ: 14.0 (CH3),

B. Madrigal et al. / Bioorg. Med. Chem. 10 (2002) 303–312
311
37.8 (CH2), 45.7 (CH), 56.2 (OCH3), 56.2 (OCH3), 56.5
(CH), 60.7 (OCH3), 105.5 (CH), 105.5 (CH), 108.8
(CH), 109.1 (CH), 120.3 (C), 120.4 (C), 122.5 (CH),
123.1 (C), 125.1 (C), 126.1 (CH), 126.3 (CH), 126.9
(CH2), 136.1 (C), 137.0 (C), 137.8 (C), 138.1 (C), 140.6
(CH), 143.6 (C), 153.3 (C), 153.3 (C), 176.7 (C¼O),
185.4 (C¼O).
(ꢁ)(2S, 3S, 4R)-2-methoxymethyl-1-oxo-4-(3,4,5-trime-
thoxyphenyl)-1,2,3,4-tetrahydrophenanthrene-3-methyl-
1
carboxylate (11e). H NMR (200 MHz, CDCl3) d 3.18
(ddd, 1H, J=4.1, 4.7, 8.3), 3.29 (s, 3H), 3.57 (s, 3H),
3.67 (s, 6H), 3.73 (dd, 1H, J=8.3, 9.9), 3.75 (dd, 1H,
J=4.7, 8.2), 3.80 (s, 3H), 4.12 (dd, 1H, J=5.4, 9.9), 5.38
(d, 1H, J=8.8), 6.28 (s, 2H), 7.47 (t, 1H, J=8.6), 7.54 (t,
1H, J=8.6), 7.80 (d, 1H, J=8.6), 7.83 (d, 1H, J=8.6),
7.90 (d, 1H, J=8.6), 8.24 (d, 1H, J=8.6); 13C NMR
(50.13 MHz, CDCl3) d 42.6 (CH), 44.0 (CH), 49.3 (CH),
Esterification reaction
Compound B-8f (225 mg, 0.46 mmols) was treated with
a saturated solution of CH2N2 in ether for 5 min. The
solvent was then evaporated and the crude product was
purified by chromatography (hexane/ethyl acetate 1:1)
affording B-10f (R=Me) (206 mg, 90%).
51.5 (OCH3), 55.4 (OCH3), 55.4 (OCH3), 58.1(OCH ),
3
60.0 (OCH3), 69.1(CH 2), 104.8 (CH), 104.8 (CH), 121.4
(CH), 124.8 (CH), 126.4 (CH), 127.5 (CH), 127.9 (CH),
128.1 (CH), 129.8 (C), 130.3 (C), 135.5 (C), 136.0 (C),
136.4 (C), 137.3 (C), 152.8 (C), 152.8(C), 172.0 (C¼O),
195.1 (C¼O); FABMS m/z (%) 465 (M+, 30).
(ꢁ)(1R,2S)-7-ethyl-3-methylen-1-(3,4,5-trimethoxyphe-
nyl)-4-oxo-1,2,3,4-tetrahydrobenzo[c]carbazole-2-methyl-
carboxylate (B-10f; R=Me). 1H NMR (200 MHz,
CDCl3) d 1.47 (t, 3H, J=6.9), 3.61(s, 3H), 3.69 (s, 6H),
3.79 (s, 3H), 3.99 (s, 1H), 4.42 (c, 2H, J=6.9), 5.54 (s,
1H), 6.36 (brs, 2H), 6.59 (brs, 2H), 7.10 (m, 2H), 7.44
(m, 2H), 7.52 (d, 1H, J=8.6), 8.44 (d, 1H, J=8,6); 13C
NMR (50.13 MHz, CDCl3) ꢀ 13.8 (CH3), 37.8 (CH2),
45.4 (CH), 52.6 (OCH3), 56.1(OCH 3), 56.1(OCH 3),
59.6 (CH), 60.9 (OCH3), 105.7 (CH), 105.7 (CH), 108.9
(CH), 109.0 (CH), 120.4 (C), 120.6 (C), 122.1 (CH),
123.1 (C), 125.0 (C), 126.0 (CH2), 126.1 (CH), 126.7
(CH), 135.9 (C), 136.8 (C), 137.3 (C), 137.3 (C), 140.5
(CH), 143.4 (C), 153.5 (C), 153.5 (C), 173.1 (C¼O),
194.7 (C¼O); EIMS m/z (%): 499 (5), 391(15).
Treatment with p-TsOH
Compound 11e (140 mg, 0.3 mmols) was treated with p-
TsOH (excess) in benzene and stirred at reflux for 7 h.
The reaction mixture was quenched with a concd solu-
tion of NaHCO3, extracted with ethyl acetate, washed
with brine, dried over Na2SO4 and the organic solvent
evaporated in vacuo affording compound 10e (140 mg,
100%).
Chlorination reaction
Compound 10e (200 mg, 0.5 mmols) in CH2Cl2 (15 mL)
was treated with dry HCl gas for 2 h. The organic sol-
vent was evaporated in vacuo and the crude product
purified in MeOH, affording 12e (180 mg, 80%).
Naphthalene series
Lactone opening reaction. Compound 6e (280 mg,
0.7 mmols) was treated with a solution of KOH in
MeOH (1%) for 1.5 h and then quenched by addition of
HCl (2 N), extracted with ethyl acetate, washed with
brine, dried over MgSO4 and evaporated in vacuo
affording a mixture (2:1) of 8e and 9e (250 mg).
(ꢁ)(2S, 3S, 4R)-2-chloromethyl-1-oxo-1-(3,4,5-trime-
thoxyphenyl)-1,2,3,4-tetrahydrophenantrene-3-methylcar-
boxylate (12e). Mp 194 ꢀC (MeOH); 1H NMR
(200 MHz, CDCl3) d 3.18 (dd, 1H, J=5.0, 8.8), 3.59 (s,
3H), 3.61(s, 6H), 3.82 (s, 3H), 3.90 (m, 2H), 4.42 (dd,
1H, J=5.0, 11.7), 5.46 (s, 1H), 6.25 (s, 2H), 7.47 (t, 1H,
J=8.4), 7.58 (t, 1H, J=8.4), 7.82 (d, 1H, J=8.4), 7.88
(d, 1H, J=8.4), 7.90 (d, 1H, J=8.4), 8.20 (d, 1H,
J=8.4); 13C NMR (50.13 MHz, CDCl3) d 43.2 (CH2),
45.3 (CH), 46.3 (CH), 50.4 (CH), 53.1(OCH 3), 56.8
(OCH3), 56.8 (OCH3), 61.4 (OCH3), 106.1 (CH), 106.1
(CH), 122.9 (CH), 126.2 (CH), 127.9 (CH), 129.2 (CH),
129.5 (CH), 129.5 (CH), 130.9 (C), 130.9 (C), 131.7 (C),
136.7 (C), 136.9 (C), 138.9 (C), 154.3 (C), 154.3 (C),
173.0 (C¼O), 195.2 (C¼O); FABMS m/z (%) 469 (M+,
10).
Esterification reaction
A mixture of 8e and 9e (200 mg) was treated with a
saturated solution of CH2N2 in ether for 5 min. The
solvent was then evaporated and the crude product was
purified by chromatography (hexane/ethyl acetate 1:1)
affording 10e (60 mg, 28%) and 11e (140 mg, 64%).
(ꢁ)(3S, 4R)-2-methylen-1-oxo-4-(3,4,5-trimethoxyphe-
nyl)-1,2,3,4-tetrahydrophenantrene-3-methylcarboxylate
(10e). Mp 218 ꢀC (MeOH); 1H NMR (200 MHz,
CDCl3) d 3.51(s, 3H), 3.68 (s, 6H), 3.75 (s, 3H), 4.04 (s,
1H), 5.41 (s, 1H), 5.58 (s, 1H), 6.27 (s, 2H), 6.43 (s, 1H),
7.50 (t, 1H, J=8.4), 7.87 (t, 1H, J=8.4), 7.89 (d, 1H,
J=8.4), 8.02 (d, 1H, J=8.4), 8.28 (d, 1H, J=8.4); 13C
NMR (50.13 MHz, CDCl3) d 44.8 (CH), 52.6 (OCH3),
55.9 (CH), 56.0 (OCH3), 56.0 (OCH3), 60.7 (OCH3),
105.1 (CH), 105.1 (CH), 122.9 (CH), 125.4 (CH), 126.7
(CH), 126.8 (CH2), 127.2 (CH), 128.5 (CH), 128.7 (CH),
130.5 (C), 131.0 (C), 136.3 (C), 136.7 (C), 136.9 (C),
137.6 (C), 140.0 (C), 153.3 (C), 153.3 (C), 172.1 (C¼O),
186.1 (C¼O); FABMS m/z (%): 433 (M+, 100).
Antitumor assays17
P-388 cells (suspension culture) were seeded into 16 mm
wells at 1ꢂ104 cells/well in 1mL aliquot of medium
MEM 10FCS containing the indicate (Table 2) con-
centration of sample. The remaining cell lines: A-549,
HT-29 and MEL-28 (monolayer cultures) were seeded
into 16 mm wells at 2ꢂ104 cells /well in 1mL aliquots of
MEM 10FCS. The day after the inoculum, media were
replaced by 1mL aliquots of MEM 10FCS containing
the different (Table 2) concentrations of sample. In both
cases a separate set of cultures without sample was

312
B. Madrigal et al. / Bioorg. Med. Chem. 10 (2002) 303–312
counted daily to ensure that the cells remained in expo-
nential growth. Cells were incubated at 37 ꢀC in a 10%
CO2 humid atmosphere. All determinations were car-
ried out in duplicate. After three days of incubation,
cells were counted and the IC50 for each sample was
determined.
Bararoain, I.; Andreu, J. M.; Medarde, M. Bioorg. Med.
Chem. Lett. 2000, 10, 2549. (b) Lopez-Perez, J. L.; Olmo, E.;
de Pascual-Teresa, B.; Merino, M.; Martın, S.; San Feliciano,
A. Tetrahedron 1995, 51, 6343.
7. Ziegler, F. E.; Schwartz, J. A. J. Org. Chem. 1978, 43, 985.
8. Medarde, M.; Ramos, A. C.; Caballero, E.; Lopez, J. L.;
Pelaez-Lamamie, R.; San Feliciano, A. Tetrahedron Lett.
1996, 37, 2663.
9. (a) Medarde, M.; Pelaez Lamamie de Clairac, R.; Tome, F.;
Lopez, J. L.; San Feliciano, A. Arch. Pharm. (Weinheim)
1994, 327, 91. (b) Medarde, M.; Pelaez Lamamie de Clairac,
R.; Tome, F.; Lopez, J. L.; San Feliciano, A.; Gravalos, D. G.
Arch. Pharm. (Weinheim) 1995, 328, 403.
Acknowledgements
Financial support came from Spanish DGICYT (SAF98-
0103) and EC (COST contract BIO4CT98-0451).
10. Ramos, A. C.; Pelaez, R.; Lopez, J. L.; Caballero, E.;
Medarde, M.; San Feliciano, A. Tetrahedron 2001, 57, 3963.
11. Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, John Wiley & Sons: New York; 1991, p. 198–207.
12. Medarde, M.; Ramos, A. C.; Pelaez, R.; Caballero, E.;
San Feliciano, A. J. Chem. Soc., Perkin Trans. 1994, 1, 45.
13. Khun, M.; von Wartburg, A. Helv. Chim. Acta 1967,
1546.
References and Notes
1. San Feliciano, A.; Lopez, J. L.; Medarde, M.; Miguel del
Corral, J. M.; de Pascual Teresa, B.; Puebla, P. Tetrahedron
1988, 44, 7255.
2. (a) Lopez-Perez, J. L.; Olmo, E.; de Pascual-Teresa, B.;
Merino, M.; Martın, S.; San Feliciano, A. Tetrahedron 1995,
51, 6343. (b) Lopez-Perez, J. L.; Olmo, E.; de Pascual-Teresa,
B.; Merino, M.; San Feliciano, A. Tetrahedron 1996, 52, 4903.
3. (a) Ter Haar, E.; Rosenkranz, H. S.; Hand, E.; Day, B. W.
Bioorg. Med. Chem. 1996, 4, 1659. (b) Tepe, J. J.; Mada-
lengoitia, J. S.; Slunt, K. M.; Werbovetz, K. W.; Spoors, P. G.;
Macdonald, T. L. J. Med. Chem. 1996, 39, 2188. (c) Cho, S. J.;
Kashiwada, Y.; Bastow, K. F.; Cheng, Y. C.; Lee, K. H. J.
Med. Chem. 1996, 39, 1396.
14. (a) Gonzalez, A. G.; de la Rosa, R.; Trujillo, J. M. Tetra-
hedron 1986, 42, 3899. (b) Boissin, P.; Dhal, R.; Brown, E.
Tetrahedron 1992, 48, 687. (c) Ogiku, T.; Yoshida, S.; Ohmizu,
H.; Iwasaki, T. J. Org. Chem. 1995, 60, 4585. (d) Pelter, A.;
Ward, R. S.; Pritchard, M. C. J. Chem. Soc., Perkin Trans. 1
1988, 1615. (e) Tomioka, K.; Ishiguro, T.; Koga, K. Chem.
Pharm. Bull. 1985, 33, 4333. (f) Brown, E.; Daugan, A. Tetra-
hedron 1986, 45, 141.
15. Vedejs, E.; Fuchs, P. J. J. Org. Chem. 1971, 36, 366.
16. Medarde, M.; Pelaez-Lamamie de Clairac, R.; Lopez,
J. L.; Gravalos, D.Ga; San Feliciano, A. Arch. Pharm.
(Weinheim) 1995, 328, 640.
4. Medarde, M.; Ramos, A. C.; Caballero, E.; Lopez, J. L.;
Pelaez, R.; San Feliciano, A. Tetrahedron Lett. 1998, 39, 2001.
5. Ramos, A. C.; Pelaez, R.; Medarde, M. Heterocycles 1999,
51, 1443.
17. Bergeron, R. J.; Cavanaugh, P. F., Jr.; Kline, S. J.;
Hughes, R. G., Jr.; Elliott, G. T.; Porter, C. W. Bioch. Bioph.
Res. Commun. 1984, 3, 121 .
6. (a) Maya, A. B.; del Rey, B.; Pelaez, R.; Caballero, E.;