Modular total synthesis of lamellarin D

Full text of DOI:10.1021/jo051083a

Modular Total Synthesis of Lamellarin D
Daniel Pla, Antonio Marchal,† Christian A. Olsen,‡
Fernando Albericio,*,§ and Mercedes AÄ lvarez*,|
Biomedical Research Institute, Barcelona Scientific Park,
University of Barcelona, 08028 Barcelona, Spain
albericio@pcb.ub.es; malvarez@pcb.ub.es
Received May 30, 2005
FIGURE 1. Structures of compounds 1-3.
In addition, Ishibashi et al. described a synthetic route
to lamellarin D6 which they used for the preparation and
biological evaluation of 10 derivatives that differ from
the natural product in the substitution pattern of the
nonheterocyclic aromatic rings. Important conclusions
about the relation between substitution and activity were
reached from that study.7
To prepare libraries of biologically active analogues of
such natural products for lead discovery and/or optimiza-
tion in medicinal chemistry, it is essential to have robust
and versatile chemistries at hand. To that end, we have
developed an efficient and highly convergent synthetic
route to lamellarin D.
A modular total synthesis of lamellarin D, a marine alkaloid
with potent cytotoxic as well as topoisomerase I inhibition
properties, has been accomplished. A sequential and regio-
selective bromination/Suzuki cross-coupling procedure was
applied for the introduction of aryl groups at positions 1 and
2 of scaffold 1. Microwave-assisted 2,3-dichloro-5,6-dicyano-
p-benzoquinone (DDQ) oxidation to yield pyrroloisoquinoline
15, followed by phenol group deprotection and subsequent
lactonization, gave lamellarin D (18% in eight steps from
1).
Lamellarin D is a potent cytotoxic agent first isolated
in 1985 by Faulkner and co-workers from the marine
prosobranch mollusc Lamellaria sp.1 Since then, a family
of more than 30 lamellarins has been isolated from
natural sources, and several synthetic strategies have
been reported.2 Recently, lamellarin D was identified as
a potent inhibitor of topoisomerase I, thus providing some
insight regarding its biological mechanism of action.3
Furthermore, a structure-activity relationship (SAR)
study with derivatives of lamellarin D afforded candi-
dates for in vivo preclinical development of their antitu-
mor activity.4 The total synthesis of lamellarin D was
achieved utilizing a modification of the strategy devel-
oped by Banwell and co-workers for the synthesis of
lamellarin K.5 This strategy allowed the preparation of
several derivatives by acylation of the free phenolic sites.
In a previous communication, we described the utility
of methyl 5,6-dihydropyrrolo[2,1-a]isoquinoline-2-carbox-
ylate 1 as a scaffold for the synthesis of open-chain
lamellarin analogue 2, which contains two methoxyphe-
nols on the pyrrole ring, and for the simplified lamellarin
analogue 3 (Figure 1).8
Derivative 2 differs from lamellarin D in lacking the
lactone ring and the aromatization of the isoquinoline
ring, structural modifications that afford conformational
flexibility, whereas compound 3 lacks the entire aryl ring
at position 2 of the scaffold. Compounds 2 and 3 were
obtained using a synthetic procedure based on selective
halogenation of the pyrrole ring followed by a Pd(0)-
catalyzed Suzuki cross-coupling reaction.
Dibromination of the scaffold was accomplished in
excellent yield using an excess of NBS in THF. Further-
more, regioselective monobromination was achieved in
excellent yield by modifying the reaction time and the
amount of NBS. Both compounds, 2 and 3, proved to have
cytotoxic effects on cancer cell lines; thus, the preparation
and evaluation of a library of related compounds is in
progress.
† Current address: Department of Inorganic and Organic Chemistry,
University of Jae´n, 23071 Jae´n, Spain.
‡ Current address: The Danish University of Pharmaceutical Sci-
ences, Copenhagen, Denmark.
§ Department of Organic Chemistry, University of Barcelona.
| Laboratory of Organic Chemistry, University of Barcelona.
(1) Andersen, R. J.; Faulkner, J.; Chun-heng, H.; Van Duyne, G.
D.; Clardy, J. J. Am. Chem. Soc. 1985, 107, 5493.
The total synthesis of lamellarin D described in the
present paper is based on the findings described for
preparation of 2 and 3. This new protocol involves two
(2) For a recent review, see: Cironi, P.; Albericio, F.; AÄ lvarez, M.
In Progress in Heterocyclic Chemistry; Gribble, G. W., Joule, J. A., Eds.;
Pergamon: Oxford, UK, 2004; Vol. 16, pp 1.
(3) Facompre´, M.; Tardy, C.; Bal-Mahieu, C.; Colson, P.; Pe´rez, C.;
Manzanares, I.; Cuevas, C.; Bailly, C. Cancer Res. 2003, 63, 7392.
(4) Tardy, C.; Facompre´, M.; Laine, W.; Baldeyrou, B.; Garc´ıa-
Gravalos, D.; Francesch, A.; Mateo, C.; Pastor, A.; Jime´nez, J. A.;
Manzanares, I.; Cuevas C.; Bailly, C. Bioorg. Med. Chem. 2004, 12,
1697.
(6) Ishibashi, F.; Miyazaki, Y.; Iwao, M. Tetrahedron 1997, 53, 5951.
(7) Ishibashi, F.; Tanabe, S.; Oda, T.; Iwao M. J. Nat. Prod. 2002,
65, 500.
(8) Olsen, C. A.; Parera, N.; Albericio, F.; AÄ lvarez, M. Tetrahedron
Lett. 2005, 46, 2041.
(5) Banwell, M.; Flynn, B.; Hockless, D. Chem. Commun. 1997, 259.
10.1021/jo051083a CCC: $30.25 © 2005 American Chemical Society
Published on Web 09/02/2005
J. Org. Chem. 2005, 70, 8231-8234
8231

SCHEME 1. Synthesis of Lamellarin D
sequential brominations and cross-coupling reactions
using differently substituted arylboronic esters, followed
by oxidation to give the aromatic isoquinoline ring, and
lactonization to afford lamellarin D (Scheme 1). Key steps
of this procedure are the regioselective bromination of
the tricyclic system 8 in the presence of two highly
activated benzene rings and the efficient preparation and
introduction of building block 11b.
NBS in THF at 70 °C for 1.5 h.12 The electrophilic
substitution occurs on the pentagonal heterocycle exclu-
sively, and no traces of bromination on the activated
benzene rings were observed in these conditions. Similar
regioselectivity was also described for the synthesis of
lamellarin G trimethyl ether.13
Preparation of the 2,4-diisopropoxy-5-methoxyphenyl-
boranes 11a and 11b needed for the second Pd(0)-
catalyzed cross-coupling reaction required optimization.
Thus, the syntheses were attempted from bromobenzene
derivative 1014 by employing bromine-lithium exchange
followed by reaction with borolane 12a or 12b, respec-
tively (Table 1, entries 1 and 2). The phenylboronic acid
derivative 11a was obtained in lower yield than the
borolane 11b, and its purification and manipulation were
more tedious. Furthermore, boronic acid 11a was inef-
ficient to introduce the second aryl moiety through cross-
coupling reaction with 9 using Pd(PPh3)4 and Na2CO3,
as in the previous reaction for the preparation of 7; thus,
coupling with 11b was investigated in detail. Preparation
of 11b was attempted by Pd(0)-catalyzed reaction be-
tween bromide 10 and pinacolborane 13 (entries 3 and
N-Alkylation of methyl pyrrole-2-carboxylate9 with
tosylate 4,10 followed by Heck cyclization, gave scaffold
1. Regioselective bromination of position 1 followed by
Pd(0)-catalyzed cross-coupling with commercially avail-
able boronic ester 6 furnished compound 7. Protection of
the phenol as an isopropoxyether was achieved by reac-
tion with 2-bromopropane in DMF using K2CO3 as a
base.11 The regioselective bromination of 8 to give com-
pound 9 was achieved in 84% yield by treatment with
(9) Methyl pyrrole-2-carboxylate was obtained from pyrrole by
acylation with trichloroacetyl chloride as described by Harbuck, J. W.;
Rapoport, H. J. Org. Chem. 1972, 37, 3618. The 2-trichloroacetylpyrrole
was treated with a solution of NaOMe in MeOH at 0 °C to furnish the
methyl ester.
(10) Tosylate 4 was obtained from the 2-bromo-5-isopropoxy-4-
methoxybenzaldehyde as detailed in Supporting Information. A Wittig
reaction afforded the styrene, while anti-Markovnikov hydroboration
and standard tosylation of the resulting 2-(2-bromo-5-isopropoxy-4-
methoxyphenyl)ethanol gave 4. Preparation of 2-(2-bromo-5-isopropoxy-
4-methoxyphenyl) ethanol was described from the same benzaldehyde
using a different procedure by: Treu, M.; Jordis, U. Molecules 2002,
7, 374.
(12) This reaction required extensive fine-tuning, as longer reaction
times and/or higher temperature resulted in complex mixtures due to
additional bromination. Furthermore, oxidation of the dihydroiso-
quinoline ring was observed by 1H NMR. The reaction could be followed
by HPLC.
(13) Handy, S. T.; Zhang, Y.; Bregman, H. J. Org. Chem. 2004, 69,
2362.
(11) 1H NMR spectra of 7 and 8 differ only in the integrated area of
the isopropoxy signals, whereas two OCH(CH3)2 signals can be
observed at 71.4 and 71.6 ppm in the 13C NMR.
(14) Obtained from 3-isopropoxy-4-methoxybenzaldehyde by Baeyer-
Villiger reaction followed by protection of the resulting phenol and
bromination; see Supporting Information for details.
8232 J. Org. Chem., Vol. 70, No. 20, 2005

TABLE 1. Preparation of Boronic Acid 11a and Boronic
Ester 11b
its steric hindrance. Ultimately, the problem was solved
by using an excess of 11b, Pd(PPh3)4 as the catalyst, and
K3PO4 as a base, instead of K2CO3, affording the biaryl
derivative 14 in high yield.17 Handy and co-workers13 also
described the use of a total of 8 equiv of a boronic acid to
lead to a 46% yield of the coupling product, in a related
structure. Applying our improved protocol with 5 equiv
of 11b, of which 3 equiv were added at the beginning of
the reaction and the last 2 equiv by syringe pump over
2.5 h, led to a yield of 87%.
The aromatization of dihydroisoquinoline 14 to give 15
was achieved using DDQ in CHCl3 in a sealed tube with
controlled microwave (MW) irradiation at 120 °C for 5
min.19 Several experimental conditions with a simple
analogue were tested for the optimization of this oxida-
tion, and the MW-assisted DDQ method proved to be
superior.20 Cleavage of the four isopropoxyether protect-
ing groups21 in 15 with AlCl3 followed by lactonization
using NaH as a base afforded lamellarin D.
In conclusion, the preparation of lamellarin D has been
accomplished in eight steps (18% overall yield) from
scaffold 1. The strategy is based on two consecutive,
regioselective bromination-Suzuki cross-coupling steps
for introducing the appropriate aryl groups in positions
1 and 2 of scaffold 1. Synthesis of an ortho-substituted
borolane (11b) and its coupling to compound 9 were
optimized, and high yields were obtained in the remain-
ing sequence leading to lamellarin D.
In addition to our previous report on open-chain
lamellarins, the methodology reported herein can be
exploited for the preparation of analogues of the natural
product by utilizing the concept of diverted total synthesis
(DTS).22
Experimental Section
2-(2,4-Diisopropoxy-5-methoxyphenyl)-4,4,5,5-tetramethyl-
1,3,2-dioxaborolane (11b). A mixture of 10 (1.59 g, 5.3 mmol),
Et3N (28.9 mL, 21 mmol), Pd(OAc)2 (62.4 mg, 0.28 mmol),
DPEphos (0.29 mg, 0.54 mmol), and 13 (2.36 mL, 15.76 mmol)
in dioxane (5.3 mL) was heated at 100 °C for 13.5 h. After cooling
to room temperature, the reaction mixture was quenched with
saturated NH4Cl, and the aqueous solution was extracted with
Et2O. The organic solution was dried, filtered, and concentrated.
The resulting oil was purified by flash chromatography (SiO2
previously deactivated with 5% Et3N). Elution with hexane/
AcOEt (70:30) afforded 11b (1.47 g, 79%) as a brown syrup: IR
(film) ν 2976, 2929, 1407, 1371, 1145, 1112, 1032. 1H NMR
(CDCl3, 400 MHz) δ 1.29 (d, 6H, J ) 6.0 Hz, 2Me); 1.33 (s, 12H,
4Me); 1.35 (d, 6H, J ) 6.0 Hz, 2Me); 3.84 (s, 3H, OMe); 4.24 (h,
1H, J ) 6.0 Hz); 4.53 (h, J ) 6.0 Hz, 1H); 6.51 (s, 1H); 7.14 (s,
1H). 13C NMR (CDCl3, 100 MHz) δ 21.9 (q, 2Me); 22.2 (q, 2Me);
24.8 (q, 4Me); 56.4 (q); 71.0 (d); 75.0 (d); 83.2 (s, C4, C5); 107.8
a In refluxing dioxane; DPEphos ) (oxydi-2,1-phenylene)bis-
(diphenylphosphine).
4). Pd(OAc)2 together with (DPEphos)15 or biphenyl
ligand [2-(dicyclohexylphosphino)biphenyl]16afforded 11b
in good yields.17,18
The Suzuki-Miyaura cross-coupling reaction between
11b and 9 was tested under several experimental condi-
tions without satisfactory yields. The lack of reactivity
of boronic ester 11b could presumably be attributed to
(19) A doublet at 9.24 ppm in the 1H NMR spectrum, characteristic
of H-5, verified the formation of 15.
(20) Oxidation of methyl 1,2-bis(2-thienyl)pyrrolo[2,1-a]isoquinoline-
3-carboxylate using either (i) DDQ in CHCl3 at reflux temperature,
(ii) MnO2 in refluxing toluene or pyridine, or (iii) Pd-C in toluene or
decalin were unsuccessful (unpublished results).
(21) Protection of phenol groups as isopropoxyethers was previously
demonstrated to be very efficient in the solid-phase synthesis of
lamellarins: Cironi, P.; Manzanares, I.; Albericio, F.; AÄ lvarez, M. Org.
Lett. 2003, 5, 2959. Marfil, M.; Albericio, F.; AÄ lvarez, M. Tetrahedron
2004, 60, 8659.
(15) Kranenburg, M.; van der Burgt, Y. E. M.; Kamer, P. C. J.; van
Leeuwn, P. W. N. M.; Goubitz, K. Fraange, J. Organometallics 1995,
14, 3081.
(16) Wolfe, J. P.; Singer, R. A.; Bryant, H. Y.; Buchwald, S. L. J.
Am. Chem. Soc. 1999, 121, 9550.
(17) Other catalysts such as PdCl2(dppf)2 were inefficient for this
transformation: Baudoin, O.; Gue´nard, D.; Gue´ritte, F. J. Org. Chem.
2000, 65, 9268.
(18) Reaction of 10 with bis(neopentyl glycolato)diboron catalyzed
by Pd(OAc)2 and DPEphos afforded the debrominated compound, 2,4-
diisopropoxy-1-methoxybenzene, instead of the desired building block
11b.
(22) For an example of DTS, see: Gaul, C.; Njadarson, J. T.; Shan,
D.; Diorn, D. C.; Wu, K.-D.; Tong, W. T.; Huang, X.-Y.; Moore, M. A.
S.; Danishefsky, S. J. J. Am. Chem. Soc. 2004, 126, 11326 and
references therein.
J. Org. Chem, Vol. 70, No. 20, 2005 8233

(d); 118.9 (d); 145.2 (s); 150.4 (s); 158.2 (s). MS(CI) 350 (M, 100);
351 (M + 1, 92); 352 (M + 2, 48); 353 (M + 3, 12).
149.5 (s); 149.9 (s); 150.2 (s); 162.9 (s, CdO). MS (MALDI-TOF)
699 (M, 100), 700 (M + 1, 59), 701 (M + 2, 17). HRMS m/z calcd
for C41H49NO9 699.3407, found 699.3402.
Methyl 2-(2,4-diisopropoxy-5-methoxyphenyl)-8-isopro-
poxy-1-(4-isopropoxy-3-methoxyphenyl)-9-methoxy-5,6-di-
hydropyrrolo[2,1-a]isoquinoline-3-carboxylate (14). 11b (3
mL, 1.88 mmol, 0.63M in DMF), Pd(PPh3)4 (145 mg, 0.13 mmol),
and 2 M K3PO4 (1.25 mL) were added to a solution of bromide
9 (350 mg, 0.63 mmol) in DMF (10 mL). The reaction mixture
was stirred at 110 °C, and another portion of 11b (5 mL, 1.25
mmol, 0.25 M in DMF) was added by syringe pump over 2.5 h.
After 6 h of heating, the solvent was evaporated, and the residue
was dissolved in AcOEt. The organic solution was washed with
sodium diethyldithiocarbamate (0.02 M solution), brine, and
water, dried, filtered, and concentrated to give a crude material
that was purified by column chromatography on silica gel.
Elution with hexane/AcOEt (60:40) gave 14 (383 mg, 87%) as a
Methyl 2-(2,4-dihydroxy-5-methoxyphenyl)-8-hydroxy-
1-(4-hydroxy-3-methoxyphenyl)-9-methoxypyrrolo[2,1-a]-
isoquinoline-3-carboxylate (16). A solution of 15 (143 mg,
0.20 mmol) and AlCl3 (148 mg, 1.06 mmol) in dry CH2Cl2 (1 mL)
was stirred at room temperature for 2 h. The reaction mixture
was then quenched with saturated NH4Cl and washed with
water and brine. The organic phase was dried, filtered, and
concentrated, and the resulting crude material was purified by
flash chromatography. Elution with hexane/AcOEt (gradient
from 25:75 to pure AcOEt) gave 16 (69.1 mg, 64%) as a light
brown solid: IR (film) ν 3426, 1679, 1380, 1268, 1242, 1210 cm-1
.
1H NMR (CDCl3, 400 MHz) δ 3.51 (s, 3H, Me); 3.55 (s, 3H, Me);
3.66 (s, 3H, Me); 3.77 (s, 3H, Me); 5.43 (br, 1H, OH); 5.56 (s,
1H, OH); 5.61 (s, 1H, OH); 5.87 (s, OH); 6.33-6.92 (m, 3H); 6.92-
7.27 (m, 5H); 9.19 (m, 1H, H5). 13C NMR (CDCl3, 100 MHz) δ
51.4 (q, Me); 55.4 (q, Me); 56.0 (q, Me); 56.4 (q, Me); 102.7 (d);
104.8 (d); 110.4 (d); 112.5 (d); 112.7 (s); 113.8 (d); 113.9 (s); 114.0
(d); 114.1 (d); 119.1 (s); 123.4 (d, C5); 124.2 (d); 124.5 (s); 131.3
(s); 140.2 (s); 144.8 (s); 145.8 (s); 146.1 (s); 146.6 (s); 146.7 (s);
148.6 (s); 162.3 (s, CdO). MS (MALDI-TOF) 531 (M, 100), 532
(M + 1, 38), 533 (M + 2, 11). HRMS m/z calcd for C29H25NO9
531.1529, found 531.1524.
reddish oil: IR (film) ν 2975, 1693, 1438, 1254, 1208, 1111 cm-1
.
1H NMR (CDCl3, 400 MHz) δ 1.31 (d, 12H, J ) 6.0 Hz, 4Me);
1.36 (d, 12H, J ) 6.0 Hz, 4Me); 3.00 (m, 2H, C5); 3.33 (s, 3H,
OMe); 3.51 (s, 3H, OMe); 3.59 (s, 6H, OMe, CO2Me) 4.09-4.14
(m, 1H, CH); 4.41-4.55 (m, 3H, CH); 4.64-4.69 (m, 2H, C6);
6.46 (s, 1H); 6.48 (s, 1H); 6.66 (br, 1H); 6.74-6.79 (m, 4H). 13C
NMR (CDCl3, 100 MHz) δ 22.0 (q, Me); 22.1 (q, Me); 22.2 (q,
Me); 29.0 (t, C6); 42.8 (t, C5); 50.8 (q, CO2Me); 55.1 (q, OMe);
55.7 (q, OMe); 56.1 (q, OMe); 71.5 (d, OCH); 71.6 (d, OCH); 71.7
(d, OCH); 107.2 (d); 109.1 (d); 114.9 (d); 115.0 (d); 116.0 (d); 116.1
(d); 119.2 (s); 119.5 (s); 121.3 (s); 121.5 (s); 123.3 (d); 125.5 (s);
127.9 (s); 129.1 (s); 130.6 (s); 144.5 (s, C-OMe); 145.7 (s); 145.8
(s); 146.2 (s); 148.5 (s); 149.4 (s, C-OMe); 150.2 (s, C-OMe);
162.7 (s, CdO). MS (MALDI-TOF) 701 (M, 100), 702 (M + 1,
39), 703 (M + 2, 8). HRMS m/z calcd for C41H51NO9 701.3564,
found 701.3558.
Lamellarin D. A mixture of NaH (60% dispersion, 25.7 mg,
0.64 mmol) and 16 (33.9 mg, 0.06 mmol) in THF (3.5 mL) was
stirred for 3 h at room temperature and for 1 h at 40 °C. The
solvent was removed under reduced pressure, and AcOEt was
added to the residue. The organic solution was washed with
saturated NH4Cl, water, and brine and then dried, filtered, and
concentrated. The residue was purified by flash chromatography.
Elution with AcOEt/MeOH (gradient from 100:0 to 80:20)
furnished lamellarin D (23.7 mg, 75%) as a pinkish-white solid.
The spectroscopic data were in accordance with previous re-
ports.1,6
Methyl 2-(2,4-diisopropoxy-5-methoxyphenyl)-8-isopro-
poxy-1-(4-isopropoxy-3-methoxyphenyl)-9-methoxypyrrolo-
[2,1-a]isoquinoline-3-carboxylate (15). A mixture of 14 (411
mg, 0.59 mmol) and DDQ (173 mg, 0.76 mmol) in dry CHCl3
(15 mL) was purged with Ar and irradiated with a microwave
at 120 °C for 5 min in a sealed vessel. The organic solution was
washed with 2 M NaOH, water, and brine. The solvent was
removed to afford a crude residue, which was purified by flash
chromatography on silica gel. Elution with hexane/AcOEt (55:
45) gave 15 (336 mg, 82%) as a pale yellow oil: IR (film) ν 1684,
1211 cm-1. 1H NMR (acetone-d6, 400 MHz) δ 1.25-1.35 (m, 24H);
3.38 and 3.40 (2d, 3H, OMe); 3.54 and 3.59 (2s, 3H, OMe); 3.60
and 3.62 (2s, 3H, OMe); 3.62 (s, OMe); 4.19-4.35 (m, 1H); 4.48-
4.59 (m, 2H); 4.72 (h, 1H, J ) 5.6 Hz); 6.56 (s, 1H); 6.60 (d, 1H,
J ) 8.8 Hz); 6.69-6.82 (m, 2H); 6.90-7.13 (m, 2H), 7.28 and
7.16 (2s, 1H); 7.29 (s, 1H); 9.24 (d, 1H, J ) 7.6 Hz, H5). 13C NMR
(CDCl3, 100 MHz) δ 21.8 (q, 2Me); 21.9 (q, 2Me); 22.1 (q, 2Me);
22.2 (q, 2Me); 50.7 (q, CO2Me); 55.2 (q, OMe); 55.5 (q, OMe);
56.1 (q, OMe); 71.2 (d); 71.4 (d); 71.5, 71.6 (d); 71.8 (d); 105.6
(d), 106.6 (d); 107.0 (s); 110.7 (d); 111.7 (d); 113.1 (s); 115.4 (d),
115.6 (d); 115.9 (d), 116.1 (d); 118.2 (s); 119.4 (s); 119.8 (s); 123.3
(d); 129.4 (s); 131.6 (s); 144.3 (s); 146.0 (s); 146.2 (s); 147.3 (s);
Acknowledgment. This work was partially sup-
ported by CICYT (BQU 2003-00089), Generalitat de
Catalunya, Barcelona Science Park, and PharmaMar S.
L., which is also gratefully acknowledged for performing
the preliminary biological tests. A.M. thanks the Junta
de Andaluc´ıa, UJA, and UB for financial support and
staying facilities. Authors thank Dr. Carmen Cuevas for
her encouragement to perform the present work.
Supporting Information Available: Materials and meth-
1
ods, experimental procedures, characterization data, and H
and 13C NMR spectra. This material is available free of charge
via the Internet at http://pubs.acs.org.
JO051083A
8234 J. Org. Chem., Vol. 70, No. 20, 2005