Further developments in the synthesis of lamellarin alkaloids via direct metal-halogen exchange

Article abstract of DOI:10.1016/S0040-4039(02)02887-3

Direct metal-halogen exchange of 2-bromopyrrole carbonate derivatives with tert-butyllithium followed by the intramolecular lactonization of the resulting 2-pyrrole anion onto the carbonate provided the corresponding lamellarins in moderate to good yield.

Full text of DOI:10.1016/S0040-4039(02)02887-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 1363–1366
Further developments in the synthesis of lamellarin alkaloids via
direct metal–halogen exchange
Poonsakdi Ploypradith,a Wiyada Jinaglueng,b Chitkavee Pavarob and Somsak Ruchirawata,c,d,
*
aChulabhorn Research Institute, Vipavadee Rangsit Highway, Bangkok 10210, Thailand
bDepartment of Pharmaceutical Chemistry, Mahidol University, Rama 6 Road, Bangkok 10400, Thailand
cDepartment of Chemistry, Mahidol University, Rama 6 Road, Bangkok 10400, Thailand
dProgramme on Research and Development of Synthetic Drugs,
Institute of Science and Technology for Research and Development, Mahidol University, Salaya Campus, Thailand
Received 28 October 2002; revised 11 December 2002; accepted 20 December 2002
Abstract—Direct metal–halogen exchange of 2-bromopyrrole carbonate derivatives with tert-butyllithium followed by the
intramolecular lactonization of the resulting 2-pyrrole anion onto the carbonate provided the corresponding lamellarins in
moderate to good yield. The lamellarin framework could be obtained from the direct metal–halogen exchange strategy in a
26–33% overall yield over 5–6 steps. © 2003 Elsevier Science Ltd. All rights reserved.
Lamellarins 1, whose structures contain polyoxy-
genated aromatics on their periphery and can be
classified as 3,4-diarylpyrroloisoquinoline lactones, are
a group of marine natural products isolated from the
prosobranch mollusc Lamellaria sp. and also from the
ascidians.1,2 Including the first four lamellarins isolated
by Faulkner in 1985, a total of 35 lamellarins have been
isolated and identified thus far.3,4
of 0.7 mg/mL (0.06 mM) and 0.4 mg/mL (0.04 mM),
respectively.3 A recent study by Faulkner also showed
that the presence of sulfate groups on the periphery
could greatly influence the selectivity of HIV-1 inte-
grase inhibition.7 More importantly, lamellarins also
act as non-toxic inhibitors of acquired multi-drug resis-
tance (MDR).8 Lamellarin I (X=OMe; R1, R2, R3, R4
and R5=Me; R6=H; Y=H) showed sensitizing effects
in
multidrug-resistant
P388/Schabel
cells
to
doxorubicin.3,9
Up to now, several studies directed towards the total
synthesis of these marine natural products have been
reported,10 notably by Steglich,11,12 Banwell,13,14 Boger9
and Ishibashi.15,16 Previously, our research group
reported an efficient synthesis of lamellarin derivatives,
as shown in Scheme 1.17 Synthesis of the lamellarin
skeleton was achieved by first condensing the appropri-
ately substituted benzylisoquinoline 2 with the phenacyl
bromide mesylate 3. The resulting 2H-3,4-disubstituted
pyrrole intermediate 4 was smoothly formylated under
Vilsmeier conditions. Following the removal of the
mesyl group, the cyclic hemiacetal (lactol) 6 was oxi-
dized to give the desired lamellarin skeleton 7.
Some of the lamellarins have been found to exhibit a
wide array of interesting and significant biological
activities including cell division inhibition, cytotoxicity,
HIV-1 integrase inhibition and immunomodulatory
activity.5,6 Lamellarin K (X=OH; R1, R2, R3 and R5=
Me; R4 and R6=H; Y=H) and lamellarin L (X=H;
R1, R3, and R6=H; R2, R4 and R5=Me; Y=H), for
example, exhibited significant cytotoxicity against P388
and A549 cultured cancer cell lines with the mean IC50s
One drawback, albeit a minor one, in our previous
Scheme was the use of a mesyl protecting group, which
added two steps to the synthesis. It occurred to us that
a better approach could be realized by using a hydroxy
protecting group on the phenacyl bromide synthon that
can act as a directing group for the remote deprotona-
Keywords: lamellarin alkaloid; metal–halogen exchange; DreM; natu-
ral products; pyrrole.
* Corresponding author. Tel.: 662-574-0622; fax: 662-574-2027;
e-mail: somsak@tubtim.cri.or.th
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02887-3

1364
P. Ploypradith et al. / Tetrahedron Letters 44 (2003) 1363–1366
Scheme 3. Reagents and conditions: (a) BF3·Et2O, Ac2O, 80–
90°C, 90%; (b) Et2NC(O)Cl, DMAP (cat.), Et3N, CH2Cl2, rt,
82% (12a) and 79% (12b); (c) NaH, EtOCOCl, THF, rt, 96%
(14a) and 83% (14b); (d) BnMe3NBr3, CH2Cl2, 0°C to rt, 82%
(13a), 70% (13b), 70% (15a) and 90% (15b).
synthesized in 90% yield from acetylation of 3,4-
dimethoxyphenol with acetic anhydride and BF3·Et2O,
as shown in Scheme 3. Use of DMAP, Et3N and
N,N-diethylcarbamoyl chloride smoothly converted 11a
and 11b into their corresponding carbamate derivatives
12a and 12b in 82 and 79% yields, respectively. How-
ever, when similar reaction conditions were used for
carbonating 11a and 11b, the desired products 14a and
14b were produced in only 66 and 49% yields, respec-
tively, since the product was often obtained as an
inseparable mixture with remaining starting material.
The use of a stronger base such as NaH in place of
Et3N and ethyl chloroformate yielded the desired car-
bonate derivatives 14a and 14b in 96 and 83% yields,
respectively, with no starting material remaining. Sub-
sequent bromination of 12a, 12b, 14a and 14b with
BnMe3NBr3 effectively provided the desired phenacyl
bromide derivatives 13a, 13b, 15a and 15b in 82, 70, 70
and 90% yields along with the dibrominated products
in approximately 8% yield.
Scheme 1. Reagents and conditions: (a) K2CO3, CH3CN,
reflux, 63%; (b) DMF, POCl3, rt, 80–82%; (c) KOH, EtOH,
reflux, 77–81%; (d) MnO2, CH2Cl2, rt, 20–54%; (e) Pd(OAc)2,
PPh3, K2CO3, DMF, PhBr, 120°C, 12 h, 80%.
tion at the C-2 position of the pyrrole as well as being
the source of the lactone group in the subsequent
lactonization of the resulting anion without the need
for a separate formyl group equivalent. This strategy
was pioneered by Snieckus and termed DreM (for
directed remote metalation).18 The directing group is
typically a carbonate or a carbamate group, as depicted
in Scheme 2. Alternatively, the 2H-pyrrole intermediate
9 could be selectively brominated at the 2-position8 of
the pyrrole to give the corresponding bromo compound
10 which could undergo metal–halogen exchange to
provide an anion similar to that from the DreM strat-
egy after initial remote deprotonation.
Both synthetic strategies required the benzylisoquino-
line 2 and the carbonate or carbamate phenacyl bro-
mide derivatives 8. Our synthesis commenced with the
preparation of 8 starting from commercially available
2-hydroxyacetophenone 11a (R=H) and 2-hydroxy-
4,5-dimethoxyacetophenone 11b (R=OMe) which was
When benzylisoquinoline 2 was reacted with the carba-
mate derivatives 13a and 13b in the presence of
NaHCO3 in refluxing acetonitrile,17 the corresponding
pyrrole carbamates 16a and 16b were obtained in 91
and 81% yields, respectively (Scheme 4). The carbonate
Scheme 2. Directed remote metalation (DreM) and metal–halogen exchange strategies for the synthesis of lamellarin skeleton 7.

P. Ploypradith et al. / Tetrahedron Letters 44 (2003) 1363–1366
1365
steps) and more efficient than our previously reported
one which provided lamellarin 19 only in 25% overall
yield in six steps and lamellarin 20 in 15% overall
yield in seven steps. Between the two strategies, the
direct metal–halogen exchange provided lamellarins
more efficiently. The two best overall yields for the
synthesis of 19 and 20 from DreM and metal–halogen
exchange strategies are 33% in five steps and 26% in
six steps, respectively.
derivatives 15a and 15b were also coupled with 2
under similar conditions to give the pyrrole carbon-
ates 17a and 17b in 72 and 60% overall yields after
subjecting the inseparable mixture of the desired car-
bonate product and the pyrrole phenols 18a and 18b
(the decarbonated products) obtained from the cou-
pling reaction to the carbonation conditions with
DMAP, Et3N and ethyl chloroformate.
With the required carbamates 16a and 16b and car-
bonates 17a and 17b in our hands, we then per-
formed a study of the DreM methodology of these
compounds. After some exploratory work, we found
that refluxing the carbonate 17a with 7 equiv. of
LDA in THF for 36 h gave the desired lamellarins 19
but in only 35% yield. In addition to the low yields,
in our hands, the DreM/cyclization reactions were not
highly reproducible and partial deprotonation of the
starting material was frequently encountered. These
problems together with the seemingly required pro-
longed reaction time have prompted us to consider
another approach. The alternative approach ideally
would feature a more effective means of generating
the C-2 pyrrole anion as well as of facilitating the
cyclization of the resulting anion onto the carbonate
or carbamate at lower temperature and with a shorter
reaction time.
Scheme 4. Reagents and conditions: (a) NaHCO3, CH3CN,
reflux, 13a, 91% (16a), or 13b, 81% (16b); (b) NaHCO3,
CH3CN, reflux, 15a or 15b; (c) DMAP, Et3N, CH2Cl2,
ClC(O)OEt, 72% (17a), 60% (17b).
We then considered a more direct way to generate
the C-2 pyrrole anion via metal–halogen exchange,
this would require the corresponding C-2 halo
pyrrole. As shown in Scheme 5, the C-2 position of
the pyrroloisoquinolines 16a, 16b, 17a and 17b could
be selectively brominated with N-bromosuccinimide
(NBS) to give the corresponding bromo pyrroles 21a,
21b, 22a19 and 22b in excellent yields (>95%). Subse-
quent lithium–halogen exchange of carbamates 21a
and 21b using tert-BuLi gave only the corresponding
2-(N,N-diethyl)amido-pyrroles 23a and 23b in virtu-
ally quantitative yield. Various attempts to affect the
ring closure of these amido-pyrroles failed.18 Lithium–
halogen exchange of carbonates 22a and 22b with
tert-BuLi,20 on the other hand, proceeded smoothly
to give the desired lamellarins 1917 and 2017 in 72 and
67% yields, respectively. From the isolation of 23a
and 23b as the product, it appears that cyclization of
the C-2 pyrrole anion may proceed via the intermedi-
acy of the corresponding 2-amido and 2-alkoxycar-
bonyl pyrroles.18
In conclusion, two approaches towards the total syn-
thesis of the lamellarin skeleton have been developed.
Both DreM and metal–halogen exchange strategies
share a similar C-2 pyrrole anion intermediate which,
upon cyclization onto a carbonate or carbamate,
gives the desired lamellarin framework. Results from
both DreM and metal–halogen exchange are summa-
rized in Table 1. From Table 1, the synthesis of
lamellarin 20, with two methoxy groups on the
periphery, is less efficient than that of lamellarin 19.
These two strategies are relatively short (only 4–6
Scheme 5. Reagents and conditions: (a) NBS, CH2Cl2, rt,
99% (21a), 99% (21b), 99% (22a), 95% (22b); (b) tert-BuLi,
THF, −78°C to rt, 99% (23a), 98% (23b), 72% (19), 67%
(20).

1366
P. Ploypradith et al. / Tetrahedron Letters 44 (2003) 1363–1366
Table 1. Summary of total syntheses of lamellarins 19 and 20
Lamellarins
DreM
Metal–halogen exchange
Carbamate yield (%)
Carbonate yield (%)
17
Carbonate yield (%)
d
d
19
20
33b
26c
–
–
a
–
a The reaction was not performed.
b The overall yield of five steps.
c The overall yield of six steps.
d The reaction gave only the amido-pyrrole intermediate.
Acknowledgements
14. Banwell, M. G.; Flynn, B. L.; Hockless, D. C. R.;
Longmore, R. W.; Rae, A. D. Aust. J. Chem. 1999, 52,
755–765.
We acknowledge the financial contribution from the
Thailand Research Fund (TRF; Grant No. RTA/07/
2544 for S.R. and PDF/82/2544 for P.P.) for the gener-
ous support of the research program and the award of
Senior Research Scholar to S.R. We also acknowledge
the facilities in the Department of Chemistry, Mahidol
University provided by the Postgraduate Education and
Research Program in Chemistry (PERCH).
15. Ishibashi, F.; Miyazaki, Y.; Iwao, M. Tetrahedron 1997,
53, 5951–5962.
16. Ishibashi, F.; Tanabe, S.; Oda, T.; Iwao, M. J. Nat. Prod.
2001, 65, 500–504.
17. Ruchirawat, S.; Mutarapat, T. Tetrahedron Lett. 2001,
42, 1205–1208.
18. Chauder, B. A.; Kalinin, A. V.; Taylor, N. J.; Snieckus,
V. Angew. Chem., Int. Ed. 1999, 38, 1435–1438.
19. 22a: Mp 88–89°C; IR (KBr): wmax 2937, 1759, 1495, 1466,
1248, 1135, 1029 cm−1 1H NMR (200 MHz, CDCl3) l
;
1.22 (t, 3H, J=7.2 Hz, OCH2CH3), 3.00–3.11 (m, 2H,
NCH2CH2Ar), 3.40, 3.60, 3.83, 3.87 (4s, 12H, OCH3),
4.09–4.26 (m, 4H, OCH2CH3 and NCH2CH2Ar), 6.71–
6.78 (m, 4H, ArH), 6.83–6.89 (m, 1H, ArH), 7.08–7.11
(m, 2H, ArH), 7.14–7.18 (m, 1H, ArH), 7.22–7.27 (m,
1H, ArH); 13C NMR (75 MHz, CDCl3) l 14.02, 29.02,
43.33, 55.22, 55.65, 55.76, 55.83, 64.25, 102.9, 107.5,
110.9, 111.0, 114.2, 119.5, 121.1, 121.7, 121.8, 122.7,
123.9, 125.4, 126.6, 127.2, 127.8, 127.9, 132.9, 147.2,
147.4, 147.6, 148.6, 149.2, 152.9; LRMS (EI) m/z (rel.
intensity) 609 (M++2, 37), 607 (M+, 43), 529 (100), 483
(23), 481 (23); HRMS (FAB) (C31H30BrNO7+H) calcd
608.1284, found 608.1282.
References
1. Andersen, R. J.; Faulkner, D. J.; He, C.-H.; Van Duyne,
G. D.; Clardy, J. J. Am. Chem. Soc. 1985, 107, 5492–
5495.
2. Davidson, B. S. Chem. Rev. 1993, 93, 1771–1791.
3. Bowden, B. F. Studies in Natural Products Chemistry
(Bioactive Natural Products (Part D)) 2000, 23, 233–283.
4. Ham, J.; Kang, H. Bull. Korean Chem. Soc. 2002, 23,
163–166.
5. Urban, S.; Hickford, S. J. H.; Blunt, J. W.; Munro, M.
H. G. Curr. Org. Chem. 2000, 4, 765–807.
6. Reddy, M. V. R.; Rao, M. R.; Rhodes, D.; Hansen, M.
S. T.; Rubins, K.; Bushman, F. D.; Venkateswarlu, Y.;
Faulkner, D. J. J. Med. Chem. 1999, 42, 1901–1907.
7. Ridley, C. P.; Venkata Rami Reddy, M.; Rocha, G.;
Bushman, F. D.; Faulkner, D. J. Bioorg. Med. Chem.
2002, 10, 3285–3290.
20. A typical procedure is as follows: To a mixture of 22a
(0.30 g, 0.49 mmol) in THF (10 mL) at −78°C was added
tert-butyllithium (0.73 mL, 1.23 mmol, c=1.7 M in pen-
tane). The mixture turned dark red immediately. The
mixture was allowed to stir at −78°C and slowly warmed
up to room temperature at which the mixture was stirred
for 16 h. The reaction was then quenched with water (5
mL) and diluted with EtOAc (5 mL). The two layers were
separated and the aqueous phase was extracted with
EtOAc (2×10 mL). The organic layers were combined,
dried over Na2SO4, filtered, and concentrated under
reduced pressure to give the crude product which was
further purified by column chromatography on silica
(50% EtOAc/hexanes) to give the desired lamellarin 19 as
a solid (0.17 g, 0.35 mmol, 72%). Spectroscopic data of
19 were identical to those of the compound synthesized
by a different approach previously reported in Ref. 17.
8. Furstner, A.; Krause, H.; Thiel, O. R. Tetrahedron 2002,
58, 6373–6380.
9. Boger, D. L.; Boyce, C. W.; Labroli, M. A.; Sehon, C.
A.; Jin, Q. J. Am. Chem. Soc. 1999, 121, 54–62.
10. Diaz, M.; Guitian, E.; Castedo, L. Synlett 2001, 1164–
1166.
11. Peshko, C.; Winklhofer, C.; Steglich, W. Chem. Eur. J.
2000, 6, 1147–1152.
12. Heim, A.; Terpin, A.; Steglich, W. Angew. Chem., Int. Ed.
Engl. 1997, 36, 155–156.
13. Banwell, M.; Flynn, B.; Hockless, D. Chem. Commun.
1997, 2259–2260.