An efficient synthesis of furano analogs of duocarmycin C1 and C2: Seco-iso-cyclopropylfurano[e]indoline-trimethoxyindole and seco- cyclopropylfurano[f]quinoline-trimethoxyindole

Article abstract of DOI:10.1016/j.tetlet.2014.04.062

seco-Iso-cyclopropylfurano[e]indoline-trimethoxyindole 1 and seco-cyclopropylfurano[f]quinoline-trimethoxyindole 2 are potent cytotoxic agents. Previous synthesis of these compounds was inefficient and the approach required separation of a mixture containing the isomeric indoline and quinoline intermediates 11 and 12, formed from radical cyclization of allylic bromide 10. Reported herein is an efficient and selective synthesis of either intermediate 11 or 12, thus product 1 or 2, as needed. The conditions of the Stobbe condensation, bromination, 5-exo-trig radical cyclization, amine-carboxylic acid coupling, and debenzylation were optimized for multi gram scale. As a result, the reaction times were shortened, the products were readily isolated, and the yields and purity improved.

Full text of DOI:10.1016/j.tetlet.2014.04.062

Accepted Manuscript
An efficient synthesis of Furano Analogs of Duocarmycin C1 and C2: seco-Iso-
cyclopropylfurano[e]indoline-trimethoxyindole and seco-Cyclopropylfura-
no[f]quinoline-trimethoxyindole
Pravin C. Patil, Moses Lee
PII:
S0040-4039(14)00674-1
DOI:
http://dx.doi.org/10.1016/j.tetlet.2014.04.062
TETL 44524
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
29 March 2014
15 April 2014
16 April 2014
Please cite this article as: Patil, P.C., Lee, M., An efficient synthesis of Furano Analogs of Duocarmycin C1 and
C2:
seco-Iso-cyclopropylfurano[e]indoline-trimethoxyindole
and
seco-Cyclopropylfurano[f]quinoline-
trimethoxyindole, Tetrahedron Letters (2014), doi: http://dx.doi.org/10.1016/j.tetlet.2014.04.062
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

An efficient synthesis of Furano Analogs of Duocarmycin C1 and C2: seco-Iso-
cyclopropylfurano[e]indoline-trimethoxyindole and seco-Cyclopropylfurano[f]quinoline-
trimethoxyindole
Pravin C. Patil,a Moses Lee,a,b,
*
aDepartment of Chemistry and Division of Natural and Applied Sciences, Hope College,
Holland, MI 49422
bCurrent address: Department of Chemistry, Georgia State University, Atlanta, GA 30303
*Corresponding author. Tel.: +1 360 9496310; fax: +1 360 6941819. E-mail address:
mosesleenf@gmail.com (M. Lee).
Graphical Abstract:
OCH3
OCH3
OCH3
OCH3
OCH3
Cl
Cl
N
N
O
O
N
H
N
H
OCH3
O
O
1
OH
2
OH
Abstract:
Seco-iso-cyclopropylfurano[e]indoline-trimethoxyindole
1
and
seco-
cyclopropylfurano[f]quinoline-trimethoxyindole 2 are potent cytotoxic agents. Previous
synthesis of these compounds was inefficient and the approach required separation of a mixture
containing the isomeric indoline and quinoline intermediates 11 and 12, formed from radical
cyclization of allylic bromide 10. Reported herein is an efficient and selective synthesis of either
intermediate 11 or 12, thus product 1 or 2, as needed. The conditions of the Stobbe condensation,
bromination, 5-exo-trig radical cyclization, amine-carboxylic acid coupling, and debenzylation
were optimized for multi gram scale. As a result, the reaction times were shortened, the products
were readily isolated, and the yields and purity improved.
Keywords: Duocarmycins, CC-1065, seco-iso-cyclopropylfurano[e]indoline, seco-
cyclopropylfurano[f]quinoline, 5-exo-trig cyclization.
Duocarmycins and CC-1065 are potent anticancer agents that contain
a
cyclopropylpyrroloindolone (CPI) DNA alkylating functionality and a non-covalently active
portion, such as the 5,6,7-trimethoxyindole-2-carbonyl moiety.1,2 Upon binding within the minor
groove of AT-rich sequences, the reactivity of the cyclopropane is enhanced for covalent
bonding with adenine-N3,1,2 leading to the killing of cancer cells through apoptosis.3 CC-1065
and duocarmycin analogs, such as adozelesin, carzelesin, KW5189, and bizelesin were
developed and subjected to clinical studies. Due to toxicity to the bone marrow clinical studies of

these compounds were discontinued.1,2,4 Consequently, there is a strong interest to design and
develop novel analogs of CC-1065 and the duocarmycins that possess antitumor activity but with
reduced toxicity toward bone marrow cells. Cyclopropylbenzo[e]indolone (CPI),5
cyclopropylpyrazolo[e]indoline
(CPzI),6
cyclopropylfurano[e]indolone
(CFI),7
and
cyclopropylindolone (CI)8 are examples of analogs that were synthesized and tested for
biological activity; their structures are depicted in Figure 1.
H3C
N
H3C
R
R
N
N
N
O
O
N
H
N
H
O
O
CPzI
CPI
R
R
N
H3CO
O
O
O
O
O
O
CFI
CI
Figure 1. Examples of known DNA alkylating subunits of duocarmycins and CC-1065.
Our group has contributed to this work by synthesizing and testing of seco-iso-
cyclopropylfurano[e]indoline-trimethoxyindole
cyclopropylfurano[f]quinoline-trimethoxyindole (seco-CFQ-TMI, 2); the structures are depicted
in
Figure 2.9, 10 Both compounds are highly cytotoxic against cancer cells in culture, and
(seco-iso-CFI-TMI,
1)
and
seco-
less toxic to murine bone marrow cells compared to the corresponding seco-CBI-TMI analog 3.9a
CC-1065 and duocarmycin analogs, such as adozelesin,11a centanamycin (an achiral analog of
3),11b-c and seco-iso-CFI-TMI, 211d are also highly cytotoxic against the growth of Plasmodium
falciparum in culture.
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
Cl
Cl
Cl
OCH3
OCH3
N
N
N
O
O
N
H
N
H
N
H
OCH3
O
O
O
1
OH
2
3
OH
OH
Figure 2. Structures of targeted compounds seco-iso-CFI-TMI 1, seco-CFQ-TMI 2, as well as
seco-CBI-TMI 3.
Despite the potency and biological activity of compounds 1 and 2, further examination of
their anticancer or anti-parasitic activity has been hindered by their limited availability. That was
due to several inefficient steps in our earlier reported synthetic approach, as generally depicted in
Scheme 1.9a First, the Stobbe condensation of furfural 4 with diethyl succinate 5 to produce
benzofuran 6 required long reaction times, and isolation of the product was difficult resulting in
poor yields. Second, the bromination of compound 8 gave multiple products and the yield was
low. Third, the 5-exo-trig radical cyclization of allylic bromide 10 yielded an unpredictable
2

mixture of indoline and quinoline intermediates 11 and 12, respectively. This required time
consuming purification by silica gel column chromatography. Fourth, the overall yields for the
conversion of the indoline and quinoline intermediates 11 and 12 to their corresponding amides
13 and 14 were low. Finally, upon scaling up the reaction from milligrams to multi gram
quantities, catalytic hydrogenation removal of the benzyl group in compound 13 yielded an over-
reduced and difficult to remove by-product 15 in addition to the desired compound 1.
Each of the aforementioned limitations has been addressed and the results are described
below. The synthesis of benzofuran 6 was adapted from the report of Toda et al.12 Furfural 4 was
reacted with diethyl succinate 5 under neat conditions, and with grinding using a mortar and
pestle. The addition of solid potassium t-butoxide was made slowly under a gentle stream of
nitrogen. The reaction afforded a readily isolable isomeric mixture of half esters. This reaction
took 20 min at room temperature, compared to 24 h under reflux in our earlier reported
procedure.9a The half esters were immediately converted into the hydroxy substituted benzofuran
6 by Friedel-Crafts acylation (AcONa, Ac2O, reflux), followed by removal of the acetate group
using potassium carbonate in ethanol. The overall yield of 6 after purification by column
chromatography was 30%, which was five times higher than the earlier reported process. This
synthetic transformation was readily adaptable to a 50 g scale.
Optimization of the bromination of compound 8 to produce bromide 9 was achieved in
the following manner. Instead of using NBS, p-TsOH, and a solvent containing a 1:1 mixture of
dry methanol and THF, which gave a 42% yield of 9,9a we discovered that the reaction in pure
methanol (7.0 g scale of 8, 1.1 equiv. NBS, 265 mL dry MeOH, 0.07 equiv. p-TsOH, -78 °C, 2
h) or THF (2.3 g scale of 8) gave 9 in yields of 93% and 85%, respectively, after purification by
flash silica gel column chromatography. Dry methanol was freshly distilled over
magnesium/iodine, and since it consistently gave a higher yield than using dry THF as the
solvent, it was selected as the solvent of choice.
Another problematic step in the original synthesis of the target compounds 1 and 2 was
the unpredictable radical 5-exo-trig cyclization of bromo-allylic chloride 10.9a In a paper
published recently from our laboratory and guided by results from theoretical calculations, we
discovered that the radical 5-exo-trig cyclization could be controlled by solvent effects.
Specifically, using degassed dry pyridine as a solvent, allylic bromide 10 was selectively
converted into the seco-iso-CFI intermediate 11 in 83% yield (500 mg of 10, 0.1 g AIBN, 0.7
mL nBu3SnH, 40 mL dry pyridine, degassed with a gentle stream of nitrogen bubbles, 90-95 °C,
2 h).13 We report herein that the seco-CFQ intermediate 12 could be produced selectively in 82%
by using non-degassed toluene as solvent (500 mg of 10, 0.1 g AIBN, 0.7 mL nBu3SnH, 40 mL
toluene, 100-110 °C, 3 h). In both cases negligible amounts of the other isomer was observed.
To complete the synthesis of target molecules 1 and 2, the BOC protecting group on
intermediates 11 and 12 was removed by treatment with a 3M solution of HCl in dry ethyl
acetate. The resulting amines were directly coupled with 5,6,7-trimethoxyindole-2-carboxylic
acid in the presence of EDCI.HCl to produce amides 13 and 14 in 90% and 83% yields,
respectively. The overall yields for the synthesis of 13 and 14 from 10 were enhanced to 74%
and 68%, compared to the earlier reported yields of 26% and 38%, respectively.9a We found that
3

using freshly prepared 3M HCl in dry ethyl acetate was crucial for improving the yield, as was
the need to use scrupulously dry and pure 5,6,7-trimethoxyindole-2-carboxylic acid.
O
O
O
O
COOEt
OH
a, b, c
OEt
d
H
+
O
O
COOEt
OR
e
OH
7
4
6
5
Cl
X
Br
NHBOC
NBOC
g
O
O
OR
OR
10
8, X = H
9, X = Br
f
h
i
Cl
Cl
NBOC
NBOC
O
O
OR
12
OR
11
j
j
OCH3
OCH3
OCH3
Cl
Cl
OCH3
OCH3
N
N
O
O
N
H
N
H
OCH3
OCH3
Cl
O
O
OCH3
OCH3
OR
k
OR
k
N
O
N
H
13, R = Bn
1, R = H
14, R = Bn
2, R = H
O
15
OH
a
Scheme 1. An improved synthesis of seco-iso-CFI-TMI 1 and seco-CFQ-TMI 2. Potassium t-
b
c
butoxide, rt, N2, grinding; acetic anhydride, sodium acetate, reflux; K2CO3, ethanol, reflux
(30% for three steps); d(i) benzyl bromide, NBu4I, DMF, K2CO3, 80 °C (70%); (ii) (a) NaOH; (b)
e
HCl (87% in two steps); DPPA; t-butanol, Et3N, reflux, overnight (62 %); NBS, p-TsOH, dry
f
g
MeOH, -78 °C (93%); NaH, DMF, 1,3-dichloro-2-propene (85%); AIBN, Bu3SnH, degassed
h
dry pyridine (N2 purge 30 min), 90-95 °C (83% of 11); iAIBN, Bu3SnH, toluene (non-degassed),
j
100-110 °C (82% of 12); (i) 3M HCl/EtOAc, dry EtOAc; (ii) 5,6,7-trimethoxyindole-2-
k
carboxylic acid, EDCI.HCl, dry DMF, rt (90 and 83% for 13 and 14, respectively); 10% Pd-C,
1,4-cyclohexadiene, 6:4 ethanol:THF, 35-40 °C [71% for 1 (99.6% pure) and 53% for 2 (> 99%
pure)].
4

Finally, the catalytic hydrogenation step required for removing the benzyl group in
compounds 13 and 14 to produce the respective desired products 1 and 2 became a challenge
when the reaction scale was increased from milligrams to multi gram quantities. That was due to
over reduction of the furan moiety during the debenzylation step thereby contaminating the final
product 1 with an almost inseparable dihydrofuran derivative 15. Accordingly, a study to
optimize the procedure for removing the benzyl group from amide 13 was carried out. The
catalyst, reducing agent, temperature, and solvent were systematically optimized. The results
from our studies demonstrated that atmospheric pressure catalytic hydrogenation of compound
13 using 10% Pd-C in methanol or 10% Pd-C and ammonium formate in methanol or acetic acid
(from 5 °C to room temperature) consistently gave the desired product 1 that was contaminated
with 20-30% of the over-reduced by-product 15 as judged by 1H-NMR and LC-MS analyses.14
Similarly, the use of Pd(OAc)2, triethylsilane, triethylamine in dichloromethane15 from 5 °C to
room temperature also gave a mixture of products (1 contaminated with about 20% of 15). The
reaction condition that worked best was 10% Pd-C (e.g., 300 mg of catalyst for a 600 mg of 13
reaction), 1,4-cyclohexadiene (0.6 mL) 6:4 ethanol:THF (60 mL) at 35-40 °C. This reaction gave
the desired product 1 in 71% yield after purification using silica gel chromatography, and 99.6%
purity according to LC-MS analysis. Interestingly, when the reaction was conducted at room
temperature, there was no reaction. The 10% Pd-C, 1,4-cyclohexadiene reaction condition was
equally effective for the conversion of quinoline 14 (900 mg scale) to the desired product 2,
which was isolated in 53% yield following chromatography, and with > 99% purity by LC-MS
analysis. The overall process described for the synthesis of duocarmycin analogs 1 and 2 has
been effectively reproduced 25 times from milligrams to multi gram scale, producing sufficient
quantities required for biological studies.
Acknowledgments
The authors thank ACS-PRF and Taiho Pharmaceutical Company (Japan) for support, and Dr.
Michael D. Seymour of Hope College for his assistance in the LC-MS studies.
References
1. (a) Boger, D. L.; Garbaccio, R. M. Acc. Chem. Res. 1999, 32, 1043-1052. (b) Boger, D. L.;
Johnson, D. S. Angew. Chem., Int. Ed. Engl. 1996, 355, 1438-1474. (c) Boger, D. L.; Johnson, D.
S. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 3642-3649. (d) Boger, D. L. Acc. Chem. Res. 1995, 28,
20-29.
2. (a) Li, L. H.; Wallace, T. L.; DeKoning, T. F.; Warpehoski, M. A.; Kelly, R. C.; Prairie, M.
D.; Drueger, W. C. Invest. New Drugs 1987, 5, 329-337. (b) Li, L. H.; Dekoning, J. F.; Kelly R.
C.; Krueger, W. C.; McGovern, J. P.; Padbury, G. E.; Petzold, G. L.; Wallace, T. L.; Ouding, R.
J.; Praire, M. D.; Gebhard, I. Cancer Res. 1992, 52, 4904-4913.
3. (a) Wrasidlo, W.; Johnson, D. S.; Boger, D. L. Bioorg. Med. Chem. Lett. 1994, 4, 631-636. (b)
Wang, Y.; Yuan, H.; Ye, W.; Wright, S. C.; Wang, H.; Larrick, J. W. J. Med. Chem. 2000, 43,
1541-1549.
5

4. Pitot, H. C. IV; Erlichman, C.; Reid, J. M.; Sloan, J. A.; Ames, M. M.; Bagniewski, P. G.;
Atherton-Skaff, P.; Adjei, A. A.; Rubin, J.; Rayson, D.; Goldberg, R. M.; Proc. Am. Assoc.
Cancer Res. 1999, 40, 91-92.
5. (a) Boger, D. L.; McKie, J. A. J. Org. Chem. 1995, 60, 1271-1275. (b) Boger, D. L.; Yun, W.;
Han, N. Bioorg. Med. Chem. 1995, 3, 1429-1453.
6. Baraldi, P. G.; Cacciari, B.; Pinedo de las Infantas, M. J.; Romagnoli, R.; Spatullo, G.; Cozzi,
P.; Mongelli, N. Anti-Cancer Drug Des. 1997, 12, 67-74.
7. (a) Mohamadi, F.; Spees, M. M.; Staten, G. S.; Garder, P.; Kipka, J. K.; Johnson, D. A.;
Boger, D. L.; Zarrinmayeh, H. J. Med. Chem. 1994, 37, 232-239. (b) Patel, V. F.; Andis, S. L.;
Enkema, J. K.; Johnson, D. A.; Kennedy, J. H.; Mohamadi, F.; Schultz, R. M.; Soose, D. J.;
Spees, M. M. J. Org. Chem. 1997, 62, 8868-8874.
8. (a) Boger, D. L. ; Munk, S. A.; Zarrinmayeh, H.; Ishizaki, T.; Haught, J.; Bina, M.
Tetrahedron 1991, 47, 2661-2682. (b) Boger, D. L.; Ishizaki, T.; Zarrinmayeh, H.; Munk, S. A.;
Kitos, P. A.; Suntorwart, O. J. Am. Chem. Soc. 1990, 112, 8961-8971.
9. (a) Howard, T. T.; Lingerfelt, B. M.; Purnell, B. L.; Scott, A. E.; Price, C. A.; Townes, H. M.;
McNulty, L.; Handl, H. L.; Summerville, K.; Hudson, S. J.; Bowen J. P.; Kiakos, K.; Hartley, J.
A.; Lee, M. Bioorg. Med. Chem. 2002, 10, 2941-2952. (b) Purnell, B.; Lingerfelt, B.; Scott, A.;
Townes, H.; Summerville, K.; Hudson, S.; Kiakos, K.; Hartley, J. A.; Lee, M. Med. Chem. 2006,
2, 139-146.
10. Pati, H.; Howard, F.; Townes, H.; Lingerfelt, B.; McNulty L.; Lee, M. Molecules 2004, 9,
125-133.
11. (a) Yanow, S. K.; Purcell, L. A.; Pradel, G.; Sato, A.; Rodriguez, A.; Lee, M.; Spithill, T. W.
J. Infect. Dis. 2008, 197, 527-534. (b) Yanow, S. K.; Purcell, L. A.; Spithill, T. W.
Mol. Biochem. Parasitol. 2006, 148, 52-59. (c) Purcell, L. A.; Wong, K. A.; Yanow, S. K.; Lee,
M.; Spithill, T. W.; Rodriguez, A. Vaccine 2008, 26, 4880-4884. (d) Good, M. F.; Reiman, J. M.;
Rodriguez, I. B.; Ito, K.; Yanow, S. K.; El-Deeb, I. M.; Batzloff, M. R.; Stanisic, D. I.;
Engwerda, C.; Spithill, T.; Hoffman, S. L.; Lee, M.; McPhun, V. J. Clin. Invest. 2013, 123,
3353–3362.
12. Tanaka, K.; Sugino, T.; Toda, F. Green Chem. 2000, 2, 303-304.
13. Patil, P.; Cousins, K.; Smith, M.; Wieskamp, S.; Ferrara, M.; Bruce, C. D.; Lee, M.
Tetrahedron Lett. 2013, 54, 4756-4759.
14. LC-MS analysis on 1 and 2. The studies were performed on an Agilent 1100 Series
LC/MSD, equipped with a single quad, positive electrospray ionization, and a scan range of 105
- 600 m/z. The effluent was also analyzed using a UV detector at 210, 254 and 280 nm in series
with the mass spectrometer. A Phenomenex Kinetex 2.6µ C18 100A, 100 x 4.6 mm HPLC
Column was used. A binary solvent system was used: A: 25% CH3CN, 75% H2O, 0.1% formic
6

acid; B: 100% CH3CN, 0.1% formic acid. The flow rate was 0.5 mL/min for entire run with the
following gradient. Time, min (%B): 0 min (0), 4 (0), 10 (85), 22 (85), 25 (0), 30 (0) stop time.
The retention times for 1 and 2 were 12.4 and 12.2 min, respectively.
15. Coleman, R. S.; Shah, J. A. Synthesis 1999, 1399-1400.
7