Synthesis of diketopiperazine-based carboline homodimers and in vitro growth inhibition of human carcinomas

Article abstract of DOI:10.1016/j.bmcl.2008.05.022

Starting from d- or l-tryptophan, we have synthesized and characterized six compounds 2.29-2.31a and b that belong to a class of nitrogen heterocycles: the carboline-based homodimers. Each individual homodimer features a 1,3-trans relationship on each side of the central diketopiperazine core, but differs in absolute stereochemistry and also in substitution on the 4′ and 4″ oxygens (-Bn, -CH₃, or -H). The in vitro cytotoxicity of the six compounds was evaluated by measuring the growth inhibition in NCI-H520 and PC-3 human carcinoma cells. Phenol 2.30a inhibited cancer cell growth approximately three times better than its enantiomer 2.30b and possessed a GI₅₀ comparable to the clinically used agent etoposide in both cell lines. We have concluded that both the stereochemistry imparted by l-tryptophan and the presence of hydroxy substituents at the 4′ and 4″ positions are necessary to generate cytotoxic properties in the homodimer class. We are now employing 2.30a as a new lead compound in our efforts to discover improved indole-based cancer chemotherapeutics.

Full text of DOI:10.1016/j.bmcl.2008.05.022

Bioorganic & Medicinal Chemistry Letters 18 (2008) 3522–3525
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of diketopiperazine-based carboline homodimers and in vitro
growth inhibition of human carcinomas
a
b
b
Amy M. Deveau a, , Nancy E. Costa , Elizabeth M. Joshi , Timothy L. Macdonald
*
a Department of Chemistry and Physics, University of New England, Biddeford, ME 04005, USA
b Department of Chemistry, University of Virginia, McCormick Road, Charlottesville, VA 22904, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Starting from
D- or
L-tryptophan, we have synthesized and characterized six compounds 2.29–2.31a and
Received 18 March 2008
Received in revised form 3 May 2008
Accepted 6 May 2008
b that belong to a class of nitrogen heterocycles: the carboline-based homodimers. Each individual
homodimer features a 1,3-trans relationship on each side of the central diketopiperazine core, but differs
in absolute stereochemistry and also in substitution on the 40 and 400 oxygens (–Bn, –CH3, or –H). The
in vitro cytotoxicity of the six compounds was evaluated by measuring the growth inhibition in NCI–
H520 and PC-3 human carcinoma cells. Phenol 2.30a inhibited cancer cell growth approximately three
times better than its enantiomer 2.30b and possessed a GI50 comparable to the clinically used agent eto-
Available online 9 May 2008
Keywords:
Carboline
poside in both cell lines. We have concluded that both the stereochemistry imparted by L-tryptophan and
Diketopiperazine
Pictet–Spengler
Indole
Anti-cancer
Heterocycle
Tryptophan
Tryprostatin
the presence of hydroxy substituents at the 40 and 400 positions are necessary to generate cytotoxic prop-
erties in the homodimer class. We are now employing 2.30a as a new lead compound in our efforts to
discover improved indole-based cancer chemotherapeutics.
Ó 2008 Elsevier Ltd. All rights reserved.
Interest in the synthesis of novel indole-based heterocycles has
increased in recent years due to the prevalence of indoles in biolog-
ically-active natural products such as the fumitremorgin class1,2
and the significance of the indole moiety in clinical chemothera-
peutics like ellipticine.3,4 As Danishefsky,5 Corey,6 Boger,7 Cook,8
Joullié9 and others10 have independently demonstrated, indoles
can be easily incorporated into a complex target molecule by start-
ing from the amino acid tryptophan (Trp). While pursuing a pro-
gram to synthesize b-carboline derivatives from D- or L-Trp–OMe
using the Pictet–Spengler reaction,11 we serendipitously discov-
ered a new class of cytotoxic carboline analogs that are similar in
structure to ellipticine,3,12 azatoxin,13 gypsetin,5 the tryprosta-
tins,14 and other15 heterocycles. This group of bivalent, nitrogen-
rich heterocycles, which we identify as the carboline homodimer
class, contains several characteristic structural features: (1) seven
consecutive fused rings (2) a central diketopiperazine (DKP) core,
and (3) indole rings that cap each end (Fig. 1). Although each
homodimer retains a 1,3-trans relationship on each side of the cen-
tral DKP ring, the six homodimers also differ by their absolute con-
figuration at C7a/C15a and C14/C6, and by their substitution at the 40
and 400 phenol oxygen (Scheme 1).
We synthesized three enantiomeric pairs of the homodimers:
2.29a and b (R = –Bn); 2.30a and b (R = –H); or 2.31a and b
* Corresponding author. Tel.: +1 207 602 2813; fax: +1 207 602 5926.
Figure 1. Structures of carboline homodimers, etoposide, and structurally-related
E-mail address: adeveau@une.edu (A.M. Deveau).
nitrogen heterocycles.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.05.022

A. M. Deveau et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3522–3525
3523
Scheme 1. Synthetic approach to synthesize carboline homodimers 2.29–2.31a and b.
(R = –CH3) as indicated in Scheme 1.16 The key trans-1,3-disubsti-
tuted b-carboline synthons 2.5a, c, e, and g were accessed via a Pic-
tet–Spengler condensation in good yields,17 and were
subsequently saponified to their respective carboline-based car-
boxylic acids 2.6a, c, and d using LiOH in THF.11 Trans-carboline
acid 2.6a was then treated with N,O-dimethylhydroxylamine
hydrochloride, DCC, and TEA in CH2Cl2 with the goal of synthesiz-
ing the Weinreb amide.18 After this reaction was complete, the two
products observed on thin-layer chromatography (TLC) were iso-
lated using silica gel column chromatography in approximately a
1:1 ratio. Subsequent spectroscopic analysis revealed that the less
polar product (Rf = 0.69; 9:1:CHCl3/acetone) was in fact the DCC-
based carboline intermediate 2.27 (Scheme 2), while the more po-
lar spot (Rf = 0.58) was a 2,5-DKP-based dimer of our carboline core
2.29a. Although 2,5-DKPs are well-documented products of -ami-
a
no acid coupling,15a,16a,19 we were surprised and intrigued by our
result; no DKP products were isolated when the reagent PyBop
was used to couple 2.6a with various amino acids during previous
experiments.11 Overall, and as detailed in Scheme 2, we think that
2.29a resulted from attack of a second carboline molecule on the
activated DCC intermediate 2.27 and that the ring closure was
facilitated by populating accessible conformations of the two ter-
tiary amides present in 2.28.20 Furthermore, to hasten the rate of
our reaction and improve the homodimer yield we added N,N-
dimethylaminopyridine (DMAP), a well known acylation catalyst.
This simple change promoted complete conversion of starting
material, improved our yield of sister compound 2.29b to 78%
and shortened our reaction time to 5 h.
With three pairs of homodimer enantiomers in hand, we next
evaluated the compounds’ abilities to inhibit the growth of lung
(NCI–H520) and prostate (PC-3) human cancer cells as a measure
of cytotoxicity. Two of the six dimers screened exhibited double-
digit micromolar cytotoxicity (Tables 1 and 2). Phenol 2.30a pos-
sessed a GI50 of 21.5 and 21.9 lM in lung and prostate lines,
respectively, which was in the same range as the GI50 for etoposide
and ds2–Tps B (Fig. 1; Tables 1 and 2).14c Enantiomer 2.30b was
approximately three times less active in the same cell line (Table
2). Because phenol 2.30a contains the same trans relationship
and absolute stereochemistry as the azatoxin and etoposide cores,
we were not surprised by its antiproliferative activity.
Table 1
In vitro GI50 data for homodimers 2.30a and b against human carcinomas
Compound
NCI–H520a
PC-3a
Etoposide
Ds2–Tps B
2.30a
13.5
11.9
21.5 ( 1.1)
61.5 ( 1.6)
12.5
12.3
21.9 ( 3.1)
55.3 ( 1.0)
2.30b
a
GI50 is defined as the drug concentration required to inhibit 50% of cell growth.
Scheme 2. Proposed intermediates in the formation of the carboline homodimers.

3524
A. M. Deveau et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3522–3525
Table 2
In vitro growth inhibition studies with carboline homodimers 2.29–2.31a and b to measure cytotoxicity in human carcinoma lines
Compound
Percent cell survivala
NCI–H520
50
Ma
PC-3
Ma
10
l
Ma
l
100
l
Ma
10
l
Ma
50
l
100 l
Ma
Etoposide
2.29a
2.29b
2.30a
2.30b
2.31a
2.31b
78.6 ( 3.3)
88.1 ( 4.3)
>100
67.3 ( 2.2)
88.7 ( 4.4)
90.3 ( 5.2)
99.8 ( 1.3)
35.2 ( 3.4)
81.3 ( 3.2)
>100
16.5 ( 1.6)
53.2 ( 2.3)
90.8 ( 4.6)
98.9 ( 1.8)
21.3 ( 2.7)
81.0 ( 4.1)
84.5 ( 3.4)
5.8 ( 3.2)
26.8 ( 1.7)
86.7 ( 6.6)
98.9 ( 1.9)
73.8 ( 2.3)
98.2 ( 5.4)
>100
80.7 ( 2.4)
>100
41.2 ( 4.2)
68.8 ( 4.4)
92.7 ( 2.3)
3.3 ( 3.5)
43.6 ( 3.7)
80.5 ( 4.4)
>100
30.8 ( 4.2)
54.5 ( 0.6)
65.1 ( 3.7)
1.6 ( 4.8)
15.7 ( 4.9)
65.0 ( 2.5)
77.4 ( 1.9)
93.0 ( 3.2)
>100
Values are reported as the standard deviation of the mean.
a
CellTiter 96 Aqueous nonradioactive cell proliferation assay (Promega) was used to determine growth inhibition. Percent inhibition values were calculated versus control
wells and were completed in quadruplicate. Control wells contained 0.2% DMSO.
Upon further examination of the homodimer growth inhibition
data in Tables 1 and 2, it is also striking to note that only the phe-
nol-based enantiomers possessed growth inhibitory properties;
placing an ether at the 40 and 400 O generates homodimers devoid
of activity. Interestingly, our findings are in contrast to Ganesan
et al.’s research on phenol 2.11a and its methyl ether 2.11b; the
methyl ether possessed a notably lower GI50 than the correspond-
ing phenol (Fig. 1).2a Our lab has also independently synthesized
2.11a did not find that it possessed significant anti-proliferative
properties against human carcinoma lines.11 This suggests that
homodimer 2.30a (seven consecutive rings fused by a central
DKP) may function via a different mechanism than glycine-based
DKP ether 2.11b (4 fused rings and a terminal DKP). Although
mechanism was not explored here, we wonder if 2.30a could be
targeting topoisomerase II (like azatoxin and etoposide), tubulin,
or interacting with DNA via groove binding or intercalation. Inter-
calation, like that found in the ellipticines,21 is possible due to the
presence of the aromatic indole moiety, and the aromatic pendant
groups on C6 and C14, respectively. Molecular modeling of homodi-
mer 2.30a clearly shows the planarity of the terminal indole, but
also suggests that the molecule can adopt a cup-like conformation
that is controlled by the flexible DKP/carboline core (Fig. 2). What-
ever the discrete molecular target or mechanism, enantiospecifici-
ty with growth inhibition data in a series of related compounds can
be associated with a preferred binding conformation and/or site.
We were also particularly interested in the carboline homodi-
mers when considering the concept of bivalency in drug design.
Carlier et al. have recently shown that dimerization of a biologi-
cally inactive precursor to the natural product huperzine A pro-
duces
a
drug with twice its potency.22 From Carlier’s and
others23 research, it has become clear that bivalency is a useful
derivatization and diversification technique for a synthetic agent
or natural product in drug discovery. Remarkable enhancements
of dimer biological activity over the individual monomer units
have been observed, but it is important to note that the effective-
ness of this technique depends entirely upon the nature of the drug
target’s binding site(s). Although most of the current bivalency lit-
erature involves use of a tether to connect two identical halves, our
application of the concept to the aforementioned symmetrical
fused-ring homodimers is also relevant. To our knowledge, this
concept has not yet been formally extended to this class of cyto-
toxic indole-based homodimers.
In summary, we serendipitously discovered a new class of cyto-
toxic carboline analogs that are similar in structure to azatoxin and
the fumitremorgin class of natural products while pursuing a pro-
gram to synthesize carboline derivatives from
D- or L-Trp. This
group of bivalent, nitrogen-rich heterocycles, the carboline
homodimer class, contains seven consecutive fused rings and is
characterized by a central DKP core and two terminal indoles that
cap each end of the molecule. We synthesized carboline homodi-
mers 2.29a and b–2.31aand b (Fig. 1) and evaluated their growth
inhibitory properties in PC-3 and NCI–H520 cell lines as a measure
of cytotoxicity. The enantiomeric pair of phenols 2.30a and b pos-
sessed GI50’s in both lung and prostate cell lines in the micromolar
range. Interestingly enantiomer 2.30a was twice as cytotoxic as
2.30b and possessed a GI50 comparable to that of etoposide and
ds2–Tps B14c in the cell lines investigated (Fig. 1). Considering
these data, we conclude that the stereochemistry around the DKP
ring junction is coupled to growth inhibition. Additionally, our data
suggest that the presence of the 4-hydroxy-3,5-dimethoxybenzene
pendant group was essential to elicit an inhibitory response. Over-
all for the homodimer class, –H ꢀ –Me ꢁ –Bn at the 40 and 400-posi-
tions, respectively. Because of 2.30a’s structural similarity to
etoposide and axatoxin, further biological testing is intended to
determine whether 2.30a may target topoisomerase II, tubulin, or
interact with DNA. Overall, we remain interested in structurally di-
verse indole-based heterocycles like the carboline homodimers.
and continue to explore the relationship between structure, activ-
ity, and mechanism in the carboline homodimer series and related
compounds.
Acknowledgment
Dr. A.M. Deveau gratefully acknowledges the support of the
University of New England for a University Faculty Mini-Grant.
Figure 2. One representative low energy cup-like conformation of homodimer 2.-
30a achieved after completing an energy minimization (Chem3D, MM2 Force Field).

A. M. Deveau et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3522–3525
3525
14. (a) Jain, H. D.; Zhang, C.; Zhou, S.; Zhou, H.; Ma, J.; Liu, X.; Liao, X.; Deveau, A.
M.; Dieckhaus, C. M.; Johnson, M. A.; Smith, K. S.; Macdonald, T. L.; Kakeya, H.;
Osada, H.; Cook, J. M. Bioorg. Med. Chem. 2008, 16, 4626; (b) Zhao, S.; Gan, T.;
Yu, P.; Cook, J. M. Tetrahedron Lett. 1998, 39, 7009; (c) Zhao, S.; Smith, K. S.;
Deveau, A. M.; Dieckhaus, C. M.; Johnson, M. A.; Macdonald, T. L.; Cook, J. M. J.
Med. Chem. 2002, 45, 1559.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2008.05.022.
15. (a) Martins, M. B.; Carvalho, I. Tetrahedron 2007, 63, 9923; (b) Siwicka, A.;
Wojtasiewicz, K.; Rosiek, B.; Leniewski, A.; Maurin, J. K.; Czarnocki, Z.
Tetrahedron: Asymmetry 2005, 6, 975.
References and notes
16. (a) Dandapani, S.; Lan, P.; Beeler, A. B.; Beischel, S.; Abbas, A.; Roth, B. L.; Porco,
J. A.; Panek, J. S. J. Org. Chem. 2006, 71, 8934; (b) During the revision of this
manuscript we discovered that Panek et al. had synthesized DKP homodimer
2.31a (7% yield) as part of a program to generate DKP-based pipecolic acid
scaffolds via parallel synthesis for use against G-protein coupled receptor
targets. Although Panek’s precedent exists the broad interest in 2.31a and
related homodimers further justifies our interest in the these compounds.
Notably, our yield of 83% for the dimerization step is also an improvement over
Panek’s previous synthesis.
17. (a) Cox, E. D.; Li, J.; Hamaker, L. K.; Yu, P.; Cook, J. M. Chem. Comm. 1996, 21,
2477; (b) Cox, E. D.; Hamaker, L. K.; Li, J.; Yu, P.; Czerwinski, K. M.; Deng, L.;
Bennett, D. W.; Cook, J. M.; Watson, W. H.; Krawiec, M. J. Org. Chem. 1997, 62,
44; (c) Cox, E. D.; Cook, J. M. Chem. Rev. 1995, 95, 1797.
18. Boumendjel, A.; Nuzillard, J. M.; Massiot, G. Bull. Chem. Soc. Fr. 1990, 127, 645.
19. (a) Dinsmore, C. J.; Beshore, D. C. Tetrahedron 2002, 58, 3297; (b) Fischer, P. M. J.
Peptide Sci. 2003, 9, 9.
20. (a) Avalos, M.; Babiano, R.; Barneto, J. L.; Bravo, J. L.; Cintas, P.; Jimenez, J. L.;
Palacios, J. C. J. Org. Chem. 2001, 66, 7275; (b) Robin, M. B.; Bovey, F. A.; Basch,
H. Molecular and Electronic Structure of the Amide Group. In The Chemistry of
1. Cui, C. B.; Kakeya, H.; Okada, G.; Onose, R.; Ubukata, M.; Takahashi, I.; Isono, K.;
Osada, H. J. Antibiot. 1995, 48, 1382.
2. (a) Wang, H.; Usui, T.; Osada, H.; Ganesan, A. J. Med. Chem. 2000, 43, 1577; (b)
Wang, H.; Ganesan, A. J. Org. Chem. 2000, 65, 4685; (c) Wang, H.; Ganesan, A.
Tetrahedron Lett. 1997, 38, 4327.
3. Juret, P.; Le Talaer, Y.; Couette, J. E.; Delozier, T. Eur. J. Cancer (Oxford) 1980,
Suppl. 1, 277.
4. Crews, C. M.; Mohan, R. Cur. Opin. Chem. Biol. 2000, 4, 47.
5. Schkeryantz, J. M.; Woo, J. C. G.; Siliphaivanh, P.; Depew, K. M.; Danishefsky, S.
J. J. Am. Chem. Soc. 1999, 121, 11964.
6. Baran, P. S.; Guerrero, C. A.; Corey, E. J. J. Am. Chem. Soc. 2003, 125,
5628.
7. (a) Boger, D. L.; Fink, B. E.; Hedrick, M. P. Bioorg. Med. Chem. Lett. 2000, 10, 1019;
(b) Boger, D. L.; Fink, B. E.; Hedrick, M. P. J. Am. Chem. Soc. 2000, 122, 6382.
8. (a) Yu, J.; Wearing, X. Z.; Cook, J. M. Tetrahedron Lett. 2004, 45, 3937; (b) Yu, J.;
Wearing, X. Z.; Cook, J. M. Tetrahedron Lett. 2003, 44, 543.
9. Chen, W.; Joullié, M. M. Tetrahedron Lett. 1998, 39, 8401.
10. (a) Priestley, N. D.; Smith, T. M.; Shipley, P. R.; Floss, H. G. Bioorg. Med. Chem.
1996, 4, 1135; (b) González-Vera, J. A.; Garc´ıa-López, M. T.; Herranz, R.
Tetrahedron 2007, 63, 9229; (c) Ohba, M.; Natsutani, I.; Sakuma, T. Tetrahedron
2007, 63, 10337.
11. Deveau, A. M.; Labroli, M. A.; Dieckhaus, C. M.; Barthen, M. T.; Smith, K. S.;
Macdonald, T. L. Bioorg. Med. Chem. Lett. 2001, 11, 1251.
12. (a) Stiborová1, M.; Sejbal, J.; Boek-Dohalská, L.; Aimová, D.; Poljaková, J.;
Forsterová, K.; Rupertová, M.; Wiesner, J.; Hudeek, J.; Wiessler, M.; Frei, E.
Cancer Res. 2004, 64, 8374; (b) Ishikura, M.; Hino, A.; Yaginuma, T.; Agata, I.;
Katagiri, N. Tetrahedron 2000, 56, 193.
Amides; Jacob, Zabicky, Ed.; John Wiley
& Sons, Interscience Publishers:
London, 1970; p 139.
21. Reha, D.; Kabelac, M.; Ryjacek, F.; Sponer, J.; Sponer, J. E.; Elstner, M.; Suhai, S.;
Hobza, P. J. Am. Chem. Soc. 2002, 124, 3366.
22. Carlier, P. R.; Du, D. M.; Han, Y. F.; Liu, J.; Perola, E.; Williams, I. D.; Pang, Y. P.
Angew. Chem., Int. Ed. Eng. 2000, 39, 1775.
23. (a) Shuker, S. B.; Hajduk, P. J.; Meadows, R. P.; Fesik, S. W. Science 1996, 274,
1531; (b) Wender, P. A.; Koehler, M. F. T.; Wright, D. L.; Irie, K. Synthesis-
Stuttgart 1999, Sp. Iss. SI, 1401.; (c) Posner, G. H.; Ploypradith, P.; Parker, M. H.;
O’Dowd, H.; Woo, S. H.; Northrop, J.; Krasavin, M.; Dolan, P.; Kensler, T. W.; Xie,
S. J.; Shapiro, T. A. J. Med. Chem. 1999, 42, 4275.
13. (a) Leteurtre, F.; Madalengoitia, J.; Orr, A.; Guzi, T.; Lehnert, E.; Macdonald, T. L.;
Pommier, Y. Cancer Res. 1992, 52, 4478; (b) Cline, S. D.; Macdonald, T. L.;
Osheroff, N. Biochemistry 1997, 36, 13095.