Synthesis of a novel tetrahydroisoquinolino[2,1-c][1,4]benzodiazepine ring system with DNA recognition potential

Article abstract of DOI:10.1016/j.bmcl.2004.07.001

The first stereospecific synthesis of a novel tetrahydroisoquinolino[2,1-c] [1,4]benzodiazepine ring system with DNA recognition potential starting from (S)-1,2,3,4-tetrahydroisoquinoline carboxylic acid has been reported. We report the first stereospecif

Full text of DOI:10.1016/j.bmcl.2004.07.001

Bioorganic & Medicinal Chemistry Letters 14 (2004) 4371–4373
Synthesis of a novel
tetrahydroisoquinolino[2,1-c][1,4]benzodiazepine ring system with
DNA recognition potential
b
Kiran Kumar Kothakondaa,b, and D. Subhas Bose
*
aDepartment of Chemistry, University of Montreal, PO Box 6128, Montreal, QC, Canada H3C 3J7
bFine Chemical Laboratory, Indian Institute of Chemical Technology, Hyderabad 500 007, India
Received 29 April 2004;revised 1 July 2004;accepted 1 July 2004
Abstract—We report the first stereospecific synthesis and reactivity of a novel tetrahydroisoquinolino[2,1-c][1,4]benzodiazepine ring
system with DNA recognition potential.
Ó 2004 Elsevier Ltd. All rights reserved.
The pyrrolo[2,1-c][1,4]benzodiazepine antitumor antibi-
otics1 are a well known class of sequence-selective
DNA binding agents derived from various streptomyces
species, collectively known as the anthramycins and rep-
resented by anthramycin, tomamycin, chicamycin, neo-
mycin, and DC-81 (1). PBDs bind to the guanine in
the minor groove of DNA in a sequence-selective cova-
lent binding manner, via an electrophilic imine or carb-
inolamine functionality at N10-C11. Their structural
activity relationships proceeding based upon CPK mod-
els have been proposed,1–3 fluorescence, high field
NMR, molecular modeling, and DNA foot printing
studies4–6 have established that the PBDs recognize with
a preference of 50-Pu–G–Pu sequence.7–9,16 The PBDs
differ in the number, type, and position of substituents
in both the aromatic A-ring,10 and also pyrrole C11 ring.
The interest in discovering and developing small mole-
cules, capable of binding to DNA in a sequence-selective
manner is growing in the area of molecular recognition.
In this context variety of PBD analogues such as linked
symmetrical, unsymmetrical dimmers, and also more re-
cently PBD-hybrids12 has been synthesized mainly with
modifications and substitutions in the aromatic A-ring.
The investigation of their anticancer activities also
necessitates the importance of the involvement to find
the modified PBD analogues with better profiles.
The C-ring varies in both the type and nature of unsat-
uration, which is either fully saturated, unsaturated at
either C2–C3 (endocyclic) or at C2 (exocyclic) position,13
SAR studies suggest that this variation influences the
activity. It is apparent from the overall twist of anthra-
mycin (45°) and tomaymycin (9°) that the degree of sat-
uration can greatly affect the shape of the molecule,
which is essential for the activity. However, all these
studies were totally on monocyclic C-ring systems. After
the keen observation and understanding the SAR stud-
ies of PBDs, we initiated a program to discover and de-
velop the DNA interactive moieties with better
recognition potential. With all these information, our ef-
forts have been directed toward the synthesis of modi-
fied novel systems, such as the replacement of pyrrole
C-ring of (1) by the corresponding 1,2,3,4-tetrahydroiso-
quinoline ring to produce the novel tetrahydroisoquino-
lino[2,1-c][1,4]benzodiazepine (2), with potential binding
affinity and sequence selectivity (Fig. 1). Not only this is
bicyclic C-ring system, this C-ring modified novel target
2 consists both the saturated and unsaturated rings,
which takes care about the shape of the molecule and
OH
H
OH
H
H
N
H
N
HO
HO
N
N
H3CO
H3CO
O
O
Keywords: Molecular recognition;DNA binding;Sequence-selective;
PBDs;Tetrahydroisoquinolino[2,1- c][1,4]benzodiazepine.
1 PBD
2
*
Figure 1. Structures of pyrrolo[2,1-c][1,4]benzodiazepine and tetra-
hydroisoquinolino[2,1-c][1,4]benzodiazepine (TBD).
Corresponding author. Tel.: +1-514-343-6111x3981;fax: +1-514-343-
6111;e-mail: kcubem@rediffmail.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.07.001

4372
K. K. Kothakonda, D. S. Bose / Bioorg. Med. Chem. Lett. 14 (2004) 4371–4373
also contains (S)-configuration at the chiral C11a posi-
tion, which provides the molecules with the right-
handed twist necessary for a snug fit within the minor
groove of DNA like all other biologically active PBDs.
compound was the downfield shift of the H11a proton,19
which was from at d 4.99, compared to H11a at d 3.92
for the DC-81. Debenzylation was effected by treatment
with 10%Pd–C in 1,4-cyclohaxadiene to afford 2 in 70%
yield.
Inspection of the target molecule (2) reveals a number of
challenges for synthesis that include (a) incorporation of
the bicyclic tetrahydroisoquinoline ring with appropri-
ate functionality, which contains both saturated and
unsaturated rings, (b) preservation of stereochemical
integrity at C11a position, (c) formation of a carbinoal-
mine or imine moiety in a mild and nonracemizing envi-
ronment. The synthetic approach described below meets
all of these criteria with good yields and also correlated
with the alternate approach.
In an alternative approach, the intermediate 8 was pre-
pared from 4 in large quantities via intermediates 11,
12, and 13 (Scheme 2). Compound 8 was exactly match-
ing with all the spectral data of the same compound
from earlier route. Ultimately, crucial intermediate 8
was converted to obtain the desired system 2, following
the reduction of nitro group, deprotection of thioacetal
resulting in concomitant ring closure and debenzylation.
In conclusion, the first stereospecific synthesis of a novel
tetrahydroisoquinolino[2,1-c][1,4]benzodiazepine ring
system has been prepared and also this opened new door
for the synthesis of a series of PBD analogues that is
both C7 and C8 linked dimers and some A-ring modified
PBDs for evaluation as potential DNA-binding ligands
and cytotoxic agents. The DNA-binding abilities and
anticancer activities of 2 under investigation, the de-
tailed mechanistic and molecular modeling studies of 2
along with other analogues of this series will be reported
elsewhere.
The synthesis was started with commercially available
(S)-1,2,3,4-tetrahydroisoquinoline-3-carboxylic acid (3),
which was protected as N-Cbz and methyl ester to give
14
4;compound 4 was reduced in presence of LiBH4 to
obtain 5. Oxidation of primary alcohol in 5 to the corre-
sponding aldehyde was problematic thus, a variety of
oxidizing protocols, such as PCC, Swern etc. were unsat-
isfactory in terms of yields and isolation. However, oxi-
dation proceeded smoothly in 76% yield upon treatment
with IBX in DMSO15,16 at room temperature to give the
aldehyde. Having obtained the desired aldehyde, the
next concern was careful protection and deprotection
of the aldehyde in a mild and nonracemizing environ-
ment. Moreover, the aldehyde was found to be rather
unstable and was therefore immediately converted into
compound 6 by treating with EtSH in presence of
BF3ÆEt2O (Scheme 1). Free amine 7 was generated in
83% yield from 6 using TMSI. Compound 7 was treated
with the 4-bezyloxy-5-methoxy-2-nitro benzoyl chlo-
ride,17,18 which was obtained from the corresponding
acid, to afford the key intermediate amide 8. Reduction
of the nitro group in 8 with SnCl2Æ2H2O in methanol un-
der reflux condition gave 9. Finally, deprotection of the
thioacetal group of 9 was performed with HgCl2/CaCO3
resulting in concomitant ring closure to give the carbi-
nolamine 10 in 79% yield. A significant feature of this
COOMe
BnO
MeO
NO2
N
COOMe
COOMe
a
c
b
NCbz
NH
4
11
O
12
CH(SEt)2
CH2OH
BnO
MeO
NO2
BnO
MeO
NO2
d
2
N
N
O
O
8
13
Scheme 2. Reagents and conditions: (a) TMSI, CH3CN, 4h, 75%;(b)
4-benzyloxy-5-methoxy-2-nitrobenzoyl chloride,17 Et3N, THF, 4h,
60%;(c) LiBH 4, THF/MeOH, 6h, 88%;(d) (i) IBX, DMSO, 6h,
81%;(ii) EtSH, BF 3ÆEt2O, CH2Cl2, 18h, 89%.
COOMe
CH(SEt)2
NCbz
COOH
CH2OH
a
b
c
NCbz
NH
NCbz
4
3
6
5
CH(SEt)2
CH(SEt)2
BnO
MeO
NH2
N
BnO
MeO
NO2
CH(SEt)2
e
f
d
N
NH
7
O
O
9
8
OH
H
H
N
N
RO
H3CO
RO
H
N
g
N
H3CO
O
O
10: R = Bn
2: R = H
h
Scheme 1. Reagents and conditions: (a) (i) BnOCOCl, NaOH (aq), 36h, 94%;(ii) SOCl 2, MeOH, reflux, 6h, 96%;(b) LiBH 4, THF/MeOH, 6h, 70%;
(c) (i) IBX, DMSO, 6h, 76%;(ii) EtSH, BF 3ÆEt2O, CH2Cl2, 18h, 66%;(d) TMSI, CH 3CN, 4h, 83%;(e) 4-benzyloxy-5-methoxy-2-nitrobenzoyl
chloride,17 Et3N, THF, 4h, 60%;(f) SnCl 2Æ2H2O/MeOH, reflux, 2.5h, 76%;(g) HgCl 2, CaCO3, CH3CN/H2O, 2.5h, 79%;(h) 10%Pd–C, 1,4-
cyclohexadiene, EtOH, 6h, 70%.

K. K. Kothakonda, D. S. Bose / Bioorg. Med. Chem. Lett. 14 (2004) 4371–4373
4373
12. Kamal, A.;Ramesh, G.;Srinivas, O.;Ramulu, P. Bioorg.
Med. Chem. Lett. 2004, 14, 471.
References and notes
13. Thurston, D. E.;Bose, D. S. Chem. Rev. 1994, 94, 433.
14. Peterson, M. L.;Vince, R. J. Med. Chem. 1991, 34, 2787.
15. Frigerio, M.;Santagostino, M. Tetrahedron Lett. 1994, 35,
8019.
16. Kiran Kumar, K.;Datta, A. Tetrahedron 1999, 55, 13899.
17. Kamal, A.;Laxman, N.;Ramesh, G.;Neelima, K.;
Kondapi, A. K. Chem. Commun. 2001, 15, 225.
18. Thurston, D. E.;Murty, V. S.;Langley, D. K.;Jones, G.
B. Synthesis, 1990, 81.
1. Thurston, D. E. Advances in the Study of Pyrrolo[2,
1-c][1,4]benzodiazepine (PBD) Antitumor Antibiotics In
Molecular Aspects of Anticancer Drug–DNA Interactions;
Neidle, S., Waring, M. J., Eds.;The Macmillan: London,
1993;Vol. 1, p 54.
2. Thurston, D. E.;Hurley, L. H. Drugs Future 1983, 8,
957.
3. Hurley, L. H.;Thurston, D. E. Pharm. Res. 1984, 2,
52.
19. All new compounds gave satisfactory spectroscopic and
analytical data. Data for 10: 1H NMR (200MHz, CDCl3)
d 7.53–7.09 (m, 9H), 6.86 (s, 1H), 6.39 (s, 1H), 5.19 (s, 2H),
5.17 (br s, 1H), 5.11–4.85 (m, 1H), 4.65–4.49 (m, 1H),
4.40–4.19 (m, 2H), 3.89 (s, 3H), 3.26–2.94 (m, 2H);
FABMS (imine): 399 [M+];[ a]D +16.5 (c 0.4, CHCl3).
4. Boyd, F. L.;Cheatham, S. F.;Remers, W.;Hill, G. C.;
Hurley, L. H. J. Am. Chem. Soc. 1990, 112, 3279.
5. Farmer, J. D.;Gustafson, G. R., Jr.;Conti, A.;Zimmt, M.
B.;Suggs, J. W. Nucleic Acids Res. 1991, 19, 899.
6. Wang, J. J.;Hill, G. C.;Hurley, L. H. J. Med. Chem. 1992,
35, 2995.
7. Hurley, L. H.;Reck, T.;Thurston, D. E.;Langley, D. R.;
Holden, K. G.;Hertzberg, R. R.;Hoover, J. R. E.;
Gallagher, G., Jr.;Faucette, L. F.;Mong, S.-M.;Johnson,
R. K. Chem. Res. Toxicol. 1998, 1, 258.
8. Hertzberg, R. P.;Hecht, S. M.;Reynolds, V. L.;
Molineuax, I. J.;Hurley, L. H. Biochemistry 1986, 25, 1249.
9. Warpehoski, M. A.;Hurley, L. H. Chem. Res. Toxicol.
1998, 1, 135.
10. Masterson, L. A.;Crocker, S. J.;Jenkins, T. C.;Howard,
P. W.;Thurston, D. E. Bioorg. Med. Chem. Lett. 2004, 14,
901.
1
Data for 9: H NMR (200MHz, CDCl3) d 7.43–6.87 (m,
9H), 6.81 (s, 1H), 6.11 (s, 1H), 5.11 (s, 2H), 4.92 (m, 1H),
4.72–4.47 (m, 2H), 4.39–4.13 (m, 2H, exchangeable with
D2O), 4.07–3.86 (m, 1H), 3.78 (s, 3H), 3.46–3.14 (m, 2H),
2.78–2.44 (m, 4H), 1.21 (t, 6H, J=6.9Hz); 13C NMR
(200MHz, CDCl3) d 150.7, 136.4, 129.2, 128.4, 127.7,
126.9, 126.3, 113.4, 102.5, 70.6, 70.5, 56.6, 52.6, 29.5, 29.4,
14.2, 14.1;FABMS calcd for C 29H34N2O3S2 (522.2) found
522.7 [M+];[ a]D ꢀ9.0 (c 0.5, CHCl3). Data for 4: 1H NMR
(200MHz, CDCl3) d 7.42–7.01 (m, 9H), 5.26–5.06 (m,
3H), 4.82–4.49 (m, 2H), 3.62–3.57 (2s, 3H), 3.24–3.16 (m,
2H);MS (EI) m/z (relative intensity): 325 ([M+], 100);[ a]D
+8.05 (c 1, CHCl3).
11. Bose, D. S.;Srinivas, P.;Gurjar, M. K. Tetrahedron Lett.
1997, 38, 5839, and references cited therein.