Synthesis and biological evaluation of an N10-Psec substituted pyrrolo[2,1-c][1,4]benzodiazepine prodrug

Article abstract of DOI:10.1016/S0960-894X(02)00150-6

The first example of an N10-protected (e.g., Psec, 15) pyrrolo[2,1-c][1,4]benzodiazepine (PBD) analogue that retains significant cytotoxicity in a number of tumour cell lines is reported. This prototype could lead to a new generation of clinically useful N10-protected PBD prodrugs.

Full text of DOI:10.1016/S0960-894X(02)00150-6

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1413–1416
Synthesis and Biological Evaluation of an N10-Psec Substituted
Pyrrolo[2,1-c][1,4]benzodiazepine Prodrug
Jane M. Berry,a Philip W. Howard,a,* Lloyd R. Kellandb and David E. Thurstona,*
aCRUK Gene Targeted Drug Design Research Group, Cancer Research Laboratories, School of Pharmaceutical Sciences,
University of Nottingham, University Park, Nottingham NG7 2RD, UK
bCRUK Centre for Cancer Therapeutics, Institute for Cancer Research, Clifton Avenue, Sutton, Surrey SM2 5PX, UK
Received 7 December 2001; revised 25 February 2002; accepted 28 February 2002
Abstract—The first example of an N10-protected (e.g., Psec, 15) pyrrolo[2,1-c][1,4]benzodiazepine (PBD) analogue that retains
significant cytotoxicity in a number of tumour cell lines is reported. This prototype could lead to a new generation of clinically
useful N10-protected PBD prodrugs. # 2002 Published by Elsevier Science Ltd.
The pyrrolo[2,1-c][1,4]benzodiazepine antitumour agents
are a family of biosynthetically derived tricyclic mol-
ecules produced by various Streptomyces species and
include such members as DC-81 and anthramycin.1
They bind selectively in the minor groove of DNA via a
covalent aminal bond between the electrophilic C11-
position of the PBD and the nucleophilic C2-amino
group of a guanine base. This adductformaiton is
thought to lead to the observed biological activity.1 The
(S)-configuration at the chiral C11a-position provides
the molecules with the necessary right-hand twist to fit
snugly within the minor groove of DNA, spanning three
base-pairs with a preference for 50-Pu-G-Pu sequences.1
are devoid of cytotoxic activity.2 However, they can act
as prodrugs in that after enzymatic removal of the N10-
protecting group, the free N10–C11 imine, carbinola-
mine or methyl ether forms (e.g., 2) may interact with
DNA and exert significant cytotoxicity against a range
of cancer cell types.2 This conceptwas exploited recently
by Thurston and co-workers who developed N10-pro-
tected PBD prodrugs suitable for use in ADEPT- and
GDEPT-type therapies.2 These contained N10-[4-
(nitrobenzyl)carbamate] protecting groups which could
be enzymatically removed in the presence of nitro-
reductase and co-factor NADH to produce the parent
cytotoxic PBD molecules.
It has been shown previously that N10-protected PBDs
(e.g., 1, Scheme 1) are unable to interact with DNA and
Encouraged by the success of this general approach, we
have now investigated the use of N10-protecting groups
Scheme 1. Activation of N10-protected PBDs (R=H or CH3).
*Corresponding authors at present address: The School of Pharmacy, University of London, 29/39 Brunswick Square, London WC1 1AX, UK.
E-mail: david.thurston@ulsop.ac.uk (D. E. Thurston); philip.howard@ulsop.ac.uk (P. W. Howard).
0960-894X/02/$ - see front matter # 2002 Published by Elsevier Science Ltd.
PII: S0960-894X(02)00150-6

1414
J. M. Berry et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1413–1416
that are removable through alternative mechanisms. In
particular, the 2-phenylsulphonylethyloxy (Psec) group
has been used by Nicolaou and co-workers3 as a trig-
gering device for enediyne antitumour agents. Protected
enediyne prodrugs can be converted into cytotoxic
forms by in situ fragmentation of the Psec group
through abstraction of the acidic protons at the a-posi-
tion to the sulphonyl group. Enediynes fitted with the
Psec triggering device display sub-nanomolar activity
againstpromeocyitc and T cell leukemia ut mour lines.
Similarly, Satyam and co-workers4 have designed a
phosphorodiamidate mustard analogue containing a
2-sulphonylethyloxycarbonyl linker which was found to
be selectively cytotoxic in cells expressing glutathione
transferase (GST). Molecular modelling studies sug-
gested that a basic tyrosine phenoxide ion located in the
active site of GST may aid deprotonation of the sul-
phonyl group, thus triggering fragmentation of the sul-
phonylethyloxycarbonyl linker to liberate the free
mustard. Based on this approach, we designed the N10-
Psec-protected PBD (15) along with the 2-phenyl-
thioethyloxycarbonyl (Ptec)-protected control (14)
which lacks the acidic sulphonyl protons, to test the
potential of this triggering device in the PBD system.
The first stage of the synthesis was to produce C-ring
acetal 7, which was synthesised5 in four steps from
commercially available N-carbobenzyloxy-l-proline
methyl ester 3 in 67% overall yield (Scheme 2). This
involved reduction of the methyl ester to secondary
alcohol 4 using lithium tetrahydridoborate in THF, fol-
lowed by oxidation to aldehyde 5 using a 3-fold excess
.
of SO3 pyridine and triethylamine in DMSO/CH2Cl2
(4:5) at ꢀ10 ꢁC. Acetal 6 was obtained by heating 5 with
thionyl chloride and trimethyl orthoformate in metha-
nol, followed by removal of the Cbz carbamate by
hydrogenolysis to give 7. This was joined to 4,5-dime-
thoxy-2-nitrobenzoic acid 8 by a standard coupling pro-
cedure using an equimolar ratio of 2-(1H-benzotriazole-
1-yl)-1,1,3,3-tetramethyluronium
tetrafluoroborate
(TBTU) and diisopropylethylamine (DIPEA) in DMF
to give 9 in 51% yield. Subsequent reduction of the
nitro functionality to give aniline 10 was achieved using
standard hydrogenolysis conditions.
ꢁ
ꢁ
ꢁ
.
Scheme 2. (a) LiBH4, THF, 0 C; (b) SO3 pyridine, Et3N, DMSO/CH2Cl2, ꢀ10 C; (c) SOCl2, HC(OCH3)3, MeOH, 60 C; (d) (i) Raney nickel,
EtOH or (ii) H2, 10% Pd/C, EtOH; (e) TBTU, DIPEA, DMF; (f) H2, 10% Pd/C, EtOH; (g) 11a/b, triphosgene, pyridine, CH2Cl2, 0 ꢁC; (h) 12a,
pyridine, CH2Cl2, 0 ꢁC; (j) (CH3CN)2PdCl2, Me2CO; (k) 12b, pyridine, CH2Cl2, 0 ꢁC; (l) 14, m-CPBA, CH2Cl2, 0 ꢁC; (m) 15, DBU, C6H6, 5 ꢁC.

J. M. Berry et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1413–1416
1415
At this stage, an attempt was made to protect the amine
as either the sulphone (13a) or sulphide (13b) carba-
mate, using the corresponding chloroformates (12a,b)
which had been freshly prepared from 2-(phenylsulph-
onyl)ethanol (11a) or 2-(phenylthio)ethanol (11b) and
triphosgene. However, on attempting to prepare the
Ptec-protected PBD (13b) from 10 using this strategy,
spontaneous cyclisation to the PBD sulphide 146 occur-
red during the introduction of the Ptec group, and so
13b could not be isolated. It is likely that HCl liberated
during the protection reaction may have caused pre-
mature hydrolysis of the acetal leading to ring closure.
Interestingly, this was not the case for the related Psec
intermediate (13a), which could be isolated and then
required acetal deprotection by treatment with trans-
bis(acetonitrile)palladium(II) chloride7 in acetone to
give 15.8 Sulphone 15 could also be obtained in 87%
yield by oxidation of sulphide 14 with 3-chloroperoxy-
benzoic acid (m-CPBA)9 in CH2Cl2. The structure of 15
was confirmed by deprotection in the presence of 1,8-
diazabicyclo[5.4.0]undec-7-ene (DBU)9 in benzene to
give PBD 1610 in 77% yield.
some cell lines, butwas approximaetly 10-fold less
active in A2780 and CH1. Sulphone 15 was also sig-
nificantly active at the sub-micromolar level in a number
of cell lines in the NCI’s 60-panel screen (Table 2).
In conclusion, this study has established that the N10-
sulphone-protected PBD 15 possesses significantcyot-
toxicity in a number of cell lines, comparable or better
in some cases (e.g., A2780cisR, CH1cis R and SKOV-3)
to the parent unprotected PBD 16. This represents the
first example of inherent cytotoxicity in an N10-pro-
tected PBD. Although the precise mechanism of
removal of the N10-Psec protecting group is presently
unproven, the role of glutathione transferase is sup-
ported by the inactivity of the N10-Ptec PBD control
14. Possible explanations for the greater cytotoxicity of
15 compared to 16 in SKOV-3 include favourable cel-
lular penetration properties of 15 in this cell line and a
higher cellular concentration of GST. Given the chemi-
cal interconvertibility of the N10-C11 imine, carbinol-
amine and carbinolamine methyl ether forms of a PBD,
it is possible that the N10-Psec-protecting group may be
useful in producing a single non-interconvertible PBD
species for clinical use. Furthermore, compared to the
PBD 16, the similar activity of 15 in both parent and
cisplatin-resistant cell lines suggests that N10-Psec pro-
tected PBDs may have potential use in drug resistant
disease.
The sulphide (14), sulphone (15) and parentPBD ( 16)
were evaluated for cytotoxicity in five ovarian cell lines,
including the matched pairs A2780/A2780cisR and
CH1/CH1cisR, and also SKOV-3 (Table 1). As antici-
pated, the Ptec-protected PBD (14) was essentially
inactive in this panel (i.e., IC50=>25 mM), as well as in
the NCI’s 60-cell line panel (data not shown). However,
the N10-Psec prodrug 15 possessed significantsub-
micromolar cytotoxicity in all cell lines examined. It had
similar (e.g., A2780cisR and CH1cisR) or better (e.g.,
SKOV-3) activity compared to the parent PBD 16 in
Acknowledgements
Cancer Research UK (CRUK) is thanked for providing
financial supportfor htis work (SP1938/0301 and
SP1938/0401 to D.E.T.). Dr. Robert Schultz (NCI) is
thanked for providing the 60-panel cell line data.
Table 1. Comparative cytotoxicity of 14, 15 and 16 in five ovarian
cell lines
Cell lines
Sulphide 14a
Sulphone 15a
PBD 16a
References and Notes
A2780
>25
>25
NA
>25
>25
NA
0.48
0.49
1
0.4
0.47
1.2
0.064
0.155
2.4
0.082
0.11
1.3
A2780cisRb
RFc
CH1
1. Thurston, D. E. Advances in the Study of Pyrrolo[2,1-
c][1,4]benzodiazepine (PBD) Antitumour Antibiotics. In
Molecular Aspects of Anticancer Drug-DNA Interactions;
Neidle, S., Waring, M. J., Eds.; Macmillan: London, 1993;
Vol. 1, p 54.
2. Sagnou, M. J.; Howard, P. W.; Gregson, S. J.; Eno-
Amouquage, E.; Burke, P. J.; Thurston, D. E. Bioorg. Med.
Chem. Lett. 2000, 10, 2083.
CH1cisRb
RFc
SKOV-3
>25
0.56
1.7
aIC50 (mM), concentration required to inhibit cell growth by 50% on
continuous exposure for 96 h.
3. Nicolaou, K. C.; Dai, W.-M.; Tsay, S.-C.; Estevez, V. A.;
Wrasidlo, W. Science 1992, 256, 1172.
bcisR denotes resistance to cisplatin.
cResistance factor=cytotoxicity of cisR/normal cell line.
4. Satyam, A.; Hocker, M. D.; Kane-Maguire, K. A.; Mor-
gan, A. S.; Villar, H. O.; Lyttle, M. H. J. Med. Chem. 1996,
39, 1736.
Table 2. In vitro cytotoxicity of 15 in a selection of NCI cell lines
5. Mori, S.; Ohno, T.; Harada, H.; Aoyama, T.; Shioiri, T.
Tetrahedron 1991, 47, 5051.
Cell lines
GI50 (mM)a
LC50 (mM)b
Lung (NCI-H552)
Colon (Colo 205)
CNS (SNB-75)
Melanoma (SK-MEL-5)
Renal (RXF 393)
Breast(MDA-MB-435)
0.11
0.19
0.72
0.29
0.24
0.45
1.38
0.71
9.00
6.78
2.97
4.33
6. Data for 14: 1H NMR (270 MHz, CDCl3) where R=OH: d
7.49–7.19 (m, 6H), 6.76 (s, 1H), 5.63 (d, 1H, J=9.52 Hz),
4.38–4.33 (m, 1H), 3.93 (s, 3H), 3.89 (s, 3H), 3.84–3.40 (m,
4H), 3.20–2.80 (m, 2H), 2.20–1.80 (m, 4H); 1H NMR
(400 MHz, CDCl3) where R=OCH3: d 7.86–7.50 (m, 5H),
7.20 (s, 1H), 7.02 (s, 1H), 5.43 (d, 1H, J=9.15 Hz), 4.66–4.63
(m, 1H), 3.96 (s, 3H), 3.94 (s, 3H), 3.93 (s, 3H), 3.60–3.20 (m,
8H), 2.20–1.90 (m, 2H); 13C NMR (67.8 MHz, CDCl3) where
R=OH: d 167.0, 156.0, 150.8, 148.4, 134.9, 129.9, 129.7,
129.1, 128.3, 126.7, 125.7, 112.6, 110.4, 86.1, 63.9, 59.9, 56.2,
aGI50, concentration required to restrict cell growth to 50% on con-
tinuous exposure for 48 h.
bLC50, concentration required to decrease cell population by 50% on
continuous exposure for 48 h.

1416
J. M. Berry et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1413–1416
46.4, 32.7, 28.7, 23.1, 14.1; 13C NMR (100.5 MHz, CDCl3)
where R=OCH3: d 167.1, 155.5, 151.0, 148.7, 138.6, 134.1,
129.5, 128.9, 128.3, 127.9, 126.2, 113.4, 110.2, 93.3, 60.0, 58.6,
56.3, 54.7, 46.2, 28.9, 23.1, 14.2; MS (EI), m/z (relative inten-
sity) 472 ([M+1]+, R=OCH3, 8), 458 ([M+1]+, R=OH, 27),
336 (6), 304 (15), 292 (9), 278 (15), 275 (33), 260 (30), 250 (11),
245 (6), 223 (6), 206 (100), 192 (8), 180 (6), 164 (12), 154 (24),
150 (11), 137 (93), 123 (28), 109 (32), 70 (24).
167.4, 155.6, 151.2, 148.8, 138.6, 134.1, 129.5, 128.3, 128.1,
126.2, 113.3, 110.2, 93.3, 60.2, 58.6, 56.4, 54.7, 46.3, 28.9, 23.2,
21.1, 14.2; MS (EI), m/z (relative intensity) 504 ([M]+, 36), 435
(64), 405 (17), 292 (8), 260 (100), 256 (7), 245 (21), 231 (14),
229 (14), 222 (18), 214 (9), 206 (20), 191 (24), 180 (8), 168 (20),
164 (29), 136 (28), 125 (64), 93 (12), 77 (53), 70 (16).
9. Nicolaou, K. C.; Maligres, P.; Suzuki, T.; Wendeborn,
S. V.; Dai, W.-M.; Chadha, R. K. J. Am. Chem. Soc. 1992,
114, 8890.
10. Data for 16: [a]2D3 +1481ꢁ (c1, CHCl3); 1H NMR
(250 MHz, CDCl3) d 7.70 (d, 1H, J=8.8 Hz), 7.43 (s, 1H), 6.85
(s, 1H), 3.96 (s, 3H), 3.93 (s, 3H), 3.90–3.48 (m, 3H), 2.40–2.29
(m, 2H), 2.12–2.05 (m, 2H); 13C NMR (62.9 MHz, CDCl3) d
165, 163, 152, 148, 141, 121, 112, 110, 57, 55, 47, 30, 25.
7. Lipshutz, B. H.; Pollart, D.; Monforte, J.; Kotsuki, H.
Tetrahedron Lett. 1985, 26, 705.
1
8. Data for 15: H NMR (270 MHz, CDCl3) d 7.86–7.50 (m,
5H), 7.22 (s, 1H), 7.03 (s, 1H), 5.44 (d, 1H, J=9.34 Hz), 4.69–
4.65 (m, 1H), 4.11–3.94 (m, 9H), 3.80–3.40 (m, 5H), 3.22–3.16
(m, 1H), 2.05–1.78 (m, 4H); 13C NMR (67.8 MHz, CDCl3) d