Cytotoxic alpha-bromoacrylic derivatives of low molecular weight.

Article abstract of DOI:10.1016/S0960-894X(02)00177-4

In vitro and in vivo activities of a small series of alpha-bromoacrylic derivatives of low molecular weight (MW) are described and compared with those of alpha-bromoacrylic derivatives of distamycin-like frames. Low MW compounds, when lacking of a strong basic moiety, are potent cytotoxics, while analogues bearing a strong basic moiety are not. This suggests the existence of an active transport mechanism for distamycin-derived cytotoxics characterized by strong basic amidino or guanidino moieties. Low MW compounds are inactive in vivo, possibly because of the metabolic lability of alpha-bromoacrylic moiety. The same moiety is however present in a series of potent anticancer distamycin-like minor groove binders, for example, PNU-166196 (brostallicin), a fact that underlines the features of the latter.

Full text of DOI:10.1016/S0960-894X(02)00177-4

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1467–1471
Cytotoxic Alpha-Bromoacrylic Derivatives of Low
Molecular Weight
Italo Beria,a,* Marina Caldarelli,a Cristina Geroni,b Nicola Mongelli,a
Benedetta Reinach,c Luisella Vignatic and Paolo Cozzia
aChemistry Department, Pharmacia Discovery Research Oncology, Viale Pasteur 10, 20014 Nerviano, Milan, Italy
bExternal Research Department, Pharmacia Discovery Research Oncology, Viale Pasteur 10, 20014 Nerviano, Milan, Italy
cGlobal Drug Metabolism Department, Pharmacia Discovery Research Oncology, Viale Pasteur 10, 20014 Nerviano, Milan, Italy
Received 14 January 2002; revised 4 March 2002; accepted 20 March 2002
Abstract—In vitro and in vivo activities of a small series of a-bromoacrylic derivatives of low molecular weight (MW) are described
and compared with those of a-bromoacrylic derivatives of distamycin-like frames. Low MW compounds, when lacking of a strong
basic moiety, are potent cytotoxics, while analogues bearing a strong basic moiety are not. This suggests the existence of an active
transport mechanism for distamycin-derived cytotoxics characterized by strong basic amidino or guanidino moieties. Low MW
compounds are inactive in vivo, possibly because of the metabolic lability of a-bromoacrylic moiety. The same moiety is however
present in a series of potent anticancer distamycin-like minor groove binders, for example, PNU-166196 (brostallicin), a fact that
underlines the features of the latter. # 2002 Elsevier Science Ltd. All rights reserved.
Introduction
Table 1. In vitro and in vivo activity of a-halogenoacrylamido
distamycin derivatives 1–8 against L1210 murine leukemia
We have recently reported a series of potent cytotoxic
antitumor agents showing an a halogenoacrylamido
moiety linked to a distamycin-like frame, some of which
modified also at the amidine terminus typical of dis-
tamycin A. The SAR of these minor groove binders was
discussed, particularly as far as the role of the length of
the polypyrrolic frame and the a halogenoacrylamido
moiety are concerned.1,2
Compd
X
n
In vitroa IC50, nM
In vivob
OD mg/kg
T/C%
1
2
3
4
5
6
7
8
Br
Br
Br
Br
Cl
Cl
F
4
3
2
1
4
3
4
4
6.3 (Æ1.3)
79.6 (Æ22.4)
1300
1.56
12.5c
nd
200
175
nd
While both a-bromo and a-chloro derivatives of the
same four or three pyrrole units frame are substantially
equipotent in vitro, the three pyrrole unit derivatives,
both bromo and chloro, are about one order of magni-
tude less cytotoxic than the corresponding four pyrrole
congeners.
>2500
nd
nd
3.8 (Æ1.4)
96.8 (Æ24.2)
>800
1.56
12.5
nd
133
117
nd
H
2000
nd
nd
IC50=50% inhibitory concentration as the meanÆSE from dose–
response curves of at least two experiments.
The same decrease of activity occurs with a-bromo
derivatives with two and one pyrrole units, which are
therefore devoid of significant activity (Table 1).
aDrug sensitivity was determined after 48 h of continuous exposure
against L1210 cells.
bFor in vivo studies cells were injected iv at day 0 and mice were trea-
ted iv the day after tumor injection.
cCells were injected ip at day 0 and mice were treated ip the day after
tumor injection; O.D.=optimal (non toxic) dose <LD10; T/C%=
median survival time of treated versus untreated miceÂ100. L1210
murine leukaemia cell lines were obtained from NCI, Bethesda,
USA.
*Corresponding author. Fax: +39-02-48303833; e-mail: italo.beria@
pharmacia.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00177-4

1468
I. Beria et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1467–1471
Table 2. In vitro cytotoxicity of a-bromoacrylic guanidino deriva-
tives 9–11
Table 3. In vitro cytotoxicity against L1210 murine leukemia cells for
some low MW compounds
Compd
R
IC50 nMa
Compd
n
In vitroa IC50 (nM)
9
10
11
4
3
2
1.8 (Æ0.1)
27.0 (Æ3.1)
>500
4
>2500
aSee footnote a in Table 1.
12
13
43.0
51.9
36.8
This feature is in accordance with what was previously
found in the case of distamycin derived nitrogen mus-
tards3 and was considered to arise from a tighter DNA
binding depending on the increased multiplicity of
interaction between the pyrrolecarboxamide units and
minor groove TA-rich sequences.4
14
aSee footnote a in Table 1.
An identical trend of cytotoxicity (Table 2) has been
more recently found for a-bromoacrylic derivatives of
four, three and two pyrrole units distamycin like frames
ending with a guanidino moiety, that is, PNU-166196
(brostallicin) 9, a potent antitumor agent undergoing
Phase II clinical trials.5
symmetric anhydride, with the suitable amines and in
presence of a base.
In particular the reaction to obtain compound 16, was
carried out in acetonitrile/H2O mixture using NaHCO3
as base, while reactions to yield compounds 12 and 14
were carried out in acetonitrile using triethylamine as
base.
As far as the role of the a halogenoacrylamido moiety is
concerned, the cytotoxicity data clearly indicated a key
role for the reactivity of the a-halogenoacrylic moiety,
that we hypothesized to be based on a first-step
Michael-type nucleophilic attack, followed by further
substitution reaction of the no more vinylic halogen.
Compound 13 was synthesized by coupling 4-[(2-bromo-
acryloyl)amino]-1-methyl-1H-pyrrole-2-carboxylic acid
chloride1 with 3-aminopropanenitrile fumarate. Com-
pound 15 and 17 were synthesized from a-bromoacryl-
oylchloride and the commercially available 4-fluoroaniline
or 4-amino-benzamidine dihydrochloride, respectively.
The reactions for the preparation of compounds 13 and
15 were carried out in methylene chloride and ace-
tonitile respectively, in presence of triethylamine as
base, while for the compound 17, a mixture of dioxane/
H2O in presence of NaHCO3 as a base was used. Com-
pound 18 was prepared by coupling the a-fluoroacrylic
acid, activated as acyl chloride with 4-amino-1-methyl-
N-phenyl-1H-pyrrole-2-carboxamide hydrochloride in
acetonitrile and triethylamine as base.
In fact while a-bromo and chloro derivatives 1, 5 show a
relevant cytotoxicity, the fluoro and acrylamido
derivatives 7, 8 appear devoid of significant activity.
Our hypothesis was supported by the study of the
reactivity toward nucleophilic attack of a-bromoacryl-
amido derivative 12 synthesized as model compound.1
Unexpectedly compound 12, tested in our routine in
vitro activity test on L1210 murine leukemia, proved to
be endowed with a significant cytotoxic activity, higher
than a-bromoacrylic derivative of distamycin 2, at var-
iance with monopyrrole analogue 4 ending with the
amidinoethyl terminus typical of distamycin. This
prompted us to synthesize and test some low MW non-
polypyrrolic a-bromoacrylic derivatives.
Results andDiscussion
The key structural feature that characterized compound
12, in comparison to inactive compound 4, was the lack
of the basic amidino moiety. Therefore, we synthesized
and tested a close analogue of 4 in which the amidino
moiety was replaced by the non basic cyano group (13)
and the simple a-bromoacrylic anilide (14) in which any
residue of the distamycin frame was absent. Both com-
pounds showed a significant cytotoxicity, closely com-
parable with that of 12 (Table 3). It must be underlined
that a-bromoacrylic acid per se is not cytotoxic (L1210:
IC50 >120 mM).7
Chemistry
The novel compounds6 reported in Table 4, except com-
pounds 13, 15, 17 and 18 were synthesized by coupling the
commercially available a bromoacrylic acid, activate as

I. Beria et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1467–1471
Table 4. In vitro and in vivo activity of low MW derivatives
1469
Compd
R
IC50 (nM)a
Calcd pKa
OD ivb mg/kg
nd
%T/C
nd
4
>2500
11.7
12
13
43.0
51.9
ns
ns
25.0
40
100
100
14
15
16
17
36.8
58.6
ns
12.5
>50
25.0
nd
117
100
100
nd
ns
54.0
5.0
8.2
>5000
aSee footnote a in Table 1.
bSee footnote b in Table 1.
These data suggest that a-bromoacrylic derivatives of
low MW are potent cytotoxics only when lacking of a
strong basic moiety. The extreme structural simplicity of
compound 14 prompted us to synthesize and test some
close analogues substituted on the phenyl ring by simple
groups able to modulate the chemico-physical features
of the molecule and the electronic effect on the a-bro-
moacrylic moiety, apparently determinant for the activ-
ity. Table 4 reports also the calculated values of pKa,8
for basic compounds. Few of these derivatives were also
tested for antileukemic activity in vivo.
Table 4 shows, moreover, unexpected lack of antileu-
kemic activity in vivo for compounds 12, 14, 15 and 16,
accompanied also by a substantial lack of significant
toxicity, particularly evident for compound 15, in spite
of their relevant cytotoxicity in vitro. A possible expla-
nation for in vivo inactivity could be a massive meta-
bolic degradation, which may depend upon the presence
of the same a-bromoacrylic moiety which is responsible
for the cytotoxicity and which is the sole reactive moiety
of a-bromoacrylic anilides. The lack of in vivo activity
for compound 15 should exclude a metabolic hydroxyl-
ation on the phenyl ring as a possible explanation of the
inactivity of 14.
The data of Table 4 show that in vitro activities of
different a-bromoacrylic anilides are, in molar terms,
substantially the same, in spite of different electronic
effects of para substituents, and equivalent to those of
neutral a-bromoacrylic pyrrole derivatives 12 and 13,
with the relevant exception of basic amidino derivative
17, which shows the same inactivity of basic amidino
derivative 4.
Again there is an apparent contradiction between the
maintenance of a potent in vivo activity for distamycin
like polypyrrolic frames with three or more pyrrole
units and the inactivity of compounds 12, 14, 15 and 16,
in spite of common presence of the a-bromoacrylic
moiety, which is apparently determinant for both cyto-
toxicity and metabolic degradation in the case of low
MW derivatives.
We hypothesize that the inactivity of compounds 4 and
17 could be explained by the lack of cellular membrane
permeability depending upon the ionization of the
molecules due to the presence of strongly basic amidino
moiety. The mild basicity of dimethylanilino derivative
16 should avoid full ionization of the molecule at pH
7.4, thus allowing cellular membrane permeability and
explaining in vitro activity.
A preliminary attempt to evaluate the cell permeability
of low MW compounds by the Caco-2 cells assay, did
not give conclusive results because of the very low mass
balance shown by many compounds, with the notable
exception of amidino compounds 4 and 17, whose very
low permeability, made reliable in view of its high mass

1470
I. Beria et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1467–1471
Table 5. Caco-2 cell permeability of low MW derivatives
a
Compd
Papp Â10À6 cm/s
Mass balanceb
83.6
IC50 (nM)
4
0.7
3.9
>2500
17
13
12
18
62.0
26.5
1.7
>5000
51.9
uc
1.2
43.0
39.6
88.0
>5000
All the compounds were incubated at a concentration of 10 mM in presence of 0.1% of dimethyl sulfoxide.
aApparent permeability coefficient determined after 120 min of incubation according to the method reported by Artusson.9
bTotal recovery of the compound calculated at the end of the experiment. UC=unable to calculate. Caco-2 cells were grown on permeable filters of
12 polycarbonate transwell for 21 days.
Table 6. Caco-2 cell permeability for a series of a-bromoacrylamidino derivatives
a
Compd
n
Papp Â10À6 cm/sa
Mass balanceb
IC50 (nM)
1
2
3
4
4
3
2
1
38.20
15.21
5.58
117
75
87
84
6.3
98.8
1300
>2500
0.69
aSee footnote a in Table 5.
bSee footnote b in Table 5.
Conclusion
balance, is an accordance with their lack of significant
cytotoxicity (Table 5).
The reported data show that low MW a-bromoacrylic
derivatives, even simple a-bromoacrylic anilides, are
endowed with relevant cytotoxic activity in vitro, pro-
vided that they are lacking of strong basic moieties.
However these compounds are inactive in vivo, a fact
apparently depending on a possible metabolic lability of
a-bromoacrylic moiety.
The low mass balance of the low MW compounds 13
and 12 should arise from the chemical reactivity of
a-bromoacrylic moiety, as confirmed by the high mass
balance shown by non-cytotoxic, unreactive1 a-fluoro-
acrylic analogue 18.
The apparent contradiction with the relevant in vitro
activity of distamycin-like polypyrrolic frames with
three or more pyrrole units, for example, compounds 1
and 2, which share with compounds 4 and 17 the strong
basic amidino moiety, could be explained by the pre-
sence of an active transport mechanism, that may be
justified by the biotic nature of distamycin, but hardly
conceivable for low MW xenobiotics, for which a pas-
sive transport should be expected.
These features underline the peculiarity of potent in vitro
and in vivo activities shown by a-bromoacrylic dis-
tamycin-like derivatives characterized by a polypyrrole
frame ending with strong basic amidino or guanidino
moieties. For these compounds an active transport
mechanism should be hypothesized, in addition to minor
grove binding capability, to explain potent cytotoxicity.
The reason why in the caco-2 cell assay, good mass balance
values were found only for a-bromoacrylic derivatives
ending with an amidino moiety is open to speculation.
Table 6 show preliminary permeability data, performed
using Caco-2 cells line on a series of polypyrrole ami-
dino derivatives, wherein there is a progressive increasing
of permeability from one to four pyrrole units apparently
in accordance with their increased cytotoxicity.
Studies aimed to clarify these issues are foreseen also in
order to better define the mechanistic profile of PNU

I. Beria et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1467–1471
1471
166196 (brostallicin). The results could also give useful
information concerning the hypothesized therapeutic
role of polyazolecarboxamides investigated by different
groups for chemical gene regulation because of their
DNA high affinity and binding specificity.10
6.94 (d, J=1.8 Hz, 1H), 6.65 (d, J=2.9 Hz, 1H), 6.20 (d, J=
2.9 Hz, 1H), 3.83 (s, 3H), 3.82 (s, 3H), 3.80 (s, 3H), 3.40–3.20
(m, 2H); (11) 10.29 (s, 1H), 9.92 (s, 1H), 8.10 (t, J=5.3 Hz,
1H), 7.59 (m, 1H), 7.40–6.90 (bs, 4H), 7.19 (d, J=1.8 Hz, 1H),
7.17 (d, J=1.8 Hz, 1H), 7.02 (d, J=1.8 Hz, 1H), 6.92 (d, J=
1.8 Hz, 1H), 6.67 (d, J=2.9 Hz, 1H), 6.20 (d, J=2.9 Hz, 1H),
3.83 (s, 3H), 3.80 (s, 3H), 3.40–3.20 (m, 2H); (12) 10.27 (s, 1H),
9.83 (s, 1H), 7.69 (m, 2H), 7.29 (m, 2H), 7.26 (d, J=1.8 Hz,
1H), 7.10 (d, J=1.8 Hz, 1H), 7.02 (m, 1H), 6.67 (d, J=2.9 Hz,
1H), 6.21 (d, J=2.9 Hz, 1H), 3.83 (s, 3H); (13) 10.29 (s, 1H),
8.39 (t, J=6.4 Hz 1H), 7.22 (d, J=1.8 Hz, 1H), 6.91 (d,
J=1.8 Hz, 1H), 6.68 (d, J=3.0 Hz, 1H), 6.23 (d, J=3.0 Hz,
1H), 3.83 (s, 3H), 3.40 (m, 2H), 2.74 (t, J=6.4 Hz, 2H); (14*)
8.34 (bs, 1H), 7.58 (m, 2H), 7.37 (m, 2H), 7.22 (m, 1H), 7.13
(d, J=1.7 Hz, 1H), 6.14 (d, J=1.7 Hz, 1H); (15*) 8.33 (bs,
1H), 7.56 (m, 2H), 7.13 (d, J=1.7 Hz, 1H), 7.10 (m, 2H),
6.14 (d, J=1.7 Hz, 1H); (16) 10.45 (s, 1H), 7.72 (m, 2H), 7.53
(bs, 2H), 6.79 (d, J=2.9 Hz, 1H), 6.30 (d, J=3.3 Hz, 1H),
3.04 (s, 6H); (17) 10.67 (s, 1H), 9.22 (bs, 2H), 8.83 (bs, 2H),
7.78–7.89 (m, 4H), 6.82 (d, J=3.1 Hz, 1H), 6.39 (d, J=3.1 Hz,
1H).
References andNotes
1. Cozzi, P.; Beria, I.; Caldarelli, M.; Capolongo, L.; Geroni,
C.; Mongelli, N. Bioorg. Med. Chem. Lett. 2000, 10, 1269.
2. Cozzi, P.; Beria, I.; Caldarelli, M.; Capolongo, L.; Geroni,
C.; Mongelli, N.; Pennella, G. Bioorg. Med. Chem. Lett. 2000,
10, 1273.
3. Cozzi, P.; Mongelli, N. Curr. Pharm. Des. 1998, 4, 195.
4. Dervan, P. B. Science 1986, 232, 464.
5. Cozzi, P. Idrugs 2001, 4, 573.
6. Tested compounds were purified by silica gel column chro-
matography giving satisfactory analytical values and 1H NMR
1
spectra in agreement with assigned structures. H NMR data
of representative novel compounds (DMSO-d6 or *CDCl3) are
given. (Varian 200 and 300 MHz spectrometer, d in ppm, TMS
as internal standard): (10) 10.28 (s, 1H), 9.93 (s, 1H), 9.89 (s,
1H), 8.09 (t, J=5.3 Hz, 1H), 7.54 (m, 1H), 7.40–6.90 (bs, 4H),
7.22 (d, J=1.8 Hz, 1H), 7.21 (d, J=1.8 Hz, 1H), 7.18 (d, J=
1.8 Hz, 1H), 7.04 (d, J=1.8 Hz, 1H), 7.02 (d, J=1.8 Hz, 1H),
7. D’Alessio, R.; Geroni, C.; Biasoli, G.; Pesenti, E.; Grandi,
M.; Mongelli, N. Bioorg. Med. Chem. Lett. 1994, 4, 1467.
8. Commercial ACD software was used. (ACD/pKa DB v.
4.0. Advanced Chemistry Department Inc. Toronto, Canada.
9. Artursson, P. J. Pharm. Sci. 1990, 79, 476.
10. Dervan, P. B. Curr. Opin. Chem. Biol. 1999, 688.