Aldehyde dehydrogenase inhibitors: α,β-Acetylenic N-substituted aminothiolesters are reversible growth inhibitors of normal epithelial but irreversible apoptogens for cancer epithelial cells from human prostate in culture

Article abstract of DOI:10.1016/j.ejmech.2007.06.004

The pharmacomodulation of the N atom of α,β-acetylenic aminothiolesters or the replacement of the thiolester moiety by more electrophilic groups did not permit any clear rationale to be established for improving the selective growth-inhibitory activity of this family of compounds over that of the previously synthesized α,β-acetylenic aminothiolesters DIMATE and MATE [G. Quash, G. Fournet, J. Chantepie, J. Gore, C. Ardiet, D. Ardail, Y. Michal, U. Reichert, Biochem Pharmacol 64 (2002) 1279-92]. Hence DIMATE and MATE were investigated more thoroughly for selectivity and growth-inhibitory activity using human prostate epithelial normal cells (HPENC) on the one hand and human prostate epithelial cancer cells (DU145) on the other. Unequivocal evidence was obtained showing that both compounds were reversible growth inhibitors of HPENC but irreversible growth inhibitors of DU145. Growth-inhibition of DU145 was due to the induction of early apoptosis as revealed by the flow cytometric analytical profile of inhibitor-treated cells, of the decrease in the redox potential and increase in superoxide anion content of their mitochondria. Of the two intracellular enzymes: aldehyde dehydrogenases 1 and 3 (ALDH1 and ALDH3) targeted by DIMATE and MATE, ALDH3 was inhibited to the same extent by both compounds whereas ALDH1 was less susceptible to inhibition by MATE. As the induction of ALDH3 by xenobiotics is hormone-dependent, MATE, the more selective of the two inhibitors, is a useful tool not only for examining the role of the ALDH3 isoform in hormone-sensitive and resistant prostate cancer cells in culture but also for investigating if it can inhibit the growth of xenografts of prostate cancer in immunodeficient mice.

Full text of DOI:10.1016/j.ejmech.2007.06.004

Available online at www.sciencedirect.com
European Journal of Medicinal Chemistry 43 (2008) 906e916
http://www.elsevier.com/locate/ejmech
Original article
Aldehyde dehydrogenase inhibitors: a,b-Acetylenic N-substituted
aminothiolesters are reversible growth inhibitors of normal
epithelial but irreversible apoptogens for cancer epithelial cells
from human prostate in culture
b
b
Gerard Quash a, Guy Fournet b, , Charlotte Courvoisier , Rosa M. Martinez ,
*
Jacqueline Chantepie a, Marie Julie Paret a, Julie Pharaboz c,
c
Marie Odile Joly-Pharaboz , Jacques Gore , Jean Andre , Uwe Reichert
b
c
d
´
´
a Laboratoire d’Immunochimie, Faculte´ de me´decine Lyon Sud, INSERM U-329, Oullins, F-69621, France
b Laboratoire de Chimie Organique 1, CNRS UMR5246, CPE, Villeurbanne, F-69622, France
c Hopital Debrousse, INSERM U-329, Lyon, F-69322, France
d Galderma R & D, Route des luciolles, BP 87, 06902 Sophia Antipolis, Valbonne, France
Received 19 January 2007; received in revised form 31 May 2007; accepted 8 June 2007
Available online 6 July 2007
Abstract
The pharmacomodulation of the N atom of a,b-acetylenic aminothiolesters or the replacement of the thiolester moiety by more electrophilic
groups did not permit any clear rationale to be established for improving the selective growth-inhibitory activity of this family of compounds
´
over that of the previously synthesized a,b-acetylenic aminothiolesters DIMATE and MATE [G. Quash, G. Fournet, J. Chantepie, J. Gore,
C. Ardiet, D. Ardail, Y. Michal, U. Reichert, Biochem Pharmacol 64 (2002) 1279e92]. Hence DIMATE and MATE were investigated more
thoroughly for selectivity and growth-inhibitory activity using human prostate epithelial normal cells (HPENC) on the one hand and human
prostate epithelial cancer cells (DU145) on the other. Unequivocal evidence was obtained showing that both compounds were reversible growth
inhibitors of HPENC but irreversible growth inhibitors of DU145. Growth-inhibition of DU145 was due to the induction of early apoptosis as
revealed by the flow cytometric analytical profile of inhibitor-treated cells, of the decrease in the redox potential and increase in superoxide anion
content of their mitochondria. Of the two intracellular enzymes: aldehyde dehydrogenases 1 and 3 (ALDH1 and ALDH3) targeted by DIMATE
and MATE, ALDH3 was inhibited to the same extent by both compounds whereas ALDH1 was less susceptible to inhibition by MATE. As the
induction of ALDH3 by xenobiotics is hormone-dependent, MATE, the more selective of the two inhibitors, is a useful tool not only for exam-
ining the role of the ALDH3 isoform in hormone-sensitive and resistant prostate cancer cells in culture but also for investigating if it can inhibit
the growth of xenografts of prostate cancer in immunodeficient mice.
Ó 2007 Elsevier Masson SAS. All rights reserved.
Keywords: Acetylenic aminothiolesters; ALDH1 and ALDH3 inhibitors; Human prostate epithelial normal cells; Human prostate cancer cells; Apoptosis
1. Introduction
irreversible inhibitor of aldehyde dehydrogenase 1 (ALDH1)
with an apparent Ki of 280 mM and an inducer of apoptosis
(IC50 4 mM) in mouse lymphoid cells BAF3 overexpressing
the antiapoptotic gene bcl2 [1]. In two rat hepatoma cell lines
in culture (JM2 and 7777) treated with DIMATE, the activities
of both ALDH1 and ALDH3 were inhibited and there was
a concomitant inhibitory effect on cell growth with IC50s of
10e25 mM [2]. On the contrary, the morpholino derivative
We have previously shown that 4-amino-4-methyl-pent-
2-ynthioic acid, S-methyl ester (ATE) was a competitive
* Corresponding author. Fax: þ33 4 72431214.
E-mail address: fournet@univ-lyon1.fr (G. Fournet).
0223-5234/$ - see front matter Ó 2007 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2007.06.004

G. Quash et al. / European Journal of Medicinal Chemistry 43 (2008) 906e916
907
MATE was a very weak inhibitor of yeast ALDH1 (15% inhi-
bition at 600 mM) [3] but retained good growth-inhibitory
activity towards BAF3 bcl2 (IC50 12 mM) [4].
our a,b-acetylenic N-substituted aminothiol esters on such
a spectrum of cells in culture was considered to be essential
before attempting any experimentation on mice in vivo.
Compounds less active than DIMATE or MATE were not
retained for further investigations. Those with increased
growth-inhibitory activity compared to that of DIMATE and
MATE were screened for selectivity on HPENC. When there
was some evidence for selectivity towards DU145 compared
to HPENC, investigations were pursued: to determine whether
growth was inhibited reversibly or irreversibly, to identify the
ALDH isoform implicated therein, to assess if apoptosis had
taken place and if so, to measure some of the biochemical
changes associated with apoptosis such as the mitochondrial
redox potential and superoxide anion levels.
On DU145 (cerebral metastases of human prostate cancer)
cells in culture, the IC50s of DIMATE and MATE were similar:
6e9 mM whereas on human embryonic lung normal fibroblasts
(MRC5), the IC50 of DIMATE was 8 mM, (not different from
that observed on DU145) but that of MATE (29 mM) showed
about a three-fold increase. From these preliminary observa-
tions, MATE appeared to show some selectivity towards cancer
compared to normal cells, but this interpretation was subject to
caution because MRC5 cells are fibroblasts whereas DU145 are
of epithelial origin; hence the difference in susceptibility to
MATE may simply be a reflection of the differences in tissue
origin. To try to improve the efficacy and selectivity of these
thiolester compounds as inhibitors of the growth of human pros-
tate cancer cells, additional derivatives were synthesized by
pharmacomodulation of the N atom and changing the thiolester
moiety. Screening for growth-inhibitory efficacy was done on
the well-characterised cell line: DU145 and for selectivity on
their true counterparts human prostate epithelial normal cells
(HPENC). DU145 does not possess the androgen receptor
(AR) and their growth is not stimulated by either androgens
or antiandrogens [5]. They also overexpress the antiapoptotic
gene bcl2 [6] involved in resistance to chemotherapy [7].
To try to make this study as pertinent as possible to the in
vivo situation, two additional cell lines were used. The MOP
cell line was chosen as representative of cells that persist in
a low androgen environment after patient castration. Indeed,
these cells were isolated after continuous culturing of LNCaP
(Lymph Node metastasis of Carcinoma of the Prostate) cells in
a steroid deprived medium for over one year. Their isolation
and growth characteristics in response to androgens and anti-
androgens have previously been described in detail [8] and
contrary to the parent cell line LNCaP, their growth in culture,
on adding androgens is partially inhibited [8]. The other cell
line was derived from MOP cells grafted subcutaneously in fe-
male Swiss nude mice. Tumours were allowed to grow to ap-
proximately 300 mm3 before the start of androgen treatment.
With weekly androgen administration, tumour regression
was almost complete after 3 weeks but then tumours reap-
peared at 5e6 weeks in spite of continuous androgen treat-
ment. Cultures of these tumours gave rise to the ME cell
line. Both cell lines possess the androgen receptor with the
same mutation which is present in the parental cell line
LNCaP [8], but from the differences in their response to an-
drogens and antiandrogens in culture they are representative
of the heterogeneity observed in a population of human pros-
tate cancer cells in vivo. The determination of the activity of
2. Materials and methods
2.1. Cell cultures
DU145, MOP and ME were all grown in RPMI 1640 sup-
plemented with 10% foetal calf serum (FCS) in the case of
DU145, and 5% charcoal-treated FCS in the case of MOP
and ME. Human prostate epithelial normal cells (HPENC),
it’s specialized medium PrEGMÔ and medium supplement
SingleQuotsÒ were obtained from BioWhittaker. FCS (10%)
was an additional supplement for growing HPENC.
All cells were maintained at 37 ꢀC in a humid atmosphere of
air/CO2 (95/5). Cells were harvested by trypsinisation, washed
once in phosphate buffered saline (PBS) by centrifugation at
300g for 5 min and finally suspended in PBS at the appropriate
density 2 ꢁ 104 per well in 48-well plates and 1 ꢁ 105 per well
in 24-well plates. Compounds were added 4 h after seeding and
growth was determined at the intervals indicated by measuring
protein content by the Lowry method [9] or by cell-count using
a Coulter counter (Coulter Electronics, Luton, UK).
2.2. Chemistry
The protocol used for the aminothiolesters’ synthesis is de-
picted in Scheme 1. It consists of a nucleophilic addition of an
amino-acetylide on carbon oxysulfide followed by treatment
with methyl iodide as described before [3]. Required prop-
argyl amines were prepared by standard methods from
3-chloro-3-methyl-1-butyne [10]. Yields and structures of
compounds 2ael are given in Table 1. The preparation of thi-
oamide 4 and it’s iminium salt 5 was carried out using
Hartke’s procedure [11] from corresponding 1-alkyne and
commercial N,N-dimethylthiocarbamoyl chloride as depicted
_
R
R'
R
R'
.I
+NMe3
Cl
N
N
O
O
RNHR'
a
d
b, c
SMe
SMe
1
2
3
Scheme 1. Aminothiolester synthesis: (a) Cu, CuCl, NEt3, Et2O see Ref. [10]; (b) BuLi, THF, ꢂ70e0 ꢀC; (c) COS, ꢂ70e0 ꢀC then MeI; (d) MeI, EtOAc, rt.

908
G. Quash et al. / European Journal of Medicinal Chemistry 43 (2008) 906e916
Table 1
Aminothiolester synthesis
R
N
R
R
R'
R
R'
N
N
O
O
N
R'
R'
SMe
SMe
Me
N
Me
2a (83%) e Ref. [3]
DIMATE
2g (85%)
2h (88%)
N
nPr
N
nPr
2b (75%)
N
O
Me
N
Ph
N
2i (93%) e Ref. [3]
MATE
2c (72%)
2d (59%)
Me
N
S
N
Ph
2j (97%)
Me
N
N
2e (59%)
2f (31%)
2k (56%)
2l (39%)
Ph
Ph
N
Ph
N
in Scheme 2 (yield: 58%). The iminium salt was obtained by
reaction of 4 with methyl iodide (yield: 86%).
2.2.3. 4-Methyl 4-(methylphenylamino)-2-pentynethioic
acid, S-methyl ester (2c)
Prepared as described for DIMATE 2a [3] by using the N-
(1,1-dimethyl-2-propynyl)-N-methyl- benzenamine [10]. Scale:
6 mmol, purification by chromatography on silica gel (petro-
leum ether/Et2O ¼ 94/6), yield: 72% e orange oil e IR
(neat): n ¼ 3080, 3060, 3020, 2200, 1640, 1580, 930, 860,
2.2.1. General remarks
For general remarks see Ref. [3], chemistry section.
2.2.2. 4-(Dipropylamino)-4-methyl-2-pentynethioic acid,
S-methyl ester (2b)
1
790, 770, 700 e H NMR (CDCl3): d ¼ 1.43 (s, 6H), 2.41
Prepared as described for DIMATE 2a [3] by using the
2-methyl-N,N-dipropyl-3-butyn-2-amine [10]. Scale: 3 mmol,
purification by chromatography on silica gel (petroleum
ether/ethyl acetate ¼ 95/5), yield: 72% e colorless oil e IR
(neat): n ¼ 2200, 1460, 1650, 1080, 870 e 1H NMR
(CDCl3): d ¼ 0.86 (s, 6H), 1.42 (s, 6H), 1.47 (m, 4H), 2.38
(s, 3H), 2.52 (m, 4H).
(s, 3H), 2.86 (s, 3H), 7.13e7.22 (m, 1H), 7.25e7.33 (m, 3H)
e
13C NMR (CDCl3): d ¼ 12.5, 29.1, 39.4, 55.0, 81.8, 96.9,
125.7, 127.6, 128.5, 150.3, 176.4 e LREIMS: 247 (Mþꢃ, 4),
204 (100), 200 (69), 148 (64), 113 (47), 107 (81), 77 (58) e
Anal.: calc. for C14H17NOS: %C ¼ 67.98, %H ¼ 6.93,
%S ¼ 12.96; found: %C ¼ 68.37, %H ¼ 6.93, %S ¼ 13.11.
O
N
O
N
O
N
S
SMe
a
b
-
.I
N
N
+
4
5
Scheme 2. Synthesis of compounds 4 and 5: (a) N,N-dimethylthiocarbamoyl chloride, Et3N, PPh3, CuI, PdCl2(PPh3)2, rt; (b) MeI, CH2Cl2, 0 ꢀC to rt.

G. Quash et al. / European Journal of Medicinal Chemistry 43 (2008) 906e916
909
2.2.4. 4-[Methyl (2-benzyl)amino]-4-methyl-2-pentynethioic
acid, S-methyl ester (2d)
Prepared as described for DIMATE 2a [3] by using the
Anal.: calc. for C11H17NOS: %C ¼ 62.52, %H ¼ 8.11; found:
%C ¼ 62.15, %H ¼ 8.11.
N-(1,1-dimethyl-2-propynyl)-N-methyl-benzylamine
[12].
2.2.8. 4-Methyl-4-(1-piperidinyl)-2-pentynethioic acid,
S-methyl ester (2h)
Scale: 5 mmol, purification by chromatography on silica gel
(petroleum ether/Et2O ¼ 95/5), yield: 90% e yellow oil e
IR (neat): n ¼ 3080, 3060, 3020, 2200, 1640, 1600 e 1H
NMR (CDCl3): d ¼ 1.55 (s, 6H), 2.20 (s, 3H), 2.43 (s, 3H),
3.63 (s, 2H), 7.20e7.50 (m, 5H) e 13C NMR (CDCl3):
d ¼ 12.5, 28.0, 36.2, 54.9, 56.6, 81.2, 96.2, 126.9, 128.3,
128.6 (2C), 140.0, 176.5 e Anal.: calc. for C15H19NOS:
%C ¼ 68.93, %H ¼ 7.33, %S ¼ 12.27; found: %C ¼ 68.70,
%H ¼ 7.30, %S ¼ 12.53.
Prepared as described for DIMATE 2a [3] by using the
1-(1,1dimethyl-2-propynyl)-piperidine [15]. Scale: 7 mmol,
purification by chromatography on silica gel (petroleum
ether/ethyl acetate ¼ 80/20), yield: 88% e yellow oil e IR
1
(neat): n ¼ 2200, 1650, 950, 900, 850, 790, 730 e H NMR
(CDCl3): d ¼ 1.32 (s, 8H), 1.50 (4H, m), 2.28 (s, 3H), 2.47
(m, 4H) ꢂ 13C NMR (CDCl3): d ¼ 12.2, 24.0, 26.2, 27.1,
47.9, 54.4, 81.1, 96.1, 176.0 e Anal.: calc. for C12H19NOS:
%C ¼ 63.96, %H ¼ 8.50; found: %C ¼ 63.85, %H ¼ 8.57.
2.2.5. 4-[Methyl (2-phenylethyl)amino]-4-methyl-2-
pentynethioic acid, S-methyl ester (2e)
2.2.9. 4-Methyl-4-(4-thiomorpholinyl)-2-pentynethioic acid,
S-methyl ester (2j)
Required
N-(1,1-dimethyl-2-propynyl)-N-methyl-phene-
thylamine was prepared from 3-chloro-3-methyl-1-butyne ac-
cording to Ref. [10], yield e 55% on 4 mmol scale, pale
yellow solid e Fus. 39e41 ꢀC e IR (KBr): n ¼ 3180, 3080,
3060, 3020, 2080, 1600, 745, 700 e 1H NMR (CDCl3):
d ¼ 1.40 (s, 6H), 2.25 (s, 1H), 2.39 (s, 3H), 2.65e2.90
(m, 4H), 7.23e7.30 (m, 5H). The 4-[methyl (2-phenylethyl)
amino]-4-methyl-2-pentynethioic acid, S-methyl ester was
prepared as described for DIMATE 2a [3] by using the N-
(1,1-dimethyl-2-propynyl)-N-methyl-phenethylamine. Scale:
2 mmol, purification by chromatography on silica gel (petro-
leum ether/Et2O ¼ 95/5), yield: 90% e colorless oil e IR
(neat): n ¼ 3080, 3060, 3020, 2200, 1640, 1600, 800, 750,
700 e 1H NMR (CDCl3): d ¼ 1.37 (s, 6H), 2.34 (s, 3H),
2.65 (m, 4H), 2.85 (m, 4H) e 13C NMR (CDCl3): d ¼ 12.5,
27.9, 35.6, 36.7, 54.6, 54.9, 80.1, 96.5, 126.1, 128.4, 128.9,
140.4, 176.5 e HRCIMS: calc. for C16H21NOS [M þ H]þ
276.1422, found 276.1424.
Required 4-(1,1-dimethyl-2-propynyl)-thiomorpholine was
prepared from 3-chloro-3-methyl-1-butyne according to
Ref. [10], (yield 20% on 8 mmol scale) e yellow solid e
Fus. 36e38 ꢀC e IR (KBr): n ¼ 3300, 3150, 2815, 2805,
2.73 (m, 4H), 2.90 (m, 4H) e 13C NMR (CDCl3): d ¼ 27.3,
28.6, 49.1, 54.3, 71.2, 86.2. The 2-pentynethioic acid, 4-
methyl-4-(4-thiomorpholinyl)-2-pentynethioic acid, S-methyl
ester was prepared as described for DIMATE 2a [3] by using
1
2080 e H NMR (CDCl3): d ¼ 1.38 (s, 6H), 2.30 (s, 1H),
the
4-(1,1-dimethyl-2-propynyl)-thiomorpholine.
Scale:
1 mmol, purification by chromatography on silica gel (petro-
leum ether/Et2O ¼ 70/30), yield: 97% e white crystals e
Fus. 42e44 ꢀC e IR (KBr): n ¼ 2200, 1640 e 1H NMR
(CDCl3): d ¼ 1.37 (s, 6H), 2.34 (s, 3H), 2.65 (m, 4H), 2.85
(m, 4H) e 13C NMR (CDCl3): d ¼ 12.3, 26.8, 28.5, 49.3,
54.7, 80.1, 95.4, 176.1 e LREIMS: 243 (Mþꢃ), 228 (100)
200 (29), 144 (26). Anal.: calc. for C11H17NOS2:
%C ¼ 54.28, %H ¼ 7.04, %S ¼ 26.35; found: %C ¼ 54.23,
%H ¼ 6.95, %S ¼ 26.97.
2.2.6. 4-Methyl-4-[bis(phenylmethyl)amino]-2-
pentynethioic acid, S-methyl ester (2f)
Prepared as described for DIMATE 2a [3] by using the N-
(1,1-dimethyl-2-propynyl)-N-(phenylmethyl)-benzenemethan-
amine [13]. Scale: 1.0 mmol, yield: 31% (not optimized) as
a yellow oil e IR (neat): n ¼ 3080, 3060, 3020, 2200, 1640,
1600, 955, 920, 820, 760, 740, 720, 700 e 1H NMR
(CDCl3): d ¼ 1.45 (s, 6H), 2.45 (s, 3H), 3.82 (s, 4H), 7.32e
7.43 (m, 10H) e 13C NMR (CDCl3): d ¼ 12.6, 29.4, 55.7,
56.5, 80.9, 97.0, 126.7, 128.1, 128.3, 141.2, 176.7 e LREIMS:
337 (Mþꢃ, 3), 322 (76), 294 (17), 91 (100) e Anal.: calc. for
C21H23NOS: %C ¼ 74.74, %H ¼ 6.87, %O ¼ 4.74; found:
%C ¼ 74.83, %H ¼ 6.94, %O ¼ 4.91.
2.2.10. 4-(2,3-Dihydro-1H-indol-1-yl)-4-methyl-2-
pentynethioic acid, S-methyl ester (2k)
Prepared as described for compound 2g by using the 1-(1,1-
dimethyl-2-propynyl)-2,3-dihydro-1H-indole [16]. Scale:
1 mmol, purification by chromatography on silica gel (petro-
leum ether/ethyl acetate ¼ 98/2), yield: 39% e yellow oil e
1H NMR (CDCl3): d ¼ 1.66 (s, 6H), 2.37 (s, 3H), 2.93 (t,
J ¼ 8.05, 2H), 3.40 (t, J ¼ 8.05, 2H), 6.71e6.79 (m, 1H),
7.04e7.12 (m, 3H) e 13C NMR (CDCl3): d ¼ 12.5, 26.5,
28.1, 49.5, 51.4, 80.1, 96.3, 111.3, 118.8, 124.5, 127.0 (4C),
131.5, 149.5, 176.6. Slow decomposition at room temperature.
2.2.7. 4-Methyl-4-(1-pyrrolidinyl)-2-pentynethioic acid,
S-methyl ester (2g)
2.2.11. 4-(1,3-Dihydro-2H-isoindol-2-yl)-4-methyl-2-
pentynethioic acid, S-methyl ester (2l)
Prepared as described for DIMATE 2a [3] by using the
1-(1,1-dimethyl-2-propynyl)-pyrrolidine [14]. Scale: 3 mmol,
purification by chromatography on silica gel (petroleum ether/
ethyl acetate ¼ 70/30), yield: 85% e slightly yellow oil e IR
Prepared as described for compound 2g by using the 2-(1,1-
dimethyl-2-propynyl)-2,3-dihydro-1H-isoindole [17]. Scale:
0.8 mmol, purification by chromatography on silica gel (petro-
leum ether/ethyl acetate ¼ 75/25), yield: 56% e beige solid e
Fus. 64 ꢀC decomposition e IR (KBr): n ¼ 3075, 3040, 3020,
1
(neat): n ¼ 2200, 1650, 1090, 920, 790 e H NMR (CDCl3):
1
d ¼ 1.45 (s, 6H), 1.81 (4H), 2.39 (s, 3H), 2.72 (m, 4H) e
2890, 2200, 1640, 920, 800, 765, 750, 650, 610 e H NMR

910
G. Quash et al. / European Journal of Medicinal Chemistry 43 (2008) 906e916
(CDCl3): 1.55 (s, 6H), 2.35 (s, 3H), 4.12 (s, 4H), 7.21 (m, 4H)
e HRCIMS: calc. for C15H18NOS [M þ H]þ 260.1109, found
260.1106.
concentration of 1 mM. After incubation for 15 min, cells
were analyzed by flow cytometry.
2.6. Measurement of the activity of ALDH1 and ALDH3
2.2.12. N,N,4-Trimethyl-4-morpholin-4-ylpent-2-
ynethioamide (4)
This was carried out on aliquots of 1 mg protein from cell
lysates which had been centrifuged at 48,000g for 30 min. For
the measurement of ALDH1 activity propanal was used as the
substrate and for ALDH3 benzaldehyde was used. Full details
of these methods have been described in Refs. [2,3].
Obtained from corresponding amine following known pro-
cedure [11] as a yellow solid. Scale: 3 mmol, purification by
chromatography on silica gel (CH2Cl2/MeOH ¼ 96/4), yield:
58% e Fus. 102 ꢀC e IR (KBr): n ¼ 2955, 2205, 1510,
1
1385, 1290, 1120 e H NMR (CDCl3): 1.45 (s, 6H), 2.66
(m, 4H), 3.45 (s, 3H), 3.53 (s, 3H) 3.74 (s, 4H) e 13C NMR
(CDCl3): d ¼ 26.9, 41.0, 44.0, 47.6, 54.8, 67.1, 82.5, 100.5,
3. Results
177.8
e
Anal.: calc. for C12H20N2OS: %C ¼ 59.96,
3.1. Pharmacomodulation of the N atom of the lead
compound: 4-amino-4-methyl-pent-2-ynthioic acid,
S-methyl ester (ampal thiolester: ATE)
%H ¼ 8.39, %N ¼ 11.65; found: %C ¼ 60.00, %H ¼ 8.46,
%N ¼ 11.78.
2.2.13. N-Methyl-N-[4-methyl-1-(methylthio)-4-morpholin-
4-ylpent-2-yn-1-ylidene] methanaminium iodide (5)
This involved: changing the pK of the amine group, varying
the steric hindrance on the N atom, modifying the lipophilic
nature of the substituents on the N atom. The importance of
the thiolester group was also assessed. All these new deriva-
tives were screened for their growth-inhibitory effect on
DU145 using DIMATE and MATE as reference compounds
and only those with IC50s (1e2 mM) substantially less than
those of DIMATE and MATE (7e9 mM) were evaluated for
selectivity using HPENC.
Prepared from 4 by treatment with methyl iodide according
to [11]. Isolated by crystallization from Et2O as a white solid e
Fus. 111 ꢀC (decomposition) C e IR (KBr): 2970, 2220, 1585
e 1H NMR (CDCl3): 1.57 (s, 6H), 2.65 (m, 4H), 2.93 (s, 3H),
3.73 (m, 4H), 3.75 (s, 3H), 3.99 (s, 3H) e Anal.: calc. for
C13H23N2OS: %C ¼ 40.84, %H ¼ 6.06, %N ¼ 7.33; found:
%C ¼ 40.88, %H ¼ 6.03, %N ¼ 11.39.
2.3. Culturing cells for analysis by flow cytometry
3.1.1. Effect of ATE derivatives on the growth of prostate
normal and cancer cells
DU145 cells (5 ml) at a density of 1 ꢁ 105/ml were seeded
in 6 cm petri dishes and allowed to fix for 4 h. At the end of
this period duplicate dishes were each treated with: 25 ml
EtOH as controls, 25 ml of 2 mM DIMATE (final concentra-
tion 10 mM), 25 ml of 4 mM DIMATE (final concentration
20 mM). After 72 h incubation, the culture medium was col-
lected and centrifuged. Adherent cells were harvested by treat-
ment with 0.2% EDTA for 5 min at 37 ꢀC, dishes were rinsed
with 1 ml PBS and centrifuged. Pellets from the culture me-
dium and from it corresponding dish were united after
suspension in fresh medium supplemented with 7.5% FCS.
It is apparent (Table 2) that the introduction of a positive
charge on the N atom as the quaternary ammonium compound
3 leads to an 80-fold decrease in inhibitory activity
(IC50 > 600 mM). It seems difficult, to correlate the basicity
of the compounds synthesized to their inhibitory activity. In-
deed, 2g and 2h have effectively similar basicities, but 2h
with a piperidine ring instead of the pyrrolidine in 2g, ex-
hibited a three-fold decrease in activity. The same remark
can be applied to 2i (MATE) and 2j, two other compounds
with comparable pKs, yet 2j showed a six-fold decrease
when morpholine was changed to a thiamorpholine ring.
The lengthening of the alkane chains by two methylenes as
in 2b, or the replacement of the two dimethyl groups of DI-
MATE by two benzyl moieties as in 2f did not lead to an im-
provement in inhibitory activity, all around 7e12 mM. It was
only when one of the methyl groups of 2a (DIMATE) was re-
placed by a benzyl as in 2d a marked increase (5-fold) in in-
hibitory activity was observed, passing from IC50 7 mM to
1.4 mM. The number of methylenes between the aromatic
ring and the N atom in 2d would also appear to be important
because growth-inhibitory activity decreased when the ring
was attached directly to the N atom as in 2c (IC50 12 mM)
or when it was separated from the N atom by two methylenes
as in 2e (IC50 6 mM).
2.4. Measurement of redox potential
EtOH treated cells were suspended at 2.4 ꢁ 106/ml, 10 mM
DIMATE-treated cells at 7.5 ꢁ 105/ml and 20 mM DIMATE-
treated cells at 6.2 ꢁ 105/ml.
To each suspension was added a DMSO solution of 3-3-
dihexyloxacarbocyanine iodide (DiOC6) from Molecular
Probes, Oregon, pre-diluted to 1 mM in EtOH to a final con-
centration of 2.5 nM. After incubation for 15 min, cells were
analyzed by a Beckman/Coulter EPICS (XL) flow cytometer.
2.5. Measurement of superoxide anion
When the N atom was part of a ring such as 2,3-dihydro-
indole or 1,3-dihydroisoindole or pyrrolidine, the correspond-
ing derivatives 2k, 2l and 2g retain inhibitory activity (IC50
6e7 mM) similar to that of MATE. With a thiamorpholine
A solution of hydroethidine (Molecular Probes, Oregon)
pre-diluted to 1 mM in EtOH was added to controls and DI-
MATE-treated cells (as described above) to
a final

G. Quash et al. / European Journal of Medicinal Chemistry 43 (2008) 906e916
911
Table 2
IC50 values of ATE derivatives on prostate normal and cancer cell growth
Compound
2a DIMATE
2b
2c
2d
2e
2f
2g
2h
2i MATE
2j
2k
2l
3 TRIMATE
4
5
DU145
HPENC
MOP
7.2
9.7
3.0
3.0
12.6
12.5
1.4
1.9
6
7
6.7
22
9.3
[10
10.0
10.0
54
7
6.2
>600
590
100
>100
ME
Values are means of three experiments, IC50 values are expressed in mM. HPENC, DU145, MOP and ME cells were seeded at a density of 1 ꢁ 105 cells in 1 ml
culture medium per well of 24-well plates. Compounds were added at 4 h after seeding when cells had adhered. After 3 days in culture, the sheet of adhering
DU145 was washed twice with PBS and harvested directly with 0.14M NaOH. Cell growth was measured by determining protein content. For MOP and ME
growth was measured by cell-count using a Coulter counter. DIMATE and MATE values are the means of four determinations and for the other compounds, means
of two determinations each done in duplicate.
and more curiously with a piperidine, the corresponding
derivatives 2j (IC50 54 mM) and 2h (IC50 22 mM) were less in-
hibitory than MATE itself (IC50 9 mM).
Taken together these results suggest that in addition to the
residual charge on N, other factors, such as steric hindrance on
the N atom or in the case of 2d stabilization by a stacking in-
teraction between the aromatic ring and an amino acid residue
of the enzyme, play a role.
and MATE. However, the values for DIMATE are consistently
lower than those of MATE, suggesting that DIMATE is more
active than MATE. With growth-inhibition established on the
hormone-sensitive and -resistant cancer cell lines it’s revers-
ibility was examined as a criterion of selectivity.
3.4. Reversibility of growth-inhibition
Changing the methyl thiolester group by other electrophilic
groups was undertaken but the acetylenic dithioester function
was excluded due to it’s instability [11]. A thioamide 4 or it’s
more electrophilic iminium salt 5 led to a sharp decrease in in-
hibitory activity with IC50s reaching 100 mM or more. It would
therefore seem important to keep the thiolester group for
inhibitory efficacy.
From this panel of 15 derivatives only 2d showed a clear
five-fold increase in inhibition over DIMATE or MATE. It
was therefore chosen for investigations on selectivity using
HPENC as controls. Unfortunately, the results (Table 2)
showed that it’s increased efficacy as an inhibitor of prostate
cancer cell growth (IC50 1.4 mM) was also accompanied by in-
creased toxicity towards prostate epithelial normal cells (IC50
1.9 mM). Further work on 2d was therefore shelved and redir-
ected to examine the selectivity of DIMATE and MATE.
This was examined by replacing on the third day of culture,
the medium containing inhibitors with fresh medium without
inhibitors. After a further 5 days in culture, cells were exam-
ined under the microscope and growth assessed quantitatively
by measuring the protein content of adhering cells. It is appar-
ent (Fig. 1) that the growth of DU145 cells treated for 3 days
with either DIMATE or MATE has been severely retarded
compared to the growth of DU145 treated with vehicle alone
(0.5% EtOH). Growth-inhibition was still as marked on day
8 with medium change at day 3 as was observed on day 8
without medium change. When the increase in cell number
on day 8 with that on day 3 was compared it was found
(Table 4) that no additional growth had taken place between
days 3 and 8.
In contradistinction to the results obtained with DU145, the
treatment of HPENC with DIMATE and MATE under condi-
tions similar to those of DU145 showed that at 3 days
MATE at 15 mM and DIMATE at 10 mM were growth-inhibi-
tory and the cell sheets were less dense under the microscope
(Fig. 2). Quantitative values for the inhibition by DIMATE and
MATE were 60% and 37%, respectively (Table 5). However,
3.2. Effect of DIMATE and MATE on the growth of
normal and cancer cells
It is apparent (Table 2) that both DIMATE and MATE with
IC50s of 7and 9 mM, respectively, are good inhibitors of the
growth of DU145 whereas on the normal counterparts
HPENC, inhibition is a maximum of 10% with MATE at
10 mM but goes up to 52% with DIMATE at an equivalent con-
centration (Table 3). This result provided the first line of evi-
dence that MATE might be a selective inhibitor of the growth
of prostate cancer cells in culture. However, before investigat-
ing selectivity we first verified the effect of DIMATE and
MATE on the hormone-sensitive cell line MOP and on it’s
hormone-resistant counterpart ME.
Table 3
Effect of DIMATE and MATE on prostate normal and cancer cell growth
Product
Concentration
(mM)
% Inhibition of growth
DU 145
HPENC
DIMATE
0
2
0
6
0
6
5
34
78
0
23
52
0
10
0
MATE
2
0
3
3.3. The effect of DIMATE and MATE on prostate cancer
cell lines
5
10
8
54
6
11
See legend to Table 2. Values for HPENC are the means of three experiments
each done in duplicate. Values for DU145 are the means of five experiments
each done in duplicate.
As can be seen from the IC50s (Table 2) the growth of all
three cell lines DU145, MOP and ME is inhibited by DIMATE

912
G. Quash et al. / European Journal of Medicinal Chemistry 43 (2008) 906e916
Day 3
0.5% EtOH
DIMATE 10 M
µ
MATE 15 M
µ
Day 8 - No medium change at day 3
0.5% EtOH
DIMATE 10 M
µ
MATE 15 M
µ
Day 8 - Fresh medium at day 3
0.5% EtOH
DIMATE 10 M
µ
MATE 15 M
µ
Fig. 1. Morphology by light microscopy at day 8 of DU145 treated with DIMATE and MATE with or without medium change at day 3. Magnification ꢁ200.
the real difference with DU145 was seen with medium change
at day 3. Indeed, HPENC restarted their proliferation in fresh
medium without inhibitor (Fig. 2) and by day 8 cell number
had increased about two-fold when pre-treated with MATE,
three-fold when pre-treated with DIMATE and three-fold
also in vehicle pre-treated cells (Table 5). These results
show that DIMATE and MATE inhibit the growth of human
prostate epithelial normal cells reversibly but that of human
prostate epithelial cancer cells irreversibly. The mechanism
of irreversible inhibition was therefore investigated.
anions (hydroethidine labelling) are increased and redox po-
tential (DIOC6) is decreased. These changes are accentuated
with 20 mM DIMATE and will be discussed further in the light
of the perturbations which take place in mitochondria during
early apoptosis.
3.6. The ALDH isoforms targeted by DIMATE
and MATE
The results of the experiments on the action of these com-
pounds on cells in culture showed that they were effective
growth inhibitors. Nevertheless, their intracellular targets had
to be different because DIMATE is a competitive irreversible
inhibitor of yeast ALDH1 with an apparent Ki of 280 mM
whereas MATE is a very weak inhibitor of yeast ALDH1
showing a maximum of 15% inhibition at 600 mM even
when pre-incubated with the enzyme for 1 h [3]. To try to
3.5. Flow cytometric analyses for cell distribution, redox
potential and superoxide anion
The results (Fig. 3) show quite clearly that in DU145 cells
treated with DIMATE, changes in these three parameters take
place. Apoptotic cells are present in H1 and H3, superoxide

G. Quash et al. / European Journal of Medicinal Chemistry 43 (2008) 906e916
913
Table 4
Effect of DIMATE and MATE on DU145 cell growth with and without
medium change at day 3
elucidate this obvious discrepancy, the ALDH1 and ALDH3
activities of DU145 cells treated for 3 days with the same mo-
lar concentrations of DIMATE or MATE were measured. The
results (Table 6) show that the inhibition of ALDH3 is similar
in both DIMATE-treated and MATE-treated cells whereas that
of ALDH1 is less marked with MATE than that with DI-
MATE. It is clear that of the two isoforms it is the inhibition
of ALDH3 enzyme activity that is better correlated with the
inhibition of growth of DU145. These results are in accor-
dance with many reports in the literature showing that
ALDH3 is involved in controlling cell growth [18].
Samples
% Growth-
inhibition
at day 3
Increase in cell number day 8/day 3
No medium
change at day3
Fresh medium
at day 3
Control
0
67
70
88
1.6
2.5
3
2.1
2.7
2.6
1.6
DIMATE 10 mM
MATE 10 mM
MATE 15 mM
1.9
Values are the means of two experiments each done in duplicate. DU145 cells
were seeded in six wells and treated with compounds as described in the leg-
end to Table 2. After 3 days in culture the medium was aspirated in two wells,
the cell sheet washed with PBS and 1 ml of fresh medium added. Cells were
incubated for a further 5 days and growth was measured on day 8. In two other
wells cells were harvested after washing and protein content was determined.
In the last two wells cells were left in the medium in which they were origi-
nally seeded and incubated for an additional 5 days. Cell growth was measured
on day 8 by protein content of adherent cells. Control wells received 5 ml
EtOH and were treated exactly as described for inhibitor-treated wells.
4. Discussion
The results obtained with this limited panel of a,b-acetyle-
nic N-substituted aminothiol esters show that no clear-cut
rationale emerged for ‘‘designing’’ growth inhibitors more ef-
fective and selective than DIMATE and MATE. Nevertheless,
Day 3
0.5% EtOH
DIMATE 10 M
µ
MATE 15 M
µ
Day 8 - No medium change at day 3
0.5% EtOH
DIMATE 10 M
µ
MATE 15 M
µ
Day 8 - Fresh medium at day 3
0.5% EtOH
DIMATE 10 M
µ
MATE 15 M
µ
Fig. 2. Morphology by light microscopy at day 8 of HPENC treated with DIMATE and MATE with or without medium change at day 3. Magnification ꢁ200.

914
G. Quash et al. / European Journal of Medicinal Chemistry 43 (2008) 906e916
Table 5
Effect of DIMATE and MATE on HPENC cell growth with and without
medium change at day 3
one moiety that contributes to growth-inhibitory activity is the
thiolester. However, it does appear that it’s contribution to
growth inhibiton is not dependent on it’s degree of electrophi-
licity since derivatives of MATE with other electrophiles such
as the thioamide 4 or it’s more electrophilic iminium salt 5
showed a sharp decline in inhibitory activity (IC50 ꢄ 100 mM).
Our results with MATE (the thiolester) confirm earlier reports
showing the importance of the ester functionality of the
methyl and isopropyl esters of chlorpropamide for the inhibi-
tion of ALDH3 [19].
Samples
% Growth-
inhibition
day 3
Increase in cell number day 8/day 3
No medium
change at day3
Fresh medium
at day 3
Control
0
60
0
0.7
0.9
0.4
1.4
2.2
2.8
1.6
2.6
DIMATE 10 mM
MATE 10 mM
MATE 15 mM
37
Values are the means of two experiments each done in duplicate. HPENC were
subjected to the identical experimental conditions described in the legend to
Table 4.
With regard to efficacy, the only criterion which improved
growth-inhibition was hydrophobicity on N as in 2d
Fig. 3. Flow cytometric analysis for cell distribution, redox potential and superoxide anion content of DU145 treated with 10 and 20 mM DIMATE for 3 days.

G. Quash et al. / European Journal of Medicinal Chemistry 43 (2008) 906e916
915
Table 6
Effect on ALDH1 and ALDH3 of DU145 incubated with DIMATE and MATE
for 3 days
When the mechanism of growth-inhibition was investi-
gated, unequivocal evidence was obtained for inhibitor-in-
duced apoptosis as shown by flow cytometric analysis of the
changes in the distribution of DIMATE-treated DU145 cells
(Fig. 3), in their redox potential and superoxide anion profiles
(Fig. 3). The decrease in the mitochondrial redox potential
(DIOC6 labelling) and increase in superoxide anion (hydroe-
thidine labelling) confirm that DIMATE (Fig. 3), which had
previously been shown to also decrease mitochondrial gluta-
thione [3], is acting on the execution phase of early apoptosis
[20] before events typical of late apoptosis such as nuclear
degradation and Annexin V positivity are manifested [20].
We can therefore conclude that DIMATE is apoptogenic on
human prostate epithelial cancer cells as had previously been
shown on mouse lymphoid cells BAF3 bcl2 [3]. Concerning
the apoptotic activity of MATE, the above mentioned bio-
chemical parameters of apoptosis were not examined, only
DNA fragmentation was measured on ME and here too, there
was evidence for an increase in low MW DNA on electropho-
resis (results not shown).
As the results obtained so far on these two inhibitors
showed that they were comparable as growth inhibitors but
with some advantage for MATE in terms of selectivity
(IC50 MATE [ 10 mM on HPENC vs IC50 DIMATE 9.7 mM on
HPENC), their effects on their intracellular putative target en-
zymes (ALDH isoforms) were investigated. Additional rea-
sons for this experiment were previous results showing that
(a) DIMATE is a good competitive irreversible inhibitor of
yeast ALDH1 whereas MATE is a very poor inhibitor of this
enzyme showing a maximum of 15% inhibition even when
pre-incubated at 600 mM with the enzyme for 1 h [3]; (b)
the inhibition of ALDH1 activity in BAF3 bcl2 cells by DI-
MATE is accompanied by the increased fragmentation of
DNA typical of apoptosis [3]. Hence, if ALDH1 in DU145
were as refractory as yeast ALDH1 to inhibition by MATE,
the growth-inhibitory effect of MATE could not be attributed
to it’s inhibition of this isoform. The results shown in Table
4 permit this issue to be clarified. Indeed, in DU145 cells in-
cubated with either inhibitor for 3 days, ALDH3 was the iso-
form that was inhibited to the same extent by both compounds.
ALDH1 was two-fold less sensitive to inhibition by MATE
than by DIMATE. This result confirms many other earlier re-
ports in the literature showing that ALDH3 is involved in con-
trolling cell growth [18,21], in protecting cells from the
deleterious effects of xenobiotics [22,23] and of endogenous
toxic aldehydes, e.g., 4-hydroxynonenal [24]. It is noteworthy
that this enzyme is expressed constitutively in some
tissues such as cornea, lung and stomach [18,25] whereas it
is inducible in others such as liver [25]. Further, the existence
of an apparently tumour-specific form of ALDH3 has been
reported [26].
Compound
Concentration mM
% Inhibition
ALDH1
ALDH3
DIMATE
2.5
5
15
40
78
3
40
62
7.5
2.5
5
100
40
MATE
20
32
60
78
7.5
Values are the means of two experiments each done in duplicate. DU145 and
HPENC were seeded as described in the legend to Table 2 and treated with
DIMATE or MATE at the concentrations indicated. After 3 days in culture,
cells were harvested by trypsinisation, centrifuged, washed twice with PBS
and the pellet lysed directly in 500 ml 60 mM sodium phosphate buffer of
pH 7 containing 1% (v/v) TritonX-100, 1 mM EDTA for 30 min at 4 ꢀC. Ly-
sates were centrifuged at 48,000g for 30 min and the supernatants were used
for measuring ALDH activity. Activity of ALDH1 was measured by adding
1 mg lysate protein to a reaction mixture containing 60 mM sodium phosphate
buffer of pH 8.5, 1 mM EDTA, 100 mM KCl, 2.35 mM NAD. The reaction
was started by the addition of 2 mM propanol and the final volume was
1 ml. ALDH activity was measured by the increase in OD340 of the NADH
formed and inhibition as the % loss of activity of activity of DIMATE-treated
vs that of non-inhibitor-treated cells. For measuring ALDH3 activity, 1 mg
protein lysate was also used but with 5 mM benzaldehyde as substrate,
2.5 mM NADP as coenzyme in the presence of 1 mM pyrazole. Activity
and percentage inhibition were determined as described for ALDH1.
(IC50 1.4 mM) but this led to a concomitant decrease in
selectivity as shown by it’s IC50 (1.9 mM) on HPENC. This
loss of selectivity for human prostate cancer cells could be
due to the increased capacity of 2d to diffuse across the hydro-
phobic cell membrane and/or it’s increased affinity for the
putative intracellular target enzyme ALDH. One argument in
favour of the latter hypothesis was the finding that the concen-
tration of 2d which inhibited yeast ALDH1 activity by 50%
was 73 mM, two-fold lower than that of DIMATE 165 mM
(results not shown). The question of diffusion will only be
answered when radiolabeled 2d becomes available.
With regard to selectivity, MATE, from the IC50s in Table 2,
appeared to be less growth-inhibitory on HPENC than DI-
MATE. However, when an additional criterion of selectivity
viz: the reversal of growth-inhibition was examined on
HPENC, it was found that HPENC, which exhibited reduced
growth after 3 days in the presence of both compounds
(Fig. 2), proliferated anew when the medium containing inhib-
itor was changed and replaced by fresh medium without inhib-
itor (Fig. 2). Both compounds can therefore be classified as
reversible growth inhibitors for human prostate epithelial nor-
mal cells, at least under the experimental conditions used here.
The effect on reversibility of longer contact times with the
inhibitors remains to be investigated.
In contradistinction to the observations made on HPENC,
the cancer cell line DU145 did not re-proliferate when me-
dium plus inhibitor was replaced by fresh medium minus in-
hibitor (Fig. 1). This strongly suggests that the inhibitors are
irreversibly cytotoxic for cancer cells; a suggestion supported
by the growth-inhibition also observed on the other two pros-
tate cancer cell lines MOP and ME (Table 2).
Of particular relevance to the work done here is the fact that
the promoter region of the ALDH3 gene contains many differ-
ent response elements among which are those for aromatic hy-
drocarbons and glucocorticoids [27]. These response elements
control ALDH3 synthesis at the transcriptional level as shown
by the differences in inducibility of the enzyme in adult life

916
G. Quash et al. / European Journal of Medicinal Chemistry 43 (2008) 906e916
[9] O.H. Lowry, N.J. Rosenbrough, A.L. Farr, R. Randall, J. Biol. Chem. 193
(1951) 265e275.
when hormones are given during puberty [28]. Indeed, it has
been reported that down-regulation occurred with progester-
one but up-regulation with cortisone, cyproterone and tamox-
ifen [28]. These observations suggest that ALDH3 could be
playing a pivotal role in hormone-dependent tumours such
as prostate and breast cancers. Hence MATE, as an inhibitor
of this enzyme could be used on the one hand for examining
this role in hormone-sensitive cells in culture and on the other
for investigating the effect of ALDH3 inhibition on the growth
of xenografts of human prostate cancer cells in vivo.
[10] G.F. Hennion, R.S. Hanzel, J. Am. Chem. Soc. 82 (1960) 4908e4912.
[11] K. Hartke, H.D. Gerber, U. Roesrath, Tetrahedron Lett. 30 (1989) 1073e
1076.
[12] R.E. McMahon, N.E. Easton, J. Med. Pharm. Chem. 4 (1961) 437e445.
[13] M. Sugawara, A. Fujii, K. Okuro, Y. Saka, N. Nagashima, K. Inoue, T.
Takeda, K. Kinoshita, T. Moroshima, Y. Fuse, Y. Ueda, PCT Int. Appl.
WO 0160795, 2001.
[14] S. Bertrand, N. Hoffmann, S. Humbel, J.P. Pete, J. Org. Chem. 65 (2000)
8690e8703.
[15] C. Polizzi, L. Lardicci, A.M. Caporusso, Gazz. Chim. Ital. 124 (1994)
241e248.
[16] H. Sugiyama, F. Yokokawa, T. Aoyama, T. Shioiri, Tetrahedron Lett. 42
(2001) 7277e7280.
[17] I. Bacso, German Offen DE 2301911, 1973.
Acknowledgments
[18] N.E. Sladek, L. Sreerama, G.K. Rekha, Adv. Exp. Med. Biol. 372 (1995)
103e114.
Financial assistance for this work was provided in part by
´
La Region Rhone Alpes and Galderma R&D, Sophia Antipo-
lis, France.
ˆ
[19] N.E. Sladek, G.K. Rekha, M. Lee, H. Nagasawa, Chem.-Biol. Interact.
130e132 (2001) 135e149.
[20] P.X. Petit, Mitochondrial implication in cell death, in: J.J. Lemasters,
A.L. Nieminen (Eds.), Mitochondria in Pathogenesis, Kluwer Aca-
demic/Plenum Publishers, New York, 2001, pp. 215e246.
[21] R. Lindahl, Aldehyde dehydrogenases and their role in carcinogenesis,
Crit. Rev. Biochem. Mol. Biol. 27 (1992) 283e335.
[22] K.D. Bunting, R. Lindahl, A.J. Townsend, J. Biol. Chem. 269 (1994)
23197e23203.
[23] L. Sreerama, N.E. Sladek, Clin. Cancer Res. 3 (1997) 1901e1914.
[24] R.A. Canuto, M. Ferro, G. Muzio, A.M. Bassi, G. Leonarduzzi,
M. Maggiora, D. Adamo, G. Poli, R. Lindahl, Carcinogenesis 15
(1994) 1359e1364.
References
[1] G. Quash, G. Fournet, C. Raffin, J. Chantepie, Y. Michal, J. Gore,
U. Reichert, in: H. Weiner, E. Maser, D. Crabb, R. Lindhahl (Eds.), Ad-
vances in Experimental Medicine and Biology, vol. 463, Kluwer Aca-
demic/Plenum Publishers, New York, 1999, pp. 97e106.
[2] R.A. Canuto, G. Muzio, R. Salvo, M. Maggiora, A. Trombetta,
J. Chantepie, G. Fournet, U. Reichert, G. Quash, Chem.-Biol. Interact.
130e132 (2001) 209e218.
[25] M. Burton, R. Reisdorph, R. Prough, R. Lindahl, in: H. Weiner, E. Maser,
D. Crabb, R. Lindahl (Eds.), Advances in Experimental Medicine and Bi-
ology, vol. 463, KluwerAcademic/Plenum Publishers, New York, 1999,
pp. 165e170.
´
[3] G. Quash, G. Fournet, J. Chantepie, J. Gore, C. Ardiet, D. Ardail,
Y. Michal, U. Reichert, Biochem. Pharmacol. 64 (2002) 1279e1292.
[4] G. Fournet, G. Quash, J. Gore, US patent 7,078,402 B2, 2006.
[5] A. Chlenski, K. Nakashiro, K. Ketels, G. Korovaitseva, R. Oyasu, Pros-
tate 47 (2001) 66e75.
[6] T.J. Mc Donnell, P. Troncoso, S.M. Brisbay, Cancer Res. 52 (1992)
6940e6944.
[26] L. Sreerama, N. Sladek, Arch Oral Biol. 41 (1996) 597e605.
[27] Y.Q. Xie, K. Takimoto, H.C. Pitot, W.K. Miskimins, R. Lindahl, Nucleic
Acids Res. 24 (1996) 4185e4195.
[28] P. Stephanou, P. Pappas, V. Vasiliou, M. Marselos, in: H. Weiner,
E. Maser, D. Crabb, R. Lindahl (Eds.), Advances in Experimental Med-
icine and Biology, vol. 463, KluwerAcademic/Plenum Publishers, New
York, 1999, pp. 143e150.
[7] T. Miyashita, J.C. Reed, Blood 81 (1993) 151e157.
[8] M.O. Joly-Pharaboz, A. Ruffion, A.M. Roch, L. Michel-Calemard,
´
J. Andre, J. Chantepie, B. Nicolas, G. Panaye, J. Steroid Biochem.
Mol. Biol. 73 (2000) 237e249.