3-Deoxy-3,4-dehydro analogs of XM462. Preparation and activity on sphingolipid metabolism and cell fate

Article abstract of DOI:10.1016/j.bmc.2012.03.073

Three analogs of the dihydroceramide desaturase inhibitor XM462 are reported. The compounds inhibit both dihydroceramide desaturase and acid ceramidase, but with different potencies depending on the N-acyl moiety. Other enzymes of sphingolipid metabolism,

Full text of DOI:10.1016/j.bmc.2012.03.073

Bioorganic & Medicinal Chemistry 20 (2012) 3173–3179
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
3-Deoxy-3,4-dehydro analogs of XM462. Preparation and activity
on sphingolipid metabolism and cell fate
Luz Camacho a, , Fabio Simbari a, , Maria Garrido a, José Luis Abad a, Josefina Casas a, Antonio Delgado b,
Gemma Fabriàs a,
⇑
a Research Unit on Bioactive Molecules (RUBAM), Department of Biomedicinal Chemistry, Institute for Advanced Chemistry of Catalonia,
Spanish Council for Scientific Research (IQAC-CSIC), Jordi Girona 18-26, 08034 Barcelona, Spain
b Unitat de Química Farmacèutica, Facultat de Farmàcia, Universitat de Barcelona (UB), Avda. Joan XXIII s/n, 08028 Barcelona, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Three analogs of the dihydroceramide desaturase inhibitor XM462 are reported. The compounds inhibit
both dihydroceramide desaturase and acid ceramidase, but with different potencies depending on the
N-acyl moiety. Other enzymes of sphingolipid metabolism, such as neutral ceramidase, acid sphingomy-
elinase, acid glucosylceramide hydrolase, sphingomyelin synthase and glucosylceramide synthase, are
not affected. The effect on the sphingolipidome of the two best inhibitors, namely (R,E)-N-(1-hydroxy-
4-(tridecylthio)but-3-en-2-yl)octanamide (RBM2-1B) and (R,E)-N-(1-hydroxy-4-(tridecylthio)but-3-en-
2-yl)pivalamide (RBM2-1D), is in accordance with the results obtained in the enzyme assays. These
two compounds reduce cell viability in A549 and HCT116 cell lines with similar potencies and both
induced apoptotic cell death to similar levels than C8-Cer in HCT116 cells. The possible therapeutic impli-
cations of the activities of these compounds are discussed.
Received 9 February 2012
Revised 26 March 2012
Accepted 30 March 2012
Available online 6 April 2012
Keywords:
Sphingolipids
Ceramides
Ceramidase
Dihydroceramide desaturase
Cancer
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
aCDase activity with increased ceramide levels and apoptotic cell
death in a wide range of cancer cell lines.1
Alterations in the cellular levels of the sphingolipid ceramide, a
potent mediator of apoptosis and an inhibitor of cell proliferation,
contribute to cancer outcome, progression, and resistance to both
chemo- and radiotherapy. The cytotoxic and/or antiproliferative ef-
fect of many antitumor strategies is at least partly due to a boost of
ceramide in a variety of tumor cell lines. Growing evidence points
to important roles of acid ceramidase (aCDase) in cancer develop-
ment and response of tumors to therapy.1,2 Likewise, increasing ef-
forts are being dedicated to the development of aCDase inhibitors
as anticancer drugs, either alone or in combination with radio- or
chemotherapy as sensitizers to overcome resistance.1 Although
the aCDase inhibitor N-oleoylethanolamine (NOE) has been exten-
sively used as a pharmacological tool in tumor biology studies, its
weak potency precludes its therapeutic use. Furthermore, NOE is
not specific for aCDase, as it also inhibits the skin alkaline ceram-
idases3 and the glucosylation of ceramides.4 Much more interest-
ing aCDase inhibitory properties have been exhibited by the
compound B13 and its analogs with modified cell targeting proper-
ties, which have shown effective and selective suppression of
The NOE backbone inspired the design and synthesis of NOE-
analogs, which exhibited interesting aCDase inhibitory proper-
ties.5,6 Among them, the corresponding aminoethanol pivaloyla-
mides and octanoylamides with a E-1-hexadecenyl moiety at C25
were among the most potent ones (compounds E, Fig. 1). These com-
pounds elicited cytotoxicity in A549 cells with higher potencies
than NOE. Cells died by apoptosis, which was accompanied by an in-
crease in ceramide levels. The absence of the characteristic sphingo-
lipid C3 stereogenic centre in these inhibitors meant a valuable
improvement from a synthetic point of view. Along this line, in this
paper we report on a new family of inhibitors arising from the for-
mal replacement of the allylic methylene unit of the above inhibi-
tors with
a sulphur bridge (compounds RBM2-1, Fig. 1).
Interestingly, the sulphur atom at C5 position of the sphingoid back-
bone, present in compounds RBM2-1, is reminiscent of our previ-
ously reported mechanism-based dihydroceramide desaturase 1
(Des1) inhibitor XM462 (Fig. 1).7 In this context, the new analogues
here reported can be regarded as hybrid compounds containing the
most relevant structural features of the above aCDase and Des1
inhibitors. Recent evidence supports that Des1 inhibition is of ther-
apeutical interest.8 It has been shown that some of the effects of
fenretinide rely on its capacity to inhibit dihydroceramide desatura-
tion.9–11 Thus, in the human neuroblastoma cells SMS-KCNR, inhibi-
tion of Des1 and accumulation of endogenous dihydroceramides
⇑
Corresponding author. Tel.: +34 93 4006100; fax: +34 93 2045904.
E-mail address: gemma.fabrias@iqac.csic.es (G. Fabriàs).
Both authors contributed equally to this work and are listed in alphabetical order.
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.03.073

3174
L. Camacho et al. / Bioorg. Med. Chem. 20 (2012) 3173–3179
of the reported assay was used,20 the previously described com-
S(CH2)12CH3
pounds (E’s, Fig. 1)5 were also included in the screening as positive
controls of inhibition and also for comparative purposes. In both
series, pivaloylamides RBM2-1B and E-tb were the most active
compounds, whose activity was higher in intact cells that in cell ly-
sates (Fig. 2). Dose–response curves afforded in vitro IC50 values of
S(CH2)12CH3 EDC, HOBt,
RCOOH, CH2Cl2
HO
HO
HN
R
1
NH2
HO
RBM2-1's
O
OH
(CH2)13CH3
S
77, 315, and 51 lM for compounds RBM2-1B, RBM2-1C and RBM2-
HO
HO
(CH2)12CH3
1D, respectively. Under the same conditions, but using RBM14C16
HN
HN
R
R
HN
HN
as substrate,21 N-pivaloylsphingosine and N-pivaloyldihydrosp-
E'sa
XM462
O
O
hingosine, both at 16
Dase activity (Fig. 2).
lM, produced an about 50% inhibition of aC-
OH
OH
The effect of compounds on neutral ceramidase (nCDase) was
also determined in both intact and lysed cells transformed for
overexpression of nCDase.22 As previously found with structurally
similar compounds,5 none of the vinylthioethers modified nCDase
activity as compared to controls (Fig. S1) either in vitro or in intact
cells.
HO
(CH2)12CH3
(CH2)12CH3
R
R
NPSA
NPSO
O
O
RBM2-1B E-c7a
XM462:
,
and
RBM2-1C and E-tbpha: R=
RBM2-1D E-tba NPSA
R=
To assess the activity of the compounds on Des1, their effect
on the formation of N-[12-[(7-nitro-2-1,3-benzoxadiazol-4-
yl)amino]hexanoyl]-D-erythro-sphingosine (CerC6NBD) from N-
NPSO:
,
,
and
R=
[12-[(7-nitro-2-1,3-benzoxadiazol-4-yl)amino]hexanoyl]-
D
-ery-
thro-dihydrosphingosine (dhCerC6NBD) was tested following the
standard assay.7 The cell line HCT116 was used, as it is a very suit-
able model to determine Des1 inhibition (Muñoz-Olaya et al.,
unpublished). As shown in Figure 3A, amounts of CerC6NBD de-
creased significantly with the three compounds, as well as with
XM462, which was used as positive control. Similar effects were
found with the A549 cell line, previously used to investigate the
activity of structurally related compounds5 (data not shown). Com-
pound RBM2-1B was the most potent one with both cell models.
Its inhibitory activity was dose dependent with IC50 values around
Figure 1. Compounds investigated in this study and their preparation. aReported in
Chem. Phys. Lipids 2008, 156, 33 and included here for comparison. XM462 was
reported in ChemMedChem 2008, 3, 946. HOBt, hydroxybenzotriazole; EDC, 1-ethyl-
3-(3-dimethylaminopropyl)carbodiimide hydrochloride; NPSA, N-pivaloyldihyd-
rosphingosine; NPSO, N-pivaloylsphingosine.
inhibited cell growth, with cell cycle arrest at G(0)/G(1) and a signif-
icant decrease in the amount of phosphorylated retinoblastoma
protein.11 In the A2780 human ovarian carcinoma cells, fenretinide
incremented the production of dihydroceramide with a concomi-
tant reduction of cell proliferation and induction of apoptosis.12,13
In a recent paper, Mao et al.14 have shown that in two different cell
models, HeLa cervical tumor and human oral squamous cell carci-
noma cells, the cytotoxicity elicited by fenretinide is not due to
dihydroceramide, but to dihydrosphingosine resulting from dihy-
droceramide hydrolysis by the Golgi alkaline ceramidase (Asah3L
or ACER2).15 Des1 is also inhibited by resveratrol16 and the cycloox-
ygenase 2 inhibitor celecoxib.17 Inhibition of Des1 and increase of
intracellular dihydroceramides has been linked to the induction of
autophagy in different cancer cell lines.9,16
Both the synthesis of the new analogues RBM2-1B/D and their
biological activity in two different human adenocarcinoma cell
lines, namely lung A459 and colon HCT116, are reported in this
article,18 and their interest as new lead structures for the discovery
of inhibitors of sphingolipid metabolizing enzymes with impact on
cancer cell death is discussed.
18 lM with both HCT116 and A549 cell models (Fig. 3C). The other
compounds produced a 50% inhibition at concentrations P 100 lM
(Fig. 3B).
The effect of the vinylthioethers RBM2-1B/D on other sphingo-
lipid metabolism enzymes was also investigated following stan-
dard protocols (see Supplementary data). Neither compound
affected the activities of either the acidic sphingomyelinase
(Fig. S2A) or glucocerebrosidase (Fig. S2B). These activities de-
creased in the presence of desipramine and N-nonyl-1-deoxynojir-
imycin (NNDNJ), which were used as the respective positive
controls of inhibition in parallel experiments. The effect of the
compounds on the metabolization of CerC6NBD into both the glu-
cosyl and the phosphocholine derivatives was also investigated. No
marked effect was elicited by any of the compounds on the
formation of either metabolite from the administered probe
2. Results
2.1. Chemistry
Preparation of amides was carried out by N-acylation of the
sphingoid backbone 119 with the corresponding carboxylic acid
in the presence of 1-ethyl-3-(3-(dimethylamino)propyl)-carbodi-
imide hydrochloride/hydroxybenzotriazole as coupling reagents
(Fig. 1). N-Pivaloyldihydrosphingosine (NPSA) and N-pivaloylsp-
hingosine (NPSO) were prepared following the same procedure.
Figure 2. Activity of compounds as aCDase inhibitors. aCDase activities were
determined in Moh. pAS AcCer10X cells with a fluorogenic substrate (16
lM).
Compounds were tested at 16 M (aCDase; S/I = 1) and the experiments were
l
performed as detailed in Section 5. Data correspond to the mean% of control ( SD)
of three experiments with triplicates. Means are statistically different from controls
at: ⁄p 6 0.05; ⁄⁄p 6 0.005 (unpaired two-tail t-test). Results are normalized to
amount of protein (in vitro) or number of cells (intact), which was similar in all
cases. Ctrl-(ꢀ) corresponds to Moh. pAS cells. NPSO, N-pivaloylsphingosine; NPSA,
N-pivaloyldihydrosphingosine.
2.2. Effect on enzymes of sphingolipid metabolism
The compounds were tested as aCDase inhibitors both in vitro
and in intact cells, using a Farber cell line transduced to overex-
press aCDase (Moh. pAS AcCer10X cell line). Since a modification

L. Camacho et al. / Bioorg. Med. Chem. 20 (2012) 3173–3179
3175
Figure 4. Effect of compounds RBM2-1B and RBM2-1D on the sphingolipidome.
Cells were incubated with the compounds (10 M) for 24 h (A549) or 14 h
l
(HCT116) and then lipids were extracted and analyzed as detailed in Section 5. Data
correspond to the mean SD of one representative experiment with triplicates.
Asterisk indicates statistically significant difference from the mean (p 6 0.05;
unpaired two-tail t-test). (So: sphingosine; Sa: dihydrosphingosine).
In the HCT116 cell line, RBM2-1B and RBM2-1D produced 8.5 and
4.5-fold increases in total dihydroceramides, respectively (Fig. 4B)
and these rises were statistically significant as compared to controls
for all the several N-acyl species (Fig. S5) (mean total pmol/106 cells:
ctrl., 11.8 1.13; RBM2-1B, 100.8 9.69, p 6 0.00009; RBM2-1D,
53.5 10.89, p 6 0.00273). These increments were accompanied
with significant increases in both dihydrosphingomyelins (dhSM’s)
and glucosyldihydroceramide(s) GlcdhCer’s (Fig. 4B), a significant
reduction in total glucosylceramide(s) (GlcCer’s) (Fig. 4B) and a mar-
ginally significant decrease of total sphingomyelins (SM’s). Although
total ceramides (Cer’s) were not modified by any compound, both
elevated significantly the production of C18-ceramide (mean pmol/
106 cells SD: ctrl., 3.9 0.3; RBM2-1B, 16.0 1.6, p 6 0.00025;
RBM2-1D, 10.8 1.7, p 6 0.0025) and reduced that of C24-Cer and
C16-SM (Fig. S5). Analysis of long chain bases showed that RBM2-
1B did not modify either sphingosine or dihydrosphingosine levels.
In contrast, the amounts of both bases exhibited a slight but signifi-
cant increase with RBM2-1D. No significant changes in the amounts
of sphingosine-1-phosphate occurred with any treatment (mean
pmol/106 cells SD of three experiments with triplicates: Ctrl.,
1.49 0.70; RBM2-1B, 1.25 0.22, p P 0.43; RBM2-1D, 2.17 0.56,
p P 0.16), while amounts of dihydrosphingosine-1-phosphate were
barely above background in all experimental groups.
Figure 3. Effect of compounds on Des1 activity. (A) Percentage of CerC6NBD over
total fluorescent species in HCT116 intact cells incubated for 4 h with both
dhCerC6NBD and the test compounds at equimolar concentrations (10
C) Dose–response data obtained with cell lysates of the specified cell lines
incubated for 4 h with dhCerC6NBD (10 M) and different concentrations of
lM). (B and
l
RBM2-1C (B), RBM2-1D (B) and RBM2-1B (C). Samples were processed and analysed
as detailed in Section 5. Data correspond to the mean SD of two (A) and three (B
and D) experiments with triplicates. In C, regression analysis of data (sigmoidal
dose–response with variable slope) affords IC50 values of 18.1
18.8 M (HCT116). In all the experiments, XM462 (8 M) was included as a positive
control of inhibition.
lM (A549) and
l
l
(Fig. S3). However, amounts of the sphingomyelin analogue were
reduced significantly to about 80% of control by incubations with
RBM2-1C and RBM2-1D. While the decreased production of the
phosphocholine derivative was accompanied with
a rise of
CerC6NBD in incubations with RBM2-1D, CerC6NBD did not accu-
mulate in the presence of RBM2-1C, but was converted into the
glucosylceramide analogue, which increased over the control by
about a 30%.
2.4. Metabolization of RBM2-1B and 1D
Ultra-performance liquid chromatography coupled to time-of-
flight mass spectrometry (UPLC-TOF) analysis of the same extracts
used to characterize sphingolipidomes revealed that both com-
pounds, RBM2-1B and RBM2-1D, are mainly metabolized to their
C1-O-phosphocholine derivatives (Table 1), which shows their
suitability as substrates of sphingomyelin synthase 1 and/or 2.
Similar metabolization was found to occur in both cells lines
examined.
2.3. Effect on the sphingolipidome
In the A549 cell line, neither RBM2-1B nor RBM2-1D produced a
significant increase in total ceramides (Figs. 4A and S4), although a
slight but significant rise occurred on C18-ceramide (mean pmol/
106 cells: ctrl., 4.8 0.5; RBM2-1B, 11.5 2.1, p 6 0.005; RBM2-
1D, 7.0 1.1, p 6 0.028). In contrast, both compounds induced high
increments of all dihydroceramide species (Figs. 4A and S4) (mean
total pmol/106 cells SD: ctrl., 6.5 0.42; RBM2-1B, 22.9 4.10,
p 6 0.002; RBM2-1D, 12.4 2.08, p 6 0.01; 3.5 and 1.9-fold
increase with RBM2-1B and RBM2-1D, respectively), while
RBM2-1C did not modify the total amounts of either ceramides
or dihydroceramides (data not shown).
2.5. Effect on cell fate
Similar effects on cell viability were elicited by the vinylthioe-
thers RBM2-1B and RBM2-1D in the two different cell lines, A549
and HCT116 (Table 2). Flow cytometry analysis with propidium
iodide and Annexin V showed that a 14 h treatment with both

3176
L. Camacho et al. / Bioorg. Med. Chem. 20 (2012) 3173–3179
Table 1
MS-Based assignments of RBM2-1B and RBM2-1C metabolites present in extracts of A549 and HCT116 cells
Compounda
Theoretical m/z
Measured m/z (error, ppm)
Amountsb (%conversionc)
A549
HCT116
A549
HCT116
RBM2-1B
PC-RBM2-1B
RBM2-1D
428.3562
593.4117
386.3105
551.3633
427.3577 (+3.5)
593.4100 (ꢀ2.9)
386.3102 (+3.2)
551.3633 (ꢀ2.8)
428.3580 (+4.2)
593.4092 (ꢀ4.3)
386.3102 (+2.4)
551.3630 (ꢀ3.3)
5645 1031
1589 360 (22)
2563 425
3645 398
965 104 (21)
1481 204
PC-RBM2-1D
1318 62 (34)
781 53 (35)
a
RBM1-1B-PC and RBM2-1D-PC refer to the phosphocholine derivatives of RBM2-1B and RBM2-1D, respectively.
The amounts (mean pmols SD, n = 3) of RBM2-1B and RBM2-1D relative to N-dodecylsphingosine and of RBM1-1B-PC and RBM2-1D-PC relative to N-dodecylsp-
b
hingosylphos-phorylcholine.
c
[PC-RBM2-1/(PC-RBM2-1 + RBM2-1)] ꢁ 100.
compounds triggered apoptotic cell death to similar levels than C8-
Cer, which was used as a positive control of apoptosis stimulation.
Moreover, the three compounds promoted also necrotic cell death
with the molecule exhibiting the best apoptosis/necrosis ratio
being RBM2-1B (Fig. 5).
No changes in cell cycle were observed in HCT116 cells after
incubation for 14 h with either RBM2-1B or RBM2-1D (20
(data not shown).
lM)
3. Discussion
Figure 5. Analysis of apoptosis. HCT116 cells were incubated without (ctrl) or with
RBM2-1B (15 M), RBM2-1D (15 M) or N-octanoylsphingosine (Cer-C8, 30 M) for
14 h. Apoptosis was determined by flow cytometry using Annexin V and propidium
iodide as dyes. Data correspond to the mean SD of 3 experiments with triplicates.
Apoptosis includes both propidium iodide (ꢀ)/Annexin V (+) and propidium iodide
(+)/Annexin V (+) cells. Asterisks denote statistical significance at p 6 0.05.
The importance of sphingolipid metabolic pathways as targets
for drug discovery has promoted the search for specific enzyme
inhibitors. The acid ceramidase has attracted special attention be-
cause of its role in cancer development and resistance to therapy.1
In this context, a number of aCDase inhibitors are currently known,
including several C2-substituted aminoethanol amides reported by
our group5 (see Fig. 1). In contrast, the interest of Des1 as a thera-
peutic target in cancer has emerged more recently after the discov-
ery that some drugs inhibit Des1 and elevate dihydroceramides
(dhCer’s) levels as part of their mode of action.8 These include fen-
retinide,11 celecoxib,17 resveratrol,16 and specific forms of vitamin
E.23,24 Besides these drugs, two rationally designed Des1 inhibitors,
namely GT1125,26 and XM462,7 have been reported. The latter is a
dihydroceramide analog with a sulfur atom replacing the natural
C5-methylene unit. Therefore, the vinylthioethers reported here
can be regarded as hybrid analogs that combine the most charac-
teristic structural features of the above aCDase and Des1 inhibitors.
Biological studies have shown that the three compounds inhibit
both enzymes, aCDase and Des1, although their selectivity is mod-
ulated by the N-acyl substituent. Thus, while the p-tert-butylphe-
nylamide is a very poor inhibitor of both aCDase and Des1, the
pivaloylamide is a better aCDase inhibitor and the linear octanoyl
group confers preferential activity over Des1.
l
l
l
difference. The aCDase inhibition data reinforce the importance of
the tert-butyl group in the amide function for inhibitory activity. In
this regard, compound DM102, the Z-isomer of E-tb, inhibited
aCDase in different prostate and breast cancer cell lines and syner-
gized with weakly cytotoxic agents at inducing apoptotic cell
death.27,28 Furthermore, Antoon et al.29 reported that N-pivaloylsp-
hingosine inhibits the viability and clonogenic survival in several
breast cancer cell lines. Similar activities were reported by the
same authors for a ceramide analog with the pivaloylamide unit.30
Although the effect of these compounds on aCDase activity was not
determined, we found that both N-pivaloylsphingosine and N-piv-
aloyldihydrosphingosine produce a 50% and 40% inhibition of aC-
Dase activity, respectively, as measured in intact Moh. pAS
AcCer10X cells with the fluorogenic substrate in the usual condi-
tions followed in our laboratories (Fig. 2A). Therefore, it is possible
that inhibition of this enzyme and the subsequent increase in cera-
mides is responsible for the effects observed by Antoon et al.29,30
Compound RBM2-1B was the most active Des1 inhibitor, show-
ing similar IC50 values in the two cell lines used as source of en-
zyme. However, its activity is lower than that of XM462.7,16 This
difference can be explained assuming a higher affinity of the latter
for the enzyme by virtue of its C4-OH group, which is not present
in RBM2-1B. On the other hand, since XM462 could arise in cells
from metabolism of RBM2-1B, this possibility was investigated
and ruled out in the sphingolipidome analyses, as no traces of
XM462 or its metabolites were detected in extracts from cells trea-
ted with RBM2-1B. Likewise, no evidence of dehydration of XM462
to RBM2-1B was obtained in the UPLC-TOF analyses of lipids ex-
tracted from cells incubated with XM462.
The pivaloylamide RBM2-1D was the most potent aCDase inhib-
itor of the three analogs. Its activity is similar in vitro, but lower in
intact cells, than that found for E-tb. The higher metabolization of
RBM2-1D (34% conversion into the C1-O-phosphocholine
metabolite) (Table 1) as compared to E-tb (16% conversion into
the C1-O-phosphocholine metabolite5) may account in part for this
Table 2
Cytotoxicity of compoundsa
Compound
A549
HCT116
RBM2-1B
RBM2-1C
RBM2-1D
19 0.2
76 0.9
24 0.4
28 0.1
81 1.4
30 1.1
The observed effects of the compounds on the sphingolipidome
were in accordance with the overall data. Thus in the two cell lines
examined, both RBM2-1B and RBM2-1D increased dihydrocera-
mide levels, although RBM2-1B produced a significantly higher
increment than RBM2-1D (A549, p 6 0.017, HCT116, p 6 0.005;
see Fig. 4B), in agreement with its higher potency as Des1 inhibitor.
a
Number of viable cells was determined with the MTT test after 24 h of treat-
ment. Data correspond to the LD50 values ( M) obtained by regression analysis of
l
the dose–response curves constructed with data from two experiments with
triplicates.

L. Camacho et al. / Bioorg. Med. Chem. 20 (2012) 3173–3179
3177
It is worth noting that 24 h treatments with compounds E-c7 and
E-tb induced an accumulation of ceramides, but not dihydrocera-
mides, in A549 cells.5 Therefore, unlike RBM2-1B, compounds E-
c7 and E-tb do not appear to inhibit Des1 in a similar concentration
range, underscoring the importance of the sulfur atom at C5 for
Des1 inhibition.
In contrast to the marked rise in total Cer content produced by
analogs of D-e-MAPP and B13,31 both RBM2-1B and RBM2-1D did
not induce Cer accumulation, which can be explained, at least in
part, by their activity as Des1 inhibitors. Interestingly, however,
both RBM2-1B and RBM2-1D provoked a significant rise in C18-
ceramide in the two cell lines studied. Whether the rise of this spe-
cific ceramide is of importance in the observed cell fate is un-
known, although it has been reported that ceramides with
different chain lengths may have distinct functions in the regula-
tion of tumor progression. Thus, in head and neck squamous cell
carcinoma tumors, C18-ceramide inhibits, while C16-ceramide in-
duces tumor growth.32
Analysis of long chain bases showed that RBM2-1D, but not
RBM2-1B, produced a small but significant increase in the levels
of both sphingosine and dihydrosphingosine (Fig. 4C). Since
RBM2-1D is a more potent aCDase inhibitor than RBM2-1B, de-
creased aCDase activity may cause up-regulation of other ceramid-
ases, such as the nCDase, the alkaline ceramidase Asah3 and/or the
alkaline dihydroceramidase (Asah3L).14 Thus, nCDase and/or Asah3
activity would augment the production of extralysosomal sphingo-
sine to counterbalance the decreased intralysosomal sphingosine
production, while elevated dihydrosphingosine levels would result
from enhanced Asah3L activity.14 Therefore, the lower dihydro-
ceramide levels present in HCT116 cells treated with RBM2-1D
as compared to RBM2-1B (p 6 0.017; see Fig. 4B) might result from
lower inhibition of Des1 together with augmented Asah3L activity.
The latter would not occur with RBM2-1B because its effect on aC-
Dase is not strong enough to up-regulate other ceramidase activi-
experimental setup. The reasons for this difference are unknown,
but two possibilities can be invoked. First, the HCT116 cell line
may not be a good model of dihydroceramide-mediated autoph-
agy, as it is less responsive to XM462 treatment in terms of dihy-
droceramide accumulation than the HGC27 cell line used in our
previous study.9,16 Second, autophagy may indeed have been in-
duced upon RBM2-1B treatment, but at earlier time points than
the one examined, in which cells were already showing signs of
death.
Flow cytometry analysis of HCT116 cells showed that apoptosis
was triggered by RBM2-1B (and RBM2-1D). Moreover, apoptosis
was induced to a similar extent by C8-Cer. Although the role of
ceramides as sphingolipid mediators of apoptotic cell death has
been extensively documented, involvement of dhCer’s in apoptosis
induction is far from proven. Most data come from experiments
using drugs that increase dhCer’s, such as c-tocopherol, c-tocotri-
enol, fenretinide, resveratrol and celecoxib.8 However, exoge-
nously added dihydroceramides have been generally used as the
inactive counterparts of pro-apoptotic ceramides.8 Likely, the proa-
poptotic action of RBM2-1B is independent of its activity as a Des1
inhibitor. Indeed, RBM2-1D, which is a less potent Des1 inhibitor
than RBM2-1B, induces apoptosis to a similar extent. Further stud-
ies are necessary to decipher the precise mechanisms involved in
the propapoptotic activity of RBM2-1B and RBM2-1D. In the case
of RBM2-1D, the possible involvement of dihydrosphingosine de-
serves further attention. Merrill and co-workers showed that
silencing the ASAH1 gene in breast cancer MCF7 cells reduced the
autophagy induced by fenretinide. In the light of these results,
the authors suggested that dihydrosphingosine accumulation con-
tributes to fenretinide cytotoxicity.35 In a recent work, Mao et al.14
proved that the Golgi alkaline ceramidase (Asah3L) is involved in
the release of dihydrosphingosine upon treatment with fenreti-
nide. The authors show that the drug increases ASAH3L expression
with enhanced hydrolysis of dhCer’s, the preferred Asah3L sub-
strate, and provokes dihydrosphingosine production and cell
death. As mentioned above, dihydrosphingosine rose upon treat-
ment of cells with RBM2-1D, but not RBM2-1B, which is a weaker
inhibitor of aCDase. This result suggests upregulation of Asah3L
and augmented hydrolysis of dhCer’s with increased formation of
dihydrosphingosine upon aCDase inhibition. RBM2-1B is not po-
tent enough at inhibiting aCDase to provoke upregulation of
Asah3L as a response.
ties.
A recent report showing that knockdown of alkaline
ceramidase ACER3 up-regulated the expression of ASAH3L33 sup-
ports that this could also occur by sustained inhibition of aCDase.
The three compounds were similarly toxic to the two cell lines
examined. Conversely, RBM2-1C, which exhibited the poorest
activity on sphingolipid metabolism, was also the least toxic com-
pound, in agreement with the similar trend found for the previ-
ously reported compounds E-c7, E-tbph and E-tb.5 This result
supports that cell death induced by RBM2-1B and RBM2-1D is
mediated by altered sphingolipid metabolism. Sphingolipidomics
data point to dihydrosphingolipids and/or C18-ceramide as the ac-
tive mediators of RBM2-1B and RBM2-1D activity. However, given
their structural similarity to ceramide, the possibility that the com-
pounds are active by themselves cannot be ruled out with the
available data. In this regard, in prostate cancer cell models,
DM102 synergized with fenretinide at decreasing cell viability by
induction of apoptosis. However, blocking ceramide generation
failed to rescue cells from cytotoxicity, which casts some doubt
on the involvement of ceramide in the cytotoxic response induced
by the cotreatment.28
Finally, metabolism of RBM2-1B and RBM2-1D occurs mainly
through the formation of the C1-O-phosphocholine derivatives in
both A549 and HCT116 cell lines, with RBM2-1D exhibiting a high-
er incorporation of phosphocholine (PC) (PC-RBM2-1D/RBM2-
1D = 35/65; PC-RBM2-1B/RBM2-1B = 22/78). Competition with
the substrate may thus be the reason why RBM2-1D reduces the
formation of the sphingomyelin derivative of CerC6NBD (Fig. S3).
Therefore, improving the metabolic stability of RBM2-1B and
RBM2-1D at C1OH would likely increase their potency in vivo. Fur-
ther research along this line is ongoing in our laboratories.
4. Conclusions
Since it has been reported that dihydroceramide accumulation
induces autophagy in a number of cell lines,9,16 the effect of
RBM2-1B on autophagy in HCT116 cells was examined. The pair
RBM2-1B/HCT116 was chosen for this study because they afforded
the highest dihydrosphingolipid increases upon treatments. In
agreement with the literature,34 higher LC3-II levels were found
in cells cultured in the absence of serum. The LC3-II increase was
enhanced by co-incubation with protease inhibitors, indicating
that the LC3-II build up results from an enhanced autophagic flux.
Furthermore, HCT116 cells responded to XM462 with autophagy
enhancement, as previously found in a gastric cancer cell line.16
However, RBM2-1B failed to induce autophagy in the same
In summary, the vinylthioether structure reported in this article
emerges as an interesting scaffold in the discovery of inhibitors of
sphingolipid metabolism. As exemplified with RBM2-1B and
RBM2-1D, selectivity towards specific enzymes can be achieved
by changing the N-acyl group. Thus, while the linear octanoyl
moiety in RBM2-1B confers selectivity against dihydroceramide
desaturase, the pivaloyl group of RBM2-1D renders the compound
selective for acid ceramidase. The pro-apoptotic activity of both
RBM2-1B and RBM2-1D in HCT116 cells warrants the interest of
further studies with both compounds in other cancer cells lines.

3178
L. Camacho et al. / Bioorg. Med. Chem. 20 (2012) 3173–3179
C22H43NO2S: C, 68.52; H, 11.24; N, 3.63. Found: C, 68.55; H,
5. Experimental
11.21; N, 3.65.
5.1. General protocol for the acylation of sphingoid bases
5.2. Cells
To a solution of the corresponding acid (0.22 mmol) and
hydroxybenzotriazole (32 mg, 0.25 mmol) in anhydrous CH2Cl2
(4 mL) was added solid 1-ethyl-3-(3-dimethylaminopropyl) carbo-
diimide hydrochloride (62 mg, 0.3 mmol) at room temperature un-
der argon atmosphere. After disappearance of the white
precipitate, the resulting solution was stirred for 5 min and then
transferred dropwise via syringe to a vigorously stirred solution
A549,36 Moh. pAS and Moh. pAS AcCer10X37 cells were cultured
as reported. HCT116 cells (clone 40–16) were maintained at 37 °C
and 5% CO2 in low glucose Dulbecco’s Modified Eagle Medium
(DMEM) supplemented with 10% FBS and 1% penicillin–streptomy-
cin. All the experiments were performed at about 80% confluence.
containing 119 (60 mg, 0.20 mmol) and Et3N (56
lL, 0.40 mmol)
5.3. Enzyme inhibition
in anhydrous CH2Cl2 (4 mL). After stirring for 1 h at room temper-
ature under argon, the mixture was diluted with CH2Cl2 (5 mL) and
washed with saturated NaHCO3 solution (5 mL) and brine (2 mL).
The organic layer was dried over anhydrous MgSO4, filtered and
solvent was evaporated to afford a crude residue, which was puri-
fied by flash chromatography (two gradients: hexane/MTBE 0–20%
and then CH2Cl2/MeOH 0–3%) to afford the pure acylated products.
aCDase activity was measured both in vitro and in intact cells as
reported20,21 using Moh. pAS AcCer10X cells37 and the fluorogenic
substrate RBM14C12.nCDase activity was determined as
described22,38 using N-[12-[(7-nitro-2-1,3-benzoxadiazol-4-yl)
amino]lauroyl]-D-erythro-sphingosine (CerC12NBD) as the sub-
strate and Moh. pAS transfected with nCDase. Twenty-four hours
before transfection, cells were plated at a density of 2.5 ꢁ 105
cells/mL in 6 well-plates (35 mm, 9.46 cm2, 1 mL/well). Then cells
were transfected with pcDNA5/TO (empty vector) or pcDNA5/TO-
5.1.1. Compound RBM2-1B (72% yield)
1H NMR (400 MHz, CDCl3): d 6.27 (d, 1H, J = 15.3 Hz, CH@CH–S),
5.74 (br s, 1H, NHCO), 5.46 (dd, 1H, J1 = 15.3, J2 = 6.8 Hz, CH@CH–S),
4.61–4.54 (m, 1H, CHNH), 3.74–3.62 (m, 2H, CH2OH), 2.67 (t, 2H,
J = 7.4 Hz, S–CH2), 2.48 (br s, 1H, CH2OH), 2.21 (t, 2H, J = 7.6 Hz,
COCH2), 1.68–1.57 (m, 4H, S–CH2CH2, COCH2CH2), 1.42–1.34 (m,
2H, S–CH2CH2CH2), 1.25 (br s, 26H, 13CH2), 0.88 (t, 6H, J = 6.8 Hz,
ASAH2 in lipofectamine 2000 (10 lL, 0.4 lg DNA/lL). Twenty-four
hours after transfection cells were collected by trypsinization and
washed twice with phosphate buffered-saline (PBS). The cell pellets
were suspended in 1 mM Tris–HCl, 0.1 mM ethylenediaminetetra-
acetic acid (EDTA), pH 7.5 (100
sonication (10 min, ultrasonic ice-cold bath), and incubated at
37 °C for 1 h with 5 M CerC12NBD in reaction buffer (25 mM
Tris–HCl, pH 7.5, 1% sodium cholate) with or without the test com-
pounds (50 M final concentration). The reaction was terminated by
the addition of methanol (1 mL) and 50 L were injected into a high
lL, 5–6 mg protein/mL), lysed by
2CH3). 13C NMR (101 MHz, CDCl3):
d 173.7 (CO), 128.7
l
(CH@CH–S), 123.0 (CH@CH–S), 65.7 (CH2OH), 53.8 (CHNH2), 37.0
(COCH2), 32.4 (SCH2), 32.0, 31.8, 29.8, 29.8, 29.8, 29.6, 29.5, 29.4,
29.3, 29.3, 29.3, 29.1, 29.0, 25.9 (14CH2), 22.8, 22.7 (2CH2CH3),
l
l
14.3, 14.2 (2CH3). ½aꢂD = ꢀ3.1 (c 0.90, CHCl3). HRMS: calculated
performance liquid chromatography with a fluorescence detector
(HPLC-FD).
To determine nCDase inhibition in intact cells, 24 h prior to
transfection, 2.5x105 cells per well were plated in 6 well-plates.
Then cells were transfected with pcDNA5/TO (empty vector) or
for C25H50NO2S (M+1)+: 428.3562; Found: 428.3560. Anal. Calcd
for C25H49NO2S: C, 70.20; H, 11.55; N, 3.27. Found: C, 70.23; H,
11.58; N, 3.31.
5.1.2. Compound RBM2-1C (73% yield)
pcDNA5/TO-ASAH2 with lipofectamine 2000 (10
lL, 0.4
lg DNA/
1H NMR (400 MHz, CDCl3): d 7.72 (d, 2H, J = 8.4 Hz, Ar-H), 7.43
(d, 2H, J = 8.4 Hz, Ar-H), 6.59 (d, 1H, J = 7.6 Hz, NHCO), 6.32 (d, 1H,
J = 15.3 Hz, CH@CH–S), 5.56 (dd, 1H, J1 = 15.3, J2 = 6.6 Hz, CH@CH–
S), 4.80–4.72 (m, 1H, CHNH), 3.76 (qd, 2H, J = 11.1, 4.5 Hz, CH2OH),
2.66 (t, 2H, J = 7.4 Hz, S–CH2), 1.60 (quint, 2H, J = 7.0 Hz,
S–CH2CH2), 1.41–1.34 (m, 2H, S–CH2CH2CH2), 1.32 (s, 9H, 3CH3),
1.25 (br s, 18H, 9CH2), 0.87 (t, 3H, J = 6.8 Hz, CH3). 13C NMR
(101 MHz, CDCl3): d 167.6 (CO), 155.4, 131.3, 128.9 (CH@CH–S),
127.0, 125.6, 122.9 (CH@CH–S), 65.6 (CH2OH), 54.2 (CHNH2), 35.1
(C(CH3)3), 32.4 (SCH2), 32.0 (CH2), 31.3 (C(CH3)3), 29.8, 29.8, 29.7,
29.6, 29.5, 29.3, 29.3, 29.0, (9CH2), 22.8 (CH2CH3), 14.3 (CH3).
lL). Twenty-four hours after transfection cells were treated with
the test compounds (50 lM) or vehicle for 4 h. Cells were collected,
lysed and nCDase activity was determined as described above.
Des1 activity was determined in both A549 and HCT116 cells,
which were seeded at a density of 2x105 cells/well in 6 well-plates.
Twenty-four hours later, the medium was removed and fresh med-
ium containing 16 lM each of both substrate and test compound
was added. After incubation for 4 h, both media and cells were col-
lected separatedly. One milliliter of MeOH was added to the med-
ium and 25
trypsinized (trypsin–EDTA, 0.4 mL), media (0.6 mL) and methanol
(1 mL) were sequentially added and 100 L were injected into
lL were injected into the HPLC-FD. Cells were
½
a D = +2.1 (c 1.25, CHCl3). HRMS: calculated for C28H48NO2S
ꢂ
l
(M+1)+: 462.3406; Found: 462.3400. Anal. Calcd for C28H47NO2S:
C, 72.83; H, 10.26; N, 3.03. Found: C, 72.84; H, 10.28; N, 3.01.
the HPLC-FD system. Equipment and analysis conditions were as
detailed previously.7 Compound XM4627 was used as positive con-
trol of Des1 inhibition.
5.1.3. Compound RBM2-1D (84% yield)
1H NMR (400 MHz, CDCl3): d 6.24 (d, 1H, J = 16.0 Hz, CH@CH–S),
5.95 (d, 1H, J = 6.8 Hz, NHCO), 5.47 (dd, 1H, J1 = 15.3, J2 = 6.7 Hz,
CH@CH–S), 4.57–4.51 (m, 1H, CHNH), 3.75–3.72 (m, 1H, CH2OH),
3.70–3.62 (m, 1H, CH2OH), 2.75 (br s, 1H, CH2OH), 2.66 (t, 2H,
J = 7.4 Hz, S–CH2), 1.61 (quint, 2H, J = 7.4 Hz, S–CH2CH2), 1.40–
1.33 (m, 2H, S–CH2CH2CH2), 1.25 (br s, 18H, 9CH2), 1.21 (s, 9H,
C(CH3)3), 0.87 (t, 3H, J = 6.8 Hz, CH3). 13C NMR (101 MHz, CDCl3):
d 179.0 (CO), 128.7 (CH@CH–S), 123.0 (CH@CH–S), 65.9 (CH2OH),
53.8 (CHNH2), 38.9 (C(CH3)3), 32.4, 32.0, 29.8, 29.8, 29.7, 29.6,
29.5, 29.3, 29.3, 29.0 (10CH2), 27.7 (C(CH3)3), 22.8 (2CH2CH3),
5.4. Lipidomics
Sphingolipid analysis by UPLC-TOF was carried out following
the standard protocol used in our laboratories.39 Long chain bases
and long chain base phosphates were analyzed with a system con-
sisting of a Waters Alliance 2690 liquid chromatography pump
equipped with an autosampler and connected to a Quattro LC tri-
ple-quadrupole mass spectrometer from Micromass (Manchester,
UK). Separation was achieved on a Purospher STAR-RP-18 column
(125 ꢁ 2 mm, 5
lm) (Merck, Darmstadt) using a gradient of two
14.3 (CH3). ½aꢂD = ꢀ2.3 (c 1.73, CHCl3). HRMS: calculated for
mobile phases: A, methanol/water/formic acid (74/25/1 v/v/v); B,
C
22H43NO2S (M+1)+: 386.3093; Found: 386.3090. Anal. Calcd for
methanol/formic acid (99/1 v/v), both also contained 5 mM ammo-

L. Camacho et al. / Bioorg. Med. Chem. 20 (2012) 3173–3179
3179
3. Houben, E.; Uchida, Y.; Nieuwenhuizen, W. F.; De Paepe, K.; Vanhaecke, T.;
Holleran, W. M.; Rogiers, V. Skin Pharmacol. Physiol. 2007, 20, 187.
4. Spinedi, A.; Di Bartolomeo, S.; Piacentini, M. Biochem. Biophys. Res. Commun.
1999, 255, 456.
5. Bedia, C.; Canals, D.; Matabosch, X.; Harrak, Y.; Casas, J.; Llebaria, A.; Delgado,
A.; Fabrias, G. Chem. Phys. Lipids 2008, 156, 33.
6. Grijalvo, S.; Bedia, C.; Triola, G.; Casas, J.; Llebaria, A.; Teixido, J.; Rabal, O.;
Levade, T.; Delgado, A.; Fabrias, G. Chem. Phys. Lipids 2006, 144, 69.
7. Munoz-Olaya, J. M.; Matabosch, X.; Bedia, C.; Egido-Gabas, M.; Casas, J.;
Llebaria, A.; Delgado, A.; Fabrias, G. ChemMedChem 2008, 3, 946.
8. Fabrias, G.; Munoz-Olaya, J.; Cingolani, F.; Signorelli, P.; Casas, J.; Gagliostro, V.;
Ghidoni, R. Prog. Lipid Res. 2011, 51, 82.
9. Zheng, W.; Kollmeyer, J.; Symolon, H.; Momin, A.; Munter, E.; Wang, E.; Kelly,
S.; Allegood, J. C.; Liu, Y.; Peng, Q.; Ramaraju, H.; Sullards, M. C.; Cabot, M.;
Merrill, A. H., Jr. Biochim. Biophys. Acta 1864, 2006, 1758.
10. Schulz, A.; Mousallem, T.; Venkataramani, M.; Persaud-Sawin, D. A.; Zucker, A.;
Luberto, C.; Bielawska, A.; Bielawski, J.; Holthuis, J. C.; Jazwinski, S. M.;
Kozhaya, L.; Dbaibo, G. S.; Boustany, R. M. J. Biol. Chem. 2006, 281, 2784.
11. Kraveka, J. M.; Li, L.; Szulc, Z. M.; Bielawski, J.; Ogretmen, B.; Hannun, Y. A.;
Obeid, L. M.; Bielawska, A. J. Biol. Chem. 2007, 282, 16718.
12. Valsecchi, M.; Aureli, M.; Mauri, L.; Illuzzi, G.; Chigorno, V.; Prinetti, A.;
Sonnino, S. J. Lipid Res. 1832, 2010, 51.
13. Illuzzi, G.; Bernacchioni, C.; Aureli, M.; Prioni, S.; Frera, G.; Donati, C.; Valsecchi,
M.; Chigorno, V.; Bruni, P.; Sonnino, S.; Prinetti, A. J. Biol. Chem. 2010, 285,
18594.
nium formate. A linear gradient was programmed as follows:
0.0 min, 50% B; 2 min, 50% B; 7 min, 100% B; 17 min, 100% B;
19 min, 50% B; 26 min, 50% B. The flow rate was 0.3 mL/min. MS/
MS detection was performed with an electrospray interface operat-
ing in the positive ion mode acquiring the following selected reac-
tion monitoring transitions: C17-dihydrosphingosine, 288-252,
collision energy 18 eV; sphingosine, 300-264, collision energy
16 eV; dihydrosphingosine, 302-266, collision energy 18 eV; C17-
dihydrosphingosine-1-phosphate, 368-252, collision energy
18 eV; sphingosine-1-phosphate, 380-264, collision energy 16 eV
and dihydrosphingosine-1-phosphate, 382-266, collision energy
18 eV.
5.5. Cell viability
Cells were seeded in 96-well plates at a density of 105 cells/mL.
After 24 h, media were removed and replaced with fresh media
with and without the test compounds at the specified concentra-
tions. The number of viable cells was determined after the specified
times with the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoli-
um bromide (MTT) test.
14. Mao, Z.; Sun, W.; Xu, R.; Novgorodov, S.; Szulc, Z. M.; Bielawski, J.; Obeid, L. M.;
Mao, C. J. Biol. Chem. 2010, 285, 29078.
15. Xu, R.; Jin, J.; Hu, W.; Sun, W.; Bielawski, J.; Szulc, Z.; Taha, T.; Obeid, L. M.; Mao,
C. FASEB J. 1813, 2006, 20.
16. Signorelli, P.; Munoz-Olaya, J. M.; Gagliostro, V.; Casas, J.; Ghidoni, R.; Fabrias,
G. Cancer Lett. 2009, 282, 238.
5.6. Flow cytometry
17. Schiffmann, S.; Sandner, J.; Schmidt, R.; Birod, K.; Wobst, I.; Schmidt, H.;
Angioni, C.; Geisslinger, G.; Grosch, S. J. Lipid Res. 2009, 50, 32.
18. Abad, J. L.; Fabrias, G.; Casas, J.; Garrido, M.; Camacho, L.; Simbari, F.; Delgado,
A. Spain Patent P201031642, 2010.
19. Nieves, I.; Garrido, M.; Abad, J. L.; Delgado, A. Synlett 2010, 2950.
20. Bedia, C.; Camacho, L.; Abad, J. L.; Fabrias, G.; Levade, T. J. Lipid Res. 2010, 51,
3542.
21. Bedia, C.; Casas, J.; Garcia, V.; Levade, T.; Fabrias, G. ChemBioChem 2007, 8, 642.
22. Kono, M.; Dreier, J. L.; Ellis, J. M.; Allende, M. L.; Kalkofen, D. N.; Sanders, K. M.;
Bielawski, J.; Bielawska, A.; Hannun, Y. A.; Proia, R. L. J. Biol. Chem. 2006, 281,
7324.
23. Jiang, Q.; Rao, X.; Kim, C. Y.; Freiser, H.; Zhang, Q.; Jiang, Z.; Li, G. Int. J. Cancer
2012, 130, 685.
Cells are seeded in a 96 well-plate (0.1 mL, 2.5 ꢁ 105 cells/mL)
in low glucose DMEM supplemented with FBS (10%) and penicil-
lin–streptomycin (1%). Twenty-four hours after seeding, medium
was replaced with fresh medium (0.1 mL) containing the test com-
pounds at the specified concentrations (Fig. 5). The same volume of
vehicle (ethanol) was added to controls. After 15 h, cells were
washed twice with PBS (0.2 mL), collected by trypsinization and
stained with the Dead Cell Apoptosis Kit from Invitrogen following
the manufacturer’s instructions. Cells were analyzed with a Guava
easyCyte 8HT (Millipore).
24. Jiang, Q.; Wong, J.; Fyrst, H.; Saba, J. D.; Ames, B. N. Proc. Natl. Acad. Sci. U.S.A.
2004, 101, 17825.
25. Triola, G.; Fabrias, G.; Casas, J.; Llebaria, A. J. Org. Chem. 2003, 68, 9924.
26. Triola, G.; Fabrias, G.; Llebaria, A. Angew. Chem., Int. Ed. Engl. 1960, 2001, 40.
27. Flowers, M.; Fabrias, G.; Delgado, A.; Casas, J.; Abad, J. L.; Cabot, M. C. Breast
Cancer Res. Treat. 2011. http://dx.doi.org/10.1007/s10549-011-1768-8.
28. Gouazé-Andersson, V.; Flowers, M.; Karimia, R.; Fabriás, G.; Delgado, A.; Casas,
J.; Cabot, M. C. Prostate 2011, 71, 1064.
29. Antoon, J. W.; Liu, J.; Ponnapakkam, A. P.; Gestaut, M. M.; Foroozesh, M.;
Beckman, B. S. Cancer Chemother. Pharmacol. 2010, 65, 1191.
30. Antoon, J. W.; Liu, J.; Gestaut, M. M.; Burow, M. E.; Beckman, B. S.; Foroozesh,
M. J. Med. Chem. 2009, 52, 5748.
31. Bielawska, A.; Bielawski, J.; Szulc, Z. M.; Mayroo, N.; Liu, X.; Bai, A.; Elojeimy, S.;
Rembiesa, B.; Pierce, J.; Norris, J. S.; Hannun, Y. A. Bioorg. Med. Chem. 2008, 16,
1032.
32. Ponnusamy, S.; Meyers-Needham, M.; Senkal, C. E.; Saddoughi, S. A.; Sentelle,
D.; Selvam, S. P.; Salas, A.; Ogretmen, B. Future Oncol. 2010, 6, 1603.
33. Hu, W.; Xu, R.; Sun, W.; Szulc, Z. M.; Bielawski, J.; Obeid, L. M.; Mao, C. J. Biol.
Chem. 2010, 285, 7964.
Acknowledgments
Financial support from the Ministerio de Ciencia e Innovación
(Projects SAF2008-00706 and SAF2011-22444) and Generalitat de
Catalunya (SGR 2009 1072), and predoctoral fellowships from Gen-
eralitat de Catalunya (SGR 2005 01063 to L.C.), Ministerio de Edu-
cación y Ciencia (F.S.) and the Spanish Council for Scientific
Research (M.G.) are acknowledged. We thank Dr. Meritxell Egido-
Gabás, Pedro Rayo and Eva Dalmau for their excellent technical
assistance and Dr. Carmen Bedia for kindly providing the
pcDNA5/TO-ASAH2 construct.
Supplementary data
34. Morselli, E.; Tasdemir, E.; Maiuri, M. C.; Galluzzi, L.; Kepp, O.; Criollo, A.;
Vicencio, J. M.; Soussi, T.; Kroemer, G. Cell Cycle 2008, 7, 3056.
35. Sims, K.; Fluke, K.; Symolon, H.; Zheng, W.; Munter, E.; Momin, A.; Pack, C.;
Haynes, C.; Kelly, S.; Allegood, J.; Wang, E.; Merrill, A. H. FASEB J. 2007, 21,
779.6. (Abstract A976).
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2012.03.073.
36. Villorbina, G.; Canals, D.; Carde, L.; Grijalvo, S.; Pascual, R.; Rabal, O.; Teixido, J.;
Fabrias, G.; Llebaria, A.; Casas, J.; Delgado, A. Bioorg. Med. Chem. 2006, 15, 50.
37. Medin, J. A.; Takenaka, T.; Carpentier, S.; Garcia, V.; Basile, J. P.; Segui, B.;
Andrieu-Abadie, N.; Auge, N.; Salvayre, R.; Levade, T. Hum. Gene Ther. 1999, 10,
1321.
38. Tani, M.; Okino, N.; Mitsutake, S.; Ito, M. J. Biochem. 1999, 125, 746.
39. Canals, D.; Mormeneo, D.; Fabrias, G.; Llebaria, A.; Casas, J.; Delgado, A. Bioorg.
Med. Chem. 2009, 17, 235.
References and notes
1. Gangoiti, P.; Camacho, L.; Arana, L.; Ouro, A.; Granado, M. H.; Brizuela, L.; Casas,
J.; Fabrias, G.; Abad, J. L.; Delgado, A.; Gomez-Munoz, A. Prog. Lipid Res. 2010, 49,
316.
2. Liu, X.; Elojeimy, S.; Turner, L. S.; Mahdy, A. E.; Zeidan, Y. H.; Bielawska, A.;
Bielawski, J.; Dong, J. Y.; El-Zawahry, A. M.; Guo, G. W.; Hannun, Y. A.; Holman,
D. H.; Rubinchik, S. Front. Biosci. 2008, 13, 2293.