Novel fluorescent ceramide derivatives for probing ceramidase substrate specificity

Article abstract of DOI:10.1016/j.bmc.2012.08.035

Ceramidases are key regulators of cell fate. The biochemistry of different ceramidases and of their substrate ceramide appears to be complex, mainly due to specific biophysical characteristics at the water-membrane interface. In the present study, we describe the design and synthesis of a set of fluorescently labeled ceramides as substrates for acid and neutral ceramidases. For the first time we have replaced the commonly used polar NBD-dye with the lipophilic Nile Red (NR) dye. Analysis of kinetic data reveal that although both the dyes do not have any noticeable preference for the substitution at acyl or sphingosine (Sph) part in ceramide towards hydrolysis by acid ceramidase, the ceramides with acyl-substituted NBD and Sph-substituted NR dyes have been found to be a better substrate for neutral ceramidase.

Full text of DOI:10.1016/j.bmc.2012.08.035

Bioorganic & Medicinal Chemistry 20 (2012) 6154–6161
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Novel fluorescent ceramide derivatives for probing ceramidase
substrate specificity
⇑
Krishna P. Bhabak, Denny Proksch, Susanne Redmer, Christoph Arenz
Humboldt Universität zu Berlin, Institute for Chemistry, Brook-Taylor-Str 2, 12489 Berlin, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Ceramidases are key regulators of cell fate. The biochemistry of different ceramidases and of their sub-
strate ceramide appears to be complex, mainly due to specific biophysical characteristics at the water-
membrane interface. In the present study, we describe the design and synthesis of a set of fluorescently
labeled ceramides as substrates for acid and neutral ceramidases. For the first time we have replaced the
commonly used polar NBD-dye with the lipophilic Nile Red (NR) dye. Analysis of kinetic data reveal that
although both the dyes do not have any noticeable preference for the substitution at acyl or sphingosine
(Sph) part in ceramide towards hydrolysis by acid ceramidase, the ceramides with acyl-substituted NBD
and Sph-substituted NR dyes have been found to be a better substrate for neutral ceramidase.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 25 May 2012
Revised 31 July 2012
Accepted 8 August 2012
Available online 31 August 2012
Keywords:
Sphingolipids
Ceramidases
Fluorescent dyes
1. Introduction
In total, five ceramidases have been described so far.9 The acid
ceramidase with a pH optimum of ꢀ4.0 was first described by Shi-
Ceramide takes a central position in sphingolipid biosynthesis
and metabolism. It is the common lipid anchor for the membrane
constituents like sphingomyelin, gangliosides and other glyco-
sphingolipids (GSLs) (Fig. 1).1 Furthermore, it is the precursor of
the lower sphingolipids involved in signaling events. The biochem-
istry of ceramide appears to be relatively complex, since ceramide
can appear in a variety of structural modifications not only in the
sphingosyl-, but also in its N-acyl part2,3 and not all forms of cera-
mide are equally recognized by the respective metabolic enzymes.
Ceramidases are lipolytic enzymes that cleave the N-acyl moi-
ety of ceramide to form the lysolipid sphingosine.4 Ceramide,
sphingosine and its derivative sphingosine-1-phosphate (S1P) are
signaling lipids triggering contradictory cellular effects. While cer-
amide is involved in inflammation and initiation of apoptosis,
sphingosine and mainly S1P induce cell proliferation and differen-
tiation. Due to this phenomenon, also referred to as the ‘sphingo-
lipid rheostat’,5 the different cellular ceramidases are regarded as
key enzymes for the determination of cell fate. For ceramide’s
mode of action there are two alternative models. According to
the first model, ceramide can interact with some intracellular pro-
teins like ceramide-activated protein phosphatases (CAPP) or
cathepsin D.6 In an alternative model, ceramide acts as an impor-
tant mediator of plasma membrane microenvironment enabling
receptor clustering within liquid ordered membrane domains,7
also referred to as ‘lipid rafts’.8
mon Gatt10 and purified from urine,11 cloned12 and biochemically
characterized13 by the groups of Sandhoff and Schuchman. This en-
zyme is mainly localized to the lysosomes, but a portion is secreted
extracellularly. A neutral ceramidase was first described in 196914
and further described by Stoffel et al.15The latter was cloned from
human (Hannun)16 and mouse (Ito).17 It mainly localizes to the
plasma membrane and mitochondriae and has an optimal pH be-
tween pH 7 and pH 9. For this enzyme, investigation of substrate
specificity revealed that, among the four possible diastereomers
only the natural D-erythro ceramide is cleaved by this enzyme.18
In contrast, dihydroceramide, phytoceramide or analogues with
shortened alkyl chains were not efficient substrates. Beside these
two ceramidases, alkaline ceramidase activity with an pH optimum
at ꢀ9 has been described.19 Alkaline ceramidases are encoded by
three different genes, ACER1,20 ACER221 and ACER3.22
Mutations in the gene for acid ceramidase can lead to a defi-
ciency in enzyme activity and a lysosomal storage disease, termed
Farber’s disease.23 Lang and co-workers have described that an
important hallmark of cystic fibrosis is an imbalance between aSM-
ase and aCDase activities caused by an elevated lysosomal pH.24
While the acid sphingomyelinase, having an activity optimum
around pH 5, is still active, the aCDase activity at this pH is signifi-
cantly reduced. This imbalance leads to an accumulation of cera-
mide and an increased cell death in lung epithelial cells. Elevated
activity of acid ceramidase in contrast, has been observed in a num-
ber cancer types and has been related to suppression of apoptotic
pathways.25 Accordingly, pharmacological inhibition of ceramidas-
es should also elevate cellular ceramide levels and lead to a pre-
domination of ceramide-associated effects over those mediated
⇑
Corresponding author. Fax: +49 30 2093 6947
E-mail address: christoph.arenz@chemie.hu-berlin.de (C. Arenz).
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.08.035

K. P. Bhabak et al. / Bioorg. Med. Chem. 20 (2012) 6154–6161
6155
by sphingosine or S1P.26 Indeed, some moderately active inhibitors
GSL degradation
Sphingomyelin
de novo Biosynthesis
of the acid ceramidase have been reported with pro-apoptotic and
anti-proliferative effects in several cancer models.9 Among the
most prominent inhibitors are B-1327 and N-oleoylethanolamine
(NOE)19 for acid ceramidase and D-erythro-N-myristoyl-2-amino-
1-phenyl-1-propanol (D-e-MAPP) for alkaline ceramidases.27,28 All
these compounds have given rise to the development of various
derivatives.9,29
O
CH3
CH3
HN
HO
Ceramide
OH
In order to gain a deeper insight into ceramidase biochemistry,
our goal is to develop fluorescent ceramide analogues as in situ
probes for intracellular ceramidase activities. Such a goal can be
achieved by the development of FRET probes with suitably substi-
tuted fluorescent moieties such as Nile Red (NR) and NBD in a sin-
gle ceramide molecule. This may enable it for real time ratio
imaging and thus in situ quantification of enzymatic activities. In
contrast to quenched fluorescent probes, real FRET probes provide
their own internal standard that allow for determining the ratio of
cleaved versus uncleaved probes at any time point, reflected by the
ratio of donor fluorescence to acceptor fluorescence.
Ceramidases
NH2
HO
CH3
CH3
Sphingosine
OH
O
+
HO
Stearic Acid
Figure 1. Schematic representation for the catalytic hydrolysis of ceramide by
ceramidases.
In the special case of ceramidases, the specific enzyme activity
is relatively low. In addition, in the case of the acid ceramidase,
high substrate specificity has been reported,18 but objective data
on structure–activity-relationship (SAR) of different ceramides as
the substrates of ceramidases is limited. Thus, a more detailed
SAR study on ceramides is necessary for a better understanding
of substrate specificity toward ceramidases. It should be men-
tioned that the determination of ceramidase activity was per-
formed previously by using natural ceramide as substrate and
the hydrolysis was quantified by the fluorescent emission intensity
of produced sphingosine upon its derivatization with o-phthalalde-
hyde (OPA).11 The ceramidase assays today are more conveniently
conducted either by using fatty acid 14C or 3H-labeled ceramides as
substrates followed by extraction or TLC-separation or alterna-
tively by using the commercially available fatty acid substituted
ceramides were tested as substrates for the recombinant acid
and neutral ceramidases in micellar assays. In addition, detailed ki-
netic studies are performed with all the substrates for the determi-
nation of different kinetic parameters such as Michaelis constant
(KM), maximum velocity (Vmax), catalytic constant (kcat) and the ra-
tio of kcat and KM, specificity constant (g) to understand their suit-
able positional selectivity. (See Fig. 2)
2. Results and discussion
In order to make use of the positive results obtained by Schultz
and Co-workers with the NR-modified phosphoglycerolipids as
substrates of phospholipase A2 (PLA2),38,39 we planned to synthe-
size NR-labeled ceramides and to evaluate their properties as cer-
amidase substrates in comparison to the commercially available
NBD-labeled ceramides. Although NR is a hydrophobic dye with a
selectivity to lipids and membranes,40 to the best of our knowledge
no NR-labeled sphingolipids have been reported so far. In contrast
to NR, NBD is a relatively polar dye, which only reluctantly inte-
grates into membranes.41,42 Unlike pyrene dyes, which are very
hydrophobic but under most conditions are only poorly fluorescing
molecules, NR has its highest quantum yield in nonpolar solvents
or lipid environments. These features suggest that NR-labeling
may be favored even for singly labeled ceramidase substrates. In
order to verify this hypothesis, we synthesized a set of three cera-
mide analogues, with the incorporation of either NBD- or NR-label-
ing in the fatty acid or in the sphingosyl part of the ceramide
molecule (Fig. 2). The set was completed for the SAR studies with
the inclusion of acyl-NBD-Cer, commercially obtained from Avanti
Polar Lipids Inc.
In order to guarantee a modular synthesis of sphingosyl-modi-
fied ceramides with a minimum number of synthetic steps, we fol-
lowed a strategy developed by Nussbaumer et al employing olefin
metathesis as a key step.43,44 Starting from Garner’s aldehyde 4
(Scheme 1), the allyl alcohol 5 was synthesized using vinylmagne-
sium bromide.45 Subsequently, the building block 5 was subjected
to olefin metathesis reaction using the 2nd generation Hoveyda-
Grubbs catalyst to form either protected sphingosine 6 or the
NR- or NBD-modified compounds 7a or 7b, respectively. Acidic
deprotection yielded natural sphingosine 8 or the NR- or NBD-
modified sphingosines 9a and 9b. Finally, acylation of the sphingo-
sine derivatives led to the formation of fluorescent ceramides 1, 3a
and 3b (Scheme 1). Probe 1 was synthesized by the reaction of
sphingosine 8 with NR-acyl-N-oxysuccinimide 11 (Scheme 2).
ceramides such as
x
-hexanoyl or dodecanoyl-labeled NBD-cera-
-dodecanoyl-labeled
mides,30 or the somewhat less common
x
BODIPY-ceramides.31 Tani et al have reported the detailed kinetic
studies for the hydrolysis of acyl-NBD-C12-Cer using different cel-
lular lysates of alkaline, neutral as well as acid ceramidases to
understand the selectivity of the substrate towards a particular en-
zyme sub-type.32 They observed that acyl-NBD-C12-Cer was much
more selective towards alkaline and neutral ceramidases but was
found to be relatively poor substrate for aCDase. However, cera-
mide analogues labeled in the sphingosine part are relatively rare.
Long before, Sandhoff and co-workers have developed fluorescent
sphingosine analogues bearing diphenylhexatrienyl (DPH) fluoro-
phore and subsequently synthesized C6-DPH-Cer. The ceramide
analogue was studied for their incorporation behavior towards cul-
tured human fibroblasts, which was similar to that of C6-NBD-
Cer.33 Nieuwenhuizen et al have described a quenched fluorescent
probe for bacterial ceramidase, in which the fatty acid was labeled
by a fluorescent pyrene and the sphingosine part was substituted
with a quenching 3,5-dinitrobenzoyl group.34 Bedia et al have re-
ported ceramidase probes in which a coumarin dye is attached to
the sphinganine part of a dihydroceramide derivative.35,36 This
probe can be used for homogenous high-throughput assays,37 but
a treatment of post-reaction periodate is necessary for the libera-
tion of the fluorescent hydroxycoumarin, making it incompatible
with real-time or in situ assays. A detailed structure-activity stud-
ies with the variation of fatty acid chain length revealed that the
cleavage rate strongly depends on fatty acid length with an opti-
36
mum at C12
.
In the present study we describe the straightforward synthesis
for the development of two novel Nile Red-substituted fluorescent
ceramide analogues. Together with the commercially available
acyl-NBD-Cer and the literature-known Sph-NBD-Cer, these

6156
K. P. Bhabak et al. / Bioorg. Med. Chem. 20 (2012) 6154–6161
OH
O
OH
O
NO2
HO
HO
C13H27
HO
C13H27
HN
HN
HN
N
H
O
O
N
7
7
O
N
N
O
Acyl-NR-Cer (1)
Acyl-NBD-Cer (2)
NEt2
O
N
OH
NO2
OH
O
O
8
HO
N
N
N
O
C15H31
8
H
HN
C15H31
O
NEt2
O
Sph-NR-Cer (3a)
Sph-NBD-Cer (3b)
Figure 2. Chemical structures of the fluorescent ceramide derivatives evaluated in this study: The sphingosine-labeled derivatives 3a and 3b and the acyl-NR-Cer analogue 1
as well as the commercially available acyl-NBD-Cer (2).
O
OH
OH
OH
iii
1
3a
3b
i
ii
iv
R
O
O
O
HO
R
N
N
N
NH2
R = -C13H27
9a R = -C8H16-O-NR
9b R = -C8H16-NH-NBD
Boc
Boc
Boc
4
6
R = -C13H27
7a R = -C8H16-O-NR
7b R = -C8H16-NH-NBD
8
5
Scheme 1. Synthetic routes to fluorescent-labeled ceramide analogues. Reagents and conditions: (i) Vinylmagnesium bromide, THF, À78 °C, 1.5 h; (ii) For 6: 1-Pentadecene,
Hoveyda-Grubb’s catalyst, CH2Cl2, 12 h, 40 °C; for 7a: 12, Hoveyda-Grubb’s catalyst, CH2Cl2, 12 h, 40 °C; for 7b: 15, Hoveyda-Grubb’s catalyst, CH2Cl2, 12 h, 40 °C; (iii) TFA,
CH2Cl2, 45–180 min; (iv) for 1: 8, 11, DIPEA, CHCl3, 40 h; for 3a: 9a, palmitic acid, EDC, HOBT, DIPEA, CH2Cl2; for 3b: 9b, palmitoyl chloride, pyridine, 1 h.
O
O
CO2R
( )10
OH
O
O
N
( )10
i
O
ii
N
O
N
O
O
NR-OH
iii
N
N
O
O
N
10,
R = Me
10a, R = H
O
O
11
N
( )8
13
N3
( )8
O
iv
NO2
N
v
( )8
N
H
15
N
O
( )8
14
NH2
N
O
O
12
N
Scheme 2. Synthetic routes to functionalized fluorescent dyes. Reagents and conditions: (i) (a) Methyl-12-bromo-dodecanoate, K2CO3, DMF, 80 °C, 4 h; (b) Me3SiOK, Et2O,
48 h; (ii) N-Hydroxysuccinimide, D-MAP, DCC, CHCl3, 0 °C-RT, 12 h; (iii) 11-Bromo-1-undecene, NaH, THF, reflux, 72 h; (iv) LiAlH4, THF, 0 °C, 30 min; (v) NBD-Cl, DIPEA,
MeOH, 15 h.
The NR-sphingosine 9a was coupled to palmitic acid using EDC/
HOBT yielding 3a and NBD-sphingosine 9b was reacted with pal-
mitoyl chloride to form 3b.
12 for metathesis reaction was formed by Williamson ether syn-
thesis with 11-bromo-1-undecene. The NBD-alkene 15 was syn-
thesized by reaction of the amine 14 with NBD chloride.47
The conjugated dyes were synthesized as shown in Scheme 2.
The precursor hydroxyl-NR (NR-OH) was synthesized as described
previously46 and coupled to methyl-12-bromo-dodecanoate in the
presence of potassium carbonate to yield 10. The methyl ester was
then hydrolyzed via silyl ester formation. The respective carboxylic
acid (10a) was isolated and purified and then reacted with N-
hydroxysuccinimide to yield the activated ester 11. The NR-alkene
Before carrying out the detailed experiments for the determina-
tion of kinetic parameters, we optimized the assay conditions with
the variation of important parameters such as the incubation time
and enzyme concentrations (Supplementary data). Incubation
times of 150 min and 60 min for aCDase and nCDase, respectively
were considered for all the kinetic studies as linear increase of
hydrolysis by ceramidases was observed for all the substrates with

K. P. Bhabak et al. / Bioorg. Med. Chem. 20 (2012) 6154–6161
6157
fixed concentrations of substrates and enzymes. Furthermore, the
concentration of both aCDase (50 g/mL) and nCDase (20 ng/mL)
was optimized by the variation of enzyme concentrations at fixed
concentrations of substrates (20 M) using the optimized incuba-
the best parameter for determining the relative specificity of NR-
and NBD-ceramides toward aCDase and nCDase in the present
study.
l
l
As shown in Table 1, the specificity constants of all the NR-Cer
and NBD-Cer were almost comparable for the hydrolysis by
aCDase, although a 1.3 times enhancement of specificity of acyl-
NR-Cer has been observed than Sph-NR-Cer, the corresponding
NBD-based ceramides did not show any positional selectivity.
However, a more pronounced effect is observed for the hydrolysis
of ceramide analogues by nCDase. For example, around 1.4 times
enhancement in the specificity has been observed upon the incor-
poration of NR group from acyl part (1) to sphingosine part (3a) of
ceramide. On the other hand, the specificity of acyl-NBD-Cer (2) is
found to be almost double of Sph-NBD (3b) analogue of ceramide.
These observations clearly suggest that incorporation of NBD-moi-
ety at the acyl part and NR-moiety at the sphingosine part could be
a better choice for the development of a FRET probe as a substrate
of nCDase.
To understand the suitability of these singly fluorescent-labeled
ceramides as substrates for the inhibition studies, we employed
two literature known inhibitors such as B-1327,48 and DP-24a49
for their effects towards aCDase-catalyzed hydrolysis of the sub-
strates in the present study (Fig. 3). As NBD and NR-based cera-
mides have different Vmax and KM values, the substrate
concentrations for the inhibition studies were chosen similar to
their KM values. A noticeable inhibition of aCDase activity was ob-
tion times. The values of these parameters are optimized for the
standard assay method to ensure a linear progress of the hydrolysis
of substrates in the presence of ceramidases. Due to the known
lipophilic properties of NR in contrast to the relatively poor lipo-
philicity of NBD, we expected that the novel NR-modified cera-
mides would better mimic the natural ceramide and thus should
be better substrates for ceramidases. However, when we tested
the modified ceramides as substrates for ceramidases (aCDase
and nCDase) under standard assay conditions, we unexpectedly
observed a significantly lower cleavage rate for acyl-NR-Cer 1 as
compared to the corresponding acyl-NBD-Cer 2 (aCDase: 1, 6%; 2,
10%; nCDase: 1, 3%; 2, 10%). On the other hand, the enzymatic
cleavage of Sph-NR-Cer was found to be almost equal or lower than
the corresponding Sph-NBD-Cer (aCDase: 3a, 6%; 3b, 6%; nCDase:
3a, 4%; 3b, 6%). In the present assay condition, the ceramidase-cat-
alyzed hydrolysis of NBD-substituted ceramides was more or at
least equal to the corresponding NR-based analogues and there
was no case in which the NR-dye was superior to the NBD-dye
with respect to the cleavage rate. However, in order to have more
generalized data and to understand the positional selectivity of
NBD and NR dyes to ceramide, we decided to evaluate dose-depen-
dency and to determine the Michaelis-Menten kinetic parameters
such as KM, Vmax, kcat and the ratio of kcat to KM
(g) values for all
served for all the substrates in the presence of 50 lM concentra-
the substrates with the variation of substrate concentrations at a
fixed concentration of enzymes.
tion of inhibitors. As shown previously for the aCDase-mediated
hydrolysis of natural ceramide,49 DP-24a has been found to be a
better inhibitor than B-13 for the aCDase-catalyzed hydrolysis of
all the NR- and NBD-substituted ceramide analogues.
The Michaelis constant KM is the substrate concentration
needed for half-maximum catalytic velocity and thus represents
the enzyme’s affinity to a given substrate. In contrast, the maxi-
mum catalytic velocity Vmax is a measure for the enzyme’s ability
to cleave a given substrate, assumed that all the existing enzyme
is bound or saturated by substrate. Thus, the ideal substrate should
have low KM and high Vmax values. Evaluation of the kinetic param-
eters however showed in most cases that low KM values were asso-
ciated with low Vmax values or vice versa. In general, KM values and
also Vmax values for NR-ceramides were lower as compared to
NBD-ceramides (Table 1). The only exception from this trend is
provided by the aCDase-mediated cleavage of acyl-modified cera-
mides. To obtain a more generalized comparison, we have further
determined the parameters such as catalytic constant (kcat) and
The limits of NBD-labeled lipids have been extensively dis-
cussed in the report by van Meer and Liskamp. The NBD is a rela-
tively polar dye and thus does not behave like many sphingolipids,
which are thought to mainly concentrate in liquid ordered do-
mains of membranes.42 In the course of searching for an alternative
dye for labeling sphingolipids we concentrated on NR, which has
been shown to readily integrating into lipid membranes.50 In our
study, labeling with NR in most cases generally leads to lower KM
but nonetheless the NBD labeled ceramides are cleaved more effi-
ciently. It may be speculated that this behavior is at least in part an
inherent result of the substrates’ biophysical properties. However,
the behavior of the NR-Cer analogues cannot be predicted as their
entry and distribution in cells has not been studied yet. As shown
in a detailed study by Sandhoff and Co-workers that glucosylcera-
mides with short lipid chains are most efficiently cleaved by the
lysosomal enzyme glucocerebrosidase.51 For long natural chain
lengths, the enzymatic activity drops down to zero, unless a
specificity constant (g) as a ratio of kcat and KM. While kcat value
represents the turn over number of an enzyme for a particular sub-
strate, the specificity constant, the ratio of kcat to KM is useful to
understand the comparative specificity of different substrates to
a particular enzyme active site. Therefore, specificity constant is
Table 1
Determination of percentage hydrolysis and kinetic parameters of fluorescent ceramides as substrates for aCDase or nCDase
Acyl-NR-Cer 1
Acyl-NBD-Cer 2
Sph-NR-Cer 3a
Sph-NBD-Cer 3b
aCDase
nCDase
Hydrolysis (%)⁄
6.1 1.1
7.2
110.3
14.3
1.3
3.3 0.2
10.0
9.9 0.8
4.4
86.9
8.82
1.0
6.3 1.1
3.9
81.8
7.8
1.0
3.9 1.0
5.3
5.6 0.4
12.2
203.6
24.4
1.2
6.2 0.2
31.8
Vmax (pmol minÀ1
)
KM
(lM)
kcat (pmol/min/
l
g of protein) Â 10À2
kcat/KM
(
g
, lit/min/
l
g of protein) Â 10À9
Hydrolysis (%)⁄
9.9 0.7
62.4
Vmax (pmol minÀ1
)
KM
(
l
M)
85.3
0.50
5.9
189.9
3.1
16.4
36.4
0.26
7.3
192.9
1.6
8.2
kcat (pmol/min/ng of protein)
kcat/KM
(g
, lit/min/ng of protein) Â 10À9
⁄Enzymatic hydrolysis of substrates (20
respectively. For the determination of kinetic parameters, the substrate concentration was varied from 5 to 150
l
M) was studied after 150 min for aCDase (60
l
g/mL) and 60 min for nCDase (20 ng/mL) at 37 °C in NaOAc buffer of pH 4.5 and 7.0,
l
M or 5 to 180 M at fixed concentration of enzymes under
l
standard assay condition. All the data shown in the Table represent Mean SD from three independent measurements. All the kinetic parameters are determined from the
Lineweaver-Burk plots and each data point in the plot represents Mean SD from three independent experiments (Supplementary data).

6158
K. P. Bhabak et al. / Bioorg. Med. Chem. 20 (2012) 6154–6161
22.3 lM; Vmax (new/old) = 62.4/4.7 pmol/min. The reason for this
18
16
14
12
10
8
2
1
3a
3b
discrepancy in the Michaelis constant is unclear, but it is very
likely not just due to the different specific activity of the used
enzymes.
In the present study, NR-ceramides exhibited lower rates of
cleavage than that of NBD-ceramides. Determination and detailed
analysis of kinetic parameters reveal the pronounced specificity
of ceramide analogues with NR moiety at the sphingosine part or
NBD group at the acyl part of singly labeled ceramide probes espe-
cially for the hydrolysis by nCDase, while the aCDase does not
show any significant specificity.
One drawback of known fluorescently labeled ceramides like
NBD-Cer or BODIPY-Cer55 is the fact that they do not integrate into
liquid ordered domains or ‘lipid rafts’ of plasma membranes.41,42 In
this regard, the polyene ceramide developed by Thiele and cowork-
ers is most promising.56 However, the latter requires two-photon-
excitation, a technique unavailable in many laboratrories.
Given the higher lipophilicity of our novel ceramides, it will be
interesting to test their behavior in living cells, which might be dif-
ferent from commercially available substrates commonly used for
such studies.
*
*
**
6
**
4
2
0
Figure 3. Hydrolysis of different substrates 1 (80
3b (160 M) by aCDase (30 g/mL) in the presence of 50 l
13 or DP-24a under standard assay condition. Control value represents the
hydrolysis of substrates in the absence of any inhibitors. Each data point in the
plot represents Mean SD from three independent experiments. Statistical signif-
icance of inhibition: ⁄, 0.01 < p < 0.05; ⁄⁄, 0.001 6 p 6 0.01.
l
M), 2 (80
lM), 3a (55 lM) and
M of inhibitors such as B-
l
l
4. Experimental
4.1. Acid ceramidase (aCDase) assays
sphingolipid activator protein (SAP) is present that probably acts as
a ‘liftase’ making the substrate available to the soluble enzyme.52
Also the measurement of membrane lateral forces support the idea
that membrane perturbation is an important hallmark of enzy-
matic reactions at the water lipid interface.53 In micellar assays,
the lateral forces are minimized but nonetheless, high lipophilicity
may still reduce the enzyme’s ability to detach the substrate from
the lipid phase, while NBD is additionally perturbing the mem-
brane surface making NBD-labeled substrates better accessible to
the enzymes. The high Vmax but also high KM values of the NBD
substrates seem to support such a theory. Another important factor
may be the difference in size of NR and NBD moieties and the effec-
tive chain length of NR and NBD-based ceramide analogues. Due to
the bigger size of the NR moiety with four fused ring systems, the
effective chain length is much higher than that of the correspond-
ing NBD analogues. Ettmayer and co-workers have reported the ef-
fect of linker chain length as well as the effect of different
fluorescent dyes in sphingosine derivatives towards the phosphor-
ylation by sphingosine kinases and their cellular distribution.54 In
the present study, instead of having the same aliphatic linker chain
lengths for both of NR- and NBD-based substrates, the overall chain
length are eventually different and therefore it would be worth to
study the effect of chain lengths on their kinetic as well as cellular
behavior.
The recombinant human aCDase was expressed from insect cells
as described previously.57 For details, see supplementary data. The
acid ceramidase assay was performed in sodium acetate buffer
(200 mM, pH 4.5) with 200 mM sodium chloride and 0.1% Triton-
X100. The stock solutions of the fluorescent substrates and all the
inhibitors were prepared in DMSO. Human recombinant acid cer-
amidase (aCDase) was used as the enzyme source for the cleavage
of the fluorescent substrates. The total volume of the assay mixture
was 100
used to study the effect of incubation times and enzyme concentra-
tions (50 g/mL). Different substrate concentrations (1 and 2:
80 M; 3a: 55 M; 3b: 160 M) were used for the inhibition stud-
ies with 50 M of inhibitors and 30 g/mL of aCDase. The substrate
lL. A final concentration of 20 lM of all the substrates was
l
l
l
l
l
l
concentration was varied at fixed concentration of enzyme for the
determination of kinetic parameters. In all the assay mixtures the
volume of DMSO was maintained less than 5% (v/v). The assay mix-
ture was incubated for 150 min at 37 °C. The enzymatic reaction
was quenched by adding 200
ture. The assay mixture was vortexed well and centrifuged to sep-
arate the organic and aqueous layers. 30 L of the organic layer
lL of chloroform/methanol (2:1) mix-
l
was transferred into new sets of vials and the total amount was
spotted on non-fluorescent silica gel TLC plates pre-coated on alu-
minium sheets (Mobile phase: For 1 and 2: Cyclohexane/Ethyl ace-
tate/Acetic acid = 40:60:2; 3a: Ethyl acetate/Acetic acid = 50:1; 3b:
Dichloromethane/Methanol/Acetic acid = 20:1:1). The fluorescent
spots of the cleaved and uncleaved substrates on the TLC plates
were detected by Fluorescent Imaging System (Kodak Image Station
4000 MM PRO) using 550 nm and 600 nm as excitation and emis-
sion wave lengths, respectively for the Nile-red linked substrates
and 430 nm and 550 nm as excitation and emission wave lengths,
respectively for the NBD-linked substrates. The percentage ceram-
idase activity and the inhibition were determined by the quantifica-
tion of the relative fluorescent intensity of the cleaved and
uncleaved substrates, which both together were set as 100%. The
reaction rate was calculated from the percentage conversion with
the known concentration of substrates and the reaction time using
the quantification software in fluorescent imaging system. Within
the dynamic range of the detector less than 0.5% substrate cleavage
were detectable. No further calibration was applied.
3. Conclusion
In the present study, for the first time, we report the design and
synthesis of fluorescent ceramide derivatives labeled with the li-
pid-selective dye Nile Red (NR). A detailed comparative SAR study
has been performed with a combined set of singly labeled NR- and
NBD-ceramide analogues for their specificity as substrates toward
aCDase and nCDase. In contrast to most other studies, exclusively
purified recombinant enzymes were used. For the known and com-
mercially available acyl-NBD-Ceramide 2, the obtained enzymatic
data (new) for aCDase matches quite well with a previous report
(old) by Tani et al.:32 KM (new/old) = 86.9/59.0
lM; Vmax (new/
old) = 4.4/0.8 pmol/min. taking into account that Vmax very much
depends on the specific activity of the used enzyme. For nCDase,
however the values significantly differ: KM (new/old) = 189.9/

K. P. Bhabak et al. / Bioorg. Med. Chem. 20 (2012) 6154–6161
6159
4.2. Neutral ceramidase (nCDase) assays
4.5. Synthesis of 10a
Human recombinant neutral ceramidase (nCDase) (R&D Sys-
tems GmbH) was used as the enzyme source for the cleavage of
the fluorescent substrates. The neutral ceramidase assay was per-
formed in sodium acetate buffer (200 mM, pH 7.0) with 200 mM
sodium chloride and 0.1% Triton-X100. The stock solutions of the
fluorescent substrates were prepared in DMSO. The total volume
To a stirred suspension of 10 (0.20 g, 0.36 mmol) in dry Et2O
(10 mL) was added an excess amount of potassium trimethylsila-
nolate (0.70 g, 5.49 mmol) under argon atmosphere. The mixture
was stirred at room temperature for 2 days. The solvent was evap-
orated and the crude mixture was purified by silica gel column
chromatography. The non-polar impurities and unreacted reac-
tants were eluted with ethyl acetate and cyclohexane mixture
and the product was eluted with 15% methanol in ethyl acetate.
The solvent was evaporated to afford the product as deep red solid.
Yield: 0.11 g (54.8%). Rf: 0.34 (100% Ethyl acetate). 1H NMR (CDCl3,
300 MHz, ppm): d = 1.24–1.30 (m, 18H), 1.46–1.54 (m, 2H), 1.59–
1.70 (m, 2H), 1.81–1.88 (m, 2H), 2.34 (t, J = 7.4 Hz, 2H), 3.47 (q,
J1 = 6.3 Hz, J2 = 13.4 Hz, 4H), 4.16 (t, J = 6.6 Hz, 2H), 6.31 (s, 1H),
6.45 (d, J = 2.5 Hz, 1H), 6.65 (dd, J1 = 2.6 Hz, J2 = 9.0 Hz, 1H), 7.16
(dd, J1 = 2.4 Hz, J2 = 8.8 Hz, 1H), 7.60 (d, J = 9.1 Hz, 1H), 8.04 (d,
J = 2.4 Hz, 1H), 8.21 (d, J = 8.8 Hz). 13C NMR (CDCl3, 75 MHz,
ppm): 12.6, 24.8, 26.0, 29.1, 29.2, 29.3, 29.4, 29.5, 29.7, 34.0,
45.1, 68.4, 96.3, 105.3, 106.6, 109.5, 118.3, 124.7, 125.5, 127.7,
131.0, 134.0, 140.0, 146.8, 150.7, 152.0, 161.9, 183.4. ESI-MS: m/z
calcd for C32H40N2O5: 533.3015 [M+H]+; observed: 533.3010.
of the assay mixture was 100
substrates and enzyme were 20
l
L. The final concentration of the
M and 20 ng/mL, respectively.
l
The substrate concentration was varied at fixed concentration of
enzyme for the determination of kinetic parameters. In all the
assay mixtures the volume of DMSO was maintained less than
5% (v/v). The mixture was incubated for 60 min at 37 °C and the
enzymatic reaction was quenched by adding 200 lL of chloro-
form/methanol (2:1) mixture. The mixture was vortexed thor-
oughly and centrifuged at 10,000 rpm for 2 min to separate the
organic layer. 30 lL of the organic layer was transferred into new
sets of vials and the percentage hydrolysis was determined as
mentioned for the aCDase activities.
4.3. General synthesis
Most of the commercially available chemicals were purchased
from Sigma–Aldrich. The acyl-NBD-Cer 2 was purchased from
Avanti Polar Lipids, Inc. Solvents were freshly distilled whenever
required for the reaction. All the moisture sensitive reactions were
carried out under dry argon atmosphere. Thin layer chromato-
graphic (TLC) studies are performed on pre-coated silica gel 60
F254 on aluminum sheets (Merck KGaA). 1H (500 MHz and
300 MHz) and 13C (125 MHz and 75 MHz) NMR spectra were ob-
tained on a Bruker Avance III 500 MHz and Bruker Avance DPX
300 MHz NMR spectrometers. Chemical shifts are cited with re-
spect to Me4Si as internal (1H and 13C) standard. Mass spectral
studies were carried out on a Hewlett-Packard GCMS 5995-A mass
spectrometer with ESI-MS mode analysis.
4.6. Synthesis of 11
The acid obtained in the first step (10a) was further trans-
formed to 11 in analogy to literature methods:47To a solution of
10a (45.0 mg, 0.084 mmol), N-hydroxysuccinimide (10.7 mg,
0.093 mmol) and a catalytic amount of D-MAP (2.0 mg) was added
DCC (20.9 mg, 0.10 mmol) in chloroform (5 mL) at 0 °C. After 1 h,
the mixture was allowed to warm to room temperature and the
reaction continued overnight. Thin layer chromatography indi-
cated almost complete disappearance of the starting acid 10a.
The solvent was evaporated and the crude mixture was used di-
rectly for the synthesis of 1 without any further purification.
4.7. Synthesis of 1
4.4. Synthesis of 10
Compound 1 was synthesized analogously to methods de-
scribed previously.47 To a solution of 11 (45.0 mg, 0.071 mmol) in
To a stirred solution of NR-OH (0.12 g, 0.35 mmol) and methyl-
12-bromo-dodecanoate (0.12 g, 0.39 mmol) in dry DMF (7 mL) was
added K2CO3 (0.15 g, 1.07 mmol) and the mixture was heated at
80 °C for 4 h. Thin layer chromatography indicated almost com-
plete disappearance of NR-OH and the appearance of a relatively
non-polar spot. The solvent was evaporated under reduced pres-
sure and the residue was diluted with ethyl acetate and washed
with water followed by brine solution. The combined organic ex-
tract was dried over sodium sulfate and the solvent was evapo-
rated to obtain the crude product as deep red solid. The product
was purified by silica gel column chromatography using ethyl ace-
tate and cyclohexane as eluents. Solvent was evaporated to obtain
the product as deep red solid. Yield: 0.17 g (88.1%). Rf: 0.55 (1:1
Ethyl acetate & Cyclohexane). 1H NMR (CDCl3, 300 MHz, ppm):
d = 1.22 (t, J = 7.0 Hz, 6H), 1.29 (br, 12H), 1.47–1.52 (m, 2H),
1.58–1.63 (m, 2H), 1.81–1.86 (m, 2H), 2.29 (t, J = 7.5 Hz, 2H), 3.40
(q, J = 7.0 Hz, 4H), 3.65 (s, 3H), 4.12 (t, J = 6.4 Hz, 2H), 6.23 (s,
1H), 6.36 (d, J = 2.7 Hz, 1H), 6.57 (dd, J1 = 2.7 Hz, J2 = 9.1 Hz, 1H),
7.11 (dd, J1 = 2.6 Hz, J2 = 8.7 Hz, 1H), 7.51 (d, J = 9.0 Hz, 1H), 7.96
(d, J = 2.5 Hz, 1H), 8.16 (d, J = 8.7 Hz). 13C NMR (CDCl3, 75 MHz,
ppm): 12.6, 24.9, 26.1, 29.1, 29.2, 29.3, 29.4, 29.5, 29.6, 34.1,
45.0, 51.4, 68.3, 96.2, 105.2, 106.5, 109.4, 118.2, 124.6, 125.4,
127.6, 131.0, 134.0, 139.8, 146.7, 150.6, 151.9, 161.8, 174.3,
183.2. ESI-MS: m/z calcd for C33H42N2O5: 547.3166 [M+H]+; ob-
served: 547.3165.
chloroform (5 mL) was added
a solution of sphingosine 8
(64.1 mg, 0.21 mmol) in chloroform (2 mL) with diisopropylethyl-
amine (DIPEA, 27.7 mg, 0.21 mmol). The resulting solution was
stirred at room temperature for 40 h. The crude product was puri-
fied by silica gel column chromatography. The unreacted com-
pounds and non-polar impurities were eluted with ethyl acetate
and cyclohexane mixture and the product was eluted with 2%
methanol in ethyl acetate. The solvent was evaporated to afford
the product as deep red solid. Yield: 26.8 mg (46.2%). Rf: 0.17
(100% Ethyl acetate). 1H NMR (CDCl3, 300 MHz, ppm): d = 0.88 (t,
J = 6.2 Hz, 6H), 1.25–1.29 (br, 40H), 1.45–1.54 (m, 2H), 1.60–1.64
(m, 2H), 1.81–1.90 (m, 2H), 2.05 (q, J1 = 6.9 Hz, J2 = 13.5 Hz, 2H),
2.18–2.23 (m, 2H), 3.48 (q, J = 7.1 Hz, 4H), 3.66–3.76 (m, 2H),
3.81–3.94 (m, 2H), 4.11–4.20 (m, 2H), 5.54–5.58 (m, 1H), 5.74–
5.83 (m, 1H), 6.30 (s, 1H), 6.40 (d, J = 7.2 Hz, 1H), 6.47 (d,
J = 2.6 Hz, 1H), 6.67 (dd, J1 = 2.7 Hz, J2 = 9.1 Hz, 1H), 7.17 (dd,
J1 = 2.6 Hz, J2 = 8.7 Hz, 1H), 7.61 (d, J = 9.1 Hz, 1H), 8.05 (d,
J = 2.5 Hz, 1H), 8.20 (d, J = 8.7 Hz). 13C NMR (CDCl3, 75 MHz,
ppm): 12.6, 14.1, 22.7, 25.8, 25.9, 29.1, 29.2, 29.3, 29.4, 29.5,
29.7, 31.9, 32.3, 36.8, 45.1, 54.6, 60.8, 62.5, 68.3, 74.6, 77.2, 96.3,
105.1, 106.7, 109.6, 118.3, 124.8, 125.4, 127.7, 128.8, 128.9,
131.1, 134.0, 134.1, 139.9, 146.9, 150.8, 152.2, 161.9, 174.0,
183.5. ESI-MS: m/z calcd for C50H75N3O6: 814.5734 [M+H]+; ob-
served: 814.5733.

6160
K. P. Bhabak et al. / Bioorg. Med. Chem. 20 (2012) 6154–6161
4.8. Synthesis of 7a
J = 7.0 Hz, 3H), 1.24–1.38 (m,40H), 1.52 (m, 2H), 1.63 (m, 2H),
1.85 (m, 2H), 2.03 (m, 2H), 2.23 (m, 2H), 3.48 (q, J = 7.1 Hz, 4H),
3.71 (dd, J1 = 3.3, J2 = 11.2 Hz, 1H), 3.91 (dt, J1 = 3.6, J2 = 7.3 Hz,
J3 = 7.4 Hz, 1H), 3.96 (dd, J1 = 3.8, J2 = 11.3 Hz, 1H), 4.18 (t,
J = 6.4 Hz, 2H), 4.32 (m, 1H), 5.53 (tdd, J1, J2 = 1.3 Hz, J3 = 6.4,
J4 = 15.5 Hz,1H), 5.77 (m, 1H), 6.33 (d, J = 7.6 Hz, 1H), 6.35 (s, 1H),
6.50 (d, J = 2.5 Hz, 1H), 6.70 (dd, J1 = 2.4, J2 = 9.1 Hz, 1H), 7.17 (dd,
J1 = 2.4, J2 = 8.7 Hz, 1H), 7.63 (d, J = 9.1 Hz, 1H), 8.06 (d, J = 2.6 Hz,
1H), 8.21 (d, J = 8.7 Hz, 1H). 13C NMR (CDCl3, 125 MHz, ppm):
d = 12.6, 14.1, 22.6, 24.7, 24.9, 26.0, 28.2, 28.3, 28.9, 29.0, 29.1,
29.1, 29.2, 29.2, 29.3,29.4, 29.4, 29.6, 29.6, 29.6, 31.9, 32.3, 33.8,
33.9, 34.2, 45.0,63.0, 68.3, 96.3, 105.2, 106.6, 109.5, 118.2, 124.7,
125.4, 127.7, 128.3, 131.0, 134.0, 139.2, 140.0, 146.8, 150.7,
152.0, 161.9, 173.9, 178.2, 183.3. ESI-MS: m/z calcd for
To
a
solution of (S)-tert-Butyl-4-((R)-1-hydroxyallyl)-2,2-
(0.15 g, 0.58 mmol)45,44,32
dimethyloxazolidine-3-carboxylate
5
and 9-(Diethylamino)-2-(undec-10-enyloxy)-5H-benzo[a]phenox-
azin-5-one 12 (0.10 g, 0.21 mmol) in dichloromethane (5 mL) was
added a catalytic amount of Hoveyda-Grubbs 2nd Generation cat-
alyst (5.0 mg). The reaction mixture was stirred for 24 h at rt. The
solvent was evaporated and the crude product was purified by sil-
ica gel column chromatography using cyclohexane and ethyl ace-
tate as eluents to afford the product as red solid. Yield: 54 mg
(33.3%). Rf: 0.23 (1:1 Ethyl acetate & Cyclohexane) 1H NMR (CDCl3,
500 MHz, ppm): d = 1.21 (t, J = 7.1 Hz, 6H), 1.39 (m, 4H), 1.50 (s,
19H), 1.59 (m, 4H), 1.87 (m, 2H), 2.07 (dd, J1 = 6.9, J2 = 13.5 Hz,
2H), 3.48 (q, J = 7.1 Hz, 4H), 3.89 (br, 1H), 4.03 (br, 2H), 4.18 (t,
J = 6.5 Hz, 2H), 4.32 (br, 1H), 5.47 (dd, J1 = 4.4, J2 = 15.7 Hz,1H),
5.76 (m, 1H), 6.31 (s, 1H), 6.46 (d, J = 2.7 Hz, 1H), 6.66 (dd,
J1 = 2.7, J2 = 9.1 Hz, 1H), 7.17 (dd, J1 = 2.6, J2 = 8.7 Hz, 1H), 7.61 (d,
J = 9.0 Hz, 1H), 8.05 (d, J = 2.6 Hz, 1H), 8.22 (d, J = 8.7 Hz, 1H). 13C
NMR (CDCl3, 125 MHz, ppm): d = 12.6, 26.0, 26.3, 28.3, 29.1, 29.2,
29.4, 29.4, 29.5, 32.2, 32.4, 45.0, 62.1, 64.6, 68.3, 94.4, 94.5, 96.3,
105.3, 106.6, 109.4, 118.2, 124.6, 125.5, 127.7, 130.9, 131.0,
134.0, 140.1, 146.8, 150.7, 152.0, 161.9, 183.3. ESI-MS: m/z calcd
for C42H57N3O7 (M+H)+: 716.4269; observed 716.4269.
C
50H75N3O6: (M+H)+ 814.5729; observed 814.5729.
Acknowledgements
We thank H. Schulze and K. Sandhoff for a donation of aCDase-
transformed baculo virus. The authors are grateful for generous
funding by the DFG (SPP 1267). K.P.B. thanks the Alexander v.
Humboldt foundation for a research fellowship.
Supplementary data
4.9. Synthesis of 9a
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2012.08.036.
To
a solution of protected NR-sphingosine 7a (120 mg,
0.17 mmol) in dichloromethane (10 mL) was dropwise added
2 ml of Trifluoroacetic acid at room temperature. The reaction mix-
ture was stirred for 45 min. The reaction mixture was poured into
50 ml ice cooled saturated NaHCO3solution and extracted with
ethyl acetate (3 Â 40 mL). The combined organic layer was washed
with saturated NaCl solution and dried over anhydrous Na2SO4.
Solvent was removed under reduced pressure and the crude prod-
uct was purified by silica gel column chromatography using dichlo-
romethane and methanol mixtures as eluents. The pure product
was obtained as wine red colored solid. Yield: 54 mg (55.3%). Rf:
0.05 (100% Ethyl acetate). 1H NMR (CDCl3, 500 MHz, ppm):
d = 1.21 (t, J = 7.2 Hz, 6H), 1.32 (t, J = 7.3 Hz, 10H), 1.45 (m, 2H),
1.78 (m, 2H), 1.97 (dd, J1 = 6.9, J2 = 13.9 Hz, 2H), 3.42 (dd, J1 = 7.1,
J2 = 14.3 Hz, 4H), 3.52 (dd, J1 = 7.3, J2 = 14.6 Hz, 2H), 3.79 (d,
J = 5.6 Hz, 1H), 4.10 (t, J = 6.4 Hz, 2H), 4.39 (m, 1H), 5.38 (dd,
J1 = 6.4, J2 = 15.5 Hz,1H), 5.76 (m, 1H), 6.24 (s, 1 H), 6.42 (d,
J = 2.5 Hz, 1H), 6.64 (dd, J1 = 2.5, J2 = 9.1 Hz, 1H), 7.10 (dd, J1 = 2.4,
J2 = 8.7 Hz, 1H), 7.56 (d, J = 9.1 Hz, 1H), 7.99 (d, J = 2.4 Hz, 1H),
8.11 (d, J = 8.7 Hz, 1H). 13C NMR (CDCl3; 125 MHz, ppm): d = 12.5,
25.9, 28.8, 29.1, 29.1, 29.2, 29.3, 29.4, 32.2, 42.0, 45.1, 52.8, 68.4,
96.2, 104.7, 106.6, 110.0, 115.3, 118.3, 126.3, 127.6, 131.2, 135.4,
139.2, 146.9, 151.1, 152.4, 162.0, 183.8. ESI-MS: m/z calcd for
References and notes
1. Kolter, T.; Sandhoff, K. Angew. Chem., Int. Ed. 1999, 38, 1532.
2. Hannun, Y. A.; Obeid, L. M. J. Biol. Chem. 2011, 286, 27855.
3. Merrill, A. H., Jr. Chem. Rev. 2011, 111, 6387.
4. Park, J. H.; Schuchman, E. H. Biochim. Biophys. Acta. 2006, 1758, 2133.
5. Spiegel, S.; Milstien, S. Nat. Rev. Mol. Cell. Biol. 2003, 4, 397.
6. Hannun, Y. A.; Obeid, L. M. Nat. Rev. Mol. Cell. Biol. 2008, 9, 139.
7. London, E. Curr. Opin. Struct. Biol. 2002, 12, 480.
8. Simons, K.; Ikonen, E. Nature 1997, 387, 569.
9. Canals, D.; Perry, D. M.; Jenkins, R. W.; Hannun, Y. A. Brit. J. Pharmacol. 2011,
163, 694.
10. Gatt, S. J. Biol. Chem. 1963, 238, 3131.
11. Bernardo, K.; Hurwitz, R.; Zenk, T.; Desnick, R. J.; Ferlinz, K.; Schuchman, E. H.;
Sandhoff, K. J. Biol. Chem. 1995, 270, 11098.
12. Li, C. M.; Hong, S. B.; Kopal, G.; He, X.; Linke, T.; Hou, W. S.; Koch, J.; Gatt, S.;
Sandhoff, K.; Schuchman, E. H. Genomics 1998, 50, 267.
13. Koch, J.; Gartner, S.; Li, C. M.; Quintern, L. E.; Bernardo, K.; Levran, O.; Schnabel,
D.; Desnick, R. J.; Schuchman, E. H.; Sandhoff, K. J. Biol. Chem. 1996, 271, 33110.
14. Nilsson, A. Biochim. Biophys. Acta. 1969, 176, 339.
15. Stoffel, W.; Melzner, I. Hoppe Seylers Z. Physiol. Chem. 1980, 361, 755.
16. El Bawab, S.; Roddy, P.; Qian, T.; Bielawska, A.; Lemasters, J. J.; Hannun, Y. A. J.
Biol. Chem. 2000, 275, 21508.
17. Tani, M.; Okino, N.; Mitsutake, S.; Tanigawa, T.; Izu, H.; Ito, M. J. Biol. Chem.
2000, 275, 3462.
18. El Bawab, S.; Usta, J.; Roddy, P.; Szulc, Z. M.; Bielawska, A.; Hannun, Y. A. J. Lipid.
Res. 2002, 43, 141.
19. Sugita, M.; Willians, M.; Dulaney, J. T.; Moser, H. W. Biochim. Biophys. Acta.
1975, 398, 125.
20. Mao, C.; Xu, R.; Szulc, Z. M.; Bielawska, A.; Galadari, S. H.; Obeid, L. M. J. Biol.
Chem. 2001, 276, 26577.
C
34H45N3O5: (M+H)+ 576.3432; observed: 576.3430.
21. Xu, R.; Jin, J.; Hu, W.; Sun, W.; Bielawski, J.; Szulc, Z.; Taha, T.; Obeid, L. M.; Mao,
C. FASEB J. 2006, 20, 1813.
4.10. Synthesis of 3a
22. Sun, W.; Xu, R.; Hu, W.; Jin, J.; Crellin, H. A.; Bielawski, J.; Szulc, Z. M.; Thiers, B.
H.; Obeid, L. M.; Mao, C. J. Invest. Dermatol. 2008, 128, 389.
23. Ehlert, K.; Frosch, M.; Fehse, N.; Zander, A.; Roth, J.; Vormoor, J. Pediatr.
Rheumatol. Online. J. 2007, 5, 15.
24. Teichgraber, V.; Ulrich, M.; Endlich, N.; Riethmuller, J.; Wilker, B.; De Oliveira-
Munding, C. C.; van Heeckeren, A. M.; Barr, M. L.; von Kurthy, G.; Schmid, K. W.;
Weller, M.; Tummler, B.; Lang, F.; Grassme, H.; Doring, G.; Gulbins, E. Nat. Med.
2008, 14, 382.
25. 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.; Szulc, Z.; Keane, T. E.; Tavassoli, M.; Norris, J. S. Front. Biosci.
2008, 13, 2293.
To a solution of NR-sphingosine 9a (34 mg, 0.059 mmol) in
dichloromethane (5 mL) was added palmitic acid (15 mg,
0.059 mmol), EDC (11.3 mg, 0.059 mmol) and HOBT (8.0 mg,
0.059 mmol) with 0.2 mL of DIPEA. The reaction mixture was stir-
red for 16 h at rt. The solution was poured into 20 mL ice-water
and extracted with ethyl acetate (3 Â 30 mL). The combined organ-
ic layer was washed with saturated NaCl solution and dried over
anhydrous Na2SO4. Solvent was removed under reduced pressure
and the crude product was purified by silica gel column chroma-
tography using cyclohexane and ethyl acetate as eluents to afford
the product as wine red solid. Yield: 16 mg (34.0%). Rf: 0.95
(100% Ethyl acetate). 1H NMR (CDCl3, 500 MHz, ppm): d = 0.87 (t,
26. Radin, N. S. Eur. J. Biochem. 2001, 268, 193.
27. Bielawska, A.; Linardic, C. M.; Hannun, Y. A. J. Biol. Chem. 1992, 267, 18493.
28. Bielawska, A.; Greenberg, M. S.; Perry, D.; Jayadev, S.; Shayman, J. A.; McKay, C.;
Hannun, Y. A. J. Biol. Chem. 1996, 271, 12646.

K. P. Bhabak et al. / Bioorg. Med. Chem. 20 (2012) 6154–6161
6161
29. Delgado, A.; Casas, J.; Llebaria, A.; Abad, J. L.; Fabrias, G. Biochim. Biophys. Acta.
2006, 1758, 1957.
30. Nikolova-Karakashian, M.; Morgan, E. T.; Alexander, C.; Liotta, D. C.; Merrill, A.
H., Jr. J. Biol. Chem. 1997, 272, 18718.
31. Dagan, A.; Agmon, V.; Gatt, S.; Dinur, T. In Methods Enzymol; Merill, A. H.,
Hannun, Y. A., Eds.; Academic Press: San Diego, 1999.
32. Tani, M.; Okino, N.; Mitsutake, S.; Ito, M. J. Biochem. 1999, 125, 746.
33. Antes, P.; Schwarzmann, G.; Sandhoff, K. Eur. J. Cell. Biol. 1992, 59, 27.
34. Nieuwenhuizen, W. F.; van Leeuwen, S.; Gotz, F.; Egmond, M. R. Chem. Phys.
Lipids. 2002, 114, 181.
45. Ojima, I.; Vidal, E. S. J. Org. Chem. 1998, 63, 7999.
46. Briggs, M. S. J.; Bruce, I.; Miller, J. N.; Moody, C. J.; Simmonds, A. C.; Swann, E. J.
Chem. Soc., Perkin. Trans. 1997, 1, 1051.
47. Novotny, J.; Pospechova, K.; Hrabalek, A.; Cap, R.; Vavrova, K. Bioorg. Med.
Chem. Lett. 2009, 19, 6975.
48. Raisova, M.; Goltz, G.; Bektas, M.; Bielawska, A.; Riebeling, C.; Hossini, A. M.;
Eberle, J.; Hannun, Y. A.; Orfanos, C. E.; Geilen, C. C. FEBS Lett. 2002, 516, 47.
49. Proksch, D.; Klein, J. J.; Arenz, C. J. Lipids 2011, 2011, 971618.
50. Mukherjee, S.; Raghuraman, H.; Chattopadhyay, A. Biochim. Biophys. Acta. 2007,
1768, 59.
35. Bedia, C.; Casas, J.; Garcia, V.; Levade, T.; Fabrias, G. ChemBioChem 2007, 8, 642.
36. Bedia, C.; Camacho, L.; Abad, J. L.; Fabrias, G.; Levade, T. J. Lipid. Res. 2010, 51,
3542.
51. Wilkening, G.; Linke, T.; Sandhoff, K. J. Biol. Chem. 1998, 273, 30271.
52. Furst, W.; Sandhoff, K. Biochim. Biophys. Acta. 1992, 1126, 1.
53. Giehl, A.; Lemm, T.; Bartelsen, O.; Sandhoff, K.; Blume, A. Eur. J. Biochem. 1999,
261, 650.
54. Ettmayer, P.; Billich, A.; Baumruker, T.; Mechtcheriakova, D.; Schmid, H.;
Nussbaumer, P. Bioorg. Med. Chem. Lett. 2004, 14, 1555.
55. He, X.; Li, C. M.; Park, J. H.; Dagan, A.; Gatt, S.; Schuchman, E. H. Anal. Biochem.
1999, 274, 264.
37. Xia, Z.; Draper, J. M.; Smith, C. D. Bioorg. Med. Chem. 2010, 18, 1003.
38. Wichmann, O.; Schultz, C. Chem. Commun. 2001, 2500.
39. Wichmann, O.; Wittbrodt, J.; Schultz, C. Angew. Chem., Int. Ed. 2006, 45, 508.
40. Greenspan, P.; Mayer, E. P.; Fowler, S. D. J. Cell. Biol. 1985, 100, 965.
41. Wang, T. Y.; Silvius, J. R. Biophys. J. 2000, 79, 1478.
42. van Meer, G.; Liskamp, R. M. J. Nat. Meth. 2005, 2, 14.
43. Nussbaumer, P.; Ettmayer, P.; Peters, C.; Rosenbeiger, D.; Hogenauer, K. Chem.
Commun. 2005, 5086.
56. Kuerschner, L.; Ejsing, C. S.; Ekroos, K.; Shevchenko, A.; Anderson, K. I.; Thiele,
C. Nat. Methods 2005, 2, 39.
57. Schulze, H.; Schepers, U.; Sandhoff, K. Biol. Chem. 2007, 388, 1333.
44. Peters, C.; Billich, A.; Ghobrial, M.; Hogenauer, K.; Ullrich, T.; Nussbaumer, P. J.
Org. Chem. 2007, 72, 1842.