Synthesis of a novel fluorescent ceramide analogue and its use in the characterization of recombinant ceramidase from Pseudomonas aeruginosa PA01

Article abstract of DOI:10.1016/S0009-3084(01)00206-7

Ceramidase (CDase) hydrolyses the N-acyl linkage of the sphingolipid ceramide. We synthesized the non-fluorescent ceramide analogue (4E,2S,3R)-2-N-(10-pyrenedecanoyl)-1,3,17-trihydroxy-17- (3,5-dinitrobenzoyl)-4-heptadecene (10) that becomes fluorescent upon hydrolysis of its N-acyl bond. This novel substrate was used to study several kinetic aspects of the recombinant CDase from the pathogenic bacterium Pseudomonas aeruginosa PA01. Maximum CDase activity was observed above 1.5 μM substrate, with an apparent K_m of 0.5±0.1 μM and a turnover of 5.5 min^-1. CDase activity depends on divalent cations without a strong specificity. CDase is inhibited by sphingosine and by several sphingosine analogues. The lack of inhibition by several mammalian CDase inhibitors such as D-erythro-MAPP, L-erythro-MAPP or N-oleoylethanolamine points to a novel active site and/or substrate binding region. The CDase assay described here offers the opportunity to develop and screen for specific bacterial CDase inhibitors of pharmaceutical interest.

Full text of DOI:10.1016/S0009-3084(01)00206-7

Chemistry and Physics of Lipids
114 (2002) 181–191
www.elsevier.com/locate/chemphyslip
Synthesis of a novel fluorescent ceramide analogue and its
use in the characterization of recombinant ceramidase from
Pseudomonas aeruginosa PA01
Willem F. Nieuwenhuizen a,b,*, Sander van Leeuwen b, Friedrich Go¨tz a,
Maarten R. Egmond b
a Microbial Genetics, Uni6ersity of Tu¨bingen, Waldha¨user Straße 70/8, 72076 Tu¨bingen, Germany
b Department of Biochemistry, Enzyme and Protein Engineering, Utrecht Uni6ersity, Padualaan 8,
3584 CH Utrecht, The Netherlands
Received 16 August 2001; received in revised form 4 December 2001; accepted 13 December 2001
Abstract
Ceramidase (CDase) hydrolyses the N-acyl linkage of the sphingolipid ceramide. We synthesized the non-fluores-
cent ceramide analogue (4E,2S,3R)-2-N-(10-pyrenedecanoyl)-1,3,17-trihydroxy-17-(3,5-dinitrobenzoyl)-4-heptadecene
(10) that becomes fluorescent upon hydrolysis of its N-acyl bond. This novel substrate was used to study several
kinetic aspects of the recombinant CDase from the pathogenic bacterium Pseudomonas aeruginosa PA01. Maximum
CDase activity was observed above 1.5 mM substrate, with an apparent Km of 0.590.1 mM and a turnover of 5.5
min−1. CDase activity depends on divalent cations without a strong specificity. CDase is inhibited by sphingosine and
by several sphingosine analogues. The lack of inhibition by several mammalian CDase inhibitors such as
D-erythro-
MAPP, -erythro-MAPP or N-oleoylethanolamine points to a novel active site and/or substrate binding region. The
L
CDase assay described here offers the opportunity to develop and screen for specific bacterial CDase inhibitors of
pharmaceutical interest. © 2002 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Ceramidase; Fluorescent assay; Kinetics; Inhibition
1. Introduction
Sphingolipids such as gangliosides, cere-
brosides, sphingomyelin, and ceramide are signifi-
cant constituents of eukaryotic cells, and occur
mainly in the plasma membrane, the Golgi ap-
paratus, and lysosomes. Furthermore, sphin-
golipids are highly bioactive compounds. They are
involved in the regulation of cell growth, the
induction of cell differentiation and apoptosis (for
* Corresponding author. Present address: TNO Nutrition
and Food Research, Department of Protein and Meat Tech-
nology, Utrechtseweg 48, 3700 AJ Zeist, The Netherlands.
Tel.: +31-30-6944-283; fax: +31-30-6944-295.
E-mail address: nieuwenhuizen@voeding.tno.nl (W.F.
Nieuwenhuizen).
0009-3084/02/$ - see front matter © 2002 Elsevier Science Ireland Ltd. All rights reserved.
PII: S0009-3084(01)00206-7

182
W.F. Nieuwenhuizen et al. / Chemistry and Physics of Lipids 114 (2002) 181–191
reviews see Liu et al., 1999; Mathias et al., 1998;
Sharma and Shi, 1999; Spiegel and Merrill, 1996).
Ceramidase (CDase) is an enzyme that acts at
the lipid–water interface like lipases do, and hy-
drolyses the N-acyl bond in ceramide to yield a
free fatty acid and free sphingosine. CDases have
been reported in eukaryotes and prokaryotes with
acid (Bernardo et al., 1995; Koch et al., 1996; Li
et al., 1998, 1999; Spence et al., 1986), neutral
(Tani et al., 2000) and alkaline (Bawab et al.,
2000; Mao et al., 2000; Nilsson and Duan, 1999;
Okino et al., 1998, 1999; Wertz and Downing,
1990; Yada et al., 1995) pH optima.
In mammals CDase plays an important role in
the control of the cellular ceramide content, and
in the regulation of intracellular signal transduc-
tion. Acid CDase has been detected in rat brain,
kidney, liver and spleen, human kidney and cere-
bellum, leukocytes, spleen, and plasma. Neutral
and alkaline CDases have been found in pig intes-
tinal mucosa, rat organs, human cerebellum and
fibroblasts, pig lens epithelium, human leukocytes,
and porcine epidermis (Hassler and Bell, 1993).
The lysosomal acid CDase is the best known
human CDase. The enzyme is activated by an-
ionic phospholipids and sphingolipid activator
proteins (Linke et al., 2001). The hereditary defi-
ciency of lysosomal acid CDase, such as in the
lethal Farber disease (Bar et al., 2001), leads to
the accumulation of ceramide in the lysosomes.
Ceramide and CDase activities have thusfar
been determined with the much debated diglyce-
ride kinase assay (Hofmann, 1999; Perry and
Hannun, 1999 and references therein), with ra-
dioactive (Misutake et al., 1997), and with fluores-
cent assays (Tani et al., 1998, 1999). These
discontinuous assays either have a low reliability,
or are laborious and include a separation step.
There is a need for a rapid CDase assay that
allows to detect and follow the enzymatic reaction
continuously in real time.
2. Experimental procedures
1-Bromo-12-dodecanol, 2,3-dihydropyran, Red-
Al (3.5 M in toluene), and butyllithium (BuLi, 1.6
M
in hexane) were from Aldrich. Lithium
acetylide ethylene diamine complex (95%), Am-
berlyst 15, Triton X-100, triethylamine, (−)-
and (+)-norephenedrin were from Fluka. Cy-
L
D
clohexanone, 3,5-dinitrobenzoylchloride, p-tolu-
ene sulfonic acid and dicyclohexylcarbodiimide
(DCC) were from Acros. 10-pyrene decanoic acid
was from Molecular Probes. (1S,2R)-
(N-myristoylamino)-1-phenyl-1-propanol (
thro-MAPP), -erythro-MAPP and N-oleoyletha-
nolamine were from CalBiochem. -sphinga-
nine, -erythro-sphingosine, 2-[N-morpholino]-et-
D
-erythro-2-
D
-ery-
L
D/L
D
hanesulfonic acid (MES), 3-[N-morpholino]-pro-
panesulfonic acid (MOPS), tris(hydroxymethyl)
aminoethane (Tris), boric acid and bovine serum
albumin were from Sigma.
was from Larodan.
Protein concentrations were determined with a
Bio-Rad DC (Detergent-Compatible) Protein As-
say (BioRad) with bovine serum albumin as
standard.
L-erythro-sphingosine
Hexanes, ethylacetate (EtOAc), methanol (Me-
OH), diethylether (Et2O), and chloroform (CH-
Cl3) were distilled prior to use. Toluene and
dichloromethane (CH2Cl2) were dried by distilla-
tion from phosphorpentoxide (P2O5). Tetrahydro-
furan (THF) and Et2O were dried by distillation
under a N2 atmosphere from LiAlH4 (5 g l−1
directly prior to use. Dimethylsulfoxide (DMSO),
pyridin and hexamethylphosphortriamide (HM-
)
,
PT) were dried over 3 A molsieves.
Column Chromatography was performed with
Merck Kieselgel 60 (230–400 mesh, ASTM). TLC
analyses were performed on Merck Kieselgel F254
DC-Fertig plates.
1H NMR spectra were recorded with a Bruker
AC 300 (300 MHz) spectrometer in CDCl3 at 300
K. 1H chemical shifts are given in ppm (l) relative
to internal TMS.
To meet these requirements we synthesized a
novel quenched fluorescent ceramide analogue
which becomes fluorescent when it is hydrolyzed
by CDase, and demonstrated its usefulness in the
characterization of the alkaline CDase from Pseu-
domonas aeruginosa PA01 (Nieuwenhuizen et al.,
to be published elsewhere).
High Resolution Fast Atom Bombardment was
carried out using a JEOL JMS SX/SX 102A
four-sector mass spectrometer coupled to a MS-
MP 9021D/UDP data system.

W.F. Nieuwenhuizen et al. / Chemistry and Physics of Lipids 114 (2002) 181–191
183
The samples were mixed with 3-nitrobenzylal-
cohol and were introduced via the direct insertion
probe into the ion source. During the High Reso-
lution FABMS measurements a resolving power
of 5000 (10% valley definition) was used. B/E
scantype mass spectra were acquired with CA
(Collision Activation) in the first field free region
of the instrument, using Ar as collision gas. The
pressure of the collision gas was adjusted to ob-
tain a 50% reduction of the beam. Resolution of
the instrument was maintained at 1000 through-
out the experiments.
the reaction mixture was concentrated in vacuo.
The residue was dissolved in dry DMSO (50 ml),
cooled on ice, and lithium acetylide ethylene di-
amine complex (1.5 g, 16 mmol) was carefully
added. The suspension was allowed to warm to
room temperature and stirred for 4 h. The reac-
tion mixture was then poured into a saturated
NaHCO3 solution (300 ml) and extracted with
CHCl3 (5×100 ml). The combined organic
phases were subsequently washed with saturated
NaCl (200 ml), water (200 ml), and dried
(Na2SO4), and then concentrated in vacuo.
Column chromatography (silica, hexanes/EtOAc,
95/5, v/v) yielded pure 2 as a colorless oil (2.20 g,
7.5 mmol, 80%).
CDase was produced in Escherichia coli (BL 21)
and isolated from inclusion bodies as described
(Nieuwenhuizen et al., to be published elsewhere).
1H NMR (CDCl3) l: 1.22–1.45 (m, 16H),
1.46–1.62 (m, 8H), 1.63–1.88 (m, 2H), 1.93 (t, H,
J=2.6 Hz), 2.17 (dt, 2H, J=6.9 Hz, J=2.6 Hz),
3.35 (m, H), 3.51 (m, H), 3.72 (m, H), 3.87 (m, H),
4.57 (m, H).
2.1. Synthesis of the quenched fluorescent
ceramide analogue 10
The synthesis of ceramide analogue 10 was
performed as outlined in Fig. 2.
2.1.1.2. tert-Butyl(4S, 1%R)-2,2-dimethyl-4-[1%-hy-
droxy-(15%-(2-tetrahydropyranyl)pentadec)-2%-ynyl]
oxazolidine-3-carboxylate (4). Compound 4 was
synthesized as described (Herold, 1988) with some
modifications.
2.1.1. 1,1-Dimethylethyl (S)-4-formyl-2,2-
dimethyl-3-oxazolidinecarboxylate (3)
Compound 3 was synthesized in four steps
from
L-serine as described (Garner and Min Park,
To a solution of compound 2 (1.82 g, 6.2
mmol) in dry THF (50 ml, −78 °C, under N2)
BuLi (1.6 M in hexane, 3.8 ml, 6.1 mmol) was
added via cannula. After 30 min dry HMPT (2.8
ml, 16 mmol) was added. A solution of compound
3 (0.90 g, 3.9 mmol) in dry THF (5 ml) was added
dropwise. The reaction mixture was stirred for 2 h
at −78 °C. The reaction mixture was then al-
lowed to warm to −20 °C, and saturated NH4Cl
(200 ml) and EtOAc (100 ml) were added. The
mixture was stirred for 18 h at room temperature.
After phase separation, the water phase was ex-
tracted with EtOAc (5×100 ml). The combined
organic phases were subsequently washed with
saturated NaCl (100 ml), water (100 ml), dried
(Na2SO4) and concentrated in vacuo. Column
chromatography (silica, hexane/EtOAc/triethy-
lamine, 800/200/1, v/v/v) yielded pure 4 (1.00 g,
1.90 mmol, 49%) as a colorless oil. The erythro
conformation of 4 was confirmed by the 11.2 Hz
coupling in compound 7 (vide infra) indicating a
trans coupling of protons a and b (see Fig. 2).
1987).
1H NMR (CDCl3) l: 1.49–1.52 (m, 15H), 4.11
(m, 2H), 4.41 (m, H), 9.58 (bd, H, J=16.3 Hz).
2.1.1.1. 1-(2-Tetrahydropyranyl) tetradecyl-13-yn
(2). 1-Bromo-12-dodecanol (2.5 g, 9.4 mmol), 1,2-
dihydropyran (4 ml, 38 mmol), and a catalytic
amount of p-toluenesulfonic acid were dissolved
in dry toluene (20 ml), and stirred for 1 h at room
temperature. Triethylamine (1 ml) was added, and
Fig. 1. The hydrolysis of CDase analogue 10 by CDase.

184
W.F. Nieuwenhuizen et al. / Chemistry and Physics of Lipids 114 (2002) 181–191
1H NMR (CDCl3) l:1.19–1.78 (m, 40H), 1.82
(m, 2H), 2.19 (dt, H, J=6.6 Hz, J=1.8 Hz), 3.39
(m, H), 3.48 (m, H), 3.72 (m, H), 3.75 (m, H), 3.90
(m, H), 4.10 (m, 3H), 4.45–4.71 (bs, H), 4.61 (m,
H).
2.1.1.3. tert-Butyl(1S, 2R)-N-[2-hydroxy-1-(hy-
droxymethyl)-3-hexadecynyl-16-ol]carbamate (5).
Compound 4 (1.00 g, 1.9 mmol) was dissolved in
MeOH (20 ml), Amberlyst 15 (500 mg) was added,
and the mixture was stirred for 55 h at room
temperature. After filtration (G4 glass filter) and
concentration in vacuo, compound 5 (690 mg, 1.7
mmol, 89%) was obtained as a light yellow oil.
1H NMR (CDCl3) l: 1.21–1.69 (m, 29H), 2.22
(dt, 2H, J=7.0 Hz, J=1.9 Hz), 2.41 (bs, H), 2.87
(bs, H), 3.64 (t, 2H, J=6.6 Hz), 3.76 (bd, 2H,
J=8.5 Hz), 4.10 (m, H), 4.60 (s, H), 5.30 (bs, H),
NH not visible.
2.1.1.4. tert-Butyl(3E, 1S, 2R)-N-[2-hydroxy-
1-(hydroxymethyl)-3-hexadecenyl-16-ol]carbamate
(6). To an ice cold solution of compound 5 (690
mg, 1.7 mmol) in dry Et2O (20 ml), Red-Al (3.5 M
in toluene, 7 ml, 22.5 mmol) was added dropwise
under a N2 atmosphere. The reaction mixture was
stirred for 30 min. The reaction mixture was then
stirred for 24 h at room temperature. The clear
reaction mixture was cooled on ice, and MeOH
(10 ml) was added slowly. Then saturated
sodium–potassium tartrate (150 ml) was added,
and the mixture was stirred for 1 h at room
temperature. After phase separation, the water
phase was extracted with Et2O (4×50 ml). The
combined organic phases were subsequently
washed with saturated sodium–potassium tartrate
(3×50 ml), saturated NaCl (2×50 ml), dried
(Na2SO4), and concentrated in vacuo. Column
chromatography (silica, hexanes/EtOAc, 1/1, v/v)
and repeated crystallization from hexanes/EtOAc
(9/1, v/v) yielded pure 6 as a white solid (263 mg,
0.66 mmol, 39%, melting point (m.p.) 66–67 °C).
1H NMR (CDCl3) l: 1.20–1.65 (m, 29H), 2.06
(m, 2H), 2.51 (bs, 2H), 3.64 (t, 2H, J=6.6 Hz),
3.70 (m, 2H), 3.93 (dd, H, J=3.7 Hz, J=11.3
Hz), 4.32 (s, H), 5.27 (bs, H), 5.53 (dd, H, J=6.6
Hz, J=15.4 Hz), 5.78 (dt, H, J=6.6 Hz, J=15.4
Hz), NH not visible.
Fig. 2. Synthesis of ceramide analogue 10. (a) I; 1,2 dihydropy-
ran and p-toluene sulfonic acid in toluene; II, lithium acetylide
ethylene diamine complex in DMSO. (b) BuLi, HMPT in
THF, −78 °C. (c) Amberlyst 15, MeOH. (d) Red-Al, Et2O,
0 °C. (e) Cyclohexanone, p-toluenesulfonic acid in refluxing
toluene. (f) 3,5-Dinitrobenzoyl chloride in CH2Cl2 and pyridin.
(g) I; HCl in EtOAc, II; Amberlyst 15 in MeOH. (h) 10-Pyrene
decanoic acid anhydride in THF.

W.F. Nieuwenhuizen et al. / Chemistry and Physics of Lipids 114 (2002) 181–191
185
2.1.1.5. tert-Butyl(3E, 1S, 2R)-N-[1,2-O-cyclo-
hexylidene-2-hydroxy-1-(hydroxymethyl)-3-hexa-
decenyl-16-ol]carbamate (7). Compound 6 (263
mg, 0.66 mmol), cyclohexanone (0.82 ml, 0.79
mmol) and a catalytic amount of p-toluenesul-
fonic acid were dissolved in dry toluene (50 ml).
The reaction mixture was refluxed for 30 min, and
the reaction was completed by distillation of 20
ml solvent and water. Triethylamine (1 ml) was
added, and the reaction mixture was concentrated
in vacuo. Column chromatography (silica, hex-
drin staining. The reaction mixture was concen-
trated in vacuo, the residue was dissolved in
MeOH (10 ml), Amberlyst 15 (500 mg) was
added, and the mixture was stirred for 24 h at
room temperature. The removal of the cyclo-
hexylidene group was checked with TLC (CHCl3/
MeOH, 8/2, v/v). After filtration (G4 glass filter),
concentration in vacuo, and column chromatogra-
phy (silica, CHCl3/MeOH, 8/2, v/v), compound 9
(100 mg, 0.21 mmol, 45%) was obtained as a waxy
white solid.
anes/EtOAc/triethylamine,
800/200/1,
v/v/v)
1H NMR (CDCl3) l: 1.16–1.52 (m, 20H), 1.79
(m, 2H), 2.01 (m, 2H), 3.27 (bs, H), 3.54 (bs, 2H),
3.81 (bd, 2H, J=4.5 Hz), 4.42 (s, H), 4.44 (t, 2H,
J=6.8 Hz), 5.46 (dd, H, J=6.2 Hz, J=15.4 Hz),
8.81 (dt, H, J=15.4 Hz, J=6.7 Hz), 9.14 (m,
2H), 9.21 (m, H).
yielded pure 7 (226 mg, 0.46 mmol, 70%) as a
colorless oil.
1H NMR (CDCl3) l: 1.15–1.62 (m, 35H), 1.80
(m, 2H), 1.98 (m, 4H), 3.41 (m, H), 3.58 (t, 2H,
J=6.6 Hz), 3.62 (m, H), 3.89 (dd, H, J=5.4 Hz,
J=11.2), 4.02 (bs, H), 4.45 (d, H, J=8.5 Hz),
5.39 (dd. H, J=7.4 Hz, J=15.4 Hz), 5.70 (dt, H,
J=6.6 Hz, J=15.4 Hz), NH not visible.
2.1.1.8. (4E, 2S,3R)-2-N-(10-Pyrenedecanoyl)-
1,3,17-trihydroxy-17-(3,5-dinitrobenzoyl)-4-hept-
adecene (10). 10-Pyrene decanoic acid (22 mg,
0.06 mmol) and DCC (400 mg, 1.9 mmol) were
dissolved in dry THF (10 ml) and stirred at room
temperature for 1 h in the dark. The anhydride
was added dropwise to a solution of compound 9
in dry THF (5 ml, 29 mg, 0.06 mmol) at room
temperature over a 1 h period. The reaction mix-
ture was stirred for 18 h in the dark at room
temperature. After addition of water (1 ml), the
reaction mixture was stirred for 15 min and con-
centrated in vacuo. Traces of water were coevapo-
rated with MeOH. Column chromatography
(silica, CHCl3/MeOH, 99/1, v/v) yielded 10 (29
mg, 35 mmol, 58%, RF 0.12) as a yellow wax.
1H NMR (CDCl3) l: 1.16–1.69 (m, 30 H), 1.79
(m, 4H), 2.01 (m, 2H), 2.22 (t, 2H, J=7.5 Hz),
2.87 (bs, 2H), 3.31 (t, 2H, J=7.8 Hz), 3.69 (bd,
H, J=9.2 Hz), 3.91 (m, 2H), 4.29 (bs, H), 4.40 (t,
2H, J=6.9 Hz), 5.52 (dd, H, J=15.4 Hz, J=6.5
Hz), 5.76 (dt, H, J=15.4 Hz, J=6.6 Hz), 6.25 (d,
H, J=7.3 Hz), 7.84 (d, H, J=7.8 Hz), 7.98 (m,
3H), 8.01–8.14 (m, 4H), 8.23 (d, H, J=9.2 Hz),
8.96 (m, 2H), 9.04 (m, H).
2.1.1.6. tert-Butyl(3E, 1S, 2R)-N-[1,2-O-cyclo-
hexylidene-2-hydroxy-1-(hydroxymethyl)-3-hexa-
decenyl-16-(3,5-dinitrobenzoyl)]carbamate (8). A
solution of 3,5-dinitrobenzoylchloride (117 mg,
0.51 mmol) in dry CH2Cl2 (10 ml) was added
dropwise to a stirred solution of compound 7 (226
mg, 0.46 mmol) and dry pyridine (1 ml) in dry
CH2Cl2 (20 ml). The reaction was stirred for 18 h
at room temperature, and was then concentrated
in vacuo. Column chromatography (silica, hex-
anes/EtOAc, 8/2, v/v) yielded pure 8 (321 mg,
0.46 mmol, 100%) as a light yellow oil.
1H NMR (CDCl3) l: 1.12–1.60 (m, 37H), 1.79
(m, 2H), 2.01 (m, 2H), 3.42 (m, H), 3.60 (m, H),
3.88 (dd, H, J=5.4 Hz, J=11.2 Hz), 4.02 (bs,
H), 4.31 (bs, H), 4.39 (t, 2H, J=6.8 Hz), 5.39 (dd,
H, J=7.3 Hz, J=15.4 Hz), 5.71 (dt, H, J=6.7
Hz, J=15.4 Hz), 9.11 (m, 2H), 9.17 (m, H).
2.1.1.7. (4E, 2S,3R)-2-Amino-1,3,17-trihydroxy-
17(3,5-dinitrobenzoyl)-4-heptadecene (9). Com-
pound 8 (321 mg, 0.46 mmol) was dissolved in
0.42 M HCl in EtOAc (30 ml, 12.6 mmol) and the
mixture was stirred for 24 h at room temperature
(Gibson et al., 1994). The removal of the tert-
butyloxycarbonyl (BOC) group was followed via
TLC (silica, hexanes/EtOAc, 1/1, v/v) and ninhy-
FAB-MS:. The cationized molecular ion [M+
H]+ at m/z 850.4635 is in good agreement with
the theoretical m/z value of 850.4643. Fragmenta-
tion of the molecular ion yielded ions at m/z
832.47 and m/z 814.47 and are assigned to frag-

186
W.F. Nieuwenhuizen et al. / Chemistry and Physics of Lipids 114 (2002) 181–191
ments [MꢀH2O]+ and [Mꢀ2H2O]+, respectively.
Ions at m/z 355.15, 371.17 and 414.27 are as-
signed to [C(O)ꢀ(CH2)9ꢀpyrene]+, [H2NꢀC(O)ꢀ
(CH2)9ꢀpyrene]+ and [HOꢀCH2ꢀCHꢀNHꢀC(O)ꢀ
(CH2)9ꢀpyrene]+, respectively.
2.2.1. Inhibitor assay
The CDase inhibitors
N-oleoylethanolamine,
D
-e-MAPP,
(−)-norephenedrin,
-sphinganine, and -ery-
-erythro-sphingosine were
L-e-MAPP,
L
D-
(+)-norephenedrin,
D
/L
D
thro-sphingosine and
L
dissolved in ethanol. In all inhibitor assays, the
final concentration of organic solvent was 0.5%
(v/v).
2.2. Characterization of CDase
Some kinetic aspects of recombinant CDase
were studied using the quenched fluorescent ce-
ramide analogue 10 as a substrate. All experi-
ments were performed at 30 °C using a Photon
Technology International Fluorimeter equipped
with a LPS-220 Lamp Power Supply, a 810-Pho-
tomultiplier Detector System, a MD-5020 Motor-
drive, and a Dell Dimensions M166a computer
with FELIX 1.11 (PTI) software. Excitation was done
at 346 (92 nm), and the reaction was followed
by monitoring the emission at 378 (92 nm). The
substrate (5 ml, 0.4 mM in MeOH, 2 mM) was
added to the buffer (1 ml) and the mixture was
allowed to equilibrate for 5 min. Then CDase (5
ml, 0.06 mg ml−1, 4.3 nM) was added, and the
reaction was allowed to proceed for 20 min. The
slope of the increase in fluorescence versus time
was used to assess the CDase activity. Above pH
9.5 the enzymatic hydrolysis was corrected for the
auto hydrolysis of the 3,5-dinitrobenoyl ester.
An average molecular weight of 624 for Triton
X-100 was used to calculate its concentration.
Curve fitting was done with KALEIDAGRAPH
3.0.2 software (Macintosh version).
3. Results
3.1. Synthesis of ceramide analogue 10
Here we describe the synthesis of a quenched
fluorescent ceramide analogue 10 (Fig. 1). The
3,5-dinitrobenzoyl group is an efficient in-
tramolecular quencher of the fluorescence of the
pyrene group. Upon hydrolysis of the N-acyl
bond by CDase the average distance between the
quenching and fluorescent groups increases, giv-
ing rise to a strong increase of the pyrene fluores-
cence, thus allowing the real time detection of the
CDase reaction.
In Fig. 2 the reaction scheme for the synthesis
of ceramide analogue 10 is shown. Compound 2
was synthesized from bromo alcohol 1 in 80%
yield, in a two-step synthesis with 2,3-dihydropy-
ran and lithiumacetylide. Acetylenic compound 2
was then coupled with BuLi to aldehyde 3, which
is a frequently used intermediate in the sphin-
golipid synthesis (Azuma et al., 2000; Duffin et
al., 2000; Garner and Min Park, 1987; Garner et
al., 1988; Herold, 1988; Yin et al., 1998). The
coupling was done in the presence of HMPT to
direct the coupling to the erythro form. Diacetal 4
was obtained in 49% yield (\95% erythro, vide
infra). The selective removal of the acetal protec-
tive groups with the use of the acidic cation
exchange resin Amberlyst 15 in MeOH yielded
triol 5 in 89% yield. The partial reduction of 5
with Red-Al yielded compound 6 only in 39%
yield after extensive recrystallisation to remove
traces of unreduced 5. In order to achieve the
selective coupling of the 3,5-dinitrobenzoyl group
to the ꢀ hydroxyl group, the free neighboring
hydroxyl groups were selectively protected with
cyclohexanone, which afforded acetal 7 in 70%
The effect of the Triton X-100 concentration on
the CDase reaction was measured in 50 mM
Tris–HCl, 4 mM CaCl2, pH 8.5.
The pH dependence of the CDase activity was
tested in 50 mM buffers (MES, MOPS, Tris, and
Borate), at 4 mM CaCl2 and 1.6 mM Triton
X-100, in the pH range from pH 5.5 to 11.0.
The influence of the Ca2+, Ba2+, and Mg2+
on the CDase activity was tested in 50 mM Tris–
HCl, 1.6 mM Triton X-100, pH 8.5. The influence
of the Zn2+ concentration was measured in 50
mM MES, 1.6 mM Triton X-100, pH 6.5.
The effect of the CDase concentration on the
reaction rate tested with 20 mM substrate in 50
mM Tris–HCl, 1.6 mM Triton X-100, 4 mM
CaCl2, pH 8.5.

W.F. Nieuwenhuizen et al. / Chemistry and Physics of Lipids 114 (2002) 181–191
187
yield. The erythro conformation was confirmed by
the 11.2 Hz coupling indicating a trans coupling
of protons a and b (see Fig. 2). The use of acetone
instead of cyclohexanone gave a 50/50 mixture of
the desired acetal and the oxazolide. The presence
of the less flexible and more bulky ring structure
probably prevents the formation of the undesired
oxazolide. The reaction of 7 with 3,5-dinitroben-
zoyl chloride gave compound 8 in quantitative
yield. The selective removal of the BOC group
was achieved with dry HCl in EtOAc. The acetal
was then removed with the use of Amberlyst 15 in
MeOH. Unfortunately the ester bond was also
partially hydrolyzed under these conditions, and
intermediate 9 was obtained in 45% yield. Finally
target compound 10 was synthesized in 58% yield,
in a reaction of 10-pyrene decanoic acid anhy-
dride with 9, making use of the higher reactivity
of the ꢀNH2 compared with the ꢀOH groups. We
successfully synthesized ceramide analogue 10 in
3% overall yield starting from intermediate 3.
100 concentration on the CDase hydrolysis reac-
tion at pH 8.5 in the presence of Ca2+ (4 mM),
using the quenched fluorescent ceramide analogue
10 (20 mM) at a fixed CDase concentration (4.3
nM). As expected a steep rise in activity was
observed at the critical micellar concentration (ꢀ
0.8 mM) of Triton X-100, and the maximum
reaction rate was observed at 1.6 mM Triton
X-100. All further experiments were, therefore,
conducted at that detergent concentration.
CDase did not hydrolyze the 3,5-dinitrobenzoyl
ester linkage in the substrate, even after 24 h
incubations, as was shown by TLC analyses. Only
one fluorescent product was formed which
coeluted with the pyrene fatty acid, indicating the
selective cleavage of the amide bond by CDase.
The pH dependence of the CDase hydrolysis
reaction at a fixed CDase concentration (4.3 nM),
and substrate (20 mM) is shown in Fig. 3. The
recombinant CDase is active over a broad pH
range, and shows more than 90% of its maximum
activity between pH 7.5 and 9.5. Note that CDase
is more active in MOPS buffer than in MES or
Tris at the same pH. To avoid possible alkaline
hydrolysis of the ester bond in the substrate, all
3.2. Kinetics of recombinant CDase from P.
aeruginosa PA01
First we tested the influence of the Triton X-
Fig. 3. pH-dependence of the CDase reaction with 4.3 nM CDase in 50 mM buffer with 1.6 mM Triton X-100, 4 mM CaCl2, 20
mM substrate at 30 °C. ꢀ MES, ꢁ MOPS, " Tris, ꢂ borate. The relative reaction rates were determined as described in Section
2.

188
W.F. Nieuwenhuizen et al. / Chemistry and Physics of Lipids 114 (2002) 181–191
Fig. 4. Effect of the CDase concentration on the reaction rate. Reaction conditions: 50 mM Tris–HCl pH 8.5, 4 mM CaCl2, 20 mM
substrate, 30 °C. The relative reaction rates were determined as described in Section 2.
min−1 (specific activity of 78 nmol min−1 mg−1
)
further experiments were performed at pH 8.5 (see
Section 2).
from a fluorescence calibration curve with free
10-pyrene decanoic acid in the same buffer. This
value compares well with the turnover of 14C-N-
palmitoyl sphingosine of 10.9 min−1 (155 nmol
min−1 mg−1) (Nieuwenhuizen et al., to be pub-
lished elsewhere).
The CDase reaction rate at pH 8.5 and 20 mM
substrate depends on the CDase concentration,
but maximum activity is reached at 4.3 nM
CDase. This kinetic behavior compares well with
that of lipases (Fig. 4).
The inhibiting potential of several ceramide and
sphingosine analogues was tested at pH 8.5 with 2
mM substrate, 1.6 mM Triton X-100 and 4 mM
CaCl2.
The known inhibitor of human alkaline CDase
D-erythro-MAPP (Bielawska et al., 1996) nor L-
When the substrate concentration was varied at
a fixed enzyme concentration (4.3 nM) at pH 8.5,
maximum CDase activity was observed above 1.5
mM substrate, with an apparent Km of 0.590.1
mM.
The ion-dependence of the CDase reaction was
tested at pH 8.5 with 2 mM substrate and 4.3 nM
CDase. The enzyme displays optimal activity at
Ca2+ concentrations above 7 mM, with a Kd value
of 1.290.4 mM. Ba2+ and Mg2+ can replace
Ca2+ as cofactors for CDase and have Kd values
of 8.590.2 and 9.890.3 mM, respectively. The
maximum activity of CDase in the presence of
Mg2+ is only 36% of that in the presence of
optimal Ca2+ or Ba2+ concentrations. The effect
of Zn2+ was measured at pH 6.5 because of the
low solubility of the Zn2+ ion at higher pH
values. CDase was as active with Zn2+ as with
Ca2+ at pH 6.5. The Kd for the reaction in the
presence of Zn2+ is 1.090.1 mM.
erythro-MAPP inhibit CDase from P. aeruginosa
PA01 even at 20 mM (i.e. ten times the substrate
concentration). N-oleoylethanolamine, an acid
CDase inhibitor (Spinedi et al., 1999 and refer-
ences therein), also did not affect the CDase
activity at 20 mM. CDase activity was not affected
by
L
(−)- or
D
(+)-norephenidrin at 20 mM.
D/L-
sphinganine, the racemic saturated sphingosine
analogue, reduced the CDase activity to 17% at
20 mM. The aparent Ki for
3.390.3 mM under the test conditions.
At 20 mM -erythro-sphingosine and
sphingosine reduced the CDase activity to 3 and
80%, respectively. The apparent Ki for -erythro-
D/L-sphinganine was
D
L-erythro-
D
The turnover of CDase (2.1 nM) with the sub-
strate (2 mM) in 50 mM Tris–HCl, 1.6 mM Triton
X-100, 4 mM CaCl2, pH 8.5 was found to be 5.5
sphingosine was 4.090.4 mM under the test con-
ditions. The apparent Ki for L-erythro-sphingosine
was not determined.

W.F. Nieuwenhuizen et al. / Chemistry and Physics of Lipids 114 (2002) 181–191
189
CDase has a strong chiral preference for the
binding of -erythro-sphingosine compared with
the -enantiomer, as is reflected in the different
inhibiting power of both enentiomers. It is ex-
pected that -sphinganine, like -erythro-sphin-
quenching moiety is linked via a more stable ether
or sulfonic ester linkage.
Our novel assay allows for the high-throughput
screening of modulators of CDase activity with
therapeutic potential.
D
L
D
D
gosine, is the most powerful inhibiting
enantiomer. Since we tested the racemic sphinga-
nine mixture and hence the concentration of the
inhibiting species is lower, this could indicate that
The metabolism of macrophage phospholipids
produce key mediators of inflammation and ma-
jor second messengers that modulate inflamma-
tory responses during sepsis. In a recent report
(Lo et al., 1999) the effect of ceramide and sphin-
gosine on LPS activated rabbit alveolar
macrophages is described. The authors found that
ceramide did not have any effect on TNF produc-
tion or TNF mRNA levels with or without LPS
activation of the macrophages, whereas sphin-
gosine inhibited TNF production and TNF
D
-sphinganine is a stronger CDase inhibitor than
D
-erythro-sphingosine.
4. Discussion
The existing CDase assays make use of radioac-
tive ceramide or a fluorescent ceramide analogue,
and depend on a separation protocol to determine
the ceramide hydrolysis (Misutake et al., 1997;
Tani et al., 1999). Another assay to determine the
ceramide concentration depends on the diglyce-
ride kinase assay, which is questioned as a reliable
assay (Hofmann, 1999; Perry and Hannun, 1999).
In contrast, our novel CDase assay does not
depend on a laborious separation step nor does it
require the error-prone use of a second enzyme.
In the present study we describe the synthesis of
a novel quenched fluorescent CDase substrate.
The versatile synthesis route enables the produc-
tion of different ceramide analogues with different
chiral centers, depending on the chirality of start-
ing compound 3 and the reaction conditions in
step b (see Fig. 2) (Garner et al., 1988; Herold,
mRNA
expression
in
LPS
stimulated
macrophages. The LPS induced NFkB activity
was reduced by sphingosine. Thus sphingosine, in
contrast to ceramide, down-regulates macrophage
activation induced by LPS stimulation.
These data may indicate that the CDase
product sphingosine helps the pathogenic Pseu-
domonas species to escape macrophages, and that
CDase may be regarded as a virulence factor in
combination with sphingomyelinase.
Since the Pseudomonas CDase is not inhibited
by the known mammalian CDase inhibitors,
D-
erythro MAPP (Bielawska et al., 1996) or N-
oleoyl ethanolamine (Spinedi et al., 1999), and
because the bacterial enzyme is the only divalent
cation-dependent CDase known to date, it is
highly likely that bacterial CDase has a different
enzymatic mechanism than its mammalian coun-
terparts. This finding together with our assay
offers the opportunity to screen for, and develop,
specific inhibitors for bacterial CDase that do not
interfere with the mammalian sphingolipid
metabolism, and which may help to fight Pseu-
domonas infections.
1988). We chose to synthesize the
D-erythro form
because its chirality is identical with that of natu-
ral ceramide.
Ceramide analogue 10 enables the real time
measurement of CDase activity and is a valuable
tool in the characterization of CDases. It allows
for rapid testing of assay conditions and the effect
of different cofactors much faster, and more reli-
ably than with the existing CDase assays.
The ester bond between the 3,5-dinitrobenzoyl
group and the sphingosine moiety is unstable at
high pH, and may be susceptible to hydrolysis by
esterases which may be present in more complex
systems (i.e. intact cells). Therefore, we plan to
synthesize ceramide analogues in which the
Acknowledgements
This research was financially supported by the
Dutch Organisation for Scientific Research (CW/
STW-project 349-5362). We gratefully acknowl-
edge J.W.H. Peeters of the Swammerdam

190
W.F. Nieuwenhuizen et al. / Chemistry and Physics of Lipids 114 (2002) 181–191
1998. Cloning and characterization of the full-length
cDNA and genomic sequences encoding murine acid ce-
ramidase. Genomics 50, 267–274.
Institute for Life Science, Department of
Biomacromolecular Mass Spectrometry of the
Amsterdam University, for recording of the mass
spectra.
Li, C.-M., Park, J.-H., He, X., Levy, B., Chen, F., Arai, K.,
Adler, D.A., Disteche, C.M., Koch, J., Sandhoff, K.,
Schuchman, E.H., 1999. The human acid ceramidase gene
(ASAH): Structure, chromosomal location, mutation anal-
ysis, and expression. Genomics 62, 223–231.
Linke, T., Wilkening, G., Sadeghlar, F., Moczall, H.,
Bernardo, K., Schuchman, E., Sandhoff, K., 2001. Interfa-
cial regulation of acid ceramidase activity: Stimulation of
ceramide degradation by lysosomal lipids and sphingolipid
activator proteins. J. Biol. Chem. 276 (8), 5760–5768.
Liu, G., Kleine, L., He´bert, R.L., 1999. Advances in the signal
transduction of ceramide and related sphingolipids. Crit.
Rev. Clin. Lab. Sci. 36 (4), 511–573.
Lo, C.-J., Fu, M., Lo, F.R., Cryer, H.G., 1999. Macrophage
TNF mRNA expression induced by LPS is regulated by
sphingomyelin metabolites. Shock 11 (6), 411–415.
Mao, C., Xu, R., Bielawska, A., Obeid, L.M., 2000. Cloning
of an alkaline ceramidase from Saccharomyces cere6iseae.
J. Biol. Chem. 275 (10), 6876–6884.
References
Azuma, H., Tamagaki, S., Ogino, K., 2000. Stereospecific total
syntheses of sphingosine and its analogues from L-serine. J.
Org. Chem. 65, 3538–3541.
Bar, J., Linke, T., Ferlinz, K., Neumann, U., Schuchman,
E.H., Sandhoff, K., 2001. Molecular analysis of acid ce-
ramidase deficiency in patients with Farber disease. Hum.
Mutat. 17 (3), 199–209.
Bawab, S.E., Roddy, P., Qian, T., Bielawska, A., Lemasters,
J.J., Hannun, Y.A., 2000. Molecular cloning and charac-
terization of a human mitochondrial ceramidase. J. Biol.
Chem. 275 (28), 21508–21513.
Bernardo, K., Hurwitz, R., Zenk, T., Desnick, R.J., Ferlinz,
K., Schuchman, E.H., Sandhoff, K., 1995. Purification,
characterization, and biosynthesis of human acid cerami-
dase. J. Biol. Chem. 270 (19), 11098–11102.
Mathias, S., Pena, L.A., Kolesnick, R.N., 1998. Signal trans-
duction of stress via ceramide. Biochem. J. 335, 465–480.
Misutake, S., Kita, K., Okino, N., Ito, M., 1997.
Bielawska, A., Greenberg, M.S., Perry, D., Jayadev, S., Shay-
[
14C]Ceramide synthesis by sphingolipid ceramide N-dea-
man, J.A., McKay, C., Hannun, Y.A., 1996. (1S,2R)-D-
cylase: New assay for ceramidase activity detection. Anal.
Biochem. 247, 52–57.
erythro-2-(N-myristoylamino)-1-phenyl-1-propanol as an
inhibitor of ceramidase. J. Biol. Chem. 271 (21), 12646–
12654.
Nilsson, A., Duan, R.-D., 1999. Alkaline sphingomyelinases
and ceramidases of the gastrointestinal tract. Chem. Phys.
Lipids 102, 97–105.
Duffin, G.R., Ellames, G.J., Hartmann, S., Herbert, J.M.,
Smith, D.I., 2000. Practical syntheses of [13C]- and [14C]-
labelled glucosphingolipids. J. Chem. Soc., Perkin Trans I.,
(14) 2237–2242.
Garner, P., Min Park, J., 1987. The synthesis and configura-
tional stability of differentially protected b-hydroxy-a-
amino aldehydes. J. Org. Chem. 52, 2361–2364.
Garner, P., Park, J.M., Malecki, E., 1988. A stereodivergent
Okino, N., Tani, M., Imayama, S., Ito, M., 1998. Purification
and characterization of a novel ceramidase from Pseu-
domonas aeruginosa. J. Biol. Chem. 273 (23), 14368–14373.
Okino, N., Ichinose, S., Omori, A., Imayama, S., Nakamura,
T., Ito, M., 1999. Molecular cloning, sequencing, and
expression of the gene encoding alkaline ceramidase from
Pseudomonas aeruginosa. J. Biol. Chem. 274 (51), 36616–
36622.
synthesis of
D-erythro-sphingosine and D-threo-sphingosine
from -serine. J. Org. Chem. 53, 4395–4398.
L
Perry, D.K., Hannun, Y.A., 1999. The use of diglyceride
kinase for quantifying ceramide. TIBS 24, 226–227.
Sharma, K., Shi, Y., 1999. The yins and yangs of ceramide.
Cell Res. 9, 1–10.
Gibson, F.S., Bergmeier, S.C., Rapoport, H., 1994. Selective
removal of an N-BOC protecting group in the presence of
a tert-butyl ester and other acid-sensitive groups. J. Org.
Chem. 59, 3216–3218.
Spence, M.W., Beed, S., Cook, H.W., 1986. Acid and alkaline
ceramidases of rat tissues. Biochem. Cell Biol. 64, 400–
404.
Hassler, D.F., Bell, R.M., 1993. Ceramidases:Enzymology and
metabolic roles. Adv. Lipid Res. 26, 49–57.
Herold, P., 1988. Synthesis of
D
-erythro- and
D-threo-Sphin-
Spiegel, S., Merrill, A.H. Jr, 1996. Sphingolipid metabolism
and cell growth regulation. FASEB J. 10, 1388–1397.
Spinedi, A., Di Bartolomeo, S., Piacentini, M., 1999. N-
oleoylethanolamine inhibits glucosylation of natural ce-
ramides in CHP-100 neuroepithelioma cells: Possible
implications for apoptosis. Biochem. Biophys. Res. Com-
mun. 255, 456–459.
Tani, M., Kita, K., Komori, H., Nakagawa, T., Ito, M., 1998.
Enzymatic synthesis of omega-amino-ceramide: Prepara-
tion of a sensitive fluorescent substrate for ceramidase.
Anal. Biochem. 263, 183–188.
gosine derivatives of
362.
Hofmann, K., 1999. Reply to Kolesnick and Hannun, and
Perry and Hannun. TIBS 24, 227.
Koch, J., Ga¨rtner, S., Li, C.-M., Quintern, L.E., Bernardo, K.,
Levran, O., Schnabel, D., Desnick, R.J., Schuchman, E.H.,
Sandhoff, K., 1996. Molecular cloning and characteriza-
tion of a full-length complementary DNA encoding human
acid ceramidase. J. Biol. Chem. 271 (51), 33110–33115.
Li, C.-M., Hong, S.-B., Kopal, G., He, X., Linke, T., Hou,
W.-S., Koch, J., Gatt, S., Sandhoff, K., Schuchman, E.H.,
L-Serine. Helv. Chim. Acta 71, 354–

W.F. Nieuwenhuizen et al. / Chemistry and Physics of Lipids 114 (2002) 181–191
191
Tani, M., Okino, N., Misutake, S., Ito, M., 1999. Specific and
sensitive assay for alkaline and neutral ceramidases involv-
ing C12-NBD-ceramide. J. Biochem. 125, 746–749.
Tani, M., Okino, N., Mori, K., Tanigawa, T., Izu, H., Ito, M.,
2000. Molecular cloning of the full-length cDNA encoding
mouse neutral ceramidase. J. Biol. Chem. 275 (15), 11229–
11234.
porcine epidermis. FEBS Lett. 268 (1), 110–112.
Yada, Y., Higuchi, K., Imokawa, G., 1995. Purification and
biochemical characterization of membrane-bound epider-
mal ceramidases from guinea pig skin. J. Biol. Chem. 270
(21), 12677–12684.
Yin, J., Liu, H., Pidgeon, C., 1998. Synthesis and immobiliza-
tion of ceramide analogs on silica particles. Bioorg. Med.
Chem. Lett. 8, 179–182.
Wertz, P.W., Downing, D.T., 1990. Ceramidase activity in