Development of an acid ceramidase activity-based probe

Article abstract of DOI:10.1039/c5cc00356c

Acid ceramidase is responsible for the ultimate step in the catabolism of (glyco)sphingolipids by hydrolysis of ceramide into sphingosine and free fatty acid. Deficiency in acid ceramidase is the molecular basis of Farber disease. Here we report the synthesis and characterization of an activity-based acid ceramidase probe.

Full text of DOI:10.1039/c5cc00356c

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Development of an acid ceramidase activity-based
probe†
Cite this: Chem. Commun., 2015,
51, 6161
Cecile M. J. Ouairy,a Maria J. Ferraz,ab Rolf G. Boot,ab Marc P. Baggelaar,a
´
Mario van der Stelt,a Monique Appelman,b Gijsbert A. van der Marel,a
Bogdan I. Florea,a Johannes M. F. G. Aerts*ab and Herman S. Overkleeft*a
Received 14th January 2015,
Accepted 13th February 2015
DOI: 10.1039/c5cc00356c
www.rsc.org/chemcomm
Acid ceramidase is responsible for the ultimate step in the cata-
bolism of (glyco)sphingolipids by hydrolysis of ceramide into sphin-
gosine and free fatty acid. Deficiency in acid ceramidase is the
molecular basis of Farber disease. Here we report the synthesis and
characterization of an activity-based acid ceramidase probe.
Human acid ceramidase (ACase) is a heterodimeric glyco-
protein that belongs to the N-terminal nucleophile (Ntn) hydro-
lase family. This cysteine amidase, deficient in the lysosomal
storage disorder, Farber disease, is instrumental in the final
step of lysosomal degradation of (glyco)sphingolipids by hydro-
lysing ceramide into free fatty acid and sphingosine1 (Fig. 1(1)).
Since the identification of this enzyme half a century ago by
Gatt,2 several studies have been conducted to elucidate its
processing and catalytic mechanism.3–6 Our research focuses
on the metabolism of (glyco)sphingolipids to develop new
diagnostic and therapeutic strategies for application in the
treatment of inherited sphingolipidoses including Gaucher
disease. This recessively inherited lysosomal storage disorder
is caused by a deficiency of glucocerebrosidase and results in
the accumulation of its substrate glucosylceramide, most promi-
nently in macrophages. Recently we reported that the accumulating
glucosylceramide is partially deacylated to glucosylsphingosine,
Fig. 1 (1) ACase mediated hydrolysis of (glycosyl)ceramide. (2) Structures
of vinyl sulfone (A) and carmofur (B) analogues as potential activity-based
acid ceramidase probes.
which is toxic and may be involved in the progress of the disease.7 repertoire of activity-based probes to include ACase, we chose to
Deacylated glycosphingolipids were also previously observed in develop an activity-based ceramidase probe. We here disclose our
Fabry disease, another glycosphingolipidosis. In Fabry tissue, high first results in this direction by the successful development of the
levels of globotriaosylceramide are present in patients’ plasma.8 first (to the best of our knowledge) activity-based probe (ABP) for
These findings raise the question whether acid ceramidase is the the in vitro and in vivo visualization of human acid ceramidase.
enzyme responsible for the deacylation of glycosylceramide. With
At the onset of our studies we considered two design
the aim to establish this, and more in general to expand the strategies that might lead to the desired ceramidase ABP. In
the first instance, we chose to design mimics of its natural
a Leiden Institute of Chemistry, Leiden University, P. O. Box 9502, 2300 RA Leiden,
substrate ceramide (Fig. 1(2A), ABP 1 and 2). Replacement of
the amide in this natural substrate by a vinyl sulfone may yield
a compound that reacts covalently and irreversibly with ACase,
which mediates hydrolysis by an N-terminal cysteine thiol
acting as a nucleophile.6 Activity-based cysteine protease
probes, which have a thiol-reactive Michael acceptor at the
position occupied by the carbonyl of the scissile amide bond in
The Netherlands. E-mail: h.s.overkleeft@chem.leidenuniv.nl,
j.m.f.g.aerts@lic.leidenuniv.nl; Tel: +31715274342
b Department of Medical Biochemistry, Academic Medical Centre, University of
Amsterdam, Meibergdreef 15, 1105 AS Amsterdam, The Netherlands
† Electronic supplementary information (ESI) available: Experimental and analy-
tical details on the synthesis of the compounds, and experimental biochemical
procedures. See DOI: 10.1039/c5cc00356c
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun., 2015, 51, 6161--6163 | 6161

View Article Online
Communication
ChemComm
the natural substrate, have proven their merits as activity-based
cysteine protease probes.9,10 At the same time, we considered
carmofur as a starting point for ABP development (Fig. 1(2B)).
Carmofur is a 5-fluorouracil (5-FU) releasing drug currently
used in the clinic to treat colorectal cancers and has been
recently reported by Piomelli and co-workers as a potent
ceramidase inhibitor.11 Probes based on both scaffolds were
synthesized as summarized below (detailed information can be
found in the ESI†).
Vinyl sulfone 1 was prepared via cross-metathesis of the
unsaturated ketone 7 and the long hydrophobic alkene 8
(Scheme 1). Subsequent Corey–Bakshi–Shibata reduction of
alpha-beta-unsaturated ketone 9 resulted in the corresponding
allylic alcohol, which was subjected to Birch reduction to obtain
diol 10. After protection of the diol function using DTBS
ditriflate, removal of the acetal followed by an HWE reaction
with the ylide prepared from 11, yielded vinyl sulfone 12. Final
deprotection using HF3-NEt3 afforded the desired vinyl sulfone
1 (Scheme 1). A similar strategy (see details in the ESI†) was
pursued to obtain vinyl sulfone 2.
Scheme 2 Reagents and conditions: (i) pyridine, rt, overnight, 84%. (ii) 15
(1 eq.), CuSO4 (0.2 eq.), sodium ascorbate (1.6 eq.), sonication, 6 h, 25%.
Fluorescent carmofur derivative 5 was synthesized by adap-
tation of the literature procedure towards the parent com-
pound.12 Reaction between 5-fluorouracil 13 and isocyanate
14 afforded carboxamide 3, which was ligated to Bodipy-alkyne
15 (ref. 13) by Cu(I)-catalysed azide–alkyne[2+3]cycloaddition
to afford the desired probe 5 (Scheme 2). 2,4-Dioxopyrimidine-
1-carboxamide 4 was obtained following a similar scheme
(see ESI† for a detailed synthesis).
Fig. 2 (A) Labeling of ASAH1 in Farber fibroblasts stably expressing WT
ASAH1 using ABP 5 (2.5 mM). Competition experiment performed with
carmofur (10 mM). (B) ABPP using 5 (2.5 mM and 1 mM) with different ASAH1
constructs at the same protein concentration (0.47 mg mLÀ1). Competi-
tion experiment with 1 (10 mM), 2 (10 mM), 3 (10 mM) and 4 (10 mM) using WT
ASAH1 constructs at the same protein concentration (0.47 mg mLÀ1).
Having the fluorescent probe 5 in hand, we assessed its
effect on human acid ceramidase in extracts of Farber fibroblasts
stably expressing the recombinant wild type acid ceramidase.
Incubation with 5 (2.5 mM) in a citric acid buffer (pH 4.5) at 37 1C applied (the presence of beta-mercaptoethanol).5,6 When carmo-
for 30 minutes followed by analysis by SDS-PAGE and fluores- fur (10 mM) was added prior to the incubation with 5 the band was
cence scanning revealed a fluorescent signal at approximately absent (Fig. 2A). Upon pre-incubation with carmofur derivatives 3
40 kDa. This molecular weight corresponds to the catalytically (10 mM) or 4 (10 mM), no band could be detected indicating that
glycosylated beta-subunit of acid ceramidase, which is separated these compounds are able to compete for 5 (Fig. 2B). However
from the alpha-subunit (13 kDa) under the reducing conditions upon pre-incubation with vinyl sulfone 1 or 2 (10 mM final
concentration), the fluorescent signal was still detected (Fig. 2B).
Since we obtained no indication that the ceramide mimics
interacted with human acid ceramidase we focused our attention
on our carmofur analogues.
A pull down experiment was performed to establish whether
the observed protein in Gaucher spleen extracts corresponds to
acid ceramidase. For this purpose biotinylated carmofur deri-
vative 6 was synthetized following a sequence similar to the one
depicted Scheme 2. After incubation with 6 (200 mM) in a citric
acid buffer (pH = 4.5) at 37 1C for 30 min, affinity purification
using streptavidin beads and tryptic digestion, mass spectro-
metric analysis revealed five peptidic fragments corresponding
to ACase. However, we were unable to identify the peptide with
Scheme 1 Reagents and conditions: (i) 8 (2 eq.), Hoveyda–Grubbs sec-
ond generation catalyst (5 mol%), DCM, AcOH (20 mol%), 40 1C, overnight,
78%. (ii) BH3ÁMe2S (2.5 eq.), (R)-MeCBS (1 eq.), THF, 0 1C, 1 h 30 min, 78%.
(iii) NH3 (l), Li, THF, 0 1C, 82%. (iv) t-Bu2Si(OTf)2 (1.8 eq.), 2,6-lutidine (3 eq.),
DMF, 0 1C, 2 h, 68%. (v) HCl (37%, aq.)/THF (1 : 3), rt, 3 h. (vi) (a) 11 (1.5 eq.),
NaH (1.5 eq.), 0 1C then rt, 1 h; (b) THF, 0 1C then rt, overnight, 34%
(2 steps). (vii) HF-NEt3 (1.9 eq.), pyridine, rt, 2 h 30 min, 69%.
the catalytic cysteine residue.
To unambiguously confirm the validity of 5 as APB and
establish its binding to the ACase nucleophile, we mutated
the catalytic cysteine to a serine (ASAH1-C122S) and stably
expressed this in Farber fibroblasts lacking endogenous
acid ceramidase. After incubation with 5 (2.5 or 1 mM) and
6162 | Chem. Commun., 2015, 51, 6161--6163
This journal is ©The Royal Society of Chemistry 2015

View Article Online
ChemComm
Communication
endogenous acid ceramidase from healthy and Gaucher spleen
tissue can be detected with carmofur-based ABPs. We found
that the activity of the enzyme appeared to be considerably
higher in Gaucher disease patients compared to tissue from
healthy individuals. Our findings thus represent a good starting
point in the study of this important enzyme and we are
currently investigating the implications of ceramidase in the
onset and progression of Gaucher disease and Farber disease.
Fig. 3 ABPP using 5 (2.5 mM) and competition with carmofur (10 mM) in
tissues (healthy and Gaucher spleen) at the same protein concentration
(1.6 mg mLÀ1).
Notes and references
1 C. R. Gault, L. M. Obeid and Y. A. Hannun, Adv. Exp. Med. Biol.,
2010, 688, 1.
2 S. Gatt, J. Biol. Chem., 1963, 238, 3131.
3 K. Bernardo, R. Wurhithz, T. Zenk, R. J. Desnick, K. Ferlinz,
E. H. Schuchman and K. Sandhoff, J. Biol. Chem., 1995, 270, 11098.
4 X. He, N. Okino, R. A. D. S. G. Dhami, H. Schulze, K. Sandhoff and
E. H. Schuchman, J. Biol. Chem., 2003, 278, 32978.
5 H. Schulze, U. Schepers and K. Sandhoff, Biol. Chem., 2007,
388, 1333.
SDS-PAGE, the fluorescent signal was absent in extracts expres-
sing this mutant form of acid ceramidase (Fig. 2B) while
prominently present in cells expressing the wild type enzyme.
These data underscore that ACase belongs to the Ntn-hydrolase
family with a cysteine residue as the active site nucleophile.6,14–16
As the next objective we set out to demonstrate the use of
ABP 5 to identify acid ceramidase in tissues homogenates. For
this purpose, we took human spleen from healthy controls and
Gaucher patients. When the homogenate was treated with ABP
5 (2.5 mM) for 1 h at 37 1C a band at the same height as
previously reported was observed (Fig. 3). Pre-incubation with
carmofur (10 mM) blocked enzymatic activity in both cases.
These data showed that endogenous levels of acid ceramidase
can be detected with 5 and thus validate this probe as an
effective activity-based acid ceramidase probe. Of note, in
extracts derived from a spleen of a type 1 Gaucher patient a
relative large quantity of active acid ceramidase molecules was
detected as compared to normal spleens.
6 N. Shtraizent, J.-H. Park, C. He, R. Shalgi and E. H. Schuchman,
J. Biol. Chem., 2008, 283, 11253.
7 N. Dekker, L. van Dussen, C. E. M. Hollak, H. Overkleeft, S. Scheij,
K. Ghauharali, M. J. van Breemen, M. F. Ferraz, J. E. M. Groener,
M. Maas, F. A. Wijburg, D. Speijer, A. Tylki-Szymanska, P. K. Mistry,
R. G. Boot and J. M. Aerts, Blood, 2011, 118, e118.
8 J. M. Aerts, J. E. Groener, S. Kuiper, W. E. Donker-Koopman, A. Strijland,
R. Ottenhoff, C. van Roomen, M. Mirzaian, G. E. Wijnburg, G. E.
Linthorst, A. C. Veder, S. M. Rombach, J. Cox-Brinkman, P. Somerhaju,
R. G. Boot, C. E. M. Hollak, R. O. Brady and N. J. Poorthuis, Proc. Natl.
Acad. Sci. U. S. A., 2008, 105, 2812.
9 M. J. Evans and B. F. Cravatt, Chem. Rev., 2006, 106, 3279.
10 D. Kato, K. M. Boatright, A. B. Berger, T. Nazif, G. Blum, C. Ryan,
K. A. H. Chehade, G. S. Salvesen and M. Bogyo, Nat. Chem. Biol.,
2005, 1, 33.
11 N. Realini, C. Solorzano, C. Pagliuca, D. Pizzirani, A. Armirotti,
R. Luciani, M. P. Costi, T. Bandiera and D. Piomelli, Sci. Rep., 2013,
3, 1035.
12 D. Pizzirani, C. Pagliuca, N. Realini, D. Branduardi, G. Bottegoni,
M. Mor, F. Fabio Bertozzi, R. Scarpelli, D. Piomelli and T. Bandiera,
J. Med. Chem., 2013, 56, 3518.
In conclusion, we have developed compounds 5 and 6 as the
first ABPs that efficiently target acid ceramidase. Using the
antineoplastic drug carmofur we have been able to develop a 13 M. Verdoes, U. Hillaert, B. I. Florea, M. Sae-Heng, M. D. P. Risseeuw,
D. V. Filippov, G. A. van der Marel and H. S. Overkleeft, Bioorg. Med.
Chem. Lett., 2007, 17, 6169.
5-FU analogue to visualize (with 5) and identify (with 6) the
desired lysosomal enzyme. Unlike some cysteine protease inhi-
14 C. Oinonen and J. Rouvinen, Protein Sci., 2000, 9, 2329.
bitors and ABPs,17–19 the vinyl sulfone moiety is not reacting 15 J. Pei and N. V. Grishin, Protein Sci., 2003, 12, 1131.
¨
16 S. F. Altschul, T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang,
with the active-site cysteine in the case of ceramidase mimics,
at least when introduced into ceramide analogues. One could
envision the introduction of alternative electrophilic traps in
the ceramide scaffold and that have intrinsic reactivity towards
cysteine thiols (alternative Michael acceptors, halomethyl
ketones, epoxyketones). We have not studied this because
W. Miller and D. J. Lipman, Nucleic Acids Res., 1997, 25, 3389.
17 J. T. Palmer, D. Rasnick, J. L. Klaus and D. Bromme, J. Med. Chem.,
1995, 38, 3193.
18 M. D. Mertens, J. Schmitz, M. Horn, N. Furtmann, J. Bajorath,
M. Mares and M. Gu¨tschow, ChemBioChem, 2014, 15, 955.
19 L. Mendieta, A. Pic, T. Tarrag, M. Teixid, M. Castillo, L. Rafecas,
A. Moyano and E. Giralt, ChemMedChem, 2010, 5, 1556.
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun., 2015, 51, 6161--6163 | 6163