Angewandte
Communications
Chemie
model of how the enzyme could be active at the neutral pH of
the cell surface has not been reported.[13] A more recent
report connects lysosomal ASM to the mediation of lysoso-
mal stability and thus the survival of cancer cells.[9,14] In the
past, we have developed several types of inhibitors with the
aim to clarify the role of ASM in biology and human
diseases.[15] Very recently, the X-ray structure of the enzyme in
complex with one of our inhibitors has been published,
enabling rational drug design.[16] Consequently, screening and
evaluation of newly synthesized potential ASM inhibitors will
soon demand for an efficient (in vivo) assay. Herein, we have
developed FRET probes that enable the in situ non-destruc-
tive detection of enzymatic cleavage by fluorescence imaging.
Encouraged by previously developed lipid probes, where
the combination of Nile Red (NR) and nitrobenzoxadiazole
(NBD) was shown to provide efficient FRET in a lipid
environment,[17] we designed a first sphingomyelin analogue
with these two fluorophores (Figure 1B). NBD is commonly
used as a covalent label for lipids despite the fact that it can
cause loop formation with acyl chains in phospholipid
membranes owing to its significant polarity.[18] To mimic
natural sphingomyelin in terms of its polarity, conformation,
and steric bulk, small and polar NBD was chosen for head
group labeling,[19] and the less polar NR moiety was used to
label the acyl chain of the envisioned sphingomyelin ana-
logue. The synthesis consisted of three steps: 1) stereoselec-
tive synthesis of the sphingosine core, 2) asymmetric phos-
phodiester formation, and 3) attachment of the fluorophores.
Starting from l-serine, a decorated sphingosine 3 with
natural stereochemistry was synthesized in eight steps in
analogy to previous reports.[20] The subsequent introduction
of the aminoethyl phosphate moiety proved to be challenging
as the use of hydrogen phosphonates and phosphoramidites
as well as phosphorylation with POCl3 seemed to be
incompatible with the present functional groups. However,
upon employing the reactive phosphorodichloridate reagent
4,[21] the desired unsymmetric phosphodiester 5 was isolated
in good yield (Scheme 1). This inexpensive reagent has
received little attention since its development,[21] but proved
to be an efficient precursor for phosphosphingolipid synthesis
in our hands. The synthesis was completed by hydrazinolysis
of the phthalimide and by nucleophilic aromatic substitution
Scheme 1. Synthesis of ASM probes. a) 1) Pyridine, CH2Cl2, 08C!RT,
overnight, 2) aq. NaHCO3, RT, 30 min, 94%; b) H4N2·H2O, MeOH,
96%; c) NBDCl, DIPEA, MeOH, RT, 150 min; d) HCl, MeOH, 708C,
3 h, quant.; e) R(CH2)5COOSu, DIPEA, CH2Cl2, 08C!RT, overnight,
34–80%. Boc=tert-butoxycarbonyl, DIPEA=diisopropylethylamine,
NBD=7-nitrobenzo-2-oxa-1,3-diazol-4-yl, Su=N-succinimidyl.
expected, the probe exhibited exclusively acceptor (NR)
fluorescence at 624 nm upon excitation of the donor dye
(NBD) at 430 nm while the absence of donor (NBD) emission
suggested the probe to have high FRET efficiency. Subse-
quently, recombinant human ASM was added to the micellar
solution of the probe, and spectra were recorded after
different time intervals. We expected that the emission of
the acceptor (NR) would decrease owing to enzymatic
cleavage of the FRET pair, with a concomitant increase in
the fluorescence of the donor dye (NBD). Indeed, upon
incubation with ASM, we measured a rapid decrease in
acceptor fluorescence. However, we did not observe the
anticipated rise in donor (NBD) fluorescence. Further
characterization of the expected cleavage product, NBD-
substituted aminoethyl phosphate, revealed its weak fluores-
cence in aqueous environments. In contrast, the NBD
fluorescence of the intact probe was much stronger when
embedded in more hydrophobic micelles. Therefore, we
concluded that the expected increase in donor (NBD)
fluorescence upon cleavage of the probe was not observed
owing to quenching of the polar NBD in aqueous solution
upon cleavage and liberation into the aqueous environment.
Ultimately, a ratio change (RC) of 2.6 was determined as the
responsiveness of the probe, a moderate result when com-
pared to other lipid FRET probes.[17a]
Next, we wondered whether we could exploit the phe-
nomena described above. The inherently disadvantageous
quenching of the donor upon cleavage might be turned into
an asset if the NBD moiety was turned into the FRET
acceptor. Based on recent publications,[23] 7-methoxycou-
marin-3-carboxylate (MCC) was chosen as a suitable FRET
donor for NBD, and incorporated into the fatty acid as
a replacement of the NR fatty acid. In contrast with the first
probe, the MCC-containing probe 2 indeed exhibited a supe-
rior performance upon incubation with recombinant human
ASM. A pronounced decrease (ꢀ87%) in the acceptor
fluorescence was accompanied by the appearance of intense
donor fluorescence (+ 900%), leading to an 80-fold ratio
change (Figure 2A). Organic extracts[24] of the reaction
mixture were analyzed by TLC and UPLC-MS and further
ꢀ
of NBD Cl with the liberated primary amine. In parallel,
a hexanoic acid bearing an NR (R1) substituent was obtained
following a procedure reported previously.[17b,22] Alterna-
tively, 6-aminohexanoic acid was coupled to 7-methoxycou-
marin-3-carboxylate (MCC, R2) using a commercially avail-
able MCC active ester. The two acids were then activated as
N-hydroxysuccinimide (HOSu) esters under Steglich condi-
tions. Following global deprotection of 5, the two labeled
acids were coupled to the NBD-substituted phosphosphingo-
sine to furnish probes 1 and 2 (Scheme 1). Using similar
methods, the respective monolabeled ceramides (NR-Cer and
MCC-Cer), which represent the expected ASM cleavage
products from probes 1 and 2, were synthesized, too (see the
Supporting Information for details).
Fluorometric characterization of probe 1 containing the
NBD/NR FRET pair was performed in aqueous buffer after
the probe had been embedded into Triton X-100 micelles. As
2
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!