A small-molecule FRET probe to monitor phospholipase A2 activity in cells and organisms

Article abstract of DOI:10.1002/anie.200500751

(Chemical Equation Presented) Watching live: The FRET reporter PENN/SATE (see picture; FRET = fluorescence resonance energy transfer) is membrane permeant and enters cells and small organisms by using a prodrug approach. Once inside cells the reporter is

Full text of DOI:10.1002/anie.200500751

Communications
Fluorescent Probes
application of a small-molecule FRET probe for phospholi-
pase A2 (PLA2). A number of fluorescent assays for monitor-
ing PLA2 activity have been previously described, albeit
without ratio imaging.[6,7] PLA2s are a group of key enzymes
in intracellular signaling and play an important role in
inflammation. One of the main functions of PLA2 is to
release arachidonic acid from phospholipids as the key
precursor for the synthesis of prostaglandins and leuko-
trienes.[8,9] Most of the enzymes exhibit activities on mem-
brane surfaces, especially those of the perinuclear region.[10–12]
The design of the probe PENN/SATE (1) was based on a
phospholipid (Figure 1). As there are many different phos-
DOI: 10.1002/anie.200500751
A Small-Molecule FRET Probe To Monitor
Phospholipase A2 Activity in Cells and
Organisms**
Oliver Wichmann, Jochen Wittbrodt, and
Carsten Schultz*
Biochemical experiments are shifting increasingly from the
test tube to living cells. This development is mainly made
possible by fluorescent probes[1,2] that, for example, replace
radioactivity-based assays. Fluorescent probes usually allow
real-time monitoring of events and, when located properly,
also give spatial information about the monitored event.
Accordingly, the demand for fluorescent probes has vastly
increased over the past decade. Although the rapid develop-
ment of fluorescent proteins from jellyfish and corals fosters
the construction of genetically encoded probes,[3] small-
molecule probes are mostly provided by chemical methods.[4]
The majority of small-molecule probes are ion sensors. A
fewsubstrates that exhibit fluorescence resonance energy
transfer (FRET) have been synthesized and used successfully
in the past to monitor enzyme activities.[5] One advantage of
these probes is their accumulative nature, which, combined
with the amplification effect of an enzymatic reaction, gives
enormously high sensitivity even if substrate properties are
moderate.[5] The disadvantage is that probes are sacrificed in
the process, and the lack of reversibility limits the observation
of dynamic processes. On the other hand, small, specifically
tailored molecules permit the choice of a larger variety of
fluorophores with physical properties more diverse than those
provided by fluorescent proteins. Furthermore, small-mole-
cule probes should generally give much larger ratio changes
owing to the close proximity of the fluorophores. This might
be important for applications with limited sensitivity, such as a
platereader. Herein we present the synthesis and biological
Figure 1. Structure of the PLA2 reporter 1. The FRET donor fluorophore
NBD is attachedto the tip of the sn-1 chain, the acceptor Nile redto
the butyric acidat sn-2. The chargedheadgroup moieties are masked
by bioactivatable S-acetyl-2-thioethyl (SATE) groups. Enzymatic hydroly-
sis of the thioester andrelease of thiirane is known to liberate charged
groups.[13] The arrow indicates where PLA2 cleaves the probe.
pholipases with various substrate specificities, we picked a
phosphatidylethanolamine (PE) backbone in which the ester
in the sn-1 position was altered to a nonhydrolyzable ether,
thus eliminating the substrate properties for PLA1 and
unspecific lipases in this position. Another important factor
was the selection and attachment of the two fluorescent dyes.
In our search for maximal overlap between FRET donor
emission and acceptor excitation, we chose the pair 7-
nitrobenzo-2-oxa-1,3-diazole amine (NBD) and the 2-hy-
droxy derivative of 9-diethylamino-5H-benzo[a]phenoxazin-
5-one (Nile red). Their optical properties were almost ideally
suited for our purposes (Figure 2). Furthermore, both fluo-
rophores are, under physiological conditions, uncharged and
thus lipophilic, which is important for insertion of the dyes
into the lipid bilayer of cells.
[*] O. Wichmann, Dr. C. Schultz
European Molecular Biology Laboratory
Gene Expression Programme
Meyerhofstrasse 1, 69117 Heidelberg (Germany)
Fax: (+49)6221-387-206
E-mail: schultz@embl.de
Nile red was originally used in histology to stain lipid
droplets. To make the fluorophore amenable to chemical
attachment, functionalized derivatives with hydroxy groups in
the 2- or 3-position were previously prepared.[14] We followed
a modified procedure to synthesize the 2-hydroxy-Nile red
ether attached to butyric acid (NRBA).[14] The latter was
introduced in the sn-2 position of the probe. We intended to
adjust the position of the two fluorophores by differing the
length of the fatty acid chain. As a result of this, the double
bonds of Nile red were more likely to mimic those of
arachidonic acid.[15] Previous probes with identical fatty acid
chain lengths, showed good quenching of the donor fluores-
Dr. J. Wittbrodt
European Molecular Biology Laboratory
Developmental Biology Programme
Meyerhofstrasse 1, 69117 Heidelberg (Germany)
[**] The authors thank Heike Stichnoth andNicole Heath for providing
cells andErika Grzebisz for the supply of Medaka eggs. We
gratefully acknowledge Dr. Christina Leslie, University of Colorado,
Boulder, for a plasmid of phospholipase A2 g, andDr. Michael Gelb,
University of Washington, Seattle, for a plasmidof phospholipase A 2
a, purifiedcytosolic phospholipase A 2 a enzyme, andhelpful
discussions.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
508
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 508 –512

Angewandte
Chemie
enter the cell. As cytosolic PLA2 usually acts at perinuclear
membranes inside cells, a membrane-permeant derivative of
the probe was necessary. We therefore masked the charged
groups of the probe with SATE groups (Figure 1). These
groups are well known bioactivatable phosphate-protecting
groups that are mostly used to improve the bioavailability of
antiviral nucleotides.[13,17] SATE groups can be introduced
through the use of premanufactured SATE-protected phos-
phoamidite reagents, which, after coupling to the alcohol, are
readily oxidized to uncharged phosphotriesters.[18] Through
this method, all the phospholipid headgroups can be coupled
to the glycerol backbone in one step. This methodology
appears to be advantageous over other prodrug approaches,
such as masking with acetoxymethyl ethers, which not only
require less-mild conditions for their introduction[19] but lead
Figure 2. Excitation (c) andemission ( a) spectra of NBD (gray)
andNile red(black) in methanol. Spectra of sonicatedTriton X-100
vesicles (3 mm) were almost identical. Overlap of NBD emission and
Nile redexcitation was perfect to permit FRET. Emission spectra were
recorded with excitation at 458 (NBD) and 543 nm (Nile red) and the
excitation spectra were recorded with detection at 540 nm and 640 nm,
respectively.
to modification of the Nile red dye[29]
.
The synthesis started from enantiomerically pure 2,3-
isopropylidene-sn-glycerol (2; Scheme 1). The sn-1 hydroxy
group was alkylated with 1,12-dibromododecane and the
ketal was subsequently removed in a one-pot reaction to give
compound 3 as was previously described.[16] The remaining
bromide was replaced by an azide and the primary alcohol
group was protected with 4,4-dimethoxytrityl (DMT) chlo-
ride. The resultant compound 4 was reduced with lithium
aluminum hydride in THF to form the amino group (92%
yield after chromatography) desired for coupling with NBD
chloride.
cence and a large increase in donor emission after hydrolysis
of the ester at the sn-2 position.[16] However, changes in the
acceptor emission were fairly feeble.
A problem when working with phospholipids in biological
experiments is their tendency to form insoluble aggregates.
Furthermore, we expected that a charged phospholipid might
stick to the outer side of the plasma membrane rather than
Scheme 1. A) Synthetic pathways to the protectedprobe PENN/SATE ( 1) andthe unprotectedprobe PENN ( 7). a) 1) 1,12-Dibromododecane
(excess), NaH, DMF, room temperature, 16 h, 2) TFA, MeCN, H2O, room temperature, 2 h; b) NaN3 (excess), DMSO, 858C, 3 h; c) DMT-Cl,
Et3N, DMAP, DMF, room temperature, 16 h; d) LiAlH, THF, 08C, 2 h; e) NBD-Cl, DIPEA, methanol, 08C!RT, 4 h; f) 4-(Nile red)-oxy-butyric acid,
TPSNT, MeIm, CH2Cl2, room temperature, 14 h; g) dowex 50 WX-8, CH2Cl2, room temperature, 3 h; h) 1) phosphoramidite 8, 4,5-dicyanoimida-
zole, DMF 08C!RT, 2) tBuOOH (excess), 08C, 3 h, 38–52% (for g–h); k) 1) phosphoramidite 9, 4,5-dicyanoimidazole, DMF 08C!RT,
2) tBuOOH (excess), 08C, 30 min, 3) 5% TFA, CH2Cl2, room temperature, 1 h, l) 1) 1,12-dibromododecane (excess), NaH, DMF, room
temperature, 16 h, 2) NaN3 (excess), DMSO, 858C, 3 h; 3) Ph3P on resin, THF, 24 h, 4) MeOH/H2O (9:1), 24 h, 5) TFA, MeCN, H2O, RT, 3 h,
6) DMT-Cl, Et3N, DMAP, DMF, room temperature, 14 h. B) Phosphitylation reagents with enzymatically (8) andchemically ( 9) cleavable protecting
groups. DMF=N,N-dimethylformamide, TFA=trifluoroacetic acid, DMSO=dimethyl sulfoxide, DMAP=4-diaminoamino pyridine, DIPEA=diiso-
propropylethylamine, Boc=tert-butyloxycarbonyl.
Angew. Chem. Int. Ed. 2006, 45, 508 –512
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
509

Communications
In an alternative approach the reduction was performed
on a triphenylphosphine-loaded resin. Through the Stau-
dinger reaction, the azide was initially bound to the resin and
all side products were removed by thorough washings. The
pure amine was eluted with wet methanol. This procedure was
advantageous for small amounts, whereas the batch proce-
dure was superior for larger quantities. After successful
coupling of NBD, the derivative 5 was esterified with Nile red
oxybutyric acid in the presence of 1-(2,4,6-triisopropylbenze-
nesulfonyl)-3-nitro-1H-1,2,4-triazole (TPSNT) and methyli-
midazole (MeIm) in dichloromethane. This method, with
yields up to 87%, was superior to coupling reactions with
carbodiimide reagents. The resultant bilabeled compound 6
was then deprotected through the addition of dowex 50 WX-8
to a solution of 6 in dichloromethane. Migration of the sn-2
ester to the nowdeprotected primary alcohol needed to be
avoided; we therefore filtered the dichloromethane solution
quickly through silica to remove DMT-containing by-prod-
ucts, evaporated the solvent, and immediately started the
subsequent phosphitylation reaction in DMF without further
purification (a sample of the primary alcohol was purified for
analysis). Phosphoramidite 8 and 4,5-dicyanoimidazole were
added under argon, and after 16 h the solution was treated
with tBuOOH for 3 h. Finally, 1 was purified through silica. In
a phosphitylation/oxidation/protection reaction with a Boc-
protected amino group and a tert-butyl group on the
phosphoramidite 9,[16] a variety of probes without bioactivat-
able protecting groups (PENN, 7) were prepared after
deprotection (Scheme 1). The latter was used for the inves-
tigation of the optical properties in an in vitro micelle assay.
Micelles that consist of about 140 molecules were made
from Triton X-100 by sonication by following standard
procedures.[20] PENN 7 was incorporated in such a way that
every tenth micelle received one molecule of the dye. Under
these conditions, the fluorescent spectra showed total quench-
ing of green NBD fluorescence and a significant amount of
emission with a maximum at 620 nm (typical for Nile red
emission), which suggests that the energy transfer was
complete (Figure 3 A). Quantum yields for 7 in ethanol
were 0.12, whereas NBD-labeled undecanoic acid alone had a
quantum yield of 0.36.[21] The value for 7 was identical to that
of a previously used FRET probe with the same fluorophores
(see Supporting Information). Photostability for 7 was
determined in micelles and HeLa cells. Although no photo-
bleaching was observed in vitro, NBD bleached in cells with
about 1.5% per exposure (see Supporting Information).
The addition of commercially available PLA2 from bee
venom resulted in an immediate increase in green fluores-
cence that was accompanied by a substantial decrease in red
fluorescence. Ratio changes reached a maximum of 30-fold
after 15–30 minutes (Figure 3 A). Reorganization of the
micelles by additional sonication showed no further change
in the ratio, indicating that the two fluorophores were already
spatially separated. Accordingly, the remaining red fluores-
cence resulted from direct excitation of Nile red. TLC analysis
of 7 treated with PLA2 for 30 min showed cleavage of more
than 90% (Figure 3B). These data suggest that the strong
ratio change makes the probe well suitable for applications
within living cells.
Figure 3. A) Addition of bee venom PLA2 to a 1 mm solution of 7 in
Triton X100 micelles (3 mm) at pH 7.5 ledto an increase in NBD
fluorescence anda decrease in Nile redfluorescence. Excitation was at
440 nm. The emission ratio change was 30-fold. Sonication after the
reaction was complete gave no further ratio change. B) TLC analysis of
a parallel reaction andreferences after 30 min.
Phosphatidylethanolamines are distributed throughout
the internal cellular membranes and on the inner side of the
plasma membrane. In contrast, cytosolic PLA2 activity is
known to be most pronounced on membranes of the
perinuclear region.[11,22,23] Therefore, it was crucial to demon-
strate that the probe was able to reach this cell compartment.
Incubation of HeLa cells or MDCK II (Maden Darby canine
kidney) epithelial cells showed that the probe was predom-
inantly located in the Golgi apparatus, the nuclear envelope,
and at higher doses, also in the endoplasmic reticulum (ER)
(Figure 4). Owing to an unknown mechanism, the plasma
Figure 4. Hydrolysis of 1 by endogenous enzymes in MDCK II cells is
inhibited by MAFP in a dose-dependent manner. Upper panel: NBD
fluorescence, middle panel: Nile red fluorescence, lower panel: overlay.
membrane stayed free of dye. The locations of dye penetra-
tion can most likely be attributed to the Nile red moiety.
Experiments with simple Nile red labeled fatty acids showed
accumulation in the same regions, whereas unmodified Nile
red stained small vesicles or oil droplets (data not shown).
Other aromatic dyes such as NBD and lissamine also seem to
guide attached compounds into perinuclear membranes.
MDCK II cells loaded with 1 (3 mm) were analyzed with a
widefield microscope (Zeiss Axiovert 200M). The endoge-
510
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 508 –512

Angewandte
Chemie
nous PLA2 activity of the cells was stimulated with fetal calf
serum (FCS) and cleaved from the probe within 30 to 60 min
(Figure 4, control). Fluorescence excitation was at 435–
485 nm for NBD fluorescence and 540–580 nm for Nile red.
Emission was measured at 515–555 nm (NBD) and 600–
660 nm (Nile red). The ratio of the two emission values
(donor/acceptor fluorescence) gave a direct readout of the
status of the probe: lowratio values reflected an intact probe
and high ratio values a cleaved probe. In Figure 4, Nile red
fluorescence intensity values are depicted in red and NBD
fluorescence is converted to green (false color). An overlay of
the two images combined the entire information in one
picture and showed the colocalization of the dyes.
Inhibition of PLA2 by standard inhibitors such as methyl
arachidonyl fluorophosphate (MAFP) and bromoenollactone
(BEL) was studied next. HeLa cells were plated in square 4-
welled coverslips (20 ” 9 mm per chamber), treated with 1
(3 mm) and either an inhibitor or a control substance for
30 min. Ratio pictures (not shown) gave a dose-dependent
inhibition of PLA2 activity with an IC50 of approximately 5 mm
for the covalent binding inhibitor MAFP and only minimal
inhibition by BEL (Figure 5).[24,25]
removed by enzymatic cleavage of the probe. Several grades
of cell confluency varying from 30!100% were tested and, as
expected, higher cell confluency gave a larger fluorescent
readout. Wells with a confluency below 80% were not
suitable for detection of enzyme activity. Wells treated with
MAFP during the probe incubation showed readouts similar
to noise levels, demonstrating that the probe was not cleaved
and, therefore, no dequenching of the FRET donor was
detectable.
Apart from analysis and screening of PLA2 inhibitors, we
were interested in the demonstration of the onset of PLA2
activity in living embryos during development. Similar
experiments have been previously performed in the gastro-
intestinal tract of Zebrafish embryos.[27] Such experiments
would tell us whether the probe 1) could be used in whole
organisms, 2) would be suitable for day-long experiments, and
3) would be toxic to a living organism. We therefore micro-
injected 1, which was dissolved in phosphate or balanced salt
solution (BSS) buffer (pH 7.4, up to 10% DMSO/Pluronic),
underneath a 3-h-old Medaka fish embryo without rupturing
the egg-yolk membrane. The embryo was then mounted on a
binocular stage that had fluorescence detection. Within 2–7 h
the probe had entirely accumulated within the embryo
(Figure 6 A), therefore, showing the effectiveness of the
Figure 6. Application of 1 to study PLA2 expression in Medaka fish
embryos. The upper panel shows an overlay of redandgreen
fluorescence, the lower panel shows only the redchannel for compar-
ison. A) The embryo (age 3 h) rapidly picked up the probe within 2–
7 h. B) At the age of 19 h it showeda partially developedgerm ring.
C) After 27 h the embryo continuedto exhibit substantial FRET, as is
depictedby the yellow color. D) andE) After 37 and48 h, respectively,
the embryo showedexclusively green staining by the NBD-labeled
lysophosphatidyl ethanolamine, whereas Nile red butyric acid diffused
into the fat-rich egg yolk, indicating the onset of PLA2 activity. The
onset of PLA2 enzyme activity shown here is in goodagreement with
results publishedfor time courses from Zebrafish embryo extracts. [15]
The embryo was observed for 5 days more until the fish hatched. This
showed that the probe was not detectably toxic to Medaka fish
embryos. Fluorescence persistedwithin the specimen until the fish
hatched.
Figure 5. Dose-response curves that result from experiments with
HeLa cells, which show diverse sensitivity to different PLA2 inhibitors
ynorm =normalizedemission ratio (515–555 nm/620 + nm), MAFP
c, BEL a. This suggests that the probe is sensitive to particular
PLA2 isoforms.
The latter inhibitor is known to be fairly specific towards
calcium-independent isoforms of PLA2 (iPLA2).[25] MAFP
sensitivity indicates a serine residue in the catalytic center of
the enzyme which hints towards the presence of PLA2
isoforms such as groups IV and VI to VIII in HeLa cells.[8]
Comparison of the MAFP IC50 with published data showed
similar results for the test tube and in living cells.[26] This
demonstrates that 1 is a suitable probe to screen for PLA2
inhibitors in living cells. This method was, therefore, shown to
be applicable for use in plate readers, thereby allowing fairly
high-throughput experiments. Initial experiments with a PE
EnVision platereader gave promising results. HeLa cells were
seeded in a 96-well plate and treated with 1 (5–50 mm) for
30 min before the cells were washed three times with HEPES
buffer (pH 7.4). For simplicity, we observed only FRET donor
dequenching, which occurred when the FRET acceptor was
bioactivatable-protecting-group approach. After 27 h the
embryo had grown significantly and exhibited yellow fluo-
rescence (Figure 6C), which is typical for the overlay of green
and red fluorescence in the intact FRET probe. After 37 h,
however, the fish embryo had turned green, whereas the egg
yolk was bright red (Figure 6D and E), suggesting that the
probe had been cleaved. The green-fluorescent-labeled lipid
backbone with the nonhydrolyzable ether bond in the sn-1
position stayed in the fish whereas the hydrolyzed Nile red
Angew. Chem. Int. Ed. 2006, 45, 508 –512
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
511

Communications
butyric acid had apparently diffused into the fat-rich egg yolk,
thereby separating the two parts of the probe.
Keywords: bioorganic chemistry · fluorescent probes ·
FRET (fluorescence resonance energy transfer) · phospholipids ·
signal transduction
.
It remains to be established which isoform of PLA2 is
responsible for the cleavage reaction, both in cultured cells
and in fish embryos. In vitro experiments with isolated
recombinant cPLA2 a showed no cleavage of 7. Overexpres-
sion of cPLA2 a and g also failed to increase the cleavage of
the probe. All experiments performed so far hint towards
PAF-AHII as the responsible enzyme for cleavage of 7 in
HeLa cells. This result is in accordance with recent data from
C. elegans, which show that PAF-AHII is essential for
epithelial morphogenesis.[28] Only future experiments with
purified enzyme can prove hydrolysis of 7 with PAF-AHII.
In conclusion, the PLA2 probe 1 is highly membrane
permeant and accumulates close to the intracellular region of
PLA2 activity. The probe was demonstrated to be useful in
monitoring enzyme activities in cultured cells and small
organisms and there is much potential for the probe to be
used in living-cell high-throughput screening experiments for
PLA2 inhibitors. Small-molecule FRET probes generated by
preparative organic chemistry usually exhibit much larger
ratio changes than genetically encoded FRET probes. There-
fore, this technology should generally be explored more
extensively in the future, especially when the probes are
membrane permeant and therefore suitable for the treatment
of large cell populations.
[1] J. Zhang, R. E. Campbell, A. Y. Ting, R. Y. Tsien, Nat. Rev. Mol.
Cell Biol. 2002, 3, 906 – 918.
[2] F. S. Wouters, P. J. Verveer, P. I. H. Bastiaens, Trends Cell Biol.
2001, 11, 203 – 211.
[3] A. Miyawaki, A. Sawano, T. Kogure, Nat. Cell Biol. 2003, 5, S1 –
S7.
[4] N. Johnsson, K. Johnsson, ChemBioChem 2003, 4, 803 – 810.
[5] G. Zlokarnik, P. A. Negulescu, T. E. Knapp, L. Mere, N. Burres,
L. X. Feng, M. Whitney, K. Rꢀmer, R. Y. Tsien, Science 1998,
279, 84 – 88.
[6] H. S. Hendrickson, Anal. Biochem. 1994, 219, 1 – 8.
[7] H. S. Hendrickson, E. K. Hendrickson, I. D. Johnson, S. A.
Farber, Anal. Biochem. 1999, 276, 27 – 35.
[8] I. Kudo, M. Murakami, Prostaglandins Other Lipid Mediators
2002, 68–69, 383 – 399.
[9] C. D. Funk, Science 2001, 294, 1871 – 1875.
[10] S. Das, W. Cho, J. Biol. Chem. 2002, 277, 23838 – 23846.
[11] S. Grewal, S. Ponnambalam, J. H. Walker, J. Cell Sci. 2003, 116,
2303 – 2310.
[12] J. H. Evans, D. M. Spencer, A. Zweifach, C. C. Leslie, J. Biol.
Chem. 2001, 276, 30150 – 30160.
[13] S. Benzaria, H. Pelicano, R. Johnson, G. Maury, J. L. Imbach,
A. M. Aubertin, G. Obert, G. Gosselin, J. Med. Chem. 1996, 39,
4958 – 4965.
[14] M. S. J. Briggs, I. Bruce, J. N. Miller, C. J. Moody, A. C.
Simmonds, E. Swann, Chem. Soc. Perkin Trans. 1 1997, 1051 –
1058.
[15] S. A. Farber, E. S. Olson, J. D. Clark, M. E. Halpern, J. Biol.
Chem. 1999, 274, 19338.
[16] O. Wichmann, C. Schultz, Chem. Commun. 2001, 2500 – 2501.
[17] M. Shafiee, S. Deferme, A. L. Villard, D. Egron, G. Gosselin,
J. L. Imbach, T. Lioux, A. Pompon, S. Varray, A. M. Aubertin, G.
Van den Mooter, R. Kinget, C. Perigaud, P. Augustijns, J. Pharm.
Sci. 2001, 90, 448 – 463.
[18] C. Dinkel, O. Wichmann, C. Schultz, Tetrahedron Lett. 2003, 44,
1153 – 1155.
[19] C. Schultz, Bioorg. Med. Chem. 2003, 11, 885 – 898.
[20] Y. A. Hannun, C. R. Loomis, R. M. Bell, J. Biol. Chem. 1985,
260, 10039 – 10043.
Experimental Section
1
3
1: H NMR (400 MHz, CDCl3, 258C, TMS): d = 8.48 (d, J = 8.6 Hz,
1H; NBD-H), 8.25 (d, 3J = 8.8 Hz, 1H; Nile red H4), 8.06 (d, 4J =
2.5 Hz, 1H; Nile red H1), 7.62 (d, 3J = 9.1 Hz, 1H; Nile red H11), 7.19
3
4
3
(dd, J = 8.6 Hz, J = 2.5 Hz, 1H; Nile red H3), 6.70 (dd, J = 9.1 Hz,
4J = 2.5 Hz, 1H; Nile red H10), 6.48 (d, 4J = 2.5 Hz, 1H; Nile red H8),
6.35 (s, 1H; Nile red H6), 6.16 (d, J = 8.3 Hz, 1H; NBD-H), 5.53 (t,
3
3J = 4.5 Hz, 0.5H; NH carbamate), 5.48 (t, 3J = 4.5 Hz, 0.5H; NH
carbamate), 5.30–5.20 (m, 1H; sn-2 CH), 4.37–4.10 (m, 12H; Nile red
OCH2, POCH2 (3 ” ), NHCOOCH2), 3.61 (d, 3J = 5.1 Hz, 1H; sn-1
CH2), 3.56–3.40 (m, 10H; sn-1 OCH2, NBD-NH-CH2, Nile red NCH2
(2 ” ), OOCNHCH2), 3.20 (t, 2.67 , 3J = 6.3 Hz, 2H; COSCH2), 3.15 (t,
3J = 6.4 Hz, 2H; COSCH2), 2.69 (t, 3J = 7.2 Hz, 2H OOCCH2), 2.38 (s,
3H; CH3COS), 2.37 (s, 3H; CH3COS), 2.26 (qi, 3J = 6.7 Hz, 2H; Nile
red O-CH2CH2), 1.79 (qi, 3J = 7.2 Hz, 2H; NBDNH-CH2CH2), 1.60–
1.19 ppm (m, 26H; Nile red NCH2CH3 (2 ” ), CH2 chain (10 ” ));
13C NMR (100 MHz, CDCl3, 258C): d = 183.2, 161.5, 152.1, 150.9,
146.9, 136.5, 134.1, 131.1, 130.9, 128.8, 127.8, 125.7, 124.7, 118.2, 109.7,
106.6, 105.2, 96.3, 71.8, 71.0, 68.3, 67.0, 66.3, 66.2, 63.4, 45.1, 44.1, 41.4,
38.7, 31.9, 30.8, 30.6, 30.4, 29.7, 29.5, 29.4, 29.4, 29.3, 29.2, 29.1, 28.9,
28.5, 28.3, 26.9, 26.0, 12.6 ppm; 31P NMR (162 MHz, CDCl3, 258C,
H3PO4): d = ꢀ1.1 ppm; HR FAB MS (NBA, positive mode): [M+H+]
calculated: 1212.4398; [M+H+] found: 1212.4366.
[21] H. Du, R. C. A. Fuh, J. Z. Li, L. A. Corkan, J. S. Lindsey,
Photochem. Photobiol. 1998, 68, 141 – 142.
[22] A. Matsuzawa, K. Hattori, J. Aoki, H. Arai, K. Inoue, J. Biol.
Chem. 1997, 272, 32315 – 32320.
[23] K. Asai, T. Hirabayashi, T. Houjou, N. Uozumi, R. Taguchi, T.
Shimizu, J. Biol. Chem. 2003, 278, 8809 – 8814.
[24] J. Balsinde, E. A. Dennis, J. Biol. Chem. 1996, 271, 6758 – 6765.
[25] J. Balsinde, I. D. Bianco, E. J. Ackermann, K. Condefrieboes,
E. A. Dennis, Proc. Natl. Acad. Sci. USA 1995, 92, 8527 – 8531.
[26] M. A. Balboa, I. Varela-Nieto, K. K. Lucas, E. A. Dennis, FEBS
Lett. 2002, 531, 12 – 17.
[27] S. A. Farber, M. Pack, S. Y. Ho, L. D. Johnson, D. S. Wagner, R.
Dosch, M. C. Mullins, H. S. Hendrickson, E. K. Hendrickson,
M. E. Halpern, Science 2001, 292, 1385 – 1388.
[28] T. Inoue, A. Sugimoto, Y. Suzuki, M. Yamamoto, M. Tsujimoto,
K. Inoue, J. Aoki, H. Arai, Proc. Natl. Acad. Sci. USA 2004, 101,
13233 – 13238.
Received: February 28, 2005
Revised: July 9, 2005
Published online: December 2, 2005
[29] C. Dinkel, C. Schultz, unpublished results.
512
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 508 –512