A versatile library of activity-based probes for fluorescence detection and/or affinity isolation of lipolytic enzymes

Article abstract of DOI:10.1016/j.chemphyslip.2006.06.018

This work describes the synthesis of a library of fluorescent and/or biotinylated alkylphosphonate inhibitors being reactive towards serine hydrolases, especially lipases and esterases. Fluorescent inhibitors can be used for sensitive and rapid detection

Full text of DOI:10.1016/j.chemphyslip.2006.06.018

Chemistry and Physics of Lipids 144 (2006) 60–68
A versatile library of activity-based probes for fluorescence
detection and/or affinity isolation of lipolytic enzymes
Heidrun Susani-Etzerodt, Hannes Schmidinger, Gernot Riesenhuber,
Ruth Birner-Gruenberger, Albin Hermetter∗
Institute of Biochemistry, Graz University of Technology, Petersgasse 12/2, A-8010 Graz, Austria
Received 2 February 2006; received in revised form 21 June 2006; accepted 28 June 2006
Available online 21 July 2006
Abstract
This work describes the synthesis of a library of fluorescent and/or biotinylated alkylphosphonate inhibitors being reactive
towards serine hydrolases, especially lipases and esterases. Fluorescent inhibitors can be used for sensitive and rapid detection of
active proteins by gel electrophoresis. Biotinylated inhibitors are applicable for the enrichment and isolation of active enzymes.
Functionality as well as the different detection methods of the synthesized inhibitors were successfully tested with an enzyme prepa-
ration, namely cholesterol esterase from porcine pancreas (ppCE). Moreover, a biotinylated inhibitor was employed to enrich ppCE
on avidin beads. Hence, our set of phosphonate inhibitors can be used for the detection and/or isolation of active serine hydrolases.
© 2006 Elsevier Ireland Ltd. All rights reserved.
Keywords: Activity-based enzyme probes; Alkylphosphonic acid ester; Biotin-labeled inhibitors; Cholesterol esterase; Fluorescent lipids; Serine
hydrolases
1. Introduction
Hamilton et al., 1998; Mannesse et al., 1995; Taylor et
al., 2001) and have been established as powerful probes
for enzyme analytics.
Specific and even less-specific inhibitors are useful
tools in enzyme biochemistry and biophysics provided
the mechanisms underlying enzyme function are known
(Campbell and Szardenings, 2003). In this context, p-
nitrophenyl or fluoroalkylphosphonic acid esters turned
out to be useful suicide inhibitors of serine hydrolases,
a large and diverse family which comprises numerous
lipases, esterases and proteases. These inhibitors have
been applied to investigate enzyme–lipid interactions
with respect to substrate- and stereoselectivity (Berman
et al., 1985; Brzozowski et al., 1991; Egloff et al., 1995;
The reaction of alkylphosphonic acid esters with
serine hydrolases depends on the activities of these
enzymes. The mechanism comprises the nucleophilic
attack of the active serine (which is part of the cat-
alytic triad Ser-His-Asp/Glu) on the phosphorus of
the inhibitor and the simultaneous release of the p-
nitrophenyl residue which is a very efficient leaving
group. The serine remains covalently and irreversibly
bound to the inhibitor, mimicking the first (tetrahe-
dral) transition state of carboxylic acid ester hydrolysis
(Bjorkling et al., 1994; Cygler et al., 1994; Moreau et
al., 1991; Ransac et al., 1997).
One bioanalytical technique using these activity-
sensing probes is based on the measurement of p-
nitrophenol release from p-nitrophenyl esters due to
∗
Corresponding author. Tel.: +43 3168736457;
fax: +43 3168736952.
E-mail address: albin.hermetter@tugraz.at (A. Hermetter).
0009-3084/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.chemphyslip.2006.06.018

H. Susani-Etzerodt et al. / Chemistry and Physics of Lipids 144 (2006) 60–68
61
nucleophilic substitution by the active serine of the
enzyme (Rotticci et al., 2000; Scheib et al., 1998; Stadler
et al., 1996). However, large amounts of enzyme are
required to generate measurable quantities of the ana-
lyte.
To overcome this limitation, much higher sensitivity
can be gained by fluorescence labeling of the inhibitor.
Fluorescent phosphonate inhibitors form covalent and
stoichiometric (1/1, mol/mol) complexes even with low
abundant enzymes in complex samples, that can easily be
visualized after separation by one- or two-dimensional
gel electrophoresis. Therefore, these probes are impor-
tant tools for functional proteomic research.
Biotin tagging of inhibitors is the method of choice
for detection of several proteins (i.e. very large and small
proteins, highly basic and acidic proteins and membrane
proteins, which are unsuitable for 2-D electrophoresis)
(Adam et al., 2001). Biotinylated probes allow enrich-
ment and purification of such proteins using avidin chro-
matography procedures.
Here, we report on a set of fluorescently labeled p-
nitrophenyl phosphonate probes for the rapid and sen-
sitive in-gel visualization (Birner-Gruenberger et al.,
2005; Oskolkova et al., 2003; Schmidinger et al., 2005;
Scholze et al., 1999) as well as a set of biotinylated
inhibitors applicable for affinity isolation (Adam et al.,
2001; Deussen et al., 2000; Kidd et al., 2003; Liu et
al., 1999) of enzyme–inhibitor complexes. In addition
to their tags (fluorophore or biotin) the phosphonates
also differ in their alcohol moieties (see insert to Fig. 2)
which mimic the structures of natural lipase and esterase
substrates and are therefore responsible for inhibitor
specificity towards various types of enzymes. This work
gives an overview of the respective lipase and esterase
inhibitors including known compounds and the novel
probes (3b and c, 4a–f, 5d–f) which are described in
detail.
spraying with phosphomolybdic acid. Flash chromatog-
raphy was performed on silica gel 60 (230–400 mesh,
MERCK). NMR spectra were recorded using a VAR-
IAN INOVA-500 spectrometer (Varian, Palo Alto, CA,
USA). Chemical shifts (δ) are given in parts per million
(ppm) using tetramethylsilane as standard. Multiplici-
ties are abbreviated as follows: s, singlet; d, doublet;
t, triplet; q, quadruplet; m, multiplet; bs, broad sig-
nal. Mass spectra were obtained using a MALDI micro
MXTM (Waters, Milford, MA, USA) with a nitrogen
UV laser (337 nm wavelength) and a time-lag focus-
ing unit. Reflectron mode was positive, source voltage
15 kV and pulse voltage were 1950 V. Matrix suppres-
sion was 600 Da and sampling rate was 0.5 ns. Calibra-
tion was performed with a suitable polyethylene gly-
col mixture, as well as with Bradykinin, Angiotensin
I, Substance-P or Glu-fibrino peptide as internal stan-
dards. Samples were prepared by mixing solutions of
matrix (10 mg/mL ␣-cyano-4-hydroxy cinnamic acid
in EtOH/acetonitrile/aqueous 0.1% TFA 495/495/10,
v/v/v) and analyte (0.01–1 mg/mL in CHCl3/MeOH 2/1,
v/v) at the ratio 10:1 (v/v). The mixture (0.5 ␮L) was
spotted on a target plate (stainless steel) and allowed to
air-drying prior to analysis.
2.2. Synthesis of NBD-HE-HP (1)
NBD-HE-HP (1) was synthesized as published
(Oskolkova et al., 2003; Schmidinger et al., 2005)
(Fig. 1).
2.3. Synthesis of inhibitor groups 3, 4 and 5
Fig. 2 depicts the general procedure for the prepara-
tion of inhibitor groups 3, 4 and 5 following published
procedures (Deussen et al., 2000; Reetz et al., 2002;
Schmidinger et al., 2005). To a solution of 62 mg
(0.1 mmol) of compound 2a (Reetz et al., 2002) in
1 mL absolute dichloromethane 40 ␮L (0.3 mmol)
bromotrimethylsilane were added. Stirred at room
temperature for 40 h. After removal of all volatile
components under reduced pressure the orange residue
2. Materials and methods
2.1. Inhibitor syntheses
Chemicals for lipid synthesis were purchased from
commercial suppliers and used without further purifica-
tion unless otherwise stated. Triethylamine was freshly
distilled over calcium hydride. TLC was carried out
on silica gel 60 F254 aluminium sheets (0.2 mm,
MERCK, Whitehouse Station, NJ, USA). The com-
pounds were visualized under UV light (360 nm), with p-
dimethylaminocinnamaldehyde (DMACA) or by char-
ring at 120 ◦C after spraying with 50% sulfuric acid.
Phosphorus containing compounds were detected after
Fig. 1. Structure of NBD-HE-HP (1).

62
H. Susani-Etzerodt et al. / Chemistry and Physics of Lipids 144 (2006) 60–68
Fig. 2. Synthetic routes for the preparation of inhibitor groups 3–5. For substituents R (a–f) see insert.
was dissolved in a mixture of 1 mL dichloromethane
and 26 ␮L (0.3 mmol) oxalyl dichloride followed by
addition of 1.5 ␮L (0.02 mmol) DMF. This mixture was
heated under reflux for 18 h. The volatile components
were removed, and the residue was dissolved in 1 mL
CH2Cl2. To this solution 0.1 mmol of the alcohol com-
ponents b–f (see insert to Fig. 2) and 18 ␮L (0.3 mmol)
of triethylamine were added and the mixture was stirred
at room temperature for 8 h. After evaporation of the
solvent, the intermediate compounds 2b–f were purified
by flash chromatography on 25 mL swollen silica
gel.

H. Susani-Etzerodt et al. / Chemistry and Physics of Lipids 144 (2006) 60–68
63
2b: yellowish oil; yield 69%; Rf = 0.3 in
CH2Cl2/EtOAc 8/1 (v/v); 2c: yellowish oil, yield
C42H72N5O9PS + Na+ requires 877.079; 1H NMR
(500 MHz, CD3OD): δ = 8.31 (d, J = 9.2 Hz, 2H, Ar),
7.45 (d, J = 8.4 Hz, 2H, Ar), 4.62 (bs, 1H, NH), 4.49 (q,
J = 4.9 and 3.0 Hz, 1H, CH CH2 biotin), 4.30 (q, J = 4.5
and 3.4 Hz, 1H, CH biotin), 4.22 − 4.10 (m, 2H), 4.05
(bs, 1H, NH), 4.00 (t, J = 6.6 Hz, 2H, decyl-CH2 O),
3.23 − 3.13 (m, 3H), 3.08 (t, J = 7.0 Hz, 2H), 2.93 (q,
J = 5.0 and 12.7 Hz, 1H), 2.71 (d, J = 12.7 Hz, 1H), 2.20
(t, J = 7.3 Hz, 2H), 2.06 (m, 2H), 1.78 − 1.22 (m, 46H),
0.89 (t, J = 6.8 Hz, 3H, CH3); 13C NMR (125 MHz,
CD3OD): δ = 174.8, 164.9, 158.2, 155.6, 145.1, 125.7,
121.4, 67.3, 64.6, 62.2, 60.5, 55.9, 40.4, 39.9, 39.1,
35.7, 31.8, 30.3, 30.2, 30.1, 29.7, 29.5, 29.4, 29.3, 29.2,
29.1, 29.0, 28.6, 28.3, 26.5, 26.3, 25.8, 25.6, 25.4, 24.5,
22.5, 22.2, 22.1, 13.3.
4c: yellowish oil; yield 49%; Rf = 0.5 in
CHCl3/MeOH 9/1 (v/v); MS: m/z = 933.180,
C46H80N5O9PS + Na+ requires 933.185; 1H NMR
(500 MHz, CDCl3): δ = 8.24 (d, J = 9.2 Hz, 2H, Ar),
7.39 (d, J = 7.3 Hz, 2H, Ar), 5.91 (bs, 1H, NH), 5.76 (bs,
1H, NH), 5.04 (bs, 1H, NH), 4.80 (bs, 1H, NH), 4.52
(t, J = 5.5 Hz, 1H, CH CH2 biotin), 4.33 (t, J = 5.9 Hz,
1H, CH biotin), 4.20 − 4.12 (m, 1H), 4.11 − 4.00 (m,
3H), 3.23 (q, J = 6.6 and 6.2 Hz, 2H), 3.19 − 3.11
(s, 3H), 2.93 (q, J = 5.0 and 7.7 Hz, 1H), 2.74 (d,
J = 12.9 Hz, 1 H), 2.25 − 2.17 (m, 2H), 1.98 − 1.89
(m, 2H), 1.80 − 1.21 (m, 52H), 0.89 (t, J = 6.8 Hz, 3H,
CH3); 13C NMR (125 MHz, CDCl3): δ = 173.2, 163.5,
156.2, 125.9, 121.2, 120.5, 67.2, 65.2, 62.1, 60.4, 40.9,
39.4, 32.2, 30.6, 30.1, 29.9, 29.8, 29.7, 29.6, 29.5, 29.3,
28.3, 26.8, 26.1, 25.7, 22.9, 14.4.
68%, Rf = 0.4 in CH2Cl2/EtOAc 8/1 (v/v); 2dsn1
:
yellowish oil, yield 50%, Rf = 0.3 in CH2Cl2/EtOAc
94/6 (v/v); 2dsn3: equivalent to 2dsn1; 2e: yellowish oil,
yield 52%, Rf = 0.4 in cyclohexane/CHCl3 3/1 (v/v); 2f:
orange oil, yield 21%; Rf = 0.4 in CHCl3/EtOAc 88/12
(v/v).
The purified intermediates 2a–f (0.03 mmol) were
dissolved in 1.5 mL dry DMF, respectively, and
0.03 mmol g, h or i (Schmidinger et al., 2005;
Vanwetswinkel et al., 1995) and 13 ␮L (0.09 mmol) tri-
ethylamine were added. The mixture was stirred at room
temperature for 2 h. After removal of the solvent in vacuo
the products (inhibitor groups 3, 4 and 5) were purified
by flash chromatography on 20 mL swollen silica gel.
Yields and solvents for flash chromatography as well as
MS and NMR data are only indicated for compounds
which have not been characterized yet.
3b: orange oil; yield 63%; Rf = 0.5 in CHCl3/EtOAc
3/1 (v/v); MS: m/z = 813.873, C38H59N6O10P + Na+
requires 813.873; 1H NMR (500 MHz, CDCl3): δ = 8.43
(d, J = 8.6 Hz, 1H, Ar), 8.16 (d, J = 9.2 Hz, 2H, Ar),
7.32 (d, J = 9.2 Hz, 2H, Ar), 6.46 (bs, 1H, NH), 6.11 (d,
J = 8.7 Hz, 1H, Ar), 4.62 (bs, 1H, NH), 4.13 − 4.06 (m,
1H), 4.04 − 3.95 (m, 3H), 3.43 (q, J = 5.8 and 5.7 Hz,
2H), 3.13 (q, J = 6.6 and 6.6 Hz, 2H), 1.91 − 1.83 (m,
2H), 1.78 − 1.14 (m, 38H), 0.81 (t, J = 6.8 Hz, 3H, CH3);
13C NMR (125 MHz, CDCl3): δ = 157.2, 156.1, 144.7,
144.5, 144.2, 136.7, 125.9, 121.1, 98.7, 67.1, 65.3, 43.9,
40.7, 32.0, 30.6, 30.5, 30.2, 29.7, 29.6, 29.5, 29.4, 29.2,
28.5, 26.7, 26.4, 26.2, 26.0, 25.7, 25.6, 22.9, 22.4, 14.3.
3c: orange oil; yield 75%; Rf = 0.5 in CHCl3/EtOAc
3/1 (v/v); MS: m/z = 869.979, C42H67N6O10P + Na+
requires 869.979.
4a: yellowish oil; yield 41%; Rf = 0.6 in
CHCl3/MeOH 7/1 (v/v); MS: m/z = 792.920,
C36H60N5O9PS + Na+ requires 792.919; 1H NMR
(500 MHz, DMSO): δ = 8.27 (d, J = 9.1 Hz, 2H, Ar),
7.73 (bs, 1H, NH), 7.45 (d, J = 9.0 Hz, 2H, Ar), 7.01
(bs, 1H, NH), 6.41 (bs, 1H, NH), 6.35 (bs, 1H, NH),
4.29 (t, J = 5.1 Hz, 1H, CH biotin), 4.18 − 4.09 (m, 3H),
3.88 (t, J = 6.6 Hz, 2H), 3.08 (m, 1H), 2.98 (q, J = 6.7
and 6.1 Hz, 2H), 2.92 (q, J = 6.6 and 6.4 Hz, 2H), 2.80
(q, J = 6.7 and 6.1 Hz, 1 H), 2.56 (d, J = 12.5 Hz, 1H),
2.02 − 1.95 (m, 4H), 1.62 − 1.15 (m, 35H); 13C NMR
(125 MHz, DMSO): δ = 172.5, 163.4, 157.5, 156.2,
144.8, 126.6, 122.1, 64.2, 63.2, 61.7, 59.9, 56.2, 40.0,
39.0, 36.0, 30.3, 30.2, 30.1, 29.8, 29.6, 29.4, 29.1, 28.9,
28.7, 26.8, 26.7, 26.1, 24.9, 22.5, 16.9.
4drac
:
white solid; yield 36%; Rf = 0.5 in
CHCl3/MeOH 10/1 (v/v); MS: m/z = 1287.769,
C69H126N5O11PS + Na+ requires 1287.795; 1H NMR
(500 MHz, CDCl3): δ = 8.23 (d, J = 9.1 Hz, 2H, Ar),
7.40 (q, J = 4.6 and 4.2 Hz, 2H, Ar), 5.96 (bs, 1H, NH),
5.86 (bs, 1H, NH), 5.10 (bs, 1H, NH), 4.81 (bs, 1H,
NH), 4.52 (t, J = 7.2 Hz, 1H, CH CH2 biotin), 4.33
(t, J = 5.6 Hz, 1H, CH biotin), 4.22 − 4.15 (m, 1H),
4.10 − 4.00 (m, 3H), 3.61 − 3.38 (m, 8H), 3.22 (m, 2H),
3.16 (m, 3H), 2.92 (q, J = 4.9 and 7.9 Hz, 1H), 2.74
(d, J = 12.9 Hz, 1H), 2.23 − 2.18 (m, 2H), 2.01 − 1.92
(m, 2H), 1.77 − 1.18 (m, 87H), 0.88 (t, J = 6.8 Hz, 6H,
CH3); 13C NMR (125 MHz, CDCl3): δ = 173.3, 163.7,
156.0, 144.7, 125.9, 121.3, 72.1, 71.0, 69.6, 66.4, 65.2,
62.0, 60.3, 55.7, 40.8, 39.5, 36.2, 32.2, 30.8, 30.7, 30.2,
30.1, 30.0, 29.9, 29.8, 29.7, 29.6, 29.3, 28.3, 26.7, 26.5,
26.3, 26.1, 25.8, 25.6, 25.0, 22.9, 22.5, 22.4, 14.4.
4e: white solid; yield 37%; Rf = 0.5 in CHCl3/MeOH
10/1 (v/v); MS: m/z = 1133.504, C61H100N5O9PS + Na+
requires 1133.505; 1H NMR (500 MHz, CDCl3):
δ = 8.23 (m, 2H, Ar), 7.39 (m, 2H, Ar), 6.23 (bs, 1H,
4b: yellowish oil; yield 32%; Rf = 0.5 in
CHCl3/MeOH 8/1 (v/v); MS: m/z = 877.079,

64
H. Susani-Etzerodt et al. / Chemistry and Physics of Lipids 144 (2006) 60–68
NH), 6.13 (bs, 1H, NH), 5.37 (bs, 1H, C CH CH2 chol),
5.29 (m, 1H), 4.86 (bs, 1H, NH), 4.51 (t, J = 5.1 Hz,
1H, CH CH2 biotin), 4.39 − 4.30 (m, 2H), 4.02 (t,
J = 6.7 Hz, 2H, decyl-CH2 O), 3.22 (d, J = 6.5 Hz, 2H),
3.15 (q, J = 7.2 and 4.8 Hz, 3H), 2.91 (q, J = 4.9 and
8.0 Hz, 1H), 2.74 (d, J = 12.8 Hz, 1H), 2.46 − 2.35 (m,
1H), 2.21 (t, J = 7.5 Hz, 2H), 2.02 − 1.03 (m, 61H),
1.00 (s, 3H), 0.90 (d, J = 6.5 Hz, 3H), 0.86 (q, J = 2.3
and 4.3 Hz, 6H), 0.66 (s, 3H); 13C NMR (125 MHz,
CDCl3): δ = 173.4, 164.0, 157.2, 156.2, 144.6, 139.3,
125.9, 123.6, 121.3, 77.9, 65.1, 62.0, 60.4, 56.8, 56.4,
55.8, 50.2, 42.5, 40.9, 40.8, 40.5, 39.9, 39.7, 39.4, 37.1,
36.6, 36.4, 36.2, 36.0, 32.1, 32.0, 30.7, 30.6, 30.2, 30.1,
29.9, 29.7, 29.6, 29.5, 29.3, 29.2, 28.4, 28.3, 28.2, 27.4,
23.1, 22.8, 22.6, 22.5, 21.2, 19.5, 18.9, 12.1.
4f: orange oil; yield 37%; Rf = 0.4 in CHCl3/MeOH
10/1 (v/v); MS: m/z = 1111.311, C52H82N9O12PS + Na+
requires 1111.290; 1H NMR (500 MHz, CD3OD):
δ = 8.53 (d, J = 8.9 Hz, 1H, Ar), 8.29 (d, J = 9.2 Hz, 2H,
Ar), 7.44 (d, J = 9.1 Hz, 2H, Ar), 6.36 (d, J = 8.9 Hz,
1H, Ar), 4.62 (bs, 1H, NH), 4.49 (q, J = 4.9 and 2.9 Hz,
1H, CH CH2 biotin), 4.30 (q, J = 4.5 and 3.4 Hz, 1H,
CH biotin), 4.21 − 4.09 (m, 2H), 4.04 (bs, 1H, NH),
3.99 (t, J = 6.6 Hz, 2H decyl-CH2 O), 3.53 (bs, 2H),
3.22 − 3.14 (m, 3H), 3.07 (t, J = 7.0 Hz, 2H), 2.92 (q,
J = 5.0 and 7.8 Hz, 1H), 2.70 (d, J = 12.7 Hz, 1H), 2.19
(t, J = 7.4 Hz, 2H), 2.06 (m, 2H), 1.78 − 1.20 (m, 61H);
13C NMR (125 MHz, CD3OD): δ = 178.1, 161.3, 155.1,
147.7, 145.9, 142.7, 128.8, 125.6, 121.4, 119.8, 110.0,
78.3, 64.6, 62.2, 60.5, 55.9, 39.9, 29.5, 29.1, 28.9, 28.7,
28.4, 25.8, 25.4.
7.39 (q, J = 9.3 Hz, 2H, Ar), 5.35 (d, J = 4.9 Hz, 1H,
C = CH CH2 chol), 4.49 (bs, 1H, CH CH2 biotin,
NH), 4.36 (bs, 1H, CH O chol), 4.04 (t, J = 5.4 Hz, 2H,
decyl-CH2 O), 3.60 (bs, 1H, CH CH, biotin), 3.27 (bs,
4H), 2.34 (t, J = 2.7 Hz, 4H), 2.00 − 0.65 (m, 89H).
5f: orange oil; yield 21%; Rf = 0.3 in
CH2Cl2/MeOH 8/1 (v/v); MS: m/z = 1260.525,
C56H89N10O13PS3 + Na+ requires 1260.527.
2.4. Activity tagging and detection procedures
2.4.1. Fluorescent inhibitors
Triton X-100 (20 nmol, stock solution 1 mM in
CHCl3) and fluorescent inhibitor 1, 3a–e, 4f or 5f
(see Figs. 1 and 2) (0.4 nmol, stock solution 0.1 mM
in CHCl3/MeOH 2/1, v/v) were mixed in Eppendorf
tubes and the organic solvent was evaporated under a
stream of nitrogen. Then, 0.5 ␮g of ppCE (Cholesterol
esterase from porcine pancreas, Fluka, Buchs, Switzer-
land) dissolved in 20 ␮L of 50 mM Tris–HCl (pH 7.4)
were added to the solvent-free lipid–detergent mixture
and incubated under protection from light at 37 ◦C and
550 rpm for 2 h. The final concentrations of fluores-
cent inhibitor and Triton X-100 in the aqueous solu-
tion were 20 ␮M and 1 mM, respectively. Reactions
were stopped by adding 5 ␮L of reducing 1D sample
buffer (20 mM KH2PO4, 6 mM EDTA, 60 mg/mL SDS,
100 mg/mL glycerol, 0.5 mg/mL bromophenol blue and
20 ␮L/mL mercaptoethanol, pH 6.8) followed by heat-
ing the samples at 95 ◦C for 3 min. Inhibitor-labeled
proteins were separated by SDS-PAGE which was per-
formed in a Tris–glycine buffer system (5% stacking
gel, 10% resolving gel) at 20 mA constant current (Mini
PROTEAN 3 or PROTEAN II xi multi-cell, BIORAD,
Hercules, CA, USA). Gels were subsequently fixed in
7.5%aceticacidand10%ethanolandvisualizedbyscan-
ning at 530 nm with a Molecular ImagerTM FX Pro Plus
(BIORAD). Excitation wavelength was 488 nm, scan-
ning resolution was 100 ␮m.
5dsn1
:
white solid; yield 42%; Rf = 0.3 in
CH2Cl2/MeOH 8/1 (v/v); MS: m/z = 1437.032,
1
C73H133N6O12PS3 + Na+ requires 1437.032; H NMR
(500 MHz, DMSO): δ = 8.22 (d, J = 9.1 Hz, 2H, Ar),
7.93 (t, J = 5.4 Hz, 1H, NH), 7.71 (t, J = 5.4 Hz, 1H,
NH), 7.42 (m, 2H, Ar), 7.18 (bs, 1H, NH), 6.42 (s,
1H, NH), 6.35 (s, 1H, NH), 4.29 (t, J = 7.4 Hz, 1H, CH
biotin), 4.12 (t, J = 5.4 Hz, 1H), 3.88 (t, J = 6.1 Hz, 2H),
3.23 (q, J = 5.6 and 6.3 Hz, 4H), 3.08 (m, 1H), 2.98 (q,
J = 6.5 and 6.2 Hz, 3H), 2.80 (q, J = 5.1 and 7.4 Hz, 1H),
2.73 (t, J = 6.6 Hz, 8H), 2.63 (s, 1H), 2.58 (d, J = 3.2 Hz,
1H), 2.55 (s, 1H), 2.53 (bs, 2H), 2.51 (t, J = 1.8 Hz,
8H), 2.46 (s, 1H), 2.36 (s, 1H), 2.05 − 2.00 (m, 6H),
1.62 − 1.12 (m, 76H), 0.80 (t, J = 6.9 Hz, 6H).
2.4.2. Biotinylated inhibitors without a disulfide
bridge
For labeling 0.5 ␮g of ppCE, 0.08 nmol of biotiny-
lated inhibitor without a disulfide bridge (inhibitor group
4, see Fig. 2) were used (final concentration 4 ␮M).
For control experiments protein samples were heated
at 95 ◦C for 5 min prior to treatment with inhibitors
to distinguish between specific (heat-sensitive) and
non-specific (heat-insensitive) protein reactivity. All
other preparation steps were performed as described
above for the fluorescent inhibitors. After SDS-PAGE,
enzyme–inhibitor complexes were transferred onto
5dsn3
CH2Cl2/MeOH 8/1 (v/v); MS and NMR equiva-
lent to 5dsn1
:
white solid; yield 56%; Rf = 0.3 in
.
5e: white solid; yield 46%; Rf = 0.6 in
CH2Cl2/MeOH 89/11 (v/v); MS: m/z = 1282.766,
1
C65H107N6O10PS3 + Na+ requires 1282.741; H NMR
(500 MHz, CDCl3): δ = 8.24 (q, J = 8.9 Hz, 2H, Ar),

H. Susani-Etzerodt et al. / Chemistry and Physics of Lipids 144 (2006) 60–68
65
nitrocellulose membranes (Protran BA85, Schleicher
& Schuell, Dassel, Germany) by electroblotting (Mini
Trans-Blot® Electrophoretic Transfer Cell, BIORAD).
The membranes were blocked in 5% (w/v) non-fat
dry milk in TBS-T (150 mM NaCl, 100 mM Tris–HCl
pH 7.4, 1% Tween®20) at room temperature for 1 h
and subsequently treated with an avidin-Alexa Fluor®
555 conjugate (synthesized as recommended by the
manufacturer (Molecular Probes, Eugene, OR, USA))
(dilution 1:2000 in TBS-T) at room temperature
for 30 min followed by three washings in TBS-T
(10 min each). For detection of fluorescent bands,
membranes were laser-scanned at an excitation wave-
length of 532 nm with fluorescence detection at 555 nm
(Molecular ImagerTM FX Pro Plus, BIORAD).
beads (suspension in 50% glycerol, 0.15 M sodium phos-
phate, pH 6.8, containing 15 ppm Kathon® CG/ICP II,
Sigma, St. Louis, MO, USA) on a rotator at room
temperature for 2 h. After removal of the supernatant,
the beads were washed twice with TBS containing
0.2% SDS (w/v) and twice with TBS alone. Bound
protein was subsequently eluted with 20 ␮L of reduc-
ing 1D sample buffer (see above) at 95 ◦C for 10 min.
Proteins in the supernatant were precipitated in 10%
trichloroacetic acid on ice for 60 min and collected by
centrifugation at 4 ◦C at 10,000 × g for 15 min. The
pellet was washed once with ice-cold acetone, resus-
pended in 20 ␮L reducing 1D sample buffer (see above),
heated at 95 ◦C for 3 min and subjected to SDS-PAGE
and Western blotting together with the eluate. Detec-
tion of enzyme–inhibitor complexes was performed as
indicated for biotinylated inhibitors without a disulfide
bridge.
2.4.3. Biotinylated inhibitors with a disulfide bridge
Biotinylated inhibitors containing a disulfide bridge
(inhibitor group 5, see Fig. 2) (0.08 nmol, final con-
centration 4 ␮M) were used for labelling of 0.5 ␮g of
ppCE. Labeling and control reactions were performed
in the same manner as described before. Reactions were
stopped by adding 5 ␮L of non-reducing 1D sample
buffer (20 mM KH2PO4, 6 mM EDTA, 60 mg/mL SDS,
100 mg/mL glycerol, 0.5 mg/mL bromophenol blue, pH
6.8) followed by heating of samples at 95 ◦C for 3 min.
Enzyme–inhibitor complexes were resolved by non-
reducing SDS-PAGE since the biotin tags would other-
wisebecleavedoff. AfterSDS-PAGEandelectroblotting
with subsequent blocking (see above) membranes were
treated with an avidin–horseradish peroxidase (HRP)
conjugate from BIORAD (dilution 1:2000 in TBS-T)
at room temperature for 30 min. After three washing
steps using TBS-T (10 min each), blots were incubated
with SuperSignal® West Pico Chemiluminescent Sub-
strate (Pierce, Rockford, IL, USA) and then exposed
to an X-ray film (Kodak, Rochester, NY, USA) for
2–45 min.
3. Results and discussion
This work provides a comprehensive library of
alkylphosphonic acid esters with a fluorescent or a
biotin tag to be used as mechanism-based inhibitors that
covalently react with their target enzymes (i.e. serine
hydrolases including lipases and esterases). The respec-
tive lipids include known compounds and novel probes
(3b–c, 4a–f, 5d–f) which are described in detail.
Fluorescent inhibitors containing NBD as a fluo-
rophore (see Figs. 1 and 2) are very sensitive enzyme
probes. The respective enzyme–inhibitor complexes can
rapidly be detected by in-gel fluorescence scanning after
proteinseparationbyone-aswellastwo-dimensionalgel
electrophoresis as the small and uncharged NBD is com-
patible with isoelectric focusing (Birner-Gruenberger et
al., 2005; Schmidinger et al., 2005). Moreover, NBD
as a reporter fluorophore has the great advantage that
absorption and emission wavelengths are long enough
compared with the unspecific fluorescence background
of many biological samples.
2.5. Avidin-based affinity purification
Biotinylated phosphonate inhibitors (see Fig. 2)
can be utilized for avidin-based enrichment and isola-
tion of low abundant and membrane-bound enzymes
which are difficult to analyze by 2D-gel electrophore-
sis. We used avidin agarose for affinity isolation of
enzymes from complex proteomes (inhibitor group 4,
Fig. 2). Biotin–avidin interactions were cleaved by heat-
ing the enzyme–inhibitor complexes under denaturing
conditions at 95 ◦C (Adam et al., 2001; Kidd et al.,
2003). If the biotinylated inhibitors contain disulfide
bridges (inhibitor group 5, Fig. 2), these can be cleaved
under mild reductive conditions according to Deussen
For affinity isolation, 5 ␮g of ppCE were incubated
with 0.1 nmol of inhibitor 4a (final concentration 4 ␮M)
and 25 nmol Triton X-100 (final concentration 1 mM) in
25 ␮L 50 mM Tris–HCl (pH 7.4) at 37 ◦C and 550 rpm
for 2 h. After addition of 300 ␮L Tris-buffered saline
(TBS, 150 mM NaCl, 100 mM Tris–HCl pH 7.4) con-
taining 0.5% SDS (w/v), the solution was heated at
95 ◦C for 10 min to ensure accessibility of biotin moi-
eties (Kidd et al., 2003). The sample was further diluted
2.5-fold with TBS (to a final concentration of 0.2% SDS,
w/v) followed by incubation with 20 ␮L avidin–agarose

66
H. Susani-Etzerodt et al. / Chemistry and Physics of Lipids 144 (2006) 60–68
et al. (2000), releasing the biotin moiety from the
enzyme–inhibitor complexes.
3.2. Application of suicide inhibitors to functional
enzyme analysis
Trifunctional inhibitors containing a fluorophore as
well as a biotin tag without (4f) and with (5f) a disul-
fide bridge (see Fig. 2) were also synthesized. These
inhibitors may facilitate the detection of enzyme–
inhibitor complexes after affinity isolation with avidin–
agarose beads avoiding time-consuming analytical pro-
cedures (Adam et al., 2002).
In order to assess functionality of the synthesized flu-
orescent and biotinylated inhibitors they were incubated
with an enzyme preparation of known lipolytic activity
and selectivity, namely cholesterol esterase from porcine
pancreas (ppCE). The labeled protein preparations were
subjected to 1D-gel electrophoresis and the labeled pro-
tein bands were detected as described under Section
2. Fig. 3 shows that each of the synthesized inhibitors
reacted more or less efficiently with this enzyme. Nei-
ther unlabeled nor thermally denatured proteins showed
signals. Thus, labeling is specific for active enzymes.
All fluorescent phosphonates displayed in Panel A effi-
ciently inhibited ppCE irrespective of polarity of the
alcohol residue. Both groups of biotinylated inhibitors
(with and without disulfide bridge) reacted similarly
(Panels B and C). The polar phosphonic acid esters
4a–c, 4f, 5a and 5f show good reactivities towards ppCE
whereas inhibitors 4drac, 5dsn1, 5dsn3 and 5e, which
contain hydrophobic alcohol moieties, display a rather
weak signal. These bulky alcohol residues could steri-
cally hinder interactions between the biotin tags of the
enzyme–inhibitor complexes and the avidin conjugates
that are necessary for signal detection. Proteins labeled
with biotinylated inhibitors containing a disulfide bridge
(Panel C) had to be separated by non-reducing SDS-
PAGE (without mercaptoethanol) in order to detect the
biotin tags by avidin blotting, which would otherwise
be cleaved off. Because detection of biotinylated phos-
phonates is very time-consuming, the major application
of these inhibitors is protein purification using solid sup-
ports (Panel D). ppCE was affinity tagged with 4a, bound
onto avidin–agarose beads and carefully washed. Pro-
teins were eluted from the beads by heating with 1D
sample buffer. Compared to the supernatant (s), which
was obtained after centrifugation of the beads, labeled
protein was highly enriched in the eluate (el). Control
experiments with unlabeled or preheated protein did not
show any unspecific reactivity with the avidin–agarose
beads or the inhibitors. Cholesterol esterase labeled by
the trifunctional probes 4f and 5f can be detected either
by NBD-fluorescence or by avidin blotting (see Panels
A–C). These results show that the trifunctional probes
are reactive towards lipolytic enzymes. Since the labeled
enzyme binds to avidin, these inhibitors are applicable
for avidin-based affinity isolation. Finally, they allow
rapid and sensitive in-gel detection of enzyme–inhibitor
complexes by NBD-fluorescence. They are promising
tools for the isolation and subsequent detection of active
serine hydrolases though it has to be emphasized that
3.1. Synthesis of p-nitrophenylphosphonate
inhibitors
NBD-HE-HP (1) (Fig. 1) was prepared by coupling
an NBD-labeled alcohol to an unlabeled phosphonic
acid dichloride (Oskolkova et al., 2003). In addition, a
set of alkylphosphonic acid esters containing the label
(fluorophore or biotin) in the phosphorus-alkyl chain
was synthesized (Fig. 2). Starting from the central
bifunctional synthon 2a (Reetz et al., 2002) the ethoxy
group was replaced by TMS-Br. The resulting inter-
mediate was subsequently reacted with oxalyl chloride
to form an activated p-nitrophenylphosphonochloridate
which was converted with the alcohol components b–f
(see insert in Fig. 2). Intermediates 2a–f have a second
reactive group, namely ␻-succinimidyl carbonate which
was reacted with the primary amines g, h and i to give
stable carbamate bonds resulting in three main groups
of alkylphosphonates. The first group (inhibitor group
3) contains 7-nitrobenz-2-oxa-1,3-diazole (NBD) as a
fluorescent tag, the second group (inhibitor group 4)
contains a biotin tag but no disulfide bridge, and the third
group (inhibitor group 5) has an affinity tag (biotin) and a
disulfide bridge. Each of the three groups shown in Fig. 2
comprises phosphonic acid esters with alcohol compo-
nents of different polarity. These components mimic the
structures of natural lipase and esterase substrates and
show therefore considerable specificity towards various
types of lipolytic enzymes. Inhibitors 3dsn1, 3dsn3, 4drac
,
5dsn1 and 5dsn3 correspond to glycerolipids but contain
ether instead of carboxylester bonds in the sn-2 and -3(1)
positions of glycerol that are not cleavable by lipase-
catalyzed hydrolysis. Therefore, the p-nitrophenyl ester
bonds in the sn-1(3) positions are the sole reaction sites
availableforthenucleophilicserineoflipolyticenzymes.
All synthesized inhibitors are racemic at phosphorus.
It is known that the SP isomers preferably react with ser-
ine hydrolases. However, the presence of the RP isomers
does not interfere with this process (Birner-Grunberger
et al., 2004; Scholze et al., 1999; Zandonella et al., 1999).
Therefore, the cheaper racemates (at P) were used for
enzyme analytics.

H. Susani-Etzerodt et al. / Chemistry and Physics of Lipids 144 (2006) 60–68
67
Fig. 3. Functional analysis of suicide inhibitors. Cholesterol esterase from porcine pancreas (ppCE) was incubated with 20 ␮M fluorescent inhibitor
(A), or with 4 ␮M biotinylated inhibitor without disulfide bridge (B), or with 4 ␮M biotinylated inhibitor with disulfide bridge (C). ppCE was also
affinity isolated by avidin–agarose beads after reaction with 4a (D). Numbering of inhibitors corresponds to Figs. 1 and 2. Labeling and detection
was performed as described in Section 2. Unlabeled (Bl) and preheated (ꢀT+) samples were used for control reactions. el: eluate; s: supernatant.
their structures are more complex than the molecules of
the natural enzyme substrates.
organs and tissues that play essential roles in murine and
human lipid metabolism.
In conclusion, we have developed a library of fluo-
rescent and/or biotinylated alkylphosphonate inhibitors
containing alcohol moieties of different polarity for
detection and identification of serine hydrolases as well
as for discrimination of lipases and esterases. Func-
tional enzyme analysis performed with the synthesized
inhibitors has shown that all of them reacted with choles-
terol esterase from porcine pancreas (ppCE). Some of the
fluorescent inhibitors (1, 3a, 3dsn1, 3dsn3 and 3e) have
already been used for screening and discrimination of
lipolytic enzymes in enzyme preparations (Schmidinger
et al., 2005) as well as for mapping the lipolytic pro-
teome of mouse adipose tissue (Birner-Gruenberger et
al., 2005). Biotinylated inhibitors turned out to be useful
for microarray screening of lipase and esterase activ-
ity (Schmidinger et al., 2006), for molecular evolution
screening of lipases by phage display (Danielsen et
al., 2001) and for enzyme enrichment (see above and
Deussen et al., 2000). They will hopefully promote the
discovery of lipolytic enzymes which may have escaped
detection by fluorescent inhibitors using 2D-gel elec-
trophoresis. The whole set of activity-based probes is
currently used for mapping the lipolytic proteomes of
Acknowledgements
This work has been financed by the projects
Genomics of Lipid-associated Disorders (GOLD II)
(http://gold.uni-graz.at/) and Click it: In vivo functional
proteomic screening and cell imaging of lipases in the
framework of GEN-AU (GENome research in AUstria)
(http://www.gen-au.at/) which is funded by the Austrian
Federal Ministry for Education, Science and Culture
(bm:bwk).
References
Adam, G.C., Sorensen, E.J., Cravatt, B.F., 2002. Trifunctional chem-
ical probes for the consolidated detection and identification of
enzyme activities from complex proteomes. Mol. Cell Proteomics
1, 828–835.
Adam, G.C., Cravatt, B.F., Sorensen, E.J., 2001. Profiling the specific
reactivity of the proteome with non-directed activity-based probes.
Chem. Biol. 8, 81–95.
Berman, H.A., Olshefski, D.F., Gilbert, M., Decker, M.M., 1985. Fluo-
rescent phosphonate labels for serine hydrolases. Kinetic and spec-
troscopic properties of (7-nitrobenz-2-oxa-1,3-diazole)aminoalkyl

68
H. Susani-Etzerodt et al. / Chemistry and Physics of Lipids 144 (2006) 60–68
methylphosphonofluoridates and their conjugates with acetyl-
cholinesterase molecular forms. J. Biol. Chem. 260, 3462–3468.
Moreau, H., Moulin, A., Gargouri, Y., Noel, J.P., Verger, R., 1991.
Inactivation of gastric and pancreatic lipases by diethyl para-
nitrophenyl phosphate. Biochemistry 30, 1037–1041.
Oskolkova, O.V., Saf, R., Zenzmaier, E., Hermetter, A., 2003. Fluores-
cent organophosphonates as inhibitors of microbial lipases. Chem.
Phys. Lipids 125, 103–114.
Ransac, S., Gargouri, Y., Marguet, F., Buono, G., Beglinger, C., Hilde-
brand, P., Lengsfeld, H., Hadvary, P., Verger, R., 1997. Covalent
inactivation of lipases. Lipases Meth. Enzymol. 286, 190–231.
Reetz, M.T., Ruggeberg, C.J., Droge, M.J., Quax, W.J., 2002. Immo-
bilization of chiral enzyme inhibitors on solid supports by
amide-forming coupling and olefin metathesis. Tetrahedron 58,
8465–8473.
Rotticci, D., Norin, T., Hult, K., Martinelle, M., 2000. An active-
site titration method for lipases. Biochim. Biophys. Acta 1483,
132–140.
Scheib, H., Pleiss, J., Stadler, P., Kovac, A., Potthoff, A.P., Haalck,
L., Spener, F., Paltauf, F., Schmid, R.D., 1998. Rational design
of Rhizopus oryzae lipase with modified stereoselectivity toward
triradylglycerols. Protein Eng. 11, 675–682.
Birner-Gruenberger, R., Susani-Etzerodt, H., Waldhuber, M., Riesen-
huber, G., Schmidinger, H., Rechberger, G., Kollroser, M., Strauss,
J.G., Lass, A., Zimmermann, R., Haemmerle, G., Zechner, R., Her-
metter, A., 2005. The lipolytic proteome of mouse adipose tissue.
Mol. Cell Proteomics 4, 1710–1717.
Birner-Grunberger, R., Scholze, H., Faber, K., Hermetter, A., 2004.
Identification of various lipolytic enzymes in crude porcine pan-
creatic lipase preparations using covalent fluorescent inhibitors.
Biotechnol. Bioeng. 85, 147–154.
Bjorkling, F., Dahl, A., Patkar, S., Zundel, M., 1994. Inhibition of
lipases by phosphonates. Bioorg. Med. Chem. 2, 697–705.
Brzozowski, A.M., Derewenda, U., Derewenda, Z.S., Dodson, G.G.,
Lawson, D.M., Turkenburg, J.P., Bjorkling, F., Huge-Jensen, B.,
Patkar, S.A., Thim, L., 1991. A model for interfacial activation
in lipases from the structure of a fungal lipase–inhibitor complex.
Nature 351, 491–494.
Campbell, D.A., Szardenings, A.K., 2003. Functional profiling of the
proteome with affinity labels. Curr. Opin. Chem. Biol. 7, 296–303.
Cygler, M., Grochulski, P., Kazlauskas, R.J., Schrag, J.D., Bouthillier,
F., Rubin, B., Serreqi, A.N., Gupta, A.K., 1994. A structural basis
for the chiral preferences of lipases. J. Am. Chem. Soc. 116,
3180–3186.
Schmidinger, H., Birner-Gruenberger, R., Riesenhuber, G., Saf, R.,
Susani-Etzerodt, H., Hermetter, A., 2005. Novel fluorescent phos-
phonic acid esters for discrimination of lipases and esterases.
Chembiochem 6, 1776–1781.
Danielsen, S., Eklund, M., Deussen, H.J., Graslund, T., Nygren, P.A.,
Borchert, T.V., 2001. In vitro selection of enzymatically active
lipase variants from phage libraries using a mechanism-based
inhibitor. Gene 272, 267–274.
Schmidinger, H., Susani-Etzerodt, H., Birner-Gruenberger, R., Her-
metter, A., 2006. Inhibitor- and protein microarrays for activity-
based recognition of lipolytic enzymes. Chembiochem 7, 527–
534.
Deussen, H.J., Danielsen, S., Breinholt, J., Borchert, T.V., 2000. A
novel biotinylated suicide inhibitor for directed molecular evolu-
tion of lipolytic enzymes. Bioorg. Med. Chem. 8, 507–513.
Egloff, M.P., Marguet, F., Buono, G., Verger, R., Cambillau, C., van
Scholze, H., Stutz, H., Paltauf, F., Hermetter, A., 1999. Fluorescent
inhibitors for the qualitative and quantitative analysis of lipolytic
enzymes. Anal. Biochem. 276, 72–80.
Stadler, P., Zandonella, G., Haalck, L., Spener, F., Hermetter, A.,
Paltauf, F., 1996. Inhibition of microbial lipases with stereoiso-
meric triradylglycerol analog phosphonates. Biochim. Biophys.
Acta 1304, 229–244.
˚
Tilbeurgh, H., 1995. The 2.46 A resolution structure of the pancre-
atic lipase–colipase complex inhibited by a C11 alkyl phosphonate.
Biochemistry 34, 2751–2762.
Hamilton, R., Walker, B., Walker, B.J., 1998. Synthesis and proteinase
inhibitorypropertiesofdiphenylphosphonateanaloguesofaspartic
and glutamic acids. Bioorg. Med. Chem. Lett. 8, 1655–1660.
Kidd, D., Liu, Y., Cravatt, B.F., 2003. Profiling serine hydrolase activ-
ities in complex proteomes. Biochemistry 40, 4005–4015.
Liu, Y., Patricelli, M.P., Cravatt, B.F., 1999. Activity-based protein
profiling: the serine hydrolases. Proc. Natl. Acad. Sci. U.S.A. 96,
14694–14699.
Taylor, C.M., O’Connor, C.J., Chamberlain, J.L., Sun, C.Q., 2001.
Design, synthesis and evaluation of two diastereoisomeric phos-
phonate inhibitors of lamb pregastric lipase. J. Mol. Catal. B:
Enzym. 15, 15–22.
Vanwetswinkel, S., Touillaux, R., Fastrez, J., Marchand-Brynaert, J.,
1995. Bifunctional activity labels for selection of filamentous bac-
teriophages displaying enzymes. Bioorg. Med. Chem. 3, 907–
915.
Mannesse, M.L.M., Boots, J.-W.P., Dijkman, R., Slotboom, A.T., van
der Hijden, H.T.W.M., Egmond, M.R., Verheij, H.M., de Haas,
G.H., 1995. Phosphonate analogues of triacylglycerols are potent
inhibitors of lipase. Biochim. Biophys. Acta 1259, 56–64.
Zandonella, G., Stadler, P., Haalck, L., Spener, F., Paltauf, F., Hermet-
ter, A., 1999. Interactions of fluorescent triacylglycerol analogs
covalently bound to the active site of a lipase from Rhizopus oryzae.
Eur. J. Biochem. 262, 63–69.