New fluorous photoaffinity labels (F-PAL) and their application in V-ATPase inhibition studies

Article abstract of DOI:10.1002/ejoc.200901463

(Trifluoromethyl)diazirines are well established photoaffinity labels (PAL) used in biochemical investigations to create covalent ligand-receptor bonds. Two new diazirinylbenzoic acids 8b, c with perfluorobutyl and perfluorooctyl chains (FPAL) were efficiently prepared from p-bromobenzaldehyde and attached to the highly potent and specific V-ATPase inhibitors 21-deoxyconcanolide A (2) and bafilomycin A₁ (5), deriving from the natural product pool from Streptomyces producer strains. The labelled derivatives 17 and 18 were efficiently purified by fluorous solid-phase extraction. Func-tional biochemical assays with the V-ATPase holoenzymes proofed, strong inhibition activities. So far, radioactive isotopes or biotin-tags have mainly been used for tracing compounds in photoaffinity studies. The C₄F₉- and C ₈F₁₇-fluorous tags aim to enable advantageous separation and identification of labelled peptide fragments by fluorous chromatography followed by MS analysis. Therefore, F-PAL represents an innovative new concept for binding site determination and should significantly accelerate and simplify such biochemical investigations.

Full text of DOI:10.1002/ejoc.200901463

FULL PAPER
DOI: 10.1002/ejoc.200901463
New Fluorous Photoaffinity Labels (F-PAL) and Their Application in
V-ATPase Inhibition Studies
Nadja Burkard,[a] Tobias Bender,[b] Johannes Westmeier,[b] Christin Nardmann,[c]
Markus Huss,[c] Helmut Wieczorek,[c] Stephanie Grond,*[a] and Paultheo von Zezschwitz*[b]
Dedicated to Professor Christian Reichardt on the occasion of his 75th birthday
Keywords: Bafilomycin / Concanamycin / Fluorous chromatography / Perfluorinated reagents / Photoaffinity labeling / V-
ATPase inhibitors
(Trifluoromethyl)diazirines are well established photoaffinity
labels (PAL) used in biochemical investigations to create co-
valent ligand-receptor bonds. Two new diazirinylbenzoic ac-
ids 8b,c with perfluorobutyl and perfluorooctyl chains (F-
PAL) were efficiently prepared from p-bromobenzaldehyde
and attached to the highly potent and specific V-ATPase in-
hibitors 21-deoxyconcanolide A (2) and bafilomycin A1 (5),
deriving from the natural product pool from Streptomyces
tional biochemical assays with the V-ATPase holoenzymes
proofed strong inhibition activities. So far, radioactive iso-
topes or biotin-tags have mainly been used for tracing com-
pounds in photoaffinity studies. The C4F9- and C8F17-fluo-
rous tags aim to enable advantageous separation and identi-
fication of labelled peptide fragments by fluorous chromatog-
raphy followed by MS analysis. Therefore, F-PAL represents
an innovative new concept for binding site determination
producer strains. The labelled derivatives 17 and 18 were ef- and should significantly accelerate and simplify such bio-
ficiently purified by fluorous solid-phase extraction. Func-
chemical investigations.
mutagenesis.[3] In addition, Zeeck and co-workers prepared
concanolide 3 for photoaffinity labelling studies.[4] The 3-
aryl-3-(trifluoromethyl)diazirine moiety at R2 served as
photoactivatable group in these experiments and led to a
covalent bond between inhibitor and binding site. Tracing
and detection was realized with the radioactive 125I at R3.
These studies strongly suggested subunit c of the V-ATPase
as the binding site of concanamycins, however, were ham-
pered by a seriously reduced inhibitory potential of com-
pound 3 (IC50 = 15–20 µ) possibly due to modification of
two hydroxy groups in concanolide analogue 3. Other trac-
ing methods like voluminous fluorescent-tagged PAL were
not suitable for these V-ATPase studies.[4c] Recently, this
problem was solved by our development of the new photo-
affinity label (PAL) 7 which contains a radioactive 14C atom
in the photosensitive substituent itself and offers the ad-
ditional advantage of the very long half-life of 14C (τ =
5730 y). Indeed, inhibitors 4 and 6, modified only at the
tetrahydropyrane rings, retain most of the high inhibitory
activities of the natural products 1 and 5.[5] Photoactivation
of these compounds in the presence of the V-ATPase holo-
enzyme from the tobacco hornworm Manduca sexta, sepa-
ration of the subunits by SDS PAGE and detection by auto-
radiography furnished further insights into involved sub-
units (details will be reported in due course). Yet, attempts
to analyze tagged, very non-polar peptide fragments by
HPLC-MS analysis proofed to be difficult – this is a general
Introduction
Fundamental research and development of new drugs
demand a detailed molecular understanding of ligand-tar-
get recognition and interaction between pharmacophore
and target molecule. The V-type ATPase – an important
molecular proton transporting machine and pH regulator
in almost all eukaryotic cells – shows a great potential as
target in osteoporosis and cancer therapy.[1] This was shown
in a number of studies with the microbial metabolites con-
canamycin A (1) from Streptomyces strain Gö 22/15, its de-
rivative 21-deoxyconcanolide A (2) or bafilomycin A1 (5)
from strain Gö 3822-F4 (Figure 1) which all belong to the
most selective and potent V-ATPase inhibitors with IC50
values at nanomolar range.[2] In order to determine the
mechanism of inhibition, Bowman et al. investigated the
binding site of these plecomacrolides using site-directed
[a] Institut für Organische Chemie der Eberhard-Karls-Universität
Tübingen,
Auf der Morgenstelle 18, 72076 Tübingen, Germany
Fax: +49-7071-295397,
E-mail: stephanie.grond@uni-tuebingen.de
[b] Fachbereich Chemie, Philipps-Universität Marburg,
Hans-Meerwein Straße, 35032 Marburg, Germany
Fax: +49-6421-2825362,
E-mail: zezschwitz@chemie.uni-marburg.de
[c] Tierphysiologie – FB Biologie/Chemie, Universität Osnabrück,
Barbarastrasse 11, 49076 Osnabrück, Germany
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.200901463.
2176
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2176–2181

New Fluorous Photoaffinity Labels (F-PAL)
problem in PAL studies since the efficiency of crosslinking unit in a single compound. Again, only one operational step
is usually low and only traces of labelled fragments need to is then necessary for their attachment to the bioactive li-
be identified among a vast excess of unlabelled ones.[6]
gand. Identification of the labelled fragments would then
be achieved by fluorous chromatography and HPLC-MS
analysis. The chemical synthesis of such photoaffinity labels
has been devised independently by Zhang et al. and us,[10]
and we here present the first successful practical use of the
new fluorous photoaffinity labels (F-PAL) in a ligand-target
binding study exemplified with the functional V-ATPase in-
hibition.
Results and Discussion
The interaction of fluorinated compounds with fluorous
stationary phases is critically dependent on the balance be-
tween their organic and fluorous character.[9] Therefore,
we decided to prepare F-PAL 8b and 8c with different
lengths of the perfluoroalkyl chains, namely perfluorobutyl
and perfluorooctyl while perfluoropropyl and perfluo-
rohexyl residues were chosen independently by Zhang et
al.[10a] Our synthesis commenced with metallation of the
respective perfluoroalkyl iodides with phenylmagnesium
bromide and addition to p-bromobenzaldehyde (9) furnish-
ing alcohols 10b,c which were oxidized using the Swern pro-
tocol or Dess–Martin periodinane (DMP, Scheme 1).
Ketones 11b,c were then transformed into diazirines 14b,c
along the common approach[11] consisting of formation of
oximes 12b,c, their tosylation and treatment with liquid am-
monia in an autoclave at room temp. Conversion to benzoic
acids 16b,c was then performed following our protocol[5] of
temporary silylation of the diaziridine moiety with trimeth-
ylsilyl triflate, bromide-lithium exchange and carboxylation
with carbon dioxide. Finally, oxidation to the diazirines 8b,c
with iodine/NEt3 completed an eight-step synthesis with
Figure 1. Structures of V-ATPase inhibitors 1–6 and photoaffinity
labels 7, 8.
An elegant possibility to overcome this difficulty is the overall yields of 3% (8b) and 11% (8c), respectively.
“fishing approach” for ligand-protein target studies.[7] Fre-
The potent microbial natural products concanamycin A
quently, a biotin molecule is attached to the bioactive ligand (1) and bafilomycin A1 (5) were isolated and purified start-
or the photosensitive group itself, e.g., to diazirine-based ing from cultures of Streptomyces sp. Gö 22/15 and
PAL. After photoactivation and digestion, labelled frag- Gö 3822-F4, respectively.[12] Preparation of the chemically
ments are easily “fished” from unlabelled peptide fragments more stable aglycon 21-deoxyconcanolide A (2) was per-
using avidin- or streptavidin-based matrices and analysed formed via methylation, reduction and deglycosylation as
by mass spectrometry. Yet, in some cases this sterically de- described.[4a] In former coupling reactions of PAL 7 and 8a
manding tracer group unfavourably interferes with the li- with these inhibitors, the hydroxy groups on the tetra-
gand-target complex.[8]
hydropyran rings of 2 (at C-23) and 5 (at C-21) had shown
Therefore, we became interested in the development of a to be most reactive and were thus selectively esterified.[5,13]
new non-voluminous tracer group and herewith bring the Accordingly, the desired derivatives 17b,c were obtained
young method of fluorous chromatography and fluorous from esterification with the respective acid chlorides ob-
solid phase extraction (FSPE) into play.[9] The outstanding tained by treatment of 8b,c with thionyl chloride
property of this separation technique is that only fluori- (Scheme 2). Derivatives 18b,c of bafilomycin A1 (5) carrying
nated compounds are retarded on the fluorinated silica gel the fluorous photoaffinity labels (F-PAL) were obtained
while all other organic material – (mostly) regardless of its from mild EDCI-mediated reactions. Direct light was ob-
physicochemical properties – is washed off a column with structed from all reactions and products. Purification of the
the appropriate organic mobile phase. Thus, we speculated crude F-PAL inhibitor fractions from standard silica gel
that substitution of the trifluoromethyl group in the non- chromatography already took advantage of fluorous solid
radioactive PAL 8a by a longer perfluoroalkyl residue phase extraction. Thus, concanamycin derivatives 17b,c
would lead to a new fluorous photoaffinity label (F-PAL) were applied to C8F17-modified silica gel and eluted with
containing both, the photosensitive moiety and the tracer MeOH/H2O (80:20) as fluorophobic, and MeOH as fluoro-
Eur. J. Org. Chem. 2010, 2176–2181
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2177

S. Grond, P. von Zezschwitz et al.
FULL PAPER
Scheme 1. Synthesis of the new fluorous photoaffinity labels (F-PAL) 8b,c.
philic solvents to afford highly pure compounds based on with the hydrophobic enzyme surface. The thus obtained
TLC and NMR analysis in yields of 0.6 mg (17b) and six F-PAL-substituted probes 17, 18 and 20 as well as the
1.0 mg (17c).[13]
native natural products concanamycin A (1) and bafilomy-
cin A1 (5) as further controls were subjected to the func-
tional V-ATPase holoenzyme inhibition assay performed as
described by Huss et al.[4b]
Table 1. Conditions of fluorous solid phase extraction (FSPE).
Compound
Solvent [fluorophobic/fluorophilic]
17b
17c
18b
18c
20b
20c
MeOH/H2O 80:20/MeOH
MeOH/H2O 80:20/MeOH
MeCN/H2O 70:30/MeCN
MeCN/H2O 70:30/MeCN
MeOH/H2O 80:20/MeOH
MeOH/H2O 80:20/MeOH
The results of the biochemical investigation are shown
in Figure 2 and Figure 3 and point out that concanamycin
analogues 17b,c have a very high inhibition potential with
IC50 values of approximately 0.06 µ. Thus, they are about
one order of magnitude inferior to the values of the natural
product concanamycin A (1) (IC50 = 0.005 µ) and of con-
canolide derivative 17a[5,14] (IC50 = 0.01 µ) modified with
the trifluoromethyl-based PAL 8a. The longer perfluoro-
alkyl chains in agents 17b,c might therefore slightly interfere
with the binding sites of the V-ATPase. However, there is
no significant inhibition of the V-ATPase by model com-
pounds 20b,c which excludes the possibility of the F-PAL
interacting unspecifically with the relevant binding site of
Scheme 2. Attachment of F-PAL 8b or 8c to V-ATPase inhibitors
2, 5 and model compound 19.
Since the macrolide bafilomycin A1 (5) is known to be the holoenzyme to give a false positive result. Contrary, the
highly unstable in MeOH solution an alternative FSPE pro- inhibition potential of bafilomycin analogues 18b,c is lower
cedure had to be established for purification of compounds than that of the natural product as drawn from the IC50
18. Among mixtures containing THF, acetone or DMF, best values, both ϾϾ 100 µ. Taking into account that intro-
results were observed with MeCN/H2O (70:30) as fluo- duction of trifluoromethyl-based PAL 8a leads to a de-
rophobic and pure MeCN as fluorophilic solvents,[13] and crease of the activity by the factor 20 (IC50 = 0.08 µ for
photoaffinity agents 18b and 18c were obtained in yields 18a),[5,14] bafilomycin A1 (5) seems to be more sensitive
of 2.2 mg and 4.6 mg, respectively (Table 1). Moreover, the than concanolide 2 for disturbing substituents like the fluo-
tetrahydronaphth-2-ol derived agents 20b,c were prepared: rous tags. Hitherto, the exact reasons for these differences
Since one of the binding sites at the V-ATPase is extremely in activity are unknown; there might be an interaction be-
apolar we aimed to unambiguously exclude a misleading tween the photoaffinity label, the fluorous tail and the mac-
inhibitory activity of the apolar fluorous tails themselves, rolide ring of bafilomycin A1 (5) in a different way com-
that is, the possibility of unspecific inhibitory interactions pared to the similar structures of the concanamycins.
2178
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2176–2181

New Fluorous Photoaffinity Labels (F-PAL)
both entities in a single molecule offers the advantage of
requiring only one operational step for functionalization of
the bioactive ligand thus causing less modification of its
structure.
Moreover, we have shown the applicability of this new
concept in two aspects: First, the F-PAL could easily be
attached to potent V-ATPase inhibitors, and as an example
the concanamycin derivatives 17b,c retained the high and
specific binding affinity. This was not obvious from the be-
ginning due to the highly apolar character of the perfluoro-
alkyl chains which could either inhibit formation of the li-
gand-receptor complex or even lead to an unspecific bind-
ing to the enzyme surface – the later possibility was ex-
cluded by the thoroughly applied control substances 20b,c.
Second, the fluorous tails on the photoaffinity label add
sufficient fluorous character to the complex photoaffinity
probes 17 and 18 with molecular weights of more than
1 kDa to enable their efficient and reliable separation and
purification by fluorous solid phase extraction of the crude
mixtures. This is a strong indication for a successful future
separation of the labelled peptide fragments after photoac-
tivation and digestion of the ligand-receptor complex, espe-
cially since van Boom et al. already described purification
of a synthetic oligopeptide of 22 amino acids using a per-
fluorooctyl anchor.[15]
Figure 2. Inhibition of the V-ATPase with concanamycin A (1),
derived F-PAL agents 17a,b,c and F-PAL control compounds
20b,c.
Thus, we expect that the concept of F-PAL will afford
significant improvements in elucidating the exact peptide
sequences of the sites of the V-ATPase to which inhibitors
bind and, by doing so, will contribute to the overall long-
term aim to unravel the function of the V-ATPase proton
pumping machinery. Additionally, linkage of the concepts
of photoaffinity labelling and fluorous chromatography is
thought to replace less convenient labels and generate a
method to make binding studies much faster and easier.
Experimental Section
Figure 3. Inhibition of the V-ATPase with bafilomycin A1 (5) and
derived F-PAL-agents 18a,b,c (control compounds 20b,c; see Fig-
ure 2).
General: See Supporting Information.
General Procedure for Esterification with 21-Deoxyconcanolide A
(2): Diazirinyl compounds were all handled under low light condi-
tions. The respective F-PAL 8b (19.0 mg, 49.9 µmol) or 8c (30.8 mg,
53.1 µmol) and thionyl chloride (0.50 mL, 6.9 mmol) were stirred
for 15 min at room temp., excess thionyl chloride was removed un-
der reduced pressure (20 mbar) and the residue was dissolved in
dichloromethane (CH2Cl2, 2 mL). Concanolide A (2) and DMAP
were dissolved in CH2Cl2 (1.5 mL) and treated with triethylamine
(35.0 mg, 346 µmol) and 4-[(3-perfluorobutyl)diazirin-3-yl]benzoyl
chloride or 4-[(3-perfluorooctyl)diazirin-3-yl]benzoyl chloride dis-
solved in CH2Cl2, respectively. The mixture was stirred for 3 h and
treated with a second portion of the respective benzoyl chloride
dissolved in CH2Cl2. The reaction was stirred for 24 h and then
diluted with water (1 mL) and ethyl ether (150 mL). The mixture
was washed with saturated aqueous NaHCO3 (1ϫ 5 mL), brine
(1ϫ 5 mL) and water (1ϫ 5 mL). The organic layer was dried with
Na2SO4 and the solvent was removed under reduced pressure. The
crude product was initially purified by column chromatography
(silica gel, CHCl3/EtOAc/MeOH, 10:5:0.5).
One can only speculate if the fluorous tail might change
the folding of the macrocyclic ring in structure 5, but any-
way, the binding of inhibitor analogues 18b,c is too low
to allow for successful photoaffinity studies. Concanamycin
derivatives 17b,c, however, are perfectly suited for this appli-
cation and first experiments to separate and identify lab-
elled protein fragments using fluorous chromatography and
MS-MS analyses are currently under way.
Conclusions
A new type of fluorous photoaffinity label (F-PAL) has
been developed equipped with both, a photoactivatable
group and a new tracer unit. While diazirines are generally
meant to be the most suitable photoreactive moiety[6] the
perfluoroalkyl chains are promising anchors for the “fishing
General Procedure for Esterification with Bafilomycin A (5): Diaziri-
technique” using fluorous chromatography. Combination of nyl compounds were all handled under low light conditions. 4-[(3-
Eur. J. Org. Chem. 2010, 2176–2181
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2179

S. Grond, P. von Zezschwitz et al.
Perfluorobutyl)diazirin-3-yl]benzoic acid (8b) or 4-[(3-perfluorooc- 28-H3), 1.66 (m, 1 H, 20-H), 1.84 (s, 3 H, 12-CH3), 1.95 (s, 3 H, 4-
FULL PAPER
tyl)diazirin-3-yl]benzoic acid (8c), respectively, were dissolved in
dry CH2Cl2 (0.8 mL) and cooled to 0 °C. EDCI was added and the
solution was stirred for 5 min. After addition of DMAP, the reac-
tion mixture was stirred for 3 min and a solution of bafilomycin
A1 (5) in CH2Cl2 (0.4 mL) was added dropwise. The reaction mix-
ture was stirred for 18 h in the dark and then diluted with ethyl
ether (150 mL). The mixture was washed with saturated aqueous
NaHCO3 (1ϫ 5 mL), brine (1ϫ 5 mL) and water (1ϫ 5 mL). The
organic layer was dried with Na2SO4 and the solvent was removed
under reduced pressure. The crude product was initially purified by
column chromatography (silica gel, cyclohexane/acetone, 6:1).
CH3), 2.01 (m, 2 H, 11-H2), 2.01 (m, 1 H, 18-H), 2.22 (m, 1 H, 22-
Hb), 2.30 (m, 1 H, 10-H), 2.68 (m, 1 H, 6-H), 3.14 (m, 1 H, 9-H),
3.22 (s, 3 H, 16-OCH3), 3.30 (m, 1 H, 25-H), 3.48 (dt, 3J = 8.0,
0.3 Hz, 1 H, 21-H), 3.54 (s, 3 H, 2-OCH3), 3.62 (dd, 3J = 12.0,
3
1.0 Hz, 1 H, 19-H), 3.68 (m, 1 H, 7-H), 3.86 (dd, J = 6.0, 6.0 Hz,
1 H, 16-H), 4.88 (m, 1 H, 23-H), 5.18 (m, 1 H, 17-H), 5.22 (m, 1
H, 15-H), 5.32 (m, 1 H, 26-H), 5.38 (m, 1 H, 27-H), 5.63 (m, 1 H,
3
13-H), 5.74 (m, 1 H, 5-H), 6.42 (s, 1 H, 3-H), 6.56 (dd, J = 14.4,
10.4 Hz, 14-H), 7.45 (d, 3J = 10.8 Hz, 2 H, PhH, 4Ј, 6Ј-H), 8.04 (d,
3J = 10.8 Hz, 2 H, PhH, 3Ј, 7Ј-H) ppm. 13C NMR (125.7 MHz,
CD2Cl2/CD3OD, 2:1): δ = 8.2 (20-CH3), 9.7 (18-CH3), 13.5 (24-
CH3), 17.7 (C-28), 29.8 (C-8Ј, mC), 35.9 (C-10), 39.8 (C-11), 41.4
(C-20), 55.8 (16-OCH3), 59.6 (2-OCH3), 70.0 (C-19), 77.8 (C-23),
83.0 (C-16), 128.5 (C-15), 129.5 (C-27), 130.4 (C-Ar), 132.4 (C-Ar),
133.5 (C-Ar), 134.0 (C-Ar), 165.6 (C-1Ј) ppm. ESI-MS (positive
ions): m/z (%) = 1261.4 [M + Na]+. ESI-MS (negative ions): m/z
(%) = 1237.5 [M – H]–. MW calcd. = 1239.1, C55H67F17N2O10. Not
all 13C-NMR signals could be detected at room temp. as described
previously.[16]
Fluorous Solid Phase Extraction: The extractions of the concana-
mycin derivatives 17b and 17c were performed with a solvent mix-
ture of MeOH/H2O (80:20) for the non-fluorous and pure MeOH
for the fluorous wash. The extractions of the bafilomycin deriva-
tives 18b and 18c were performed with a solvent mixture of MeCN/
H2O (70:30) for the non-fluorous and pure MeCN for the fluorous
wash.[13]
23-O-[4-(3-Perfluorobutyl-3H-diazirin-3-yl)benzoyl]-21-deoxycon-
canolide (17b): The reaction was performed as described above with
reduced reaction time of 5 h employing 21-deoxyconcanolide (2)
(9.7 mg, 14 µmol), the benzoyl chloride of F-PAL 8b (21.0 mg,
52.7 µmol) and DMAP (11.7 mg, 95.7 µmol) to furnish 0.65 mg
(0.63 µmol, 4.5%) of 17b and 5.8 mg (8.5 µmol, 60%) of the re-
isolated starting material 2. Rf = 0.48 (cyclohexane/acetone, 6:1).
1H NMR (600 MHz, CD2Cl2/CD3OD, 2:1): δ = 0.86 (m, 3 H, 6-
CH3), 0.86 (m, 3 H, 8-CH2CH3), 0.86 (m, 3 H, 18-CH3), 0.86 (m,
3 H, 20-CH3), 0.86 (m, 3 H, 24-CH3), 1.05 (m, 3 H, 10-CH3), 1.15
(m, 1 H, 22-Ha), 1.19 (m, 1 H, 24-H), 1.24 (m, 2 H, 8-CH2CH3),
1.46 (m, 1 H, 8-H), 1.62 (dd, 3J = 7.2, 1.2 Hz, 3 H, 28-H), 1.66 (m,
1 H, 20-H), 1.84 (s, 3 H, 12-CH), 1.94 (s, 3 H, 4-CH3), 1.99 (m, 2
H, 11-H2), 2.06 (m, 1 H, 18-H), 2.26 (m, 1 H, 22-Hb), 2.30 (m, 1
H, 10-H), 2.68 (m, 1 H, 6-H), 3.14 (m, 1 H, 9-H), 3.22 (s, 3 H, 16-
OCH3), 3.25 (m, 1 H, 25-H), 3.48 (dt, 3J = 8.0, 0.3 Hz, 1 H, 21-
H), 3.54 (s, 3 H, 2-OCH3), 3.62 (dd, 3J = 12.0, 1.0 Hz, 1 H, 19-H),
3.68 (m, 1 H, 7-H), 3.86 (dd, 3J = 6.0, 6.0 Hz, 1 H, 16-H), 4.88 (m,
1-H, 23-H), 5.15 (m, 1 H, 17-H), 5.19 (m, 1 H, 15-H), 5.32 (m, 1
H, 26-H), 5.63 (m, 1 H, 27-H), 5.74 (m, 1 H, 5-H), 5.76 (m, 1 H,
21-O-[4-(3-Perfluorobutyl-3H-diazirin-3-yl)benzoyl]bafilomycin A1
(18b): The reaction was performed as described above employing
bafilomycin A1 (5) (17.3 mg, 27.8 µmol), DMAP (8.0 mg,
65.5 µmol), EDCI (11.4 mg, 59.5 µmol) and label 8b (8.5 mg,
22 µmol). The reaction time was elongated to 22 h to give 2.2 mg
1
(2.2 µmol, 8.0%) of 18b. Rf = 0.32 (cyclohexane/acetone, 6:1). H
NMR (600 MHz, CD2Cl2): δ = 0.76–0.82 (m, 3 H, 22-CH3), 0.76–
0.82 (m, 3 H, 16-CH3), 0.76–0.82 (m, 3 H, 25-H3), 0.96 (m, 3 H,
8-CH3), 0.96 (m, 3 H, 26-H3), 1.01 (d, 3J = 10.0 Hz, 3 H, 18-CH3),
3
1.04 (d, J = 10.0 Hz, 3 H, 6-CH3), 1.24 (m, 1 H, 20-Ha), 1.56 (m,
1 H, 18-H), 1.68 (m, 1 H, 8-H), 1.76 (m, 1 H, 22-H), 1.90 (m, 1 H,
24-H), 1.92 (m, 1 H, 9-Ha), 1.93 (s, 3 H, 10-CH3), 1.96 (d, 3J =
1.0 Hz, 3 H, 4-CH3), 2.12 (m, 1 H, 9-Hb), 2.12 (m, 1 H, 16-H),
3
2.41 (dd, J = 13.2, 7.0 Hz, 1 H, 20-Hb), 2.54 (m, 1 H, 6-H), 3.10
(s, 3 H, 14-OCH3), 3.24 (m, 1 H, 7-H), 3.62 (s, 3 H, 2-OCH3), 3.64
3
(m, 1 H, 23-H), 3.86 (dd, J = 9.0, 9.0 Hz, 1 H, 14-H), 4.12 (m, 1
3
H, 17-H), 4.94 (d, J = 9.0 Hz, 1 H, 15-H), 5.18 (m, 1 H, 13-H),
3
5.18 (m, 1 H, 21-H), 5.53 (m, 1 H, 19-OH), 5.80 (d, J = 9.0 Hz, 1
H, 5-H), 5.81 (d, 3J = 10.0 Hz, 1 H, 11-H), 6.48 (dd, 3J = 13.0,
10.0 Hz, 1 H, 12-H), 6.66 (s, 1 H, 3-H), 7.40 (d, 3J = 10.8 Hz, 2 H,
PhH, 4Ј, 6Ј-H), 8.11 (d, 3J = 10.8 Hz, 2 H, PhH, 3Ј, 7Ј-H) ppm.
13C NMR (125.7 MHz, CD2Cl2): δ = 7.1 (18-CH3), 9.9 (16-CH3),
12.5 (22-CH3), 14.1 (4-CH3), 14.4 (24-CH3), 17.3 (6-CH3), 20.2 (10-
CH3), 21.1 (24-CH3), 21.7 (8-CH3), 29.7 (C-24), 36.7 (C-6), 37.2
(C-16), 55.6 (14-OCH3), 60.1 (2-OCH3), 70.6 (C-17), 75.3 (C-21),
75.6 (C-23), 77.6 (C-15), 81.2 (C-7), 82.2 (C-14), 98.8 (C-19), 124.7
(C-11), 128.0 (C-13), 130.0 (C-Ar), 133.2 (C-4), 142.9 (C-2), 167.8
(C-1) ppm. 19F NMR (282 MHz, CD2Cl2): δ = –80.88 (m, 3 F,
CF3), –109.41 (m, 2 F, CF2), –121.14 (m, 2 F, CF2), –126.16 (m, 2
F, CF2) ppm. MW calcd. = 985.0, C47H61F9N2O10. Not all 13C-
NMR signals could be detected.[12]
3
13-H), 6.42 (s, 1 H, 3-H), 6.58 (dd, J = 14.4, 10.4 Hz, 14-H), 7.45
3
3
(d, J = 10.8 Hz, 2 H, PhH, 4Ј, 6Ј-H), 8.04 (d, J = 10.8 Hz, 2 H,
PhH, 3Ј, 7Ј-H) ppm. 13C NMR (125.7 MHz, CD2Cl2/CD3OD, 2:1):
δ = 8.2 (20-CH3), 9.8 (18-CH3), 13.5 (24-CH3), 14.1 (4-CH3), 17.7
(C-28), 23.1 (8-CH2CH3), 29.7 (C-8Ј, mC), 36.0 (C-10), 40.0 (C-20),
41.5 (C-11), 55.9 (16-OCH3), 60.0 (2-OCH3), 70.0 (C-19), 76.9 (C-
21), 77.9 (C-23), 83.0 (C-25), 128.6 (C-15), 129.5 (C-27), 130.4 (C-
Ar), 132.5 (C-Ar), 133.54 (C-Ar), 134.0 (C-Ar), 165.6 (C-1Ј) ppm.
ESI-MS (positive ions): m/z (%) = 1061.5 [M + Na]+. ESI-MS
(negative ions): m/z (%) = 1083.5 [M + HCOO]–. MW calcd. =
1039.1, C51H67F9N2O10. Not all 13C-NMR signals could be de-
tected at room temp. as described previously.[16]
21-O-[4-(3-Perfluorooctyl-3H-diazirin-3-yl)benzoyl]bafilomycin A1
(18c): The reaction was performed as described above employing
bafilomycin A1 (5) (10.4 mg, 16.7 µmol), DMAP (4.8 mg, 39 µmol),
EDCI (6.9 mg, 36 µmol) and label 8c (7.8 mg, 13 µmol) to furnish
4.6 mg (3.9 µmol, 23%) of 18c. Rf = 0.34 (cyclohexane/acetone,
6:1). 1H NMR (600 MHz, CD2Cl2): δ = 0.76–0.89 (m, 3 H, 22-
CH3), 0.76–0.89 (m, 3 H, 16-CH3), 0.76–0.89 (m, 3 H, 25-H3), 0.90
23-O-[4-(3-Perfluorooctyl-3H-diazirin-3-yl)benzoyl]-21-deoxycon-
canolide (17c): The reaction was performed as described above em-
ploying 21-deoxyconcanolide (2) (8.6 mg, 13 µmol), the benzoyl
chloride of F-PAL 8c (20 mg, 33 µmol) and DMAP (9.9 mg,
81 µmol) to furnish 0.95 mg (0.77 µmol, 5.9%) of 17c and 4.8 mg
(7.1 µmol, 56%) of the re-isolated starting material 2. Rf = 0.52
1
3
(cyclohexane/acetone, 6:1). H NMR (600 MHz, CD2Cl2/CD3OD,
(m, 3 H, 8-CH3), 0.90 (m, 3 H, 26-CH3), 0.98 (d, J = 10.0 Hz, 3
3
2:1): δ = 0.86 (m, 3 H, 6-CH3), 0.86 (m, 3 H, 8-CH2CH3), 0.86 (m, H, 18-CH3), 1.02 (d, J = 10.0 Hz, 3 H, 6-CH3), 1.26 (m, 1 H, 20-
3 H, 18-CH3), 0.86 (m, 3 H, 20-CH3), 0.86 (m, 3 H, 24-CH3), 1.05 Ha), 1.58 (m, 1 H, 18-H), 1.72 (m, 1 H, 8-H), 1.80 (m, 1 H, 22-H),
(m, 3 H, 10-CH3), 1.15 (m, 1 H, 22-Ha), 1.20 (m, 1 H, 24-H), 1.28
1.90 (m, 1 H, 24-H), 1.90 (m, 1 H, 9-Ha), 1.93 (s, 3 H, 10-CH3),
3
(m, 2 H, 8-CH2CH3), 1.38 (m, 1 H, 8-H), 1.60 (dd, 3J = 7.0, 1.0 Hz, 1.98 (d, J = 1.0 Hz, 3 H, 4-CH3), 2.13 (m, 1 H, 9-Hb), 2.13 (m, 1
2180
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2176–2181

New Fluorous Photoaffinity Labels (F-PAL)
3
[2] a) S. Dröse, K. U. Bindseil, E. J. Bowman, A. Siebers, A.
Zeeck, K. Altendorf, Biochemistry 1993, 32, 3902; b) B. J. Bow-
man, E. J. Bowman, J. Biol. Chem. 2002, 277, 3965–3972; c)
J. A. Beutler, T. C. McKee, Curr. Med. Chem. 2003, 10, 787–
796; d) M. Huss, H. Wieczorek, J. Exp. Biol. 2009, 212, 341–
346.
[3] a) E. J. Bowman, L. A. Grahams, T. H. Stevens, B. J. Bowman,
J. Biol. Chem. 2004, 279, 33131–33138; b) B. J. Bowman, M. E.
McCall, R. Baertsch, E. J. Bowman, J. Biol. Chem. 2006, 281,
31885–31893.
[4] a) G. Ingenhorst, K. U. Bindseil, C. Boddien, S. Dröse, M.
Gaßel, K. Altendorf, A. Zeeck, Eur. J. Org. Chem. 2001, 4525–
4532; b) M. Huss, G. Ingenhorst, S. König, M. Gaßel, S. Dröse,
A. Zeeck, K. Altendorf, H. Wieczorek, J. Biol. Chem. 2002,
277, 40544–40548; c) S. Dröse, C. Boddien, M. Gassel, G. In-
genhorst, A. Zeeck, K. Altendorf, Biochemistry 2001, 40, 2816–
2825.
[5] T. Bender, M. Huss, H. Wieczorek, S. Grond, P. Zezschwitz,
Eur. J. Org. Chem. 2007, 3870–3878.
[6] General reviews: a) J. Brunner, Annu. Rev. Biochem. 1993, 62,
483–514; b) F. Kotzyba-Hibert, I. Kapfer, M. Goeldner, Angew.
Chem. 1995, 107, 1391–1408; Angew. Chem. Int. Ed. Engl. 1995,
34, 1296–1312; c) A. Blencowe, W. Hayes, Soft Matter 2005, 1,
178–205.
[7] a) Y. Hatanaka, Y. Sadakane, Curr. Top. Med. Chem. 2002, 2,
271–288; b) Y. Sadakane, Y. Hatanaka, Anal. Sci. 2006, 22,
209–218; c) M. Hashimoto, Y. Hatanaka, Eur. J. Org. Chem.
2008, 2513–2523.
[8] J. Zotzmann, L. Hennig, P. Welzel, D. Müller, C. Schäfer, S.
Zillikens, H. Pusch, H. G. Glitsch, R. Regenthal, Tetrahedron
2000, 56, 9625–9632.
H, 16-H), 2.39 (dd, J = 13.2, 7.0 Hz, 1 H, 20-Hb), 2.54 (m, 1 H,
6-H), 3.24 (s, 3 H, 14-OCH3), 3.24 (m, 1 H, 7-H), 3.62 (s, 3 H, 2-
OCH3), 3.64 (dd, 3J = 12.0, 2.4 Hz, 1 H, 23-H), 3.90 (dd, 3J = 9.0,
9.0 Hz, 1 H, 14-H), 4.15 (m, 1 H, 17-H), 4.91 (d, 3J = 9.0 Hz, 1 H,
15-H), 5.12 (m, 1 H, 13-H), 5.14 (m, 1 H, 21-H), 5.44 (m, 1 H, 19-
3
3
OH), 5.78 (d, J = 9.0 Hz, 1 H, 5-H), 5.82 (d, J = 10.0 Hz, 1 H,
11-H), 6.54 (dd, 3J = 13.0, 10.0 Hz, 1 H, 12-H), 6.68 (s, 1 H, 3-H),
3
3
7.44 (d, J = 10.8 Hz, 2 H, PhH, 4Ј, 6Ј-H), 8.02 (d, J = 10.8 Hz,
2 H, PhH, 3Ј,7Ј-H) ppm. 13C NMR (125.7 MHz, CD2Cl2): δ = 7.2
(18-CH3), 9.9 (16-CH3), 12.5 (22-CH3), 14.1 (4-CH3), 14.3 (24-
CH3), 17.4 (6-CH3), 20.2 (10-CH3), 21.3 (24-CH3), 21.8 (8-CH3),
28.3 (C-24), 37.1 (C-6), 37.6 (C-16), 40.5 (C-22), 41.6 (C-8), 41.6
(C-9), 42.3 (C-18), 55.7 (14-OCH3), 60.2 (2-OCH3), 71.0 (C-17),
75.6 (C-21), 76.0 (C-23), 77.9 (C-15), 81.2 (C-7), 82.7 (C-14), 99.1
(C-19), 125.3 (C-11), 127.1 (C-13), 128.3 (C-Ar), 130.2 (C-Ar),
132.6 (C-Ar), 133.1 (C-4), 133.4 (C-12), 133.8 (C-3), 141.4 (C-2),
143.4 (C-5), 143.5 (C-10), 167.5 (C-1) ppm. 19F NMR (282 MHz,
CD2Cl2): δ = –81.12 (m, 3 F, CF3), –109.48 (m, 2 F, CF2), –120.35
(m, 2 F, CF2), –121.90 to –123.10 (m, 8 F, CF2), –126.37 (m, 2 F,
CF2) ppm. ESI-MS (positive ions): m/z (%) = 1207.4 [M + Na]+.
ESI-MS (negative ions): m/z (%) = 1183.4 [M – H]–. MW calcd. =
1185.0, C51H61F17N2O10. Not all 13C-NMR signals could be de-
tected.
Supporting Information (see also the footnote on the first page of
this article): Experimental procedures and analytical data for the
synthesis of F-PAL 8b,c and compounds 20b,c, details of the purifi-
1
cation by fluorous solid phase extraction and H NMR spectra of
compounds 17b,c.
[9] a) D. P. Curran, Synthesis 2001, 1488–1496; b) D. P. Curran
in: Handbook of Fluorous Chemistry (Eds.: J. A. Gladysz, D. P.
Curran, I. T. Horvath), Wiley-VCH, Weinheim, 2004, pp. 101–
127.
Acknowledgments
[10] a) Z. Song, Q. Zhang, Org. Lett. 2009, 11, 4882–4885; b) N.
Burkard, M. Huss, Book of abstracts, 3. Göttinger Chemie-
Forum, Göttingen, 03. July 2009; see also http://www.uni-goet-
tingen.de/de/117812.html.
[11] a) M. Nassal, Liebigs Ann. Chem. 1983, 1510–1523; b) Y. Hat-
anaka, H. Nakayama, Y. Kanaoka, Heterocycles 1993, 35, 997–
1004.
[12] T. Schuhmann, S. Grond, J. Antibiot. 2004, 57, 655–661.
[13] Details are described in the Supporting Information.
[14] Compounds 17a and 18a are the non-radioactive analogues of
compounds 4 and 6, respectively.
This work was financially supported by the Bundesministerium für
Bildung und Forschung (GenoMik+: MetabolitGenoMik) and the
Otto Römer Stiftung (to S. G.) as well as the Deutsche Forschungs-
gemeinschaft (DFG) (SFB 431, to H. W., M. H.). Thanks go to
Hans-Jörg Langer (from S. G.) for excellent technical assistance
and his prolific scientific encouragement and ideas and to Bastian
Weinert (from P. Z.) for preparation of compound 8c. H. W. and
M. H. thank Martin Dransmann for his excellent technical assist-
ance in purifying the V-ATPase.
[15] D. V. Filippov, D. J. van Zoelen, S. P. Oldfield, G. A.
van der Marel, H. S. Overkleeft, J. W. Drijfhout, J. H.
van Boom, Tetrahedron Lett. 2002, 43, 7809–7812.
[16] K. U. Bindseil, A. Zeeck, Liebigs Ann. Chem. 1994, 3, 305–312.
Received: December 15, 2009
[1] a) C. Farina, S. Gagliardi, Curr. Pharm. Des. 2002, 8, 2033–
2048; b) H. Izumi, T. Torigoe, H. Ishiguchi, H. Uramoto, Y.
Yoshida, M. Tanabe, T. Ise, T. Murakami, T. Yoshida, M. No-
moto, K. Kohno, Cancer Treat. Rev. 2003, 29, 541–549; c) K.
Beyenbach, H. Wieczorek, J. Exp. Biol. 2006, 209, 577–589.
Published Online: March 4, 2010
Eur. J. Org. Chem. 2010, 2176–2181
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2181