Synthesis of photoactive α-mannosides and mannosyl peptides and their evaluation for lectin labeling

Article abstract of DOI:10.1002/ejoc.200600449

Adhesion to the glycosylated surface of eukaryotic cells, mediated by lectins for example, plays an important role in inflammation and other cellular processes of living organisms. To elucidate the mechanisms involved in the adhesion to cell surfaces and

Full text of DOI:10.1002/ejoc.200600449

FULL PAPER
DOI: 10.1002/ejoc.200600449
Synthesis of Photoactive α-Mannosides and Mannosyl Peptides and Their
Evaluation for Lectin Labeling
Michaela Wiegand[a] and Thisbe K. Lindhorst*[a]
Keywords: Carbohydrates / Lectins / Photoaffinity labeling / Mannosylation / Biotinylation
Adhesion to the glycosylated surface of eukaryotic cells, me- labeled mannosides with an azide, a diazirine, and a benzo-
diated by lectins for example, plays an important role in in-
flammation and other cellular processes of living organisms.
To elucidate the mechanisms involved in the adhesion to cell
surfaces and their biological consequences, the investigation
of the molecular interactions between carbohydrate recogni-
tion domains of lectins and their ligands is of relevance. In
this work, we have selected the photoaffinity labeling tech-
nique for the exploration of the ligand binding to mannose-
phenone moiety were synthesized. Their crosslinking activity
was investigated by photolysis in the presence of six different
amino acids and with three model peptides, angiotensin II,
PTHIKWGD, and pentaglycine as well. The crosslinked ad-
ducts so obtained were analyzed by mass spectrometry. In
addition, difunctionalized mannosides were sought that con-
tained a photolabel and a biotin marker to facilitate the isola-
tion and the eventual identification, respectively, of the pho-
specific lectins, particularly the α-mannose-specific adhesin tolabeled peptides and proteins. To realize this concept, we
FimH, which is expressed at the tips of type 1 fimbriae of
Escherichia coli bacteria. We have designed and synthesized
a series of mannosides and glycopeptides derived thereof
that are equipped with a photoactive functional group. It was
our goal to compare the properties and labeling potencies of
different types of photolabile residues, and therefore, photo-
have employed orthogonally functionalized glycoamino acid
building blocks, which could be utilized as scaffold mole-
cules for the synthesis of our bifunctional target molecules.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
rial appendages, named fimbriae or pili,[5] for adhesion and
colonization of host cell surfaces. Uropathogenic Escher-
ichia coli express so-called type 1 fimbriae, which have been
shown to play a crucial role in adhesion, infection, invasion,
and pathogenicity.[6] The lectin domain of type 1 fimbriae
is represented by a protein called FimH,[7] which is specific
for α-mannosyl residues. The molecular details of carbo-
hydrate binding to FimH are not yet fully understood. X-
ray studies have revealed that FimH has a carbohydrate re-
cognition domain (CRD) at its tip, which is compatible with
mannose binding;[8] however, binding experiments with
various multivalent mannosyl clusters[9] as well as theoreti-
cal studies[10] have suggested that there might be additional
carbohydrate binding domains on FimH.
We have envisaged photoaffinity labeling of FimH with
suitable α-mannoside derivatives equipped with a suitable
photoreactive functional group (PAFG) to further investi-
gate the interactions of FimH with carbohydrates.[11] The
photoaffinity labeling technique has experienced consider-
able development and improvement during the last years,
and various optimized photolabile residues, such as aryl az-
ides,[12] diazirines,[13] and benzophenones,[14] have been suc-
cessfully used. Since it can hardly be predicted which of
these photoactive groups is most suited for photoaffinity
labeling in a particular case,[15] it became our goal to syn-
thesize and investigate a collection of α-mannosides that
carry different photolabels and compare the resulting pho-
Introduction
Photoaffinity labeling is a straightforward technique
used to explore molecular interactions of biomolecules such
as ligand binding to carbohydrate-specific proteins. Intro-
duced over 40 years ago,[1] the general principle of the pho-
toaffinity methodology still holds. Upon irradiation, the
photoprobe – a ligand that carries a photolabile moiety –
is converted into a highly reactive intermediate, which is
able to form a covalent bond with its receptor by insertion
into C–H, N–H, or O–H bonds.[2] The resulting adduct can
be analyzed by mass spectrometry after enzymatic digestion
of the labeled protein.
It is our goal to utilize photoaffinity-labeled carbo-
hydrates to assist in the investigation of the biological pro-
cesses that occur on cell surfaces, such as cell–cell com-
munication and bacterial adhesion. Involved in most of
these adhesion processes are carbohydrate-binding proteins
called lectins and selectins.[3] The interaction of lectins with
the cell surface carbohydrates influences cellular events
such as cell division or immune response.[4] Bacteria utilize
their own lectins, which are assembled as part of the bacte-
[a] Otto Diels Institute of Organic Chemistry, Christiana Albertina
University of Kiel,
24098 Kiel, Germany
Fax: +49-0431-880-7410
E-mail: tklind@oc.uni-kiel.de
Eur. J. Org. Chem. 2006, 4841–4851
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4841

M. Wiegand, T. K. Lindhorst
FULL PAPER
toactive mannosides in a series of irradiation experiments Therefore, 3 was first activated with hydroxylamine hydro-
to elucidate their photochemical properties and labeling po- chloride to yield corresponding oxime 4, and subsequently
tencies. To the best of our knowledge, this is the first study reacted with p-toluenesulfonyl chloride to give tosylated ox-
on photolabeling of simple protected amino acids (threo- ime 5, which is equipped with a highly efficient leaving
nine, serine, tyrosine, isoleucine, lysine, and arginine) on the group. Tosyloxime 5 was then reacted with liquid ammonia
one hand and model peptides (angiotensin II, PTHIKWGD overnight in dry ethyl ether at –60 °C to form the unstable
and pentaglycine) on the other with the employment of diaziridine, which was not isolated as reported[13a–c] but oxi-
various photophores in different ratios with the respective dized in situ with iodine in MeOH to afford tert-butyldi-
target molecule.
methylsilyl-protected diazirine 6 in good yield. Removal of
the TBDMS-protecting group with 10% hydrochloric acid
in dry MeOH gave the desired hydroxy-functionalized [p-
(hydroxymethyl)phenyl](trifluoromethyl)diazirine 7. Analyt-
ical data for 7 were in accordance with the literature; its
purification was possible by flash chromatography.
Results and Discussion
As photolabels (PALs), a diazirine ring, an aryl azide
moiety, and a benzophenone substructure were envisaged,
which would generate carbenes, nitrenes and radicals,
respectively, upon photolysis.[12–14] On the basis of molecu-
lar modeling studies[16] and with the use of the published
3D structure of our target protein, FimH,[8] we decided to
introduce these photolabels in the aglycon part of α-manno-
side ligands. Furthermore, we synthesized mannosyl pep-
tides as bifunctional scaffold molecules that allow for both
photolabeling and biotinylation. Finally, investigation of
synthesized molecules 12, 14, 16, and 19 in different photo-
lysis experiments is described here.
Scheme 1. a) TBDMS-Cl, imidazole, dry DMF, room temp., 87%;
b) nBuLi, dry THF, argon, –30 °C, N-trifluoroacetyl piperidine,
–50 °C, argon, 89%; c) hydroxylamine hydrochloride, dry EtOH,
argon, 60 °C, 60%; d) tosyl chloride, N,N-DMAP, triethylamine,
dry CH2Cl2, argon, 0 °C, room temp., 65%; e) NH3, dry diethyl
ether, –60 °C, triethylamine, 10% I2 in MeOH, 0 °C Ǟ room temp.,
55%; f) 10% HCl in dry MeOH, room temp., 77%.
Synthesis of Hydroxy-Functionalized Photolabels 7 and 9
For the synthesis of photolabeled mannose derivatives,
hydroxy-functionalized photoactive groups were needed in
order to allow for the introduction of a mannosyl residue
by glycosylation of the hydroxy group. We selected a diazir-
ine ring as the first photoactive group to be introduced into
our target molecules. We sought a diazirine substructure
Another photoactive group we aspired to incorporate
lacking α-hydrogens in order to avoid hydrogen abstraction into the mannosides is the aryl azide moiety, which is
after irradiation. This is a notorious problem when diazir- among the most common photoactive reagents.[12,13] How-
ine derivatives with α-hydrogens are used for photoaffinity ever, it has been reported that phenyl azides show almost
labeling.[13d,17] The carbenes, which are delivered after irra- no insertion products after irradiation in hydrocarbon sol-
diation of diazirines with UV light, often do not insert into vents at room temperature.[21] Instead, the singlet nitrenes
C–H, N–H, or O–H bonds of the receptors under investiga- that are generated by photolysis undergo rapid ring expan-
tion. Instead, the carbenes undergo intramolecular hydro- sion to form dehydroazepines[22] that are either intercepted
gen abstraction, which leads to unreactive alkenes. To cir- by nucleophiles to give amines, or in the absence of nucleo-
cumvent this undesired side reaction, as well as to suppress philes undergo polymerization.[23] When polyfluorinated
the isomerization to the diazo analog, we selected 3-trifluo- aryl azides such as pentafluorophenyl azides are used as
romethyl-3-phenyl diazirine[13,18] as the photolabel, which photoactive reagents, irradiation leads to a stabilized singlet
had to be functionalized in order to allow for the subse- nitrene, which is less prone to undesired ring expansion,
quent introduction of a mannosyl residue. Thus, p-bro- and is more readily trapped by nucleophiles, such as amines
mobenzyl alcohol (1) was selected as the starting material, or alcohols and even inactivated alkanes, to form stable co-
and it was protected as a silyl ether according to the litera- valent adducts.[21,24] Thus, we synthesized an aryl azide
ture[13] to form 2 (Scheme 1). It was then converted to tri- from pentafluorobenzaldehyde, which was reacted with so-
fluoroacetate 3 with freshly prepared N-trifluoroacetyl pi- dium azide in acetone–water to form p-azidobenzaldehyde
peridine after dehalogenation with butyllithium.[19] Ketone 8 after regioselective nucleophilic aromatic substitution in
3 was used as the starting material for the synthesis of the an improved yield as compared to the literature[21]
desired diazirine. This is normally obtained after the con- (Scheme 2). Chemoselective reduction of the aldehyde
version of a ketone into a diaziridine intermediate[13b,19,20] group in 8 was then accomplished with the dimethylamine–
followed by oxidation. However, in the case of ketone 3, the borane complex in acetic acid[25] to provide the literature-
corresponding diaziridine did not form possibly because of known benzyl alcohol 9[21] in a yield of 89% after chroma-
the stereoelectronic effects of the aromatic ring system. tographic purification.
4842
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4841–4851

Photoactive α-Mannosides and Mannosyl Peptides and Evaluation for Lectin Labeling
FULL PAPER
Mannosides 12 and 14 showed similar potency as inhibi-
tors of type 1 fimbriae-mediated bacterial adhesion as the
standard inhibitor of this carbohydrate-protein interaction
p-nitrophenyl α--mannoside.[28] Thus, they can be re-
garded as suitable candidates for the photoaffinity labeling
of FimH. Next, we wanted to explore a benzophenone sub-
structure as the third photolabel in our study as it has been
reported that benzophenones are also advantageous photo-
active groups.[14a] Photolabile benzophenones have been
employed as photophores either to functionalize remote C–
H bonds in biomolecules or to map binding sites and con-
formational properties of nucleotides or proteins.[14a] They
have three major chemical and biochemical advantages;
first, they show higher stability than arylazides and diazir-
ines; second, benzophenones can be used under ambient
light without any decomposition, and protein damage is
avoided because photolysis is achieved at wavelengths of
350-360 nm, third, the benzophenone group reacts prefer-
entially with unreactive C–H bonds in hydrophobic parts
of the biomolecules even in the presence of water or bulky
nucleophiles.
To introduce the benzophenone photolabel, we envisaged
the peptide coupling of 4-benzoylbenzoic acid with amino-
functionalized mannoside 15,[29] which is easily obtained
from peracetylated mannose in three steps.[30] This turned
out to be a most convenient synthetic route which led to
photoactive mannoside 16 in a HATU-mediated peptide
coupling reaction[31] (Scheme 4). The pure product was ob-
tained after purification on silica gel followed by RP-HPLC
in a yield of 52%.
Scheme 2. a) NaN3, acetone–water (1:1), reflux (100 °C), 95%; b)
dimethylamine–borane complex, 98% AcOH, 55 °C, 89 %.
Synthesis of Photolabeled Mannosides 12 and 14
With the employment of photolabels 7 and 9, mannos-
ides with a photoactive aglycon moiety could be obtained
by glycosylation with the use of acetyl-protected mannosyl
trichloroacetimidate 10 under Lewis acid catalysis.[26]
Azido-functionalized benzyl alcohol 9 was mannosylated
with TMSOTf to yield acetyl-protected mannoside 11 in
69% yield after column chromatography. This was sub-
sequently deprotected according to Zemplén[27] with freshly
prepared sodium methoxide in methanol (Scheme 3). After
purification by flash chromatography on reverse-phase (RP)
silica, desired photolabile α-mannoside 12, which carries
the azide functionality, was obtained in a yield of 89%. The
analogous procedure that starts with diazirine 7 led to acet-
ylated α-mannoside 13 in 69% after purification. Removal
of the acetyl protecting groups of 13 was successful with
Zemplén reaction conditions and freshly prepared sodium
methoxide reagent. After RP-flash chromatography, desired
photolabile α-mannoside 14 equipped with the diazirine
moiety was obtained in a yield of 88%.
Our next consideration was dedicated to the isolation
and analysis of the photoaffinity-labeled adducts. Photoly-
sis of a photoprobe–protein mixture followed by proteolytic
digestion of the photolabeled protein can lead to a complex
mixture of nonlabeled fragments together with the desired
photocrosslinked adduct as well as a number of undesired
side products. MS analysis of such a mixture is complicated
because of the small mass differences of the receptor frag-
ments in the obtained mixture. To facilitate the detection as
well as the isolation of the photocrosslinked products, we
envisaged a strategy that would allow for the introduction
of a biotin tag into our photoprobes.[32] An incorporated
biotin tag would allow for both the determination of the
concentration of the insertion products by avidin-biotin af-
finity chromatography and the detection of the products
through the combination of SDS gel electrophoresis and
Western blotting technique.[33–35]
A bifunctional mannosyl peptide was selected as the
scaffold for the synthesis of a biotinylated photoprobe. Am-
inoethyl mannoside 15 was coupled with an orthogonally
protected lysine derivative in a HATU-mediated reaction
Scheme 3. a) 0.3 equiv. TMSOTf, dry CH2Cl2, argon, room temp.,
69% for 11 (56 h), 69% for 13 (5 d); b) 1  NaOMe in dry MeOH,
argon, room temp., 89% for 12, 88% for 14.
The mannosylation reactions which afford 11 and 13 re- to afford bifunctional glycopeptide 17, which carries two
quired modified reaction conditions because of the lower orthogonally protected amino groups in the aglycon moiety
reactivity of corresponding benzyl alcohols 7 and 9. The (Scheme 4). Diamine 17 formed the starting material for the
reactivity of 7 turned out to be even lower than that of 9. introduction of the photoactive group as well as the biotin
To obtain complete conversion of both acceptor alcohols 7 tag. In situ removal of the N-tert-butyloxycarbonyl protect-
and 9, reaction temperature, reaction time, and the amount ing group followed by its peptide coupling with (+)-biotin
of catalyst had to be increased.
led to tagged glycopeptide 18 in 69% yield after purification
Eur. J. Org. Chem. 2006, 4841–4851
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4843

M. Wiegand, T. K. Lindhorst
FULL PAPER
Scheme 4. a) 4-Benzoylbenzoic acid, HATU, DIPEA, dry DMF, argon, room temp., 52%; b) Boc-Lys-Fmoc-OH, HATU, DIPEA, dry
DMF, argon, room temp., 87% (crude); c) 80% TFA, room temp.; d) (+)-biotin, HATU, DIPEA, dry DMF, argon, room temp., 69%
(two steps); e) 20% piperidine, DMF; f) 4-benzoylbenzoic acid, HATU, DIPEA, dry DMF, argon, room temp., 40% from 15.
on silica gel. The α-amino function of lysine was then de- quired to transfer the photolabile residues into their respec-
protected with 20% piperidine in DMF and the subsequent tive reactive intermediates (Table 1). All tested photolabile
introduction of 4-benzoylbenzoic acid by standard peptide compounds exhibit absorption maxima at wavelengths
coupling techniques in DMF led to the desired bifunctional Ͼ320 nm; thus, they all qualify as photoprobes for the in-
photoactive mannoside. Purification was accomplished by vestigation of biomacromolecules by photoaffinity labeling
two consecutive flash chromatographic separations on silica without decomposition of the target receptors.
gel to provide pure 19 in 40% overall yield from 15.
Table 1. UV spectroscopic data for the synthesized photolabels 12,
14, 16, and 19.
Irradiation Experiments
Compound (PAL)
Solvent
λmax
[nm]
Disappearance of λmax
[min][a]
Photoaffinity labeling photophores are incubated with a
target peptide or protein in the dark and then irradiated to
be crosslinked with their receptors. The resulting covalent
adduct can be analyzed with MS techniques such as
MALDI-TOF. Synthesized photolabeled mannosides 12,
14, and 16 and 19 should form highly reactive intermediates
such as nitrenes, carbenes, and radical ions, respectively,
upon irradiation with UV light. These are able to insert
into C–H, N–H or O–H bonds of biomacromolecules, the
protein FimH for instance. To check this ability we per-
formed a series of different photolysis experiments.
12 (azide)
14 (diazirine)
16 (benzophenone) doubly distilled H2O 330.0
19 (benzophenone) doubly distilled H2O 327.8
MeOH[b]
doubly distilled H2O 355.5
343.3
30
10
[c]
[c]
[a] After irradiation at wavelengths Ն320 nm; upon irradiation, UV
spectra were recorded in 5 min intervals. [b] Azide 12 is only poorly
soluble in water. [c] No decrease in λmax because of the relaxation
of the excited triplet radical to the ground state.
MS analyses of the irradiated solutions showed the for-
For all PALs, absorption maxima (λmax) were deter- mation of insertion products with the solvent in case of 12
mined, and upon irradiation a decrease in λmax was ob- and 14. Benzophenone-labeled derivatives 16 and 19 did not
served with time. Therefore, stock solutions of photoprobes show insertion products of any sort, which indicates their
12, 14, 16, and 19 were prepared in doubly distilled water lower reactivity. This is due to the relaxation of the excited
or MeOH. Because of its poor solubility in water, arylazide triplet diradical that is initially formed during irradia-
12 had to be dissolved in MeOH. To monitor the decrease tion.[14a]
in λmax, each sample was irradiated with UV light Ն320 nm
A number of amino acids were then selected for photola-
up to 60 min and UV spectra were recorded in 5 min inter- beling with PALs 12, 14, 16, and 19. Table 2 summarizes
vals to obtain information about the irradiation time re- the details of the labeling experiments with six protected
4844
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4841–4851

Photoactive α-Mannosides and Mannosyl Peptides and Evaluation for Lectin Labeling
FULL PAPER
amino acid derivatives derived from threonine, serine, tyro- irradiation. This assumption is supported by the fact that
sine, isoleucine, lysine, and arginine. Irradiation of pho- 16 and 19 do not form insertion products with water
toprobes 14, 16, and 19 with all amino acids was carried out (Table 1) or polar amino acids such as threonine or serine
in doubly distilled water, whereas experiments with azide (Table 2). Furthermore, all investigated PALs seem to be
12 as the photophore required MeOH as the solvent. The able to form insertion products, as irradiation with certain
concentration of each PAL solution was chosen as 0.1 m side-chain-functionalized amino acids led to crosslinked
and irradiation experiments were carried out with four dif- products that were detected by mass spectrometric methods.
ferent amino acid concentrations (0.05 m, 0.2 m, 0.5 m,
and 1.0 m) for each amino acid.
We found that insertion products of all the PALs investi-
gated were formed with one of the amino acids, whereas
MS analyses of the irradiated PAL-amino acid samples their exact structure remained unknown. However, at this
revealed if an insertion product had been formed, as well point we were not interested in the structural analysis of
as the mass of the formed insertion product. However, nei- the formed insertion products, but instead we wanted to
ther the chemoselectivity nor the regioselectivity of the pho- know if our photoactive mannosides, which were successful
tolabeling reaction could be deduced from the MS data. with amino acids would also work with peptides. Therefore,
For example, it cannot be determined whether the carbene two octapeptides and pentaglycine were selected as model
derived from diazirine 14 inserted into the O–H bond or a peptides and irradiated. Angiotensin II (H-Asp-Arg-Val-
C–H single bond of threonine. According to the litera- Tyr-Ile-His-Pro-Phe-OH) and PTHIKWGD (H-Pro-Thr-
ture,[11,12] carbenes tend to insert into polar functionalities His-Ile-Lys-Trp-Gly-Asp-OH) were chosen to reflect a
such as hydroxy groups or amino groups rather than into broad variety of amino acid side chains, whereas pentagly-
nonpolar C–H bonds. On the other hand, benzophenone cine provided only unreactive C–H-bonds for insertion.
can be assumed to produce C–H insertion products upon Sample mixtures containing 10 equiv. of peptide (1.0 m)
Table 2. Results of the photolabeling experiments of various protected amino acids with PALs 12, 14, 16, and 19 in solution as detected
by mass spectrometry.
[a] Limited solubility of amino acid component. [b] Decarboxylation of the amino acid was observed. [c] Atmospheric oxidation of the
biotin sulfur during photolysis formed the respective biotin sulfone.
Eur. J. Org. Chem. 2006, 4841–4851
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4845

M. Wiegand, T. K. Lindhorst
Table 3. Monoisotopic masses of the insertion products formed from the reaction of PALs with three different model peptides.
FULL PAPER
Peptide
Crosslinked products
Crosslinked products
Crosslinked products
Crosslinked products
with 12
with 14
with 16
with 19
Angiotensin II
(M = 1046.54)
PTHIKWGD
(M = 953.07)
Pentaglycine
m/z = 1415.490[a]
m/z = 1308.689
–
m/z = 1396.641[a]
m/z = 1326.588
–
no insertion
no insertion
m/z = 733.848
m/z = 1848.192[b]
no insertion
(M = 303.27)
no insertion
[a] Further analysis by fragmentation with the use of ESI-FT-ICR-MS–MS. [b] Atmospheric oxidation of the biotin sulfur atom during
photolysis formed the respective biotin sulfone.
and one equiv. of photolabel (0.1 m) in doubly distilled
water were prepared and irradiated with UV light Ն320 nm
Conclusions
for 10 min (diazirine 14), 30 min (azide 12), and 60 min
With the aim to investigate carbohydrate binding to the
(benzophenones 16 and 19). Formed insertion products mannose-specific lectin FimH, we have synthesized manno-
were concentrated by “ZipTipping”®[36] and analyzed with sides 12, 14, and 16 equipped with three different photo-
MALDI-TOF-MS and ESI-MS. MS analyses of the irradi- active functional groups: a diazirine, an azide, and a benzo-
ation experiments with those model peptides showed phenone moiety, respectively. Furthermore, to assist the
crosslinked products in the case of angiotensin II with 12, photoaffinity technique by biolabeling, we have synthesized
14, and 19, and in case of PTHIKWGD with 12 and 14. biotin derivative 19. We then compared the potential of
Photoprobe 16 showed no reaction with these two peptides these three photophores in photoaffinity labeling.
but was crosslinked with pentaglycine (Table 3).
All of the synthesized photoprobes exhibit absorption
Diazirine 14 and azide 12 were fragmented in an ESI- maxima at wavelengths Ͼ320 nm and can therefore be used
FT-ICR-MS–MS experiment to further characterize the in the investigation of biomacromolecules such as peptides
crosslinked products formed upon their reaction with irra- and proteins without degradation. During their photolysis
diated angiotensin II. In particular, we were interested to in water or MeOH, photophores 12 and 14 formed covalent
learn if different photoprobes attack different positions of insertion products. Further studies that include illumination
the investigated model peptides, and to determine whether experiments in the presence of six different amino acids as
they show preferences for certain functional groups. MS– well as the three model peptides, angiotensin II,
MS analysis of 12-angiotensin II, the insertion product re- PTHIKWGD, and pentaglycine, allowed further conclu-
sulting from the irradiation of azide 12 with the peptide, sions about the features of the investigated PALs. To obtain
and 14-angiotensin II, the insertion product resulting from the highest amount of photocrosslinked product it was de-
the irradiation of diazirine 14 and the peptide, provided termined that the optimal ratio of photolabel to target
fragmentation patterns that are compatible with the corre- molecule is 1:10. Although each of the synthesized PALs is
sponding data found in the ProteinProspector Database. able to crosslink to biomolecules, the exact position of the
However, it was obvious that the covalent linkage between formed covalent bond could not be elucidated unequivo-
the respective PAL and the peptide was cleaved before frag- cally.
mentation of the peptide backbone and consequently, no
According to our irradiation experiments, we found that
information about the photolabeling position could be ob- the diazirine group of PAL 14 seemed to be the most suit-
tained from this analysis. We are currently developing this able photoprobe for the photolabeling of peptides. The
technology further to eventually allow for the accurate de- short irradiation time, an absorption maximum Ͼ320 nm,
termination of labeling sites in biomolecules.
its good solubility in water, and its preference for the inser-
As the irradiation of model peptides with 12, 14, and 16 tion into polar groups such as amines and OH groups
delivered a high amount of unlabeled peptide and only a makes the diazirine moiety an advantageous photolabel for
low yield of crosslinked adduct, a method to concentrate studies in biological chemistry. Benzophenones on the other
and purify insertion products was required to facilitate the hand showed some disadvantages, such as steric hindrance,
analytical potential of this technology. Therefore, manno- increased irradiation times, and limited reactivity. However,
side 19 containing the benzophenone photolabel and a bio- its low reactivity, which is due to the relaxation of the ex-
tin tag was employed for the photoaffinity labeling experi- cited state to the ground state, also limits undesired side
ment. After its irradiation with angiotensin II, a biotin- reactions with the solvent for instance. Moreover, the
tagged insertion product could be isolated and concentrated hydrophobic benzophenone photolabel preferentially in-
with a combination of cationic and affinity chromatography serts into apolar C–H bonds.
and an avidin cartridge facilitated MALDI-TOF-MS analy-
Tetrafluorinated aryl azide 12 is less suited for photoaf-
sis of the concentrated crosslinked adduct. Again, mass finity labeling. Because of its poor solubility in water, it is
spectrometric analysis indicated the oxidation of the biotin less useful for the investigation of larger peptides or pro-
sulfur atom to form the respective biotin sulfone.
teins under physiological conditions. An increased tendency
4846
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4841–4851

Photoactive α-Mannosides and Mannosyl Peptides and Evaluation for Lectin Labeling
FULL PAPER
argon atmosphere. p-Azidotetrafluorobenzyl alcohol (9, 0.83 g,
3.90 mmol) was then added under argon, followed by the dropwise
addition of a freshly prepared TMSOTf solution (0.3  in dry
DCM, 3.3 mL) at room temp. The resulting reaction mixture was
kept under an argon atmosphere and stirred for 56 h in the dark.
Evaporation of the solvent and purification of the crude product by
flash chromatography (toluene/ethyl acetate, 2:1) delivered acetyl-
protected mannoside 11 as a pure solid (1.32 g, 2.40 mmol, 69%).
to unspecific side reactions, which complicate the analysis
of the irradiation products combined with very long irradia-
tion times are central disadvantages of the azide photolabel.
These results form the basis for our future work, which
will be directed towards the optimization of the photoaffin-
ity methodology, and the determination of the crosslinking
sites in lectin CRDs. As part of this endeavor, we have syn-
thesized biotin-labeled photoactive mannoside 19, which
has facilitated the isolation and identification of photo-
crosslinked peptides.
R = 0.55 (toluene/ethyl acetate, 2:1). FTIR (KBr): ν = 2975.7,
˜
f
2124.7, 1739.9, 1654.5 cm–1. H NMR (500 MHz, CDCl3, 25 °C):
1
δ = 5.25 (dd, J = 9.9, 3.0 Hz, 1 H, 4-H), 5.23 (dd, J = 3.7, 3.2 Hz,
1 H, 3-H), 5.15 (dd, J = 3.6, 1.7 Hz, 1 H, 2-H), 4.85 (d, J = 1.7 Hz,
1 H, 1-H), 4.75 (dt, J = 11.0, 1.8 Hz, 1 H, benzyl-CHH,), 4.54 (dt,
J = 11.0, 1.6 Hz, 1 H, benzyl-CHH), 4.23 (dd, J = 12.3, 5.4 Hz, 1
H, 6a-H), 4.03 (dd, J = 12.3, 2.4 Hz, 1 H, 6b-H), 3.92 (ddd, J =
9.9, 5.5, 2.4 Hz, 1 H, 5-H,), 2.15, 2.07, 1.98, 1.92 (4 s, 12 H, 4 CH3,
4 OAc) ppm. 13C NMR (125.75 MHz, MeOD, 25 °C): δ = 170.62,
169.96, 169.82, 169.72 (4 C=O, 4 OAc), 146.60 (d, JC,F = 156.2 Hz,
Experimental Section
General Remarks: Optical rotations were measured with a Perkin–
Elmer polarimeter (22 °C, 589 nm, length of cell = 1 dm). Reac-
tions were monitored by TLC on silica gel GF254 (Merck) with
detection under UV light and by charring with 10% sulfuric acid
in ethanol or with anisaldehyde and subsequent heating. Flash col-
umn chromatography was performed on silica gel 60 (40-63 µm,
Merck) and for RP-MPLC a Merck Licroprep RP-18 column
(Büchi) was used. Preparative HPLC was performed with a Shim-
adzu LC-8a (LiChrosorb RP-8, HIBAR). NMR spectra were re-
corded with Bruker AMX 400, Bruker DRX 500, or Bruker Avance
600 instruments. Chemical shifts are relative to TMS or the solvent
peaks of CDCl3 (δ =7.24 ppm for 1H NMR, 77.0 ppm for 13C
NMR) or MeOD (δ =3.35 ppm and 4.78 ppm for 1H NMR,
49.3 ppm for 13C NMR). Where necessary, assignments were based
on 2D experiments (COSY, HSQC, HMBC, or NOESY). IR spec-
tra were taken with a Perkin–Elmer FT IR Paragon 1000. MALDI-
TOF mass spectra were measured with a Bruker Biflex III with a
19 kV acceleration voltage. 4-Hydroxy-α-cyanocinnamic acid
(HCCA) was used as the matrix, either as a saturated solution in
a solvent mixture (33% MeCN/doubly distilled water and 0.1%
TFA) or as a saturated solution in acetone. Ionization was effected
with a nitrogen laser at 337 nm. ESI-MS spectra were measured
with an Applied Biosystems Mariner ESI-TOF 5280. ESI-FT-ICR-
MS/MS spectra were measured with a Bruker 7 Tesla Apex II mass
spectrometer at the Research Center Borstel. Fragmentation pat-
terns and sequence analysis were determined with ProteinProspec-
tor as the protein sequence database. Millipore C18-pipette tips
were used for ZipTipping® and for cationic and affinity chromatog-
raphy ICAT®-kits from Applied Biosystems were used.
aryl-C-F), 144.62 (d, JC,F = 164.2 Hz, aryl-C-F), 141.41 (d, JC,F
=
106.7 Hz, aryl-C-F), 139.42 (d, JC,F = 115.4 Hz, aryl-C-F), 120.96
(t, JC,F = 84.3 Hz, aryl-C-N3), 110.07 (t, JC,F = 126.1 Hz, aryl-C-
CH2,), 97.39 (C-1), 69.31 (C-2), 69.09 (C-5), 68.78 (C-3), 65.95 (C-
4), 62.35 (C-6), 56.61 (benzyl-C), 20.82, 20.65, 20.60, 20.42 (4 CH3,
4 OAc) ppm. MALDI-TOF-MS: m/z = 574.3 [M + Na]+, 590.3
[M + K]+. ESI-HRMS: calcd. for C21H21F4N3NaO10 [M + Na]+
574.1055; found 574.1039.
p-Azidotetrafluorobenzyl α-D-Mannopyranoside (12): Mannoside 11
(600 mg, 1.10 mmol) was dissolved in dry methanol (12 mL), and
a freshly prepared mixture of sodium methoxide in dry methanol
(1 , 0.5 mL) was carefully added. The resulting mixture was
stirred at room temp. for 4 h, and then acidic ion exchange resin
(Amberlite IR-120) was added in portions until pH 6 was reached.
The mixture was filtered, and the filtrate evaporated in vacuo. Puri-
fication of the crude product by flash chromatography with RP-
silica gel (ethyl acetate/methanol/water, 6:2:1) delivered the title
compound (371 mg, 0.97 mmol, 89%) as light orange crystals. Rf
= 0.66 (ethyl acetate/methanol/water, 6:2:1). M.p. 159 °C. [α]D = +
36.5 (c = 2.0, MeOH). UV: λmax(1) = 343.3 nm (c = 0.10 mM,
MeOH), ε(1) = 1500 Lmol–1 cm–1, λmax(2) = 250.6 nm (c =
0.10 mM, MeOH), ε(1) = 16300 Lmol–1 cm–1. FTIR (KBr): ν =
˜
3464.6, 3417.4, 3370.2, 3275.6, 2925.0, 2896.2, 2120.0, 1489.7,
1240.8, 1067.2 cm–1. 1H NMR (500 MHz, MeOD, 25 °C): δ = 4.92
(d, J = 1.6 Hz, 1 H, 1-H), 4.89 (dt, J = 11.6, 1.8 Hz, 1 H, benzyl-
CHH), 4.70 (dt, J = 11.6, 1.7 Hz, 1 H, benzyl-CHH), 3.82 (dd, J
= 11.8, 2.4 Hz, 1 H, 6a-H), 3.81 (dd, J = 3.0, 1.7 Hz, 1 H, 2-H),
3.77 (dd, J = 11.8, 5.5 Hz, 1 H, 6b-H), 3.67 (dd, J = 9.8, 2.7 Hz, 1
H, 4-H), 3.66 (dd, J = 2.9, 2.6 Hz, 1 H, 3-H), 3.56 (ddd, J = 9.8,
5.5, 2.3 Hz, 1 H, 5-H) ppm. 13C NMR (125.75 MHz, MeOD,
For assignment of the NMR spectroscopic data 18 and 19 were
numbered according to the following drawing:
25 °C): δ = 147.90 (d, JC,F = 179.5 Hz, aryl-C-F), 145.93 (d, JC,F
=
212.0 Hz, aryl-C-F), 142.86 (d, JC,F = 179.0 Hz, aryl-C-F), 140.89
(d, JC,F = 227.5 Hz, aryl-C-F), 121.78 (t, JC,F = 222.7 Hz, aryl-C-
N3), 112.91 (t, JC,F = 130.0 Hz, aryl-C-CH2), 101.96 (C-1), 75.09
(C-5), 72.04 (C-4), 71.99 (C-2), 68.40 (C-3), 62.79 (benzyl-CH2),
57.50 (C-6) ppm. MALDI-TOF-MS: m/z = 406.3 [M + Na]+, 421.3
[M + K]+. ESI-MS: m/z = 406.06 [M + Na]+, 789.13 [2M + Na]+.
ESI-HRMS: calcd. for C13H13F4N3NaO6 [M + Na]+ 406.0633;
found 406.0678. C13H13F4N3O6 (383.25): calcd. C 40.72, H 3.42,
N 10.97; found C 39.78, H 3.02, N 11.37.
When helpful, the mannosyl substructure is abbreviated as “man”
and in complicated cases lengthy IUPAC names were substituted
by simplified terms.
3-Trifluoromethyl-3-[p-(2Ј,3Ј,4Ј,6Ј-tetra-O-acetyl-α-
pyranosyloxymethyl)phenyl]diazirine (13): Acetyl-protected manno-
D-Manno- syl trichloroacetimidate 10 (0.62 g, 1.26 mmol) was dried in vacuo
D-manno-
p-Azidotetrafluorobenzyl-2,3,4,6-tetra-O-acetyl α-
pyranoside (11): Acetyl-protected mannosyl trichloroacetimidate 10
for 20 min prior to use and dissolved in dry DCM (15 mL) under
(1.70 g, 3.50 mmol) was dissolved in dry DCM (20 mL) under an
an argon atmosphere. Diazirine derivative 7 (0.32 g, 1.50 mmol)
Eur. J. Org. Chem. 2006, 4841–4851
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4847

M. Wiegand, T. K. Lindhorst
FULL PAPER
was then added, followed by the dropwise addition of a freshly
prepared TMSOTf solution (0.1  in dry DCM, 4.5 mL) under a
flow of argon. The reaction mixture was stirred in the dark at room
temp. for 5 d. Evaporation of the solvent and purification of the
crude product by flash chromatography (toluene/ethyl acetate, 2:1)
delivered the title compound as a yellowish powder (566 mg,
6:2:1) followed by gradient RP-HPLC to obtain the product
(201 mg, 52%) as a light yellow lyophilisate. Rf = 0.59 (ethyl ace-
tate/methanol/water, 6:2:1). HPLC: retention time = 40.89 min,
gradient = MeCN/H2O, 38:62. [α]D = + 29.0 (c = 2, H2O). UV:
λmax(1) = 330.0 nm (c = 0.25 mM, doubly distilled H2O), ε(1) =
266.7 Lmol–1 cm–1, λmax(2) = 258.1 nm (c = 0.25 mM, doubly dis-
tilled H2O), ε(2) = 14000 Lmol–1 cm–1. 1H NMR (600 MHz,
1
1.04 mmol, 69%). Rf = 0.62 (toluene/ethyl acetate, 2:1). H NMR
(300 MHz, CDCl3, 25 °C): δ = 7.31 (dd, J = 8.6, 2.2 Hz, 2 H, H- MeOD, 25 °C): δ = 8.00 (dd, J = 8.4, 1.8 Hz, 2 H, benzophenone-
aryl-H), 7.14 (dd, J = 8.0, 2.3 Hz, 2 H, aryl-H), 5.30 (dd, J = 9.9,
3.3 Hz, 1 H, 3Ј-H), 5.24 (dd, J = 9.7, 9.7 Hz, 1 H, 4Ј-H), 5.22 (dd,
m-H), 7.87 (dd, J = 8.1, 1.8 Hz, 2 H, benzophenone-o-H), 7.83 (dd,
J = 7.8, 1.2 Hz, 2 H, benzophenone-o-H), 7.71 (dd, J = 7.4, 1.2 Hz,
J = 3.2, 1.8 Hz, 1 H, 2Ј-H), 4.80 (d, J = 1.6 Hz, 1 H, 1Ј-H), 4.66 1 H, benzophenone-p-H), 7.59 (td, J = 7.6, 1.6 Hz, 2 H, benzophe-
(d, J = 12.3 Hz, 1 H, benzyl-CHH), 4.50 (d, J = 12.3 Hz, 1 H,
benzyl-CHH), 4.21 (dd, J = 12.3, 5.2 Hz, 1 H, 6Јa-H), 4.00 (dd, J
= 12.3, 2.5 Hz, 1 H, 6Јb-H-), 3.91 (ddd, J = 9.5, 5.1, 2.5 Hz, 1
H, 5Ј-H), 2.13, 2.07, 1.89, 1.85 (4 CH3, 4 OAc) ppm. 13C NMR
(75.45 MHz, CDCl3, 25 °C): δ = 170.67, 170.07, 169.96, 169.75 (4
none-m-H), 4.86 (d, J = 1.6 Hz, 1 H, 1-H), 3.93 (dd, J = 10.3,
4.8 Hz, 1 H, manOCH2CHH), 3.87 (dd, J = 3.1, 1.7 Hz, 1 H, 2-
H), 3.86 (dd, J = 12.0, 2.1 Hz, 1 H, 6a-H), 3.76 (dd, J = 6.2, 3.5 Hz,
1 H, 3-H), 3.75 (dd, J = 10.9, 4.6 Hz, 1 H, manOCH2CHH), 3.73
(dd, J = 11.7, 5.8 Hz, 1 H, 6b-H), 3.71 (dd, J = 8.8, 7.3 Hz, 1 H,
C=O, 4 OAc), 163.45 (aryl-C-N=N), 137.95 (aryl-C-CH2), 129.05 manOCHHCH2), 3.66 (dd, J = 9.1, 7.3 Hz, 1 H, manOCHHCH2),
(aryl-C), 128.30 (2 aryl-C), 126.74 (2 aryl-C), 122.20 (q, JC,F 3.65 (dd, J = 9.2, 6.9 Hz, 1 H, 4-H), 3.62 (ddd, J = 9.5, 5.9, 2.2 Hz,
274.7 Hz, CF3), 96.79 (C-1Ј), 77.22 (C-5Ј), 69.43 (C-3Ј), 69.02 (C- 1 H, 5-H) ppm. 13C NMR (150.90 MHz, MeOD, 25 °C): δ = 197.73
=
2Ј), 68.81 (benzyl-CH2), 66.04 (C-4Ј), 62.41 (C-6Ј), 20.91, 20.78,
20.73, 20.72 (4 CH3, 4 OAc) ppm. MALDI-TOF-MS: m/z = 519.4
[M – N2 + H]+. ESI-HRMS: calcd. for C23H25F3N2NaO10 [M +
Na]+ 569.1525; found 569.1446.
(aryl-C=O), 169.46 (benzophenone-C=O), 141.37 (aryl-C),139.24
(aryl-C) 138.42 (aryl-C), 134.15 (aryl-C), 131.06 (2 aryl-C), 130.97
(2 aryl-C), 129.66 (2 aryl-C), 128.49 (2 aryl-C), 101.86 (C-1), 74.87
(C-5), 72.60 (C-3), 72.15 (C-2), 68.64 (C-4), 67.26 (manOCH2CH2),
62.93 (C-6), 41.07 (manOCH2CH2) ppm. MALDI-TOF-MS: m/z
= 454.3 [M + Na]+. ESI-MS: m/z = 454.12 [M + Na]+. ESI-HRMS:
calcd. for C22H25NNaO8 [M + Na]+ 454.1472; found 454.1452.
3-Trifluoromethyl-3-[p-(α-D-mannopyranosyloxymethyl)phenyl]dia-
zirine (14): Photolabile acetyl-protected mannoside 13 (297 mg,
0.50 mmol) was dissolved in dry methanol (14 mL) and a freshly
prepared solution of sodium methoxide in dry methanol (2 mM,
1.1 mL) was added. The resulting mixture was stirred at room
temp. for 1 h, acidic ion exchange resin (Amberlite IR-120) was
then added in small portions until a pH 6.5 was reached. The mix-
ture was filtered and the filtrate evaporated in vacuo. The product
was purified by two consecutive flash column chromatography sep-
arations on RP-silica gel (ethyl acetate/methanol/water, 6:2:1) to
yield pure photolabile mannoside 14 (166 mg, 0.44 mmol, 88%) as
a light yellow powder. Rf = 0.68 (ethyl acetate/methanol/water,
6:2:1). [α]D = + 44.5 (c = 1.9, H2O). UV: λmax(1) = 355.5 nm (c =
0.1 mM, doubly distilled H2O), ε(1) = 1500 Lmol–1 cm–1, λmax(2) =
265.5 nm (c = 0.1 mM, doubly distilled H2O), ε(2) = 5200
Lysine Derivative 17: 2Ј-Aminoethyl α--mannoside (15) (375 mg,
1.65 mmol), HATU (590 mg, 1.65 mmol), and Fmoc-Lys(Boc)-OH
(630 mg, 1.35 mmol) were mixed in a Schlenk flask and dried in
vacuo for 15 min. The solid mixture was then dissolved in dry
DMF (16 mL) and DIPEA (0.53 mL, 2.70 mmol) was carefully
added. This reaction mixture was stirred under an argon atmo-
sphere at room temp. for 18 h, and the solvent was evaporated.
Purification of the crude product was accomplished by flash
chromatography (ethyl acetate/methanol/water, 6:2:1) to yield the
desired product as colorless powder (980 mg, 87%) with Rf = 0.61
(ethyl acetate/methanol/water, 6:2:1), which was directly used in the
next step without further characterization.
Lmol–1 cm–1. FTIR (KBr): ν = 3431.8, 2929.3, 1663.7, 1342.0,
˜
Biotinylated Lysine Derivative 18: Compound 17 (550 mg,
0.82 mmol) was dissolved in trifluoroacetic acid (80% aq., 20 mL)
and stirred at room temp. for 30 min. The solvent was evaporated,
and the resulting crude product was then coevaporated with tolu-
ene (6×10 mL), and neutralized with basic ion exchange resin
(Amberlite IRA-402). The desired deprotected product (Rf = 0.19;
ethyl acetate/methanol/water, 6:2:1) was obtained as a colorless
powder. (+)-Biotin (208 mg, 0.85 mmol) and HATU (316 mg,
0.81 mmol) were added to the crude amine (468 mg, 0.81 mmol),
and the solid mixture was dried in vacuo for 30 min, and then dis-
solved in dry DMF (20 mL). DIPEA (0.45 mL, 2.25 mmol) was
carefully added and the resulting reaction mixture was stirred un-
der an argon atmosphere at room temp. for 18 h. The solvent was
evaporated, and the crude product was purified by flash
chromatography (ethyl acetate/methanol/water, 6:2:1). After lyophi-
lization, compound 18 resulted as a colorless, fluffy powder
(496 mg, 69% over two reaction steps). Rf = 0.40 (ethyl acetate/
1
1154.1, 831.9 cm–1. H NMR (500 MHz, MeOD, 25 °C): δ = 7.52
(d, J = 8.5 Hz, 2 H, aryl-H), 7.28 (d, J = 8.1 Hz, 2 H, aryl-H), 4.87
(d, J = 1.7 Hz, 1 H, 1-H), 4.83 (d, J = 12.4 Hz, 1 H, benzyl-CHH),
4.61 (d, J = 12.4 Hz, 1 H, benzyl-CHH), 3.89 (dd, J = 3.4, 1.8 Hz,1
H, 2-H), 3.87 (dd, J = 11.8, 2.3 Hz, 1 H, 6b-H), 3.77 (dd, J = 9.4,
3.4 Hz, 1 H, 3-H), 3.75 (dd, J = 11.8, 5.9 Hz, 1 H, 6a-H), 3.67 (t,
J = 9.5 Hz, 1 H, 4-H), 3.60 (ddd, J = 9.5, 6.0, 2.3 Hz, 1 H, 5-H)
ppm. 13C NMR (125.75 MHz, MeOD, 25 °C): δ = 166.26 (aryl-
C-N=N), 141.54 (aryl-C-CH2), 129.53 (2 aryl-C), 129.24 (aryl-C),
127.64 (2 aryl-C), 123.63 (q, JC,F = 273.8 Hz, CF3), 100.94 (C-1),
75.03 (C-5), 72.60 (C-3), 72.11 (C-2), 69.02 (benzyl-CH2), 68.61 (C-
4), 62.89 (C-6) ppm. ESI-MS: m/z
= 473.18 [M – N2 +
2 MeOH]+, 401.08 [M + Na]+, 373.08 [M – N2 + Na]+. ESI-
HRMS: calcd. for C15H17F3N2NaO6 [M + Na]+ 401.0931; found
401.0841.
Benzophenone Derivative 16: 2Ј-Aminoethyl α--mannoside 15
(200 mg, 0.90 mmol), HATU (330 mg, 0.90 mmol), and 4-benzo- methanol/water, 6:2:1). 1H NMR (500 MHz, MeOD, 25 °C): δ =
ylbenzoic acid (225 mg, 1.00 mmol) were mixed in a Schlenk flask
and dried in vacuo for 30 min. The solid mixture was then dissolved
7.84 (d, J = 7.5 Hz, 2 H, Fmoc-H), 7.72 (dd, J = 10.4, 7.7 Hz, 2
H, Fmoc-H), 7.43 (t, J = 7.4 Hz, 2 H, Fmoc-H), 7.34 (td, J = 7.5,
in dry DMF (12 mL) and DIPEA (0.34 mL, 2.00 mmol) was care- 1.2 Hz, 2 H, Fmoc-H), 4.81 (d, J = 1.4 Hz, 1 H, 1-H), 4.44 (dd, J
fully added. The mixture turned yellow, and was stirred at room
temp. for 17 h under an argon atmosphere. The solvent was then
evaporated, and the resulting crude product was purified by flash
chromatography on RP-silica gel (ethyl acetate/methanol/water,
= 9.0, 5.7 Hz, 1 H, Fmoc-CHCH2), 4.38 (ddd, J = 7.9, 4.2, 0.8 Hz,
1 H, biotin-NHCHCHalkyl), 4.18 (dd, J = 7.9, 4.5 Hz, 1 H, biotin-
NHCHCH2S), 3.72 (dd, J = 7.2, 3.9 Hz, 1 H, 6a-H), 3.71 (dd, J =
2.8, 1.4 Hz, 1 H, 2-H), 3.68 (dd, J = 6.8, 4.1 Hz, 1 H, 6b-H), 3.62
4848
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4841–4851

Photoactive α-Mannosides and Mannosyl Peptides and Evaluation for Lectin Labeling
FULL PAPER
(dd, J = 8.9, 3.4 Hz, 1 H, 3-H), 3.57 (dd, J = 11.7, 5.7 Hz, 1 H,
manOCHHCH2), 3.50 (t, J = 9.2 Hz, 1 H, 4-H), 3.47 (dd, J = 11.5,
5.1 Hz, 1 H, manOCH2CHH), 3.45 (dd, J = 11.6, 4.7 Hz, 1 H,
3.5 Hz, 1 H, 16a-H), 1.81 (dd, J = 9.8, 5.1 Hz, 1 H, 19b-H), 1.78
(dd, J = 6.2, 3.3 Hz, 1 H, 16b-H), 1.76 (dd, J = 9.3, 4.5 Hz, 1 H,
18a-H), 1.74 (dd, J = 9.2, 4.5 Hz, 1 H, 13a-H), 1.73 (dd, J = 10.1,
4.6 Hz, 1 H, 19a-H), 1.61 (dd, J = 14.1, 7.6 Hz, 1 H, 17a-H), 1.59
manOCH2CHH), 3.40 (dd,
J = 11.6, 5.2 Hz, 1 H, man-
OCHHCH2), 3.31 (dd, J = 6.4, 4.3 Hz, 1 H, 12a-H), 3.30 (dd, J = (dd, J = 9.8, 8.0 Hz, 1 H, 11a-H), 1.57 (dd, J = 13.0, 5.3 Hz, 1 H,
7.7, 1.1 Hz, 1 H, biotin-CHH), 3.29 (ddd, J = 10.5, 6.1, 4.0 Hz, 1
H, 5-H), 3.28 (dd, J = 6.3, 4.3 Hz, 1 H, 12b-H), 3.26 (dd, J = 4.5,
1.0 Hz, 1 H, biotin-CHH), 3.17 (ddd, J = 10.3, 7.9, 4.5 Hz, 1 H,
biotin-NHCHCHalkyl), 3.11 (dd, J = 9.0, 6.6 Hz, 1 H, 13a-H),
3.09 (ddd, J = 9,0, 7.0, 2.1 Hz, 1 H, 10-H), 3.06 (dd, J = 9.3, 6.0 Hz,
1 H, 13b-H,), 2.80 (dd, J = 12.8, 5.0 Hz, 1 H, Fmoc-CHCHH),
biotin-NHCHCHHS), 1.55 (dd, J = 10.0, 7.7 Hz, 1 H, 11b-H), 1.53
(dd, J = 14.9, 7.4 Hz, 1 H, 12a-H), 1.51 (dd, J = 9.3, 6.9 Hz, 1 H,
13b-H), 1.47 (dd, J = 13.8, 7.8 Hz, 1 H, 17b-H), 1.36 (dd, J = 9.7,
6.5 Hz, 1 H, 18b-H), 1.31 (dd, J = 15.2, 7.6 Hz, 1 H, 12b-H) ppm.
13C NMR (125.75 MHz, MeOD, 25 °C): δ = 197.70 (aryl-C=O),
176.07 (C-15), 174.59 (C-9), 169.37 (benzophenone-C=O) 166.09
2.59 (d, J = 12.7 Hz, 1 H, Fmoc-CHCHH), 2.08 (t, J = 7.4 Hz, 2 [(NH)2-C=O], 141.51 (2 aryl-C), 138.73 (2 aryl-C), 138.37 (aryl-C),
H, 16a/b-H), 1.85 (dd, J = 10.2, 5.8 Hz, 1 H, 14a-H), 1.82 (dd, J
= 10.0, 5.7 Hz, 1 H, 14b-H), 1.78 (dd, J = 7.9, 3.7 Hz, 1 H, 17a-
134.19 (aryl-C), 131.09 (2 aryl-C), 130.95 (2 aryl-C), 129.67 (2 aryl-
C), 128.78 (2 aryl-C), 101.69 (C-1), 74.75 (C-5), 72.60 (C-2), 72.11
H), 1.76 (dd, J = 7.4, 3.9 Hz, 1 H, 17b-H), 1.61 (dd, J = 7.7, 6.1 Hz, (C-3), 68.78 (C-4), 67.08 (C-6), 63.38 (biotin-NHCHCHalkyl),
1 H, 19a-H), 1.59 (dd, J = 7.5, 3.2 Hz, 1 H, 19b-H), 1.53 (dd, J = 62.98 (C-14), 61.63 (biotin-NHCHCH2), 57.02 (biotin-
15.5, 7.9 Hz, 1 H, 18a-H), 1.46 (dd, J = 9.0, 7.2 Hz, 1 H, 11a-H), NHCHCHalkyl), 55.72 (C-10), 41.05 (manOCH2CH2), 40.42
1.43 (dd, J = 7.1, 2.5 Hz, 1 H, 11b-H), 1.39 (dd, J = 15.7, 7.5 Hz,
(manOCH2CH2), 39.99 (C-16), 35.79 (C-19), 32.74 (C-11), 30.12
1 H, 18b-H) ppm. 13C NMR (125.75 MHz, MeOD, 25 °C): δ = (C-13), 29.76 (biotin-CH2), 29.45 (C-17), 26.86 (C-18), 24.52 (C-
176.03 (C-9), 175.09 (C-15), 166.10 [(NH)2-C=O], 158.47 (Fmoc- 12) ppm. UV: λmax(1) = 327.8 nm (c = 0.75 mM, doubly distilled
C=O), 151.37 (2 Fmoc-C), 148.57 (2 Fmoc-C), 128.84 (2 Fmoc-C),
H2O), ε(1) = 1013.3 Lmol–1 cm–1, λmax(2) = 290.0 nm (c =
128.21 (2 Fmoc-C), 126.26 (2 Fmoc-C), 120.96 (2 Fmoc-C), 101.70 0.75 mM, doubly distilled H2O), ε(2) = 3613.3 Lmol–1 cm–1, λmax(3)
(C-1), 74.77 (C-5), 72.60 (C-3), 72.11 (C-2), 68.74 (C-4), 67.94 (C- = 242.8 nm (c = 0.75 mM, doubly distilled H2O), ε(2) = 4213.3
6), 67.11 (manOCH2CH2), 63.38 (biotin-NHCHCH2), 62.98 (man-
OCH2CH2), 61.64 (biotin-NHCHCHalkyl), 57.01 (C-10), 56.76
(biotin-NHCHCHalkyl), 55.57 (Fmoc-CH), 41.04 (biotin-CH2),
40.32 (C-14), 36.79 (C-12), 33.00 (C-13), 30.02 (C-16), 29.75 (C-
17), 29.47 (C-18), 26.86 (C-19), 24.35 (C-11) ppm. MALDI-TOF-
MS: m/z = 822.9 [M + Na]+, 838.9 [M + K]+. ESI-HRMS: calcd.
for C39H53N5NaO11S [M + Na]+ 822.92; found 822.34.
Lmol–1 cm–1. MALDI-TOF-MS: m/z = 808.7 [M + Na]+, 824.7 [M
+K]+. ESI-MS: m/z = 808.32 [M + Na]+. ESI-HRMS: 808.3184 [M
+ Na]+.
Irradiation Experiments
Evaluation of the Decomposition Time of the Photolabile Residues
and Detection of Insertion Products: The experimental setup for the
irradiation tests consisted of a UV lamp (Peschl-Consulting Mainz,
150 Watt) and a metal box to exclude external light. A conical glass
vial equipped with a septum to assure an inert gas atmosphere was
used as the reactor. The distance between the light source and the
sample was kept constant at 12 cm. To obtain the optimal wave-
length range appropriate glass filters (Schott AG) were attached
between the light source and the sample. Sample volumes for pho-
tolysis were 5.0 mL and 0.5 mL (Table 1).
Benzophenone-Substituted Biotinylated Lysine Derivative 19: Com-
pound 18 (279 mg, 0.35 mmol) was dissolved in dry DMF (8 mL)
and piperidine (2 mL), and the resulting solution was stirred under
an argon atmosphere at room temp. for 30 min. The mixture was
then concentrated, and the crude product (280 mg, 0.35 mmol) was
dissolved in DMF (12 mL) and added to a mixture of 4-benzo-
ylbenzoic acid (68.0 mg, 0.30 mmol) and HATU (130 mg,
0.35 mmol), which had been dried in vacuo for 30 min. To the re-
sulting solution, DIPEA (0.7 mL, 0.70 mmol) was carefully added
and the mixture was stirred at room temp. under an argon atmo-
sphere for 20 h. The solvent was evaporated and the crude product
was purified by two subsequent flash chromatographic separations
on RP-silica gel (ethyl acetate/methanol, 1:1; ethyl acetate/meth-
anol/water, 6:2:1) to obtain photolabile derivative 19 (110 mg, 40%
from 15) as light yellow crystals. Rf = 0.33 (ethyl acetate/methanol/
water, 6:2:1). M.p. 134 °C. [α]D = + 17.8 (c = 2.29, H2O). 1H NMR
(500 MHz, MeOD, 25 °C): δ = 7.92 (ddd, J = 8.6, 1.9, 1.0 Hz, 2
H, benzophenone-m-H), 7.74 (ddd, J = 8.6, 1.9, 1.0 Hz, 2 H,
benzophenone-o-H), 7.70 (dd, J = 8.4, 1.3 Hz, 2 H, benzophenone-
o-H), 7.57 (mc, J = 7.5, 1.5 Hz, 1 H, benzophenone-p-H), 7.45 (mc,
J = 7.6, 1.6, 0.4 Hz, 2 H, benzophenone-m-H), 4.68 (d, J = 1.6 Hz,
1 H, 1-H), 4.44 (dd, J = 9.0, 5.7 Hz, 1 H, 10-H), 4.38 (ddd, J = 7.9,
5.0, 0.8 Hz, 1 H, biotin-NHCHCH2S), 4.18 (dd, J = 7.9, 4.5 Hz, 1
H, biotin-NHCHCHalkyl), 3.72 (dd, J = 7.2, 6.9 Hz, 1 H, 14a-H),
3.71(dd, J = 3.4, 1.6 Hz, 1 H, 2-H), 3.68 (dd, J = 6.4, 4.1 Hz, 1 H,
6a-H), 3.62 (dd, J = 8.9, 3.4 Hz, 1 H, 3-H), 3.57 (dd, J = 13.7,
5.7 Hz, 1 H, manOCHHCH2), 3.50 (t, J = 9.2 Hz, 1 H, 4-H), 3.41
(dd, J = 7.0, 6.7 Hz, 1 H, 14b-H), 3.40 (dd, J = 6.6, 4.2 Hz, 1 H,
6b-H), 3.29 (ddd, J = 9.2, 6.4, 4.3 Hz, 1 H, 5-H), 3.11 (dd, J = 13.8,
7.9 Hz, 1 H, manOCHHCH2), 3.07 (ddd, J = 10.3, 5.9, 4.5 Hz, 1
H, biotin-NHCHCHalkyl), 2.80 (dd, J = 12.8, 5.0 Hz, 1 H, man-
OCH2CHH), 2.58 (d, J = 12.7 Hz, 1 H, manOCH2CHH), 1.83 (dd,
To determine the absorption maxima and the decomposition time,
sample solutions were prepared by the dissolution of the photola-
bile compounds in doubly distilled H2O or MeOH at concentra-
tions of 0.1 m (12 and 14), 0.25 m (16) and 0.75 m (19). Pho-
tolysis experiments were carried out with UV light at wavelengths
Ն320 nm (total sample volume 5.0 mL). UV spectra were recorded
prior to irradiation and to monitor the decrease in the absorption
maximum upon photolysis; UV spectra were recorded every 5 min
during irradiation. The covalent insertion products formed with the
respective solvents were detected by MALDI-TOF-MS or ESI-MS:
Photolysis of Diazirine 14 in MeOH: The insertion product of the
generated carbene and MeOH was formed. MALDI-TOF-MS: m/z
= 381.4 [M – N2 + MeOH]+. ESI-MS: m/z = 405.11 [M – N2
+
MeOH + Na]+.
Photolysis of Diazirine 14 in Doubly Distilled H2O: The insertion
product of the generated carbene and water was formed. MALDI-
TOF-MS: m/z = 367.4 [M – N2 + H2O]+. ESI-MS: m/z = 391.10
[M – N2 + H2O + Na]+.
Photolysis of Azide 12 in MeOH: The insertion product of the gen-
erated nitrene and MeOH was formed. ESI-MS: m/z = 410.17 [M –
N2 + MeOH + Na]+.
Photolysis of Azide 12 in Doubly Distilled H2O: No insertion prod-
J = 13.8, 6.2 Hz, 1 H, biotin-NHCHCHHS), 1.82 (dd, J = 6.1, ucts were discernable.
Eur. J. Org. Chem. 2006, 4841–4851
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4849

M. Wiegand, T. K. Lindhorst
FULL PAPER
Irradiation Experiments with Various Amino Acids: Irradiation ex-
periments were carried out in doubly distilled H2O in the case of
PALs 14, 16, and 19 and in MeOH for12. The total sample volume
was 0.5 mL with PAL/amino acid ratios of 1:2, 1:5, 1:10, and 2:1.
The concentration of each PAL was chosen as 0.1 m; the concen-
trations of the amino acids were 0.2 m, 0.5 m, 1.0 m, and
0.05 m. The samples were irradiated for 30 min (12), 10 min (14),
or 60 min (16 and 19). After photolysis, the samples were concen-
trated by ZipTipping® to allow for MS analysis (Table 2).
would like to thank our referees for their effort and valuable com-
ments.
[1] A. Singh, E. R. Thornton, F. H. Westheimer, J. Biol. Chem.
1962, 237, 3006–3008.
[2] a) Y. Hatanaka, Y. Sadakane, Curr. Top. Med. Chem. 2002, 2,
271–288; b) O. Jahn, K. Eckart, H. Tezval, J. Spiess, Anal. Bio-
anal. Chem. 2004, 378, 1031–1036.
[3] a) H. Lis, N. Sharon, Chem. Rev. 1998, 98, 637–647; b) D. L.
Evers, K. G. Rice, Glycoscience 2001, 2, 1779–1816; c) M. Am-
brosi, N. R. Cameron, B. G. Davis, Org. Biomol. Chem. 2005,
3, 1593–1608.
[4] a) M. W. Turner, Mol. Immunol. 2003, 40, 423–429; b) E. Arce,
P. M. Nieto, V. Díaz, R. G. Castro, A. Bernad, J. Rojo, Bioconj.
Chem. 2003, 14, 817–823.
[5] a) P. Klemm, K. A. Krogfelt in Fimbriae: Adhesion, Genetics,
Biogenesis and Vaccines (Ed.: P. Klemm), CRC Press, Boca Ra-
ton, 1994, pp. 9–26; b) I. Connell, W. Agace, P. Klemm, M.
Schembri, S. Marild, C. Svanborg, Proc. Natl. Acad. Sci. USA
1996, 93, 9827–9832.
Insertion Products Found with Azide 12: MALDI-TOF-MS: m/z =
707.3 [C34H36F4N2O10 – H]+, covalent adduct with Fmoc-Ile-OH.
Insertion Products Found with Diazirine 14: MALDI-TOF-MS: m/z
= 568.5 [C24H34F3NO11 – H]+, covalent adduct with Boc-Thr-OH;
523.3 [C21H31F3N4O8 – H]+, covalent adduct with H-Arg-OH;
631.4 [C29H37F3N2O10 + H]+, covalent adduct with Z-Lsy-OH.
Insertion Products Found with Benzophenone 16: MALDI-TOF-
MS: m/z = 710.8 [C36H45N3O12 – H]+, covalent adduct with Z-Lys-
OH.
[6] a) A. Siitonen, Ann. Med. 1994, 26, 229–231; b) F. G. Sauer,
M. A. Mulvey, J. D. Schilling, J. J. Martinez, S. J. Hultgren,
Curr. Opin. Microbiol. 2000, 3, 65–72.
Insertion Products Found with Benzophenone 19: MALDI-TOF-
MS: m/z = 1066.3 [C52H71N7O15S], covalent adduct with Z-Lys-
OH.
[7] a) P. Klemm, M. Schembri, Adv. Exp. Med. Biol. 1996, 408,
193–195; b) E. V. Sokurenko, V. Chesnokova, D. E. Dykhu-
idzen, I. Ofek, X.-R. Wu, K. A. Krogfelt, C. Struve, M. A.
Shembri, D. L. Hasty, Proc. Natl. Acad. Sci. USA 1998, 95,
8922–8926; c) M. Vetsch, C. Puorger, T. Spirig, U. Grauschopf,
E.-U. Weber-Ban, R. Glockshuber, Nature 2004, 431, 330–332.
[8] a) S. D. Knight, M. Mulvey, J. Pinkner, Acta Crystallogr. Sect.
A 1997, 53, 207–210; b) D. Choudhury, A. Thompson, V. Sto-
janoff, S. Langermann, J. Pinkner, S. J. Hultgren, S. D. Knight,
Science 1999, 285, 1061–1066; c) C.-S. Hung, J. Bouckaert, D.
Hung, J. Pinkner, C. Widberg, A. Defusco, C. G. Auguste, R.
Strouse, S. Langermann, G. Waksmann, S. J. Hultgren, Mol.
Microbiol. 2002, 44, 903–915; d) J. Bouckaert, J. Berglund, M.
Schembri, E. De Genst, L. Cools, M. Wuhrer, C. S. Hung, J.
Pinkner, R. Slättergård, A. Zavialov, D. Choudhury, S. Langer-
mann, S. J. Hultgren, L. Wyns, P. Klemm, S. Oscarson, S. D.
Knight, H. De Greve, Mol. Microbiol. 2005, 55, 441–455.
[9] a) Th. K. Lindhorst, S. Kötter, U. Krallmann-Wenzel, S.
Ehlers, J. Chem. Soc. Perkin Trans. 1 2001, 823–831; b) N.
Röckendorf, O. Sperling, Th. K. Lindhorst, Austr. J. Chem.
2002, 55, 87–93; c) M. M. K. Boysen, K. Elsner, O. Sperling,
Th. K. Lindhorst, Eur. J. Org. Chem. 2003, 4376–4386; d) N.
Nagahori, R. T. Lee, S.-I. Nishimura, D. Page, R. Roy, Y. C.
Lee, ChemBioChem 2004, 3, 836–844.
Irradiation Experiments in the Presence of Model Peptides Angioten-
sin II and PTHIKWGD and Pentaglycine: To investigate the poten-
tial of PALs 12, 14, 16 and 19 in the presence of peptides, sample
mixtures of the various PALs with a 10-fold excess of investigated
peptide in doubly distilled H2O (total sample volume of 0.5 mL)
were irradiated with UV light Ն320 nm for 30 min (12), 10 min
(14), or 60 min (16 and 19). The concentrations were chosen as
0.1 m for each PAL and 1.0 m for each peptide. For work-up,
the ZipTip® technique[33] was used prior to MALDI-TOF-MS
analysis to reduce the content of cations such as Na+ and K+ and
to increase the concentration of the insertion products in the re-
spective sample (Table 3).
Irradiation of Azide 12 with Angiotensin II: MALDI-TOF-MS: m/z
= 1415.5 [C63H83F4N15O18 + H]+. ESI-MS: m/z = 707.29.
Irradiation of Azide 12 with PTHIKWGD: MALDI-TOF-MS: m/z
= 1308.7 [C57H77F4N13O18]+.
Irradiation of Diazirine 14 with Angiotensin II: MALDI-TOF-MS:
m/z = 1396.6 [C65H88F3N13O18 + H]+. ESI-MS: m/z = 698.82
[C65H88F3N13O18 + H]+.
[10] F. G. Sauer, M. Barnhart, D. Choudhury, S. D. Knight, G.
Waksman, S. J. Hultgren, Curr. Opin. Struct. Biol. 2000, 10,
548–556.
[11] a) M. Walter, Th. K. Lindhorst, Synthesis 2006, 952–958; b)
M. Walter, M. Wiegand, Th. K. Lindhorst, Eur. J. Org. Chem.
2006, 719–728.
Irradiation of Diazirine 14 with PTHIKWGD: MALDI-TOF-MS:
m/z = 1326.6 [C59H81F3N12O18 + Na]+. ESI-MS: m/z = 671.29
[C59H81F3N12O18 +K]+.
Irradiation of Benzophenone 16 with Pentaglycine: MALDI-TOF-
MS: m/z = 733.8 [C32H42N6O14 – H]+.
[12] a) A. Reiser, H. M. Wagner in The Chemistry of the Azido
Group (Ed.: S. Patai), Wiley, New York 1971, pp. 441–501; b)
R. A. Abramovitch in Organic Reactive Intermediates (Ed.:
S. P. McManus), Academic Press, New York 1973, pp. 127–192;
c) G. B. Schuster, M. S. Platz, Adv. Photochem. 1992, 17, 69–
107.
[13] a) J. Brunner, H. Senn, F. M. Richards, J. Biol. Chem. 1980,
255, 3313–3318; b) M. Nassal, Liebigs Ann. Chem. 1983, 1510–
1523; c) L. B. Shih, H. Bayley, Anal. Biochem. 1985, 144, 132–
141; d) Y. Hatanaka, M. Hashimoto, H. Kurihara, H. Nakay-
ama, Y. Kanaoka, J. Org. Chem. 1994, 59, 383–387; e) A. Blen-
cowe, W. Hayes, Soft Matter 2005, 1, 178–205.
Irradiation of Biotinylated PAL 19 with Angiotensin II: MALDI-
TOF-MS: m/z = 1848.2 [C88H122N18O24S]+.
Acknowledgments
This work was financed by the Deutsche Forschungsgemeinschaft
(DFG) in the frame of the collaborative network SFB 470. We are
indebted to Prof. Dr. U. Seydel and PD Dr. B. Lindner (Research
Center Borstel, Division of Biophysics) for measuring the fragmen-
tation spectra of labeled model peptides using ESI-FT-ICP-MS and
their generous support of our work. We are also grateful to PD Dr.
U. Wille (School of Chemistry, University of Melbourne) for the
donation of some irradiation equipment. Last, but not least, we
[14]
a) G. D. Prestwich, G. Dormán, Biochemistry 1994, 33, 5661–
5673; b) P. Mirelis, R. Brossmer, Bioorg. Med. Chem. Lett.
1995, 5, 2809–2814; c) J. J. Tate, J. Persinger, B. Bartholomew,
Nucleic Acids Res. 1998, 26, 1421–1426.
4850
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4841–4851

Photoactive α-Mannosides and Mannosyl Peptides and Evaluation for Lectin Labeling
FULL PAPER
[15] P. J. A. Weber, A. G. Beck-Sickinger, J. Pept. Res. 1997, 49,
375–383.
[16] Th. K. Lindhorst, A. Fuchs, unpublished results.
[17] a) M. S. Platz, H. Huang, F. Ford, J. Toscano, Pure Appl.
Chem. 1997, 69, 803–807; b) J. Brunner, Ann. Rev. Biochem.
1993, 62, 483–514.
[29] a) A. Y. Chernyak, G. V. M. Sharma, L. Kononov, P. R.
Krishna, A. B. Levinsky, N. K. Kochetkov, Carbohydr. Res.
1992, 223, 303–309; b) R. Roy, J. M. Kim, Angew. Chem. 1999,
111, 380–384; Angew. Chem. Int. Ed. 1999, 38, 369–372.
[30] J. Dahmén, T. Frejd, G. Grönberg, T. Lave, G. Magnusson, G.
Noori, Carbohydr. Res. 1983, 116, 303–307.
[18] Y. Hatanaka, M. Hashimoto, Y. Kanaoka, Bioorg. Med. Chem.
1994, 2, 1367–1373.
[19] H. Schuttenberg, R. C. Schulz, Angew. Chem. 1976, 24, 848–
849; Angew. Chem. Int. Ed. Engl. 1976, 15, 777–778.
[20] R. F. R. Church, M. J. Weiss, J. Org. Chem. 1970, 35, 2465–
2471.
[21] a) J. F. Keana, S. X. Cai, J. Org. Chem. 1989, 55, 3640–3647;
b) K. Strømgaard, D. R. Saito, H. Shindou, S. Ishii, T. Shim-
izu, K. Nakanishi, J. Med. Chem. 2002, 45, 4038–4046.
[22] a) T. Autrey, G. B. Schuster, J. Am. Chem. Soc. 1987, 109,
5814–5820; b) I. Woelfle, B. Sauerwein, T. Autrey, Photochem.
Photobiol. 1988, 47, 497–501; c) M. E. Sigman, T. Autrey,
G. M. Schuster, J. Am. Chem. Soc. 1988, 110, 4297–4305; d)
M. J. T. Young, M. S. Platz, Tetrahedron Lett. 1989, 30, 2199–
2202.
[31] L. A. Carpino, J. Am. Chem. Soc. 1993, 115, 4397–4398.
[32] a) M. Hashimoto, Y. Liu, K. Fang, H.-y. Li, G. Campiani,
K. Nakanishi, Bioorg. Med. Chem. 1999, 7, 1181–1194; b) Y.
Hatanaka, U. Kempin, P. Jong-Jip, J. Org. Chem. 2000, 65,
5639–5643.
[33] a) G. Lauc, M. Flögel, W. E. G. Müller, Z. Naturforsch., C
1994, 49, 843–848; b) G. D. Prestwich, G. Dormán, J. T. Elliott,
D. M. Marecak, A. Chaudhary, Photochem. Photobiol. 1997,
65, 222–234; c) F. Koumanov, J. Yang, A. E. Jones, Y. Hat-
anaka, G. D. Holman, Biochem. J. 1998, 330, 1209–1215; d) G.
Lauc, R. T. Lee, J. Dumic´, Y. C. Lee, Glycobiology 2000, 10,
357–364; e) M. Daghish, L. Hennig, M. Findeisen, S. Giesa, F.
Schumer, H. Hennig, A. G. Beck-Sickinger, P. Welzel, Angew.
Chem. 2002, 114, 2404–2408; Angew. Chem. Int. Ed. 2002, 41,
2293–2297.
[23] a) N. Soundararajan, M. S. Platz, J. Org. Chem. 1990, 55,
2034–2044; b) M. J. T. Young, M. S. Platz, J. Org. Chem. 1991,
56, 6403–6406.
[24] a) R. Poe, K. Schnapp, M. J. T. Young, J. Grayzar, M. S. Platz,
J. Am. Chem. Soc. 1992, 114, 5054–5067; b) J. Michalak, H. B.
Zhai, M. S. Platz, J. Phys. Chem. 1996, 100, 14028–14036.
[25] J. H. Billman, J. W. McDowell, Contribution from the Chemistry
Laboratory of Indiana University 1961, 26, 1437–1440.
[26] R. R. Schmidt, W. Kinzy, Adv. Carbohydr. Chem. Biochem.
1994, 50, 21–123.
[34] a) Y. Hatanaka, M. Hashimoto, Y. Kanaoka, J. Am. Chem.
Soc. 1997, 120, 453–454; b) A. Skerra, Biospektrum 2003, 2,
189–192; c) T. Hosoya, T. Hiramatsu, T. Ikemoto, M. Nakani-
shi, H. Aoyama, A. Hosoya, T. Iwata, K. Maruyama, M.
Endo, M. Suzuki, Org. Biomol. Chem. 2004, 2, 637–641.
[35] a) Y. Hatanaka, U. Kempin, P. J. Jip, J. Org. Chem. 2000, 65,
5639–5643; b) M. Völkert, K. Uwai, A. Tebbe, B. Popkirova,
M. Wagner, J. Kuhlmann, H. Waldmann, J. Am. Chem. Soc.
2003, 125, 12749–12758.
[36] Millipore Corporation, Technical Note 2003 (ZipTip Protocol).
[27] G. Zemplén, E. Pascu, Ber. Dtsch. Chem. Ges. 1929, 62, 1613–
1614.
Received: May 23, 2006
Published Online: August 31, 2006
[28] Th. K. Lindhorst, A. Fuchs, O. Sperling, M. Wiegand, unpub-
lished results.
Eur. J. Org. Chem. 2006, 4841–4851
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4851