Semisynthesis of a glycosylphosphatidylinositol-anchored prion protein

Article abstract of DOI:10.1002/anie.200802161

Pinning down the role of the anchor: The chemical synthesis of a cysteine-modified glycosylphosphatidylinositol (GPI) anchor provides access to homogeneous GPI-anchored prion protein through expressed protein ligation (see scheme). By this method, it should be possible to investigate the influence of the complex posttranslational GPI modification on protein structure and function. (Figure Presented)

Full text of DOI:10.1002/anie.200802161

Angewandte
Chemie
DOI: 10.1002/anie.200802161
Synthetic Prion Protein
Semisynthesis of a Glycosylphosphatidylinositol-Anchored Prion
Protein**
Christian F. W. Becker,* Xinyu Liu, Diana Olschewski, Riccardo Castelli, Ralf Seidel, and
Peter H. Seeberger*
Proteins are often modified posttranslationally by glycosyla-
tion and lipidation.[1] Glycosylphosphatidylinositol (GPI)
anchors combine both types of modification and link many
proteins to the cell surface.[2] Advances in solid-phase peptide
synthesis (SPPS) and recombinant protein engineering, in
combination with the development of native chemical ligation
(NCL) and expressed protein ligation (EPL), have resulted in
numerous total syntheses and semisyntheses of proteins.[3]
These approaches facilitate access to homogeneous glyco-
and lipoproteins, which serve as defined molecular probes to
elucidate the effects of glycosylation and lipidation on the
biophysical properties of proteins.[4] Synthetic GPI glycans
and lipidated GPI anchors[5] have emerged as valuable tools
that allow for the precise dissection of their biological
relevance in infectious and metabolic diseases.[6] Efforts
towards the assembly of chemically defined GPI-anchored
proteins have focused on model studies; no synthetic GPI-
anchored protein has been reported to date.[7,8]
pathogenic isoform PrP scrapie (PrPSc). However, the spec-
ulation that GPI anchoring might contribute to the pathoge-
nicity of PrP is controversial.[10] As the isolation of homoge-
neous GPI-anchored PrP has not yet been possible, the
majority of in vitro studies on the function, structure, folding,
and stability of PrP have been carried out with recombinant
protein lacking the GPI anchor, simple GPI-anchor mimics,
or heterogeneous protein preparations from mammalian cell
lines.[11] Thus, the exact function of the GPI anchor could not
be assessed directly. Homogenous GPI-anchored proteins
would be ideally accessed by chemical synthesis. Herein we
report the development of a general strategy for the synthesis
of homogeneous GPI-anchored proteins, with a particular
focus on the prion protein.
We envisioned a general solution based on EPL to the
construction of defined GPI-anchored proteins. We antici-
pated that the synthetic GPI anchor 2, with a cysteine residue
on the 2-aminoethyl phosphate moiety, would undergo
ligation with peptides and proteins with a C-terminal
Ca thioester to give GPI-anchored proteins 1 (Scheme 1).
Several synthetic approaches can be proposed for the
construction of a cysteine-containing GPI anchor of this
type. The direct coupling of cysteine to a native GPI anchor
through an amide linkage is plausible. However, the difficul-
ties in handling native GPI anchors, as well as the instability
of lipid esters under basic conditions, render this approach
less appealing. A more realistic solution is the installation of a
protected cysteine residue on the GPI anchor prior to global
deprotection. The cysteine ethanolamine phosphate residue
will be incorporated into the glycan backbone at the final
stage of the chemical synthesis of GPI. The thiol and amino
groups of the cysteine residue would be protected with acid-
labile groups, such as tert-butyl and tert-butoxycarbonyl (Boc)
groups (e.g. in 3). The benzyl groups would be removed by
hydrogenolysis, and treatment with acid would then liberate
the cysteine residue to furnish the cysteine-tagged GPI anchor
2.
A prominent example of a GPI-anchored protein is the
prion protein (PrP).[9] Numerous studies have indicated the
strong influence of membrane association through the GPI
anchor on the conversion of cellular PrP (PrPC) into its
[*] Prof. Dr. C. F. W. Becker,[+] Dr. D. Olschewski, Dr. R. Seidel
Department of Physical Biochemistry
Max-Planck-Institut fꢀr molekulare Physiologie
Otto-Hahn-Strasse 11, 44227 Dortmund (Germany)
E-mail: christian.becker@ch.tum.de
Homepage: http://www.ch.tum.de/proteinchemie
Dr. X. Liu,[#] R. Castelli, Prof. Dr. P. H. Seeberger
Laboratory for Organic Chemistry
Swiss Federal Institute of Technology (ETH) Zꢀrich
Wolfgang-Pauli-Strasse 10, 8093 Zꢀrich (Switzerland)
Fax: (+41)44-633-1235
E-mail: seeberger@org.chem.etz.ch
Homepage: www.seeberger.ethz.ch
[+] Current address: Technische Universitꢁt Mꢀnchen
Department of Chemistry, Laboratory of Protein Chemistry
Lichtenbergstrasse 4, 85747 Garching (Germany) and
Center for Integrated Protein Science Munich (CIPSM)
Fax: (+49)89-289-13345
Two key transformations had to be studied carefully prior
to executing the synthesis. The incorporation of the phosphate
diester relies on the H-phosphonate method, which requires
the oxidation of phosphorus(III) to phosphorus(V) with
iodine in pyridine and water.[12] It was unclear whether the
thioether would also be oxidized. Moreover, the cleavage of
benzyl ethers by hydrogenolysis requires the use of the
heterogeneous catalyst Pd/C, and thioethers, although less
troublesome than thiols, can poison heterogeneous cata-
lysts.[13] To address these concerns, the two transformations
were evaluated with a model compound (Scheme 2).
[#] Current address: Harvard Medical School
Department of Biological Chemistry and Molecular Pharmacology
240 Longwood Avenue, Boston, MA, 02115 (USA)
[**] Financial support from the Max Planck Society, the Deutsche
Forschungsgemeinschaft, the Fonds der chemischen Industrie, the
Swiss National Science Foundation (SNF grant 205321-107651),
and ETH Zꢀrich is gratefully acknowledged. The authors thank J.
Tatzelt and M. Engelhard for helpful discussions.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200802161.
Angew. Chem. Int. Ed. 2008, 47, 8215 –8219
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8215

Communications
H-phosphonate monoester 5 was mediated by phosphonic
acid in the presence of pivaloyl chloride.[14] We chose the
a(1!2) dimannoside 6, which resembles a portion of the GPI
glycan, as a model to evaluate the phosphorylation and global
deprotection strategy (Scheme 2). The phosphonylation of
the C6 hydroxy group of 6 was mediated by pivaloyl chloride.
The selective in situ oxidation to phosphorus(V) with iodine
at 08C was complete within 1 h, and no product of oxidation
at the sulfur atom was observed. We carefully surveyed
conditions for hydrogenolysis in the presence of thioethers.[15]
A solution of formic acid in methanol in combination with
Pd/C and hydrogen proved effective for the rapid cleavage of
all benzyl ether groups present in 7 within 2 h to furnish 8 in
quantitative yield (Scheme 2). The N-Boc and S-tBu protect-
ing groups were removed with Hg(OTFA)2 in trifluoroacetic
acid (TFA), and excess mercury salts were precipitated by
treatment with 2-sulfanylethanol.[13b] Purification with a
sephadex G-25 column gave 9 as a mixed disulfide with a
2-mercaptoethanol group.
The dimannoside 9 was tested as a substrate for NCL with
recombinant PrP (rPrP) containing a C-terminal MESNa
thioester functionality (MESNa = 2-mercaptoethane sulfona-
te).[11f] The treatment of the rPrP thioester with 9 (1.5 equiv)
in the presence of thiophenol produced the glycan–rPrP
conjugate 10 in 80% yield. The ligation was complete within
12 h at room temperature, as indicated by reversed-phase
(RP) HPLC (Figure 1A). The homogeneity of the product
was assessed by electrospray ionization mass spectrometry
(ESIMS), and the observed mass of 16929.5 Da agrees well
with the calculated mass of 10 (16927 Da; Figure 1B).
On the basis of these results, we proceeded to assemble a
complete GPI anchor. Native prion GPI anchors contain a
core pseudopentasaccharide glycan common to mammalian
GPI anchors; this core glycan is amended with an oligosac-
charide branch up to the trisaccharide Neu-Gal-GalNAc.[16]
Scheme 1. General strategy for the semisynthesis of a GPI-anchored
protein by native chemical ligation (R’=lipid chain, R’’=alkyl or aryl
group). Bn=benzyl, Boc=tert-butoxycarbonyl.
The synthesis commenced with the preparation of the H-
phosphonate 5 (Scheme 2). Commercially available N-Boc-S-
tBu-l-cysteine (4) was activated with succinimide and treated
with ethanolamine at 08C. Subsequent conversion into the
Scheme 2. Evaluation of the synthetic strategy. Synthesis of a GPI
dimannoside fragment: a) N-hydroxysuccinimide, DIPC, THF, room
temperature; b) ethanolamine, THF, DMF, 90% (2 steps); c) H3PO3,
PivCl, pyridine, room temperature, 75%; d) 5, PivCl, pyridine; e) I2,
pyridine/H2O, 08C, 94% (2 steps); f) Pd/C, H2, 4% HCOOH in
MeOH, quantitative; g) Hg(OTFA)2, TFA/anisole, 08C; then
HSCH2CH2OH, AcOH, H2O, room temperature, 93%; h) PhSH (1%,
v/v), pH 7.8 (R=(CH2)2SO3ꢀ). DIPC=diisopropylcarbodiimide,
DMF=N,N-dimethylformamide, Piv=pivaloyl.
Figure 1. A) RP-HPLC chromatogram for the NCL reaction of the rPrP
thioester with 9 after 12 h. B) ESI mass spectrum of material from the
fraction with the HPLC peak in (A).
8216
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 8215 –8219

Angewandte
Chemie
Six glycoforms of prion protein GPI have been reported;
however, in all cases, the exact linkage position and anomeric
configuration of the oligosaccharide branch were not
defined.[17] No details of the lipid composition of prion
protein GPIs are known. In view of the structural uncertainty
associated with the prion protein GPI glycan, we selected the
core GPI pseudopentasaccharide 2 as the initial synthetic
target.
tert-butyl groups in TFA with Hg(OTFA)2. Excess mercury
salts were precipitated with 2-sulfanylethanol to give the
cysteine-tagged GPI anchor 22 as a heterodisulfide, as
identified by ESIMS (see the Supporting Information).
The crude GPI anchor 22 was subjected directly to NCL
with recombinant PrP (rPrP) containing a C-terminal MESNa
The convergent assembly of 2 relied on a [3+2] glyco-
sylation strategy to connect a trimannoside with a glucos-
amine–inositol pseudodisaccharide (Scheme 3). The selective
coupling of the mannosyl trichloroacetimidate 12 and the
known dimannoside 11[18] was mediated by TMSOTf in
diethyl ether. The allyl group at the reducing terminus of 13
was removed with PdCl2 in acetate buffer. Treatment with
trichloroacetonitrile and DBU then furnished the trimannosyl
trichloroacetimidate 14. The union of 14 and pseudodisac-
charide 15 in the presence of a catalytic amount of TMSOTf
led to the differentially protected pseudopentasaccharide,
which was treated with sodium methoxide to give 16 in 86%
yield over two steps. Benzylation of the central mannose unit
and removal of the allyl protecting group on the inositol
residue furnished 17. Subsequent phosphorylation was carried
out in
a one-pot, two-step sequence by using the
H-phosphonate 18 as a diacyl glycerol phosphate surrogate
to give 19.[19] The C6 silyl ether of the terminal mannose
residue was removed with HCl generated in situ in methanol.
Further phosphorylation with 5 gave the fully protected
lipidated GPI anchor 21.
The stepwise final deprotection was carried out under the
conditions described for the model dimannoside
(Scheme 4). Hydrogenolysis with Pd/C in formic acid and
methanol was followed by removal of the acid-labile Boc and
9
Scheme 4. Completion of the synthesis of the cysteine-tagged GPI
pseudopentasaccharide and native chemical ligation with the rPrP
thioester.
Scheme 3. Assembly of the protected GPI pseudopentasaccharide 21 with a 2-(cysteinylamino)phosphate moiety: a) TMSOTf, CH2Cl2, 95%;
b) PdCl2, AcONa, AcOH, AcOEt, room temperature; c) Cl3CCN, DBU, CH2Cl2, 68% (2 steps); d) TMSOTf, CH2Cl2, 08C; e) NaOMe, MeOH, 508C,
86% (2 steps); f) BnBr, NaH, DMF, 08C!RT; then PdCl2, AcONa, AcOH, room temperature, 76% (2 steps); g) 1) 18, PivCl, pyridine, room
temperature; 2) I2, pyridine/water, quantitative (2 steps); h) HCl, MeOH, 08C!RT, 88%; i) 1) 5, PivCl, pyridine, room temperature; 2) I2, pyridine/
water, quantitative (2 steps). All=allyl, Bz=benzoyl, DBU=1,8-diazabicyclo[5.4.0]undec-7-ene, TIPS=triisopropylsilyl, TMSOTf=trimethylsilyl
trifluoromethanesulfonate.
Angew. Chem. Int. Ed. 2008, 47, 8215 –8219
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8217

Communications
thioester.[11f] The ligation of the rPrP thioester with 22
(1.5 equiv) proceeded in 6m guanidine hydrochloride at
pH 7.8 in the presence of 1% (v/v) thiophenol to produce
23 in 50% yield (Scheme 4). The GPI-anchored rPrP 23 was
purified by RP HPLC and analyzed by MS (calculated mass:
17705 Da; Figure 2A,B).[20] Excess GPI anchor (derived from
22) without the 2-mercaptoethanol protecting group was
recovered and recycled to improve the overall efficiency of
the synthesis. No addition of detergents or lipids was required
to solubilize 22 or 23 during the ligation, in contrast to
described ligations of lipidated peptides.[7b,16,21]
The GPI anchor 22 confers lipophilic properties to
proteins; at the same time, the large hydrophilic head group
ensures solubility in water. The folding of 23 into its native
form was promoted by rapid 10-fold dilution with a buffer
containing 20 mm NaOAc at pH 5.5, followed by gel filtra-
tion.[22] The folded rPrP–GPI 23 was obtained in 90% yield.
SDS-PAGE analysis confirmed the homogeneity of 23, with
the observation of a sharp band at approximately 17 kDa
(Figure 2C). CD spectroscopy indicated an a-helical secon-
dary structure, which is characteristic of PrPC. The observed
molar ellipticities for folded 23 are comparable to those
measured for PrPC obtained from expression in bacterial and
eukaryotic systems and indistinguishable from those of folded
rPrP without a GPI anchor (Figure 3A).[11b,e,f] Folded 23 was
soluble in aqueous buffers without the addition of detergents
or lipids.
~
*
Figure 3. A) CD spectrum of folded 23 ( ) and rPrP ( ) at
0.2 mgmLꢀ1 in NaOAc buffer at pH 5.5. B) Immunoblotting analysis of
a vesicle pull-down assay of folded 23 (upper panel) and rPrP (lower
panel) with the PrP-specific antibody A7. DOPC vesicles with an
average diameter of 80 nm were mixed with 23. Lane 1: pellet resulting
from the low-spin centrifugation of 23 and rPrP solutions before
addition to vesicles; lane 2: supernatant of high-spin centrifugation of
vesicles incubated with PrP samples; lane 3: pellet resulting from the
high-speed centrifugation of vesicles incubated with PrP samples.
The influence of GPI anchoring on the membrane
association of rPrPC was studied by incorporating 23 into
small unilamellar 1,2-dioleoyl-sn-glycero-3-phosphocholine
(DOPC) vesicles (SUVs) in a vesicle-spin-down assay. Low-
spin centrifugation of 23 removed any aggregated protein
prior to transfer into the vesicles (Figure 3B, lane 1). No
precipitation of 23 was observed during this first centrifuga-
tion step. Fully soluble 23 was diluted 10-fold in a solution of
SUVs at a concentration of 10 mgmLꢀ1 and incubated for
30 min at room temperature. Subsequent high-speed centri-
fugation precipitated the SUVs and all vesicle-associated
protein. Analysis of the pellets and the supernatant revealed
that 23 was located exclusively in the pellet (Figure 3B,
lanes 2 and 3). This observation indicates quantitative attach-
ment of 23 to the vesicles, although the synthetic GPI anchor
contains only one C18 lipid chain. In contrast, peptide-based
GPI-anchor mimics require at least two lipid chains for
efficient membrane attachment. rPrP without a lipid anchor
remained in the supernatant under the same conditions
(Figure 3B, lane 2). These observations emphasize the power
of GPI anchors in the membrane association of proteins.
In summary, we have developed a general synthetic
strategy for the preparation of homogeneous GPI-anchored
proteins. Our approach is based on the native chemical
ligation of a synthetic cysteine-tagged GPI anchor with a
recombinant protein containing
a C-terminal thioester.
Access to the lipidated GPI anchor relies on the incorporation
of the cysteine residue into the GPI backbone prior to global
deprotection and on the judicious selection of protecting
Figure 2. A) ESI mass spectrum of 23. B) Deconvoluted mass spec-
trum of 23 (MWcalcd: 17704 Da). C) SDS-PAGE of purified 23 (right
lane); left lane: molecular-weight marker.
8218
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 8215 –8219

Angewandte
Chemie
[8] For a GPI-glycan–peptide synthesis, see: N. Shao, B. Xue, Z. W.
groups. This strategy was applied to the semisynthesis of a
GPI-anchored prion protein. Synthetic access to lipidated
GPI anchors in milligram quantities for selective protein
modification by native chemical ligation paves the way for
detailed analysis of the influence of this complex posttransla-
tional modification on protein structure and function. In vitro
and in vivo experiments with this GPI-anchored PrP should
be helpful in the elucidation of the influence of GPI-mediated
membrane association on the conversion of PrPC into
pathogenic PrPSc.
Guo, Angew. Chem. 2004, 116, 1595 – 1599; Angew. Chem. Int.
Ed. 2004, 43, 1569 – 1573.
[9] N. Stahl, D. R. Borchelt, K. Hsiao, S. B. Prusiner, Cell 1987, 51,
229 – 240.
[10] For recent examples of this controversy, see: a) P. A. Lewis, F.
Properzi, K. Prodromidou, A. R. Clarke, J. Collinge, G. S.
Jackson, Biochem. J. 2006, 395, 443 – 448; b) B. Chesebro, M.
Trifilo, R. Race, K. Meade-White, C. Teng, R. LaCasse, L.
Raymond, C. Favara, G. Baron, S. Priola, B. Caughey, E.
Masliah, M. Oldstone, Science 2005, 308, 1435 – 1439; c) G.
Legname, I. V. Baskakov, H. O. B. Nguyen, D. Riesner, F. E.
Cohen, S. J. DeArmond, S. B. Prusiner, Science 2004, 305, 673 –
676.
Received: May 8, 2008
Published online: September 22, 2008
[11] For examples, see: a) L. Breydo, Y. Sun, N. Makarava, C. I. Lee,
V. Novitskaia, O. Bocharova, J. P. Y. Kao, I. V. Baskakov,
Biochemistry 2007, 46, 852 – 861; b) M. R. Hicks, A. C. Gill,
I. K. Bath, A. K. Rullay, I. D. Sylvester, D. H. Crout, T. J. T.
Pinheiro, FEBS J. 2006, 273, 1285 – 1299; c) D. R. Taylor, N. M.
Hooper, Mol. Membr. Biol. 2006, 23, 89 – 99; d) P. Critchley, J.
Kazlauskaite, R. Eason, T. J. T. Pinheiro, Biochem. Biophys. Res.
Commun. 2004, 313, 559 – 567; e) H. Eberl, P. Tittmann, R.
Glockshuber, J. Biol. Chem. 2004, 279, 25058 – 25065; f) D.
Olschewski, R. Seidel, M. Miesbauer, A. S. Rambold, D.
Oesterhelt, K. F. Winklhofer, J. Tatzelt, M. Engelhard, C. F. W.
Becker, Chem. Biol. 2007, 14, 994 – 1006; g) A. Gorodinsky,
D. A. Harris, J. Cell Biol. 1995, 129, 619 – 627.
[12] a) C. B. Reese, Q. L. Song, J. Chem. Soc. Perkin Trans. 1 1999,
1477 – 1486; b) R. H. Hall, A. Todd, R. F. Webb, J. Chem. Soc.
1957, 3291 – 3296.
[13] a) S. A. Hutchinson, S. P. Baker, J. Linden, P. J. Scammells,
Bioorg. Med. Chem. 2004, 12, 4877 – 4884; b) Y. Okada, N. Ohta,
Chem. Pharm. Bull. 1982, 30, 581 – 585.
Keywords: glycosylphosphatidylinositol ·
native chemical ligation · protein semisynthesis · proteins ·
total synthesis
.
[1] C. T. Walsh, S. Garneau-Tsodikova, G. J. Gatto, Jr., Angew.
Chem. 2005, 117, 7508 – 7539; Angew. Chem. Int. Ed. 2005, 44,
7342 – 7372.
[2] For reviews, see: a) H. Ikezawa, Biol. Pharm. Bull. 2002, 25,
409 – 417; b) A. S. Campbell in Glycoscience: Chemistry and
Biology, Vol. 2 (Eds.: B. Fraser-Reid, K. Tatsuta, J. Thiem),
Springer, Berlin, 2001, pp. 1696 – 1777; c) M. A. J. Ferguson, J.
Cell Sci. 1999, 112, 2799 – 2809.
[3] For reviews, see: a) P. E. Dawson, S. B. H. Kent, Annu. Rev.
Biochem. 2000, 69, 923 – 960; b) T. W. Muir, Annu. Rev. Bio-
chem. 2003, 72, 249 – 289; c) V. Muralidharan, T. W. Muir, Nat.
Methods 2006, 3, 429 – 438.
[14] C. Lin, H. Fu, G. Tu, Y. Zhao, Synthesis 2004, 509 – 513.
[15] a) A. M. Felix, M. H. Jimenz, J. Meienhofer, Org. Synth. Coll.
Vol. 59, 159; b) B. ElAmin, G. M. Anantharamaiah, G. P. Royer,
G. E. Means, J. Org. Chem. 1979, 44, 3442 – 3444; c) A. M. Felix,
M. H. Jimenez, T. Mowles, J. Meienhofer, Int. J. Pept. Protein
Res. 1978, 11, 329.
[16] N. Stahl, M. A. Baldwin, R. Hecker, K. M. Pan, A. L. Burlin-
game, S. B. Prusiner, Biochemistry 1992, 31, 5043 – 5053.
[17] M. A. Baldwin, Methods Enzymol. 2005, 45, 172 – 186.
[18] a) Y. U. Kwon, R. L. Soucy, D. A. Snyder, P. H. Seeberger,
Chem. Eur. J. 2005, 11, 2493 – 2504; b) X. Liu, Y.-U. Kwon, P. H.
Seeberger, J. Am. Chem. Soc. 2005, 127, 5004 – 5005.
[19] The structure of the lipid moieties in prion GPI anchors has not
been fully elucidated (see Refs. [16, 17]). It is presumed that they
are derived from glycerol 1-phosphate; however, this presump-
tion includes the possibility of diacyl or dialkyl glycerol lipids, 1-
acyl or 1-alkyl 2-lysoglycerol lipids, or other glycerol lipids.
Therefore, as a starting point for our study, we chose a single
lipid chain (stearoyl phosphate), which should be long enough to
interact with membranes.
[4] For reviews, see: a) M. J. Grogan, M. R. Pratt, L. A. Marcaur-
elle, C. R. Bertozzi, Annu. Rev. Biochem. 2002, 71, 593 – 634;
b) A. Brik, S. Ficht, C. H. Wong, Curr. Opin. Chem. Biol. 2006,
10, 638 – 644; c) L. Brunsveld, J. Kuhlmann, K. Alexandrov, A.
Wittinghofer, R. S. Goody, H. Waldmann, Angew. Chem. 2006,
118, 6774 – 6798; Angew. Chem. Int. Ed. 2006, 45, 6622 – 6646.
[5] For reviews on GPI syntheses, see: a) Z. W. Guo, L. Bishop, Eur.
J. Org. Chem. 2004, 3585 – 3596; b) D. Martin-Lomas, J. Lopez-
Prados in Comprehensive Glycoscience, Vol. 1 (Eds.: J. P.
Kamerling, G.-J. Boons, Y. Lee, A. Suzuki, N. Taniguchi,
A. G. J. Voragen), Elsevier, Dordrecht, 2007, pp. 748 – 795.
[6] For selected biological applications of synthetic GPIs, see:
a) N. W. Brattig, K. Kowalsky, X. Liu, G. D. Burchard, F.
Kamena, P. H. Seeberger, Microb. Infect. 2008, 10, 885 – 891;
b) F. Kamena, M. Tamborrini, X. Liu, Y. U. Kwon, F. Thompson,
G. Plusche, P. H. Seeberger, Nat. Chem. Biol. 2008, 4, 238 – 240;
c) T. K. Smith, A. Crossman, J. S. Brimacombe, M. A. J. Fergu-
son, EMBO J. 2004, 23, 4701 – 4708; d) L. Schofield, M. C.
Hewitt, K. Evans, M. A. Siomos, P. H. Seeberger, Nature 2002,
418, 785 – 789; e) T. K. Smith, A. Crossman, C. N. Borissow, M. J.
Paterson, A. Dix, J. S. Brimacombe, M. A. J. Ferguson, EMBO J.
2001, 20, 3322 – 3332; f) M. Martꢀn-Lomas, N. Khiar, S. Garcia,
J.-L. Koessler, P. M. Nieto, T. W. Rademacher, Chem. Eur. J.
2000, 6, 3608 – 3621; g) W. Frick, A. Bauer, J. Bauer, S. Wied, G.
Mueller, Biochemistry 1998, 37, 13421 – 13436.
[20] The RP HPLC analysis and mass analysis of 23 were aggravated
by the amphiphilic nature of this compound. MALDI-TOF
analysis confirmed the identity of 23 (see Figure S4 in the
Supporting Information).
[21] T. Durek, K. Alexandrov, R. S. Goody, A. Hildebrand, I.
Heinemann, H. Waldmann, J. Am. Chem. Soc. 2004, 126,
16368 – 16378.
[7] For model synthetic studies on GPI-anchored proteins, see: a) Y.
Tanaka, Y. Nakahara, H. Hojo, Y. Nakahara, Tetrahedron 2003,
59, 4059 – 4067; b) M. J. Grogan, Y. Kaizuka, R. M. Conrad, J. T.
Groves, C. R. Bertozzi, J. Am. Chem. Soc. 2005, 127, 14383 –
14387; c) M. G. Paulick, A. R. Wise, M. B. Forstner, J. T. Groves,
C. R. Bertozzi, J. Am. Chem. Soc. 2007, 129, 11543 – 11550.
[22] R. Riek, S. Hornemann, G. Wider, M. Billeter, R. Glockshuber,
K. Wꢁthrich, Nature 1996, 382, 180 – 182.
Angew. Chem. Int. Ed. 2008, 47, 8215 –8219
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8219