Synthesis of an isotopically-labelled antarctic fish antifreeze glycoprotein probe

Article abstract of DOI:10.1071/CH10464

Antifreeze glycoproteins (AFGPs) are glycosylated polypeptides produced by Antarctic and Arctic fishes, which allow them to survive in seawater at sub-zero temperatures. An investigation into the postulated enteric uptake of AFGP synthesized in the exocrine pancreas of Antarctic fishes required a custom-prepared AFGP probe that incorporated seven isotopically-labelled Ala residues for detection by mass spectrometry. The AFGPs are composed of a repetitive three amino acid unit (Ala-Ala-Thr), in which the threonine residue is glycosylated with the disaccharide β-d-Gal-(1→3) α-d-GalNAc. The synthesis of isotopically-labelled AFGP8 (1), as well as the optimized synthesis of the protected glycosylated amino acid building block 2, is reported.

Full text of DOI:10.1071/CH10464

RESEARCH FRONT
Full Paper
CSIRO PUBLISHING
www.publish.csiro.au/journals/ajc
Aust. J. Chem. 2011, 64, 723–731
Synthesis of an Isotopically-labelled Antarctic Fish
Antifreeze Glycoprotein Probe
A
B
C
Joanna M. Wojnar, Clive W. Evans, Arthur L. DeVries,
A D
,
and Margaret A. Brimble
A
Department of Chemistry, The University of Auckland, 23 Symonds Street,
Auckland 1142, New Zealand.
B
School of Biological Sciences, The University of Auckland, 3a Symonds Street,
Auckland 1142, New Zealand.
C
Department of Animal Biology, University of Illinois at Urbana-Champaign,
524 Burrill Hall, 407 S Goodwin, Urbana, IL 61801, USA.
D
Corresponding author. Email: m.brimble@auckland.ac.nz
Antifreeze glycoproteins (AFGPs) are glycosylated polypeptides produced by Antarctic and Arctic fishes, which allow
them to survive in seawater at sub-zero temperatures. An investigation into the postulated enteric uptake of AFGP
synthesized in the exocrine pancreas of Antarctic fishes required a custom-prepared AFGP probe that incorporated seven
isotopically-labelled Ala residues for detection by mass spectrometry. The AFGPs are composed of a repetitive three
amino acid unit (Ala-Ala-Thr), in which the threonine residue is glycosylated with the disaccharide b-D-Gal-(1-3)-
a-D-GalNAc. The synthesis of isotopically-labelled AFGP8 (1), as well as the optimized synthesis of the protected
glycosylated amino acid building block 2, is reported.
Manuscript received: 20 December 2010.
Manuscript accepted: 28 January 2011.
Introduction
and are thought to operate by adsorbing onto specific faces of
ice crystals, inhibiting crystal growth.[1b,3] AFGPs consist pre-
dominantly of a repetitive three amino acid unit (Ala-Ala-Thr),
with the threonine hydroxyl bearing the disaccharide b-D-
galactose-(1-3)-a-D-N-acetyl-galactosamine (identified clini-
cally as the Thomsen-Friedenrich (TF) antigen; CD176). The
different AFGPs identified to date vary in the number of repeat
units present (4–50, equivalent to molar masses between 2.6 and
33.7 kDa), and the smaller variants sometimes have a proline
in place of the first alanine in the trimer.[1c,2] The isoforms are
Antifreeze proteins and glycoproteins are a class of polypeptides
that inhibit ice crystal growth, allowing certain vertebrates,
insects, or plants that produce them to survive at sub-zero
temperatures.[1] Specifically, Antarctic notothenioid fish and
northern cod produce antifreeze glycoproteins (AFGPs) that
lower the freezing point of the blood and permit the fishes to
live at temperatures close to the freezing point of seawater
(,ꢀ1.98C).[1a,2] The AFGPs are present in very high quantities
(up to 3.5% (w/v) or 35 mg mLꢀ1) in the plasma of the fish,[2]
Margaret Brimble graduated from the University of Auckland with an MSc (First Class Honours) in Chemistry in 1983. She
was subsequently awarded a UK Commonwealth Scholarship to undertake her Ph.D. studies at Southampton University, UK.
In 1986, she was appointed as a lecturer at Massey University. After a Visiting Professorship at the University of California,
Berkeley, she moved to the University of Sydney in 1995 as a Reader in Organic Chemistry. She returned to New Zealand in
1999 to take up the Chair of Organic Chemistry at the University of Auckland. She established New Zealand’s first BSc (Hons)
degree in medicinal chemistry and has been Director of the Medicinal Chemistry Programme since 2002. Professor Brimble’s
research focusses on the synthesis of novel bioactive natural products, peptides, and peptidomimetics as potential therapeutic
agents. She developed the drug candidate NNZ2566 for Neuren Pharmaceuticals that is in phase 2b clinical trials for traumatic
brain injury in partnership with the US Army Walter Reed Institute. She is currently the Chairperson of the Rutherford
Foundation, Principal Investigator of the Maurice Wilkins Centre for Molecular Biodiscovery Titular Member of the
International Union of Pure and Applied Chemistry (IUPAC) Organic and Biomolecular Chemistry Division and a member of
international advisory boards for several international chemistry journals. She is the past convenor of the Marsden Fund –
Chemistry, Physics and Biochemistry Panel, a member of the Marsden Fund Council and the past President of the International
Society of Heterocyclic Chemistry. She has published over 250 papers and is a member of the Editorial Board for Organic and
Biomolecular Chemistry and a member of the International Advisory Board for Synthesis, Synlett, Natural Product Reports,
and several other journals. In 2004, Professor Brimble was conferred the Queen’s honour Member New Zealand Order of
Merit (MNZM). She was named the 2007 L’Ore´al-UNESCO Women in Science Laureate for Asia-Pacific in Materials Science
and the recipient of the 2010 Natural Product Chemistry Award from the Royal Society of Chemistry.
Ó CSIRO 2011
10.1071/CH10464
0004-9425/11/060723

RESEARCH FRONT
724
J. M. Wojnar et al.
HO
HO
(a), (b)
87%
OH
OAllyl
OAc
O
H2N
FmocHN
AcO
AcO
O
O
(e)
3
ꢀ
N3
O
72%
OH
OAc
O
OH
HO
HO
HO
AcO
AcO
OAllyl
O
(c), (d)
68%
FmocHN
O
O
HO
OH
OH
NH2
N3
OH
OAc
5
.HCl
4
(i), (j)
56%
OAc
AcO
AcO
O
OCH3
OH
OAc
(f)
86%
9
(k)
71%
O
O
OAc
O
AcO
AcO
OAc
AcO
AcO
(g), (h)
70%
O
O
ꢀ
HO
AcHN
AcHN
AcO
O
CCl3
NH
O
O
OAllyl
OAllyl
8
FmocHN
FmocHN
O
O
7
6
(l)
63%
OCH3
O
O
OAc
O
OAc
O
OAc
O
OAc
O
OAc
O
AcO
AcO
AcO
AcO
AcO
AcO
O
AcO
O
(o)
95%
(m), (n)
70%
O
AcO
O
AcO
AcHN
AcO
AcHN
AcO
AcHN
O
O
O
OAllyl
OAllyl
OH
O
FmocHN
FmocHN
FmocHN
10
11
2
O
O
Scheme 1. Synthesis of the Gal-GalNAc-Thr building block 2. Reagents and conditions: (a) Fmoc-OSu, Na2CO3, 1,4-dioxane, RT, 16 h; (b) allyl bromide,
NaHCO3/H2O, nBu4NþBrꢀ, RT, 20 h; (c) imidazole-1-sulfonyl azide,[16] Na2CO3, CuIISO4, MeOH/H2O, RT, 2.5 h; (d) Ac2O/Et3N, RT, 3 h; (e) BF3ꢁEt2O,
CH2Cl2, 408C, 78 h; (f) Zn, CuSO4, THF/Ac2O/AcOH, RT, 30 min; (g) AcCl, MeOH, 408C, 36 h; (h) 4-MeOPhCH(OMe)2, cat. pTsOH, DMF, 558C, 4 h,
˚
20 mbar; (i) cat. HClO4, Ac2O, 1 h; (j) Cu(OAc)2, MeOH/H2O, reflux, 4 h; (k) Cl3CCN, K2CO3, DCM, RT, 16 h; (l) TMSOTf, 4 A MS, CH2Cl2, ꢀ108C to RT,
2 h; (m) HOAc (80%), RT, 1.5 h; (n) Ac2O/pyridine, RT, 16 h; (o) Pd(PPh3)4, N-methylaniline, THF, RT, 2 h.
grouped into eight classes (AFGP1–8) based on their size (large
to small) and electrophoretic properties.
The preparation of such large and complex molecules is a
formidable challenge, hence most synthetic efforts to date have
focussed on simpler AFGP analogues (see ref. [6] for a recent
review). Two main approaches are usually considered for the
synthesis of glycopeptides: direct glycosylation of a full-length
peptide, or the incorporation of a pre-formed glycosylated
amino acid building block into stepwise solid-phase synthesis
of the peptide. While direct glycosylation is more convergent, it
often gives poor yields due to the low reactivity of the peptide
side chain hydroxyl groups.[7] The Nishimura group have used
another alternative: solution-phase polymerization of a glyco-
sylated tripeptide building block to make the glycopeptide.[8]
Unfortunately, this method lacks fine control over the peptide
chain length, resulting in a distribution of molecular weights
rather than a single isoform, and does not permit the incorpora-
tion of proline at specific sites.
As the liver is the main source of secreted plasma proteins, it
was also thought to be the site of biosynthesis of the abundant
plasma AFGPs. Recently, however, Cheng et al.[4] have shown
the exocrine pancreas is the major site of AFGP synthesis, with
additional contributions from the oesophagus and anterior por-
tion of the stomach. Since all three of these organs secrete into
the gastrointestinal tract and not into the circulatory system, the
possibility arises that AFGPs reach the blood by re-absorption
intact through the intestinal wall.[5] In order to prove this
hypothesis, we wish to introduce isotopically-labelled AFGPs
directly into the gut (to mimic the natural secretory pathway)
and monitor their subsequent uptake by mass spectrometry of
blood samples. For this purpose, we require a custom-prepared
AFGP that incorporates an isotopically-labelled residue (or
preferably multiple labelled residues) in order for the introduced
AFGPs to be distinguished from the endogenous glycoproteins
by mass spectrometry.
Tseng et al.[9] were the first to use a glycosylated amino acid
building block in solid-phase peptide synthesis (SPPS) to gen-
erate natural AFGPs, showing that homogenous peptides can be

RESEARCH FRONT
Synthesis of Isotopically-labelled AFGP8 (1)
725
OAc
O
OAc
O
AcO
AcO
O
Fmoc Ala
O
HMPBA
AcO
Thr* ꢁ
Fmoc removal (a)
Amino acid coupling (b), (c), or (d)
AcO
AcHN
O
O
N
H
O
H2N-Ala-Ala-Thr*-Ala-Ala-Thr*-Pro-Ala-Thr*-Ala-Ala-Thr*-Pro-Ala-O
Cleavage from resin (e)
WANG
Ala ꢁ15N-Ala
H2N-Ala-Ala-Thr*-Ala-Ala-Thr*-Pro-Ala-Thr*-Ala-Ala-Thr*-Pro-Ala-OH
12
Removal of acetate protecting groups (f)
OH
O
OH
O
HO
HO
HO
O
OH
O
OH
O
HO
HO
HO
O
O
H
OH
O
AcHN
O
N
N
O
H
15N
H
15N
O
O
O
OH
H
N
OH
AcHN
O
H
15N
H
15N
N
O
O
O
15N
H
H
N
H
O
O
15N
N
15N
N
O
AcHN
OH
O
H
H
O
O
O
H
O
H
HO
HO
O
AcHN
OH
O
O
HO
HO
O
HO
OH
OH
O
1
HO
OH
OH
Scheme 2. Synthesis of labelled AFGP8 peptide (1). Reagents and conditions: (a) 20% piperidine/DMF, MW 758C, 30 s then 3 min; (b) Fmoc-Pro, HBTU,
iPr2EtN, DMF, MW 758C, 5 min (double coupling); (c) Fmoc-15N-Ala, HBTU, iPr2EtN, DMF, MW 758C, 15 min; (d) glycoside 2, HATU, HOAt, collidine,
DMAP (cat.), DMF, MW 758C, 20 min; (e) TFA/TIPS/H2O (95:2.5:2.5), RT, 2 h; (f) 1 M NaOMe/MeOH, pH 9–10, RT, 3 h.
obtained by this method. In a similar fashion to Tseng et al.,
we have employed the glycosylated building block strategy for
the synthesis of the labelled AFGP probe. We decided on an
alternate route to the glycosylated amino acid, however, in order
to also obtain access to the clinically relevant Tn-antigen
(CD175). The Tn antigen (a-D-N-acetyl-galactosamine) conju-
gated to serine/threonine is the most prevalent of the O-linked
glycoprotein motifs,[7] and, along with the TF antigen, they
are often overexpressed in oncogenically transformed cells.
Glycosylated building blocks of this nature are therefore attrac-
tive synthetic targets both for the synthesis of cancer vaccine
candidates,[10] and other glycopeptides or glycoproteins.
Herein we report the synthesis of AFGP8 (1), the most
abundant variant of the smallest natural AFGP,[11] containing
seven 15N-Ala residues. The glycopeptide probe was prepared
through the use of the protected glycosylated amino acid
building block 2 in microwave-assisted SPPS (see Schemes 1
and 2).
Previous work in our laboratory[13] has also established that
tetra-O-acetyl-2-azido-deoxygalactose (4) was the most suitable
glycosyl donor to effect glycosylation of the protected threonine.
Glycosyl donor 4 was previously made[14] from galactose via
the glycal, which was converted into 2-azido-2-deoxygalactosyl
nitrate using cerium ammonium nitrate and sodium azide. Regret-
tably, this route suffers from persistent low yields obtained
from the critical azidonitration step (yields of 30–40% are
common).[14]
An alternative route to 2-azido-2-deoxy galactose makes use
of a diazotransfer reaction,[15] using trifluoromethanesulfonyl
azide as the ‘diazo donor’ to transform an amine (e.g. galacto-
samine) into the corresponding azide. TfN3 has several pro-
blems associated with its use: it is explosive and has to be
handled with care, has relatively poor shelf-life, provides
inconsistent yields during preparation (which requires the
expensive precursor trifluoromethanesulfonic anhydride), and
also necessitates specialized workup to remove the sulfonamide
by-product.[16] More recently, Goddard-Borger et al.[16] synthe-
sized a new diazotransfer reagent, the inexpensive, shelf-stable
imidazole-1-sulfonyl azide hydrochloride, which can be pre-
pared in a simple one-pot procedure. We therefore used the
imidazole-1-sulfonyl azide hydrochloride as the diazo transfer
agent to effect the conversion of galactosamine to azide 4.
The glycosyl donor 4 thus obtained was then coupled to the
amino acid glycosyl acceptor 3 using boron trifluoride-diethyl
etherate with gentle heating for 78 h. The progress of the
glycosylation reaction was often difficult to interpret by TLC
due to the similar Rf values of the reagents. The reaction was
Results and Discussion
The synthetic route to the glycosylated amino acid building
block 2 is shown in Scheme 1. First, the amino acid L-threonine
was converted into a suitable glycosyl acceptor (3) for the initial
glycosylation reaction. The amino group was protected with
Fmoc using Fmoc-N-hydroxysuccinimide, while the carboxyl
group was protected as an allyl ester (using allyl bromide with a
phase transfer catalyst), as we have found this to be the most
reliable protecting group.[12]

RESEARCH FRONT
726
J. M. Wojnar et al.
(a)
800
WVL: 210 nm
mAU
%C: 0.0%
2 ꢂ 15.188
[M ꢂ 2Ac ꢀ 3H]3ꢀ
95.0
1193.7833
600
400
200
0
[M ꢀ 3H]3ꢀ
*
1221.4549
1125.4270
1 ꢂ 14.807
[M ꢀ 2H]2ꢀ
1831.6709
1000
1100
1200
1300
1000
1500
2000
MeCN/TFA: 5.0%
Flow: 1.000 mL minꢂ1
min
ꢂ200
5.0
10.0
15.0
20.0
25.0
(b)
500
WVL: 210 nm
mAU
%C: 0.0%
1 ꢂ 9.574
[M ꢀ 3H]3ꢀ
885.3519
95.0
[M ꢀ H ꢀ Na]2ꢀ
1338.4077
200
0
2 ꢂ 9.862
800
1000
1200
1400
MeCN/TFA: 5.0%
Flow: 1.000 mL minꢂ1
min
25.0
ꢂ300
5.0
10.0
15.0
20.0
Fig. 1. Analytical data for (a) the crude, protected AFGP8 peptide 12 and (b) the final AFGP8 peptide 1. MS data given in the insets.
*Unidentified by-product.
therefore monitored by analytical RP-HPLC (isocratic flow,
60% acetonitrile in water, with 0.1% TFA). Pleasingly, the
reaction proceeded with high a-selectivity, to give the azido-
glycoside 5 in 72% yield after purification.
The subsequent reductive acetylation of the azide group of
the glycosylated threonine intermediate 5 has previously[13]
been performed using thioacetic acid and pyridine,[17] a reaction
that although high yielding, involves an unpleasant workup. In
the present work, the azide was reductively acetylated using zinc
powder in THF/acetic anhydride/acetic acid (with addition
of CuSO4 to activate the zinc) in 86% yield to give Fmoc-
[D-GalNAc(OAc3)(a1-O)]Thr-O-Allyl (6).[18]
four steps. At this point, if the Tn antigen building block is
desired, the building block can be deallylated to give a free acid
ready for coupling in SPPS.[13] Alternatively, further manipula-
tions can be performed to form more complex di- and
oligosaccharides.
For the purpose of the AFGP8 probe, the disaccharide Gal-b-
(1-3)-GalNAc was required, hence a second glycosylation
reaction was performed. Compound 6 was transformed into a
glycosyl acceptor suitable for glycosylation at the 3-OH position
by removal of the acetates, followed by selective protection of
the 4 and 6 hydroxyl groups. Deprotection of acetates was first
attempted through the use of NaOMe/MeOH solution as pre-
viously reported.[8a,10b,19] Unfortunately, some loss of the Fmoc
protecting group was observed, as well as significant transester-
ification to the threonine methyl ester, resulting in poor yields
Our second generation synthetic protocol has resulted in a
significant improvement in the yield of intermediate 6 – from
4.9% over seven steps[13] to an improved yield of 42% over only

RESEARCH FRONT
Synthesis of Isotopically-labelled AFGP8 (1)
727
220
mAU
WVL: 210 nm
Fmoc-protected building block 2 in 95% yield for this step, and
in 12.3% overall yield from D-galactosamine.
The disaccharide building block 2 was then utilized in Fmoc
solid-phase glycopeptide synthesis (Scheme 2). The synthesis
was performed on aminomethyl resin (prepared[25] in house),
to which an Fmoc-Ala-HMPBA linker was coupled using DIC.
Chain elongation proceeded using a CEM Liberty microwave
peptide synthesizer. For the non-glycosylated amino acids
(Pro/15N-Ala), the coupling was carried out using HBTU in
150
t ꢁ 3h
t ꢁ 2h
t ꢁ 1h
5
100
50
0
4
3
t ꢁ 30min
2
i
the presence of Pr2EtN; for incorporation of the glycosylated
t ꢁ 0
1
building block 2 (1.25 equiv.), HATU/HOAt with 4-N,N-
(dimethylamino)pyridine and collidine was used, as developed
by Kowalczyk et al.[26] A Kaiser test after coupling of the
building block revealed that despite the low excess of reagent,
complete reaction was observed after 20 min reaction time in
the microwave synthesizer. In all cases, the Fmoc group was
removed using 20% piperidine in DMF. When chain elongation
was complete, the peptide was cleaved from the resin with 95%
TFA, with TIPS and H2O (2.5% each) as scavengers, giving the
desired crude peptide in high purity (Fig. 1a). The major by-
product was determined to be the desired peptide but with a loss
of two acetate groups.
Finally, the acetate protecting groups of the sugar hydroxyl
groups were removed using NaOMe in MeOH (pH ,9). The
reaction was monitored by analytical RP-HPLC (Fig. 2) until
completion (approximately 3 h). The solution was then carefully
neutralized by addition of 1 M HCl, and lyophilized from 1:1
acetonitrile/water to give AFGP8 as a white powder. As can be
seen in Fig. 1b, the final product was of sufficient purity for
biological testing and hence was not purified further.
min
17.0
7.0
10.0
12.0
14.0
Fig. 2. LC data with detection at 210 nm for the deacetylation reaction.
for this step. Consequently, an acidic deprotection method was
employed using 50 mol-% AcCl.[20] Although the reaction was
much slower, no Fmoc cleavage was observed, and transester-
ification to the methyl ester was only observed after 3 days when
the reaction was essentially complete. The resultant triol was
then selectively protected using p-anisaldehyde dimethyl acetal
under p-toluenesulfonic acid (pTsOH) catalysis[10b] to afford 7
in 70% yield over two steps.
With the p-methoxybenzyl (Pmb) protected glycosyl accep-
tor 7 in hand, the Helferich glycosylation procedure (galactosyl
bromide with Hg(CN)2) was initially attempted.[10b] Unfortu-
nately, despite considerable effort, the reaction failed to produce
the desired disaccharide, with only starting material being iso-
lated. Consequently, attention turned to the use of the Schmidt
glycosylation,[21] with galactosyl trichloroacetimidate 8 as the
galactosyl donor and trimethylsilyl trifluoromethanesulfonate
(TMSOTf) as the Lewis acid promoter.
The synthesis of imidate 8 required peracetylation of galac-
tose, followed by the selective deprotection of the anomeric
hydroxyl to give 2,3,4,6-tetra-O-acetyl-D-galactopyranose (9),
which can be further functionalized to provide 8. Although
many routes to 1-hydroxy sugars are available,[22] all suffer from
low selectivity and yield, and often require anhydrous condi-
tions, toxic reagents, or harsh bases. Recently, Bhaumik et al.[23]
reported a simple and mild method for anomeric deprotection
using copper(II) acetate and methanol/water (9:1). Although the
selectivity is still somewhat poor, the simplicity and mildness
of the method makes this an attractive alternative. Using this
procedure, the 2,3,4,6-tetra-O-acetyl-D-galactopyranose (9) was
prepared from D-galactose in 56% yield over two steps. Sub-
sequent reaction with trichloroacetonitrile in the presence of
K2CO3[24] afforded the imidate (8) in 71% yield.
Acceptor 7 was coupled with the imidate glycosyl donor 8
using TMSOTf as activator, and the reaction was complete
within 2 h, providing the b-(1-3)-linked disaccharide 10
exclusively (63% yield). The use of excess Lewis acid was
avoided, as this leads to partial cleavage of the Pmb protecting
group and subsequent undesirable glycosylations at the 4 and 6
positions.
Conclusion
In summary, the isotopically-labelled AFGP8 probe was syn-
thesized in good yield and purity, and is currently being tested
in vivo in Antarctic fish; biological results will be published at a
later date. Importantly, the synthetic protocol developed during
this study provides facile access to both the Tn and TF antigen
building blocks for the SPPS of glycopeptides of clinical
relevance.
Experimental
General
All reagents were purchased as reagent grade and used without
further purification. Solvents were dried and purified before use
according to standard methods.[27] Solvents for RP HPLC were
purchased as HPLC grade and used without further purification.
15N-labelled Fmoc-Ala was purchased from Cambridge Isotope
Laboratories, Inc. Analytical TLC was performed using 0.2 mm
plates of Kieselgel F254 (Merck). Compounds were visualized
by UV fluorescence and/or by staining with 5% H2SO4/EtOH
and heating. Flash chromatography was carried out using Kie-
selgel F254 S 63–100 mm silica gel with the solvents indicated.
Analytical RP HPLC was performed on a Dionex Ultimate 3000
System or a Dionex P680 System using an analytical column
˚
Finally, several protecting group manipulations were
required to convert 10 into a suitable building block for SPPS.
The Pmb group was removed with acetic acid, and the emergent
hydroxyl groups were acetylated with a mixture of acetic
anhydride in pyridine to give 11 in 70% yield over two steps.
Lastly, deallylation using tetrakis(triphenylphosphine)palla-
dium(0) with N-methylaniline as scavenger afforded the desired
(Phenomenex Gemini C18, 110A, 150 mm ꢂ 2.0 mm, 5 mm) at a
flow rate of 1 mL minꢀ1. Analytical HPLC analysis of peptides
was performed using a linear gradient (5–95%) of buffer B
in buffer A over 15 min (buffer A ¼ 0.1% trifluoroacetic acid
(TFA) in water; buffer B ¼ 0.1% TFA in acetonitrile). Optical
rotations were determined at the sodium D line (589 nm) at
208C with a Perkin–Elmer 341 polarimeter. IR spectra were

RESEARCH FRONT
728
J. M. Wojnar et al.
obtained using a Perkin–Elmer Spectrum 100 FT-IR spectro-
1
N-(9-Fluorenylmethoxycarbonyl)-3-O-(3,4,6-tri-O-acetyl-
2-azido-2-deoxy-a-D-galactopyranosyl)-L-threonine
Allyl Ester 5
meter. H NMR (400 MHz) and 13C NMR (100 MHz) spectra
were recorded on a Bruker AVANCE 400 spectrometer. All
samples were run in CDCl3 and the chemical shifts are reported
in ppm relative to the tetramethylsilane signal recorded at dH
0.00. The 13C values were referenced to the residual chloroform
signal at dC 77.00. Mass spectra were recorded on a Bruker
micrOTOFQII mass spectrometer coupled with a Dionex
Ultimate 3000 HPLC with autosampler. Carrier phase consisted
of 50:50 acetonitrile/water containing 0.1% formic acid at a
flow rate of 0.1 mL minꢀ1. For HRMS, external mass calibration
was performed with Agilent Tunemix. Data were acquired for
3 min and averaged over the region of maximum sample-ion
intensity.
The glycosyl donor 1,3,4,6-tetra-O-acetyl-2-azido-2-deoxy-D-
galacto-pyranose (4) (1.86 g, 5 mmol) and the glycosyl acceptor
Fmoc-threonine allyl ester (3) (2.0 g, 5.2 mmol) were dried on
the high vacuum for several hours. Anhydrous DCM (15 mL)
was added, and the reaction flask was cooled in ice. BF3ꢁEt2O
(3 mL, 5 equiv.) was added slowly, then the reaction was heated
with stirring at 408C under an atmosphere of N2. Periodically,
5-mL samples were removed from the reaction, and a mini
workup was performed: the sample was diluted with 500 mL of
DCM, and washed with 500 mL NaHCO3. The organic layer was
separated and evaporated, then re-dissolved in 200 mL of 50:50
ACN/H2O for analysis by HPLC. After 78 h, the reaction was
judged complete and cooled to RT. A saturated solution of
NaHCO3 was added to quench the reaction, and the organic
layer was diluted with DCM. The organic layer was washed
with more NaHCO3 solution, then brine, then H2O, dried over
MgSO4, filtered, and concentrated under reduced pressure. The
dark residue was purified by silica flash chromatography (2%
MeOH/DCM) to yield 2.48 g (72%) of compound 5 as a pale
yellow solid; Rf 0.45 (1:2 EtOAc/hexane). 1H NMR data was in
agreement with that reported in the literature.[29]
N-(9-Fluorenylmethylcarbonyl)-L-threonine Allyl Ester 3
L-Threonine (24.0 g, 201.4 mmol) was suspended in a mixture of
10% aqueous Na2CO3 (330 mL) and 1,4-dioxane (150 mL) with
stirring in an ice-bath. Fmoc-OSu (71.44 g, 211.5 mmol) was
dissolved in 1,4-dioxane (150 mL) with gentle heating and the
solution was added dropwise to the L-threonine solution. The
reaction mixture was allowed to warm to room temperature
overnight and the solvent was subsequently removed under
reduced pressure. The suspension was diluted with H2O and
extracted three times with EtOAc. The aqueous phase was
acidified with citric acid (approximately 45 g) to pH 3.5 (pH
paper). Filtration and lyophilization yielded 61.1 g (89%) of the
Fmoc-threonine as a white solid that was used in the next step
without further purification.
N-(9-Fluorenylmethoxycarbonyl)-O-(2-acetamido-
3,4,6-tri-O-acetyl-2-deoxy-a-D-galactopyranosyl)-
L-threonine Allyl Ester 6
To a solution of compound 5 (1 g, 1.5 mmol) in 50 mL of THF/
Ac2O/AcOH solution (3:2:1) was added zinc dust (1.64 g,
25 mmol) and 2.5 mL of sat. aq. CuSO4, which served to activate
the zinc. After 30 min of stirring, the solution was filtered
and concentrated under reduced pressure. Purification by flash
chromatography (2:1 EtOAc/hexane) yielded compound 6
(860.5 mg, 84%) as a pale yellow oil; Rf 0.55 (4:1 EtOAc/
hexane). 1H NMR data was in agreement with that reported in
the literature.[29]
A solution of N-(9-fluorenylmethoxycarbonyl)-L-threonine
(2.34 g, 6.84 mmol) and NaHCO3 (0.58 g, 6.93 mmol) in H2O
(22 mL) was added to a solution of tetrabutylammonium bro-
mide (2.16 g, 6.71 mmol) and allyl bromide (3.12 mL, 3.90 g,
36.05 mmol) in DCM (40 mL). The suspension was stirred
vigorously at room temperature for 20 h under nitrogen. H2O
(50 mL) was added to the reaction mixture and the suspension
was extracted with DCM (3 ꢂ 30 mL). The combined organic
layers were dried over Na2SO4, filtered, concentrated under
reduced pressure and the crude product (5.43 g) was purified by
flash chromatography (2:3-3:2-1:1 EtOAc/hexane) to yield
2.57 g (98%) of compound 3 as a white solid (lyophilized from
N-(9-Fluorenylmethoxycarbonyl)-O-(2-acetamido-
2-deoxy-4,6-O-p-methoxybenzylidine-a-D-
galactopyranosyl)-L-threonine Allyl Ester 7
1
tBuOH); Rf 0.51 (1:1 EtOAc/hexane). H NMR data was in
agreement with that reported in the literature.[28]
Compound 6 (1.0 g, 1.4 mmol) was dissolved in 10 mL of
MeOH, and 50 mL (0.7 mmol) of AcCl was added, and the
reaction was stirred at 408C for 36 h. The reaction was quenched
by the addition of NaHCO3 (sat.) solution, and diluted with
DCM. The aqueous layer was extracted with DCM (2ꢂ), and
the organic layers were dried over MgSO4, filtered, and con-
centrated under reduced pressure, to give 794.6 mg of the crude
deprotected compound. This material was then dissolved in
10 mL of DMF, and 464 mL (2.7 mmol) of p-anisaldehyde
dimethyl acetal was added. A catalytic amount (approximately
20 mg) of pTsOH was added, and the reaction was stirred on
the rotary evaporator at 558C and ,20 mbar. After 2 h, a further
464 mL (2.7 mmol) of p-anisaldehyde dimethyl acetal was
added. After 4 h, the reaction was neutralized by the addition
1,3,4,6-Tetra-O-acetyl-2-azido-2-deoxy-
D-galactopyranose 4
Imidazole-1-sulfonyl azide hydrochloride[16] (1.2 equiv., 3.27 g,
15.6 mmol) was added to D-galactosamine hydrochloride
(2.72 g, 12.6 mmol), K2CO3 (4.80 g, 34.7 mmol), and
CuSO4ꢁH2O (31.5 mg, 126 mmol) in MeOH/water (5:3, 80 mL)
and the mixture stirred at room temperature for 2.5 h. The
mixture was co-evaporated with toluene and filtered through a
pad of CeliteÒ (Rf of 2-azido-2-deoxy-D-galactose intermediate:
0.43 (0.5:2.5:7 AcOH/MeOH/DCM)). Acetic anhydride
(50 mL) was added to the residue in dry NEt3 (25 mL) and
the mixture stirred for 3 h. The mixture was concentrated under
reduced pressure and co-evaporated with toluene. Purification
by flash chromatography (1:2 EtOAc/hexane), afforded 3.04 g
of compound 4 as a yellow oil in 68% yield (ratio of a:b 1:3);
i
of 5 mL of Pr2EtN, and the yellow solution was partitioned
between EtOAc/H2O. The aqueous layer was washed with
EtOAc, and the combined organic layers were dried over
MgSO4, filtered, and concentrated under reduced pressure. The
residue was dissolved in MeOH/DCM, silica gel was added, and
the whole dried to a powder. This powder was then loaded onto a
1
Rf 0.45 (1:1 EtOAc/hexane). H NMR data was in agreement
with that reported in the literature.[15a]

RESEARCH FRONT
Synthesis of Isotopically-labelled AFGP8 (1)
729
flash column, and eluted with 3:1 EtOAc/hexane to yield 690 mg
(70% over two steps) of the Pmb protected glycosylated amino
acid 7 as a white solid; Rf 0.23 (3:1 EtOAc/hexane). [a]D þ40.6
(c 0.14 in CHCl3). nmax(film)/cmꢀ1 3323br (O-H), 3068w
(Ar C-H), 2937m (C-H), 1721s (C¼O), 1653s (vinyl C¼C),
1520s (Ar C¼C), 1449m (C-O-H), 1249s (C-O-C), 1045s (C-O),
742m (Ar C-H). dH (400 MHz) 7.77 (d, J 7.7, 2H, Fmoc ArcH),
7.63 (d, J 6.9, 2H, Fmoc ArCH), 7.44 (d, J 8.7, 2H, Pmb ortho),
7.40 (d, J 7.5, 2H, Fmoc ArCH), 7.31 (t, J 7.0, 2H, Fmoc ArCH),
6.87 (d, J 8.7, 2H, Pmb meta), 5.86 (ddt, J 16.9, 10.5, 6.0, 1H,
allyl CH), 5.50 (s, 1H, Pmb acetal), 5.33 (dd, J 17.5, 1.1, 1H,
allyl ¼CH2), 5.27 (dd, J 9.9, 1.1, 1H allyl ¼CH2), 4.94 (d, J 3.5,
1H, H1), 4.65 (dd, J 12.9, 5.8, 1H, allyl CH2), 4.58 (dd, J 12.9,
5.9, 1H, allyl CH2), 4.44 (m, 1H, Thr Ha), 4.42 (m, 2H, Fmoc
CH2), 4.41 (m, 1H, H2), 4.34 (qd, J 6.3, 1.5, 1H, Thr Hb), 4.25
(m, 1H, Fmoc CH), 4.21 (m, 1H, H6a), 4.15 (m, 1H, H4), 4.03 (d,
J 12.5, 1H, H6b), 3.87 (m, 1H, H3), 3.71 (br s, 1H, H5), 3.79 (s,
3H, Pmb OMe), 2.06 (s, 3H, NAc), 1.29 (d, J 6.3, 3H, Thr Hg).
dC (100 MHz) 171.7 (NAc), 170.8 (Thr C¼O), 160.1 (Pmb
para), 156.9 (Fmoc urethane), 143.7, 141.3 (Fmoc C), 131.0
(allyl CH), 130.1 (Pmb ipso), 127.74 (Pmb ortho), 127.69,
127.0, 125.1, 120.0 (Fmoc ArCH), 119.7 (allyl ¼CH2), 113.5
(Pmb meta), 101.1 (Pmb acetal), 100.5 (C1), 76.3 (Thr Cb), 75.6
(C4), 69.2 (C6), 68.3 (C3), 67.1 (Fmoc CH2), 66.3 (allyl CH2),
63.4 (C5), 58.8 (Thr Ca), 55.3 (Pmb OMe), 50.2 (C2), 47.2
(Fmoc CH), 23.1 (NAc), 18.9 (Thr Cg). m/z (HR-ESI) 703.2851
[M þ H]þ. C38H43N2O11 requires 703.2861, D1.0 ppm.
N-(9-Fluorenylmethoxycarbonyl)-O-(2-acetamido-2-
deoxy-4,6-O-p-methoxybenzylidine-3-O-[2,3,4,6-tetra-
O-acetyl-b-D-galactopyranosyl]-a-D-galactopyranosyl)-
L-threonine Allyl Ester 10
The glycosyl acceptor 7 (997.4 mg, 2 mmol) and glycosyl donor
8 (690.2 mg, 0.98 mmol) were combined and dried on the high
vacuum for 3 h. They were dissolved in anhydrous DCM, and
˚
4 A molecular sieves (spatula tip) were added. The reaction flask
was cooled in an ice/salt bath to approximately –108C. TMSOTf
(45 mL, 0.25 mmol) was then added to the flask. The reaction
was allowed to warm to RT. After 2 h, the reaction was quenched
by the addition of 100 mL of Et3N, was filtered, and concentrated
to dryness under reduced pressure. Purification by silica flash
chromatography (3:1 EtOAc/hexane) yielded 637 mg (63%) of
disaccharide 10 as a white amorphous solid; Rf 0.31 (3:1 EtOAc/
hexane). [a]D þ63.0 (c 0.092 in CHCl3). nmax(film)/cmꢀ1
3352m, 3290m, 3186m (N-H), 2927m, 2852m (C-H), 1722s
(C¼O), 1601m (C¼C), 1371m, 1226s (C-O), 1046s (C-O-C),
827s (Ar C-H). dH (400 MHz) 7.77 (d, J 7.3, 2H, Fmoc ArCH),
7.44 (d, J 8.8, 2H, Pmb ortho), 7.40 (m, 2H, Fmoc ArCH), 7.61
(br d, J 6.2, 2H, Fmoc ArCH), 7.31 (td, J 7.6, 2.4, 2H, Fmoc
ArCH), 6.88 (d, J 8.7, 2H, Pmb meta), 5.85 (m, 1H, allyl CH),
5.73 (d, J 8.6, 1H, GalNAc NH), 5.67 (d, J 9.6, 1H, Thr NH),
5.49 (s, 1H, Pmb acetal), 5.37 (dd, J 3.4, 1.0, 1H, Gal H4),
5.33 (br d, J 17.1, 1H, allyl ¼CH2), 5.28 (dq, J 10.4, 1.1, 1H,
allyl ¼CH2), 5.17 (dd, J 10.5, 8.0, 1H, Gal H2), 4.96 (m, 1H, Gal
H3), 4.95 (d, J 2.9, GalNAc H1), 4.73 (d, J 7.8, 1H, Gal H1), 4.64
(m, 1H, allyl CH2), 4.60 (m, 1H, GalNAc H2), 4.55 (m, 1H, allyl
CH2), 4.51 (m, 2H, Fmoc CH2), 4.40 (d, J 10.5, 1H, Thr Ha),
4.27 (m, 1H, Thr Hb), 4.26 (m, 1H, GalNAc H4), 4.24 (m, 1H,
Fmoc CH), 4.21 (m, 1H, Gal H6a), 4.19 (m, 1H, GalNAc H6a),
4.11 (m, 1H, Gal H6b), 4.02 (br d, J 12.0, 1H, GalNAc H6b),
3.89 (m, 1H, Gal H5), 3.87 (m, 1H, GalNAc H3), 3.79 (s, 3H,
Pmb OMe), 3.64 (br s, 1H, GalNAc H5), 2.13, 2.03, 2.00, 1.99,
1.96 (s, 15H, 5ꢂ Ac), 1.26 (3H, Thr Hg). dC (100 MHz) 170.5,
170.4, 170.2, 170.1, 169.8, 169.4 (5ꢂ Ac, 1ꢂ C¼O), 159.9 (Pmb
para), 156.5(Fmocurethane), 143.5, 141.3 (Fmoc C), 130.8(allyl
CH), 130.1 (Pmb ipso), 127.8 (Fmoc ArCH), 127.5 (Pmb ortho),
127.0, 124.8 (Fmoc ArCH), 120.0 (allyl ¼CH2), 119.9 (Fmoc
ArCH), 113.5 (Pmb meta), 100.9 (Gal C1), 100.6 (Pmb acetal),
100.4 (GalNAc C1), 76.4 (Thr Cb), 75.4 (GalNAc C4), 73.8
(GalNAc C3), 70.9 (Gal C5), 70.8 (Gal C3), 69.0 (Gal C2), 68.8
(GalNAc C6), 67.0 (Gal C4), 67.0 (Fmoc CH2), 66.3 (allyl CH2),
63.5 (GalNAc C5), 61.3 (Gal C6), 58.6 (Thr Ca), 55.2 (Pmb
OMe), 48.0 (GalNAc C2), 47.2 (Fmoc CH), 23.3, 21.0, 20.6 (ꢂ2),
20.5 (5ꢂ Ac), 18.6 (Thr Cg). m/z (HR-ESI) 1033.3839 [Mþ H]þ.
C52H61N2O20 requires 1033.3812, D2.7 ppm.
2,3,4,6-Tetra-O-acetyl-D-galactopyranose 9
To a stirred suspension of D-galactose (500 mg, 2.7 mmol)
in Ac2O (20 mL) was added dropwise 60% perchloric acid
(120 mL). Additional D-galactose (4.5 g, 25.0 mmol) was added
in small portions. During the additions, the temperature was kept
below 408C by occasional cooling in an ice-water bath. After the
addition was complete, the reaction was stirred for 2 h at room
temperature. The resulting mixture was then diluted with DCM
and water, and washed twice with NaHCO3 (sat) solution. The
organic layer was dried over MgSO4 and concentrated. The
crude material (10.92 g) was then dissolved in 150 mL MeOH/
H2O (9:1). Copper (II) acetate dihydrate (5.6 g, 12.8 mmol) was
added, and the reaction was heated until reflux for 6 h (mon-
itored by TLC). The reaction was cooled, filtered, and the filtrate
was concentrated under reduced pressure. The dark residue was
diluted with water and extracted twice with EtOAc. Flash
chromatography (1:1 EtOAc/hexane) yielded 4.67 g (56%) of 9
as a white foam (ratio of a:b 5:1); Rf 0.20 (1:1 EtOAc/hexane).
1H NMR data was in agreement with that reported in the
literature.[30]
N-(9-Fluorenylmethoxycarbonyl)-O-(2-acetamido-4,6-
di-O-acetyl-2-deoxy-3-O-[2,3,4,6-tetra-O-acetyl-b-D-
galactopyranosyl]-a-D-galactopyranosyl)-L-threonine
Allyl Ester 11
2,3,4,6-Tetra-O-acetyl-D-galactopyranosyl
Trichloroacetimidate 8
Compound 9 (3.27 g, 9.4 mmol) was dried on the high vacuum,
and then dissolved in 40 mL of anhydrous DCM. To this solution
was added K2CO3 (3.90 g, 28.2 mmol) and trichloroacetonitrile
(3.77 mL, 37.6 mmol) and the reaction was stirred at room
temperature under an inert atmosphere overnight. The reaction
mixture was then filtered and concentrated under reduced
pressure. Subsequent purification by flash chromatography (2:3
EtOAc/hexane with 0.5% Et3N) afforded the imidate 8 (3.27 g,
71%) as a white foam (a/b 1:1); Rf 0.65 (a) and 0.34 (b) (1:1
EtOAc/hexane). 1H NMR data was in agreement with that
reported in the literature.[19,31]
Compound 10 (614.6 mg, 0.595 mmol) was dissolved in 10 mL
80% HOAc and the solution was stirred at room temperature for
1.5 h. The solution was neutralized with a solution of NaHCO3
(sat) and diluted with EtOAc. The aqueous layer was further
extracted with EtOAc (2ꢂ) and the combined organic layers
were washed with NaHCO3 (sat) solution. The organic phase
was dried over MgSO4 and concentrated. The resulting solid was
dissolved in 10 mL of 1:1 pyridine/Ac2O, and the reaction
was left to stir at room temperature overnight. The reaction was
quenched by the addition of ice, and was then diluted with DCM.

RESEARCH FRONT
730
J. M. Wojnar et al.
The aqueous layer was extracted twice more with DCM, and the
combined organic phases were washed twice with NaHCO3
(sat.), then with brine, dried over MgSO4 and concentrated.
Flash chromatography (3:1 EtOAc/hexane) afforded 556.0 mg
(0.557 mmol, 94% over two steps) of 11 as a clear glass;
Rf 0.30 (3:1 EtOAc/hexane). [a]D þ62.6 (c 0.099 in CHCl3).
143.7, 141.3 (Fmoc C), 127.8, 127.1, 124.8, 120.0 (Fmoc
ArCH), 100.4 (Gal C1), 99.5 (GalNAc C1), 77.2 (Thr Cb), 72.5
(GalNAc C3), 70.7 (Gal C3), 70.6 (Gal C5), 68.79, 68.75 (Gal
C2, Gal C4), 67.7 (GalNAc C5), 66.8 (GalNAc C4), 66.7 (Fmoc
CH2), 62.8 (GalNAc H6), 60.9 (Gal C6), 58.5 (Thr Ca), 49.1
(GalNAc C2), 47.2 (Fmoc CH), 22.7, 20.57 (ꢂ2), 20.62 (ꢂ2),
20.5, 20.0 (7ꢂ Ac), 18.3 (Thr Hg). m/z (HR-ESI) 981.3109
[M þ H]þ. C45H54N2NaO21 requires 981.3111, D1.1 ppm.
n
max(film)/cmꢀ1 3335m (N-H), 2981w, 2941w (C-H), 1747s
(C¼O), 1663m (vinyl C¼C), 1536w (Ar C¼C), 1372m, 1227s
(C-O), 743m (Ar C-H). dH (400 MHz) 7.79 (d, J 7.5, 2H, Fmoc
ArCH), 7.61 (d, J 7.7, 2H, Fmoc ArCH), 7.42 (t, J 7.4, 2H, Fmoc
ArCH), 7.34 (t, J 6.9, 2H, Fmoc ArCH), 5.87 (m, 1H, allyl CH),
5.78 (d, J 9.2, 1H, GalNAc NH), 5.54 (d, J 9.6, 1H, Thr NH),
5.37 (m, 1H, Gal H4), 5.36 (m, 1H, GalNAc H4), 5.32 (m, 2H,
allyl ¼CH2), 5.10 (dd, J 10.5, 7.9, 1H, Gal H2), 4.95 (dd, J 10.3,
3.0, 1H, Gal H3), 4.86 (d, J 3.1, 1H, GalNAc H1), 4.67 (dd, J
12.4, 6.0, 1H, allyl CH2), 4.59 (m, 1H, allyl CH2), 4.58 (m, 1H,
Gal H1), 4.53 (m, 2H, Fmoc CH2), 4.48 (m, 1H, GalNAc H2),
4.41 (d, J 9.8, 1H, Thr Ha), 4.26 (m, 1H, Fmoc CH), 4.25 (m, 1H,
Thr Hb), 4.15 (m, 1H, Gal H6b), 4.16 (m, 2H, GalNAc H6), 4.12
(m, 1H, Gal H5), 3.98 (dd, J 11.0, 7.1, 1H, Gal H6a), 3.88 (t, J
6.5, 1H, GalNAc H5), 3.81 (dd, J 10.5, 2.7, 1H, GalNAc H3),
2.16, 2.13, 2.06, 2.02, 2.02, 2.01, 1.97 (s, 21H, 7ꢂ Ac), 1.31 (d, J
6.2, 3H, Thr Hg). dC (100 MHz) 171.1, 170.4 (ꢂ2), 170.3,
170.14, 170.10, 169.9, 169.6 (7ꢂ Ac, 1ꢂ C¼O), 156.4 (Fmoc
urethane), 143.5, 141.3 (Fmoc C), 130.8 (allyl CH), 127.8, 127.1,
124.8, 120.1 (Fmoc ArCH), 120.0 (allyl ¼CH2), 100.7 (Gal C1),
100.1 (GalNAc C1), 77.2 (Thr Cb), 73.0 (GalNAc C3), 70.73
(GalNAc C5), 70.70 (Gal C3), 68.8 (Gal C2), 68.7 (Gal C4), 67.9
(Gal C5), 67.0 (Fmoc CH2), 66.7 (GalNAc C4), 66.4 (allyl CH2),
62.9 (Gal C6), 61.0 (GalNAc C6), 58.6 (Thr Ca), 48.6 (GalNAc
C2), 47.2 (Fmoc CH), 23.3, 20.8 (ꢂ2), 20.7 (ꢂ3), 20.5 (7ꢂ Ac),
18.3 (Thr Cg). m/z 999.3633 (HR-ESI) [M þ H]þ. C48H59N2O21
requires 999.36055, D1.8 ppm.
Solid-phase Synthesis of Protected AFGP8 12
Solid-phase synthesis of AFGP8 was carried out on Fmoc-
Ala-Wang polystyrene resin (1.178 mmol gꢀ1 loading,
0.1 mmol scale), prepared by coupling Fmoc-Ala-O-CH2-Ph-
OCH2CH2COOH (2 equiv.), using N,N0-diisopropylcarbodi-
imide (DIC) (2 equiv.) in DCM, 2 h at RT to aminomethyl resin
which had been prepared according to published protocol.[25]
The peptide was assembled with the use of a CEM Liberty
microwave synthesizer using the following conditions: (a) Fmoc
removal: 20% piperidine/DMF, MW 758C, 30 s then 3 min;
(b) Fmoc-Pro coupling: Fmoc-Pro (5 equiv.), 2-(1H-benzo-
triazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate
i
(HBTU) (4.5 equiv.), Pr2EtN (10 equiv.) in DMF, MW 758C,
5 min, repeated twice (double coupling); (c) 15N-Ala
i
(1.5 equiv.), HBTU (1.45 equiv.), Pr2EtN (29 equiv.) in DMF,
MW 758C, 15 min; (d) compound 2 (1.25 equiv.), 2-(7-aza-1H-
benzotriazole-1-yl)-1,1,3,3-tetramethyluronium
hexafluoro-
phosphate (HATU) (1.2 equiv.), 1-hydroxy-7-azabenzotriazole
(HOAt) (1.25 equiv.), collidine (4.5 equiv.) in DMF, MW 758C,
20 min. The peptide was cleaved from the resin using TFA/
triisopropylsilane (TIPS)/H2O (95:2.5:2.5), precipitated using
Et2O, and isolated by centrifugation. The resulting pellet was
dissolved in 50:50 ACN/H2O (with 0.1% TFA) and lyophilized,
to give the protected AFGP8 peptide (12) as a fluffy powder
(206 mg, 57% crude yield). m/z (ESI-MS) 1221.4 [M þ 3H]3þ
,
N-(9-Fluorenylmethoxycarbonyl)-O-(2-acetamido-4,6-
di-O-acetyl-2-deoxy-3-O-[2,3,4,6-tetra-O-acetyl-b-D-
galactopyranosyl]-a-D-galactopyranosyl)-L-threonine 2
requires 1221.8.
Removal of Acetate Protecting Groups
Compound 11 (556.0 mg, 0.56 mmol) was dried on the high
vacuum for 2 h, then dissolved in 10 mL anhydrous THF and
degassed with argon for 30 min. N-Methylaniline (60 mL,
0.56 mmol) then Pd(PPh3)4 (6.5 mg, 0.056 mmol) were added,
and then the reaction was stirred at RT under argon. After
completion of the reaction (2 h), the sample was concentrated,
and purified by flash chromatography (1:1 MeOH/EtOAc) to
yield 506 mg (95%) of 2 as a film; Rf 0.1 (0.1% HOAc/EtOAc),
0.6 (1:1 MeOH/EtOAc). [a]D þ127.4 (c 0.11 in CHCl3).
The protected AFGP8 peptide 12 (50 mg, 0.014 mmol) was
dissolved in 10 mL MeOH. Freshly prepared 1 M NaOMe/
MeOH solution was added until pH ,9–10. After 3 h at RT,
the reaction was neutralized by addition of 1 M HCl, and con-
centrated to dryness. The residue was dissolved in 50:50 ACN/
H2O, and lyophilized to give AFGP8 (1) as a white powder
(42.3 mg, 117% apparent yield due to NaCl salts). m/z (ESI-MS)
885.4 [M þ 3H]3þ, requires 885.6.
n
max(film)/cmꢀ1 3355br (O-H), 3072w (Ar C-H), 2930w (C-H),
Acknowledgements
1749s (C¼O), 1658m (C¼C), 1372m, 1228s (C-O), 1060s (C-
O-C), 742m (Ar C-H). dH (400 MHz) 7.41 (t, J 7.3, 2H, Fmoc
ArCH), 7.32 (t, J 7.3, 2H, Fmoc ArCH), 7.63 (d, J 6.7, 2H, Fmoc
ArCH), 7.79 (d, J 7.5, 2H, Fmoc ArCH), 4.58 (d, J 7.5, 1H, Gal
H1), 5.00 (d, J 3.0, 1H, GalNAc H1), 4.40 (m, 1H, Thr Hb), 3.86
(m, 1H, GalNAc H3), 4.95 (dd, J 10.5, 3.4, 1H, Gal H3), 3.86 (m,
1H, Gal H5), 5.08 (dd, J 10.2, 7.9, 1H, Gal H2), 5.37 (br d, J 2.2,
1H, Gal H4), 4.15 (m, 1H, GalNAc H5), 5.35 (d, J 3.0, 1H,
GalNAc H4), 4.51 (m, 2H, Fmoc CH2), 4.00 (m, 2H, GalNAc
H6), 4.17 (m, 1H, Gal H6a), 4.12 (m, 1H, Gal H6b), 4.40 (m, 1H,
Thr Ha), 4.39 (m, 1H, GalNAc H2), 4.25 (m, 1H, Fmoc CH),
2.15, 2.12, 2.11, 2.05, 2.00, 1.99, 1.97 (s, 21H, 7ꢂ Ac), 1.29 (d, J
6.1, 3H, Thr Hg), 6.26 (d, J 8.9, 1H, Thr NH), 5.72 (d, J 8.9, 1H,
GalNAc NH). dC (100 MHz) 170.5, 170.4, 170.3, 170.1 (ꢂ2),
170.0, 169.9, 169.6 (7ꢂ Ac, 1ꢂ C¼O), 156.6 (Fmoc urethane),
This research was supported by the Human Frontier Science Program (grant
RGP003/2009-C). The authors thank Dr Nicole Miller for assistance in the
synthesis of tetra-O-acetyl-2-azido-2-deoxygalactose.
References
[1] (a) A. L. DeVries, Science 1971, 172, 1152. doi:10.1126/SCIENCE.
172.3988.1152
(b) R. N. Ben, ChemBioChem 2001, 2, 161. doi:10.1002/1439-7633
(20010302)2:3o161::AID-CBIC16143.0.CO;2-F
(c) M. M. Harding, P. I. Anderberg, A. D. J. Haymet, Eur. J. Biochem.
2003, 270, 1381. doi:10.1046/J.1432-1033.2003.03488.X
[2] A. L. DeVries, Annu. Rev. Physiol. 1983, 45, 245. doi:10.1146/
ANNUREV.PH.45.030183.001333
[3] J. A. Raymond, A. L. DeVries, Proc. Natl. Acad. Sci. USA 1977, 74,
2589. doi:10.1073/PNAS.74.6.2589

RESEARCH FRONT
Synthesis of Isotopically-labelled AFGP8 (1)
731
[4] C.-H. C. Cheng, P. A. Cziko, C. W. Evans, Proc. Natl. Acad. Sci. USA
2006, 103, 10491. doi:10.1073/PNAS.0603796103
[5] C. W. Evans, V. Gubala, R. Nooney, D. E. Williams, M. A. Brimble,
A. L. DeVries, Antarct. Sci. 2011, 23, 57. doi:10.1017/
S0954102010000635
[6] R. Peltier, M. A. Brimble, J. M. Wojnar, D. E. Williams, C. W. Evans,
A. L. DeVries, Chem. Sci. 2010, 1, 538. doi:10.1039/C0SC00194E
[7] M. Cudic, G. D. Burstein, Methods Mol. Biol. 2008, 494, 187.
doi:10.1007/978-1-59745-419-3_11
[8] (a) Y. Tachibana, N. Matsubara, F. Nakajima, T. Tsuda, S. Tsuda,
K. Monde, S.-I. Nishimura, Tetrahedron 2002, 58, 10213. doi:10.1016/
S0040-4020(02)01359-5
[17] T. Rosen, I. M. Lico, D. T. W. Chu, J. Org. Chem. 1988, 53, 1580.
doi:10.1021/JO00242A051
[18] K. A. Winans, D. S. King, V. R. Rao, C. R. Bertozzi, Biochemistry
1999, 38, 11700. doi:10.1021/BI991247F
[19] V. R. Krishnamurthy, A. Dougherty, M. Kamat, X. Song, R. D.
Cummings, E. L. Chaikof, Carbohydr. Res. 2010, 345, 1541.
doi:10.1016/J.CARRES.2010.05.004
[20] C.-E. Yeom, S. Y. Lee, Y. J. Kim, B. M. Kim, Synlett 2005, 1527.
[21] (a) R. R. Schmidt, J. Michel, Angew. Chem. Int. Ed. Engl. 1980, 19,
731. doi:10.1002/ANIE.198007311
(b) R. R. Schmidt, Angew. Chem. Int. Ed. Engl. 1986, 25, 212.
doi:10.1002/ANIE.198602121
(b) T. Tsuda, S.-I. Nishimura, Chem. Commun. 1996, 2779. doi:10.1039/
CC9960002779
(c)Y. Tachibana, K. Monde,S.-I. Nishimura, Macromolecules2004, 37,
6771. doi:10.1021/MA049756Z
[22] (a) A. T. Tran, S. Deydier, D. Bonnaffe, C. Le Narvor, Tetrahedron
Lett. 2008, 49, 2163. doi:10.1016/J.TETLET.2008.01.106
(b) P. Tiwari, A. K. Misra, Tetrahedron Lett. 2006, 47, 3573.
doi:10.1016/J.TETLET.2006.03.050
[9] P.-H. Tseng, W.-T. Jiaang, M.-Y. Chang, S.-T. Chen, Chemistry 2001,
7, 585. doi:10.1002/1521-3765(20010202)7:3o585::AID-CHEM5854
3.0.CO;2-6
[10] (a) S. D. Kuduk, J. B. Schwarz, X.-T. Chen, P. W. Glunz, D. Sames,
G. Ragupathi, P. O. Livingston, S. J. Danishefsky, J. Am. Chem. Soc.
1998, 120, 12474. doi:10.1021/JA9825128
[23] K. Bhaumik, P. D. Salgaonkar, K. G. Akamanchi, Aust. J. Chem. 2003,
56, 909. doi:10.1071/CH03057
[24] M. A. Brimble, R. Kowalczyk, P. W. R. Harris, R. Dunbar, V. J. Muir,
Biopolymers 2007, 88, 539.
[25] A. R. Mitchell, S. B. H. Kent, M. Engelhard, R. B. Merrifield, J. Org.
Chem. 1978, 43, 2845. doi:10.1021/JO00408A022
(b) S. Dziadek, C. Brocke, H. Kunz, Chemistry 2004, 10, 4150.
doi:10.1002/CHEM.200400228
[26] R. Kowalczyk, P. W. R. Harris, R. P. Dunbar, M. A. Brimble, Synthesis
2009, 2210.
(c) B. L. Wilkinson, L. R. Malins, C. K. Y. Chun, R. J. Payne, Chem.
Commun. 2010, 46, 6249. doi:10.1039/C0CC01360A
[11] K. F. Geoghegan, D. T. Osuga, A. I. Ahmed, Y. Yeh, R. E. Feeney,
J. Biol. Chem. 1980, 255, 663.
[12] M. A. Brimble, R. Kowalczyk, P. W. R. Harris, P. R. Dunbar, V. J.
Muir, Org. Biomol. Chem. 2008, 6, 112. doi:10.1039/B712926B
[13] D. J. Lee, P. W. R. Harris, R. Kowalczyk, P. R. Dunbar, M. A. Brimble,
Synthesis 2010, 763.
[27] W. L. F. Armarego, C. L. L. Chai, Purification of Laboratory
Chemicals, 5th edn 2003 (Elsevier: Burlington, MA).
[28] (a) R. Krishnamoorthy, L. D. Vazquez-Serrano, J. A. Turk, J. A.
Kowalski, A. G. Benson, N. T. Breaux, M. A. Lipton, J. Am. Chem.
Soc. 2006, 128, 15392. doi:10.1021/JA0666250
(b) S. Ficht, R. J. Payne, R. T. Guy, C.-H. Wong, Chemistry 2008, 14,
3620. doi:10.1002/CHEM.200701978
[29] (a) R. J. Payne, S. Ficht, S. Tang, A. Brik, Y.-Y. Yang, D. A. Case,
C.-H. Wong, J. Am. Chem. Soc. 2007, 129, 13527. doi:10.1021/
JA073653P
[14] (a) R. U. Lemieux, R. M. Ratcliffe, Can. J. Chem. 1979, 57, 1244.
doi:10.1139/V79-203
(b) O. Srinivas, B. Muktha, S. Radhika, T. N. Guru Row, N. Jayaraman,
Carbohydr. Res. 2004, 339, 1447. doi:10.1016/J.CARRES.2004.
03.025
(b) T. Vuljanic, K.-E. Bergquist, H. Clausen, S. Roy, J. Kihlberg,
Tetrahedron 1996, 52, 7983. doi:10.1016/0040-4020(96)00369-9
[30] (a) M. Filice, R. Fernandez-Lafuente, M. Terreni, J. M. Guisan,
J. M. Palomo, J. Mol. Catal., B Enzym. 2007, 49, 12. doi:10.1016/
J.MOLCATB.2007.07.002
(b) A. K. Misra, G. Agnihotri, Carbohydr. Res. 2004, 339, 885.
doi:10.1016/J.CARRES.2003.12.020
[31] P.-H. Amvam-Zollo, P. Sinay, Carbohydr. Res. 1986, 150, 199.
doi:10.1016/0008-6215(86)80016-7
(c) M. Liu, V. G. Young, S. Lohani, D. Live, G. Barany, Carbohydr.
Res. 2005, 340, 1273. doi:10.1016/J.CARRES.2005.02.029
[15] (a) A. Vasella, C. Witzig, J. L. Chiara, M. Martin-Lomas, Helv. Chim.
Acta 1991, 74, 2073. doi:10.1002/HLCA.19910740842
(b) R.-B. Yan, F. Yang, Y. Wu, L.-H. Zhang, X.-S. Ye, Tetrahedron
Lett. 2005, 46, 8993. doi:10.1016/J.TETLET.2005.10.103
[16] E. D. Goddard-Borger, R. V. Stick, Org. Lett. 2007, 9, 3797.
doi:10.1021/OL701581G