Design and synthesis of lipopeptide-carbohydrate assembled multivalent vaccine candidates using native chemical ligation

Article abstract of DOI:10.1071/CH09065

Development of a synthetic vaccine against group A streptococcal infection is increasingly paramount due to the induction of autoimmunity by the main virulent factor M protein. Peptide vaccines, however, are generally poorly immunogenic, necessitating adm

Full text of DOI:10.1071/CH09065

RESEARCH FRONT
Full Paper
CSIRO PUBLISHING
www.publish.csiro.au/journals/ajc
Aust. J. Chem. 2009, 62, 993–999
Design and Synthesis of Lipopeptide–Carbohydrate
Assembled Multivalent Vaccine Candidates
Using Native Chemical Ligation
Wei Zhong,A Mariusz Skwarczynski,A Yoshio Fujita,A Pavla Simerska,A
Michael F. Good,B and Istvan TothA,C
AThe University of Queensland, School of Chemistry and Molecular Biosciences,
St. Lucia, Qld 4072, Australia.
BQueensland Institute of Medical Research, Herston, Qld 4029, Australia.
CCorresponding author. Email: i.toth@uq.edu.au
Development of a synthetic vaccine against groupA streptococcal infection is increasingly paramount due to the induction
of autoimmunity by the main virulent factor – M protein. Peptide vaccines, however, are generally poorly immunogenic,
necessitating administration with carriers and adjuvants. One of the promising approaches to deliver antigenic peptides is
to assemble peptides on a suitable template which directs the attached peptides to form a well defined tertiary structure. For
self-adjuvanting human vaccines, the conjugation of immunostimulatory lipids has been demonstrated as a potentially safe
method. This study describes the design and optimized synthesis of two lipopeptide conjugated carbohydrate templates
and the assembling of peptide antigens. These lipopeptide–carbohydrate assembled multivalent vaccine candidates were
obtained in high yield and purity when native chemical ligation was applied. Circular dichroism studies indicated that
the template-assembled peptides form four α-helix bundles. The developed technique extends the use of carbohydrate
templates and lipopeptide conjugates for producing self-adjuvanting and topology-controlled vaccine candidates.
Manuscript received: 2 February 2009.
Manuscript accepted: Accepted without change.
Introduction
has been demonstrated as a potentially safe method to adjuvant
otherwise poorly immunogenic peptides.[9–12]
Group A streptococcal (GAS) infection is associated with a
broad range of diseases including common pharyngitis and life-
treating post-infectious diseases such as rheumatic fever (RF).[1]
The cell surface M protein, an α-helical coiled-coil protein, is
a major virulence factor in GAS infection.[2] Protective immu-
nity to GAS is mediated by serotype-specific opsonic antibodies
against the hypervariable N-terminal regions of M protein,
as well as antibodies specific to the conserved C-terminal
regions.[1,3] Development of vaccines against GAS face substan-
tial obstacles. There are more than 180 identified serotypes that
require serotype-specific opsonising antibodies for protection.[4]
Furthermore, the conserved C-terminal region contains epi-
topes which are capable of inducing autoimmunity.[5,6] These
problems may be addressed by the development of subunit
vaccineswhichcontainnon-cross-reactiveB-cellepitopes. How-
ever, subunit vaccines containing short peptides are generally
poorly immunogenic due to lack of either the correct three-
dimensional structure or an appropriate helper T-cell epitope.[7]
Theimmunogenicitycanbeenhancedbyincorporatinganappro-
priate adjuvant, which facilitates the uptake and presentation
of antigen by antigen-presenting cells.[7] Because humans are
the only host of the GAS pathogen, an effective and non-toxic
adjuvant is required. Current approved adjuvants for human use
mainly contain alum-based adjuvants.[8] This limited choice has
spawned a need to develop new adjuvants. The covalent conju-
gation of lipids (self-adjuvanting moiety) to synthetic peptides
Over the past decade, our laboratory has performed exten-
sivestudiesofsubunitvaccinecandidatesagainstGAS.[13–16] We
have also developed a unique lipid core peptide (LCP) system[17]
featuring a lipopeptide moiety (inbuilt adjuvant), a polylysine
branching core[18] (carrier), and multiple peptides (antigen) in
one molecular entity.The LCP system gives greater flexibility in
the physico-chemical properties, such as lipophilicity and mul-
tiplicity of the incorporated peptides, which can improve their
biological activities.[19] The aim of this study was to accom-
modate peptide epitopes in a spatial arrangement on a suitable
template to afford new LCPs with well-defined antigen sec-
ondary structure (Fig. 1). We have chosen carbohydrates as the
templates for assembling the antigenic peptides as the distinct
conformations and the rigid ring make them promising scaf-
folds to control the topology of the assembled peptides.[20–23]
Although such branched multivalent peptides could be synthe-
sized using stepwise solid phase peptide synthesis (SPPS),[24]
synthetic errors relating to SPPS itself and the purification
of high molecular weight multivalent peptides present a clear
challenge.[18] Therefore, we examine here the potential of
chemoselective ligation to provide highly pure vaccine candi-
dates that allow accurate conformational studies and provide
potential vaccines suitable for use in humans. Native chemi-
cal ligation (NCL) previously enabled the synthesis of several
lipopeptide-polylysine-antigen constructs and the technique,
© CSIRO 2009
10.1071/CH09065
0004-9425/09/090993

RESEARCH FRONT
994
W. Zhong et al.
Antigen
to overnight with excess of reagents. However, by in situ con-
verting allyl bromide to allyl iodide with potassium iodide and
performing the reaction at room temperature for 5 h, 2a was
obtained in 78% yield. Reduction of 2a was achieved using
stannous chloride.[31] The following coupling with adipic acid
monobenzyl ester[32] using HBTU/DIPEA method afforded 3a
in 47% overall yield (two steps). Free-radical reaction of 3a with
tert-butyl-N-(2-mercaptoethyl) carbamate (Boc-cysteamine)[33]
in 1,4-dioxane under UV irradiation at 254 nm[34] proceeded
efficiently and gave 4a in 69% yield. Subsequent benzyl-
deprotection was achieved by saponification instead of stan-
dard hydrogenation due to the risk of hydrogenation catalysts
deactivation by sulfur-containing compound 4a.
The preparation of galactose derivatives 2b–5b was previ-
ously reported.[31] Modified procedures for the synthesis of 2b
and 3b were used. Instead of performing the allylation using
allyl bromide at 50◦C for 4 h, the reaction was carried out at
room temperature for 5 h and the allyl bromide was converted to
allyl iodide in situ, leading to an increased yield (78% instead
of 60%[31]). The synthesis of compound 3b was also more effi-
cient than previously reported (43% instead of 32%[31]) when
the crude corresponding amine was reacted directly with adipic
acid monobenzyl ester.
Theintroductionofthelipopeptidemoiety(C12-G-C12-C12-
G) and free thiol functionalities (cysteine) to the carbohydrate
templates 5a and 5b were achieved by stepwise SPPS. In order
to improve the efficiency of synthesis of resin-bound lipopeptide
6,[35] the lipo-amino acid was pre-activated (HBTU/DIPEA) for
5 min or N-methyl pyrrolidinone was used as a co-solvent. By
pre-activating the template 5a or 5b (2 mol. equiv.) for 5 min
and coupling to 6 for 6 h, the coupling efficiency reached 99.8%
according to the quantitative ninhydrin test[36] for both 7a and
7b. Four copies of Boc-Cys(MBzl)-OH were then coupled to 7a
or 7b. The lipopeptide–carbohydrate templates 8a and 8b were
cleaved from the resin with anhydrous hydrofluoric acid (HF).
During and after HF cleavage, to prevent thiazolidine forma-
tion of the N-terminal free thiols, the use of acetone for washing
glassware or lyophilization was avoided.[37] Following purifica-
tion by preparative reverse phase (RP)-HPLC, 8a and 8b were
obtained with a total yield of 28% and 36%, respectively.
The examined epitope (DNGKAIYERARERALQELGP),[3]
derived from the N-terminal of M protein of commonly circu-
lated GAS 8830 strain, was chemically synthesized as a thioester
9[25] using Boc-based SPPS. Both standard and microwave-
assisted SPPS were investigated. Microwave-assisted SPPS
increased the yield by 1.5-fold. To perform ligation between
building blocks 8a or 8b and 9, the selection of solvents
was carefully assessed due to poor solubility of 8a and 8b
in aqueous ligation buffer (phosphate buffer). It was previ-
ously reported[16,26] that attempts to solubilize Cys-lipopeptide
constructs by the addition of organic solvent such as 10%
DMF and 50% trifluoroethanol resulted in slow and incom-
plete ligations, although these co-solvents were used to improve
NCL yields.[38–40] Nevertheless, the ligation between a Cys-
containing lipopeptide-polylysine core and a thioester was suc-
cessful in the presence of SDS, with a higher yield and a shorter
reaction time when performing the ligation at 37◦C.[25] Co-
lyophilization of the poorly water soluble peptide with SDS
significantly improved solubility of the peptide in aqueous lig-
ation buffer. Thus, the hydrophobic peptides 8a and 8b were
dissolved in SDS solution, lyophilized, and ligated to peptide
thioester 9 in 0.1 M phosphate buffer, pH 7.6, containing sodium
2-mercaptoethanesulfonate (MESNA) as a thiol additive and
O
H
N
C12:
Lipidic adjuvant
O
O
O
O
Spacer 2
C12-Gly-C12-C12-Gly-NH2
O
Carbohydrate template
Fig. 1. Carbohydrate-based lipid core peptide system.
using sodium dodecyl sulfate (SDS) as a solubilizing agent,
enabled the ligation of the highly hydrophobic lipid core to the
highly hydrophilic peptide.[25,26] Therefore, this method was
examined for the synthesis of lipopeptide–carbohydrate assem-
bled multivalent peptides.
This paper describes the design and optimized synthesis
of two lipopeptide conjugated, carbohydrate templates and the
assembling of peptide antigens onto the carbohydrate templates
using NCL. This work also investigated the secondary structure
of the obtained vaccine candidates by circular dichroism (CD)
measurements. Important comparisons were drawn between
the conformational characteristics of the tetravalent peptides,
based on different carbohydrate templates, and the free antigenic
peptide.
Results and Discussion
Each vaccine candidate is composed of (i) multiple copies of
an antibody-inducing peptide epitope, (ii) a self-adjuvanting
lipopeptide moiety (C12-Gly-C12-C12-Gly; C12: 2-(R/S)-
aminododecanoic acid[27]), and (iii) a functionalized monosac-
charide template acting as a branching core and a topology
controlling scaffold. Two different monosaccharide (d-glucose
and d-galactose) derivatives were developed as templates. In all
casestheantigenicpeptidesareconjugatedtothefourfunctional-
ized non-anomeric positions of the sugar ring and the lipopeptide
moiety is conjugated to the functionalized anomeric posi-
tion. The convergent synthetic approach to vaccine candidates
was designed based on chemoselective ligation of the peptide
thioester and the Cys-containing lipopeptide–carbohydrate tem-
plates.The covalent linkage of lipopeptide and carbohydrate was
achieved using Boc-based SPPS.To afford this, hydroxyl groups
of β-d-glycosyl azides (1a and 1b)[28] were suitably modified to
5a and 5b which possess four Boc-protected amine groups at
the non-anomeric positions and a free carboxylic acid group at
the anomeric position of the sugar ring (Scheme 1).
The glucose derivative 5a was synthesized from readily avail-
able azide 1a. However, initial allylation of 1a with NaH/allyl
bromideinTHFaccordingtotheliterature[29] wasnotsuccessful,
resulting in recovery of the starting material.Allylation was then
carried out using the procedure described by McGeary et al.[30]
The reaction led to a complex mixture of di-, tri-, and tetra-
allylated derivatives even when the reaction time was extended

RESEARCH FRONT
Lipopeptide–Carbohydrate Assembled Multivalent Vaccine Candidates
995
R1
R1
O
OH
O
O
O
R2
R2
(i) SnCl2·H2O, MeOH/H2O
(i) NaH/DMF, 0°C
N3
N3
O
HO
(ii) Adipic acid monobenzyl ester
HBTU, DIPEA, THF
(ii) Allyl bromide, KI
OH
1a R1 ϭ H; R2 ϭ OH
1b R1 ϭ OH; R2 ϭ H
2a R1 ϭ H; R2 ϭ OCH2CHϭCH2 (78%)
2b R1 ϭ OCH2CHϭCH2; R2 ϭ H (78%)
S
R1
O
R1
NHBoc
O
O
O
O
O
O
R2
R2
H
N
H
N
Boc-cysteamine
O
O
S
OBn
OBn
BocHN
O
S
1,4-Dioxane, UV
O
O
BocHN
4a R1 ϭ H; R2 ϭ O(CH2)3S(CH2)2NHBoc (69%)
4b R1 ϭ O(CH2)3S(CH2)2NHBoc; R2 ϭ H (77%)
3a R1 ϭ H; R2 ϭ OCH2CHϭCH2 (47%)
3b R1 ϭ OCH2CHϭCH2; R2 ϭ H (43%)
S
R1
O
NHBoc
Boc-C12-OH, Boc-Gly-OH
p-MBHA resin
O
O
R2
H
LiOH·H2O
N
S
O
OH
SPPS
BocHN
O
S
MeOH/H2O, pH 13
O
BocHN
Boc-C12-Gly-C12-C12-Gly-NH
5a R1 ϭ H; R2 ϭ O(CH2)3S(CH2)2NHBoc (85%)
5b R1 ϭ O(CH2)3S(CH2)2NHBoc; R2 ϭ H (87%)
6
SPPS
S
R1
NHBoc
O
O
O
O
R2
H
N
O
S
N
H
C12-Gly-C12-C12-Gly-NH
BocHN
S
O
BocHN
7a R1 ϭ H; R2 ϭ O(CH2)3S(CH2)2NHBoc
7b R1 ϭ O(CH2)3S(CH2)2NHBoc; R2 ϭ H
(i) Boc-Cys(MBzl)-OH
SPPS
(ii) HF cleavage
S
R1
O
NH-Cys
O
O
O
R2
H
N
S
O
N
C12-Gly-C12-C12-Gly-NH2
Cys-HN
H
S
O
Cys-HN
8a
8b
R
R
1 ϭ H; R2 ϭ O(CH2)3S(CH2)2NHCys (28%)
1 ϭ O(CH2)3S(CH2)2NHCys; R2 ϭ H (36%)
Scheme 1. Synthesis of lipopeptide–carbohydrate templates 8a and 8b.
tris-2-carboxyethylphosphine hydrochloride (TCEP) to prevent
disulfide bond formation (Scheme 2). It was reported that the use
of MESNA, instead of commonly used thiol additive thiophe-
nol, may facilitate the ligation when SDS was used to solubilize
hydrophobic peptides.[41] The reaction proceeded efficiently and
completed within 24 h. SDS kept the peptides in ligation buffer
throughout the course of reaction. Following purification by
preparative RP-HPLC, tetravalent peptides 10a and 10b were
obtained in 75% and 78% yield respectively. The HPLC profiles
and the electrospray ionization-mass spectrometry (ESI-MS)
spectra (Fig. 2) showed the high purity of 10a and 10b and
verified the correct primary structure of the synthetic vaccine
candidates.
potentialanditwasfoundthatthefreepeptideexistsinphosphate
buffer (pH 7.3) in an almost random structure (Fig. 3). In
contrast, the CD spectra of 10a and 10b recorded in phos-
phate buffer (pH 7.3) showed that the constructs contained an
α-helices, with the template itself (8a or 8b) showing very
low mean residue ellipticity values (Fig. 3). The CD spectra
of 10a and 10b are characterized by two negative minima at
207 nm and 222 nm which are typical for the onset of a heli-
cal structure. Because the accuracy of the calculated α-helical
content relies on the accurate determination of tetravalent pep-
tide concentration, the UV absorbance of stock solutions of 10a
and 10b (both containing four Tyr residues) was measured at
280 nm.[42] The peptide α-helical content was calculated to be
40% and 36% for 10a and 10b, respectively from the mean
residue ellipticity at 222 nm (MRE, [θ]222) according to the
The free antigenic peptide 11 (DNGKAIYERARERALQ
ELGP), synthesized using standard SPPS, was tested for helical

RESEARCH FRONT
996
W. Zhong et al.
8a, 8b
Ac-[8830]-Gly
S
Leu NH2
9
Ligation
1% (w/v) SDS, TCEP, 0.1 M phosphate buffer pH 7.6, 37°C
O
Thiol exchange
MESNA, 0.1 M phosphate buffer pH 7.6
O
OϪ
Naϩ
S
Ac-[8830]-Gly
S
S
R1
O
NH-Cys-Gly-[8830]-Ac
O
O
9a
O
O
R2
H
N
S
O
N
C12-Gly-C12-C12-Gly-NH2
Ac-[8830]-Gly-Cys-HN
H
S
O
Ac-[8830]-Gly-Cys-HN
10a R1 ϭ H; R2 ϭ O(CH2)3S(CH2)2NHCysGly[8830]Ac (75%)
10b R1 ϭ O(CH2)3S(CH2)2NHCysGly[8830]Ac; R2 ϭ H (78%)
8830: DNGKAIYERARERALQELGP
Scheme 2. Synthesis of tetravalent vaccine candidates 10a and 10b.
5000
ϩ11
ϩ10
1.0
0.8
0.6
0.4
0.2
0.0
10a
ϩ12
ϩ13
ϩ9
0
ϩ8
Ϫ5000
10a
10b
8a
Ϫ10000
0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30
Minutes
Ϫ15000
Ϫ20000
8b
11
ϩ11
ϩ10
1.0
0.8
0.6
0.4
0.2
0.0
200
210
220
230
240
250
260
ϩ12
Wavelength [nm]
10b
ϩ9
Fig. 3. Circular dichroism spectra of 8a, 8b, 10a, 10b, and 11.
ϩ13
ϩ8
ϩ14
peptide was observed.This result suggests that the two monosac-
charide molecules are promising templates for constructing
α-helix bundles that may be useful as mimics of conformational
epitopes for vaccine development. The use of appropriate liga-
tion techniques provided multivalent immunogens in highly pure
state suitable for use in humans.The design and synthesis extend
the use of carbohydrate templates and LCP system for producing
self-adjuvanting and topology-controlled vaccine candidates.
0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30
Minutes
Fig. 2. Reverse phase-HPLC and electrospray ionization-mass spectro-
metry profiles of 10a and 10b.
published formula.[43] The glucose-based and galactose-based
tetravalent peptides showed comparable helicity. Concentration
dependences of helicity of peptide 10a and 10b over a range
of concentrations from 20 to 33 µM were investigated. It was
found that the CD spectra of 10a and 10b were concentration-
independent within the range studied, suggesting that there was
no oligomerization.
Experimental
General Procedures
N-α-Boc-l-amino acids were purchased from Novabiochem
(Laufelfingen, Switzerland) or GL Biochem (Shanghai, China).
p-MBHA resin was purchased from Peptides International
(Louisville, Kentucky, USA). Peptide synthesis grade TFA,
DIEA, DMF, and HBTU were purchased from Auspep
(Melbourne, Australia) or GL Biochem. HPLC grade ACN
and IPA were purchased from Honeywell-Burdick & Jackson
(Morristown, NJ, USA) or Labscan (Dublin, Ireland).
Microwave-assisted SPPS was performed in a SPS mode CEM
Discovery reactor (CME Corporation, Matthews, NC, USA)
equipped with an external FO temperature sensor. 1H NMR and
13C NMR spectra were recorded on a Bruker Avance 500 MHz
Conclusions
In conclusion, we have demonstrated the utility of chemo-
selective ligation for the synthesis of two carbohydrate template
assembled vaccine candidates, featuring four copies of epi-
topes and a self-adjuvanting lipopeptide moiety. Significantly
enhanced α-helicity of the attached peptide antigens over the free

RESEARCH FRONT
Lipopeptide–Carbohydrate Assembled Multivalent Vaccine Candidates
997
spectrometer (Bruker Biospin, Germany). ESI-MS was per-
formed on either a Perkin–Elmer-Sciex API3000 instrument or
a LC ESI MS (Agilent 1200 HPLC (Waldbronn, Germany) and
API QSTAR pulsar i ESI-MS). Analytical RP-HPLC was per-
formed using Shimadzu (Tokyo, Japan) instrumentation (Class
Vp 6.12 software, SCL-10AVp controller, SIL-10A autoinjec-
tor, LC-10AT pump, LC-10AD pump, Waters 486 tuneable
absorbance detector or SPD-6A UV detector). Peptide purifica-
tion was performed on either a Vydac preparative C18 column
or a Vydac preparative C4 column. Analytical RP-HPLC was
performed in gradient mode using 0.1% TFA/H2O as solvent A,
and 90%ACN/0.1%TFA/H2O (B1) or 90% IPA/0.1%TFA/H2O
(B2) or 90% MeOH/0.1% TFA/H2O (B3) as solvent B. Separa-
tion was achieved on either a Vydac analytical C18 column or
a Vydac analytical C4 column using a gradient of 0–100% sol-
vent B over 30 min unless otherwise specified. CD spectra were
measured on a JASCO-810 (CD-ORD) spectropolarimeter using
quartz cuvette of 1 mm path length at 23◦C.
(EtOAc/hexane 2:3) gave 3a (71.3 mg, 47%) as a white solid.
δH (500 MHz, CDCl3) 1.61–1.67 (m, 4H, 2 × CH2CH2CH2),
2.15–2.19 (m, 2H, CH2CO), 2.33–2.36 (m, 2H, CH2CO), 3.11
(t, 1H, J 9.0, H2), 3.36–3.49 (m, 3H, H3, H4, H5), 3.58 (dd,
1H, J 3.1, H6a), 3.62 (dd, 1H, J 2.1, H-6b), 3.91–4.28 (m,
8H, 4 × CH2CH=CH2), 5.02 (t, 1H, J 9.3, H-1), 5.08 (s, 2H,
CH2OPh), 5.08–5.28 (m, 8H, 4 × CH=CH2), 5.77–5.97 (m,
4H, 4 × CH=CH2), 7.27–7.35 (m, 5H, 5 ×ArH). δC (125 MHz,
CDCl3) 24.2, 24.5, 33.8, 36.2, 66.2, 67.9, 72.4, 73.5, 73.7,
74.2, 76.2, 78.7, 80.4, 85.1, 116.8, 116.8, 117.4, 117.4, 128.1,
128.5, 134.4, 134.7, 134.8, 134.9, 135.9, 172.4, 173.1. m/z (ESI)
558.3 [M + H]+, 575.4 [M + NH4]+, 580.4 [M + Na]+. HRMS
(ESI, [M + Na]+) Found: 580.2861. Calc. for C31H43NO8Na:
580.2881.
6-[2,3,4,6-Tetra-O-[3-(2-tert-
butoxycarbonylaminoethylthio)-propyl]-β-D-
glucopyranosylamino]-6-oxohexanoic Acid
Benzyl Ester 4a
2,3,4,6-Tetra-O-allyl-β-D-glucopyranosyl Azide 2a
Compound 3a (78.9 mg, 140 µmol) and Boc-cysteamine
(400 mg, 2.26 mmol) were dissolved in dried and degassed 1,4-
dioxane (1.2 mL) and the mixture was irradiated at 254 nm for
4 h at 0◦C. The solvent was then evaporated under reduced
pressure and the residue was purified by flash chromato-
graphy (EtOAc:hexane 3:2, 5% Et3N) to give 4a (123 mg,
69%) as a colourless oil. δH (500 MHz, CDCl3) 1.41 (s, 36H,
4 × C(CH3)3), 1.64–1.87 (m, 12H, 6 × CH2CH2CH2), 2.22 (br
t, 2H, J 7.6, CH2CO), 2.35 (br t, 2H, J 6.9, CH2CO), 2.52–
2.60 (m, 16H, 8 × CH2S), 3.01 (t, 1H, J 8.5, H2), 3.23–3.29 and
3.45–3.79 (2m, 21H, 4 × CH2N, 5 × CH2O, 3 × CHO), 5.00 (t,
1H, J 9.3, H1), 5.07 (s, 2H, CH2OPh), 7.27–7.35 (m, 5H, ArH).
δC (125 MHz, CDCl3) 24.4, 24.6, 27.6, 28.1, 28.4, 28.4, 29.9,
30.1, 30.2, 30.5, 31.9, 32.0, 32.1, 33.8, 36.1, 39.9, 66.1, 68.9,
69.2, 70.7, 70.8, 71.7, 76.2, 77.4, 78.3, 78.8, 79.4, 79.5, 81.6,
85.4, 128.1, 128.5, 135.9, 155.7, 155.8, 156.0, 172.9, 173.2.
m/z (ESI) 1266.9 [M + H]+, 1283.8 [M + NH4]+. HRMS (ESI,
[M + Na]+) Found: 1288.6138. Calc. for C59H103N5O16S4Na:
1288.6175.
NaH (60% in dispersion oil, 1.17 g, 29.3 mmol) was added por-
tion wise to a solution of β-d-glucopyranosyl azide 1a (1.0 g,
4.9 mmol) at 0◦C in DMF (80 mL). After stirring for 45 min, KI
(3.4 g, 19.5 mmol) was added, followed by drop wise addition of
a solution of allyl bromide (3.25 mL, 39 mmol) in DMF (8 mL).
The mixture was stirred at room temperature under an inert atmo-
sphere. After 5 h, the excess NaH was quenched by addition of
MeOH (15 mL).The solvent was evaporated and the residue was
dissolved in DCM (120 mL), washed with water (3 × 100 mL),
dried (MgSO4), filtered, and concentrated under reduced pres-
sure. Purification of the crude residue by flash chromatography
(EtOAc:hexane 1:7) gave 2a (1.39 g, 78%) as a colourless oil.
δH (500 MHz, CDCl3) 3.10 (t, 1H, J 8.8, H2), 3.33–3.41 (m, 3H,
H3, H4, H5), 3.61 (dd, 1H, J 4.0, 11.0, H6a), 3.68 (dd, 1H, J 1.4,
11.0, H6b), 3.97–4.31 (m, 8H, 4 × CH2CH=CH2), 4.47 (d, 1H, J
8.6, H1), 5.12–5.28 (m, 8H, 4 × CH=CH2), 5.83–5.96 (m, 4H,
4 × CH=CH2). δC (125 MHz, CDCl3) 68.3, 76.8, 74.3, 73.8,
73.8, 72.5, 77.0, 80.9, 84.3, 90.0, 117.4, 117.2, 117.0, 116.7,
134.9, 134.6, 134.4. m/z (ESI) 338.3 [MH-CH=CH2]+, 383.3
[M + NH4]+.
6-[2,3,4,6-Tetra-O-[3-(2-tert-
butoxycarbonylaminoethylthio)-propyl]-β-D-
glucopyranosylamino]-6-oxohexanoic Acid 5a
6-[2,3,4,6-Tetra-O-(2-propenyl)-β-D-glucopyranosylamino]-
6-oxohexanoic Acid, Benzyl Ester 3a
Compound 4a (1.0 g, 790 µmol) was dissolved in H2O/MeOH
mixture (1:4) (200 mL) in which the pH had been adjusted
to 13 with LiOH·H2O and the reaction mixture was stirred
at room temperature overnight. The solvent was then evapo-
rated under reduced pressure and the residue was dissolved in
EtOAc (100 mL). The organic layer was washed with cold 0.1 M
citric acid solution (2 × 100 mL) and brine (3 × 50 mL), dried
(MgSO4), filtered, and concentrated.The residue was washed via
a plug of silica gel with CHCl3/MeOH (9:1) to give 5a (791 mg,
85%) as a colourless oil. δH (500 MHz, CDCl3) 1.41 (s, 36H,
4 × C(CH3)3), 1.59–1.86 (m, 12H, 6 × CH2CH2CH2), 2.25 (br
t, 2H, J 7.0, CH2CO), 2.33 (br t, 2H, J 6.6, CH2CO), 2.50–2.66
(m, 16H, 8 × CH2S), 3.01 (t, 1H, J 8.9, H2), 3.24–3.28 and 3.45–
3.79 (2m, 21H, 4 × CH2N, 5 × CH2O, 3 × CHO), 5.00 (t, 1H, J
9.3, H1). δC (125 MHz, CDCl3) 24.2, 24.6, 27.6, 28.0, 28.2, 28.3,
29.7, 30.1, 30.2, 30.5, 31.8, 31.8, 32.1, 33.4, 36.2, 39.5, 39.7,
68.8, 69.2, 70.8, 71.7, 76.2, 77.4, 78.8, 79.3, 79.5, 79.7, 81.5,
85.4, 155.7, 156.0, 173.0; m/z (ESI) 1176.7 [M + H]+, 1193.9
[M + NH4]+. HRMS (ESI, [M + Na]+) Found: 1198.5661.
Calc. for C52H97N5O16S4Na: 1198.5705.
SnCl2·H2O (620 mg, 2.7 mmol) in MeOH (5 mL) with a few
drops of deionized water was added to a solution of 2a (100 mg,
270 µmol) and the mixture was stirred overnight at room temper-
ature under an inert atmosphere.The solvent was then evaporated
under reduced pressure and the residue was diluted with EtOAc
(30 mL). The solution was basified by saturated NaHCO3 solu-
tion (30 mL) and filtered. The organic phase of the filtrate was
separated and basified by saturated NaHCO3 solution (30 mL)
and filtered. The organic phase was separated, dried (MgSO4),
filtered, and evaporated. The residue was dissolved in dry
THF (10 mL) to which adipic acid monobenzyl ester (62 mg,
270 µmol), HBTU (100 mg, 270 µmol), and DIPEA (46 µL,
270 µmol) in dry THF (5 mL) were added with stirring. The
reaction mixture was stirred overnight under an inert atmo-
sphere. The solvent was then evaporated under reduced pressure
and the residue was dissolved in EtOAc (50 mL). The organic
phase was washed with 5% HCl solution (3 × 30 mL), saturated
NaHCO3 (2 × 30 mL), and brine (50 mL), dried (MgSO4), fil-
tered, and concentrated. Purification by flash chromatography

RESEARCH FRONT
998
W. Zhong et al.
Synthesis of 6, 7a, 7b, 8a, and 8b
was then formed by the 2 × 10 min coupling of Boc-Gly-OH
(800 µmol, 4 equiv.).The sequence of the peptide epitope (8830:
DNGKAIYERARERALQELGP) was then C-terminally cou-
pled by stepwise SPPS to the Boc-Gly-TAMPAL resin. Each
amino acid coupling cycle consisted of Boc-deprotection with
neat TFA (2 × 1 min), a 1 min DMF flow wash, followed by a
2 × 10 min coupling with pre-activated amino acid. Amino acid
activation was achieved by dissolving amino acids (600 µmol,
3 equiv.) in 0.5 M HBTU/DMF solution (1.1 mL, 550 µmol)
to which DIPEA (210 µL, 1.2 mmol) was added and activa-
tion proceeded for 1 min. All couplings were carried out under
microwave irradiation (SPS mode, power 20W, temperature
60◦C, ꢀT = 1◦C). After coupling glutamine residue, the resin
was washed with DCM before and after Boc-deprotection to pre-
vent pyrrolidone carboxylic acid formation.[44] Spectroscopic
data and purity of 9 are identical to previously reported.[25]
Stepwise solid peptide synthesis of resin-bound lipopeptide 6
was carried out on p-MBHA resin (0.45 mmol NH2/g, 250 µmol
scale) using HBTU/DIPEA in situ neutralization procedure.[44]
Each amino acid coupling cycle consisted of Boc-deprotection
with neatTFA (2 × 1 min), a 1 min DMF flow wash, followed by
a 1–4 h coupling with pre-activated amino acid.Amino acid acti-
vation was achieved by dissolving amino acids (1.1 mmol, 4.4
equiv.) in 0.5 M HBTU/DMF solution (2 mL, 1 mmol) to which
DIPEA (350 µL, 2 mmol) was added and activation proceeded
for 5 min. Coupling efficiency was monitored by quantitative
ninhydrin test.[42] Where necessary, couplings were repeated to
give coupling efficiency greater than 99.7%. Following cleavage
of the N-terminal Boc group, carbohydrate template 5a or 5b was
C-terminally coupled to 6 (100 µmol scale).Activation of 5a, 5b
was achieved by dissolving the compound (200 µmol, 2 equiv.)
in 0.5 M HBTU/DMF solution (380 µL, 190 µmol) to which
DIPEA (70 µL, 400 µmol) was added and activation proceeded
for 5 min. Coupling proceeded for 6 h at which time coupling
efficiencywasgreaterthan99.8%togive 7a and 7b, respectively.
Following cleavage of the N-terminal Boc group, four copies
of Boc-Cys(MBzl)-OH were then coupled to the carbohydrate
template residue. Boc-Cys(MBzl)-OH activation was achieved
by dissolving the amino acid (1.76 mmol, 17.6 equiv.) in 0.5 M
HBTU/DMF solution (3.16 mL, 1.58 mmol) to which DIPEA
(522 µL, 3 mmol) was added and activation proceeded for 1 min.
Three couplings were performed to give coupling efficiency
greater than 99.9%. Following removal of the N-terminal Boc
groups, the peptidyl resin was washed with DMF (3 × 10 mL),
DCM (3 × 10 mL) and MeOH (3 × 10 mL), and dried under
vacuum before cleavage with HF. HF cleavage (10 mL HF/g
resin) was performed for 2 h at 0◦C in the presence of 5% (v/v)
p-cresol and 5% (v/v) thiocresol as scavengers. Following HF
cleavage, peptides were precipitated by washing with ice-cold
ether and dissolved in 50% aqueous ACN containing 0.1% TFA
and lyophilized. The crude peptides were purified by prepar-
ative RP-HPLC on a C4 column using a gradient of 40–70%
solvent B1 over 45 min. Fractions were analyzed by ESI-MS
and combined where appropriate to give 8a (52.3 mg, 28%) and
8b (68.5 mg, 36%) following lyophilization. The characteristic
data for the Cys-containing lipopeptide–carbohydrate templates
8a and 8b are shown below.
Ligation of Thioester 9 to Lipopeptide–Carbohydrate
Templates 8a and 8b
Peptide 8b (8.35 mg; 4.41 µmol) was dissolved in 1% (w/v)
aqueous SDS (10 mL), frozen, and lyophilized. The obtained
powder was then dissolved in 0.1 M phosphate buffer pH 7.6
(5 mL), followed by addition ofTCEP (15.2 mg; 52.9 µmol) and
the pH adjustment to 7.5 with 0.1 M sodium phosphate dibasic
(0.8 mL). Peptide 9 (91.2 mg; 35.3 µmol) was dissolved in 0.1 M
phosphate buffer pH 7.6 (4.2 mL) to which MESNA (115 mg;
710 µmol) was added. The thioester exchange was proceeded
for 1 h, and then analyzed by ESI-MS (thiol exchange product
9a: m/z 1256.6 [M + 2H]2+ (calc. 1255.4), 837.5 [M + 3H]3+
(calc. 837.3); MW 2508.77 g mol−1). This solution was then
transferred into the solution of lipopeptide 8b. The ligation
was performed at 37◦C under an inert atmosphere and mon-
itored by analytical RP-HPLC (solvent B1, C4 column). The
ligation appeared complete within 24 h. The ligation mixture
was purified by preparative RP-HPLC (30–70% solvent B1 over
40 min, C4 column). The appropriate fractions were combined
and lyophilized to give 10b (39 mg, 78%). Peptide 10a (9.8 mg,
75%) was synthesized and purified according to the procedure
as for 10b. The characteristic data for 10a and 10b are shown
below.
10a: HPLC: tR = 19.2 min (solvent B1, C4 column),
tR = 16.3 min (solvent B2, C4 column); m/z (ESI) 1421.5
[M + 8H]8+ (calc. 1421.5), 1263.7 [M + 9H]9+ (calc. 1263.7),
1137.4 [M + 10H]10+ (calc. 1137.4), 1034.1 [M + 11H]11+
(calc. 1034.1), 948.0 [M + 12H]12+ (calc. 948.0), 875.2
8a: HPLC: tR = 21.3 min (solvent B1, C4 column),
tR = 23.5 min (solvent B3, 30–100% B over 30 min, C4 column);
m/z (ESI) 1893.2 [M + H]+ (calc. 1894.8), 948.1 [M + 2H]2+
(calc. 947.9); MW 1893.79 g mol−1
.
[M + 13H]13+ (calc. 875.1); MW 11364.09 g mol−1
.
8b: HPLC: tR = 20.8 min (solvent B1, C4 column),
tR = 18.3 min (solvent B2, C4 column); m/z (ESI) 1894.4
[M + H]+ (calc. 1894.8), 948.1 [M + 2H]2+ (calc. 947.9); MW
10b: HPLC: tR = 19.2 min (solvent B1, C4 column),
tR = 16.5 min (solvent B2, C4 column); m/z 1421.5 [M + 8H]8+
(calc. 1421.5), 1263.6 [M + 9H]9+ (calc. 1263.7), 1137.4
[M + 10H]10+ (calc. 1137.4), 1034.1 [M + 11H]11+ (calc.
1034.1), 948.0[M + 12H]12+ (calc. 948.0), 875.1[M + 13H]13+
(calc. 875.1), 812.7 [M + 14H]14+ (calc. 812.7); MW
1893.79 g mol−1
.
Microwave-Assisted Synthesis of 8830 Thioester 9
Microwave-assisted synthesis of peptide 9 was carried out using
a trityl-associated mercaptopropionic acid-leucine (TAMPAL)
linker.[45] The TAMPAL linker was synthesised on p-MBHA
resin by coupling of Boc-Leu-OH (600 µmol, 3 equiv.)
(2 × 10 min) and S-trityl-mercaptopropionic acid (800 µmol,
4 equiv.) (1 × 10 min) using the standard HBTU/DIPEA in situ
neutralization procedure.[44] Following TAMPAL synthesis, the
trityl protecting group was removed by three 2 min treatments
with 2.5% (v/v) triisopropylsilane, 2.5% (v/v) H2O, 95% (v/v)
TFA, followed by a 1 min DMF flow wash. The thioester bond
11364.09 g mol−1
.
CD Measurements
CD measurements were performed on a Jasco J-710 spectro-
polarimeter at 23◦C using a circular 0.1 cm path length quartz
cell. Stock solutions of 10a, 10b, and 11 were prepared by dis-
solving the peptide in a phosphate buffer (10 mM, pH 7.3). The
actual concentration of the peptide was measured by UV absorp-
tion and calculated according to the published formula based on

RESEARCH FRONT
Lipopeptide–Carbohydrate Assembled Multivalent Vaccine Candidates
999
Tyr absorption at 280 nm.[42] Spectral measurements of the pep-
tides 10a and 10b were obtained over a range of concentrations
(20–33 µM). The samples 8a and 8b were dissolved in phos-
phate buffer (10 mM, pH 7.3). Blank was subtracted from the
CD spectra. Content of helicity was calculated according to the
published formula.[43]
[19] C. Olive, M. R. Batzloff, I. Toth, Expert Rev. Vaccines 2004, 3, 43.
doi:10.1586/14760584.3.1.43
[20] L. X. Wang, J. Ni, S. Singh, Bioorg. Med. Chem. 2003, 11, 159.
doi:10.1016/S0968-0896(02)00339-5
[21] J. H. Ni, R. Powell, I.V. Baskakov,A. DeVico, G. K. Lewis, L. X.Wang,
Bioorg. Med. Chem. 2004, 12, 3141. doi:10.1016/J.BMC.2004.04.008
[22] J. Brask, K. J. Jensen, Bioorg. Med. Chem. Lett. 2001, 11, 697.
doi:10.1016/S0960-894X(01)00035-X
[23] J. Brask, J. M. Dideriksen, J. Nielsen, K. J. Jensen, Org. Biomol. Chem.
2003, 1, 2247. doi:10.1039/B301948A
[24] P. Simerska,A. B. M.Abdel-Aal,Y. Fujita, P. M. Moyle, R. P. McGeary,
M. R. Batzloff, C. Olive, M. F. Good, I. Toth, J. Med. Chem. 2008, 51,
1447. doi:10.1021/JM701410P
[25] P. M. Moyle, Y. Hari, N. Huang, C. Olive, M. F. Good, I. Toth, Tetra-
hedron Lett. 2007, 48, 4965. doi:10.1016/J.TETLET.2007.05.129
[26] Y. Fujita, P. M. Moyle, S. Hieu, P. Simerska, I.Toth, Biopolymers 2008,
90, 624. doi:10.1002/BIP.21002
[27] W. A. Gibbons, R. A. Hughes, M. Charalambous, M. Christodoulou,
A. Szeto, A. E. Aulabaugh, P. Mascagni, I. Toth, Liebigs Ann. Chem.
1990, 1175. doi:10.1002/JLAC.1990199001215
[28] Z. Györgydeak, L. Szilágyi, H. Paulsen, J. Carbohydr. Chem. 1993,
12, 139. doi:10.1080/07328309308021266
[29] C. Li, A. S. Peters, E. L. Meredith, G. W. Allman, P. B. Savage, J. Am.
Chem. Soc. 1998, 120, 2961. doi:10.1021/JA973881R
[30] R. P. McGeary, I. Jablonkai, I. Toth, Tetrahedron 2001, 57, 8733.
doi:10.1016/S0040-4020(01)00852-3
[31] P. Y. Hayes, B. P. Ross, B. G. Thomas, I. Toth, Bioorg. Med. Chem.
2006, 14, 143. doi:10.1016/J.BMC.2005.08.004
[32] A. R. English, D. Girard, V. J. Jasys, R. J. Martingano, M. S. Kellogg,
J. Med. Chem. 1990, 33, 344. doi:10.1021/JM00163A055
[33] S. Lopez, S. S. Simons Jr, J. Med. Chem. 1991, 34, 1762.
doi:10.1021/JM00110A002
Acknowledgements
This work was supported by the National Health and Medical Research
Council (NHMRC) of Australia and the Australian National Heart Founda-
tion (NHF). The author would like to acknowledge Chris Wood (University
of Queensland, SCMB) for LC-ESI-MS analysis.
References
[1] M. R. Batzloff, K. S. Sriprakash, M. F. Good, Curr. DrugTargets 2004,
5, 57. doi:10.2174/1389450043490721
[2] V. A. Fischetti, Clin. Microbiol. Rev. 1989, 2, 285.
[3] E. R. Brandt, K. S. Sriprakash, R. I. Hobb, W.A. Hayman, W. G. Zeng,
M. R. Batzloff, D. C. Jackson, M. F. Good, Nat. Med. 2000, 6, 455.
doi:10.1038/74719
[4] P. R. Smeesters, P. Mardulyn,A.Vergison, R. Leplae, L.Van Melderen,
Vaccine 2008, 26, 5835. doi:10.1016/J.VACCINE.2008.08.037
[5] E. H. Beachey, M. Bronze, J. B. Dale, W. Kraus, T. Poirier, S. Sargent,
Vaccine 1988, 6, 192. doi:10.1016/S0264-410X(88)80027-6
[6] S. Pruksakorn, B. Currie, E. Brandt, C. Phornphutkul,
S. Hunsakunachai, A. Manmontri, J. H. Robinson, M. A. Kehoe,
A. Galbraith, M. F. Good, Int. Immunol. 1994, 6, 1235. doi:10.1093/
INTIMM/6.8.1235
[7] C. Olive, I. Toth, D. Jackson, Mini Rev. Med. Chem. 2001, 1, 429.
doi:10.2174/1389557013406666
[8] C. Olive, M. Batzloff, A. Horváth, T. Clair, P. Yarwood, I. Toth, M. F.
Good, Infect. Immun. 2003, 71, 2373. doi:10.1128/IAI.71.5.2373-
2383.2003
[9] P. M. Moyle, C. Olive, M. F. Good, I. Toth, J. Pept. Sci. 2006, 12, 800.
doi:10.1002/PSC.815
[10] B. Y. Chua, E. M. Eriksson, L. E. Brown, W. Zeng, E. J. Gowans,
J. Torresi, D. C. Jackson, Vaccine 2008, 26, 4866. doi:10.1016/
J.VACCINE.2008.03.032
[34] C. Kieburg, M. Dubber, T. K. Lindhorst, Synlett 1997, 1447.
[35] A. Horváth, C. Olive, A. Wong, T. Clair, P.Yarwood, M. Good, I. Toth,
Lett. Pept. Sci. 2002, 8, 285.
[36] V. K. Sarin, S. B. H. Kent, J. P. Tam, R. B. Merrifield, Anal. Biochem.
1981, 117, 147. doi:10.1016/0003-2697(81)90704-1
[37] P. E. Dawson, Methods Enzymol. 1997, 287, 34. doi:10.1016/S0076-
6879(97)87005-X
[38] B. Wu, J. H. Chen, J. D. Warren, G. Chen, Z. H. Hua, S. J. Danishefsky,
Angew. Chem. Int. Ed. 2006, 45, 4116. doi:10.1002/ANIE.200600538
[39] R. Balambika, T. Inui, H. Sargsyan, B. Arshava, L. S. Cohen, F. X.
Ding, J. M. Becker, F. Naider, Int. J. Pept. Res. Ther. 2007, 13, 251.
doi:10.1007/S10989-006-9076-6
[11] A. B. M. Abdel-Aal, M. R. Batzloff, Y. Fujita, N. Barozzi, A. Faria,
P. Simerska, P. M. Moyle, M. F. Good, I. Toth, J. Med. Chem. 2008,
51, 167. doi:10.1021/JM701091D
[12] Y. Fujita, A. B. Abdel-Aal, N. Wimmer, M. R. Batzloff, M. F. Good,
I. Toth, Bioorg. Med. Chem. 2008, 26, 4866.
[40] F. I. Valiyaveetil, R. MacKinnon, T. W. Muir, J. Am. Chem. Soc. 2002,
124, 9113. doi:10.1021/JA0266722
[13] C. Olive, K. Hsien, A. Horváth, T. Clair, P. Yarwood, I. Toth, M. F.
Good, Vaccine 2005, 23, 2298. doi:10.1016/J.VACCINE.2005.01.041
[14] C. Olive, M. F. Ho, J. Dyer, D. Lincoln, N. Barozzi, I.Toth, M. F. Good,
J. Infect. Dis. 2006, 193, 1666. doi:10.1086/504266
[41] T. Sato, Y. Saito, S. Aimoto, J. Pept. Sci. 2005, 11, 410. doi:10.1002/
PSC.634
[42] C. N. Pace, F. Vajdos, L. Fee, G. Grimsley, T. Gray, Protein Sci. 1995,
4, 2411. doi:10.1002/PRO.5560041120
[15] C. Olive, S. Hsien Kuo, H. Mei-Fong, J. Dyer, A. Horváth, I. Toth,
M. F. Good, J. Infect. Dis. 2006, 194, 316. doi:10.1086/505580
[16] P. M. Moyle, C. Olive, M. F. Ho, M. F. Good, I. Toth, J. Med. Chem.
2006, 49, 6364. doi:10.1021/JM060475M
[17] I. Toth, M. Danton, N. Flinn, W. A. Gibbons, Tetrahedron Lett. 1993,
34, 3925. doi:10.1016/S0040-4039(00)79265-3
[43] P. C. Lyu, J. C. Sherman,A. Chen, N. R. Kallenbach, Proc. Natl.Acad.
Sci. USA 1991, 88, 5317. doi:10.1073/PNAS.88.12.5317
[44] M. Schnolzer, P. Alewood, A. Jones, D. Alewood, S. B. Kent, Int. J.
Pept. Protein Res. 1992, 40, 180.
[45] T. M. Hackeng, J. H. Griffin, P. E. Dawson, Proc. Natl.Acad. Sci. USA
1999, 96, 10068. doi:10.1073/PNAS.96.18.10068
[18] J. P. Tam, Proc. Natl. Acad. Sci. USA 1988, 85, 5409. doi:10.1073/
PNAS.85.15.5409