Synthesis and evaluation of linear CuAAC-oligomerized antifreeze neo-glycopeptides

Article abstract of DOI:10.1039/c4md00013g

Antifreeze glycoproteins (AFGPs) are important naturally occurring biological antifreezes that lower the freezing point of a solution, thereby preventing uncontrolled ice growth. These compounds also inhibit ice recrystallization. Described in this paper is a synthetic antifreeze glycopeptide-based polymer synthesized from an azide/alkyne glycopeptide building block by partial reduction of the azide and subsequent copper catalyzed azide alkyne cycloaddition (CuAAC) polymerization to obtain linear oligomers. To compare the activity with native AFGPs, a linear dodecapeptide (oligomer with four repeating units) was synthesized and isolated which had a comparable length to AFGP-8, the lowest molecular mass glycoprotein AFGP found in nature. In terms of ice recrystallization inhibition (IRI) activity, the triazole-based oligomers displayed only modest IRI activity compared with AFGP-8 and a previously described carbon-linked AFGP analogue. However, CD spectroscopy showed that the triazole-based tetramer possessed a similar secondary structure to the related amide based carbon-linked AFGP tetramer based on AFGP-8. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4md00013g

MedChemComm
View Article Online
View Journal
CONCISE ARTICLE
Synthesis and evaluation of linear CuAAC-
oligomerized antifreeze neo-glycopeptides†
Cite this: DOI: 10.1039/c4md00013g
Steffen van der Wal,ab Chantelle J. Capicciotti,c Stamatia Rontogianni,ab
c
ab
*
Robert N. Ben and Rob M. J. Liskamp
Antifreeze glycoproteins (AFGPs) are important naturally occurring biological antifreezes that lower the
freezing point of a solution, thereby preventing uncontrolled ice growth. These compounds also inhibit
ice recrystallization. Described in this paper is a synthetic antifreeze glycopeptide-based polymer
synthesized from an azide/alkyne glycopeptide building block by partial reduction of the azide and
subsequent copper catalyzed azide alkyne cycloaddition (CuAAC) polymerization to obtain linear
oligomers. To compare the activity with native AFGPs, a linear dodecapeptide (oligomer with four
repeating units) was synthesized and isolated which had a comparable length to AFGP-8, the lowest
molecular mass glycoprotein AFGP found in nature. In terms of ice recrystallization inhibition (IRI)
activity, the triazole-based oligomers displayed only modest IRI activity compared with AFGP-8 and a
previously described carbon-linked AFGP analogue. However, CD spectroscopy showed that the
triazole-based tetramer possessed a similar secondary structure to the related amide based carbon-
linked AFGP tetramer based on AFGP-8.
Received 9th January 2014
Accepted 25th February 2014
DOI: 10.1039/c4md00013g
www.rsc.org/medchemcomm
between melting and freezing points is known as thermal
hysteresis (TH).5 This effect can lower the freezing point of a
Introduction
biological uid enough to allow for survival in Arctic envi-
ronments. The other “antifreeze” effect of the AFGPs is the
ability to inhibit ice recrystallization. Specically, this is the
ability to prevent the enthalpy driven process of large ice
crystals growing larger at the expense of small ice crystals.6,7
Large ice crystals are highly damaging to cells and tissues,
and as such the prevention of the formation of these large
crystals is a good strategy in the cryopreservation of biolog-
ical materials (cells, tissues, etc.).8 Previously tested anti-
freeze glycoprotein analogues have shown clinical potential
by rivalling the efficacy of the oen-used cryoprotectant
DMSO in the preservation of embryonic liver cells, when used
at a lower concentration.8 Using larger molecular weight
oligomers or more potent analogues could further increase
the potential of AFGPs in the cryoprotection of sensitive
tissues and cells.
Antifreeze glycoproteins (AFGPs) are naturally occurring anti-
freeze compounds found in several deep sea teleost sh. These
proteins play an important role in the survival of these organ-
isms living in sub-zero temperatures.1 In contrast to antifreeze
proteins (AFPs) that show a relative diversity in their structures,2
the structure of AFGPs is relatively conserved. A key feature of
these proteins is that they are mostly composed of repeating
tripeptide units (Fig. 1).3,4
The tripeptide repeat consists of two L-alanine residues and
one L-threonine which is glycosylated with b-Gal-(1,3)a-GalNAc.
AFGPs occur in different lengths and are named based on their
size or molecular mass. The shortest fragment, named AFGP-8,
consists of 4 tripeptide repeats, while the longest fragment,
AFGP-1, contains approximately 50 repeats.3
AFGPs exhibit two distinct “antifreeze” properties. One is
a lowering of the freezing point by the antifreeze compound,
while the melting point remains unchanged. The difference
aMedicinal Chemistry and Chemical Biology, Utrecht Institute for Pharmaceutical
Sciences (UIPS), University Utrecht, Universiteitsweg 99, Utrecht, The Netherlands.
E-mail: R.M.J.Liskamp@uu.nl; Fax: +31(0)30-253 6655
bSchool of Chemistry, Joseph Black Building, University of Glasgow, University Avenue,
Glasgow, G12 8QQ, UK
cDepartment of Chemistry, University of Ottawa, D'Iorio Hall, 10 Marie Curie, Ottawa,
Canada
† Electronic supplementary information (ESI) available: Full experimental details,
1H and 13C NMR spectra of all synthesized intermediates plus 1H NMR and ESI
data of the nal compounds. See DOI: 10.1039/c4md00013g
Fig. 1 General structure of AFGPs with the repeating structure H–
(Ala-Ala-Thr(b-D-Gal-(1,3)a-D-GalNAc))n–OH.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
It is thought that the interaction of the sugar hydroxyl
moieties with the water–ice interface is responsible for the
antifreeze properties of AFGPs. However, model AFGPs have
recently been investigated, showing that the most active
compound in terms of IRI activity showed a distinct folding,
thereby creating a hydrophobic pocket that may be responsible
for its specic water–ice interface interactions.9
The use of antifreeze glycoproteins and their analogues for
cryopreservation applications has been limited by the low
availability of these compounds from natural sources. Conse-
quently, effort has been undertaken to obtain these compounds
synthetically. The total synthesis of natural AFGPs has been
described in the literature,10–12 and while elegant, it is not a
viable large scale synthesis of antifreeze compounds on a cost-
effective basis. Therefore, previous work has sought to simplify
the structure of native AFGP-8 while maintaining the antifreeze
activity, in particular the ability to inhibit ice recrystallization. To
date, the best examples of this are the carbon-linked antifreeze
glycoprotein analogues13 (C-AFGPs), e.g. OGG-Gal analogue 1
shown in Fig. 2. Interestingly, these studies have demonstrated
Fig. 3 (Left) The proposed structure of the glycopeptide monomer.
(Right) Comparison between the published AFGP mimic by Ben et al.
(top) and the resulting polymer obtained from the proposed
monomer.
that glycopeptides containing the monosaccharide galactose are isostere into various structure-containing peptides such as in
very effective ice recrystallization inhibitors,14 despite the fact a-helical peptides,20 small cyclic peptides21 and b-sheet-forming
that the monosaccharide is less structurally complex than the peptides.22 Moreover, the strategy of incorporating azide and
disaccharide moiety found in native AFGPs.
alkyne termini proved feasible in the synthesis of Ab peptide-
One of the most easily accessible structures was found to be a based oligomers23 and click chemistry was used successfully in
galactose derived carboxylic acid conjugated via the side chain an approach to biodegradable polymeric systems.24–26
of an ornithine residue instead of a threonine residue
Unfortunately, this method usually generates a signicant
(1, Fig. 2).15 Furthermore, it was found that replacing the two amount of cyclic molecules due to intramolecular cycloaddition.
alanine residues in the natural AFGP by glycine did not result in Although cyclic glycoproteins have been synthesized before and
loss of IRI activity.
are a potentially interesting group of molecules,27 for direct
This compound, which is already more synthetically acces- comparison it was decided to focus on smaller linear oligomers
sible than natural AFGP, was chosen as the starting point for to better compare the IRI activity of the CuAAC oligomerized
creating a glycopeptide monomer that can be polymerized into antifreeze glycoproteins to native antifreeze glycoproteins.
longer compounds that resemble natural AFGPs and synthetic
C-linked AFGPs.
While the CuAAC approach has been utilized previously to
link carbohydrate moieties to polypeptide backbones of various
The idea of treating the tripeptide repeat as a monomer has AFGP analogues, this is the rst instance where it has been used
been explored before, using peptide coupling reagents as poly- to replace the amide bonds in the polypeptide backbone of an
merization agents.16 To simplify the polymerization and increase AFGP (analogue).28–30
the compatibility with more functional groups, the orthogonal
One of the key steps in generating linear oligomers is blocking
CuAAC reaction was employed in our approach (Fig. 3). The 1,2,3- one of the reactive azide or alkyne endgroups of a (growing)
triazole ring has previously been used as a nonconventional oligomer chain (see Fig. 4). In the case of azide/alkyne monomers
bioisostere of an amide bond17–19 and incorporated as an amide this can be accomplished either by having a fraction of the
monomers protected on the alkyne, for example by the TIPS
protecting group, or by reduction of a fraction of the azide to the
amine. Both are potentially feasible strategies, but in the end the
latter was the method of choice, as reduction of an azide to an
amine can be performed with phosphine reducing agents that are
compatible with CuAAC click chemistry.31 Since some of the
starting azide/alkyne monomers can be converted into amino/
alkyne monomers in the same pot by addition of a certain amount
of phosphine, the two discrete starting monomers do not need to
be synthesized separately, which is very advantageous.
The main aim of this research was to evaluate the design of
the resulting oligomers by investigating whether the triazole
moiety does not reduce IRI activity and whether the amide
bonds in the polypeptide backbone of these analogues are not
an essential structural feature of IRI activity. Therefore a
Fig. 2 Structure of the galactose based ornithine-Gly-Gly AFGP 1 by
Ben et al.8,13,15
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
Fig. 4 Schematic overview of obtaining AFGPs by linear polymerization of an azido–alkyne monomer using CuAAC chemistry.
synthetic oligomer with a similar length to natural AFGP-8 was
First, commercially available Boc-sidechain protected orni-
synthesized.32 This was accomplished by tuning the amount of thine was transformed by diazotransfer into the corresponding
phosphine to monomer ratio to 3 : 1. The resulting distribution a-azido compound 2 using imidazolium sulfonyl azide, in a
of linear oligomers can then, when the hydroxyl groups of the slightly modied procedure.33 Compound 2 was puried by
carbohydrate are protected, be separated by HPLC. Separation crystallization as the dicyclohexylamine salt from diethyl ether
was possible because every oligomer has one positive charge in excellent yield. Next, commercially available Boc-glycine was
and a hydrophobic C-glycosidic moiety; the physical properties coupled to propargylamine using the BOP reagent. The resulting
of short oligomers are likely to be more dominated by the dipeptide 3a could be puried by crystallization in good yield
positive charge whereas longer oligomers become more apolar (82%). Precursor 4a was assembled from these two building
due to the amount of hydrophobic moieties per molecule. The blocks by removal of the Boc protecting group of 3a followed by
oligomers were deprotected and the isolated tetramer was BOP coupling with 2 and was obtained in an acceptable yield
ꢀ
compared with the previously described C-AFGP 1 (shown in (55%). Solid compound 4a could be stored at 4 C for 2 weeks
Fig. 2) by evaluation of TH and IRI activity and determination of without any signicant auto-polymerization, but prolonged
the secondary structure by CD spectroscopy.
storage or at room temperature led to the formation of a mixture
of oligomeric impurities. For smaller scale reactions this
stability of compound 4a was satisfactory; however, it was
insufficient for synthesizing larger stocks. Thus, the triisopro-
pylsilyl (TIPS) protected alkyne 4b was also synthesized. The
TIPS protecting group is one of the more stable silyl protecting
groups and is tolerant to treatment with triuoroacetic acid.34
Synthesis of TIPS protected azido/alkyne tripeptide 4b was
similar to the synthesis of the unprotected peptide 4a, but uses
TIPS protected propargylamine instead of propargylamine.35
Since the intermediate 3b and nal product 4b were oils, puri-
cation was carried out using silica chromatography. The
resulting protected azide/alkyne tripeptide could be stored for
months at room temperature without signicant oligomeriza-
tion or degradation. In view of its increased stability as
compared with precursor 4a, 4b was chosen as the starting
material for the synthesis of glycopeptide monomer 5a. Glyco-
sylated TIPS-protected monomer 5b was synthesized by Boc-
deprotection of 4b using TFA, followed by a coupling with the
N-hydroxysuccinimide ester of the galactosylic acid 6 to obtain
compound 5b in good yield (Scheme 2).
Results and discussion
Synthesis of the azido–alkyne glycopeptide monomer
The azido/alkyne tripeptide backbone 4a required for the
synthesis of the glycopeptide monomer was synthesized from
two building blocks (Scheme 1).
Scheme
1 (a) Imidazole-1-sulfonyl-azide hydrochloride, CuSO4,
MeOH$H2O, 98%; (b) BOP, DiPEA, CH2Cl2, 82%[a], 99%[b]; (c) (1) TFA,
CH2Cl2 and (2) 2, BOP, DiPEA, CH2Cl2, 55%[a], 75%[b].
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
chosen, which has also the advantage that its aza–ylide should
be more prone to hydrolysis due to the more polar nature of the
phosphorus substituents. Indeed, it was observed that upon
addition of TCEP, nitrogen gas evolved immediately, and aer 2
hours at room temperature, both the reaction and hydrolysis of
the ylide appeared to be complete as was shown by HPLC and
ESI-MS analyses.
This reaction mixture was used directly for the CuAAC
polymerization: copper(II)sulfate was added followed by the
reducing agent sodium ascorbate. It was allowed to polymerize
for two hours at room temperature, aer which the reaction
mixture was diluted and acidied. HPLC analysis showed a
distinct distribution of peaks with little of the starting materials
present (Fig. 5, top).
As expected, the smaller linear oligomers are more polar and
eluted rst and the higher molecular weight oligomers eluted
later. The degree of polymerization could be directed by adding
different amounts of TCEP. Small amounts of mono-acetyl
deprotected species were observed eluting slightly before the
major peaks. Cyclic peptides however, apart from a small
amount cyclic dimer, were not observed as major compounds.
Separation of the main oligomer peaks of 5a (7a, 8a and 9a) by
preparative HPLC proved feasible and yielded these smaller
oligomers (three to ve tripeptide repeats) in good purity (Fig. 5,
bottom). The identity of these compounds was conrmed by
MALDI-TOF. Small amounts of impurities eluting slightly
Scheme 2 (a) (1) TFA/CH2Cl2 and (2) 6, DiPEA, CH2Cl2 84% and (b)
TBAF, phenol, THF, 90%.
Deprotection of the TIPS–alkyne 5b was rst attempted with
tetrabutylammonium uoride trihydrate (TBAF) in THF. Equi-
molar or slightly higher amounts of TBAF gave sluggish reac-
tions. To prevent deacetylation of the C-glycoside moiety when a
larger excess of TBAF was used, acetic acid was added to reduce
the basic conditions of the reaction.36,37 Unfortunately, even when
5 equivalents of TBAF and acetic acid were used, the reaction was
still sluggish, so presumably acetic acid was too strong an acid,
thereby reducing the nucleophilicity of Fꢁ considerably. To our
delight, when instead an equimolar amount of the weaker acid
phenol was added to TBAF, complete deprotection to 5a was
observed in a relatively short time (1–3 hours) without any
detectable deacetylation of the carbohydrate moiety.
Linear CuAAC oligomerization of the azido–alkyne monomer
The linear oligomerization of monomer 5a consists of two steps
in a one-pot procedure (Scheme 3). In the rst step, a trisub-
stituted phosphine is added to a solution of the azido/alkyne
monomer to reduce a fraction of the azide to amine. Next, to the
resulting mixture of this amino/alkyne and azido/alkyne, cop-
per(I) is introduced to start polymerization. By performing this
reaction at high concentration, intermolecular cycloaddition
may outcompete intramolecular cyclization, which would
otherwise result in cyclic peptides. Initially triphenylphosphine
was chosen as the phosphine to reduce a fraction of the azide of
5a due to its widespread use in the Staudinger reaction.38
Although the initial stage of the reduction of the azide to the
phosphorus-aza-ylide was rapid, hydrolysis of the ylide to the
amine was sluggish (data not shown).
Instead, a more reactive and less sterically hindered phos-
phine, tris(2-carboxyethyl)phosphine hydrochloride (TCEP) was
Fig. 5 (Top) HPLC trace of the crude polymeric mixtures of 5a with
three different TCEP amounts, using a 48 min gradient. (Bottom) HPLC
traces of the purified 0.25 equiv. TCEP polymerization with 7a
Scheme
3 (a) (1) TCEP (0.25 eq.), DiPEA, DMF/H2O and (2) (continuous), 8a (dotted) and 9a (semi continuous) on a 20 min
CuSO4$5H2O, sodium ascorbate, H2O/DMF and (b) NaOMe in MeOH. gradient, offset on the Y-axis.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
earlier than the oligomers could be attributed by MS to acetyl
hydrolysis during the purication and subsequent lyophilisa-
tion. The higher oligomers (>7 tripeptide units) were collected
together as a mixed oligomeric fraction.
Removal of the acetyl groups was performed with sodium
methoxide in methanol, which caused precipitation of the
deprotected oligomers. Water was later added to solubilize the
suspension and cation exchange resin was used to neutralize the
reaction mixture. Aer ltration of the resin and lyophilization,
the oligomers 7b, 8b and 9b were obtained in sufficient amounts
for testing of antifreeze activity.
C-terminal variation of the synthetic AFGPs
The synthesized triazole-linked oligomers have a C-terminal
alkyne moiety, compared to a C-terminal acid functionality of the
natural AFGP-8 and earlier described OGG-Gal 1 (Fig. 1 and 2,
Scheme 5 (Top left) Reference compound OGG-Gal (1), (top right)
carboxamide terminated version of reference compound OGG-Gal
respectively). Although oligomers 7b–9b possess a free N- (11), (bottom left) triazole linked mimic of reference OGG-Gal (10), and
(bottom right) tetramer 8b, which is a triazole linked mimic of
compound 11.
terminal amine, the C-terminus is an alkyne, whereas both the
natural and earlier developed AFGPs are zwitterions by having in
addition to the amine also a C-terminal carboxylic acid. Therefore
the role of the zwitterionic nature of the AFGP on the structure of
these triazole-linked oligomers was also studied. Thus, puried
tetramer 8a, containing a C-terminal alkyne was coupled to ethyl
azidoacetate by CuAAC, which aer deprotection yielded the
structural triazole analogue of OGG-Gal, 10 (Scheme 4).
Similarly, a C-terminal variation of OGG-Gal 1 was synthesized
wherein the C-terminal glycine with a free carboxylic acid moiety
was omitted, and a carboxamide moiety was introduced, leading
to 11, which corresponds to the structure of the triazole alkyne
tetramer 8b. This was accomplished by solid phase synthesis on a
Rink-linker instead of the preloaded glycine–Wang resin as was
used for the synthesis of OGG-Gal 1. The resulting four
Fig. 6 Ice crystal habit in the presence of (A) triazole tetramer,
compounds, that is the C-terminal acids 1 and 10 and non-
C-terminal alkyne 8b; (B) triazole tetramer, C-terminal acid 10; (C)
charged C-terminal amides 8b and 11 of both the peptidic and
peptide tetramer, C-terminal carboxamide 11; (D) triazole trimer,
triazole tetramers of the synthetic AFGP are depicted in Scheme 5.
C-terminal alkyne 7b; (E) triazole pentamer, C-terminal alkyne 9b; (F)
triazole polymeric fraction (n ¼ 6–17). All compounds were assayed in
water at 10 mg mLꢁ1
.
Evaluation of the antifreeze activity of the oligomers
All glycopeptides were assessed for thermal hysteresis (TH)
activity using nanoliter osmometry39 and ice recrystallization
inhibition (IRI) activity using a “splat cooling” assay as was
described earlier.40,41 No TH activity or dynamic ice shaping
capabilities were observed for these oligomers, as the water
froze in spherical form (center of the images), indicating that
similar to the reference compound OGG-Gal these
compounds do not interact with the ice lattice (Fig. 6).15
1
The IRI activity of the glycopeptides in Scheme 5 is shown in
Fig. 7. The triazole containing oligomer 8b was only moderately
IRI active at 5.5 mM and was substantially less active than the
related C-AFGP OGG-Gal 1, even at a 1000-fold higher concen-
tration. OGG-Gal 1 exhibits equipotent IRI activity to native AFGP-
8 at a signicantly lower concentration of 5.5 mM. These results
indicate that the replacement of the peptide bonds with triazole
linkages and loss of the C-terminal negative charge in the poly-
peptide backbone of OGG-Gal 1 is detrimental and greatly reduces
the potent IRI activity of this C-AFGP analogue. From the data
presented in Fig. 7, it is also clear that the C-terminus also plays a
dominant role in the IRI activity. Compound 11, which contains a
C-terminal carboxamide and is slightly shorter than the C-
terminus of compound OGG-Gal 1, displayed a similar reduction
Scheme 4 (a) Ethyl azidoacetate, CuSO4$5H2O, sodium ascorbate,
H2O/THF and (b) NaOMe in MeOH.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
Fig. 8 Far-UV CD spectrogram comparisons: compound 1 (blue) and
11 (red); compound 10 (green) and 8b (purple).
the case of mimics 1 and 11 to 202 nm in the case of 8b and 10.
Calculation of the secondary structure element contributions to
the CD spectrum was performed using the CDPro package.9,42
There was no strong consensus among the three soware tools,
suggesting difficulty in matching the measurements to the
datasets. The SELCON3 tool, which gave the most consistent
results, showed similar secondary structures for compounds 8b,
10 and 11, while the more active compound 1 displayed a
slightly less unordered structure and helical contributions and
slightly more sheet-, turn- and PP2 structural contributions
than the other three compounds as analysed by comparison
with two datasets of standard proteins. Although further
different conformational organizations of these peptides on the
ice interface upon freezing cannot be excluded, a large struc-
tural difference when substituting amide bonds by triazole
moieties is the loss of a hydrogen bond donor per substituted
amide bond. Potentially this loss leads to a different degree of
hydration and/or different interactions of the backbone with the
carbohydrate part of the antifreeze glycopeptide.
Fig. 7 Ice recrystallization inhibition (IRI) activity of oligomers 8b, 10,
and 11 at 5.5 mM, and OGG-Gal 1 and AFGP-8 at 5.5 mM in PBS. The %
MGS (mean grain size) relative to a PBS positive control is shown for all
compounds.
in IRI activity to the triazole with an alkyne terminus 8b. Therefore
we investigated compound 10, which is a better mimic of refer-
ence compound 1, since it also contains the C-terminal glycine-
like moiety. Unfortunately, this modication did not restore the
IRI activity back to a similar level to compound 1, and it still
displayed a similar reduced IRI activity to compound 8b, sug-
gesting that the presence of the triazole rings in AFGP-mimics 8b
and 10 is responsible for the majority of IRI activity loss. The
higher oligomeric fraction of 8b (n > 6) yielded no signicant
increase in IRI activity over the tetramer 8b (data not shown),
suggesting that the limited size alone was not the reason for lower
activity of the triazole-containing oligomers.
Conclusions
The synthesis of the simplied azide/alkyne glycopeptide
building block described here was found to be efficient and
simple. The use of the TIPS protecting group allowed long term
CD spectroscopy
To investigate whether this loss of IRI activity was caused by a storage and its removal was very facile.
change in the conformation of the backbone of the triazole
The oligomerization approach with the in situ generation of a
containing AFGP, CD spectroscopy was performed. In view of small amount of terminating monomer worked conveniently and
the environment in which these peptides are normally active, separation of the individual length oligomers by HPLC was
measurements were performed in water at 4 ꢀC.11 Interestingly, feasible. While such separation might not be necessary or even
in spite of the differences in IRI activity, the CD spectra of the ultimately desirable for use as antifreeze, for the purpose of this
four AFGP mimics 8a, 10, 11 and 1 were very similar (Fig. 8).
study the focus was a comparison with natural AFGP-8 and earlier
Strikingly, the peptide OGG-Gal 1 and the amidated mimic made analogues. When a strictly dened length of the oligomers is
11 showed a nearly identical CD spectrum, even though these not required, the deacetylated derivative of 5a could in principle be
compounds differed by 3 orders of magnitude in IRI activity. used as a monomer for the polymerization, eliminating the need
Similarly, the difference in the CD spectrum between the two for the more laborious deactylation of the resulting polymers.
triazole linked AFGPs 8b and 10 is small. From this, it can be
Comparison of 8b with natural AFGP-8, or the previously pub-
concluded that the negative charge on the C-terminus does not lished OGG-Gal 1, showed modest IRI activity of the C-terminal
inuence the secondary structure of the peptide to a large alkyne triazole analogue 8b. Interestingly, the C-terminus plays a
degree. This is in contrast to the observed effect of the dominant role in the IRI activity of C-AFGPs, and removal of the C-
C-terminus on the IRI activity, especially in the case of the terminal acid in the reference compound OGG-Gal 1 caused a 1000
mimics 1 and 11. On the whole, the conformational effect of the fold reduction in IRI activity, similar to that observed in the alkyne-
triazole on the structure of the AFGPs seemed limited; never- triazole compound 8b. The introduction of a C-terminal charge on
theless a shi of the maximum [q] was observed from 212 nm in the triazole analogue 8b to give structure 10 unfortunately did not
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
restore the activity. Although the secondary structure of mimics 8b 18 B. H. M. Kuijpers, S. Groothuys, A. C. Soede, P. Laverman,
and 10 resembled the structure of the parent AFGPs 1 and 11 as
assayed with CD spectroscopy, small differences in the solution
O. C. Boerman, F. L. van Del and F. P. J. T. Rutjes,
Bioconjugate Chem., 2007, 18, 1847–1854.
conformation, molecular structure or physical interaction during 19 I. E. Valverde, A. Bauman, C. a. Kluba, S. Vomstein,
the organization on the water–ice interfaces can apparently lead to
a different IRI activity that is hard to predict based on CD
measurements or structural features.43
M. a. Walter and T. L. Mindt, Angew. Chem., Int. Ed. Engl.,
2013, 52, 8957–8960.
20 W. S. Horne, M. K. Yadav, C. D. Stout and M. R. Ghadiri,
J. Am. Chem. Soc., 2004, 126, 15366–15367.
21 V. D. Bock, D. Speijer, H. Hiemstra and J. H. van Maarseveen,
Org. Biomol. Chem., 2007, 5, 971–975.
Acknowledgements
We gratefully acknowledge Britta Ramakers (Radboud University 22 N. G. Angelo and P. S. Arora, J. Org. Chem., 2007, 72, 7963–
Nijmegen, The Netherlands) for her kind assistance with CD 7967.
measurements, carried out in the Bio-organic chemistry group of 23 R. C. Elgersma, M. van Dijk, A. C. Dechesne, C. F. van
Prof. J. C. M. van Hest. We also gratefully acknowledge Dr Johan
Kemmink (Medicinal Chemistry, Universiteit Utrecht) for his
Nostrum, W. E. Hennink, D. T. S. Rijkers and
R. M. J. Liskamp, Org. Biomol. Chem., 2009, 7, 4517–4525.
assistance with NMR measurements. C.J.C. acknowledges 24 M. van Dijk, M. L. Nollet, P. Weijers, A. C. Dechesne,
Canadian Blood Services (CBS) for a Graduate Fellowship Award.
C. F. van Nostrum, W. E. Hennink, D. T. S. Rijkers and
R. M. J. Liskamp, Biomacromolecules, 2008, 9, 2834–2843.
25 M. van Dijk, K. Mustafa, A. C. Dechesne, C. F. van Nostrum,
W. E. Hennink, D. T. S. Rijkers and R. M. J. Liskamp,
Biomacromolecules, 2007, 8, 327–330.
Notes and references
1 A. L. DeVries and D. E. Wohlschlag, Science, 1969, 163, 1073–
1075.
26 M. van Dijk, C. F. van Nostrum, W. E. Hennink,
D. T. S. Rijkers and R. M. J. Liskamp, Biomacromolecules,
2010, 11, 1608–1614.
2 S. Venketesh and C. Dayananda, Crit. Rev. Biotechnol., 2008,
28, 57–82.
3 A. L. DeVries, S. K. Komatsu and R. E. Feeney, J. Biol. Chem., 27 M. Hachisu, H. Hinou, M. Takamichi, S. Tsuda, S. Koshida
1970, 245, 2901–2908.
and S.-I. Nishimura, Chem. Commun., 2009, 1641–1643.
4 A. L. DeVries, J. Vandenheede and R. E. Feeney, J. Biol. Chem., 28 C. J. Capicciotti, J. F. Trant, M. Leclere and R. N. Ben,
1971, 246, 305–308.
Bioconjugate Chem., 2011, 22, 605–616.
5 J. A. Raymond and A. L. DeVries, Proc. Natl. Acad. Sci. U. S. A., 29 A. S. Norgren, C. Budke, Z. Majer, C. Heggemann, T. Koop
`
1977, 74, 2589–2593.
6 C. Knight and J. Duman, Cryobiology, 1986, 262, 256–262.
7 C. J. Capicciotti, M. Doshi, and R. N. Ben, in Recent
and N. Sewald, Synthesis, 2009, 488–494.
30 N. Miller, G. M. Williams and M. A. Brimble, Org. Lett., 2009,
11, 2409–2412.
Developments in the Study of Recrystallization, ed. P. Wilson, 31 M. Meldal and C. W. Tornøe, Chem. Rev., 2008, 108, 2952–3015.
2013, pp. 177–224.
32 V. R. Bouvet, G. R. Lorello and R. N. Ben, Biomacromolecules,
`
8 M. Leclere, B. K. Kwok, L. K. Wu, D. S. Allan and R. N. Ben,
2006, 7, 565–571.
Bioconjugate Chem., 2011, 22, 1804–1810.
33 E. D. Goddard-Borger and R. V Stick, Org. Lett., 2007, 9, 3797–
9 R. Y. Tam, C. N. Rowley, I. Petrov, T. Zhang, N. A. Afagh,
3800.
T. K. Woo and R. N. Ben, J. Am. Chem. Soc., 2009, 131, 34 C. Ruecker, Chem. Rev., 1995, 1009–1064.
15745–15753.
35 I. E. Valverde, A. F. Delmas and V. Aucagne, Tetrahedron,
2009, 65, 7597–7602.
10 B. L. Wilkinson, R. S. Stone, C. J. Capicciotti, M. Thaysen-
Andersen, J. M. Matthews, N. H. Packer, R. N. Ben and 36 C. Hayward, D. Yohannes and S. Danishefsky, J. Am. Chem.
R. J. Payne, Angew. Chem., Int. Ed. Engl., 2012, 51, 3606–3610.
Soc., 1993, 115, 9345–9346.
11 Y.Tachibana,G.L.Fletcher,N.Fujitani,S.Tsuda,K.Mondeand 37 A. Smith III, S. Chen, F. Nelson, J. Reichert and B. Salvatore,
S.-I. Nishimura, Angew. Chem., Int. Ed. Engl., 2004, 43, 856–862.
J. Am. Chem. Soc., 1995, 7863, 10935–10946.
12 R. Izumi, T. Matsushita, N. Fujitani, K. Naruchi, H. Shimizu, 38 E. Scriven and K. Turnbull, Chem. Rev., 1988, 88, 297–368.
S. Tsuda, H. Hinou and S.-I. Nishimura, Chem. – Eur. J., 2013, 39 A. Chakrabartty and C. L. Hew, Eur. J. Biochem., 1991, 202,
19, 3913–3920.
1057–1063.
13 R. N. Ben, A. A. Eniade and L. Hauer, Org. Lett., 1999, 1, 40 J. Jackman, M. Noestheden, D. Moffat, J. P. Pezacki,
1759–1762.
S. Findlay and R. N. Ben, Biochem. Biophys. Res. Commun.,
2007, 354, 340–344.
41 C. A. Knight, J. Hallett and A. L. DeVries, Cryobiology, 1988,
25, 55–60.
42 N. Sreerama, S. Y. Venyaminov and R. W. Woody, Anal.
Biochem., 2000, 287, 243–251.
14 A. K. Balcerzak, S. S. Ferreira, J. F. Trant and R. N. Ben,
Bioorg. Med. Chem. Lett., 2012, 22, 1719–1721.
15 P. Czechura, R. Y. Tam, E. Dimitrijevic, A. V Murphy and
R. N. Ben, J. Am. Chem. Soc., 2008, 130, 2928–2929.
16 T. Tsuda and S.-I. Nishimura, Chem. Commun., 1996, 2779–2780.
17 T. Pirali, F. Pagliai, C. Mercurio, R. Boggio, P. L. Canonico, 43 L. Corcilius, G. Santhakumar, R. S. Stone, C. J. Capicciotti,
G. Sorba, G. C. Tron and A. A. Genazzani, J. Comb. Chem.,
2008, 10, 624–627.
S. Joseph, J. M. Matthews, R. N. Ben and R. J. Payne,
Bioorg. Med. Chem., 2013, 21, 3569–3581.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.