Synthetic Receptors for the High-Affinity Recognition of O-GlcNAc Derivatives

Article abstract of DOI:10.1002/anie.201510611

The combination of a pyrenyl tetraamine with an isophthaloyl spacer has led to two new water-soluble carbohydrate receptors ("synthetic lectins"). Both systems show outstanding affinities for derivatives of N-acetylglucosamine (GlcNAc) in aqueous solution. One receptor binds the methyl glycoside GlcNAc-β-OMe with K_a≈20 000 m^-1, whereas the other one binds an O-GlcNAcylated peptide with K_a≈70 000 m^-1. These values substantially exceed those usually measured for GlcNAc-binding lectins. Slow exchange on the NMR timescale enabled structural determinations for several complexes. As expected, the carbohydrate units are sandwiched between the pyrenes, with the alkoxy and NHAc groups emerging at the sides. The high affinity of the GlcNAcyl-peptide complex can be explained by extra-cavity interactions, raising the possibility of a family of complementary receptors for O-GlcNAc in different contexts. Two carbohydrate receptors (see scheme) were prepared directly from a pyrenyl tetraamine and an isophthaloyl derivative. With extended aromatic surfaces linked by polar spacers, they are well suited to binding all-equatorial monosaccharides, showing especially high affinities for β-linked N-acetylglucosamine (GlcNAc).

Full text of DOI:10.1002/anie.201510611

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201510611
German Edition: DOI: 10.1002/ange.201510611
Biomimetic Hosts
Synthetic Receptors for the High-Affinity Recognition of O-GlcNAc
Derivatives
Pablo Rios, Tom S. Carter, Tiddo J. Mooibroek,* Matthew P. Crump, Micke Lisbjerg,
Michael Pittelkow, Nitin T. Supekar, Geert-Jan Boons, and Anthony P. Davis*
Abstract: The combination of a pyrenyl tetraamine with an
isophthaloyl spacer has led to two new water-soluble carbohy-
drate receptors (“synthetic lectins”). Both systems show out-
standing affinities for derivatives of N-acetylglucosamine
(GlcNAc) in aqueous solution. One receptor binds the
methyl glycoside GlcNAc-b-OMe with Ka ꢀ 20000mÀ1
,
whereas the other one binds an O-GlcNAcylated peptide
with Ka ꢀ 70000mÀ1. These values substantially exceed those
usually measured for GlcNAc-binding lectins. Slow exchange
on the NMR timescale enabled structural determinations for
several complexes. As expected, the carbohydrate units are
sandwiched between the pyrenes, with the alkoxy and NHAc
groups emerging at the sides. The high affinity of the
GlcNAcyl–peptide complex can be explained by extra-cavity
interactions, raising the possibility of a family of complemen-
tary receptors for O-GlcNAc in different contexts.
GlcNAc or O-GlcNAc) unit 1. This moiety is a dynamic post-
translational modification of proteins that is involved in many
cellular processes and has been linked to major diseases such
as diabetes and Alzheimerꢀs disease.[5] There is much interest
in agents that bind O-GlcNAc for use in detection and
separation methods for modified proteins.[6] A few years ago,
we reported that the tricyclic receptor 4 (Figure 1) binds the
simplest O-GlcNAc model 2 with Ka = 630mÀ1, and the
O-GlcNAcylated peptide 3 with Ka = 1000mÀ1 [4e]
These
.
affinities were encouraging and competitive with the
GlcNAc-binding lectin wheat germ agglutinin (WGA), but
probably insufficient for most biological applications. We now
describe two new systems, which show strongly increased
binding to 2 and 3. These new synthetic lectins are accessible
in just eight synthetic steps and may point to a practical
solution to O-GlcNAc recognition in glycobiology.
B
inding carbohydrates in water is a notoriously difficult
challenge.[1] Saccharides are polar units that are both hydro-
philic (therefore happy to remain in water) and hydromimetic
(hard to distinguish from aqueous solvent). As a result, both
natural and synthetic receptors tend to show low affinities,
especially towards neutral carbohydrates. Lectins, the major
class of carbohydrate-binding proteins, often bind monosac-
The conceptual basis for this work is summarized in
Figure 1. Synthetic lectin 4 is composed of parallel aromatic
units (blue) held apart by polar spacers (red). The former can
experience hydrophobic/CH–p interactions with axial CH
groups in all-equatorial substrates while the spacers can
hydrogen-bond to equatorial substituents. The cartoon used
to describe this arrangement (Figure 1, top) inspired the
“temple” description for this family of molecules.[1f] In 4, the
hydrophobic surfaces are provided by biphenyls, which are
readily introduced and provide adequate CH–p interactions.
However, biphenyls are prone to twisting, which disrupts
contact with axial CH groups, while condensed aromatic
compounds are planar.[4a,b] A 1,3,6,8-tetrasubstituted pyrene
is geometrically equivalent to the biphenyls in 4, so tricyclic
cage 5 was identified as a promising design. Based on recent
work,[3a,4a] receptor 5 was provided with more powerful water-
solubilizing groups than 4; pyrenes are strongly inclined to
self-associate in water, and this needed to be countered for
reliable binding characterization.
charides with Ka < 103 mÀ1 [2]
Biomimetic analogues (“syn-
.
thetic lectins”) have shown Ka values of approximately 104 mÀ1
for charged substrates,[3] but well-characterized binding to
neutral monosaccharides in water is generally much weak-
er.[1b–g]
We have previously described synthetic lectins that bind
all-equatorial monosaccharides (glucose and close relatives)
with affinities from approximately 10 to 600mÀ1 [3a,4]
A
.
particular target has been the b-N-acetylglucosaminyl (b-
[*] P. Rios, T. S. Carter, Dr. T. J. Mooibroek, Prof. M. P. Crump,
M. Lisbjerg, Prof. A. P. Davis
School of Chemistry, University of Bristol
Cantock’s Close, Bristol, BS8 1TS (UK)
E-mail: tiddomooibroek@gmail.com
Anthony.Davis@bristol.ac.uk
M. Lisbjerg, Prof. M. Pittelkow
Department of Chemistry, University of Copenhagen
Universitetsparken 5, 2100 Copenhagen Ø (Denmark)
Whereas the design of 5 might seem obvious, the synthesis
appeared problematic. Retrosynthesis leads to tetraamine 7
and an isophthaloyl derivative such as 8, but in the forward
direction this combination could lead to a variety of products
including a second cage 9 (see Figure 1). It was not clear
whether 5 and 9 could be formed in good yield, then
separated, and distinguished from each other. Nonetheless,
N. T. Supekar, Prof. G.-J. Boons
Complex Carbohydrate Research Center
University of Georgia
315 Riverbend Road, Athens, GA 30602 (USA)
Supporting information and ORCID(s) from the author(s) for this
article are available on the WWW under http://dx.doi.org/10.1002/
anie.201510611.
Angew. Chem. Int. Ed. 2016, 55, 3387 –3392
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3387

Angewandte
Communications
Chemie
Figure 1. Design and synthetic strategy. Top middle: Cartoon showing interactions in “temple” synthetic lectins for all-equatorial carbohydrates.
Anticlockwise from the top left: The successful design 4 can be improved by converting the biphenyls into pyrenes, giving receptor 5. Receptor 5
can be deconstructed into 7 and 8, but in the forward direction, this combination leads to both 5 and 9. The external groups X1 (in 4) were
expanded to X2 (in 5 and 9) to ensure water solubility.
in the absence of an alternative, we chose to make an attempt.
As shown in Scheme 1, tetraamine 7 was prepared as its
hydrochloride salt in 42% overall yield by a five-step
procedure from pyrene 11.[7] Tetraamine 7 was insoluble in
organic solvents, but the hydrochloride 7·4HCl could be
dissolved in THF/water (5:1). This medium was used to
convert 7 into 12 and thence to a mixture of the cages 6 and
10, which were separable by HPLC. As anticipated, the NMR
spectra of the two cages were almost identical. However,
a NOESY spectrum of a 1:1 mixture showed a difference
between the pyrene p2/p4 cross peaks for the two isomers (for
numbering, see Figure 1 and also the Supporting Information,
Scheme S1). For the more quickly eluted isomer, this cross
peak was significantly larger, as expected for “staggered” cage
10, and provisional assignments were made on this basis.[7]
Samples of 6 and 10 were treated with TFA, dried, and
dissolved in NaOD/D2O (pH 7) to give solutions of 5 and 9
for NMR analysis.[7] 1H NMR spectra of eclipsed receptor 5
appeared broadened at 258C and unexpectedly complex,
implying a less symmetric structure than the D2h cage. The
spectra were unchanged between 16 and 500 mm, suggesting
that 5 is monomeric within this concentration range. Diffu-
sion-ordered NMR spectroscopy (DOSY) implied a hydro-
dynamic diameter of 2.1–3.8 nm, which is also consistent with
monomeric 5. Upon warming to 808C the spectra sharpened
and simplified, becoming consistent with the D2h structure.
We concluded that the room-temperature spectra are affected
Scheme 1. Synthesis of the protected receptors 6 and 10. Reaction
by slow conformational exchange, with a ground-state
conformation of low symmetry. The behavior of 9 was similar,
except that the spectra at 258C showed minor concentration
dependence. However, any aggregation appeared to be
conditions: a) Br2, PhNO2, 96%; b) BuOH, CO (30 bar), DIPEA,
Pd(OAc)2, BINAP, xylenes, 89%; c) LiBH4, THF, MeOH, ca. 100%;
d) (PhO)2PON3, DMF, DBU, 52%; e) Ph3P, THF, H2O; then HCl (aq.),
93%; f) DIPEA, H2O, THF, 47%; g) DIPEA, H2O, THF, 57%.
3388
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 3387 –3392

Angewandte
Communications
Chemie
limited as DOSY was again consistent with the presence of
a monomeric receptor (hydrodynamic diameter: 2.6–3.5 nm).
Both 5 and 9 were studied as carbohydrate receptors by
1H NMR titrations in D2O.[7] In most cases, the host spectra
showed significant changes upon carbohydrate addition. For
many substrates, the spectra implied complex formation at
1
fast or intermediate rates on the H NMR timescale. Given
the complexity of the host spectra, the changes were difficult
to interpret, and no attempt was made to quantify binding in
these cases. However, for some substrates, a new set of signals
appeared, implying binding with slow exchange on the NMR
timescale. In these cases, integration of the new signals could
be used with confidence to follow the concentrations of the
complexes. The data could be fitted to a 1:1 binding model to
give the binding constants listed in Table 1.[7,8] Supporting
values were obtained by isothermal titration calorimetry
(ITC).[7] The NMR spectra, binding analysis curves, and ITC
data for staggered receptor 9 and GlcNAc derivative 2 are
shown in Figure 2. A few combinations that were quantified
solely by ITC are also included in Table 1.
Table 1: Association constants (Ka) for 1:1 complexes of the receptors 5
and 9 with carbohydrates in water, as determined by 1H NMR
spectroscopy and ITC. Data for 4 are listed for comparison.[a]
Carbohydrate
Ka [mÀ1] determined by NMR (ITC)
4[b]
5
9
GlcNAc-b-OMe (2)
GlcNAc-a-OMe (13)
GlcNAc (14)
methyl b-d-glucoside (15)
d-glucose (16)
630
2100 (2200)
18200 (16600)
1550 (1520)
520 (520)
1180 (1230)
– (190)
[d]
24[c]
56
28
9
–
–
[d]
– (1440)
– (120)
[a] T=298 K, errors estimated from curve fitting: ꢁ5%.[7] NMR experi-
ments were performed in D2O and ITC titrations in H2O. Values for
reducing sugars are weighted averages of those for the two anomers, as
discussed in Ref. [4c]. [b] See Ref. [4e]. [c] Measured by induced circular
dichroism. [d] Could not be determined accurately.
Figure 2. Binding studies of receptor 9 with GlcNAc-b-OMe (2).
a) Partial spectra from the 1H NMR titration in D2O. Signals due to 9
were replaced by new ones assigned to the complex. b) Binding curve
based on integrating the signal at 8.8 ppm versus those in the region
of 7–9.5 ppm. Ka =18200mÀ1. c) ITC output and binding analysis plot
for a titration in water. Ka =16600mÀ1
.
As shown in Table 1, the binding studies with 5 and 9
yielded some exceptionally high affinities. As expected, the
eclipsed cage 5 (our initial target) was found to be a good
receptor for the simple O-GlcNAc model compound 2,
record).[4a] Surprisingly, a-GlcNAc derivative 13, with an axial
OMe group, was also bound fairly strongly. The selectivity of
5 and 9 against other common monosaccharides was difficult
to assess, as affinities for the latter could not be measured by
NMR spectroscopy (see above). However, ITC measure-
ments on 5 and 9 with mannose, galactose, N-acetylgalactos-
amine, and N-acetylmannosamine gave low thermal output,
suggesting weak binding at best.[10]
As illustrated in Figure 2a, binding of carbohydrates by 5
and 9 with slow exchange on the NMR timescale yields
complex spectra owing to desymmetrization of the receptors
by the asymmetric substrates. We have previously shown for
4 + 2 that such spectra can be exploited to obtain detailed 3D
structures.[4e] Here, we were interested both in determining
the mode of binding and confirming the identities of the
showing
a three-fold increase in binding power over
biphenyl-based prototype 4. However, the staggered cage 9
proved even more effective. Receptor 9 bound O-GlcNAc
model compound 2 with a Ka value of 18200mÀ1, which is
30 times larger than that of 4 and 25 times larger than that of
the natural lectin WGA (Ka = 730mÀ1).[4e, 9] To the best of our
knowledge, this is the highest affinity that has been measured
for a biomimetic receptor binding a small, electrically neutral
monosaccharide derivative in water. Receptor 9 also showed
good affinities for GlcNAc 14 (10 times higher than 4)[4e] and
methyl b-d-glucoside (15; 5 times higher than the previous
Angew. Chem. Int. Ed. 2016, 55, 3387 –3392
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3389

Angewandte
Communications
Chemie
receptors. Desymmetrization of the receptor cores would
allow connections to be followed within the macropolycyclic
frameworks, distinguishing between the two possibilities.
Complexes 5·2, 9·2, and 9·13 were subjected to detailed
NMR analysis by NOESY, TOCSY, and COSY.[7] In all cases,
full assignments could be obtained for the carbohydrate and
receptor core protons. The structures of 5 and 9 were indeed
confirmed unambiguously. In the former case, the p4 and s2
signals (see Figure 1) formed two discrete networks, whereas
in the latter case, all of these signals were connected in
a single sequence. Integrated NOESY signals were employed
to build models of all three complexes. For each structure,
> 70 cross peaks were employed to obtain inter- and intra-
molecular distances. These were used as constraints to obtain
the energy-minimized structures shown in Figure 3.[11] As
expected, the carbohydrates are sandwiched between the
pyrene units with the NHAc and OMe groups emerging from
the portals. This arrangement is reflected in remarkable
upfield shifts for the carbohydrate protons, approaching
5 ppm for some CH protons (see Table S2). Interestingly,
the data for 9·13 are most consistent with a twist-boat
structure for the carbohydrate, which is consistent with the
all-equatorial preference of the temple architecture (see
Figure 1).
As mentioned earlier, we had previously shown that
biphenyl-based receptor 4 is capable of binding glycopeptide
3 with Ka = 103 mÀ1. It was interesting to determine whether 5
and 9 would show higher affinities for this substrate. An NMR
titration of 9 and 3 suggested binding that was fast on the
NMR timescale and could not be quantified. However,
addition of 3 to eclipsed receptor 5 in D2O at pH 7 yielded
new signals, implying binding with slow exchange.[7] The
growth of these signals was analyzed as above to give
a remarkable binding constant of Ka = 67000mÀ1 [12]
An
.
NMR structure was determined for the core region of the
complex (macrotricycle+GlcNAc) based on 89 integrated
signals from a 900 MHz NOESY spectrum (see Figure 4).[7,10]
As expected, the orientation of the O-GlcNAc unit in the
cavity is essentially similar to that in 5·2. The positioning of
the peptide backbone could not be determined, as the signals
could not be assigned owing to overlap between complex and
unbound glycopeptide. However, Figure 4 shows a reasonable
option derived from energy minimization. The structure
reveals that the peptidic portion of 3 can undergo various
interactions with the receptor, including the dendritic side
chains, presumably accounting for the extra affinity.
The high affinity of 5 for 3 is notable on two counts. First,
it approaches the level required for practical applications,
such as staining O-GlcNAcylated proteins in Western blotting
(estimated at ca. 105 mÀ1).[13] Second, it shows that binding to
O-GlcNAc is context-selective; the binding to simple model
compound 2 is much weaker, and it is only when the
additional interactions to the peptide are included that
these high levels are achieved. Receptors 5 and 9 provide
scope for modifying the side chains, and the synthetic method
should be adaptable to related core structures with alternative
roof/floor combinations. This raises the prospect of a family of
O-GlcNAc-binding receptors with useful affinities and com-
plementary selectivities.
In conclusion, we have shown that polycyclic synthetic
lectins may be synthesized from rigid, condensed aromatic
components by a short and efficient procedure. The method
may tend to yield mixtures, but separation was feasible in the
present case, and all cage products are likely to have
interesting binding properties. The pyrene-based cages pre-
pared in this work are outstanding receptors for O-GlcNAc
derivatives. Staggered receptor 9 binds the simple glycoside 2
with Ka ꢀ 20000mÀ1, a 25-fold advance on the lectin most
commonly used to bind GlcNAc units (WGA). Eclipsed
receptor 5 is less effective for 2 but shows even stronger
binding to O-GlcNAc glycopeptide 3 (Ka ꢀ 70000mÀ1). These
results further demonstrate that biomimetic carbohydrate
receptors can show affinities comparable to those of natural
lectins and may point to a general strategy for discovering
tools for studying O-GlcNAc and other forms of glycosyla-
tion.
Figure 3. NMR structures of a) eclipsed receptor 5 and GlcNAc-b-OMe
(2), b) staggered receptor 9 and GlcNAc-b-OMe (2), and c) staggered
receptor 9 and GlcNAc-a-OMe (13). Pyrene units are shown in cyan
and carbohydrates in pink, with OMe and NHAc groups highlighted in
yellow. Hydrogen bonds are shown as green dashed lines. The water-
solubilizing side chains are omitted.
3390
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 3387 –3392

Angewandte
Communications
Chemie
Figure 4. NMR structure of the complex between eclipsed receptor 5 and glycopeptide 3. Color coding as in Figure 3 except that the peptide
backbone is shown in orange and the receptor side chains in pale green. The positioning of the glycopeptide is indicated in the drawing.
[3] a) H. Destecroix, C. M. Renney, T. J. Mooibroek, T. S. Carter,
Acknowledgements
P. F. N. Stewart, M. P. Crump, A. P. Davis, Angew. Chem. Int. Ed.
2015, 54, 2057; Angew. Chem. 2015, 127, 2085; b) Y. Jang, R.
Natarajan, Y. H. Ko, K. Kim, Angew. Chem. Int. Ed. 2014, 53,
1003; Angew. Chem. 2014, 126, 1021; c) M. Rauschenberg, S.
Bandaru, M. P. Waller, B. J. Ravoo, Chem. Eur. J. 2014, 20, 2770;
This research was supported by the Engineering and Physical
Sciences Research Council (EP/I028501/1), a studentship
funded by the Bristol Chemical Synthesis Doctoral Training
Centre (EP/G036764/1) to T.S.C., and by the Wellcome Trust
(099185/Z/12/Z). M.L. and M.P. thank the Augustinus
Fonden, the Oticon Fonden, the Lundbeck Foundation, and
COST Action CM1005 for financial support. The Secretaría
General de Universidades, Investigación y Tecnología (Junta
de Andalucía) is acknowledged for a Talentia Scholarship to
P. R. We also thank HWB-NMR at the University of
Birmingham for providing open access to their Wellcome
Trust funded 900 MHz spectrometer and Dr. Sara Whittaker
for assistance in using this instrument. M.P. thanks the Danish
Ministry of Science, Innovation and Higher Education for
a Sapere Aude Fellowship (DFF-4181-00206).
d) C. Geffert, M. Kuschel, M. Mazik, J. Org. Chem. 2013, 78, 292;
e) C. Schmuck, M. Schwegmann, Org. Lett. 2005, 7, 3517.
[4] See, for example: a) T. J. Mooibroek, J. M. Casas-Solvas, R. L.
Harniman, C. M. Renney, T. S. Carter, M. P. Crump, A. P. Davis,
Nat. Chem. 2016, 8, 69; b) C. Ke, H. Destecroix, M. P. Crump,
A. P. Davis, Nat. Chem. 2012, 4, 718; c) N. P. Barwell, M. P.
Crump, A. P. Davis, Angew. Chem. Int. Ed. 2009, 48, 7673;
Angew. Chem. 2009, 121, 7809; d) F. Corzana, A. Fernandez-
Tejada, J. H. Busto, G. Joshi, A. P. Davis, J. JimØnez-Barbero, A.
Avenoza, J. M. Peregrina, ChemBioChem 2011, 12, 110; e) Y.
Ferrand, E. Klein, N. P. Barwell, M. P. Crump, J. JimØnez-
Barbero, C. Vicent, G. J. Boons, S. Ingale, A. P. Davis, Angew.
Chem. Int. Ed. 2009, 48, 1775; Angew. Chem. 2009, 121, 1807;
f) Y. Ferrand, M. P. Crump, A. P. Davis, Science 2007, 318, 619;
g) E. Klein, M. P. Crump, A. P. Davis, Angew. Chem. Int. Ed.
2005, 44, 298; Angew. Chem. 2005, 117, 302.
[5] G. W. Hart, M. P. Housley, C. Slawson, Nature 2007, 446, 1017;
W. B. Dias, G. W. Hart, Mol. BioSyst. 2007, 3, 766; L. Wells, K.
Vosseller, G. W. Hart, Science 2001, 291, 2376.
[6] A. S. Vercoutter-Edouart, I. El Yazidi-Belkoura, C. Guinez, S.
Baldini, M. Leturcq, M. Mortuaire, A. M. Mir, A. Steenackers,
V. Dehennaut, A. Pierce, T. Lefebvre, Proteomics 2015, 15, 1039;
E. J. Kim, ChemBioChem 2015, 16, 1397; C. F. Teo, S. Ingale,
M. A. Wolfert, G. A. Elsayed, L. G. Not, J. C. Chatham, L. Wells,
G. J. Boons, Nat. Chem. Biol. 2010, 6, 338.
Keywords: biomimetic hosts · carbohydrates ·
molecular recognition · receptors · supramolecular chemistry
How to cite: Angew. Chem. Int. Ed. 2016, 55, 3387–3392
Angew. Chem. 2016, 128, 3448–3453
[1] a) B. Wang, G.-J. Boons, Carbohydrate Recognition: Biological
Problems, Methods and Applications, Wiley, Hoboken, 2011;
b) D. Solís, N. V. Bovin, A. P. Davis, J. JimØnez-Barbero, A.
Romero, R. Roy, K. Smetana, H. J. Gabius, Biochim. Biophys.
Acta Gen. Subj. 2015, 1850, 186; c) C. E. Miron, A. Petitjean,
ChemBioChem 2015, 16, 365; d) S. Jin, Y. F. Cheng, S. Reid,
M. Y. Li, B. H. Wang, Med. Res. Rev. 2010, 30, 171; e) M. Mazik,
Chem. Soc. Rev. 2009, 38, 935; f) A. P. Davis, Org. Biomol.
Chem. 2009, 7, 3629; g) A. P. Davis, R. S. Wareham, Angew.
Chem. Int. Ed. 1999, 38, 2978; Angew. Chem. 1999, 111, 3160.
[2] E. J. Toone, Curr. Opin. Struct. Biol. 1994, 4, 719.
[7] For further details see the Supporting Information.
[8] The NMR titrations with 5 were complicated by an unusual
phenomenon whereby some receptor molecules seemed initially
inactive, but were converted into active form(s) very slowly in
the presence of the guest. As discussed in the Supporting
Information, this behavior did not prevent the quantification of
the binding to the active conformer(s) and should have little
effect on the accuracy of the results.
Angew. Chem. Int. Ed. 2016, 55, 3387 –3392
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3391

Angewandte
Communications
Chemie
[9] A. K. Allen, A. Neuberger, N. Sharon, Biochem. J. 1973, 131,
155; I. J. Goldstein, S. Hammarstrom, G. Sundblad, Biochim.
Biophys. Acta Protein Struct. 1975, 405, 53.
a timescale of weeks. As discussed in the Supporting Informa-
tion, the measured binding constant should not be affected by
this phenomenon.
[10] Binding constants in the range of 20–30mÀ1 were estimated for
9+mannose, galactose, and N-acetylgalactosamine. However,
these low values, unsupported by another technique, must be
considered uncertain.
[13] T. S. Hnasko, R. M. Hnasko in ELISA, Methods in Molecular
Biology, Vol. 1318 (Ed.: R. Hnasko), Springer, New York, 2015,
p. 87.
[11] 3D structures (as cif files) are provided in the Supporting
Information.
[12] Again, the spectra implied that some receptor molecules were
inactive during the titration, but were converted into 5·3 over
Received: November 18, 2015
Published online: January 28, 2016
3392
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 3387 –3392