Protein Glycosylation through Sulfur Fluoride Exchange (SuFEx) Chemistry: The Key Role of a Fluorosulfate Thiolactoside

Article abstract of DOI:10.1002/chem.201803912

Protein glycosylation is the most complex post-translational modification process. More than 50 % of human cells proteins are glycosylated, whereas bacteria such as E. coli do not have this modification machinery. Indeed, the carbohydrate residues in natural proteins affect their folding, immunogenicity, and stability toward proteases, besides controlling biological properties and activities. It is therefore important to introduce such structural modification in bioengineered proteins lacking the presence of carbohydrate residues. This is not trivial as it requires reagents and conditions compatible with the protein's stability and reactivity. This work reports on the introduction of lactose moieties in two natural proteins, namely ubiquitin (Ub) and l-asparaginase II (ANSII). The synthetic route employed is based on the sulfur(VI) fluoride exchange (SuFEx) coupling of a lactose tethered arylfluorosulfate (Lact-Ar-OSO₂F) with the ?-NH₂ group of lysine residues of the proteins. This metal-free click SuFEx reaction relies on the properties of the fluorosulfate employed, which is easily prepared in multigram scale from available precursors and reacts chemoselectively with the ?-NH₂ group of lysine residues under mild conditions. Thus, iterative couplings of Lact-Ar-OSO₂F to Ub and ANSII, afforded multiple glycosylations of these proteins so that up to three and four Lact-Ar-OSO₂ groups were introduced in Ub and ANSII, respectively, via the formation of a sulfamoyl (OSO₂-NH) linkage.

Full text of DOI:10.1002/chem.201803912

A Journal of
Accepted Article
Title: Protein Glycosylation via Sulfur Fluoride Exchange (SuFEx)
Chemistry: The Key Role of a Fluorosulfate Thiolactoside
Authors: Cristina Nativi, Alberto Marra, Jiajia Dong, Tiancheng Ma,
Stefano Giuntini, Elisa Crescenzo, Linda Cerofolini, Marco
Martinucci, Claudio Luchinat, Marco Fragai, and Alessandro
Dondoni
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201803912
Link to VoR: http://dx.doi.org/10.1002/chem.201803912
Supported by

10.1002/chem.201803912
Chemistry - A European Journal
FULL PAPER
Protein Glycosylation via Sulfur Fluoride Exchange (SuFEx)
Chemistry: The Key Role of a Fluorosulfate Thiolactoside
A. Marra, [b] J. Dong, [c] T. Ma, [c] S. Giuntini, [a],[d] E. Crescenzo, [a],[d] L. Cerofolini, [a],[d] M. Martinucci, [a] C.
Luchinat, [a],[d] M. Fragai, *[a],[d] C. Nativi*[a] and A. Dondoni*[e]
Dedicated to Professor K. Barry Sharpless on the occasion of his 77th birthday
Abstract: Protein glycosylation is the most complex post-translational
modification process. More than 50% of human cells proteins are
glycosylated, while bacteria such as E. coli do not have this
modification machinery. Indeed, the carbohydrate residues in natural
proteins affect their folding, immunogenicity, and stability toward
proteases, besides controlling biological properties and activities. It is
therefore important to introduce such structural modification in
bioengineered proteins lacking the presence of carbohydrate residues.
This is not a trivial as it requires reagents and conditions compatible
with protein’s stability and reactivity. We report herein on the
introduction of lactose moieties in two natural proteins, namely
ubiquitin (Ub) and L-asparaginase II (ANSII). The synthetic route
employed is based on the Sulfur(VI) Fluoride Exchange (SuFEx)
coupling of a lactose tethered arylfluorosulfate (Lact-Ar-OSO2F) with
the -NH2 group of lysine residues of the proteins. This metal free click
SuFEx reaction relies on the properties of the fluorosulfate employed
which is easily prepared in multigram scale from available precursors
and reacts chemoselectively with the -NH2 group of lysine residues
under mild conditions. Thus, iterative couplings of Lact-Ar-OSO2F to
Ub and ANSII, afforded multiple glycosylations of these proteins so
that up to three and four Lact-Ar-OSO2 groups were introduced in Ub
and ANSII, respectively, via the formation of a sulfamoyl (OSO2-NH)
linkage.
Introduction
Protein glycosylation is an important post-translational
modification in eukaryotic organisms and is biosynthetically
mediated by numerous glycosyltransferases.[1] The extensive
occurrence of such modification is supported by the fact that 50%
of proteins in human cells are glycosylated[2] and 70% of the total
of therapeutic proteins, that are currently in clinical trials, are
glycoproteins.[3] Indeed the carbohydrate residues are not just
decorative elements as they profoundly affect protein folding,
immunogenicity, and stability toward proteases besides
controlling biological properties and activities.[4] The latter
functions operate mainly in cellular processes, such as cellular
recognition and adhesion, cell growth and differentiation.[5] In this
way glycoproteins are involved in vital biological events, either
detrimental (inflammation, cancer metastasis, viral and bacterial
infection),[6] or beneficial (immune response, fertilization). Studies
on these issues are made problematic because the isolation of
glycoproteins with a well-defined carbohydrate structure from
natural sources is difficult and even with current techniques
virtually impossible. This is due to two main factors, that is: i) the
oligosaccharide micro heterogeneity produced by the presence of
various glycoforms, and ii) the very labile N- and O-acetal bonds,
which are the most diffuse connections between the terminal
carbohydrate of the oligosaccharide fragment and asparagine or
serine/threonine residue, respectively. Therefore, it is difficult to
identify the specific oligosaccharide structures that are
responsible of an individual biological process or protein function.
Hence there is an urgent need for the preparation of non-natural
glycoproteins exhibiting robust anomeric linkages and a well-
defined structure and composition. Another reason for such
request comes from impact in medicine. In fact, it is not surprising
that a substantial fraction of the currently approved protein
pharmaceuticals need to be properly glycosylated to exhibit
optimal therapeutic efficacy.[7] In some cases the glycosylation of
natural proteins lacking carbohydrate fragments can lead to new
properties as precursors to valuable drugs or it is used to produce
glycovaccines.[8] Moreover, several studies demonstrated that
glycosylation induces various effects on the stability of protein
pharmaceuticals.[9] Therefore substantial efforts are being made
[a]
Dr. S. Giuntini, E. Crescenzo, Dr. L. Cerofolini, M. Martinucci, Prof.
C.Luchinat, Prof. M. Fragai, Prof. C. Nativi
Department of Chemistry
University of Florence
via della Lastruccia, 3-13 Sesto F.no (FI) 50019 Italy
E-mail: fragai@cerm.unifi.it; cristina.nativi@unifi.it
Prof. A. Marra
Institut des Biomolécules Max Mousseron (IBMM), UMR 5247,
Université de Montpellier, CNRS, Ecole Nationale Supérieure de
Chimie de Montpellier
[b]
[c]
8 Rue de l’Ecole Normale, 34296 Montpellier- cedex 5, France
Prof. J. Dong, T. Ma
Key Laboratory of Organofluorine Chemistry, Center for Excellence
in Molecular Synthesis, Shanghai Institute of Organic Chemistry
University of Chinese Academy of Sciences, Chinese Academy of
Sciences
345 Lingling Road, Shanghai 200032, P. R. China
Prof. C. Luchinat, Prof. M. Fragai, Dr. S. Giuntini, E. Crescenzo, Dr.
L. Cerofolini
CERM and CIRMMP
via Luigi Sacconi, 6, 50019 Sesto F.no (FI), Italy
Prof. A. Dondoni
[d]
[e]
Interdisciplinary Center for the Study of Inflammation
University of Ferrara,
Italy
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803912
Chemistry - A European Journal
FULL PAPER
for the efficient glycosylation of proteins,[10] and for developing
biophysical methodologies to characterize these protein
derivatives.[11] Actually glycosylation of proteins is not a trivial task
as it requires a chemoselective chemical ligation that introduces
the carbohydrate moieties under conditions which are compatible
with the protein environments such as room temperature and
aqueous medium, as well as the absence of a metal catalyst. The
wise choice in the arsenal of click reactions may offer suitable
solutions to those requests. Indeed, in addition to the prototypical
click reaction,[12] represented by the copper catalysed alkyne
azide cycloaddition,[13] many other efficient click processes are
available,[14] For instance, a few years ago another good process
for click chemistry was revived from an unrecognized state[15] by
Sharpless and co-workers.[16] This methodology is based on the
reactivity of sulfonyl fluorides and fluorosulfates with O- and N-
nucleophiles and therefore was named as sulfur fluoride
exchange (SuFEx) (Scheme 1).
protein.[20] Another modification of ANSII that turned out to be
beneficial on its properties as a drug is the glycation by lactose
mediated by sodium cyanoborohydride, i.e. a reductive coupling,
which was reported forty years ago.[21] In fact, the glycated
enzyme showed increased thermal stability and resistance to
proteolytic cleavage. These promising results support our
program of performing an approach for the glycosylation of ANSII
using new chemistry based on the click SuFEx strategy.
Results and Discussion
The initial part of our work consisted in the preparation of a
glycosylated reagent containing the SO2F functionality suitable for
the SuFEx reaction with the -amino group of lysine residues
natively present in Ub and ANSII. As recently the SuFEx reaction
of a polyethylene glycol fluorosulfate (PEG-OSO2F) with bovine
serum albumin (BSA) has been reported,[22] we decided to employ
a glycosylated fluorosulfate for our SuFEx approach .[23] No
examples of protein functionalization by fluorosulfate derivatives
of biomolecules (sugar, amino acids, etc.) are known to date.
Despite its strong potential, the SuFEx approach has been
scarcely exploited for the functionalization of proteins. In 2016,
the coupling of a single polyethylene glycol fluorosulfate (PEG-
OSO2F) with the lysine residues of commercially available bovine
serum albumin (BSA) was reported by Averick and co-workers.[22]
In the same year, the Sharpless group demonstrated that PEG-
fluorosulfate derivatives can react[23a] chemoselectively with the
phenolic hydroxyl of tyrosine residues within the binding site of
intracellular lipid binding proteins and, more recently, that various
aryl fluorosulfates show
a similar reactivity toward other
proteins.[23b] Therefore, no examples of protein functionalization
by fluorosulfate derivatives of biomolecules (e.g. sugar and amino
acids) are known to date. Since fluorosulfate bearing
carbohydrates have never been described, we envisaged the
synthesis of a reagent featuring a sugar unit anomerically linked
to the fluorosulfate functionality through a suitable aryl tether.
Hence the lactose fluorosulfate 3 was prepared in 82% yield by
photoinduced thiol-ene coupling (TEC)[24] of the 4-allyloxyphenyl
fluorosulfate[25] 2 with the readily accessible 1-thio--D-lactose[26]
Scheme 1. SuFEx reactions of fluoro derivatives with silyl ether and amine.
Applications of SuFEx chemistry dealing with small molecules
and biomacromolecules have been reported in recent papers.[17]
Based on these premises and given our interest in peptide and
protein glycosylation by click chemistry,[18] we report below on the
coupling of a glycosylated fluorosulfate to lysine residues of
ubiquitin (Ub) and E. Coli L-asparaginase II (ANSII). Ub is a 76-
amino acid protein, that incorporates seven lysine residues, and
which is well known for its role in post-translational modification
(PTM) of many proteins. The use of glycosylated Ub would
expand the scope of the ubiquitination as the modified protein
would be also affected by the presence of the carbohydrate
moiety. Also ANSII is a lysine rich enzyme, approved for the
clinical use against acute lymphoblastic leukaemia (ALL). This
tetrameric protein featuring a molecular mass of about 140 kDa,
is well known for its ability in catalysing the hydrolysis of
asparagine, an essential amino acid for protein biosynthesis in
leukemic cells, into ammonia and aspartate. Two main
disadvantages in the therapeutic use of this biological drug are
the need for frequent intramuscular injections and the very high
rate of allergic reactions. Fortunately, the modification of the drug
by pegylation (PEG) has been shown to reduce these
1
in the presence of 2,2-dimethoxy-2-phenylacetophenone
(DPAP) as photoinitiator (Scheme 2). Quite remarkably the
reagent 3 featured the sugar residue linked to the arylfluorosulfate
moiety through anomeric carbon-sulfur bond. This linkage would
provide the glycosylated products from the SuFEx coupling with
high stability toward hydrolases. Moreover, it turned out that the
lactose fluorosulfate 3 showed sufficient water solubility and
stability[27] to be safely used in reactions with the target proteins
in aqueous medium under mild basic conditions and at room
temperature.
inconveniences
while
retaining
the
anti-leukemic
effectiveness[.[19] and the original structural features of the
This article is protected by copyright. All rights reserved.

10.1002/chem.201803912
Chemistry - A European Journal
FULL PAPER
Scheme 4. Synthesis of the glycosylated peptide 7.
Scheme 2. Synthesis of the lactose fluorosulfate 3.
Conjugation of the lactose fluorosulfate with the proteins.
Having demonstrated that the fluorosulfate 3 is enough reactive
with the free -NH2 of lysine residues thus leading to a sulfamoyl
linkage, the SuFEx reaction of 3 was then carried out on Ub and
ANSII, as both these proteins bear lysine residues. In both cases,
the reactions were carried out in phosphate buffer (pH 8) and at
room temperature with an excess of fluorosulfate 3. The
glycosylation was monitored by gel electrophoresis and the
excess of 3 removed by dialysis. The product of reaction between
3 and Ub was investigated through MALDI mass spectrometry
and NMR for the determination of the number of derivative
molecules attached to each protein molecule, and the conjugation
pattern of each lysine residue.
In order to prove the SuFEx reaction of 3 with the -amino group
of lysine residues, two model reactions were carried out. The first
experiment involved the coupling of 3 with protected lysine.
Indeed, the reaction of 3 (2 equiv.) with N-acetyl-L-lysine methyl
ester hydrochloride 4 in the presence of Et3N (3 equiv.) in DMF as
the solvent (r.t., 48 h), followed by acetylation, afforded the
sulfamate derivative
5 in 16% isolated yield. This result
demonstrated the occurrence of an effective SuFEx coupling of
the OSO2F group of 3 with the -NH2 of lysine.
The mass spectrum of ubiquitin after the reaction with 3 displays
three main peaks (Figure S1): i) one peak corresponds to the
mass of the free protein, ii) a second peak to the mass of ubiquitin
functionalized with a single molecule of disaccharide derivative,
and iii) a third to the mass of ubiquitin with two molecules of
disaccharide derivative. The most intense peak among the three
is that corresponding to the free protein. This result suggests the
presence in solution of a heterogeneous mixture of functionalized
protein. In particular, from zero up to two molecules of
disaccharide can be covalently linked to each molecule of Ub,
involving the different solvent exposed NH2 groups. The 2D 1H-15N
HSQC NMR spectrum collected on the ubiquitin sample after the
derivatization reaction confirm the presence of the heterogeneous
Scheme 3. Synthesis of the glycosylated lysine 5.
Another evidence for the above chemoselectivity came from the
coupling of 3 (10 equiv.) with the octapeptide Octreotide[28] bis-
acetate 6 featuring a lysine residue in its body. Indeed, this
reaction carried out at room temperature in the presence of Et3N
afforded the product 7 (Scheme 4). The structure of 7 derived
from the SuFEx reaction of 3 with the -NH2 of lysine residue is
in agreement with that of the product formed in the aza-Michael
1
population of isomers (Figure 1). In particular, the H-15N cross-
peaks corresponding to the unreacted protein are still present and
intense; concomitantly new H-N cross-peaks (one or more for
each original cross-peak) at different values of chemical shift
appeared. These new signals can be attributed to the H-N
backbone atoms of the new species generated after the reaction
with 3. For most of the cross-peaks, the intensity of the new
signals is lower than that of the unreacted species, in agreement
with the results of the mass spectrometry analysis. Unfortunately,
the new cross-peaks corresponding to the sulfamate protons of
the reacted lysine sidechains could not be detected because of
the pH of the buffered solution (pH = 8) that ensures the chemical
stability of the conjugate, and the low pKa value of the sulfamate
moiety.
coupling
of
Octreotide
with
N-phenylvinylsulfonamide
derivatives.[29] Notably the introduction of a lactose residue in
Octreotide bears its own value as this modification may affect the
pharmacokinetic properties of the peptide, that is in fact employed
for the treatment of acromegaly, a pathology that is characterized
by hormonal disorder in adult age patients.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803912
Chemistry - A European Journal
FULL PAPER
Figure 2. 1D 1H NMR spectra of free ubiquitin (A), ubiquitin after one SuFEx
reaction with 3 (B), and ubiquitin after three successive SuFEx reactions with
the disaccharide derivative (C). The assignment of the methyl group (Qδ1) of
Leu-50 of free ubiquitin is reported in the spectrum. The spectra were recorded
at 298 K and 950 MHz.
Figure 1. Superimposition of the 2D 1H-15N HSQC spectra of free ubiquitin (red)
and ubiquitin after one SuFEx reaction with the disaccharide derivative (black).
In the latter spectrum signals corresponding to the unreacted ubiquitin are still
visible. The spectra were recorded at 298 K and 950 MHz.
1
The 2D H-15N HSQC spectrum acquired on the protein sample
after three successive SuFEx reactions with the disaccharide
derivative shows a similar pattern of cross-peaks of that after a
single SuFEx reaction, but with different relative signal intensity
(Figure S4). In particular, the new cross-peaks are increased in
intensity. This phenomenon is sizable for some lysine residues
(Lys-11, Lys-33 and Lys-48). For Lys-11 and Lys-33 the cross-
peak of the unreacted species is almost completely disappeared
after three successive SuFEx reactions, while the signal intensity
of the new cross-peaks corresponding to the reacted lysine
residue are sizable increased. Conversely, the cross-peak
corresponding to the reacted form of Lys-48 is barely
distinguishable in the spectrum after a single SuFEx reaction, but
it shows the same intensity of the peak of the unreacted species
after three SuFEx reactions (Figure S5). This is in agreement with
what observed also for the methyl group of Leu-50 in the
monodimensional spectra (see above).
A more detailed analysis on the conjugation sites of Ub has been
provided by the analysis of the chemical shift perturbation. As
expected, the protein residues experiencing the largest effects
are the lysines and residues in their close proximity (Figure S6-
S7). For a semi-quantitative evaluation of the conjugation degree
of each lysine residue, the relative intensity of the cross-peaks
corresponding to the reacted and unreacted species within the
same spectrum has been analyzed in detail and graphically
reported in Figure 3.11
Considering the presence of unreacted lysine residues still
available for functionalization on the protein surface, the SuFEx
reaction was repeated two more times on the same protein
sample. The mass spectrum of the protein, subjected to three
successive SuFEx reactions with 3, shows four peaks (Figure S2):
the same three as after a single SuFEx reaction, and an additional
one corresponding to ubiquitin with three molecules of
disaccharide derivative. Interestingly, the relative intensity of
these peaks changes after each SuFEx reaction, and finally, the
most intense peak in the mass spectrum corresponds to Ub with
one molecule of disaccharide derivative conjugated to the protein.
To shed light on the conjugation degree of each lysine residue,
an NMR analysis has been carried out on samples of Ub before
1
and after the SuFEx reactions. In the 1D H NMR spectrum of the
protein after three successive SuFEx reactions with 3, the signals
are broader than the signals in the spectrum of the free protein
(Figure S3) and of the protein after a single reaction. This can be
explained by an increase of both the hydrodynamic volume and
the heterogeneity of the conjugated protein population. A more
detailed analysis of the aliphatic region of the spectrum around 0
ppm indicates the presence of an additional signal at the
resonance frequency of -0.05 ppm (Figure 2). This new peak,
nearby the signal assigned to the methyl group (Qδ1) of Leu-50,
corresponds to the same methyl group experiencing a different
chemical environment because of the conjugation of the
neighboring Lys-48 with disaccharide chains. The presence in the
spectra of two signals with the same intensity and corresponding
to the free and reacted forms of the protein indicates for Lys-48 a
ratio of about 1:1 between the reacted and non-reacted species.
Of note, this peak after only a single SuFEx reaction is barely
visible.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803912
Chemistry - A European Journal
FULL PAPER
glycosides (Figure S8). Interestingly, the signal of the free ANSII
is completely missing in MALDI spectrum and from the broad
signal, the peaks of the protein glycosylated with one, two, three
and four disaccharide units per monomer of ANSII can be
distinguished.
The introduction of lactose residues is expected to improve the
stability of ANSII against proteolytic enzymes. To assess the
improved stability of the glycoconjugated protein, samples of
ANSII functionalized with 3 have been incubated at 37 °C with
trypsin or with the catalytic domain of human matrix
metalloproteinase 1 (MMP-1) and monitored up to 3 days (see
Supporting Information). The effect of the proteolytic enzymes
has been monitored by SDS-PAGE gel. From the visual
inspection of the gels, it was immediately evident that the
conjugation with 3 prevents the proteolytic degradation of ASNII
by trypsin that conversely takes place when the free protein is
incubated with trypsin under the same experimental conditions
(Figure S10 and S11). Moreover, the conjugation with Lact-Ar-
OSO2 groups is also capable to slow down significantly the
degradation of ASNII treated with MMP-1, a very efficient
proteolytic enzyme involved in the proteolytic degradation of
extracellular proteins and glycoproteins (Figure S12).
Figure 3. (A) Graphical representation of the intensity ratio per residue of the
reacted species with lactose fluorosulfate 3 and the unreacted one present in
the same spectrum after three successive SuFEx reactions with 3. The lysine
residues are in magenta and the residues experiencing the largest variation are
in blue. (B) Enlargement of the 2D 1H-15N HSQC spectrum of ubiquitin after
three successive SuFEx reactions with 3. On the spectrum is indicated an
example of the cross-peaks corresponding to the functionalized and non-
functionalized Lys-48 residue used to evaluate the signal intensity ratio. (C)
Surface representation of ubiquitin (pdb code: 1UBQ) with the lysine residues
and the N-terminus color coded according their conjugation degree.
Conclusions
A similar integrated strategy has been used to analyze the
pharmacologically relevant ANSII conjugated with the same thio-
lactose derivative. The comparison of the 1D 1H spectra collected
before and after each SuFEx reaction of 3 with the protein, shows
a group of signals between 3 and 4.5 ppm, which increases
progressively in intensity after each reaction (Figure 4).
In summary, we have developed an efficient protocol for the
introduction of lactose groups on pharmaceutical relevant
biomolecules. This was made possible by using a typical SuFEx
process involving the coupling of a lactose bearing fluorosulfate
with the free -amino groups of lysine moieties incorporated into
the proteins. In this way, multiple glycosylation of Ub and ANSII
were carried out being the lactose residues linked to the lysine
fragments through a sulfamoyl tether. The glycosylation of ANSII
is expected to slow down the clearance of the enzyme in vivo,
thus increasing its half-life with a beneficial reduction of the drug
administration frequency. Due to a remarkable versatility, the
SuFEx might be suitable for the decoration of challenging proteins
with a panel of different saccharides or biologically relevant
residues.
Experimental Section
Reactions were monitored by TLC on silica gel 60 F254 with detection by
charring with sulfuric acid. Flash column chromatography was performed
on silica gel 60 (40-63 m). Optical rotations were measured at 20 ± 2 °C
in the stated solvent; []D values are given in deg mL g-1 dm-1. 1H (400 MHz),
13C (100 MHz), and 19F (376 MHz) NMR spectra were recorded in the
stated solvent at room temperature unless otherwise specified. MALDI-
TOF (Bruker Ultraflex III) and high resolution mass spectrometry (Waters
Micromass Q-TOF) analyses of compounds 3 and 5 were performed at the
Laboratoire de Mesures Physiques, IBMM – University of Montpellier. The
photoinduced thiol-ene reaction was carried out in a glass vial located 2.5
cm away from the household UV-A lamp apparatus equipped with four 15
W tubes (1.5 x 27 cm each). The commercially available photoinitiator 2,2-
dimethoxy-2-phenylacetophenone (DPAP) was used without further
purification.
Figure 4. 1D 1H NMR spectra of free ANSII (A), ANSII after one (B), two (C)
and three (D) successive SuFEx reactions with 3. The red square highlight the
region between 3 and 4.5 ppm, where particularly emerge some of the protons
of the lactose moiety. The spectra were recorded at 310 K and 900 MHz.
Unfortunately, the very large molecular weight of ANSII makes it
unfeasible the analysis of the products of the reactions by 2D 1H-
15N HSQC. However, as previously observed for Ub, the analysis
carried out by mass spectrometry shows that the SuFEx reaction
on ANSII results in a heterogeneous mixture of positional
This article is protected by copyright. All rights reserved.

10.1002/chem.201803912
Chemistry - A European Journal
FULL PAPER
Lactose fluorosulfate 3. A solution of thiol 1 (1.433 g, 4.00 mmol),
fluorosulfate 2 (1.207 g, 5.20 mmol), and DPAP (102 mg, 0.40 mmol) in
DMF (5 mL), partially concentrated under vacuum (ca. 0.1 mbar) to remove
the traces of Me2NH, and H2O (1 mL) was irradiated (max 365 nm) under
vigorous stirring at room temperature for 1 h and then concentrated. The
residue was triturated with CH2Cl2 (2 x 10 mL) and the precipitate,
recovered by decantation, afforded, after drying under vacuum, 3 (1.937 g,
82%) as an amorphous solid; []D = -15.3 (c 1.1, H2O). 1H NMR (D2O): δ
7.38 and 7.07 (2 d, 4H, J = 9.0 Hz, Ar), 4.47 (d, 1H, J1,2 = 10.1 Hz, H-1),
4.37 (d, 1H, J1’,2’ = 7.6 Hz, H-1’), 4.36-4.14 (m, 2H, CH2CH2CH2O), 3.88-
3.42 (m, 10H), 3.34-3.29 (m, 1H), 2.96-2.79 (m, 2H, CH2S), 2.11-2.04 (m,
2H, CH2CH2CH2). 13C NMR (D2O): δ 158.0 (C), 143.4 (C), 121.8 (CH),
115.6 (CH), 103.0 (CH), 85.5 (CH), 78.5 (CH), 78.2 (CH), 75.9 (CH), 75.3
(CH), 72.6 (CH), 72.2 (CH), 70.9 (CH), 68.6 (CH), 66.8 (CH2), 61.0 (CH2),
60.3 (CH2), 29.3 (CH2), 26.7 (CH2). 19F NMR (D2O): δ 36.31. HRMS (ESI/Q-
TOF) m/z calcd for C21H32FO14S2 (M+H)+ 591.1218, found 591.1220.
upon magnetic stirring. The suspension was sonicated alternating 30 sec
of sonication and 3 min of resting for 10 cycles, then centrifuged at 40000
rpm for 30 min at 277 K. The supernatant was treated with small aliquots
of an HClO4 solution to adjust the pH down to 4.5. The solution was again
centrifuged at 40000 rpm for 30 min at 277 K to remove the precipitate.
The protein in solution was then purified by cationic-exchange
chromatography using a HiPrep SP FF 16/10 column (GE Healthcare Life
Science) preliminarly equilibrated with 20 mM AcONa, pH 4.5 buffer. The
protein was eluted in the same buffer with a linear 0-500 mM NaCl gradient.
Fractions containing pure ubiquitin were identified by Coomassie staining
SDS-PAGE gels. A further purification was performed by size-exclusion
chromatography using a HiLoad 16/60 Superdex 75 pep grade column
(GE Healthcare Life Science). The elution was achieved in 50 mM
phosphate, pH 8.0 buffer.
Expression and purification of ANSII Escherichia coli C41(DE3) cells
were transformed with pET-21a(+) plasmid encoding ANSII gene. The
cells were cultured in LB medium supplemented with 0.1 mg mL−1 of
ampicillin, grown at 310 K, until A600 reached 0.6, then induced with 1 mM
isopropyl β-D-1-thiogalactopyranoside. They were further grown at 310 K
overnight and then harvested by centrifugation at 7500 rpm for 15 min at
277 K. The pellet was suspended in 10 mM Tris-HCl, pH 8.0, 15 mM EDTA,
20% sucrose buffer (60 mL per liter of culture) and incubated at 277 K for
20 min upon magnetic stirring. The suspension was centrifuged at 10000
rpm for 30 min and the supernatant discarded. The recovered pellet was
re-suspended in H2O milli-Q (60 mL per liter of culture) and newly
incubated at 277 K for 20 min under magnetic stirring. Again the
suspension was centrifuged at 10000 rpm for 30 min. The pellet was
discarded, and the supernatant treated with ammonium sulfate. Still under
magnetic stirring solid ammonium sulfate was added in aliquots up to 50%
saturation. The precipitate was removed by centrifugation, then further
ammonium sulfate was added up to 90% saturation to trigger the
precipitation of ANSII, which was recovered again by centrifugation. The
precipitated ANSII was re-dissolved in a minimal amount of 20 mM Tris-
HCl, pH 8.6 buffer and dialyzed extensively against the same buffer. ANSII
was purified by anionic-exchange chromatography using a HiPrep Q FF
16/10 column (GE Healthcare Life Science). The protein was eluted in 20
mM Tris-HCl, pH 8.6 buffer with a linear 0-1 M NaCl gradient. Fractions
containing pure ANSII were identified by Coomassie staining SDS-PAGE
gels, then pooled and further purified by size-exclusion chromatography
using a HiLoad 26/60 Superdex 75 pep grade column (GE Healthcare Life
Science). The elution was performed in 50 mM phosphate, pH 8.0 buffer.
Lactosyl-lysine 5. To a solution of N-acetyl-L-lysine methyl ester
hydrochloride 4 (24 mg, 0.10 mmol) and lactose fluorosulfate 3 (118 mg,
0.20 mmol) in DMF (250 L), partially concentrated under vacuum (ca. 0.1
mbar) to remove the traces of Me2NH, was added Et3N (42 L, 0.30 mmol).
The solution was kept at room temperature for 48 h and then concentrated.
A solution of the residue in pyridine (1 mL) and acetic anhydride (1 mL)
was kept at room temperature for 14 h and then concentrated. The residue
was eluted from a column of silica gel with 1:1 cyclohexane-AcOEt, then
AcOEt, to give 5 (17.5 mg, 16%) as a colorless syrup; []D = +2.7 (c 0.6,
CHCl3). 1H NMR (CDCl3): δ 7.16-7.12 and 6.94-6.90 (2 m, 4H, Ar), 6.05 (d,
1H, J = 7.8 Hz, NH), 5.38 (dd, 1H, J3’,4’ = 3.4 Hz, J4’,5’ = 0.9 Hz, H-4’), 5.24
(dd, 1H, J2,3 = J3,4 = 9.2 Hz, H-3), 5.13 (dd, 1H, J1’,2’ = 7.8 Hz, J2’,3’ = 10.5
Hz, H-2’), 4.98 (dd, 1H, H-3’), 4.97 (dd, 1H, J1,2 = 10.0 Hz, H-2), 4.59 (ddd,
1H, J = 5.2, 7.5, 7.8 Hz, CHNH), 4.52 (d, 1H, H-1), 4.51 (d, 1H, H-1’), 4.19-
4.07 (m, 4H, 2 H-6, 2 H-6’), 4.06 (t, 2H, J = 6.0 Hz, CH2CH2CH2O), 3.90
(ddd, 1H, J5’,6’a = 6.2 Hz, J5’,6’b = 7.6 Hz, H-5’), 3.80 (dd, 1H, J4,5 = 9.8 Hz,
H-4), 3.77 (s, 3H, OCH3), 3.66-3.62 (m, 3H), 2.94-2.79 (m, 2H, CH2S), 2.49,
2.18, 2.13, 2.09, 2.076, 2.071, 2.070, 2.05, and 1.99 (9 s, 27H, 9 Ac), 2.12-
2.06 (m, 2H, CH2CH2CH2), 1.87-1.53 and 1.36-1.23 (2 m, 6H). 13C NMR
(CDCl3): δ 172.9 (C), 170.4 (C), 170.3 (C), 170.2 (C), 170.1 (C), 169.9 (C),
169.8 (C), 169.7 (C), 169.1 (C), 158.1 (C), 142.6 (C), 122.9 (CH), 115.6
(CH), 101.1 (CH), 83.8 (CH), 77.2 (CH), 76.8 (CH), 76.2 (CH), 73.7 (CH),
71.0 (CH), 70.7 (CH), 70.2 (CH), 69.1 (CH), 66.6 (CH), 66.5 (CH2), 62.1
(CH2), 60.8 (CH2), 52.5 (CH3), 52.0 (CH), 47.8 (CH2), 31.9 (CH2), 29.4
(CH2), 28.2 (CH2), 27.0 (CH2), 24.8 (CH3), 23.2 (CH3), 22.2 (CH2), 20.86
(CH3), 20.82 (CH3), 20.76 (CH3), 20.67 (CH3), 20.65 (CH3), 20.53 (CH3).
HRMS (ESI/Q-TOF) m/z calcd for C46H65N2O25S2 (M+H)+ 1109.3318,
found 1109.3315.
Functionalization of Ub and ANSII with lactose fluorosulfate 3. The
SuFEx reaction of functionalization with 3 was achieved by mixing 800 μL
of a 500 mM phosphate, pH 8.0, solution containing 6.2 mg mL−1 of the
protein (725 μM Ub or 180 μM ANSII) with 200 μL of a 0.50 M solution of
3 in water. Each reaction was performed at 298 K for 48 hours, then the
mixture was washed with 50 mM phosphate buffer, pH 8.0.
Lactosyl-Octreotide 7. To a solution of Octreotide bis-acetate 6 (8.0 mg,
7.0 mol) and lactose fluorosulfate 3 (41 mg, 70.0 mol) in H2O (100 L)
and CH3CN (50 L) was added Et3N (4 L, 28.0 mol). The solution was
kept at room temperature for 72 h and then concentrated. The residue was
purified by HPLC (Zorbax SB-C18 column, eluent H2O2-CH3CN, gradient
from 80:20 to 20:80). The pure derivative 7 (5 mg) was isolated as glassy
solid in 45 % yield.
NMR measurements Solution NMR spectra of Ub and ANSII were
recorded on Bruker AVANCEIII-HD NMR spectrometers, operating at 900
and 950ꢀMHz 1H Larmor frequency, equipped with triple resonance cryo-
probes, at 298 K and 310 K, respectively. Protein samples were in water
buffered solution [50 mM phosphate buffer, pH 8] with a protein
concentration of ~ 0.5 and 0.25 mM for Ub and ANSII, respectively.
However, after consecutive reactions with lactose fluorosulfate 3, the
Expression and purification of 15N-labeled Ubiquitin. Escherichia coli
BL21(DE3) Gold cells were transformed with pET-21a(+) plasmid
encoding ubiquitin gene. The cells were cultured in a 15N-labeled minimal
medium supplemented with 0.1 mg mL−1 of ampicillin, grown at 310 K,
until A600 reached 0.6, then induced with 1 mM isopropyl β-D-1-
thiogalactopyranoside. Cells were further grown at 310 K overnight and
then harvested by centrifugation at 7500 rpm for 15 min at 277 K. The
pellet was suspended in 20 mM Tris-HCl, pH 7.0 buffer (20 mL per liter of
culture) supplemented with 12.5 μg mL−1 DNAse, 500 μg mL−1 lysozyme,
20 mM MgSO4 and protease inhibitors, and incubated at 277 K for 15 min
1
1
protein concentration was slightly reduced. 1D H and 2D H-15N HSQC
NMR spectra were recorded on Ub samples after one and three SuFEx
reactions; whereas only 1D 1H NMR spectra were performed on ANSII
samples after a single, two and three SuFEx reactions. All the spectra were
processed with the Bruker TopSpin software package, and analyzed with
the program CARA.[30] The assignment of the spectra of Ub was based on
This article is protected by copyright. All rights reserved.

10.1002/chem.201803912
Chemistry - A European Journal
FULL PAPER
[15] V. Gembus, F. Marsais, V. Levacher, Synlett 2008, 2008,
1463–1466.
the data reported in the Biological Magnetic Resonance Data Bank under
the accession code 6457.[31]
[16] J. Dong, L. Krasnova, M. G. Finn, K. B. Sharpless, Angew.
Chem. Int. Ed. Engl. 2014, 53, 9430–9448.
[17] For a collection of references dealing with the application of
SuFEx (a) or related group (b) on chemical biology, see: a) T. Guo,
G. Meng, X. Zhan, Q. Yang, T. Ma, L. Xu, K. B. Sharpless, J. Dong,
Angew. Chem. Int. Ed. Engl. 2018, 57, 2605–2610; b) C. Baggio,
L. Gambini, P. Udompholkul, A. F. Salem, A. Aronson, A. Dona,
E. Troadec, F. Pichiorri, M. Pellecchia, J. Med. Chem. 2018, DOI
10.1021/acs.jmedchem.8b00810.
[18] a) A. Dondoni, A. Massi, P. Nanni, A. Roda, Chemistry 2009,
15, 11444–11449; b) M. Lo Conte, S. Staderini, A. Marra, M.
Sanchez-Navarro, B. G. Davis, A. Dondoni, Chem. Commun.
(Camb.) 2011, 47, 11086–11088; c) A. Dondoni, A. Marra, Chem
Soc Rev 2012, 41, 573–586.
Acknowledgements
The authors acknowledge the support and the use of resources
of Instruct-ERIC, a landmark ESFRI project, and specifically the
CERM/CIRMMP Italy center, as well as the EC Contracts iNext
No. 653706. The authors acknowledge the support of the
University of Florence CERM-TT and the Recombinant Proteins
JOYNLAB. The FP7 WeNMR (project No. 261572) and H2020
West-Life (project No. 675858) are also acknowledged.
[19] M. L. Graham, Adv. Drug Deliv. Rev. 2003, 55, 1293–1302.
[20] a) E. Ravera, S. Ciambellotti, L. Cerofolini, T. Martelli, T.
Kozyreva, C. Bernacchioni, S. Giuntini, M. Fragai, P. Turano, C.
Luchinat, Angew. Chem. Int. Ed. Engl. 2016, 55, 2446–2449; b)
S. Giuntini, L. Cerofolini, E. Ravera, M. Fragai, C. Luchinat, Sci
Rep 2017, 7, 17934.
[21] J. W. Marsh, J. Denis, J. C. Wriston, J. Biol. Chem. 1977,
252, 7678–7684.
[22] S. Li, D. Cohen-Karni, L. T. Beringer, C. Wu, E. Kallick, H.
Edington, M. J. Passineau, S. Averick, Polymer 2016, 99, 7–12.
Keywords: SuFEx reaction • glycosylation • Ubiquitin •
Asparaginase • flurosulfate thiolactoside
[1] a) J. Marth, B. Lowe, A. Varki in Essentials of Glycobiology,
(Eds.: A. Varki, R. D. Cummings, J. D. Esko, H. H. Freeze, G. W.
Hart, J. Marth), Cold Spring Harbor Laboratory Press, New York,
1999, Chapters 8 and 17; b) P. Sears, C. H. Wong, Cell. Mol. Life
Sci. 1998, 54, 223–252; c) H.-J. Gabius, S. André, H. Kaltner, H.-
C. Siebert, Biochim. Biophys. Acta 2002, 1572, 165–177.
[2] a) R. Apweiler, H. Hermjakob, N. Sharon, Biochim. Biophys.
Acta 1999, 1473, 4–8; b) C.-H Wong J. Org. Chem. 2005, 70,
4219-4225.
[23] a) Chen, W.; Dong, J.; Plate, L.; Mortenson, D. E.; Brighty,
G. J.; Li, S.; Liu, Y.; Galmozzi, A.; Lee, P. S.; Hulce, J. J.; Cravatt,
B. F.; Saez, E.; Powers, E. T.; Wilson, I. A.; Sharpless, K. B.; Kelly,
J. W. J. Am. Chem. Soc. 2016, 138, 7353-7364. b) Mortenson, D.
E.; Brighty, G. J.; Plate, L.; Bare, G.; Chen, W.; Li, S.; Wang, H.;
Cravatt, B. F.; Forli, S.; Powers, E. T.; Sharpless, K. B.; Ian A.
Wilson, I. A.; Kelly, J. W. J. Am. Chem. Soc. 2018, 140, 200-210.
[24] A. Dondoni, Angew. Chem. Int. Ed. Engl. 2008, 47, 8995–
8997.
[25] Compound 2 was prepared by reaction of 4-(allyloxy)phenol
with sulfuryl fluoride (SO2F2) gas (see Supplementary
Information) according to the method described in ref. 16. For the
preparation of fluorosulfates, see also:E. Zhang, J. Tang, S. Li, P.
Wu, J. E. Moses, K. B. Sharpless, Chem. Eur. J. 2016, 22, 5692–
5697.
[26] N. Floyd, B. Vijayakrishnan, J. R. Koeppe, B. G. Davis,
Angew. Chem. Int. Ed. Engl. 2009, 48, 7798–7802.
[27] A solution of 3 in D2O containing NaHCO3 remained
unaltered after 96 hours at room temperature.
[28] Octreotide 6 is produced by Bioindustria, Laboratorio
Italiano Medicinali, LIM (e-mail: DT@bioindustria.it) and
commercialized as a drug against acromegaly.
[3] N. Sethuraman, T. A. Stadheim, Curr. Opin. Biotechnol.
2006, 17, 341–346.
[4] O. Seitz, Chembiochem 2000, 1, 214–246.
[5] a) A. Varki, Glycobiology 1993, 3, 97–130; b) R. A. Dwek,
Chem. Rev. 1996, 96, 683-720; c) C. R. Bertozzi, L. L. Kiessling,
Science 2001, 291, 2357–2364; d) P. M. Rudd, T. Elliott, P.
Cresswell, I. A. Wilson, R. A. Dwek, Science 2001, 291, 2370–
2376.
[6] S. Kornfeld, J. Clin. Invest. 1998, 101, 1293–1295.
[7] a) A. M. Sinclair, S. Elliott, J Pharm Sci 2005, 94, 1626–
1635; b) D. T.-Y Liu, Trends Biotechnol., 1992, 10, 114-119.
[8] P. Costantino, R. Rappuoli, F. Berti, Expert Opin Drug
Discov 2011, 6, 1045–1066.
[9] R. J. Solá, K. Griebenow, J Pharm Sci 2009, 98, 1223–1245.
[10] a) B. G. Davis, Chem. Rev. 2002, 102, 579–602; b) S. I. van
Kasteren, H. B. Kramer, H. H. Jensen, S. J. Campbell, J.
Kirkpatrick, N. J. Oldham, D. C. Anthony, B. G. Davis, Nature
2007, 446, 1105–1109; c) D. P. Gamblin, E. M. Scanlan, B. G.
Davis, Chem. Rev. 2009, 109, 131–163.
[11] S. Giuntini, E. Balducci, L. Cerofolini, E. Ravera, M. Fragai,
F. Berti, C. Luchinat, Angew. Chem. Int. Ed. Engl. 2017, 56,
14997–15001.
[29] H. Chen, R. Huang, Z. Li, W. Zhu, J. Chen, Y. Zhan, B. Jiang,
Org. Biomol. Chem. 2017, 15, 7339–7345.
[30] R. Keller, The Computer Aided Resonance Assignment
Tutorial (CARA), CANTINA Verlag, 2004.
[31] G. Cornilescu, J. L. Marquardt, M. Ottiger, A. Bax, J. Am.
Chem. Soc. 1998, 120, 6836–6837.
[12] a) H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem.
Int. Ed. Engl. 2001, 40, 2004–2021; b) H. C. Kolb, K. B. Sharpless,
Drug Discov. Today 2003, 8, 1128–1137.
[13] a) V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B.
Sharpless, Angew. Chem. Int. Ed. Engl. 2002, 41, 2596–2599;
b) C. W. Tornøe, C. Christensen, M. Meldal, J. Org. Chem.
2002, 67, 3057–3064.
[14] R. K. Iha, K. L. Wooley, A. M. Nyström, D. J. Burke, M. J.
Kade, C. J. Hawker, Chem. Rev. 2009, 109, 5620–5686.
This article is protected by copyright. All rights reserved.

10.1002/chem.201803912
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
Layout 1:
FULL PAPER
The metal free click SuFEx reaction
is here employed for the coupling of
a lactosyl fluorosulfate with the free
-amino groups of lysine moieties
incorporated into Ubiquitin (Ub) and
into the validated drug
A. Marra, [b] J. Dong, [c] T. Ma, [c] S.
Giuntini, [a],[d] E. Crescenzo, [a],[d] L.
Cerofolini, [a],[d] M. Martinucci, [a] C.
Luchinat, [a],[d] M. Fragai, *[a],[d] C.
Nativi*[a] and A. Dondoni*[e]
Asparaginase II (ANSII). The
glycosylation of ANSII slowing down
the clearance of the enzyme in vivo,
increases its half-life with a
beneficial reduction of its
administration frequency.
Page No. – Page No.
Protein Glycosylation via Sulfur
Fluoride Exchange (SuFEx)
Chemistry: The Key Role of a
Fluorosulfate Thiolactoside
This article is protected by copyright. All rights reserved.