A trisulfide-linked glycoprotein

Article abstract of DOI:10.1039/b706682a

The first member of a novel class of chemoselective reagents, glycosyl methanedithiosulfonates, has been synthesized, identified and employed in the first examples of chemical, site-selective construction of a trisulfide-modified protein with complete conversion. The Royal Society of Chemistry.

Full text of DOI:10.1039/b706682a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
A trisulfide-linked glycoprotein{
Gonc¸alo J. L. Bernardes,a Justin P. Marston,b Andrei S. Batsanov,b Judith A. K. Howardb and
Benjamin G. Davis*a
Received (in Cambridge, UK) 3rd May 2007, Accepted 11th June 2007
First published as an Advance Article on the web 4th July 2007
DOI: 10.1039/b706682a
amounts of naturally-occurring protein containing, for example,
Cys–CH2SSSCH2–Cys trisulfides is either not possible or difficult
and low yielding (,1%).4–8 To our knowledge, until now, no
examples of either intermolecular protein trisulfides or trisulfide
glycoproteins exist. Here we present the discovery of a novel thiol-
selective modification agent for the direct and quantitative
formation of protein trisulfides.
The first member of a novel class of chemoselective reagents,
glycosyl methanedithiosulfonates, has been synthesized, identi-
fied and employed in the first examples of chemical, site-
selective construction of a trisulfide-modified protein with
complete conversion.
Trisulfides are rare but important naturally occurring linkages in
biomolecules with functions ranging from subtle modulation of
conformation1–3 and stability in proteins,4–8 such as human
superoxide dismutase4,5 and human growth hormone6,7 and
peptides,9–11 through to enediyne antitumor antibiotics.12,13 The
trisulfide motif can convey intriguing biological properties such
as fine-tuning of receptor interaction14 in hormones such as
vasopressin14 and oxytocin10 through to the trisulfide bridge
‘trigger’ that relieves strain and ultimately initiates15 radical-
mediated site-specific double-stranded DNA cleavage16 in cali-
cheamicins and esperamicins.12,13
Methanethiosulfonates (MTS, R–S–SO2CH3)33 including glyco-
syl methanethiosulfonates (glyco-MTS)25 such as Glc-MTS 3 are
powerful reagents for protein modification through disulfide
formation. However, the use, by logical extension, of methane-
dithiosulfonates (MDTS, R–S–S–SO2CH3) has surprisingly not
been investigated, perhaps due to their apparent instability.34
Glyco-MTS reagents are readily synthesized23,25,35 by displacement
of halide from the appropriate peracetohalohexose derivative with
methanethiosulfonate, typically sodium methanethiosulfonate.{36
The efficiency and stereoselectivity of this displacement can be
quite dramatically affected by conditions.35 During the course of
an investigation of the effect of solvent, temperature and additives
upon the reaction of glucosyl bromide 237 with NaMTS to form
Glc-MTS 3 we observed and isolated, initially, very small amounts
(,2%) of an interesting by-product, tentatively assigned as the
glucosyl methanedithiosulfonate (Glc-MDTS) 4, of very similar
polarity to the expected product Glc-MTS 3. Variation of
conditions allowed some improvement of the yield to a workable
level (13%) capable of producing sufficient Glc-MDTS (Scheme 1)
using increased equivalents of NaMTS (5 equiv.) and dioxane at
70 uC for 70 h. Prolonged treatment or the use of other solvents
was found to give substantially lower yields.
Glycosylation is the most diverse of all post-translational
modifications in Nature.17–19 It is estimated that over 50% of
all mammalian proteins carry carbohydrate appendages,20 which
affect biopharmacological properties, stability and signalling
functions of the proteins.21 As glycosylation is not under direct
genetic control, glycoproteins tend to exist as complex mixtures
of glycoforms;22 proteins with the same protein backbone, but
varying in the nature and sites of glycosylation. This complexity
has plagued the structure–activity studies for specific glycoforms as
current separation procedures do not allow for the attainment of
sufficient quantities of glycoproteins for detailed studies. Chemical
glycosylation of proteins has therefore been an attractive strategy
for accessing single glycoform targets.18 Various approaches exist
for the precise incorporation of glycans at selected positions in
protein backbones.18 One such method relies on the attachment of
carbohydrate through a disulfide linkage onto the thiol of the
cysteine (Cys) sidechain;23–29 cysteines are particularly attractive
target ‘tags’ due to their low natural abundance, especially in their
free-thiol form.30
Despite the availability of some methods for the formation of
small molecules containing polysulfides2,14,31,32 these are neither
selective for trisulfides, complete nor compatible with or applicable
to proteins as substrates. Moreover, the isolation of the small
aDepartment of Chemistry, University of Oxford, Chemistry Research
Laboratory, 12 Mansfield Road, Oxford, UK OX1 3TA.
E-mail: Ben.Davis@chem.ox.ac.uk; Fax: +44 1865 275674;
Tel: +44 1865 275652
bDepartment of Chemistry, University of Durham, South Road, Durham,
UK DH1 3LE
{ Electronic supplementary information (ESI) available: Syntheses,
characterizations, modifications X-ray crystallographic data for 3 and 4.
See DOI: 10.1039/b706682a
Scheme 1 Synthesis and X-ray crystal structures of glyco-MTS 3 and
glyco-MDTS 4.
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 3145–3147 | 3145

View Article Online
We are still investigating the mechanism behind the unusual
formation of 4. Although conditions that increased the yield of 4
led to lower concomitant yield of 3, we do not observe direct
conversion of 3 to 4 when re-exposed to the reaction conditions.
Not only was this formation of this rare compound type
surprising, the isolation of 4 is notable given the facile route for
ejection of molecular sulfur from polythiosulfonates.34 We
speculate that isolation and subsequent successful use of 4 as a
reagent may due to enhanced stability that is a consequence of the
presence of a more hindered, secondary (glycosyl) thiyl moiety.38
Regardless, the identities of closely related compounds 3 and 4,
was unambiguously confirmed by full characterization and X-ray
crystallography§ (Scheme 1); the latter is the first known 3D
structure of an MDTS. Interestingly, the MDTS moiety appears to
be a powerful chirochromophore with significant ORD at 589 nm
in a manner consistent with Hudson’s isorotation rules:39 the
specific rotation of Na D-line light ([a]2D5 = 2235 (CHCl3)) by b-D-
4 is in the same direction but an order of magnitude greater than
that of b-D-3 despite only differing by one sulfur atom. In the
infrared spectrum of 4 two additional strong absorbances are also
observed (1258, 1226 cm21) due to S–S stretching.
Fig. 2 UV spectrum of SBL–SSS–4. Both protein samples were recorded
at 1 mg mL21
.
simply to the presence of trisulfide.4 We are currently exploring the
spectral profiles of intra- and inter-trisulfide proteins for the direct
monitoring of their presence, stability and kinetics (interaction and
reaction).
Next, the potential of 4 as a novel class of trisulfide-forming,
protein modification reagent was investigated. In aqueous buffer
at pH 9.5, 4 cleanly and completely (.95%) converted cysteine
mutant protein SBL-S156C (Scheme 2 and Fig. 1) into trisulfide
glycoprotein S156C–SSS–4, an intriguing and novel class of
glycoconjugate. This first, complete and site-selective chemical
construction of a trisulfide-containing protein has also allowed us
to clarify aspects of the long-standing speculation surrounding
proteins that contain ‘excess sulfur atoms’.4 Such proteins4,5,40 can
display unusual spectral properties due to the presence of an
intramolecular Cys–CH2SSSCH2–Cys trisulfide. S156C–SSS–4
displayed a broad increase in absorption over unmodified protein
in the region y250–270 nm (Fig. 2) consistent with the additional
presence of an unstrained SS bond that gives rise to n A s*
transition but no absorption at 325 nm. These observations
confirm that transitions at 325 nm observed in other trisulfide–
containing proteins4,5,40 are due primarily to the dihedral strain
induced in an intramolecular trisulfide strained S–S5 and not due
In summary, this communication describes the synthesis of a
novel and potent chemoselective reagent, Glc-MDTS 4 and its
successful use in the first site-selective construction of a trisulfide-
linked (glyco)protein. This method not only suggests strategies for
accessing this class of rare natural products but allows the creation
of novel (glyco)protein probes with unique reactivities and proteins
with manipulable spectral signatures, the full properties of which
are currently being determined.
We gratefully acknowledge the Fundac¸a˜o para a Cieˆncia e a
Tecnologia, Portugal for the award of a doctoral fellowship
(G.J.L.B., grant BD/16997/2004), and Sander I. van Kasteren and
Dr Elizabeth J. Grayson for technical support.
Notes and references
{ Sodium methanethiosulfonate 1 for reaction with glycosyl halide was
prepared through direct reaction of sodium sulfide with methanesulfonyl
chloride in refluxing water.36
§ X-Ray data collection: Siemens SMART 1 K CCD area detector, Mo-
˚
Ka radiation (l = 0.71073 A), narrow-frame v scans covering full sphere of
the reciprocal space up to 2h ¡ 58u; structure solution (direct methods) and
least-squares refinement against F2 using SHELXTL 6.14 programs
(Bruker AXS, Madison, Wisconsin, USA, 2003). The absolute structures
were determined from anomalous dispersion of sulfur by Flack parameter
(x).41 Crystal data: 3, C15H22O11S2, M = 442.45, T = 108 K, monoclinic,
˚
space group P21 (no. 4), a = 8.790(1), b = 12.774(2), c = 9.607(1) A, b =
113.60(2)u, U = 988.4(2) A , Z = 2, m = 0.33 mm21, 12 148 reflections (2731
3
˚
unique + 2448 Friedels, Rint = 0.028), R(F) = 0.026 on 5038 data with I ¢
2s(I), wR(F2) = 0.069 on all data, x = 0.06(4). CCDC 645866.
4, C15H22O11S3, M = 474.51, T = 120 K, monoclinic, space group P21
˚
(no. 4), a = 5.6609(6), b = 20.850(2), c = 9.2785(10) A, b =105.45(3)u, U =
Scheme 2 Conditions: 70 mM CHES, 5 mM MES, 2 mM CaCl2, pH
9.5, 50 equiv. of 4," protein concentration 2 mg mL21
1055.6(2) A , Z = 2, m = 0.41 mm21, 11 166 reflections (2886 unique + 2405
3
˚
.
Friedels, Rint = 0.019), R(F) = 0.028 on 5119 data with I ¢ 2s(I), wR(F2) =
0.070 on all data, x = 20.07(4). CCDC 645867. For crystallographic data
in CIF or other electronic format see DOI: 10.1039/b706682a
" This level of excess compares well with typical excesses used in protein
modification of 500–1000 equiv. 50 equiv. was chosen here to ensure
complete conversion and we are currently exploring the use of even less
reagent.
1 C. Andersson, P. O. Edlund, P. Gellerfors, Y. Hansson, E. Holmberg,
C. Hult, S. Johansson, J. Kordel, R. Lundin, I. Mendel-Hartvig,
B. Noren, T. Wehler, G. Widmalm and J. Ohnman, Int. J. Pept. Protein
Res., 1996, 47, 311–321.
Fig. 1 ESI-MS spectrum of SBL–SSS–4. (Observed mass 27111,
calculated mass 27110).
3146 | Chem. Commun., 2007, 3145–3147
This journal is ß The Royal Society of Chemistry 2007

View Article Online
2 D. A. Erlanson and J. A. Wells, Tetrahedron Lett., 1998, 39, 6799–6802.
3 E. Strandberg, J. Koerdel, R. Lundin, T. Wehler and G. Widmalm,
J. Pept. Res., 1997, 49, 254–260.
4 J. M. McCord and I. Fridovich, J. Biol. Chem., 1969, 244, 6049–6055.
5 R. G. Briggs and J. A. Fee, Biochim. Biophys. Acta, 1978, 537, 100–109.
6 A. M. Jespersen, T. Christensen, N. K. Lausen, P. F. Nielsen and
H. H. Sorensen, Eur. J. Biochem., 1994, 219, 365–373.
7 M. K. Thomsen, B. S. Hansen, P. Nilsson, J. Nowak, P. B. Johansen,
P. D. Thomsen and J. Christiansen, Pharmacol. Toxicol., 1994, 74,
351–358.
8 E. Canova-Davis, I. P. Baldonado, R. C. Chloupek, V. T. Ling,
R. Gehant, K. Olson and B. L. Gillece-Castro, Anal. Chem., 1996, 68,
4044–4051.
9 M. Moutiez, G. Lippens, C. Sergheraert and A. Tartar, Bioorg. Med.
Chem. Lett., 1997, 7, 719–724.
10 L. Chen, I. Zoulikova, J. Slaninova and G. Barany, J. Med. Chem.,
1997, 40, 864–876.
11 V. Windisch, F. DeLuccia, L. Duhau, F. Herman, S.-Y. Tang and
M. Vuilhorgne, J. Pharm. Sci., 1997, 86, 359–364.
12 K. C. Nicolaou and W.-M. Dai, Angew. Chem., Int. Ed. Engl., 1991, 30,
1387–1416.
13 J. S. Thorson, E. L. Sievers, J. Ahlert, E. Shepard, R. E. Whitwam,
K. C. Onwueme and M. Ruppen, Curr. Pharm. Des., 2000, 6,
1841–1879.
14 M. Moutiez, G. Lippens, C. Sergheraert and A. Tartar, Bioorg. Med.
Chem. Lett., 1997, 7, 719–724.
22 C. Neususs, U. Demelbauer and M. Pelzing, Electrophoresis, 2005, 26,
1442–1450.
23 B. G. Davis, R. C. Lloyd and J. B. Jones, J. Org. Chem., 1998, 63,
9614–9615.
24 W. M. Macindoe, A. H. van Oijen and G.-J. Boons, Chem. Commun.,
1998, 847–848.
25 B. G. Davis, M. A. T. Maughan, M. P. Green, A. Ullman and
J. B. Jones, Tetrahedron: Asymmetry, 2000, 11, 245–262.
26 D. P. Gamblin, P. Garnier, S. J. Ward, N. J. Oldham, A. J. Fairbanks
and B. G. Davis, Org. Biomol. Chem., 2003, 1, 3642–3644.
27 D. P. Gamblin, P. Garnier, S. van Kasteren, N. J. Oldham,
A. J. Fairbanks and B. G. Davis, Angew. Chem., Int. Ed., 2004, 43,
828–833.
28 G. J. L. Bernardes, D. P. Gamblin and B. G. Davis, Angew. Chem., Int.
Ed., 2006, 45, 4007–4011.
29 S. I. van Kasteren, H. B. Kramer, H. H. Jensen, S. J. Campbell,
J. Kirkpatrick, N. J. Oldham, D. C. Anthony and B. G. Davis, Nature,
2007, 446, 1105–1109.
30 M. N. Fodge and S. Al-Karadaghi, Protein Eng., 2002, 15, 353–358.
31 R. H. L. Lundin, B. E. Noren and P. O. Edlund, Tetrahedron Lett.,
1994, 35, 6339–6342.
32 L. Chen and G. Barany, Lett. Pept. Sci., 1996, 3, 283–292.
33 G. L. Kenyon and T. W. Bruice, Methods Enzymol., 1977, 47, 407–430.
34 D. N. Harpp, D. K. Ash and R. A. Smith, J. Org. Chem., 1979, 44,
4135–4140.
35 E. J. Grayson, S. J. Ward, A. L. Hall, P. M. Rendle, D. P. Gamblin,
A. S. Batsanov and B. G. Davis, J. Org. Chem., 2005, 70, 9740–9754.
36 Z. Shaked, R. P. Szajewski and G. M. Whitesides, Biochemistry, 1980,
19, 4156–4166.
37 P. G. Scheurer and F. Smith, J. Am. Chem. Soc., 1954, 76, 3224.
38 O. H. Drummer, L. Routley and N. Christophidis, Biochem.
Pharmacol., 1987, 36, 1197–1201.
39 J. K. Dale and C. S. Hudson, J. Am. Chem. Soc., 1930, 52, 2534.
40 A. Okado-Matsumoto, Z. Q. Guan and I. Fridovich, Free Radical Biol.
Med., 2006, 41, 1837–1846.
15 J. J. De Voss, C. A. Townsend, W. D. Ding, G. O. Morton, G. A.
Ellestad, N. Zein, A. B. Tabor and S. L. Schreiber, J. Am. Chem. Soc.,
1990, 112, 9669–9670.
16 N. Zein, A. M. Sinha, W. J. McGarren and G. A. Ellestad, Science,
1988, 240, 1198–1201.
17 A. Varki, Glycobiology, 1993, 3, 97–130.
18 B. G. Davis, Chem. Rev., 2002, 102, 579–601.
19 R. A. Dwek, Chem. Rev., 1996, 96, 683–720.
20 R. Apweiler, H. Hermjakob and N. Sharon, Biochim. Biophys. Acta,
1999, 1473, 4–8.
21 P. M. Rudd, T. Elliott, P. Cresswell, I. A. Wilson and R. A. Dwek,
Science, 2001, 291, 2370–2376.
41 H. D. Flack, Acta Crystallogr., Sect. A: Found. Crystallogr., 1983, 39,
876.
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 3145–3147 | 3147