Glutathione-complexed iron-sulfur clusters. Reaction intermediates and evidence for a template effect promoting assembly and stability

Article abstract of DOI:10.1039/c3cc43620a

Assembly and stabilization of a glutathione-complexed [2Fe-2S] cluster is promoted by aggregation of glutathione. The cluster core selects the tetramer species from a collection of equilibrating solution aggregate species, and in turn the core is stabilized toward hydrolytic degradation. Studies of glutathione derivatives, in combination with mass spectrometric and Moessbauer investigations provide insight on reaction intermediates during formation of [2Fe-2S](GS)₄^2-.

Full text of DOI:10.1039/c3cc43620a

ChemComm
View Article Online
View Journal
COMMUNICATION
Glutathione-complexed iron–sulfur clusters. Reaction
intermediates and evidence for a template effect
promoting assembly and stability†
Wenbin Qi,a Jingwei Li,b C. Y. Chain,c G. A. Pasquevich,c A. F. Pasquevichc and
J. A. Cowan*ab
Cite this: DOI: 10.1039/c3cc43620a
Received 14th May 2013,
Accepted 29th May 2013
DOI: 10.1039/c3cc43620a
www.rsc.org/chemcomm
Assembly and stabilization of a glutathione-complexed [2Fe–2S] cluster
Herein we evaluate the solution factors that promote the
is promoted by aggregation of glutathione. The cluster core selects the formation and stability of the glutathione-stabilized Fe–S cluster.
tetramer species from a collection of equilibrating solution aggregate These studies suggest a natural template effect, similar to the
species, and in turn the core is stabilized toward hydrolytic degradation. function of a dynamic combinatorial library, where the assembling
Studies of glutathione derivatives, in combination with mass spectro- cluster selects from an equilibrating pool of glutathione aggregates
metric and Mo¨ssbauer investigations provide insight on reaction inter- that in turn are stabilized by intermolecular salt-bridge formation.
2À
mediates during formation of [2Fe–2S](GS)4
.
Furthermore, we describe the detection of the glutathione-
complexed Fe–S cluster and reaction intermediates, the evolution
Glutathione is considered an important cellular redox buffering of iron oxidation state, and kinetics of formation of this complex by a
¨
agent as a result of its high cellular concentration and low reduction novel application of mass spectrometric and Mossbauer techniques.
potential.1 Glutathione has also been implicated in cytosolic iron–
The glutathione iron–sulfur cluster complex was synthesized
sulfur cluster biosynthesis2,3 with a recent study suggesting an by mixing reduced glutathione, ferric ion and sulfide in water at
essential role in cellular iron–sulfur cluster assembly, but only pH 8.6.7 Formation of this cluster complex was evidenced by
¨
serving as a backup to thioredoxin for maintenance of cytosolic electronic absorption, NMR, Mossbauer and electrochemical
reduction potential.4 The finding that glutaredoxins form an iron– experiments.7 Mass spectrometry was also demonstrated to be
sulfur cluster bridged dimer by incorporating two molecules of a valuable aid in cluster characterization. Although several
glutathione also supports the involvement of glutathione in cellular iron–sulfur cluster proteins have been studied by ESI-MS the
iron–sulfur cluster assembly pathways.5 Previously we reported general instability of non-protein-bound iron–sulfur cluster
glutaredoxin to undergo cluster exchange with the scaffold protein compounds has rendered the characterization of these small
ISU, consistent with glutaredoxin and glutathione playing a complexes by mass spectrometry to be challenging and very
regulatory role in cellular iron–sulfur cluster assembly.3 While small rare.9,10 Nevertheless, when the reaction mixture described
molecule iron–sulfur cluster compounds are typically stable only in herein was analyzed by ESI-MS (Fig. 1) an exact mass peak at
non-nucleophilic solvents,6 we have demonstrated the [2Fe–2S] core m/z = 1425.3 is clearly observed that is consistent with a cluster
to be stabilized in aqueous solution following coordination by four carrying two ferric ions, [(GSÀ)4[2Fe–2S]2+ + 2H+ + Na+]+.
molecules of glutathione.7 The resulting cluster complex is stable at Exchange with K+ results in the expected mass shifts (Fig. S6,
physiological pH and demonstrates exchange with the ISU scaffold ESI†). The peaks observed at 1405.3 and 1427.3 in Fig. 1
protein.7 Consequently we have proposed the glutathione iron– correspond to aggregates of glutathione in the reaction mix.
sulfur cluster complex to be a potential substrate candidate for Aggregation is discussed later in the paper and provided
the mitochondrial ABC7-type iron–sulfur cluster exporter, as well as important insight on the factors that promote the stability of
a component of the cellular labile iron pool.7,8
a complex that should not be stable, based on prior literature
precedent for ligand complexes of iron–sulfur centers.
The intensities of the peaks at 1425.3 and 1426.3 (Fig. 1)
were observed to increase as the reaction continued. In the
theoretical isotopic distribution profile, the pure oxidized
complex has the ratio 0.48. Thus, the exact mass peak at
m/z = 1426.3 represents both an isotopic peak from the oxidized
complex as well as the mixed valence complex [(GSÀ)4[2Fe–2S]+ +
3H+ + Na+]+. Based on the theoretical peak intensity ratio, we
were able to deconvolute the intensities of peaks at 1425 and
a Ohio State Biochemistry Program, The Ohio State University, 100 West 18th Ave,
Columbus, OH 43210, USA. E-mail: cowan@chemistry.ohio-state.edu
b Department of Chemistry and Biochemistry, The Ohio State University,
100 West 18th Ave, Columbus, OH 43210, USA
c IFLP-CONICET, Facultad de Ciencias Exactas, Universidad Nacional de La Plata,
Argentina
† Electronic supplementary information (ESI) available: Experimental details
concerning the synthesis and characterization of glutathione derivatives, and
¨
NMR, Mossbauer and ESI-MS experiments. See DOI: 10.1039/c3cc43620a
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun.

View Article Online
Communication
ChemComm
Fig. 2 (left) Plot of the peak intensity of the mixed valence Fe3+/Fe2+ form at
m/z = 1425.3, and the Fe3+/Fe3+ form at m/z = 1426.3, versus reaction time;
(right) Mossbauer spectrum of the isolated cluster in the solid-state recorded at
¨
room temperature. The solid line corresponds to the fitting curve. In order to
better visualize the three interactions, their contribution to the mean nuclear
cross section are presented as colored peaks (see ESI† for more details on the
interactions and the relation between the absorption cross section and the
absorption profile – the fitting curve).
Table 1 Mossbauer parameters of solid-state form cluster and assigned oxida-
¨
tion states and geometry
Contribution
Species (%)
DEQ
d (mm sÀ1) (mm sÀ1
)
Identity Geometry
I
II
III
56 Æ 1
23 Æ 1
21 Æ 1
0.21 Æ 0.02 0.55 Æ 0.03 Fe3+
1.10 Æ 0.04 2.54 Æ 0.08 Fe2+
0.33 Æ 0.02 0.79 Æ 0.03 Fe3+
4-coord, Td
5- or 6-coord
4-coord, Td
Fig. 1 Analysis of [2Fe–2S](GS)4 formation by ESI mass spectrometry. (Top) ESI-MS of
fully oxidized cluster is the most stable solution state for the
cluster, the conditions of the initial reaction mixture (with
excess iron and sulfide and thiols) maintain a thermodynamic
fraction of reduced cluster.
GSH buffer solution at pH 8.6, showing exact mass peaks at 1405.3 and 1427.3,
corresponding to the glutathione tetramer species, [(GS2À
)
+ 8Na+ + H+]+ and
4
[(GS2À
)
4
+ 9Na+]+, respectively, each with a deprotonated thiol and carboxyl (GS2À).
(Middle) ESI-MS of a reaction mixture of GSH, Fe3+, and S2À at 2 min. The exact mass
peak at m/z = 1425.3 is consistent with [(GSÀ)4[2Fe–2S]2+ + 2H+ + Na+]+ where GSÀ is
the thiolate form of glutathione, and the peak at m/z = 1426.3 is consistent with both
[(GSÀ)4[2Fe–2S]+ + 3H+ + Na+]+ and an isotopic peak of [(GSÀ)4[2Fe–2S]2+ + 2H+ + Na+]+.
(Bottom) ESI-MS of the reaction mixture of GSH, Fe3+, and S2À at 24 min. Exact mass
peaks at both m/z = 1425.3 and m/z = 1426.3 are greater, showing the formation of
the cluster in both bis-ferric Fe3+/Fe3+ and mixed valence Fe3+/Fe2+ forms.
The major difference between the solid-state and solution
spectra is species II (Table 1) which shows a feature with d B
1.10 mm sÀ1 that is assigned to either a five or six coordinate Fe2+
center in the mixed-valent cluster based on similarity to published
data, allowing for differences in sample temperature.15–18 Moreover,
the Gaussian standard deviation sigma of the interaction corre-
sponding to Fe2+ is larger than the other peaks, showing that the
atomic arrangement around Fe2+ is less defined. Expanded coordi-
nation reflects both the larger HS ferrous ion and the presence of an
intramolecular chelate effect through carboxylate ligation from one
or two a-carboxylates on two glutamate residues (Fig. 3), and is
consistent with prior reports of ferrous centers in binuclear iron15–18
as well as proton relaxation studies that indicated the glutamate
side chains to lie in close proximity to the cluster center.7
Significantly, carboxylate coordination does not impact the
expected m/z values in ESI-MS experiments.
The mass spectral peaks at 1413.3 and 1435.3 (Fig. 1)
correspond to an intermediate species, which has a cluster
center of [2Fe3+–S2À] with four coordinated molecules of
glutathione. This species is most likely an intermediate formed
on the reaction pathway because cluster alone does not yield
these peaks. A recent report described an intermediate cluster
assembly product, in a complex of the sulfur donor and cluster
scaffold proteins (IscS-IscU) with fully oxidized iron and a
related bridging persulfide (Fe2S–SÀ) adduct.19 Accordingly,
the [(GSÀ)4[2Fe3+–S2À] species that we observe could represent
an intermediate awaiting delivery of the second sulfide to
complete the cluster core.
1426 and calculate the actual intensity of bis-ferric Fe3+/Fe3+ and
mixed valence Fe3+/Fe2+ species, respectively. By plotting calculated
peak intensity versus time and fitting to first order exponential
kinetics, the observed reaction rate constants for the formation of
each of the Fe3+/Fe3+ and Fe3+/Fe2+ species are 0.16 Æ 0.03 minÀ1
and 0.34 Æ 0.10 minÀ1, respectively (Fig. 2).
It is clear that after an initial phase, the solution quickly
reaches an equilibrium in which there is a steady-state ratio of
fully oxidized cluster (species I signals) and mixed valence
cluster (species II and III signals). When a sample of reaction
solution is ethanol precipitated and the solid material then
¨
analysed by Mossbauer spectroscopy, both the mixed valence
Fe3+/Fe2+ and fully oxidized Fe3+/Fe3+ cluster species are evident
(Fig. 2, right). Isomeric and quadrupolar shifts of III agree with
literature values of ferric iron in the reduced 2Fe–2S cluster.11–14
The fitted spectra also indicated an equimolar ratio of species II
and III, and support the identity of the one-electron reduced diiron
¨
cluster. Mossbauer experiments show that when the freshly synthe-
sized cluster complex is isolated by ethanol precipitation and
redissolved with excess glutathione then only fully oxidized cluster
is present, exhibiting the NMR and other physical characteristics of
2À
the [2Fe–2S](GS)4 cluster.7 These results indicate that while the
c
Chem. Commun.
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
macrocycle that stabilizes and promotes formation of the iron–
sulfur cluster complex.
Characterization of small molecule iron–sulfur cluster com-
plexes in aqueous solution has proven to be a challenge as a
result of their hydrolytic instability. Herein we have obtained
experimental support for a hypothesis that explains the stability
associated with glutathione-complexed iron–sulfur complexes.
In particular, the propensity of glutathione to aggregate, apparently
through intermolecular salt bridge and hydrogen bond formation,
yields a pre-assembled tetrameric species that forms a stable
binding pocket for a [2Fe–2S] cluster core. Moreover, it was possible
to identify certain reaction intermediates and monitor the kinetics
Fig. 3 A two-dimensional representation of a glutathione-complexed cluster
aggregate. Salt bridge formation between carboxylates and protonated amines
appear to favor aggregation of glutathione and is supported by the ionic strength
dependence of cluster stability (Fig. S7, ESI†) and the effect of acetylation and
esterification that effectively eliminate multimer formation (Fig. S4 and S5, ESI†). In
glutathione solutions the tetrameric oligomer is the most abundant species, as
reflected by mass spectra (Fig. 4) and appears to create a pocket for cluster binding
that is optimal in size, relative to other aggregate forms. Presumably this pocket
mimics a protein binding site by both providing a pre-assembled ligand set, as well
as providing a measure of protection from solvent access in the folded state.
¨
of cluster formation by ESI-MS and Mossbauer experiments.
The formation of glutathione iron–sulfur cluster complex was
confirmed by the appearance of exact mass peaks at 1425.3 and
1426.3, which correspond to fully oxidized and mixed valence
species, respectively. Reaction kinetics was studied by following
peak intensities in ESI-MS spectra and apparent first order reaction
constants were obtained. The structural model that we propose can
be viewed as nature’s equivalent of a dynamic combinatorial
selection experiment from a pool of equilibrating glutathione
oligomers. In this case a [2Fe–2S] core selectively binds and
stabilizes a tetrameric macrocyclic aggregate, and in turn is
stabilized toward hydrolysis.
This work was supported by a grant from the National
Institutes of Health [AI072443].
Notes and references
1 A. Pompella, A. Visvikis, A. Paolicchi, V. De Tata and A. F. Casini,
Biochem. Pharmacol., 2003, 66, 1499–1503.
2 K. Sipos, H. Lange, Z. Fekete, P. Ullmann, R. Lill and G. Kispal,
J. Biol. Chem., 2002, 277, 26944–26949.
3 W. Qi and J. A. Cowan, Chem. Commun., 2011, 47, 4989–4991.
4 C. Kumar, A. Igbaria, B. D’Autreaux, A. G. Planson, C. Junot,
E. Godat, A. K. Bachhawat, A. Delaunay-Moisan and
M. B. Toledano, EMBO J., 2011, 30, 2044–2056.
Fig. 4 Solution aggregates of glutathione observed by ESI-MS, with evidence of
trimers, tetramers, pentamers, etc. . . . formation. These solutions were not pH
adjusted with NaOH and do not show the Na+ adducts evident in Fig. 1 (top).
5 C. Johansson, K. L. Kavanagh, O. Gileadi and U. Oppermann, J. Biol.
Chem., 2007, 282, 3077–3082.
6 J. H. Enemark, J. J. Cooney, J. J. Wang and R. H. Holm, Chem. Rev.,
2004, 104, 1175–1200.
7 W. Qi, J. Li, C. Y. Chain, G. A. Pasquevich, A. F. Pasquevich and
J. A. Cowan, J. Am. Chem. Soc., 2012, 134, 10745–10748.
8 G. Kuhnke, K. Neumann, U. Muhlenhoff and R. Lill, Mol. Membr.
Biol., 2006, 23, 173–184.
9 Y. Petillot, M. P. Golinelli, E. Forest and J. Meyer, Biochem. Biophys.
Res. Commun., 1995, 210, 686–694.
Crystallographic studies have earlier revealed a hydrogen
bond network in crystals of glutathione20 and it is of significant
interest that a solution of glutathione by itself shows evidence of
substantial aggregation, with trimers, tetramers, pentamers,
etc. . . ., that are clearly visible in mass spectra (Fig. 4); especially
in the lower m/z range where such aggregates are better distin-
guished from the sodium adducts exhibited in Fig. 1 (top). The
(GSH)4 tetramer is the most abundant species evident in Fig. 4.
Aggregation is essentially eliminated by carboxyl ester formation
10 P. V. Rao and R. H. Holm, Chem. Rev., 2004, 104, 527–559.
11 E. Munck, P. G. Debrunner, J. C. M. Tsibris and I. C. Gunsalus,
Biochemistry, 1972, 11, 855–863.
or amine acetylation, respectively (Fig. 4 versus Fig. S4 and S5, 12 W. R. Dunham, A. J. Bearden, I. T. Salmeen, G. Palmer, R. H. Sands,
W. H. Orme-Johnson and H. Beinert, Biochim. Biophys. Acta, 1971,
253, 134–152.
13 R. E. Anderson, W. R. Dunham, R. H. Sands, A. J. Bearden and
ESI†), while increasing ionic strength also yields the pronounced
decrease in cluster stability (Fig. S7, ESI†) that is expected when
the salt bridges are disrupted. Fig. 3 illustrates a likely inter-
molecular salt bridge/H-bonding network for glutathione tetra-
mer that appears to be of the correct size to serve as a
preassembled iron–sulfur cluster chelate, ready to accept free
iron and sulfide to form the stable cluster complex. No other
glutathione aggregates are observed in the presence of cluster.
H. L. Crespi, Biochim. Biophys. Acta, 1975, 408, 306–318.
14 J. Meyer, M. D. Clay, M. K. Johnson, A. Stubna, E. Mu¨nck, C. Higgins
and P. Wittung-Stafshede, Biochemistry, 2002, 41, 3096–3108.
15 Y. Z. Zheng, W. Xue, M. L. Tong, X. M. Chen, F. Grandjean and
G. J. Long, Inorg. Chem., 2008, 47, 4077–4087.
16 S. Yoon and S. J. Lippard, J. Am. Chem. Soc., 2005, 127, 8386–8397.
´
17 V. Paredes-Garcıa, V. Venegas-Yazigi, R. O. Latorre and E. Spodine,
Polyhedron, 2006, 25, 2026–2032.
Apparently there is a synergic interaction with the tetramer 18 R. G. Burns, Hyperfine Interact., 1994, 9, 739–745.
19 E. N. Marinoni, J. S. de Oliveira, Y. Nicolet, E. C. Raulfs, P. Amara,
species selected by the cluster core, which in turn is stabilized
by the glutathione aggregate. Through hydrogen bonding and
D. R. Dean and J. C. Fontecilla-Camps, Angew. Chem., Int. Ed., 2012,
51, 5439–5442.
salt-bridge formation, glutathione forms an apparent tetrameric 20 W. B. Wright, Acta Crystallogr., 1958, 11, 632–642.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun.