Protonation and proton-coupled electron transfer at S-ligated [4Fe-4S] clusters

Article abstract of DOI:10.1002/chem.201500152

Biological [Fe-S] clusters are increasingly recognized to undergo proton-coupled electron transfer (PCET), but the site of protonation, mechanism, and role for PCET remains largely unknown. Here we explore this reactivity with synthetic model clusters. Protonation of the arylthiolate-ligated [4Fe-4S] cluster [Fe₄S₄(SAr)₄]^2- (1, SAr=S-2,4-6-(iPr)₃C₆H₂) leads to thiol dissociation, reversibly forming [Fe₄S₄(SAr)₃L]^1- (2) and ArSH (L=solvent, and/or conjugate base). Solutions of 2+ArSH react with the nitroxyl radical TEMPO to give [Fe₄S₄(SAr)₄]^1- (1_ox) and TEMPOH. This reaction involves PCET coupled to thiolate association and may proceed via the unobserved protonated cluster [Fe₄S₄(SAr)₃(HSAr)]^1- (1-H). Similar reactions with this and related clusters proceed comparably. An understanding of the PCET thermochemistry of this cluster system has been developed, encompassing three different redox levels and two protonation states. Transfer target: S-Ligated [4Fe-4S] clusters undergo proton-coupled electron transfer involving the reassociation of a thiol ligand (see figure). An understanding of the proton-coupled electron-transfer (PCET) thermochemistry of this cluster system has been developed, encompassing three different redox levels and two protonation states.

Full text of DOI:10.1002/chem.201500152

DOI: 10.1002/chem.201500152
Full Paper
&
Electron Transfer
Protonation and Proton-Coupled Electron Transfer at S-Ligated
[
4Fe-4S] Clusters
[a, b]
[c]
[c]
[a, c]
Caroline T. Saouma,*
Wesley D. Morris, Julia W. Darcy, and James M. Mayer*
1
À
Abstract: Biological [Fe-S] clusters are increasingly recog-
nized to undergo proton-coupled electron transfer (PCET),
but the site of protonation, mechanism, and role for PCET
remains largely unknown. Here we explore this reactivity
with synthetic model clusters. Protonation of the arylthio-
TEMPO to give [Fe S (SAr) ]
(1 ) and TEMPOH. This
ox
4
4
4
reaction involves PCET coupled to thiolate association and
may proceed via the unobserved protonated cluster
1À
[Fe S (SAr) (HSAr)] (1-H). Similar reactions with this and
4
4
3
related clusters proceed comparably. An understanding of
the PCET thermochemistry of this cluster system has been
developed, encompassing three different redox levels and
two protonation states.
2
À
late-ligated [4Fe-4S] cluster [Fe S (SAr) ] (1, SAr=S-2,4-6-
4
4
4
(
iPr) C H ) leads to thiol dissociation, reversibly forming
3 6 2
1À
[Fe S (SAr) L] (2) and ArSH (L=solvent, and/or conjugate
4
4
3
base). Solutions of 2+ArSH react with the nitroxyl radical
Introduction
understanding metalloenzyme mechanisms as well as catalytic
synthetic [4Fe-4S] systems.
Iron-sulphur ([Fe-S]) clusters are ubiquitous as electron-transfer
(
ET) cofactors in biology, and recent studies indicate that they
[1–4]
can engage in proton-coupled electron transfer (PCET). This Results and Discussion
is most noted in His-ligated clusters such as those in Rieske
[
1]
Cluster protonation
and mitoNEET proteins, but has also been shown to occur at
2
À
the all S-ligated [3Fe-4S] ferredoxin from Azotobacter vinelan-
O’Sullivan and Millar’s [4Fe-4S] cluster [Fe S (SAr) ] (1), where
4
4
4
[2a]
[13]
dii. PCET from the nitrogenase FeMo-co cluster inevitably
SAr is the bulky thiolate ligand S-2,4,6-(iPr) C H (ArSH), has
3
6
2
À
plays a role in reducing N to NH , though the mechanism and
good stability for these studies. The 1Àe oxidized and re-
2
3
[
5]
1À
3À
4
site(s) of protonation are still uncertain. Synthetic S-ligated
duced congeners, [Fe S (SAr) ] (1 ) and [Fe S (SAr) ] (1red)
4 4 4 ox 4 4
À
+
[6]
[
4Fe-4S] can mediate several multi-e /H transformations,
were prepared by addition of a suitable oxidant or reductant
[
7]
+
[6b,8]
[13,14]
including the reduction of CO to formate,
H
to H2,
acet-
to 1, respectively.
Thus, the redox states for both high-
2
[
6b,9]
[10]
ylene to ethylene,
and nitrobenzene to aniline. Related
potential [Fe-S] proteins and ferredoxins can be modelled by
this cluster system.
reactivity is known for chalcogels that incorporate [4Fe-4S]
[
11]
clusters. Similar to the biological cases, little is known about
the site of protonation in synthetic systems, and mechanistic
advances have been made primarily through analyses of kinet-
[
6b,12]
ic data.
Herein we describe a spectroscopic and thermo-
chemical study of PCET and proton transfer (PT) reactions of
synthetic [4Fe-4S] clusters, which begin to establish the nature
of such reactions at [4Fe-4S] clusters. This has implications in
[
a] Dr. C. T. Saouma, Prof. J. M. Mayer
Department of Chemistry, University of Washington
Box 351700, Seattle, WA 98195 (USA)
Protonation of cluster 1 in d -MeCN with 1 equivalent of the
3
strong acid [(OEt ) H]BArF (BArF =tetrakis(3,5-bis(trifluoro-
2
2
24
24
1
methyl)phenyl)borate) was monitored by H NMR spectroscopy
Eq. (1)]. A shift of the ligand resonances was observed; for in-
stance, the m-H resonance of 1 shifted and broadened from
.05 ppm to 8.25 ppm (see the Supporting Information,
[
b] Dr. C. T. Saouma
[
Department of Chemistry, University of Utah
3
15 S 1400 E Room 2020, Salt Lake City, UT 84112 (USA)
E-mail: caroline.saouma@utah.edu
8
[
c] W. D. Morris, J. W. Darcy, Prof. J. M. Mayer
Department of Chemistry, Yale University
Section V.1). In addition, a new set of resonances appeared in
the spectrum, corresponding to free thiol, ArSH, confirmed by
addition of ArSH to the sample. From the 3:1 integration of
bound ligand to free ArSH, the new cluster is assigned as
2
25 Prospect St., PO Box 208107, New Haven, CT 06520 (USA)
E-mail: james.mayer@yale.edu
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201500152.
1À
[Fe S (SAr) L] (2) in which the thiolate has been replaced by
4
4
3
Chem. Eur. J. 2015, 21, 9256 – 9260
9256
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
+
[18]
+
solvent or the conjugate base of the added acid (L). These re-
sults parallel Holm and co-workers’ report that additions of
Et NH (also in MeCN). Since the pK of Et NH in MeCN is
3
a
3
18.9, their study implies pK values for the protonated clusters
a
2À
electrophiles to [Fe S (SR) ] clusters result in thiolate release;
of >18.9. This discrepancy could be due to the different thio-
late ligands used (SPh vs. SC H iPr ), although the >5.5 unit
4
4
4
in their studies there is only limited discussion on the reaction
6
2
3
[
15]
with acid due to instability.
Addition of the strong base
pK difference seems large for a change of three alkyl substitu-
a
tBuNP(pyrr) (pyrr=pyrrolidine) to solutions of 2+ArSH, regen-
erates 1 quantitatively (by H NMR spectroscopy). The same
ents. Complex 1-H has not been observed so we have no infor-
mation about its structure, but thiolate protonation seems
likely given the Dance and Henderson results and the final for-
mation of free thiol. Protonation at a bridging sulfide cannot
be excluded, and if the formation of 1-H involves bond
3
1
reactivity is observed for this cluster in CD Cl , and for the rel-
2
2
2
À
ated less bulky arylthiolate clusters [Fe S (SC H tBu) ] and
4
4
6
4
4
2À
[
Fe S (SC H CF ) ] in d -MeCN. With these latter clusters, how-
4 4 6 4 3 4 3
3
2
ever, competitive ArS–SAr formation limits the reversibility (see
the Supporting Information, Section V.2). The dissociation of
ArSH shows that protonation eventually occurs at a thiolate,
likely a thiolate ligand of the [4Fe-4S] cluster, but our results
cleavage (e.g., change from a m -S to a m -SH), then our
estimate would be an apparent pK rather than ‘true’ pK .
a
a
Protonation of the reduced cluster 1red by [(OEt ) H]BArF in
2 2
24
d -MeCN gives, after ꢀ20 min, free thiol (1 equiv per 1red) and
3
[
2a,12]
1
do not rule out initial protonation at a bridging sulfide.
cluster 1 (ꢀ65% yield) by H NMR spectroscopy. No other clus-
Cluster 2 shows little degradation over ꢀ20 min at room
temperature, but a black precipitate forms over the course of
hours. Protonation of synthetic [Fe-S] clusters is known to facil-
ter resonances are observed. Similar results are obtained with
other acids such as [NH ]OTf and [py-H]OTf. The presence of
4
both 1 and thiol suggests at least two reactions are occurring
[
12,14,15]
1
3À
itate ligand exchange and degradation.
H NMR spectra
in the presence of acid. The related cluster, [Fe S (SPh) ] , is
4
4
4
[6]
of d -MeCN solutions of 2+ArSH show no broadening in the
known to oxidize and liberate H2 upon addition of acid.
3
thiol resonances, so chemical exchange between the thiolate
ligands and free thiol is slow on the NMR timescale.
Though our experiments were not done at sufficient concen-
1
tration to detect H by H NMR spectroscopy, this seems like
2
1À
The identity of L in [Fe S (SAr) L] (2) was probed by gener-
a plausible reaction to occur and account for the presence of
1. The formation of thiol likely occurs from successive
protonation/ligand exchange in the reduced cluster, akin to
what Holm has noted for the di-ferric/di-ferrous clusters in the
4
4
3
ating it with a variety of acids in d -MeCN. The resulting
3
1
H NMR spectra of 2 are similar when [DMF-H]OTf, [py-H]OTF
(
py=pyridine), and [H(OEt ) ]BArF are used, suggesting that L
2 2 24
[15]
is not the conjugate base or the counteranion (see the Sup-
presence of excess electrophiles.
porting Information, Section V.1). The reaction of 1 with
[
16]
[
NH ]OTf (pK =16.46),
forms an equilibrium mixture of
4
a
À4
1
and 2 [Eq. (2)], with K =3.4”10 m (see the Supporting In-
PCET studies
2
formation, Section VIII). This equilibrium does not shift upon
Mixtures of 2+ArSH (from addition of 1 equiv [py-H]OTf to 1)
addition of 1 or 10 equiv of [nBu N]OTf, further indicating that
4
react with 1 equiv of the nitroxyl radical TEMPO to generate
triflate coordination is not significant. However, when 1 is
treated with [py-H]OTf or [DMF-H]OTf, py and DMF are only
observed by NMR spectroscopy upon addition of base
1
1ox (ꢀ30%) and TEMPOH (27%), as ascertained by H NMR
spectroscopy (Scheme 1). The 2+ArSH mixture also reacts
·
with 2,4,6-tri-tert-butylphenoxyl radical, tBu ArO, (1 equiv) or
3
tBuNP(pyrr) , which suggests reversible coordination of the py
3
2
,5-ditBu-p-benzoquinone (tBu Q, 0.5 equiv) to give 1 and
2
o
x
and DMF. We conclude that L is likely a coordinating
tBu ArOH or the hydroquinone (tBu H Q), respectively. For all
3
2
2
solvent such as d -MeCN, but it can be the conjugate base of
3
reactions, an immediate color change from brown to red is ob-
served upon addition of the oxidant, and the reactions are
the acid in some cases (see the Supporting Information,
Section V.1).
K2
þ
ƒ
!
ð2Þ
1
þNH4
ƒ2 þNH þArSH
3
Equations (1) and (2) provide an upper limit for the pK of
a
1À
the unobserved protonated cluster, [Fe S (SAr) (ArSH)] (1-H)
4
4
3
of 14.4 in d -MeCN. This is a firm upper boundary because 1-H
3
would have been observed in Equation (2) if it were lower in
+
energy than 1+NH4 (see the discussion in the Supporting In-
formation, Section VIII and the free energy diagrams in Fig-
ure S31). The estimated pK indicates that 1-H is a stronger
a
acid than ArSH by ꢁ9.8 pK units. This is consistent with the
a
x+
prior report that C H SH coordination to (NH ) Ru increases
2
5
3 5
its acidity by 2.8 and 19 pK units for x=2 and 3, respective-
a
[17]
ly. This result that pK (1-H)<14.4 contrasts with very recent
a
conclusions by Dance and Henderson, from computational and
experimental (kinetic) studies, that indicates complete initial
3
Scheme 1. Reactions of 2+ArSH with H-atom acceptors (d -MeCN). L is
a neutral (e.g., solvent) ligand; see text.
2À
protonation of a thiolate ligand [Fe S (SR) ] (R=Et, Ph) by
4
4
4
Chem. Eur. J. 2015, 21, 9256 – 9260
www.chemeurj.org
9257
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
complete within ꢀ5 min (when NMR spectra are collected).
the more stable dianionic 1, oxidation of 1 forms the less
stable 1ox.
The modest yields of 1 in these reactions is likely due to the
ox
[
19]
instability of 1ox in MeCN. Addition of an oxidant such as
The aforementioned reactions are net PCET and ligand asso-
ciation; to our knowledge, they are the first examples of PCET
at synthetic S-ligated [Fe-S] clusters. Thus, net PCET from S-
ligated [4Fe-4S] clusters can occur from two oxidation states,
which correspond to that of ferredoxins and high-potential
proteins.
ferrocenium to solutions of 1 in d -MeCN does not cleanly
3
generate 1 , but rather gives mixtures of 1, 1 , thiol, and
ox
ox
other unidentifiable resonances, many of which are present in
the PCET reactions. Though the observed yields of 1ox are
modest, the yields of tBu ArOH (90%) and tBu H Q (90%) are
3
2
2
high, suggesting that 1ox degradation occurs after the PCET
reaction.
Thermochemical considerations
Competing reactions may also affect the observed yields.
The control reactions of ArSH+TEMPO (Æ[py-H]OTf) gives
The free energy changes for interconversion of the more
oxidized clusters are summarized in Scheme 3. This includes an
<
5% TEMPOH, even after monitoring for 24 h. The reaction of
1
+ArSH+TEMPO gives TEMPOH (12% yield after 15 min; see
the Supporting Information). Thus, these side reactions are not
·
the primary path to TEMPOH. ArSH also reacts with tBu ArO
3
within minutes to give ꢀ40% tBu ArOH, which again cannot
3
account for the high yields of phenol observed.
The heightened instability of oxidized clusters with less
bulky thiolate ligands precluded related PCET studies from
2À
2À
[
Fe S (SC H tBu) ] and [Fe S (SC H CF ) ] .
4 4 6 4 4 4 4 6 4
3 4
Mixtures of the trianionic cluster, 1red
, and 1 equiv
[
(OEt ) H]BArF (d -MeCN) also reduces H-atom acceptors.
2
2
24
3
Addition of TEMPO, tBu Q or tBu ArOC ꢀ10 s after protonation,
2
3
results in formation of 1 and TEMPOH, tBu H Q, or tBu ArOH,
2
2
3
respectively (Scheme 2). The yields for these PCET reactions
increase with less electron-rich/bulky clusters (Scheme 2 and
the Supporting Information, Section V.3). For all reactions that
commence with a trianionic cluster, such as 1red, the cluster
yields more closely mirror those of the oxidant than was found
in the reactions of 1. This is likely due to 1red being oxidized to
nÀ
Scheme 3. Square scheme for [Fe
4
S
4
(SAr)4Àx(HSAr)
x
]
showing the thermo-
chemical values that interconvert the cluster congeners and thiol/thiolate in
MeCN. The clusters shown in grey have not been observed. Potentials refer
to E1/2 (see the Supporting Information, Section IV). See the Supporting
Information, Section VIII for calculation of the upper limit for
DG8(1-H!2+ArSH).
upper limit for the bond dissociation free energy (BDFE) of 1-
À1
H, ꢂ64.1 kcalmol . Though this cluster is not observed, this
limit can be obtained from the reduction potential of 1 and
ox
the upper estimate for the pK of 1-H. [BDFE=23.06E (1 /
a
1/2 ox
À1
[20]
1
)+1.37 pK (1-H)+C ; C =54.9Æ1.0 kcalmol in MeCN].
a
G
G
À1
By contrast, the DG8 is 75.6 kcalmol for the conversion of
+ArSH to 2 +ArS +HC, which can be termed an “effective
BDFE.” This is determined from the pK of ArSH, and the re-
À
2
ox
[20]
a
duction potential of [Fe S (SC H iPr ) L] (2 ) (bottom of
Scheme 2. Reactions of 1red with acid then H-atom acceptors. Yields for
4
4
6
2
3 3
ox
reactions with analogous clusters are indicated below.
Scheme 3 and the Supporting Information, Sections IV and
Chem. Eur. J. 2015, 21, 9256 – 9260
www.chemeurj.org
9258 ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[
15]
VII). Thus, donation of HC from 2+ArSH without thiolate as-
binding of ArSH to give 1-H (step iii, Figure 1, DG8<À1.9 kcal
À1
À1
+
À
sociation is ꢁ11.5 kcalmol less favorable than HC donation
mol ). Then 1-H could transfer H /e by concerted bimolecu-
À1
from 1-H. This difference is due to the different oxidized prod-
lar PCET to give 1ox (DG8<À2.4 kcalmol , step iv, Figure 1).
ucts, as PCET from 1-H gives 1 while PCET from 2+ArSH
This analysis shows that 1-H is a reasonable intermediate in
the PCET reactions in Scheme 1, both energetically and in
terms of avoiding termolecular steps. A similar analysis for the
reaction of 2+ArSH+tBu ArOC to give 1 +tBu ArOH is shown
ox
À
À1
gives 2 +ArS . The ꢁ11.5 kcalmol corresponds to the dif-
ox
ference in free energy change associated with ligand exchange
(
l.e.),
DDGl.e.8
(DDG 8=DG 8(2À1-H)ÀDG 8(1 À2 );
l.e.
l.e.
l.e.
ox
ox
3
ox
3
Figure 1 and the Supporting Information, Section VIII).
in Figure 1 (bottom).
A similar thermochemical discussion is not possible for the
interconversion of 2red with 1, due to the limited stability of
the relevant cluster congeners. However, an upper limit of the
+
“
effective BDFE” of “1red +DMAP-H ” can be obtained from the
Pr
2À
reactions in Scheme 4. The [2Fe-2S] cluster [Fe S ( bbim) ] (3)
2
2
2
Figure 1. (top) Free energy diagram for the reaction of 2+ArSH+TEMPO to
ox +TEMPOH, showing two possible intermediates. The thick grey arrow in-
1
dicates that the DG8(1-H+L+TEMPO) is a lower limit. Step i) multiple-site
À
(
ms) PCET (ET from 2, PT from ArSH). ii) L/ArS ligand exchange in 2ox/1ox
.
À1
The DG8 is from Hess’ law and is similar to the DG8=13.9 kcalmol for
À
L/ArS ligand exchange in 2/1 (from K
2
, see the Suporting Information, Sec-
tion VIII). iii) L/ArSH ligand exchange in 2/1 (see the Supporting Information,
[21]
Section VIII). iv) PCET from 1-H to TEMPO; DG8 is an upper limit.
bottom) Free energy diagram for the reaction of 2+ArSH+tBu
ArOC to
ox +tBu ArOH, showing two possible intermediates. The steps are analo-
gous to those in the reaction with TEMPO. The energy of the intermediate
-H+L+tBu
ArOC is a lower limit.
(
1
3
Scheme 4. Equilibration studies to determine the “effective BDFE” for
3
_
+
Pr
1red +DMAP-H . NN= bbim.
1
3
Pr
(
bbimH =4,4-bis(benzimidazol-2-yl)heptane) is a weak H-
2
Pr
Pr
2À
The reaction of 2+ArSH+TEMPO to give 1 +TEMPOH cor-
atom acceptor, as the corresponding [Fe S ( bbim)( bbimH)]
2 2
ox
À1 [20,22]
responds to PCET and ligand exchange. It is overall exergonic
species (3red-H) has a BDFE of 60.5 kcalmol .
Addition of
À1
by at least À4.3 kcalmol (Figure 1, top). If PCET precedes
stoichiometric [DMAP-H]OTf to a d -MeCN solution containing
3
ligand exchange, the initial termolecular step is 2+ArSH+
equimolar 1red and 3 results in complete reduction and proto-
nation of 3 to 3redH and oxidation of 1red to 1, as ascertained
by NMR spectroscopy. The formation of NMR-silent 3red-H was
confirmed by addition of TEMPO to this solution to give 3 and
À
À1
TEMPO!2 +ArS +TEMPOH, uphill by 9.1 kcalmol (from
ox
+
the difference in BDFEs, step i in Figure 1). In this step, the H
À
transferred comes from ArSH and the e originates from 2 (an
example of multiple-site PCET; ms-PCET). By Hess’s law, the
[
22]
TEMPOH in ꢀ93% yield, as previously described.
Also,
+
subsequent ligand exchange to give the observed 1 is then
mixing 3 -H (generated from 3+DMAP-H ) and 1 gives no
net reaction. This shows that the BDFE of “1 +DMAP-H ” is
ox
red
À1
+
downhill by >À13.4 kcalmol (step ii, Figure 1). Alternatively,
red
À1
ligand exchange could precede PCET, through pre-equilibrium
less than 60.5 kcalmol .
Chem. Eur. J. 2015, 21, 9256 – 9260
www.chemeurj.org
9259
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Conclusion
Keywords: [Fe-S] cluster · enzyme mechanism · iron · PCET ·
electron transfer
In sum, we have shown that S-ligated [4Fe-4S] clusters can un-
dergo PCET reactions from two distinct oxidation states, which
correspond to those found in ferredoxins and high-potential
proteins. Protonation of 1 results in loss of ArSH, suggesting
that protonation occurs at a thiolate (akin to a Cys residue)
rather than at a bridging sulfide. The unobserved cluster 1-H is
[
1] a) Y. Zu, M. M. J. Couture, D. R. J. Kolling, A. R. Crofts, L. D. Eltis, J. A. Fee,
J. Hirst, Biochemistry 2003, 42, 12400–12408; b) D. W. Bak, J. A. Zuris,
M. L. Paddock, P. A. Jennings, S. J. Elliott, Biochemistry 2009, 48, 10193–
10195.
[
2] a) R. Camba, Y.-S. Jung, L. M. Hunsicker-Wang, B. K. Burgess, C. D. Stout,
J. Hirst, F. A. Armstrong, Biochemistry 2003, 42, 10589–10599; b) J. L. C.
Duff, J. L. J. Breton, J. N. Butt, F. A. Armstrong, A. J. Thomson, J. Am.
Chem. Soc. 1996, 118, 8593–8603.
À1
+
À
ꢁ
11.5 kcalmol more reducing—that is, a stronger HC (H /e )
+
À
donor—than mixtures of 2 and ArSH (which donate H /e to
À
[3] W. N. Lanzilotta, J. Christiansen, D. R. Dean, L. C. Seefeldt, Biochemistry
998, 37, 11376–11384.
give 2 +ArS ). This suggests that PCET at biological [Fe-S]
ox
1
clusters may modulate the “effective BDFE” by binding of the
acidic species to the cluster. This effective BDFE is the key
[
4] S. J. Lippard, J. M. Berg, Principles of Bioinorganic Chemistry, University
Science Books, Mill Valley, 1994.
[
20]
energetic parameter for a PCET active site. Related ligand ex-
change/protonation upon reduction of the nitrogenase P-clus-
[5] a) P. E. Doan, J. Telser, B. M. Barney, R. Y. Igarashi, D. R. Dean, L. C. See-
feldt, B. M. Hoffman, J. Am. Chem. Soc. 2011, 133, 17329–17340;
b) B. M. Hoffman, D. Lukoyanov, D. R. Dean, L. C. Seefeldt, Acc. Chem.
Res. 2013, 46, 587–595.
À
+
ter has been proposed to help control the flux of e /H to the
[
23]
active-site. Despite 1-H not being observed, thermochemical
considerations indicate that it is a reasonable intermediate in
the PCET reaction of 2+ArSH with TEMPO. To our knowledge,
these are the first examples of PCET at S-ligated [Fe-S] clusters.
These findings suggest that biological S-ligated [Fe-S] clusters
may facilitate PCET, and that this reactivity should be consid-
ered in mechanistic hypotheses for enzymes that feature these
clusters.
[
6] a) H. Seino, M. Hidai, Chem. Sci. 2011, 2, 847–857; b) K. L. C. Grçnberg,
R. A. Henderson, K. E. Oglieve, J. Chem. Soc. Dalton Trans. 1998, 3093–
3104.
[
[
7] M. Tezuka, T. Yajima, A. Tsuchiya, Y. Matsumoto, Y. Uchida, M. Hidai, J.
Am. Chem. Soc. 1982, 104, 6834–6836.
8] T. Yamamura, G. Christou, R. H. Holm, Inorg. Chem. 1983, 22, 939–949.
[9] R. S. McMillan, J. Renaud, J. G. Reynolds, R. H. Holm, J. Inorg. Biochem.
979, 11, 213–227.
10] H. Inoue, M. Shirai, E. Haruki, J. Chem. Soc. Chem. Commun. 1987, 674–
75.
1
[
[
6
11] Y. Shim, B. D. Yuhas, S. M. Dyar, A. L. Smeigh, A. P. Douvalis, M. R. Wasie-
lewski, M. G. Kanatzidis, J. Am. Chem. Soc. 2013, 135, 2330–2337.
Experimental Section
[12] R. A. Henderson, Coord. Chem. Rev. 2005, 249, 1841–1856.
[13] T. O’Sullivan, M. M. Millar, J. Am. Chem. Soc. 1985, 107, 4096–4097.
[14] P. Venkateswara Rao, R. H. Holm, Chem. Rev. 2004, 104, 527–560.
[15] R. W. Johnson, R. H. Holm, J. Am. Chem. Soc. 1978, 100, 5338–5344.
[16] K. Izutsu, Acid-Base Dissociation Constants in Dipolar Aprotic Solvents,
Blackwell Scientific Publications, Oxford, 1990.
See Supporting Information for full details. The clusters were
prepared following literature procedures.
[13,14]
Protonation and PCET Studies (NMR spectroscopy)
[
17] C. C. Kuehn, H. Taube, J. Am. Chem. Soc. 1976, 98, 689–702.
In a typical experiment, in the glovebox, d -MeCN solutions of clus-
[18] a) I. Dance, R. A. Henderson, Dalton Trans. 2014, 43, 16213–16226; b) A.
Alwaaly, I. Dance, R. A. Henderson, Chem. Commun. 2014, 50, 4799–
3
ter (1–5 mm) were prepared with ꢀ1 mm hexamethyldisilazane
4802.
(
HMDS) as an internal standard. d -MeCN solutions of acid and
3
[
19] H. L. Blonk, O. Kievit, E. K. H. Roth, J. Jordanov, J. G. M. van der Linden,
J. J. Steggerda, Inorg. Chem. 1991, 30, 3231–3234.
base were also prepared (15–60 mm). The cluster solution was ali-
quoted out (400–500 mL) into separate NMR tubes. Unless noted,
[
[
20] J. J. Warren, T. A. Tronic, J. M. Mayer, Chem. Rev. 2010, 110, 6961–7001.
21] For the purposes of this Figure, we assume L is a molecule of solvent.
Under the reaction conditions of millimolar reagents, the 2+ArSHÐ1-
H+L equilibrium lies strongly to the left. The reason that the minimum
energy of the 1-H+L state is still placed lower than 2+ArSH in
Figure 1 (top) is that the diagram shows the energies at standard state
(1m concentrations). This equilibrium involves two molecules going to
one (the solvent L not being included in the equilibrium expression), so
the 2+ArSH side is more favored at lower concentrations. This leads to
the minimum value of DG8l.e. of À1.9 kcalmolÀ1 based on the observa-
tion that Kl.e. !1.
1
equiv of acid was then added by syringe or pipettor, the tube
capped, inverted twice, immediately removed from the glovebox,
parafilm added to the cap, and frozen with liquid nitrogen. If base
(
to show reversibility) or an oxidant (to show PCET reactivity) was
added, this was done in the glovebox after inversion and prior to
removal. The NMR samples were thawed at the probe, and allowed
to temperature equilibrate ꢀ5 min prior to collection of spectra.
Spectra of the cluster (with internal standard) were also obtained
to quantify yields.
[
22] C. T. Saouma, W. Kaminsky, J. M. Mayer, J. Am. Chem. Soc. 2012, 134,
7
293–7296.
Acknowledgements
[
23] J. W. Peters, M. H. B. Stowell, S. M. Soltis, M. G. Finnegan, M. K. Johnson,
D. C. Rees, Biochemistry 1997, 36, 1181–1187.
The authors gratefully acknowledge the U.S. National Institute
of Health, GM50422 to J.M. and 1F32M099316 to C.S., for
financial support.
Received: January 13, 2015
Published online on May 12, 2015
Chem. Eur. J. 2015, 21, 9256 – 9260
www.chemeurj.org
9260
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim