Tuning the rate and pH accessibility of a conformational electron transfer gate

Article abstract of DOI:10.1021/ic0603712

Methods to fine-tune the rate of a fast conformational electron transfer (ET) gate involving a His-heme alkaline conformer of iso-1-cytochrome c (iso-1-Cytc) and to adjust the pH accessibility of a slow ET gate involving a Lys-heme alkaline conformer are described. Fine-tuning the fast ET gate employs a strategy of making surface mutations in a substructure unfolded in the alkaline conformer. To make the slow ET gate accessible at neutral pH, the strategy involves mutations at buried sequence positions which are expected to more strongly perturb the stability of native versus alkaline iso-1-Cytc. To fine-tune the rate of the fast His 73-heme ET gate, we mutate the surface-exposed Lys 79 to Ala (A79H73 variant). This mutation also simplifies ET gating by removing Lys 79, which can serve as a ligand in the alkaline conformer of iso-1-Cytc. To adjust the pH accessibility of the slow Lys 73-heme ET gate, we convert the buried side chain Asn 52 to Gly and also mutate Lys 79 to Ala to simplify ET gating (A79G52 variant). ET kinetics is studied as a function of pH using hexaammineruthenium(II) chloride (a₆Ru ²⁺) to reduce the variants. Both variants show fast direct ET reactions dependent on [a₆Ru²⁺] and slower gated ET reactions that are independent of [a₆Ru²⁺]. The observed gated ET rates correlate well with rates for the alkaline-to-native state conformational change measured independently. Together with the previously reported H73 variant (Baddam, S.; Bowler, B. E. J. Am. Chem. Soc. 2005, 127, 9702-9703), the A79H73 variant allows His 73-heme-mediated ET gating to be fine-tuned from 75 to 200 ms. The slower Lys 73-heme (15-20 s time scale) ET gate for the A79G52 variant is now accessible over the pH range 6-8.

Full text of DOI:10.1021/ic0603712

Inorg. Chem. 2006, 45, 6338−6346
Tuning the Rate and pH Accessibility of a Conformational Electron
Transfer Gate
Saritha Baddam and Bruce E. Bowler*
Department of Chemistry and Biochemistry, UniVersity of DenVer, 2190 East Iliff AVenue, DenVer,
Colorado 80208-2436
Received March 6, 2006
Methods to fine-tune the rate of a fast conformational electron transfer (ET) gate involving a His
conformer of iso-1-cytochrome c (iso-1-Cytc) and to adjust the pH accessibility of a slow ET gate involving a
Lys heme alkaline conformer are described. Fine-tuning the fast ET gate employs a strategy of making surface
−heme alkaline
−
mutations in a substructure unfolded in the alkaline conformer. To make the slow ET gate accessible at neutral pH,
the strategy involves mutations at buried sequence positions which are expected to more strongly perturb the
stability of native versus alkaline iso-1-Cytc. To fine-tune the rate of the fast His 73−heme ET gate, we mutate the
surface-exposed Lys 79 to Ala (A79H73 variant). This mutation also simplifies ET gating by removing Lys 79,
which can serve as a ligand in the alkaline conformer of iso-1-Cytc. To adjust the pH accessibility of the slow Lys
73
−
heme ET gate, we convert the buried side chain Asn 52 to Gly and also mutate Lys 79 to Ala to simplify ET
gating (A79G52 variant). ET kinetics is studied as a function of pH using hexaammineruthenium(II) chloride
Ru2+) to reduce the variants. Both variants show fast direct ET reactions dependent on [a Ru2+] and slower
gated ET reactions that are independent of [a
Ru2+]. The observed gated ET rates correlate well with rates for the
alkaline-to-native state conformational change measured independently. Together with the previously reported H73
variant (Baddam, S.; Bowler, B. E. J. Am. Chem. Soc. 2005, 127, 9702 9703), the A79H73 variant allows His
heme-mediated ET gating to be fine-tuned from 75 to 200 ms. The slower Lys 73 heme (15 20 s time scale)
ET gate for the A79G52 variant is now accessible over the pH range 6 8.
(a
6
6
6
−
73
−
−
−
−
Introduction
in both manipulating protein function and engineering
molecular electronics devices.
Electron transfer (ET) is among the simplest of chemical
processes occurring in living organisms, yet certainly one
of the most critical. Many biological processes, such as
photosynthesis and respiration, are regulated by ET reac-
tions.1-4 It has become apparent that conformational gating
of biological ET reactions can regulate metabolic processes,
and an increasing number of examples of gated ET in
proteins is being reported.3,5 Conformational gating also
appears to be an important aspect of protein ET at electrode
interfaces and thus will be key to developing protein-based
molecular electronics devices.4,6 Therefore, the ability to tune
Changes in the coordination sphere of a metal can result
in significant changes in redox potential and thus can favor
7
ET from one coordination environment versus another. In
proteins, changes in coordination environment are expected
to require changes in the conformation of the protein. Thus,
the rate of a conformational change which controls protein
3
(
5) (a) Blazyk, J. L.; Gassner, G. T.; Lippard, S. J. J. Am. Chem. Soc.
2005, 127, 17364-17376. (b) Hoffman, B. M.; Celis, L. M.; Cull, D.
A.; Patel, A. D.; Seifert, J. L.; Wheeler, K. E.; Wang, J.; Yao, J.;
Kurnikov, I. V.; Nocek, J. M. Proc. Natl. Acad. Sci. U.S.A. 2005,
1
02, 3564-3569. (c) Di Bilio, A. J.; Dennison, C.; Gray, H. B.;
the properties of conformational ET gates will be essential
Ramirez, B. E.; Sykes, A. G.; Winkler, J. R. J. Am. Chem. Soc. 1998,
1
20, 7551-7556. (d) Kosti c´ , N. M.; Ivkoviæ-Jensen, M. M. Biochem-
*
To whom correspondence should be addressed. Phone: (303) 871-
istry 1997, 36, 8135-8144. (e) Mei, H.; Wang, K.; Peffer, N.;
Weatherly, G.; Cohen D. S.; Miller, M.; Pielak, G.; Durham, B.; Millet,
F. Biochemistry 1999, 38, 6846-6854.
2
985. Fax: (303) 871-2254. E-mail: bbowler@du.edu.
(
(
1) Gray, H. B.; Winkler, J. R. Q. ReV. Biophys. 2003, 36, 341-372.
2) Gray, H. B.; Winkler, J. R. Proc. Natl. Acad. Sci. U.S.A. 2005, 102,
(6) (a) Chi, Q.; Farver, O.; Ulstrup, J. Proc. Natl. Acad. Sci. U.S.A. 2005,
102, 16203-16208. (b) Fujita, K.; Nakamura, N.; Ohno, H.; Leigh,
B. S.; Niki, K.; Gray, H. B.; Richards, J. R. J. Am. Chem. Soc. 2004,
126, 13954-13961. (c) Avila, A.; Gregory, B. W.; Niki, K.; Cotton,
T. M. J. Phys. Chem. B 2000, 104, 2759-2766.
3
534-3539.
(
3) (a) Davidson, V. L. Biochemistry 2002, 34, 14633-14636. (b)
Davidson, V. L. Acc. Chem. Res. 2000, 33, 87-93.
4) Jeuken, L. J. C. Biochim. Biophys. Acta 2003, 1604, 67-76.
(
6338 Inorganic Chemistry, Vol. 45, No. 16, 2006
10.1021/ic0603712 CCC: $33.50
© 2006 American Chemical Society
Published on Web 06/30/2006

Metal-Ligand Effects on Gated Electron Transfer
Beyond these convenient properties for manipulating gated
ET, a number of studies suggest that the alkaline conformer
of Cytc may play a role in its function in mitochondrial
1
3
electron transport.
In mitochondrial cytochromes c, the midpoint pH for
formation of the alkaline state of Cytc ranges from 8.5 to
1
4
1
1. To make this conformer accessible near physiological
pH, we have exploited the well-understood thermodynamics
of this protein. Cytc is known to have five cooperative
substructures, as outlined in Figure 1, that unfold sequen-
1
5
tially. In other words the thermodynamics of Cytc are
hierarchical. The two least stable substructures (shown in
gray and red in Figure 1) unfold in the alkaline conformer.1
For either of the alkaline state ligands, Lys 73 or Lys 79, to
displace Met 80, the red substructure must be disrupted (see
Figure 1). Since the gray substructure is less stable than the
red substructure and unfolding of the substructures of Cytc
is sequential, mutations which destabilize either of these
substructures should destabilize the native state relative to
the alkaline state of Cytc. In previous work on yeast iso-1-
Cytc, we have shown that stabilizing mutations at position
Figure 1. Yeast iso-1-cytochrome c in the oxidized state is shown with
the substructure classifications of horse cytochrome c according to ref 15.
The substructures are color-coded gray (N-yellow substructure, residues
6-18
4
0-57), red (residues 71-85), yellow (residues 37-39, 58-61, in back of
structure), green (60’s helix and 20’s-30’s loop), and blue (N- and
C-terminal helices) in order of increasing stability. The heme (blue, iron in
red) and its ligands (magenta), His 18 and Met 80, are shown as stick
models. The Lys 73 and Lys 79 heme ligands in the alkaline conformer
and Asn 52, which is mutated to Gly in the A79G52 variant, are shown as
space-filling models (R-carbon and side-chain atoms) in the color of the
substructure in which they are contained. Lys 79 is mutated to Ala in both
the A79G52 and A79H73 variants so that there is only one possible heme
ligand in the alkaline conformer, Lys 73 and His 73, respectively. The figure
was prepared with DS ViewerPro software and the Protein Data Bank file,
5
2 in the gray substructure (see Figure 1) stabilize the
1
2
18d
2
YCC.
alkaline conformer relative to the native state. Previous
studies on the effects of mutations at position 52 on the global
stability of iso-1-Cytc show that the stability of the variant
proteins depends on the size of the amino acid at this
ET can be modulated by both the ligands involved and the
8
nature of the conformational change. An early example of
conformationally gated protein ET involving a metal-ligand
exchange reaction was observed for the alkaline conformer
of cytochrome c (Cytc).9 More recently, metal-ligand
exchange reactions have been implicated in gated ET for
plastocyanin.5
19
position. In particular, an Asn 52 f Gly mutation
significantly decreased the global stability of iso-1-Cytc. On
this basis, we created an A79G52 (Lys 79 f Ala and Asn
5
2 f Gly mutations) variant of iso-1-Cytc. Our thermody-
namic and kinetic analyses have verified that A79G52 iso-
-Cytc makes the alkaline conformer accessible near neutral
c
Conformational gates that operate near neutral pH are
desirable both for engineered function and as components
in molecular electronics devices. The ability to tune the rate
of an ET gate over narrow ranges, as well as over orders of
magnitude, is also important. Cytc provides a particularly
useful scaffold for tuning these properties for a conforma-
tionally gated ET reaction. In the alkaline conformer, lysine
ligation leads to an ∼0.5 V decrease in the affinity of yeast
iso-1-cytochrome c (iso-1-Cytc) for an electron,10 whereas
histidine ligation is expected to cause a decrease of ∼0.25
1
20
pH (midpoint pH ∼7.4). The Lys 79 f Ala mutation
simplifies gating kinetics by making Lys 73 the only heme
ligand in the alkaline conformer of this variant. As discussed
below, this latter mutation might also be expected to
modestly perturb the dynamics of the conformational change
from the native to the alkaline state. However, since the
kinetics of formation of the lysine 73 and 79 alkaline
conformers cannot be determined individually when both are
1
1
V, relative to the native heme-Met 80 ligation. Thus,
conformationally gated ET is expected for both alkaline
states. For iso-1-Cytc, the heme ligands are known to be
(12) Berghuis, A. M.; Brayer, G. D. J. Mol. Biol. 1992, 223, 959-976.
(
13) (a) D o¨ pner, S.; Hildebrant, P.; Rosell, F. I.; Mauk, A. G. J. Am. Chem.
Soc. 1998, 120, 11246-11255. (b) D o¨ pner, S.; Hildebrant, P.; Rosell,
F. I.; Mauk, A. G.; von Walter, M.; Buse, G.; Soulimane, T. Eur. J.
Biochem. 1999, 261, 379-391.
10a
lysines 73 and 79 in the alkaline conformer (see Figure 1),
allowing heme ligation in this state to be readily manipulated.
(14) Moore, G. R.; Pettigrew, G. W. Cytochrome c: EVolutionary,
Structural and Physicochemical Aspects; Springer-Verlag: New York,
1
990; pp 184-196.
(
7) (a) Rorabacher, D. B. Chem. ReV. 2004, 104, 651-697. (b) Meagher,
N. E.; Juntunen, K. L.; Salhi, C. A.; Ochrymowycz, L. A.; Rorabacher,
D. B. J. Am. Chem. Soc. 1992, 114, 10411-10420. (c) Wijetunge, P.;
Kulatilleke, C. P.; Dressel, L. T.; Heeg, M. J.; Ochrymowycz, L. A.;
Rorabacher, D. B. Inorg. Chem. 2000, 39, 2897-2905.
(15) (a) Bai, Y.; Sosnick, T. R.; Mayne, L.; Englander, S. W. Science 1995,
269, 192-197. (b) Krishna, M. M. G.; Lin, Y.; Rumbley, J. N.;
Englander, S. W. J. Mol. Biol. 2003, 331, 29-36.
(16) Hoang, L.; Maity, H.; Krishna, M. M. G.; Lin, Y.; Englander, S. W.
J. Mol. Biol. 2003, 331, 37-43.
(
(
8) Baddam, S.; Bowler, B. E. J. Am. Chem. Soc. 2005, 127, 9702-9703.
9) (a) Greenwood, C.; Palmer, G. J. Biol. Chem. 1965, 240, 3660-3663.
(17) Assfalg, M.; Bertini, I.; Dolfi, A.; Turano, P.; Mauk, A. G.; Rosell,
F. I.; Gray, H. B. J. Am. Chem. Soc. 2003, 125, 2913-2922.
(18) (a) Godbole, S.; Dong, A.; Garbin, K.; Bowler, B. E. Biochemistry
1997, 36, 119-126. (b) Godbole, S.; Bowler, B. E. Biochemistry 1999,
38, 487-495. (c) Nelson, C. J.; Bowler, B. E. Biochemistry 2000, 39,
13584-13594. (d) Kristinsson, R.; Bowler, B. E. Biochemistry 2005,
44, 2349-2359.
(
1
b) Wilson, M. T.; Greenwood, C. Eur. J. Biochem. 1971, 22, 11-
8. (c) Hodges, H. L.; Holwerda, R. A.; Gray, H. B. J. Am. Chem.
Soc. 1974, 96, 3132-3137.
10) (a) Rosell, F. I.; Ferrer, J. C.; Mauk, A. G. J. Am. Chem. Soc. 1998,
(
(
1
20, 11234-11245. (b) Barker, P. D.; Mauk, A. G. J. Am. Chem.
Soc. 1992, 114, 3619-3624.
(19) Hickey, D. R.; Berghuis, A. M.; Lafond, G.; Jaeger, J. A.; Cardillo,
T. S.; McLendon, D.; Das, G.; Sherman, F.; Brayer, G. D.; McLendon,
G. J. Biol. Chem. 1991, 266, 11686-11694.
11) (a) Raphael, A. L.; Gray, H. B. J. Am. Chem. Soc. 1991, 113, 1038-
1
040. (b) Feinberg, B. A.; Liu, X.; Ryan, M. D.; Schejter, A.; Zhang,
C.; Margoliash, E. Biochemistry 1998, 37, 13091-13101.
(20) Baddam, S.; Bowler, B. E. Biochemistry 2006, 45, 4611-4619.
Inorganic Chemistry, Vol. 45, No. 16, 2006 6339

Baddam and Bowler
present, it is not possible to assess the magnitude of this
effect. Thus, the primary goal for ET studies on the A79G52
variant is to make the slower conformational gating due to
the lysine-heme alkaline conformer accessible near neutral
pH. Here, we use reaction of the Lys 73-heme alkaline state
of A79G52 iso-1-Cytc with hexaammineruthenium(II)
Commercial [Ru(NH
3
)
6
]Cl
3
(Strem Chemicals) was reduced with
24
zinc by the method of Fergusson and Love. The [Ru(NH
)
3 6
]Cl
2
+
(
a
6
Ru2 ) was dried in a vacuum desiccator which was then refilled
with argon. The product was stored at -20 °C, and care was taken
to minimize exposure of [Ru(NH ]Cl to air and moisture to
]Cl was confirmed
)
3 6
2
prevent oxidation. The formation of [Ru(NH
3
)
6
2
by infrared spectroscopy (1217 cm- band characteristic of Ru(II)
1
2+
6
(a Ru ) to verify that gated ET occurs near neutral pH with
complex).25
this variant. We find that gated ET is observable from pH 6
to 8 on an ∼15-20 s time scale.
Experiments were done with the following buffers: 10 mM MES
in 0.1 M NaCl, pH 6.0 and 6.5; 10 mM Tris buffer in 0.1 M NaCl,
pH 7.5 and 8.0. The pH of buffers was adjusted with NaOH(aq) or
HCl(aq), as appropriate.
The second goal of this report is to develop means of fine-
tuning the rate of an ET gate. We have recently reported
that replacement of Lys 73 with His (H73 variant) dramati-
cally increases the rate of conformationally gated ET from
the alkaline conformer simply by changing the heme ligand
Argon gas (high purity grade, 99.99%) that was further purified
with an oxy-trap column (Alltech, Inc.) was used to degas all
solutions for anaerobic work.
8
in the alkaline conformer from a lysine to a histidine. The
Anaerobic Stopped-Flow Kinetics Measurements. Stopped-
rate of the gate could also be increased by a factor of nearly
flow mixing of a Ru2 with the oxidized form of the A79H73 and
+
8
2
.0 by decreasing the pH from 7.5 to 6. Furthermore, the
6
A79G52 variants was carried out using an Applied Photophysics
gated ET rates as a function of pH correlated well with the
rate of the conformational change from the His 73-heme
alkaline conformer to the native state measured independently
by pH jump methods.8,21 As a strategy to further fine-tune
π*-180 spectrometer operating in kinetics mode, as previously
8
described. Briefly, the day before experiments were done, the flow
cell and the drive syringes were soaked overnight in 50 µM
riboflavin, reduced with 1 mM EDTA and light, to ensure anaerobic
conditions.26 Before the experiments were started, the riboflavin
solution was driven out of the flow cell and syringes and then the
appropriate buffer, degassed with Ar, was used to wash out any
the rate of gated ET due to the His 73-heme alkaline
conformer, a logical approach is to make mutations of surface
residues in the red substructure which is unfolded in the
alkaline conformer. Surface mutations are expected to more
remaining riboflavin. Iso-1-Cytc variants were oxidized with K
3
-
2
2
modestly affect the stability and dynamics of a protein.
[
Fe(CN) ] followed by G-25 chromatography to remove the reagent
6
As a first attempt at this strategy, we have added the Lys 79
f Ala mutation to the H73 variant to create the A79H73
variant. Besides providing for a modest perturbation to the
conversion from the alkaline conformer to the native
conformer, this variant also removes Lys 79 as an alkaline-
state ligand, simplifying the kinetics of ET gating. We have
also carried out thorough kinetic and thermodynamic analyses
of the alkaline conformational transition of the A79H73
variant which demonstrate that the dynamics of the native
to His 73-heme alkaline conformer have been modestly
perturbed, as predicted.23 This previous work will allow for
direct correlation of the pH dependence of the conformational
change from the His 73-heme alkaline state to the native
state measured by pH jump methods with gated ET measured
and change the protein into the appropriate buffer. Oxidized protein
was diluted to 10 µM with Ar-degassed buffer and then further
2+
6
degassed on a dual Ar/vacuum manifold. All a Ru solutions were
prepared immediately before use. A known volume of Ar-degassed
3 6 2
buffer was cannula-transferred onto solid [Ru(NH ) ]Cl in an Ar-
degassed vessel. The degassed solutions of protein and a Ru2+ were
6
mixed 1:1 with the π*-180 spectrometer producing final solutions
2+
containing 5 µM protein and a Ru at 1.25, 2.5, or 5 mM in 10
6
mM buffer containing 100 mM NaCl. Reduction of the heme of
the iso-1-Cytc variants was monitored at 550 nm.
At each concentration of a Ru2+, a minimum of four trials was
6
done and a total of 1000 points was collected over 2 s on a
logarithmic time scale. Data were also collected on a longer time
scale for the A79G52 variant (50-100 s, 5000 data points on a
logarithmic time scale). The dead time for stopped-flow mixing
was measured using ascorbate reduction of 1,6-dichlorophenolin-
dophenol27 and was found to be 1.0 ms under our mixing conditions.
The data were corrected for the dead time of the experiment.
2+
here by reaction with a
to modulate gated ET over almost a 3-fold time scale, from
5 to 200 ms using the combined effects of pH and
6
Ru . We find that we are now able
7
mutagenesis with the A79H73 and H73 variants of iso-1-
Cytc.
Analysis of the data was done using the curve fitting program,
SigmaPlot (v. 7.0). The data were fit to single or double exponential
rise to maximum equations depending on the protein and the time
scale.
Experimental Section
Proteins and Reagents. The A79G52 and A79H73 variants of
iso-1-Cytc were isolated and purified, as previously described,
20,23
from the Saccharomyces cereVisiae cell-line GM-3C-2 (Cytc
deficient) transformed with the pRS/C7.8 vector carrying the
appropriately mutated iso-1-Cytc gene. Both variants also carry the
mutation Cys 102 f Ser to prevent formation of intermolecular
disulfide dimers during physical studies.
Results
Kinetic Model for Gated ET. The kinetics of conforma-
tionally gated ET can be analyzed with a square scheme
(
24) Fergusson, J. E.; Love, J. L. Inorg. Synth. 1972, 13, 208-213.
(
(
(
21) Martinez, R. E.; Bowler, B. E. J. Am. Chem. Soc. 2004, 126, 6751-
(25) Allen, A. D.; Senoff, C. V. Can. J. Chem. 1966, 45, 1337-1341.
(26) Ramsey, R. R.; Koerber, S. C.; Singer, T. P. Biochemistry 1987, 26,
3045-3050.
(27) Tonomura, B.; Nakatani, H.; Ohnishi, M.; Yamaguchi-Ito, J.; Hiromi,
K. Anal. Biochem. 1978, 84, 370-383.
6758.
22) Alber, T.; Dao-pin, J. A.; Muchmore, D. C.; Matthews, B. W.
Biochemistry 1987, 26, 3754-3758.
23) Baddam, S.; Bowler, B. E. Biochemistry 2005, 44, 14956-15968.
6340 Inorganic Chemistry, Vol. 45, No. 16, 2006

Metal-Ligand Effects on Gated Electron Transfer
Figure 2. ET square scheme for the reaction of oxidized iso-1-cytochrome c with hexaammineruthenium(II) (a6Ru2+). Horizontal equilibria are ET equilibria,
and vertical equilibria are conformational equilibria. The oxidized and reduced alkaline conformers of iso-1-Cytc are shown in the upper row of the ET
square scheme. The alkaline state iron ligand, L, is His 73 for the A79H73 variant and Lys 73 for the A79G52 variant. The oxidized and reduced native
2
+
conformers of iso-1-Cytc are shown in the lower row of the ET square scheme. Reduction of the oxidized alkaline conformer of iso-1-Cytc by a6Ru can
3+
occur via either path A or path B as described in the text. The rate constants controlling the conformational change in the oxidized (Fe ) state of iso-1-Cytc
2
+
are kLM3 and kML3 and in the reduced (Fe ) state of iso-1-Cytc are kLM2 and kML2. The rate constants controlling ET in the alkaline state are kET(Alk) and
k-ET(Alk) and in the native state are kET(Native) and k-ET(Native). Path B is expected to be dominant (see text) for the reaction of the oxidized alkaline state of
2
+
iso-1-Cytc with a6Ru . Therefore, gated ET will be controlled by the rate constant kLM3 and direct ET by the rate constant kET(Native).
model (Figure 2).7,28 In the square scheme model, the upper
left-hand conformer of Cytc is taken to be more stable in
the oxidized state (oxidized alkaline state with His 73 or
Lys 73 bound to the heme) and the lower right-hand
conformer of Cytc is taken to be more stable in the reduced
state (reduced native state with Met 80 bound to heme). Thus,
in reducing the alkaline conformer of Cytc, two pathways
are possible. In path A, the alkaline conformer is reduced
first followed by a conformational change to the more stable
native reduced state of Cytc. In path B, the conformational
change to the native conformer of the oxidized state occurs
first followed by reduction of the oxidized native conformer.
In the alkaline conformer of the A79H73 variant, His 73
will be bound to the heme whereas Lys 73 will be bound to
the heme in the alkaline state of the A79G52 variant. Since
the reduction potential of the native state of wild type (WT)
< kLM3/kML3 < 10), both direct intermolecular ET to the
oxidized native state (kET(native) in Figure 2) and conforma-
tionally gated ET (kLM3) from the oxidized alkaline state
should be detectable.
With path B dominant, the rate law for the bimolecular
2
+
III
6
reaction between a Ru and the Fe state of the A79H73
7
and A79G52 variants has the form given in eqs 1 and 2,
where the rate
III
III
2+
-
d[Fe heme]/dt ) k [Cytc(Fe -L)][ a Ru ]
obs 6
2
+
kobs ) {kET(Native)kLM3/(kET(Native)[a Ru ] + kML3)}
6
III
constants are defined in Figure 2 and Cytc(Fe -L) is the
alkaline conformer of the A79H73 and A79G52 variants with
the alkaline state heme ligand, L, being His 73 and Lys 73,
2
+
iso-1-Cytc is near 290 mV10a versus NHE, whereas the
reduction potential of the His 73-heme and Lys 73-heme
6
respectively. If kET(Native)[a Ru ] . kML3, then the rate law
2
+
6
in eq 1 becomes independent of [a Ru ] and kobs becomes
11
the rate constant of conversion from the alkaline to the native
conformer of oxidized Cytc (kLM3 in Figure 2, i.e., kLM3 is
the rate constant for ET gating). Thus, when both conforma-
alkaline conformers are near 0 mV and -205 mV versus
10
NHE, respectively, reduction of the A79H73 and A79G52
2
+
29
variants by a
6
Ru (E° ≈ 50 mV versus NHE) is expected
2
+
tions are present, a fast phase dependent on [a
6
Ru ] (direct
to proceed from the native conformer. Thus, in Figure 2,
reduction of alkaline conformers of these iso-1-Cytc variants
should proceed via path B, and the ET reaction should be
gated by the conformational change back to the native state
ET with kET(Native)) and a slower gated ET phase (kLM3 in
2
+
6
Figure 2) independent of [a Ru ] should be observed.
Anaerobic Stopped-Flow Experiments on the A79H73
Iso-1-cytochrome c Variant. To distinguish intermolecular
(k
LM3 rate constant in Figure 2). Under pH conditions where
2
+
ET reactions of the A79H73 variant with a
6
Ru from
both conformers are present in the oxidized state (i.e., 0.1
conformationally gated ET, reduction of the A79H73 variant
2+
(
28) (a) Hoffman, B. M.; Ratner, M. A.; Wallin, S. A. In AdV. Chem. Ser.;
Johnson, M. K., King, R. B., Kurtz, D. M., Jr., Kutal, C., Norton, M.
L., Scott, R. A., Eds.; American Chemical Society, Washington, D.
C., 1990; Vol. 226, pp 125-146. (b) Brunschwig, B. S.; Sutin, N. J.
Am. Chem. Soc. 1989, 111, 7454-7465.
6
was carried out at three different a Ru concentrations. Our
previously reported pH jump stopped-flow studies have
shown that the rate of conversion of the His 73-heme
alkaline conformer to the native Met 80-heme conformer
(
29) Wherland, S.; Gray, H. B. In Biological Aspects of Inorganic
Chemistry; Addison, A. W., Cullen, W. R., Dolphin, D., James, B.
R., Eds.; Wiley: New York, 1977; pp 289-368.
2
3
of the A79H73 variant varies with pH. Therefore, the
+
reduction of A79H73 iso-1-Cytc with a
6
Ru2 was carried
Inorganic Chemistry, Vol. 45, No. 16, 2006 6341

Baddam and Bowler
the His 73-heme alkaline conformer to the native state of
the A79H73 variant which we have measured independently
2
3
by pH jump stopped flow methods.
The amplitude of the slow phase (see Table 1) increases
with pH in synch with the growth of the population of the
His 73-heme alkaline conformer of the A79H73 variant
observed in our equilibrium studies over the pH range 6.0-
2
3
7
.5. In contrast, the amplitude of the fast phase decreases
over this pH range, consistent with the decreasing population
of the native conformer of the A79H73 variant. The match
between the amplitude data and the thermodynamics of the
conformational change from the native to the His 73-heme
alkaline conformer also provide strong support for our
2
+
assignment of the fast [a
6
Ru ]-dependent kinetic phase to
direct bimolecular ET to the native state (kET(Native) in Figure
Figure 3. Plot of absorbance at 550 nm versus time (on a logarithmic
scale) for the reduction of the A79H73 variant of iso-1-Cytc with
hexaammineruthenium(II) chloride at concentrations of, from top to bottom,
2
+
2
) and the slow [a
tionally gated ET (kLM3 in Figure 2).
Plots of the fast phase versus [a Ru ] (Figure 3, inset)
6
Ru ]-independent phase to conforma-
5
, 2.5, and 1.25 mM. Data at 550 nm were collected for 2 s in 0.1 M NaCl
2+
6
with 10 mM MES, pH 6.0, as buffer. The temperature was 25 °C, and the
protein concentration was ∼5 µM. The gray dots are the data at 550 nm,
and the solid black curves are fits of the data to a double exponential rise
to maximum equation. Inset: Plot of kobs(fast) versus [a6Ru2 ] for the
reduction reaction of the A79H73 variant at pH 6.0. The solid curve is a fit
can be used to extract the bimolecular electron-transfer rate
constant for direct ET via path B, kET(Native) (see Figure 2),
+
2+
6
for the reaction of a Ru with the native state of the A79H73
5
-1
2
+
variant. The value of kET(Native) is (2.22 ( 0.05) × 10 M
to kobs(fast) ) kET(Native)[a6Ru ]. The error bars are the standard deviation
-
1
5
-1 -1
of kobs(fast).
s
(
at pH 6.0, (2.36 ( 0.01) × 10 M
s
at pH 6.5, and
5
-1 -1
1.33 ( 0.09) × 10 M
s
at pH 7.5. The significance of
out at three pH values to definitively demonstrate that gated
ET is due to the alkaline conformer of this protein. Figure 3
shows data acquired at pH 6 on a 2 s time scale. A very fast
phase is complete in less than 10 ms, and a slow phase is
observed on a time scale of ∼100 ms. These two kinetic
the decrease in kET(Native) at pH 7.5 may be questionable given
the low amplitude of the fast phase at pH 7.5 and the
proximity of its rate to the dead time of the instrument.
Anaerobic Stopped-Flow Experiments on the A79G52
Variant. Our thermodynamic studies on formation of the
alkaline state of the A79G52 variant indicate that it converts
8
phases are analogous to those observed for the H73 variant.
2+
6
The fast phase is clearly dependent on [a Ru ] (Figure 3,
from the native state to the Lys 73-heme alkaline conformer
inset), as for the H73 variant, and thus can be assigned to
direct intermolecular ET to the native state (heme-Met 80
ligation) of the A79H73 variant. The slower time scale phase
20
with an apparent pK
a
of ∼7.4. Thus, we expect that gated
ET from its Lys 73-heme alkaline conformer should be
observable near neutral pH. Therefore, we have studied
2
+
is independent of [a
6
Ru ] and thus is due to conformation-
2+
6
intermolecular ET of the A79G52 variant with a Ru over
8
ally gated ET. As for the H73 variant, a very slow phase is
observed on a 100 s time scale assignable to proline
the range pH 6-8. Figure 4 shows data obtained at pH 6.5.
A fast ET kinetic phase (Figure 4A), dependent on the
concentration of the reducing agent, was observed from data
collected on a 2 s time scale, consistent with direct
intermolecular ET (kET(Native) in path B of Figure 2) to the
heme of the native state of the A79G52 variant. As with the
A79H73 variant, this direct ET phase is ∼5-fold faster than
3
0
isomerization.
Table 1 summarizes the rate constants for reduction of
the Fe(III) state of the A79H73 variant as a function of pH
2
+
and [a
6
Ru ]. At all pH values, the fast phase is dependent
2
+
2+
on [a Ru ] and the slow phase is independent of [a Ru ].
6
6
2+
Thus, at all pH values, the fast phase can be assigned to
intermolecular ET and the slow phase to conformationally
gated ET. The observed rate constant for the fast phase,
6
observed for the H73 variant at a given [a Ru ]. A slow
6
phase (Figure 4B), independent of the concentration of [a -
2
+
Ru ] is seen in the data collected on a 100 s time scale,
consistent with conformationally gated ET (kLM3 in path B
of the ET square scheme in Figure 2).
2
+
k
obs(fast), is ∼5-fold faster for a given [a
6
Ru ] for the
8
A79H73 variant than for the H73 variant. Thus, direct ET
via path B in the square scheme (see Figure 2) is much faster
for the A79H73 variant than for the H73 variant. On the
other hand, the rate constant for the slower gated ET
Rate constants and amplitudes as a function of pH and
2
+
[a Ru ] are provided in Table 2. Rate constants for the fast
6
phase could not be extracted from the data at pH 7.5 and
2
+
(
k
obs(slow) in Table 1; kLM3 in path B of Figure 2) is 50%
8.0 for a Ru at 2.5 and 5.0 mM concentrations due to the
6
8
larger for the H73 variant than for the A79H73 variant
studied here. The rate constant for the slow gated ET phase
is also dependent on pH, as observed for the H73 variant.
low amplitude of this phase under these conditions. Inspec-
tion of the data in Table 2 suggests that there may be a slight
8
2+
dependence of kobs for the slow phase on [a Ru ] at pH 6.5
6
This observation is expected on the basis of the pH
dependence of the rate of the conformational change from
and 7.5; however, at pH 6.0 and 8.0 no such dependence is
observed. In eqs 1 and 2, if direct ET and gated ET phases
2+
are not well-separated, the assumption that kET(Native)[a
6
Ru ]
(30) Baddam, S. M.S. Thesis, University of Denver, 2005.
. kML3 will not be completely valid and a slight concentra-
6342 Inorganic Chemistry, Vol. 45, No. 16, 2006

Metal-Ligand Effects on Gated Electron Transfer
Table 1. Rate Constants and Amplitudes for the Reduction of A79H73 iso-1-Cytc by a6Ru2+ at 25 °C as a Function of pH and [a6Ru2+
]
pH
[a6Ru2+], mM
kobs(fast), s-1
amplitude (fast), au
kobs(slow), s-1
amplitude (slow), au
6.0
6.5
7.5
1.25
343 ( 12
638 ( 26
1179 ( 64
274 ( 8
571 ( 10
1158 ( 88
255 ( 1
455 ( 11
760 ( 92
0.065 ( 0.002
0.065 ( 0.005
0.066 ( 0.001
0.053 ( 0.002
0.056 ( 0.003
0.0586 ( 0.0003
0.0308 ( 0.0004
0.033 ( 0.001
0.022 ( 0.006
8.8 ( 0.2
8.8 ( 0.4
9.0 ( 0.1
6.3 ( 0.2
6.3 ( 0.2
6.6 ( 0.2
5.0 ( 0.1
5.1 ( 0.1
5.18 ( 0.04
0.026 ( 0.001
0.025 ( 0.002
0.026 ( 0.001
0.035 ( 0.001
0.033 ( 0.001
0.0352 ( 0.0002
0.0451 ( 0.0004
0.044 ( 0.001
0.0434 ( 0.0002
2
5
.5
.0
1.25
2
5
.5
.0
1.25
2
5
.5
.0
2
+
tion dependence will be observed for the gated ET phase.
However, the data in Figure 4 clearly demonstrate that the
fast and slow ET phases are well separated for the A79G52
variant. The standard deviations in Table 2 at each pH and
the different pH/[a
6
Ru ] experiments. Therefore, the overall
average and standard deviation for kobs for the slow phase at
different pH/[a Ru ] values (0.06 ( 0.01 s ) provide a
2
+
-1
6
better estimate of the value of and the precision with which
we know the ET gating rate constant for this variant.
The amplitude of the fast phase (Table 2) decreases as
pH increases, consistent with the progressive decrease in the
population of the native state as pH increases from 6 to 8,
as expected on the basis of our thermodynamic studies on
2
+
[
a
6
Ru ] are based on multiple stopped-flow experiments
acquired immediately sequentially. Thus, the precision for
2
+
6
each pH/[a Ru ] experiment for the relatively long data
acquisitions required for the slow phase of the A79G52
variant might be expected to be higher than that between
2
0
the A79G52 variant. Similarly, the slow gated ET phase
increases in amplitude over the same pH range, as the Lys
7
3-heme alkaline conformer becomes the dominate form
of the A79G52 variant. As for the ET data for the A79H73
variant, the pH dependence of the amplitudes for the
2+
6
reduction of the A79G52 variant of iso-1-Cytc by a Ru
confirms the assignment of the fast phase to direct bimo-
lecular ET to the native state of the protein and the slow
phase to gated ET involving the Lys 73-heme alkaline
conformer.
2+
Plots of the fast phase versus [a
6
Ru ] (see Figure 4, inset)
can be used to extract the bimolecular ET rate constant,
2+
k
6
ET(Native), for the direct ET reaction of a Ru with the native
state of the A79G52 variant. The value of kET(Native) is (1.6
5
-1 -1
5
-1
(
s
0.4) × 10 M
s
at pH 6.0 and (2.2 ( 0.1) × 10 M
-
1
at pH 6.5.
Discussion
Comparison of Conformationally Gated ET with the
Kinetics of the Conversion of the Alkaline Conformer to
the Native Conformer. In studies on the H73 variant of
8
iso-1-Cytc, the pH dependence of the rate constant for gated
ET paralleled that for the formation of the native state from
the His 73-heme alkaline conformer measured indepen-
dently. On this basis, it was possible to directly assign the
conformational change controlling the rate of ET to the
conversion of the His 73-heme alkaline conformer to the
native state (kLM3 in Figure 2). We have also determined kLM3
independently by pH jump stopped-flow mixing for the
conversion of the His 73-heme alkaline conformer of the
Figure 4. Plot of absorbance at 550 nm versus time (on a logarithmic
scale) for reduction of the A79G52 variant of iso-1-Cytc with hexaam-
mineruthenium(II) chloride at pH 6.5. Panel A shows data collected over 2
s with hexaammineruthenium(II) chloride concentration at, from top to
bottom, 5, 2.5, and 1.25 mM. The inset to panel A shows a plot of kobs(fast)
23
A79H73 variant to the native state. Figure 5 shows a plot
of kLM3 versus pH obtained by pH jump stopped-flow
2
3
methods, along with the average values for gated ET for
the A79H73 variant at pH 6.0, 6.5, and 7.5 (kobs(slow) in
Table 1). The values for kobs(slow) increase as pH decreases,
correlating well with the pH dependence for kLM3 measured
independently. Thus, the His 73-heme alkaline conformer
of the A79H73 variant is clearly responsible for ET gating.
In the case of the A79G52 variant, within error, the rate
2
+
versus [a6Ru ] for the reduction reaction with the A79G52 variant at pH
2
+
6
.5. The solid curve is a fit to kobs(fast) ) kET(Native)[a6Ru ]. The error
bars are the standard deviation of kobs(fast). Panel B shows data collected
over 100 s with hexaammineruthenium(II) chloride concentration at 2.5
mM. Data were collected in 0.1 M NaCl with 10 mM MES, pH 6.5, as
buffer. The temperature was 25 °C, and the protein concentration was ∼5
µM. The gray dots are the data at 550 nm, and the solid black curves are
fits of the data to a single-exponential rise to maximum equation.
Inorganic Chemistry, Vol. 45, No. 16, 2006 6343

Baddam and Bowler
Table 2. Rate Constants and Amplitudes for the Reduction of A79G52 iso-1-Cytc by a6Ru2+ at 25 °C as a Function of pH and [a6Ru2+
]
pH
[a6Ru2+], mM
kobs(fast), s-1
amplitude (fast), au
kobs(slow), s-1
amplitude (slow), au
6.0
6.5
7.5
8.0
1.25
231 ( 4
563 ( 9
844 ( 27
276 ( 11
602 ( 37
1127 ( 51
142 ( 3
-
0.101 ( 0.001
0.118 ( 0.009
0.11 ( 0.01
0.081 ( 0.004
0.10 ( 0.01
0.085 ( 0.013
0.032 ( 0.001
-
0.063 ( 0.002
0.087 ( 0.005
0.063 ( 0.005
0.0549 ( 0.0004
0.059( 0.002
0.065 ( 0.002
0.0481 ( 0.0004
0.056 ( 0.001
0.065 ( 0.001
0.0518 ( 0.0004
0.059 ( 0.001
0.0583 ( 0.0003
0.0075 ( 0.0003
0.0074 ( 0.0001
0.0080 ( 0.0001
0.0162 ( 0.0003
0.0162 ( 0.0002
0.0159 ( 0.0001
0.0607 ( 0.0005
0.0613 ( 0.0004
0.0639 ( 0.0001
0.089 ( 0.001
2
5
.5
.0
1.25
2
5
.5
.0
1.25
2
5
.5
.0
-
-
1.25
177 ( 11
-
0.015 ( 0.001
-
-
2
5
.5
.0
0.0895 ( 0.0001
0.091 ( 0.001
-
of gated ET is invariant with pH. The average and standard
deviation of kobs(slow) in Table 2 for all pH values and
changed by nearly 3 orders of magnitude by changing the
ligand involved in the alkaline state of iso-1-Cytc from Lys
to His. While large changes are useful, the ability to fine-
tune the rate of a conformational ET gate over small ranges
is equally important. Over the pH range 6.0-7.5, the rate
constant for the ET gate of the H73 variant varied from 7 to
2
+
-1
concentrations of a
LM3 for the conversion of the Lys 73-heme alkaline
conformer of the A79G52 variant to the native state obtained
6
Ru is 0.06 ( 0.01 s . The value of
k
-
1
by pH jump stopped-flow measurements is 0.05 ( 0.01 s
and is also independent of pH.20 The good agreement
between these independently determined values demonstrates
that the Lys 73-heme alkaline conformer of the A79G52
variant is responsible for the gated ET observed during its
-1 8
13 s . Over the same pH range, the addition of the Lys 79
f Ala mutation to produce the A79H73 variant shifts the
-1
variation in the rate constant of the ET gate to 5-9 s . This
allows the lifetime of the ET gate to be varied from 75 to
200 ms in discrete steps (H73: pH 6, 75 ms; pH 6.5, 122
ms; pH 7.5, 145 ms. A79H73: pH 6, 113 ms; pH 6.5, 156
ms, pH 7.5, 195 ms.) by varying both pH and the amino
acid sequence of the portion of the protein involved in the
conformational change. Thus, the strategy of using surface
mutations to make modest perturbations to the thermody-
namics and kinetics of a conformational change provides a
facile means of fine-tuning the flow of an electron through
a metalloprotein.
2+
6
reduction by a Ru .
There are few cases where conformationally gated ET can
be directly correlated with the rate of a conformational
change, and these cases are primarily for small molecule
7
systems. Gated ET in proteins is typically assigned on the
basis of methods such as the viscosity dependence of the
3
ET rate constant which provide little insight into the nature
of the conformational change gating the ET reaction. The
ability to directly correlate gated ET with a particular
conformational change, as in our model system, provides
an entr e´ e to rationale manipulation of gated ET.
Fine-Tuning the Rate of an ET Gate. In previous work,8
we have demonstrated that the rate of an ET gate can be
Tuning the pH Accessibility of an ET Gate. As noted
in the Introduction, the naturally occurring alkaline conformer
of Cytc which is stabilized by lysine-heme ligation generally
populates at pH values between 8.5 and 11.14 For the
purposes of engineering protein function related to confor-
mational ET gates in proteins or for developing protein-based
6
molecular electronics, ET gates that operate near neutral
pH are preferable. For Cytc from horse heart, it has been
demonstrated that the alkaline conformer can act as an ET
9
gate, albeit at high pH. In previous work, ET gating
mediated by Lys-heme alkaline conformers is only margin-
8
ally observable for WT iso-1-Cytc at pH 7.5. Here, we
demonstrate that the slow ET gate due to Lys 73-heme
ligation can be made accessible at neutral pH and below by
using the known substructure hierarchy of Cytc to rationally
stabilize the Lys 73-heme alkaline conformer of iso-1-
1
8d,20
Cytc.
Our data on the A79G52 variant (Figure 4, Table
2
6
), show that this ET gate is observable over the pH range
.0-8.0, becoming the dominant mode of electron flow
Figure 5. Comparison of kobs(slow) for the reduction of A79H73 iso-1-
Cytc by a6Ru to the pH dependence of kLM3 for the conformational change
through this protein at pH 7.5 and 8.0. Therefore, this much
slower ET gate, which has a lifetime of 15-20 s can now
be used for ET gating near neutral pH. As discussed in the
Introduction, the large effect on the stability of the Lys 73-
heme alkaline conformer likely results in part from the
decrease in the size of the side chain when Asn is mutated
to Gly at position 52 which is buried in the native state of
2+
III
III
from the Cytc(Fe -His 73) alkaline conformer to the Cytc(Fe -Met 80)
native state. The solid circles and error bars are the average and standard
deviation of kobs(slow) for all [a6Ru ] at each pH. The solid line is the pH
dependence for kLM3 measured directly via pH jump stopped-flow methods
in 100 mM NaCl and 10 mM buffer. The dashed lines show the average
standard deviation of the data points of the pH jump stopped-flow data
used to define the dependence of kLM3 on pH.
2
+
2
3
6344 Inorganic Chemistry, Vol. 45, No. 16, 2006

Metal-Ligand Effects on Gated Electron Transfer
the protein.19 We have also demonstrated that the effects of
mutations at position 52 on the buried hydrogen bond
network of iso-1-Cytc can have substantial effects on the
cooperative stabilization of this protein.3 Thus, the desta-
bilization due to the Gly 52 mutation is also likely due in
part to its impact on the buried hydrogen-bond network of
this protein.
and small-molecule reagents.33,36 Changes in the binding site
or sites caused by this mutation might be expected to affect
both K and ket (since perturbations to the binding site will
likely affect the ET distance). For inorganic reagents, three
sites have been defined around the heme edge where these
1
33a,36d
reagents bind, prior to ET reactions with Cytc.
For
(phen )
3
+/2+
positively charged reagents such as Co(phen)
3
1
1
,10-phenanothroline), Site II which includes Val 11, Ala
5, Thr 19, and lysines 7, 25, and 27 in horse heart Cytc
Several variants at position 82 of iso-1-Cytc have previ-
ously been reported which shift the alkaline conformational
transition to near neutral pH,32 although no studies on ET
has been implicated as the primary binding site in intermo-
lecular ET reactions with Cytc.
3
6b,c,d
The presence of Glu
gating have been carried out on these variants. Position 82
21 adjacent to this site is believed to be important for binding
is buried in the three-dimensional structure of iso-1-Cytc and
3
6d
12
positively charged metal complexes. A prominent en-
hancement of the oxidation rate of Cytc by Co(phen)
packs against the heme. Thus, mutations at buried sequence
positions in the two least stable substructures of Cytc (shown
in gray and red in Figure 1) both appear to be appropriate
for manipulating gated ET in iso-1-Cytc as predicted by the
hierarchical thermodynamics of Cytc (see Figure 1, caption).
3+
3
upon
chemical modification of Lys 27 with reagents that change
its charge is a primary reason for this assignment.3 Site II
is significantly modified in iso-1-Cytc relative to the horse
heart protein (Val 11 f Lys, Ala 15 f Leu, Lys 7 f Thr,
and Lys 25 f Pro), so changes in the ET binding site for
positively charged metal complexes might be expected. The
net effect of the mutations relative to the horse protein is to
reduce the positive charge near this site, and both Lys 27
and Glu 21 are conserved in yeast iso-1-Cytc. Thus, in
principle, one might expect an increase in kET(Native) relative
to the horse protein. However, for WT iso-1-Cytc, we did
not observe such an increase in kET(Native) relative to the horse
6b,c
Modulation of the Rate of Direct ET by the Lys 79 f
Ala Mutation. In our studies on the reduction of WT and
2+ 8
H73 iso-1-Cytc by a
ET rate constants, kET(Native), were (4.8 ( 0.4) × 10 and (3.9
0.5) × 10 M s , respectively. Both the A79H73 and
A79G52 variants have substantially higher kET(Native) values
6
Ru , we found that the bimolecular
4
4
-1 -1
(
5
-1 -1
of (1.9 ( 0.5) × 10 M s . Thus, kET(Native) has increased
by a factor of about 3-5. For bimolecular ET between small-
molecule reagents and proteins, a minimal mechanism
8
protein in our previous work.
involving a binding equilibrium (K ) k
-1 are the association and dissociation rate constants,
respectively) followed by electron transfer (ket) is generally
1 1
/k-1, where k and
Lys 79 is not implicated in any of Sites I-III;33a,36d
however, studies on ferricyanide oxidation of reduced horse
Cytc show that blocking the amino groups of either Lys 79
or Lys 72 (Site III) decreases the ET rate by similar
k
3
3
invoked. For c-type cytochromes, the limiting condition
3b
k
et , k-1 appears to apply.3 Therefore, kET(Native) is best
3
6a
amounts.
Thus, Lys 79 clearly affects ET of small-
expressed as the product of the binding equilibrium constant,
molecule reagents around the heme edge. It is also note-
worthy that the differences in binding site preferences are
small. For Co(phen)
K, and a microscopic ET rate constant, i.e., kET(Native) ) Kket
.
Thus, a number of factors may be responsible for the increase
in kET(Native) caused by the Lys 79 f Ala mutation.
3
+
3
, the ET rate enhancement due to a
charge change blocking group at Lys 27 (Site II) is ∼7
Since ket will depend on the driving force, it is possible
that a change in the reduction potential of iso-1-Cytc caused
by the Lys 79 f Ala mutation is a factor. Electrochemical
studies on a Lys 79 f Ala variant of iso-1-Cytc demonstrate
that the reduction potential does not change within error
relative to wild-type iso-1-Cytc.34 Thus, driving force does
not appear to be a factor.
whereas at Lys 72 (Site III) it is ∼3 at an ionic strength of
36c
0
.1. Since Lys 79 forms a salt bridge to heme propionate
12
D in iso-1-Cytc, replacement of Lys 79 with Ala would
uncover this negatively charged group. Thus, this mutation
might be expected to have similar effects on intermolecular
ET, as observed for the charge change blocking groups in
3+
36b,c
the Co(phen)
3
studies.
The 3-5-fold enhancement we
Both variants have in common the mutation of Lys 79 f
Ala, which is located at the exposed heme edge of iso-1-
Cytc.12 A number of studies have implicated the exposed
see in kET(Native) for the A79H73 and A79G52 variants relative
to wild-type iso-1-Cytc for reduction with a Ru near an
ionic strength of 0.1 is certainly in line with the effects of
blocking Lys 27 or Lys 72 with charge-change reagents on
2+
6
35
heme edge of Cytc as the site of reaction with both protein
3
+
Co(phen)
3
oxidation of horse Cytc at the same ionic
2
+
(
31) (a) Wandschneider, E.; Hammack, B. N.; Bowler, B. E. Biochemistry
strength. Thus, a Ru clearly reacts at the heme edge in a
6
2
003, 42, 10659-10666. (b) Redzic, J. S.; Bowler, B. E. Biochemistry
2005, 44, 2900-2908.
(
32) Pearce, L. L.; G a¨ rtner, A. L.; Smith, M.; Mauk, A. G. Biochemistry
(36) (a) Ahmed, A. J.; Millet, F. J. Biol. Chem. 1981, 256, 1611-1615.
(b) Butler, J.; Davies, D. M.; Sykes, A. G.; Koppenol, W. H.;
Osherhoff, N.; Margoliash, E. J. Am. Chem. Soc. 1981, 103, 469-
471. (c) Butler, J.; Chapman, S. K.; Davies, D. M.; Sykes, A. G.;
Speck, S. H.; Osherhoff, N.; Margoliash, E. J. Biol. Chem. 1983, 258,
6400-6404. (d) Drake, P. L.; Hartshorn, R. T.; McGinnis, J.; Sykes,
A. G. Inorg. Chem. 1989, 28, 1361-1366. (e) Speh, S.; Elias, H. J.
Biol. Chem. 1994, 269, 6370-6375. (f) Cheddar, G.; Meyer, T. E.;
Cusanovich, M. A.; Stout, C. D.; Tollin, G. Biochemistry 1989, 28,
6318-6322.
1
989, 28, 3152-3156.
33) (a) Bowler, B. E.; Raphael, A. L.; Gray, H. B. Prog. Inorg. Chem.
990, 38, 259-322. (b) Cusanovich, M. A.; Tollin, G. In Cytochrome
(
1
c: A Multidiscplinary Approach; Scott, R. A., Mauk, A. G., Eds.;
University Science Books: Sausalito, CA, 1996; pp 489-513.
34) Ferrer, J. C.; Guillemette, J. G.; Bogumil, R.; Inglis, S. C.; Smith,
M.; Mauk, A. G. J. Am. Chem. Soc. 1993, 115, 7507-7508.
(
(
35) Margoliash, E.; Bosshard, H. R. Trends Biochem. Sci. 1983, 8, 316-
320.
Inorganic Chemistry, Vol. 45, No. 16, 2006 6345

Baddam and Bowler
manner analogous to Co(phen) 3
+. The preference among
These strategies permit key properties of ET gates to be
manipulated to modulate protein function and to aid in the
development of tailored components for protein-based mo-
lecular electronics devices.
3
Sites I, II, and III is unclear from the current work.
Conclusions. We have demonstrated that the rate of an
ET gate due to a His 73-heme alkaline conformer of iso-
1-Cytc can be fined tuned over the range 75-200 ms by
varying pH and introducing mutations at surface-exposed
sites in the substructure (red in Figure 1) of the protein
involved in this conformational change. By exploiting the
known hierarchical conformational thermodynamics of Cytc,
we have also designed a variant of iso-1-Cytc that makes
the slower Lys 73-heme ET gate accessible at neutral pH.
Acknowledgment. This work was supported by NSF
Grant No. CHE 0316378 (B.E.B.). The Applied Photophysics
π*-180 spectrometer was purchased with NIH Grant No. 1
S10 RR16632-01.
IC0603712
6346 Inorganic Chemistry, Vol. 45, No. 16, 2006