Báo cáo khoa học: Atomic-resolution structure of reduced cyanobacterial cytochrome c6 with an unusual sequence insertion pot

Atomic-resolution structure of reduced cyanobacterial
cytochrome c
6
with an unusual sequence insertion
Wojciech Bialek
1
, Szymon Krzywda
2
, Mariusz Jaskolski
2,3
and Andrzej Szczepaniak
1
1 Department of Biophysics, Faculty of Biotechnology, University of Wroclaw, Poland
2 Department of Crystallography, Faculty of Chemistry, A. Mickiewicz University, Poznan, Poland
3 Center for Biocrystallographic Research, Institute of Bioorganic Chemistry, Polish Academy of Sciences, Poznan, Poland
Introduction
During photosynthesis, electron transfer between two
membrane-bound complexes, cytochrome b
6
f and
photosystem I, can be accomplished by the copper-
containing protein plastocyanin (PC) or the heme
protein cytochrome c
6
[1,2]. PC is found as a unique
electron carrier in higher plants. Some algae and
cyanobacteria express either PC or cytochrome c
6
,
whereas others are able to produce both proteins,
depending on copper availability [3]. Cytochromes c

6
are water-soluble, low-spin, heme-containing proteins
Keywords
cyanobacteria; cytochrome; high-resolution
structure; photosynthesis; Synechococcus
sp. PCC 7002
Correspondence
A. Szczepaniak, Department of Biophysics,
Faculty of Biotechnology, University of
Wroclaw, Przybyszewskiego 63 ⁄ 77, 51-148
Wroclaw, Poland
Fax: +48 71 3756234
Tel: +48 71 3756236
E-mail:
M. Jaskolski, Department of
Crystallography, Faculty of Chemistry, A.
Mickiewicz University, Grunwaldzka 6,
60-780 Poznan, Poland
Fax: +48 61 829 1505
Tel: +48 61 829 1274
E-mail:
Database
Atomic coordinates and structure factors are
available from the Protein Data Bank under
the accession code 3DR0
(Received 5 May 2009, revised 9 June
2009, accepted 11 June 2009)
doi:10.1111/j.1742-4658.2009.07150.x
The structure of the reduced form of cytochrome c
6

from the mesophilic
cyanobacterium Synechococcus sp. PCC 7002 has been determined at 1.2 A
˚
and refined to an R-factor of 0.107. This protein is unique among all
known cytochromes c
6
, owing to the presence of an unusual seven-residue
insertion, KDGSKSL(44–50), which differs from the insertion found in the
recently discovered plant cytochromes c
6A
. Furthermore, the present pro-
tein is unusual because of its very high content (36%) of the smallest resi-
dues (glycine and alanine). The structure reveals that the overall fold of the
protein is similar to that of other class I c-type cytochromes, despite the
presence of the specific insertion. The insertion is located within the most
variable region of the cytochrome c
6
sequence, i.e. between helices II and
III. The first six residues [KDGSKS(44–49)] form a loop, whereas the last
residue, Leu50, extends the N-terminal beginning of helix III. Several spe-
cific noncovalent interactions are found inside the insertion, as well as
between the insertion and the rest of the protein. The crystal structure con-
tains three copies of the cytochrome c
6
molecule per asymmetric unit, and
is characterized by an unusually high packing density, with solvent occupy-
ing barely 17.58% of the crystal volume.
Abbreviation
PC, plastocyanin.
4426 FEBS Journal 276 (2009) 4426–4436 ª 2009 The Authors Journal compilation ª 2009 FEBS

involved in the high-potential electron transport
chain. They are characterized by low molecular mass
(80–90 amino acid residues in the mature protein), and
have a covalently bound heme group. Once believed
to be absent in plants, cytochrome c
6
-like proteins,
now designated cytochrome c
6A
, have recently also
been discovered in plants [4,5] and in the green alga
Chlamydomonas reinhardtii [6].
In general, cytochromes c
6
are characterized by high
redox potential (E
m,7
of 330–360 mV). However, a
number of low redox potential cytochrome c
6
-like pro-
teins have been identified. These include plant and algal
cytochrome c
6A
(E
m,7
of $ 70 mV [7]), and cyanobacte-
rial cytochrome c
6C
(E

m,7
of $ 150 mV [8]) and cyto-
chrome c
6B
, for which a low redox potential has been
postulated as well [8]. Members of these recently identi-
fied cytochrome c
6
subfamilies are characterized by the
presence of Val, Ile or Leu in place of a conserved Gln
residue located inside the cytochrome c
6
heme pocket.
Because of its low redox potential and the presence of
two additional Cys residues, cytochrome c
6A
has been
postulated to play a role in the formation of disulfide
bridges in thylakoid lumen proteins, rather than func-
tioning in the photosynthetic electron transport chain
[9]. Interestingly, the replacement of the conserved Gln
residue by Val resulted in a decrease of the cyto-
chrome c
6
redox potential by $ 100 mV [cytochrome c
6
from Phormidium laminosum [7] and Synechococcus
sp. PCC 7002 (Bialek, unpublished results)], whereas
the opposite Val fi Gln substitution in cytochrome c
6A

from Arabidopsis thaliana increased the midpoint
potential by approximately 100 mV [7].
The structures of cytochromes c
6
from three cyano-
bacteria (Arthrospira maxima [10], Synechococ-
cus elongatus [11], and P. laminosum [7]), one from a
red alga (Porphyra yezoensis [12]), one from a brown
alga (Hizikia fusiformis [13]), and four from green
algae (Monoraphidium brauni [14], C. reinhardtii [15],
Scenedesmus obliquus [16], and Cladophora glomerata
[17]) have been determined, as has the structure of
cytochrome c
6A
from A. thaliana [18]. Together, these
data represent a large volume of information about
the sequences and structures of cytochrome c
6
. How-
ever, the cytochrome c
6
molecule from the mesophilic
cyanobacterium Synechococcus sp. PCC 7002 (E
m,7
of
$ 320 mV [8]) is an exceptional protein in this class, as
it contains an unusual heptapeptide insertion beginning
after Tyr43. Moreover, the petJ1 gene, which encodes
cytochrome c
6

in Synechococcus sp. PCC 7002, could
not be inactivated or replaced with petJ (cyto-
chrome c
6
), petE (PC) or cytM (cytochrome c
M
) from
Synechocystis [19], unlike the petJ gene from Synecho-
cystis [20]. Consequently, it appears that only the
native cytochrome c
6
of Synechococcus can function
properly in electron transport and ensure cell viability.
To elucidate the characteristics of this unusual cyto-
chrome c
6
molecule, and in particular to shed light on
the role of the heptapeptide insertion, the crystal struc-
ture of the protein in its reduced form has been deter-
mined at a resolution of 1.2 A
˚
. In addition, this study
is a contribution towards a better understanding of the
molecular determinants of the spectroscopic and
electron transfer properties of this class of proteins.
Results and Discussion
Structure description
Overall, the structure of Synechococcus sp. PCC 7002
cytochrome c
6

reveals the characteristic properties of
other class I cytochromes c. The molecule is composed
of a single polypeptide chain wrapped around the
heme group, which is bound to the protein by two
thioether links at Cys14 and Cys17, as well as by
His18 and Met65, which are the axial ligands in the
coordination sphere of the iron atom (Fig. 1). The
polypeptide chain consists of four a-helices (I–IV) con-
nected by loops. The long N-terminal helix I (Ala3–
His18) has a characteristic kink at Cys14. The two Cys
residues forming the covalent links to the heme group
and the fifth iron-coordinating ligand, His18, are part
of this helix. As in other cytochrome c
6
molecules, the
three consecutive amino acids Ala19-Gly-Gly21 form a
3
10
helix followed by an X-loop ranging from Asn22
to Lys32. The X-loop separates helix I from the short
helix II (Ala33–Tyr39). Residues Leu40–Tyr43 and
Gly46–Ser49 form type IV and type VIII b-turns,
respectively [21]. The latter turn is formed entirely
within the sequence of the heptapeptide insert that is
characteristic of this cytochrome c
6
protein only.
Leu50, which is also part of the insert, belongs to
helix III, formed by Leu50–Asn60. Gln62–Met65 and
Gly69–Leu72 in the following region are type II¢ and

type IV b-turns, respectively [21]. Helix IV, which is
nearly perpendicular to helix I, runs from Asp74 to
Glu89. Two residues, Asp86 and Lys92, located near
the C-terminus, form a salt bridge stabilizing this part
of the protein fold.
Conformation of the heptapeptide insertion
In general, high overall structural similarity is observed
among all the compared cytochrome c
6
molecules
(Fig. 2A, Table S1). The most striking structural
difference between the present Synechococcus
sp. PCC 7002 cytochrome c
6
and all other cyto-
W. Bialek et al. Atomic-resolution structure of cytochrome c
6
FEBS Journal 276 (2009) 4426–4436 ª 2009 The Authors Journal compilation ª 2009 FEBS 4427
chrome c
6
molecules is the presence of the specific
insertion between helices II and III (Fig. 2C). The
insertion consists of seven, mostly polar, amino acids
KDGSKSL(44–50). Its major part [GSKS(46–49)]
forms a type VIII b-turn. Multiple hydrogen bonds
formed within this structure stabilize its conformation.
These, as well as hydrogen bonds between the insert
residues and residues from other regions of the pro-
tein, are listed in Table 1. Additionally, the cationic
side chain of Lys48 is also involved in cation–p inter-

actions with Tyr56, which, in turn, is adjacent to
Gln57, the residue that is implicated in the control of
the heme redox potential. Specifically, a structural
water molecule inside the heme pocket forms a hydro-
gen bond with the Gln57 side chain, and is thus
responsible for positioning of the c-amide towards the
heme group and within a van der Waals distance of a
methine carbon atom (CHA) of the porphyrin ring. In
consequence, Gln57 is capable of tuning the redox
potential of the heme group [7,8].
An insert of three consecutive Asp residues has been
found in the same region in cytochrome c
6
from
A. maxima [10], but its length, shape and chemical
character are significantly different from those charac-
terizing the present insertion of cytochrome c
6
from
Synechococcus sp. PCC 7002 (Fig. 2A,C). It is of note
that an insertion has also been found in the sequence
of the protein from A. thaliana. However, the insertion
in that cytochrome c
6A
protein is not only longer but
also located differently, namely between helices III and
IV, and is thus entirely different from the insertion dis-
covered in the present protein. Importantly, the cyto-
chrome c
6A

loop contains two Cys residues that are
also found in other cytochrome c
6A
molecules. In
neither case has the function of the insertions been
completely elucidated.
The heme prosthetic group
Figure 3 depicts the hydrophobic heme pocket of the
present cytochrome c
6
molecule. By analogy with other
cytochrome c prosthetic heme groups, the porphyrin
moiety is slightly distorted into a saddle-shaped form
in all three molecules in the asymmetric unit. The edge
of the pyrrole ring C and the propionic group D are
exposed towards the solvent. The heme group is cova-
lently linked to the polypeptide chain through thioe-
ther bonds from the Cys14 and Cys17 sulfhydryl
groups to the two vinyl side chains. The iron of the
heme group is in the FeII oxidation state, maintained
by the presence of 20 mm dithionate in the mother
liquor. The iron ion is coordinated through His18 N
e2
and Met65 S
d
, located at axial positions. The latter
residue has a gauche conformation along the C
b
–C
c

bond (torsion angle C
a
–C
b
–C
c
–S
d
of 50.0°, 48.2° and
50.4° in molecules A, B and C, respectively). As in
other cytochrome c
6
structures, a hydrogen bond
between the N
d1
H donor of the axial His18 and the
carbonyl oxygen atom of Asn22 serves to maintain the
required orientation of the His ring with respect to the
heme plane. Interestingly, this is the only His present
in the protein. The interactions of the nitrogen atoms
of the imidazole ring (N
e2
–Fe and N
d1
–HÆO) confirm
that it is electrically neutral.
The structure of reduced as well as oxidized cyto-
chrome c
6
from S. obliquus has been described previ-

ously [16]. Comparison of both redox states revealed a
different conformation of Pro62, which is adjacent to
the Met61 axial ligand of the heme group. The posi-
tion of C
c
of this Pro was found to change from exo
to endo upon oxidation. The protein used in the pres-
ent study was reduced with a 20-fold molar excess of
dithionite in the crystallization buffers. Pro66, which is
A
B
Fig. 1. (A) Three-dimensional structure of the reduced cyto-
chrome c
6
from Synechococcus sp. PCC 7002. The 3
10
helix (yel-
low) is followed by the X-loop (red). The specific insertion is shown
in magenta. The N-terminus and C-terminus, as well as the two
Cys residues with covalent links to the heme group, are indicated.
(B) Electron density omit map contoured at 3.2r, showing the
heme group. All structural figures were prepared in
PYMOL [39].
Atomic-resolution structure of cytochrome c
6
W. Bialek et al.
4428 FEBS Journal 276 (2009) 4426–4436 ª 2009 The Authors Journal compilation ª 2009 FEBS
equivalent to Pro62 of the S. obliquus protein, has the
exo pucker, confirming the reduced state of the mole-
cule (Fig. S1). Although the conformation of this Pro

may serve as an indicator of the protein redox state,
the mechanism of this change is currently unknown.
Its elucidation will require a comparison of the present
structure with a model of the oxidized molecule deter-
mined at comparably high resolution.
In total, four water molecules are present around
the two propionate substituents (at positions 6 and 7)
of each heme group (Fig. 3). Two of them (water
20 ⁄ water 33, water 18 ⁄ water 45 and water 24 ⁄ water 30
in molecules A, B, C, respectively) are located between
K44
L50
G46
A
C
B
Fig. 2. (A) Structural alignment of 10 cytochrome c
6
and cytochrome c
6A
molecules: Synechococcus sp. PCC 7002 (green, this work,
3DR0); P. yezoensis (light pink, 1GDV); H. fusiformis (black, 2ZBO); M. braunii (blue, 1CTJ); S. obliquus (yellow, 1C6O); C. glomerata
(magenta, 1LS9); C. reinhardtii (cyan, 1CYI); A. maxima (red, 1F1F); P. laminosum (gray, 2V08); and A. thaliana (orange, 2DGE). The char-
acteristic insertions present in the proteins from Synechococcus sp. PCC 7002 and A. maxima (bottom left) and A. thaliana (top middle)
are shown. (B) Side chains of the KDGSKSL(44–50) insertion shown in ball-and-stick representation. (C) Alignment of cytochrome c
6
and
cytochrome c
6A
sequences prepared in ESPRIPT [40]. White letters on a red background represent strictly conserved residues; similar resi-

dues are shown as purple letters on a white background in boxes. Insertions of Synechococcus sp. PCC 7002, A. maxima and A. thaliana
are shown in green, red and orange, respectively. Above the sequences, the secondary structure elements of the present cytochrome c
6
molecule are shown.
Table 1. Hydrogen bonding of amino acids of the insertion (bold),
as calculated by
WHATIF [38]. Distances are given in A
˚
.
Donor Acceptor Molecule A Molecule B Molecule C
Lys44 N Leu40 O 2.84 2.89 2.93
Gly46 N Tyr43 O 3.11 3.51 3.21
Ser47 N Asp45 O
d1
3.15 3.25 3.14
Ser47 N Gly42 O 2.97 3.04 2.97
Ser47 O
c
Gly42 O 3.18 3.41 3.27
Ser47 O
c
Asp45 O
d1
2.66 2.70 2.60
Lys48 N Gly42 O 3.08 3.08 3.16
Glu52 N Ser49 O
c
3.16 3.07 3.17
Ala53 N Ser49 O 2.93 2.88 2.93
Val54 N Leu50 O 2.86 2.86 2.86

W. Bialek et al. Atomic-resolution structure of cytochrome c
6
FEBS Journal 276 (2009) 4426–4436 ª 2009 The Authors Journal compilation ª 2009 FEBS 4429
the propionate groups and create a network of hydro-
gen bonds involving the propionates. Propionate-6 is
also hydrogen-bonded with Lys29 and Gln62, whereas
propionate-7 has a hydrogen bond contact with Thr30.
Additionally, a hydrogen bond between water 20 and
the Gln57 side chain is observed. Table 2 presents a
complete list of hydrogen bonds within the three heme
pockets.
Electrostatic surface potential
The low isoelectric point of the molecule, calculated as
pI 3.8 [8], results from the ratio of acidic (nine) to
basic (seven) amino acids (Fig. 4). The only Arg,
Arg71, is present in all cytochrome c
6
molecules,
except for the protein from S. obliquus. Arg71 creates
a minor basic patch on the otherwise highly negative
protein surface. The basic centers are predominantly
Lys residues. Interestingly, two of the six Lys residues
in the protein sequence of Synechococcus sp. PCC 7002
cytochrome c
6
, Lys44 and Lys48, are located within
the unique insertion. Their long side chains point away
from a bulge with negative surface potential created by
the atoms between them (Fig. 4). All of the negative
charge centers in this region (Lys44 O, Asp45 O, O

d1
and O
d2
, Gly46 O, Ser47 O, and O
c
) point towards the
surface of the protein and contribute to the overall
negative surface potential. The major negative patch,
however, is created mainly by the side chains of
Glu51, Asp86, and Glu89. The surface potential of the
molecule shows an additional slightly positive region,
near the heme crevice, created by Lys29 and Lys44.
Fig. 3. The heme pocket of superposed molecules A (red), B
(green), and C (blue). Hydrogen bonds involving the depicted resi-
dues are listed in Table 2. Structural water molecules associated
with protein molecules A, B and C are shown in the corresponding
colors.
Table 2. Hydrogen bond distances (A
˚
) within the heme pockets of
molecules A, B, and C. Asterisks indicate water molecules located
between the heme propionate groups.
Molecule
A
Molecule
B
Molecule
C
His18 N
d1

Asn22 O 2.86 2.87 2.84
Heme O2D Water 34 ⁄ 496 ⁄ 137 2.61 2.87 2.64
Heme O2D Water 227 ⁄ 390 ⁄ 69 3.17 2.69 2.51
Heme O1D Water 227 ⁄ 390 ⁄ 69 3.22 3.38 3.16
Heme O1D Gln62 N
e2
3.40 3.35 3.37
Heme O1D Lys29 N
f
2.78 2.77 2.91
Heme O1D Water 20 ⁄ 18 ⁄ 24 2.69 2.70 2.70
Heme O2A Water 33 ⁄ 45 ⁄ 30 2.85 2.80 2.83
Heme O1A Thr30 O 2.66 2.65 2.63
*Water
20 ⁄ 18 ⁄ 24
*Water 33 ⁄ 45 ⁄ 30 2.58 2.61 2.61
Gln57 O
e1
Water 20 ⁄ 18 ⁄ 24 2.70 2.70 2.75
Gln57 N
e2
Gln62 O
e1
2.94 2.97 3.03
Fig. 4. Electrostatic surface potential distribution for PetJ1. The red
zones correspond to negative potential, and blue zones correspond
to positive potential. The color range spans )2 to 2 kT. Side chains
are shown as sticks. The side chain of Glu89 assumes two confor-
mations. The pale blue island on the visible red face of the molecule
corresponds to the NH

3
+
group of the N-terminus. The figure was
prepared in
PYMOL using default settings of the APBS plugin [41].
Atomic-resolution structure of cytochrome c
6
W. Bialek et al.
4430 FEBS Journal 276 (2009) 4426–4436 ª 2009 The Authors Journal compilation ª 2009 FEBS
Interactions of aromatic residues
In total, there are seven aromatic residues in the pro-
tein sequence, not including the sole His, which acts as
a heme ligand. The side chains of the three conserved
aromatic residues, Phe10, Tyr83, and Trp93, form a
triangular aromatic packing motif (Fig. 5A) buried in
the hydrophobic core of the molecule. In this motif,
C–H bonds at the edges of Tyr83 and Trp93 point to
the ring center of Phe10 (C–ring centroid distances
4.1–4.2 A
˚
) (Table S2). Similar C-H–p interactions are
also present in other cytochromes c
6
with known crys-
tal structures. In some cytochrome c
6A
molecules,
including that from A. thaliana, whose crystal structure
is known, the Tyr of this ‘aromatic triangle’ is replaced
by Phe. It is interesting to note that whereas the side

chain of Trp93, owing to these interactions, is tightly
wedged into a nearly ideally shaped pocket, the C-ter-
minus of its main chain shows significant flexibility
and has been modeled in two distinct conformations in
all three molecules (Fig. 5B).
Energetically significant cation–p interactions have
been detected between Tyr43, Tyr56 and Phe68 and,
respectively, Lys37, Lys48 and Arg71 by the capture
software [22]. Tyr56, although pointing with its OH
group towards bulk solvent, is firmly held in place by
two N-H–p interactions on both sides of the ring, from
Lys48 (described above) and from the side chain amide
of Asn60. The latter interaction, which is very short
(Table S2), may be especially important for stabiliza-
tion of Tyr56. Similar interactions stabilizing a key Tyr
residue were noted previously in the ultrahigh-resolu-
tion structure of bovine pancreatic trypsin inhibitor
[23]. In the present case, the amide N–p distance (3.29–
3.37 A
˚
) is even shorter than in the bovine pancreatic
trypsin inhibitor structure (3.44–3.58 A
˚
). The stability
of Tyr56 may be important for the conformation of
Gln57, and consequently for tuning the redox potential
of the protein. Phe68 forms part of a conserved motif,
Pro66-Ala-Phe68. This motif is adjacent to the second
heme ligand, Met65, and Pro66 undergoes conforma-
tional changes upon oxidation ⁄ reduction [16]. Whereas

in cyanobacterial, plant and red and brown algal cyto-
chrome c
6
this Phe is conserved, it is replaced by Trp in
green algae (Fig. 2C). It is of note that Phe68 interacts
with the single Arg found in cytochromes c
6
, Arg71
(Fig. 5C). This Arg is required for cytochrome f oxida-
tion [24] and photosystem I reduction [25]. In addition,
Phe68 forms an evident C–H–p interaction with the
heme group itself (Table S2). All of the remaining aro-
matic residues (Phe10, Tyr39, Tyr83, and Trp93) are
involved in C–H–p interactions (Table S2).
Comparison of existing models of cytochrome c
6
The three independent molecules of Synechococcus
cytochrome c
6
in the asymmetric unit are similar, with
pairwise rmsd values of 0.30 (A ⁄ C), 0.32 (B ⁄ C) and
0.36 A
˚
(A ⁄ B) for the Ca atoms (Fig. 6). In general, the
structures superpose very closely, but there are two
regions with larger discrepancies. The first one is the
single Gly63, and the second is a stretch of six residues
from Gly70 to Ala75, with the largest difference of
1.68 A
˚

between Gly70 of chains B and C. This large
A
B
C
Fig. 5. Interactions of aromatic residues. (A) The three aromatic
residues involved in a ‘triangular’ interaction in the protein core. (B)
A pocket around Trp93. Note the tightly wedged aromatic side
chain of Trp93 and the dual conformation of its C-terminal group.
(C) N–H p interaction between the only Arg and Phe68. (A) and (C)
show 2F
o
À F
c
electron density contoured at 1.2r.
W. Bialek et al. Atomic-resolution structure of cytochrome c
6
FEBS Journal 276 (2009) 4426–4436 ª 2009 The Authors Journal compilation ª 2009 FEBS 4431
difference can be attributed to a peculiarity of the
packing arrangement involving molecules B and C, in
which the side chain of Asp74 of molecule C accepts
two strong hydrogen bonds from the main chain N–H
groups of Gly70 and Arg71 of molecule B. In this
interaction, the carbonyl oxygen of Gly69 of mole-
cule B flips by 180°, bringing the main chain nitrogen
atom of residue 70 much closer to the O
d
atom of the
adjacent Asp74. The changed geometry of this frag-
ment should be classified as reverse turn type III. The
corresponding fragment in the other two molecules, as

in all known X-ray structures of c
6
-type cytochromes,
is classified as b-turn type II. The higher lability of this
region in the otherwise very rigid and well-ordered
structure is not surprising, because, in general, the
region on the C-side of the heme moiety (i.e. including
residues from the loop between helices II and III to
the C-terminus) is more flexible, and the acidic region
Asp73-Glu74-Asp75-Glu76 is a hot spot in the struc-
ture of cytochrome c
6
from M. braunii [14].
There are 13 side chains that do not align in molecu-
lar superpositions, adopting different rotamers in one
or more of the three molecules. With two exceptions,
all of these side chains are on the surface, and some of
them are clearly perturbed by crystal contacts. The
first exception is Pro27, which in molecule C has the
exo pucker, whereas in the other two cases it is endo.
The second exception is Ile84, which is buried inside
the molecular core but in molecule B has C
d
in two
conformations. Only 10 of the 261 side chains (3.8%)
in the asymmetric unit have the side chains in double
conformation. Such a small number of alternative con-
formations is related to the unusual amino acid com-
position, where Ala accounts for 25.8% and Gly for
9.7% of all the residues in the amino acid sequence of

this cytochrome c
6
.
Table S1 shows both the percentage of sequence
identity and the overall Ca rmsd values for pairs of
different cytochrome c
6
structures. The lowest Ca
backbone deviations are observed in the regions of
helices I, III, and IV, as well as the 3
10
helix and the
X-loop. consurf analysis [26] has revealed several resi-
dues that are most conserved (data not shown). These
residues are in the heme pocket area and are involved
in electron transfer. Whereas they are located within
regions of the lowest Ca discrepancies, the shortest
helix II is sequentially the least conserved element
among the cytochrome c
6
molecules. The largest devia-
tions are observed around the specific insertion of this
cytochrome c
6
molecule, i.e. in the loop connecting
helices II and III.
Intermolecular contacts
As a consequence of the very low solvent content
(17.58%) of the Synechococcus cytochrome c
6

crystals,
their structure is very densely packed, and a number of
surface residues are involved in direct contacts with
neighboring protein molecules. The contacts between
molecules A and B involve three hydrogen bonds from
the main chain amide of Ala75 and the side chains of
Thr28 and Lys29 to the main chain carbonyl groups
of, respectively, Asp45 (B), Gly70 (A), and Arg71 (A).
There are only two direct hydrogen bonds between
molecules A and C involving main chain atoms of
molecule A, namely oxygen of Asn60 and nitrogen of
Gly63, and the side chains of Lys29 and Tyr39 of mol-
ecule C. Three of the four direct hydrogen bonds
between molecules B and C are formed by the carbox-
ylic group of Asp74 of molecule C, which interacts
with the N of Gly70, and both the N
g
atoms of
Arg71. The fourth bond is created between the Asp74
nitrogen atom (molecule B) and Gln52 O
e2
.
There are five direct hydrogen bonds between the
three molecules in the asymmetric units and their sym-
metry counterparts. Three of them are created by
Asn91, which uses its N
d
(two bonds) and O for inter-
actions with, respectively, the main chain atoms of
10

30
10
Fig. 6. Stereoview of the superposition of
the main chain traces of the three indepen-
dent molecules in the asymmetric unit, with
residue numbering added for orientation.
The side chains of the two iron-coordinating
residues (His18 and Met65) are shown as
sticks. Color code as in Fig. 3.
Atomic-resolution structure of cytochrome c
6
W. Bialek et al.
4432 FEBS Journal 276 (2009) 4426–4436 ª 2009 The Authors Journal compilation ª 2009 FEBS
Ala7 and Gly20, and the side chain of Tyr83. The
latter side chain is also a hydrogen bond donor to the
Gln89 O. Gln87 and Asn90 are linked through an
N
e
ÆO
d
hydrogen bond.
Because of the low solvent content, the surface area
buried on intermolecular contacts (including asym-
metric unit and symmetry-related protein molecules) is
quite high, 41.7% (molecule A), 45.8% (molecule B),
and 47.1% (molecule C).
Crystal packing and water structure
With three independent molecules of cytochrome c
6
in

the crystallographic asymmetric unit, the volume-to-
mass ratio V
M
is 1.49 A
˚
3Æ
Da
)1
, corresponding to a
solvent content of merely 17.58%. There are only four
protein structures deposited in the Protein Data Bank
that have lower reported solvent content (V
S
) (1XEK,
2J70, 2DUY, and 3C00). However, the number of
wrongly attributed Matthews coefficients and solvent
volumes in structures deposited in the Protein Data
Bank is astonishingly high. Searching the Protein Data
Bank for protein structures with V
S
between 5.00%
and 17.57% gave 29 structures. Calculations of the V
M
and V
S
values (Matthews program, CCP4 [27]) for
those structures returned only four cases with V
S
within the search interval. Therefore, we would not be
surprised if the actual number of structures with lower

V
S
was higher.
The solvent molecules modeled in the electron den-
sity maps correspond to 252 H
2
O sites with full occu-
pancy and 272 with partial occupancy, which are
equivalent to 404 effective water molecules. Although
the number of partial water molecules may seem high,
the average water occupancy is 0.76, which is very
close to the value of 0.78 determined for a protein
structure at ultimate resolution (0.65 A
˚
) where solvent
occupancies were refined [28]. Considering the low sol-
vent content of this crystal, one can conclude that
essentially all water molecules in the solvent region
have been accounted for in this atomic model of the
crystal structure. Additionally, there are two partially
occupied sulfate ions. One of them forms two hydro-
gen bonds with the main chain nitrogen atoms of
Asp74 (2.80 A
˚
) and Ala67 (2.86 A
˚
) and with a water
molecule (2.43 A
˚
). The other one forms two hydrogen

bonds with N
g2
(3.12 A
˚
) and N
e
(3.03 A
˚
) of Arg71
and with two water molecules (2.79 and 2.77 A
˚
).
Conclusions
In summary, cytochrome c
6
from Synechococcus
sp. PCC 7002 has an overall structure that is similar
to that of other previously characterized cyto-
chrome c
6
molecules. However, unlike other cyto-
chrome c
6
molecules, the Synechococcus sp. PCC 7002
protein has a unique insertion of seven amino acid
residues [KDGSKSL(44–50)], which is entirely differ-
ent from the insertion found in type c
6A
cyto-
chromes. The presence of the insertion in the most

variable region of cytochrome c
6
might suggest that
it does not play any role in the cell. On the other
hand, the inability to inactivate or replace the gene
encoding this cytochrome c
6
by PC or cytochrome c
6
from Synechocystis [19] is a strong indication that
the presence of the insertion is essential for this
organism. The question of whether the Synechococcus
sp. PCC 7002 insertion does indeed have a biological
function is currently being studied by site-directed
mutagenesis.
Experimental procedures
Protein expression and purification
Escherichia coli strain DH5a was cotransformed with
pUCJ1 and pEC86 plasmids. The former harbors a gene
encoding mature cytochrome c
6
from Synechococcus
sp. PCC 7002, whereas the latter harbors the heme matura-
tion genes [29]. Protein was expressed basically as described
elsewhere [8]. Briefly, 5 mL of overnight culture was used
to inoculate 1.7 L of LB medium supplemented with 1 mm
FeCl
3
in a 2 L flask. Cultures were grown for 72–96 h at
30 °C with agitation at 150 r.p.m. and harvested, and the

periplasmic proteins were released by lysozyme treatment
(1 mgÆmL
)1
). Subsequently, cytochrome c
6
was purified as
outlined in [8].
Protein crystallization
The hanging-drop vapor diffusion method was used for the
crystallization experiments. Cytochrome c
6
(1 mm in 10 mm
Tris buffer, pH 7.5) was reduced by the addition of sodium
dithionate to a final concentration of 20 mm. Drops con-
taining equal volumes (1 lL) of protein and reservoir solu-
tions were equilibrated against 1 mL of reservoir solution
in 24-well Linbro plates at 19 °C. Red crystals suitable for
X-ray analysis were obtained from 10 mm sodium Hepes
(pH 6.2) and 2.2 m ammonium sulfate over a period of
1 week.
X-ray diffraction data collection
A single crystal was scooped through a solution of 30%
glycerol and 70% reservoir solution (v ⁄ v) and placed in a
stream of cold N
2
gas at 100 K. Diffraction data were col-
lected on the X11 synchrotron beamline (k = 0.8170 A
˚
)at
W. Bialek et al. Atomic-resolution structure of cytochrome c

6
FEBS Journal 276 (2009) 4426–4436 ª 2009 The Authors Journal compilation ª 2009 FEBS 4433
the EMBL Outstation, c ⁄ o DESY, Hamburg, using a
MAR 165 mm CCD detector. Two passes were collected,
using different crystal-to-detector distances, oscillations (1°
and 0.5°) and exposure times to ensure reliable measure-
ment of high-resolution data as well as of the strong
low-angle reflections. The Synechococcus sp. PCC 7002
cytochrome c
6
crystals belong to space group P3
2
with unit
cell dimensions a = 82.88 A
˚
and c = 28.28 A
˚
and diffract
to at least 1.226 A
˚
. The mean overall redundancy is 5.0,
and it is 2.8 in the highest-resolution shell (1.260–1.226 A
˚
).
The data are 98.7% complete with an overall R
int
of 0.076.
Post-refinement of the crystal parameters during scaling
indicated low mosaicity throughout the data collection
(0.29°). The diffraction data were indexed, integrated and

scaled using the hkl2000 [30] package, and the final statis-
tics are included in Table 3.
Structure solution and refinement
The structure was solved by the molecular replacement
method as programmed in molrep [31], using diffraction
data in the 20.0–3.5 A
˚
resolution range and the cyto-
chrome c
6
model from the green alga M. brauni [14] (Protein
Data Bank accession number: 1CTJ) as the molecular probe.
Analysis of the Matthews volume [32] indicated the pres-
ence of two or three protein molecules in the asymmetric
unit (2.24 or 1.49 A
˚
3Æ
Da
)1
, solvent content 45.05% or
17.58%). Molecular replacement calculations confirmed
that there were three independent protein chains. The cor-
rect space group enantiomorph, P3
2
, was deduced from a
comparison of the translation function solutions in the two
alternative space groups (P3
2
⁄ P3
1

with CC = 0.430 ⁄ 0.235,
R = 0.454 ⁄ 0.526).
The structure was refined anisotropically using the maxi-
mum-likelihood algorithm as implemented in refmac [33],
including the TLS parameters [34]. For free R-factor calcu-
lations [35], 1274 reflections were randomly selected from
the full resolution range of 50.0–1.226 A
˚
. After the first
round of refmac refinement (isotropic, no TLS parame-
ters), the model was rebuilt in quanta [36]. The final
refined model is characterized by an R-factor of 0.107 and
an R
free
of 0.138 for all 62 908 reflections between 50 and
1.226 A
˚
, and has an rmsd from ideal bond lengths of
0.020 A
˚
. The very high quality of the electron density maps
allowed the identification of 524 (including 252 fully occu-
pied) water sites and of two partially occupied sulfate ions.
Stereochemical analysis of the final model using
procheck [37] indicated that there are no residues with
generously allowed or unfavorable backbone dihedral
angles, and that 87.1% of all residues are in the core region
of the Ramachandran plot. The statistics of the refinement
are shown in Table 3.
Acknowledgements

Some of the calculations were carried out in the Poz-
nan Metropolitan Supercomputing and Networking
Center. This work was funded in part by grants
N N204 245635 and N N303 3856 33 from the Minis-
try of Science and Higher Education.
References
1 Cramer WA, Furbacher PN, Szczepaniak A & Tae GS
(1991) Electron Transport between Photosystem II and
Photosystem I. Academic Press, San Diego.
2 Kallas T (1994) The cytochrome b6f complex. In
The Molecular Biology of Cyanobacteria: Advances in
Photosynthesis and Respiration (Bryant DA, ed.),
pp. 559–579. Springer, Dordrecht.
3 Herva
´
s M, Navarro JA & De la Rosa MA (2003) Elec-
tron transfer between membrane complexes and soluble
Table 3. Data collection and refinement statistics.
Data collection
Radiation source X11, EMBL Hamburg
Wavelength (A
˚
) 0.8170
Temperature of measurements (K) 100
Space group P3
2
Cell dimensions (A
˚
) a = 82.88, c = 28.28
Mosaicity (°) 0.29

Resolution range (A
˚
) 50.00–1.226
(1.260–1.226)
a
Crystal-to-detector distance (mm) 160 ⁄ 100
Oscillation step (°) 1.0 ⁄ 0.5
No. of images 120 ⁄ 240
Reflections collected 313 876
Unique reflections 62 960 (5567)
Completeness (%) 98.7 (87.7)
Redundancy 5.0 (2.8)
<I>=<rI> 23.1 (2.4)
R
int
b
0.076 (0.389)
Refinement
Independent protein molecules 3
Matthews coefficient (A
˚
3Æ
Da
)1
) 1.49
No. of reflections in work ⁄ test set 61 634 ⁄ 1274
Rejection criterion None
R
c
⁄ R

free
(%) 0.107 ⁄ 0.138
Number of atoms (protein ⁄ solvent)
rmsd from ideal
2708 (2174 ⁄ 534)
Bond lengths (A
˚
) 0.020
Bond angles (°) 1.76
Chiral volumes (A
˚
3
) 0.104
Average B-factor (protein ⁄ solvent) (A
˚
2
) 11.40 (9.59 ⁄ 18.83)
Residues in Ramachandran plot (%)
Most favored 86.7
Additional 13.3
a
Values in parentheses correspond to the last resolution shell.
b
R
int
¼ R
h
R
j
I

hj
À <I
h
>




=R
h
R
j
I
hj
, where I
hj
is the intensity of obser-
vation j of reflection h.
c
R ¼ R
h
F
o
jjÀF
c
jjjj=R
h
F
o
jjfor all reflec-

tions, where F
o
and F
c
are observed and calculated structure
factors, respectively. R
free
is calculated analogously for the test
reflections, randomly selected and excluded from the refinement.
Atomic-resolution structure of cytochrome c
6
W. Bialek et al.
4434 FEBS Journal 276 (2009) 4426–4436 ª 2009 The Authors Journal compilation ª 2009 FEBS
proteins in photosynthesis. Accounts Chem Res 36, 789–
805.
4 Wastl J, Bendall DS & Howe CJ (2002) Higher plants
contain a modified cytochrome c6. Trends Plant Sci 7,
244–245.
5 Gupta R, He Z & Luan S (2002) Functional relation-
ship of cytochrome c6 and plastocyanin in Arabidopsis.
Nature 417, 567–571.
6 Wastl J, Purton S, Bendall DS & Howe CJ (2004) Two
forms of cytochrome c6 in a single eukaryote. Trends
Plant Sci 9, 474–476.
7 Worrall JA, Schlarb-Ridley BG, Reda T, Marcaida MJ,
Moorlen RJ, Wastl J, Hirst J, Bendall DS, Luisi BF &
Howe CJ (2007) Modulation of heme redox potential in
the cytochrome c6 family. J Am Chem Soc 129, 9468–
9475.
8 Bialek W, Nelson M, Tamiola K, Kallas T & Szczepa-

niak A (2008) Deeply branching c6-like cytochromes of
cyanobacteria. Biochemistry 47, 5515–5522.
9 Schlarb-Ridley BG, Nimmo RH, Purton S, Howe CJ &
Bendall DS (2006) Cytochrome c6A is a funnel for thiol
oxidation in the thylakoid lumen. FEBS Lett 580,
2166–2169.
10 Sawaya MR, Krogmann DW, Serag A, Ho KK, Yeates
TO & Kerfeld CA (2001) Structures of cytochrome
c-549 and cytochrome c6 from the cyanobacterium
Arthrospira maxima. Biochemistry 40, 9215–9225.
11 Beissinger M, Sticht H, Sutter M, Ejchart A, Haehnel
W & Rosch P (1998) Solution structure of cyto-
chrome c6 from the thermophilic cyanobacterium Syn-
echococcus elongatus. EMBO J 17, 27–36.
12 Yamada S, Park SY, Shimizu H, Koshizuka Y, Kadok-
ura K, Satoh T, Suruga K, Ogawa M, Isogai Y, Nishio
T et al. (2000) Structure of cytochrome c6 from the red
alga Porphyra yezoensis at 1.57 A
˚
resolution. Acta Crys-
tallogr D Biol Crystallogr 56, 1577–1582.
13 Akazaki H, Kawai F, Chida H, Matsumoto Y, Hiray-
ama M, Hoshikawa K, Unzai S, Hakamata W, Nishio
T, Park SY et al. (2008) Cloning, expression and purifi-
cation of cytochrome c6 from the brown alga Hizikia
fusiformis and complete X-ray diffraction analysis of
the structure. Acta Crystallogr Sect F Struct Biol Cryst
Commun 64, 674–680.
14 Frazao C, Soares CM, Carrondo MA, Pohl E, Dauter
Z, Wilson KS, Hervas M, Navarro JA, De la Rosa MA

& Sheldrick GM (1995) Ab initio determination of the
crystal structure of cytochrome c6 and comparison with
plastocyanin. Structure 3, 1159–1169.
15 Kerfeld CA, Anwar HP, Interrante R, Merchant S &
Yeates TO (1995) The structure of chloroplast cyto-
chrome c6 at 1.9 A
˚
resolution: evidence for functional
oligomerization. J Mol Biol 250, 627–647.
16 Schnackenberg J, Than ME, Mann K, Wiegand G,
Huber R & Reuter W (1999) Amino acid sequence,
crystallization and structure determination of reduced
and oxidized cytochrome c6 from the green alga Scene-
desmus obliquus. J Mol Biol 290
, 1019–1030.
17 Dikiy A, Carpentier W, Vandenberghe I, Borsari M,
Safarov N, Dikaya E, Van Beeumen J & Ciurli S (2002)
Structural basis for the molecular properties of cyto-
chrome c6. Biochemistry 41, 14689–14699.
18 Marcaida MJ, Schlarb-Ridley BG, Worrall JA, Wastl J,
Evans TJ, Bendall DS, Luisi BF & Howe CJ (2006)
Structure of cytochrome c6A, a novel dithio-cyto-
chrome of Arabidopsis thaliana, and its reactivity with
plastocyanin: implications for function. J Mol Biol 360,
968–977.
19 Nomura C & Bryant DA (1999) Cytochrome c6 from
Synechococcus sp. PCC 7002. In The Phototrophic
Prokaryotes (Peschek G, ed.), pp. 269–274. Kluwer
Academic ⁄ Plenum, New York.
20 Zhang L, Pakrasi HB & Whitmarsh J (1994) Photoau-

totrophic growth of the cyanobacterium Synechocystis
sp. PCC 6803 in the absence of cytochrome c553 and
plastocyanin. J Biol Chem 269, 5036–5042.
21 Laskowski RA (2009) PDBsum new things. Nucleic
Acids Res 37, D355–D359.
22 Gallivan JP & Dougherty DA (1999) Cation-pi interac-
tions in structural biology. Proc Natl Acad Sci USA 96,
9459–9464.
23 Addlagatta A, Krzywda S, Czapinska H, Otlewski J &
Jaskolski M (2001) Ultrahigh-resolution structure of a
BPTI mutant. Acta Crystallogr D Biol Crystallogr 57,
649–663.
24 Albarran C, Navarro JA, Molina-Heredia FP, Murdoch
Pdel S, De la Rosa MA & Hervas M (2005) Laser
flash-induced kinetic analysis of cytochrome f oxidation
by wild-type and mutant plastocyanin from the cyano-
bacterium Nostoc sp. PCC 7119. Biochemistry 44,
11601–11607.
25 Molina-Heredia FP, Hervas M, Navarro JA & De la
Rosa MA (2001) A single arginyl residue in plastocya-
nin and in cytochrome c6 from the cyanobacterium An-
abaena sp. PCC 7119 is required for efficient reduction
of photosystem I. J Biol Chem 276, 601–605.
26 Landau M, Mayrose I, Rosenberg Y, Glaser F, Martz
E, Pupko T & Ben-Tal N (2005) ConSurf 2005: the pro-
jection of evolutionary conservation scores of residues
on protein structures. Nucleic Acids Res 33, W299–
W302.
27 (1994) The CCP4 suite: programs for protein crystallog-
raphy. Acta Crystallogr D Biol Crystallogr 50, 760–763.

28 Wang J, Dauter M, Alkire R, Joachimiak A & Dauter
Z (2007) Triclinic lysozyme at 0.65 A
˚
resolution. Acta
Crystallogr D Biol Crystallogr 63, 1254–1268.
29 Arslan E, Schulz H, Zufferey R, Kunzler P & Thony-
Meyer L (1998) Overproduction of the Bradyrhizobium
japonicum c-type cytochrome subunits of the cbb3 oxi-
dase in Escherichia coli. Biochem Biophys Res Commun
251, 744–747.
W. Bialek et al. Atomic-resolution structure of cytochrome c
6
FEBS Journal 276 (2009) 4426–4436 ª 2009 The Authors Journal compilation ª 2009 FEBS 4435
30 Otwinowski Z & Minor W (1997) Processing of X-ray
diffraction data collected in oscillation mode. Methods
Enzymol 276, 307–326.
31 Vagin A & Teplyakov A (1997) MOLREP: an auto-
mated program for molecular replacement. J Appl Crys-
tallogr 30, 1022–1025.
32 Matthews BW (1968) Solvent content of protein crys-
tals. J Mol Biol 33, 491–497.
33 Murshudov GN, Vagin AA & Dodson EJ (1997)
Refinement of macromolecular structures by the maxi-
mum-likelihood method. Acta Crystallogr D Biol Crys-
tallogr 53, 240–255.
34 Winn MD, Isupov MN & Murshudov GN (2001) Use
of TLS parameters to model anisotropic displacements
in macromolecular refinement. Acta Crystallogr D Biol
Crystallogr 57, 122–133.
35 Brunger AT (1992) Free R value: a novel statistical

quantity for assessing the accuracy of crystal structures.
Nature 355, 472–475.
36 Oldfield TJ (1994) A semi-automated map fitting proce-
dure. Proceedings of the CCP4 Study Weekend. From
First Map to Final Model. (Bailey S, Hubbard R &
Waller D, eds). Daresbury Laboratory, Warrington,
pp. 15–16.
37 Laskowski RA, MacArthur MW, Moss DS & Thornton
JM (1993) PROCHECK: a program to check the
stereochemical quality of protein structures. J Appl
Crystallogr 26, 283–291.
38 Hooft RW, Sander C & Vriend G (1996) Positioning
hydrogen atoms by optimizing hydrogen-bond networks
in protein structures. Proteins 26, 363–376.
39 DeLano WL (2002) The PyMOL Molecular Graphics
System. DeLano Scientific, Palo Alto, CA, USA.
40 Gouet P, Courcelle E, Stuart DI & Metoz F (1999)
ESPript: analysis of multiple sequence alignments in
PostScript. Bioinformatics 15, 305–308.
41 Baker NA, Sept D, Joseph S, Holst MJ & McCammon
JA (2001) Electrostatics of nanosystems: application to
microtubules and the ribosome. Proc Natl Acad Sci
USA 98, 10037–10041.
Supporting information
The following supplementary material is available:
Fig. S1. (A) 2F
o
À F
c
electron density around Pro66 in

molecule A, contoured at 1.3r. (B) The Pro66 residues
of the present PetJ1 molecules A (red), B (green) and
C (blue) aligned with oxidized cytochrome c6 from
S. obliquus (yellow).
Table S1. Sequence identity values and rmsd values
of Ca superpositions for crystallographically
characterized cytochromes from Synechococcus
sp. PCC 7002 (Protein Data Bank code 3DR0, this
work), C. reinhardtii (1CYI), S. obliquus (1C6O),
C. glomerata (1LS9), M. braunii (1CTJ), A. maxima
(1F1F), P. yezoensis (1GDV), A. thaliana (2DGE),
P. laminosum (2V08), and H. fusiformis (2ZBO).
Table S2. X–H–p interactions of the aromatic residues.
This supplementary material can be found in the
online article.
Please note: As a service to our authors and readers,
this journal provides supporting information supplied
by the authors. Such materials are peer-reviewed and
may be re-organized for online delivery, but are not
copy-edited or typeset. Technical support issues arising
from supporting information (other than missing files)
should be addressed to the authors.
Atomic-resolution structure of cytochrome c
6
W. Bialek et al.
4436 FEBS Journal 276 (2009) 4426–4436 ª 2009 The Authors Journal compilation ª 2009 FEBS