Structure of a trypanosomatid mitochondrial
cytochrome c with heme attached via only one thioether
bond and implications for the substrate recognition
requirements of heme lyase
Vilmos Fu
¨
lo
¨
p
1
, Katharine A. Sam
2
, Stuart J. Ferguson
2
, Michael L. Ginger
3
and James W. A. Allen
2
1 Department of Biological Sciences, University of Warwick, Coventry, UK
2 Department of Biochemistry, University of Oxford, UK
3 School of Health and Medicine, Division of Biomedical and Life Sciences, Lancaster University, UK
The principal physiological role of mitochondrial cyto-
chrome c is electron transfer from the cytochrome bc
1
complex to cytochrome aa
3
oxidase during oxidative
phosphorylation. c-Type cytochromes form a large
family in bacteria, archaea, mitochondria and chlorop-
lasts, in which the iron cofactor heme is covalently
bound to the polypeptide chain. Such cytochromes

with heme attached through only one thioether bond [to an (A ⁄ F)XXCH
motif]; the implications of this for the cytochrome structures are unclear.
Here we present the 1.55 A
˚
resolution X-ray crystal structure of cyto-
chrome c from the trypanosomatid Crithidia fasciculata. Despite the funda-
mental difference in heme attachment and in the cytochrome c biogenesis
machinery of the Euglenozoa, the structure is remarkably similar to that of
typical (CXXCH) mitochondrial cytochromes c, both in overall fold and,
other than the missing thioether bond, in the details of the heme attach-
ment. Notably, this similarity includes the stereochemistry of the covalent
heme attachment to the protein. The structure has implications for the
maturation of c-type cytochromes in the Euglenozoa; it also hints at a dis-
tinctive redox environment in the mitochondrial intermembrane space of
trypanosomes. Surprisingly, Saccharomyces cerevisiae cytochrome c heme
lyase (the yeast cytochrome c biogenesis system) cannot efficiently mature
Trypanosoma brucei cytochrome c or a CXXCH variant when expressed in
the cytoplasm of Escherichia coli, despite their great structural similarity to
yeast cytochrome c, suggesting that heme lyase requires specific recognition
features in the apocytochrome.
Abbreviations
Ccm, cytochrome c maturation; IMS, intermembrane space; NCS, noncrystallographic symmetry; SHAM, salicylhydroxamic acid.
2822 FEBS Journal 276 (2009) 2822–2832 ª 2009 The Authors Journal compilation ª 2009 FEBS
always through two thioether bonds between the vinyl
groups of heme and the thiols of cysteine residues that
occur in a CXXCH amino acid motif; the histidine
serves as the proximal ligand to the heme iron atom
[2–4]. However, in one group of eukaryotes, the mem-
bers of the protist phylum Euglenozoa, biochemical,
spectroscopic and genetic evidence suggests that,

are present in the several trypanosomatid species for
which complete genome sequences are available
[9,10,12]. The presence throughout the Euglenozoa of
unique single cysteine mitochondrial cytochromes c,
coupled with the absence from trypanosomatids of any
known cytochrome c biogenesis proteins, points
towards a novel maturation apparatus for all eugleno-
zoan mitochondrial cytochromes c [9,10].
The reason for the loss of one thioether bond from
the mitochondrial cytochromes c of euglenozoans is a
longstanding puzzle. It could be a means of altering
the structure of these cytochromes c and ⁄ or a conse-
quence of some biosynthetic demand. A high-resolu-
tion structural comparison between a euglenozoan
mitochondrial cytochrome c and a typical cyto-
chrome c (with heme bound by two thioether bonds) is
therefore important. Moreover, in all the available,
diverse structures of c-type cytochromes, there is an
invariant stereospecific arrangement of the heme [1,13];
there is no a priori reason to expect this to be the same
in the single thioether bond euglenozoan cytochromes.
Finally, it is likely that (at least for cytochrome c bio-
genesis Systems I and II) folding of holocytochromes c
mainly takes place after covalent attachment of
heme to the polypeptide chain. Thus, it is not axiom-
atic that anchoring the heme to protein through only a
single thioether bond would result in the same local
structure that is characteristic of heme attachment to a
CXXCH motif. Therefore, we have determined the
X-ray crystal structure of mitochondrial cytochrome c

presence of 200 lm salicylhydroxamic acid (SHAM), a
specific inhibitor of alternative oxidase, had little effect
on the growth rate of C. fasciculata when compared
with control cultures. Addition of 2 lgÆmL
)1
antimycin
A (a cytochrome bc
1
complex inhibitor), however,
resulted in no growth over 72 h (from a starting inocu-
lum of 10
5
cellsÆmL
)1
). Similarly, addition of SHAM
to a concentration of 3 mm exerted no effect on
oxygen consumption by C. fasciculata as measured
using a Clark oxygen electrode, whereas following
V. Fu
¨
lo
¨
p et al. Structure of Crithidia fasciculata cytochrome c
FEBS Journal 276 (2009) 2822–2832 ª 2009 The Authors Journal compilation ª 2009 FEBS 2823
addition of antimycin to 2 lgÆmL
)1
, oxygen consump-
tion by C. fasciculata effectively ceased within a few
seconds (Fig. 1). A very similar result was obtained if
2mm KCN was added instead of antimycin A; cyanide

e
of the histidine of the ‘AXXCH’ (actually AAQCH)
heme-binding motif, and the sulfur of Met91. The
heme iron–ligand distances were restrained to 2 A
˚
(his-
tidine) and 2.3 A
˚
(methionine), and the model fits well
with these values. The vinyl a-carbon–Cys28 distances
were restrained to 1.8 A
˚
, but these refined to consider-
ably longer bond lengths: 2.25, 1.98 and 2.00 A
˚
,
respectively, for the three subunits of the asymmetric
unit. Residue 83 is a trimethyllysine [5]; a moderate fit
to the electron density suggests that the trimethyl
group is flexible.
The structure of C. fasciculata cytochrome c (in red)
is overlaid with that of S. cerevisiae iso-1-cytochrome c
(in blue) in Fig. 3. The two cytochromes have 48%
amino acid identity and 72% similarity. The structures
are remarkably similar overall, and in the details of
the heme attachment and heme position. The rmsd
between the structure of C. fasciculata cytochrome c
and S. cerevisiae iso-1-cytochrome c is 0.94 A
˚
for the

resolution. The molecule is shown rainbow colored, from
the N-terminus (residue 5) in blue to the C-terminus (residue 114)
in red. The heme cofactor is shown in ball and stick representation,
as are the methionine and histidine side chains that coordinate the
heme iron, and the cysteine side chain that forms a thioether bond
between heme and protein. Also shown is the methyl group of the
alanine of the AXXCH heme-binding motif, which is found in place
of the first cysteine of a typical c-type cytochrome CXXCH heme-
binding motif.
Structure of Crithidia fasciculata cytochrome c V. Fu
¨
lo
¨
p et al.
2824 FEBS Journal 276 (2009) 2822–2832 ª 2009 The Authors Journal compilation ª 2009 FEBS
identical between the two structures. These observa-
tions concur with those of Wuthrich et al., who used
proton NMR to investigate the heme environment in
cytochrome c from another trypanosomatid species in
the 1970s, and concluded that the heme crevice and
heme electronic structure were very similar to those in
mammalian (CXXCH) mitochondrial cytochrome c
[21,22]. (Notably, these authors also correctly predicted
the overall structure of the trypanosomatid cyto-
chrome.) The stereochemistry of heme attachment
through the thioether bond is conserved in C. fascicu-
lata cytochrome c (Figs 3 and 4); it therefore remains
the same (i.e. S-stereochemistry) as in all known c-type
cytochrome structures [1,13]. Strikingly, the methyl
group of Ala25 in C. fasciculata cytochrome c (the

Fig. 4. Detail of the heme-binding site in C. fasciculata cyto-
chrome c from two angles. The
SIGMAA [50] weighted 2mF
o
)DF
c
electron density, using phases from the final model of the half-
reduced form, is contoured at the 1.5r level, where r represents the
rms electron density for the unit cell. Contours more than 1.5 A
˚
from
any of the displayed atoms have been removed for clarity. Thin lines
indicate heme axial ligand coordination and hydrogen bonds. Sulfur
atoms (in methionine and cysteine) are colored yellow, nitrogen
blue, oxygen red, and iron purple. The methyl group of the alanine of
the AXXCH heme-binding motif is green, and the unsaturated vinyl
group of heme is cyan. Pro41 is conserved in class I c-type cyto-
chromes, and its main chain carbonyl is hydrogen bonded to the N
d
atom of the heme axial histidine side chain; this interaction main-
tains the correct orientation of the histidine ring to the heme iron.
V. Fu
¨
lo
¨
p et al. Structure of Crithidia fasciculata cytochrome c
FEBS Journal 276 (2009) 2822–2832 ª 2009 The Authors Journal compilation ª 2009 FEBS 2825
of the alanine of the AXXCH motif (Ala25) (in
green) and the unsaturated vinyl group of the heme
(cyan) are separated by 3.41 A

wild-type [24,25]. Thus, we coexpressed S. cerevisiae
cytochrome c heme lyase with either T. brucei cyto-
chrome c or a CXXCH variant in the cytoplasm of
E. coli (the cytochromes c from C. fasciculata and
T. brucei have 84% sequence identity and 92% simi-
larity, and both have an AAQCH heme-binding
motif). As a control, we also coexpressed heme lyase
with S. cerevisiae iso-1-cytochrome c. Cells expressing
the yeast cytochrome were bright red, and were shown
by absorption spectroscopy to have produced
$ 1.4 mg of cytochrome c per gram of wet cells
(assuming a reduced Soret band extinction coefficient
of 130 000 m
)1
Æcm
)1
[20]). However, neither wild-type
T. brucei cytochrome c nor the CXXCH variant was
matured at levels immediately detectable by spectros-
copy or by staining of SDS ⁄ PAGE gels for proteins
with covalently bound heme. Expression of the protein
was, however, readily confirmed by western blotting of
the E. coli cytoplasmic extracts using a polyclonal
antibody raised against recombinant T. brucei
CXXCH holocytochrome c (Fig. 5A), the latter
matured by the E. coli Ccm system [9]; this antibody
is sensitive to both T. brucei holocytochrome c and
T. brucei apocytochrome c (J. W. A. Allen, unpub-
lished observation). Following concentration of the
E. coli cytoplasmic extracts, both T. brucei wild-type

chrome c. Lane 10: as the positive control, purified CXXCH variant
T. brucei holocytochrome c matured in the periplasm of E. coli by
the E. coli Ccm apparatus. (B) SDS ⁄ PAGE gel of concentrated cyto-
plasmic extracts from E. coli coexpressing heme lyase and cyto-
chrome c, stained for proteins containing covalently bound heme.
Lane 1: molecular mass markers. Lane 2: wild-type T. brucei cyto-
chrome c. Lane 3: the CXXCH variant of T. brucei cytochrome c.
Lane 4: as the positive control, purified CXXCH variant T. brucei
holocytochrome c matured in the periplasm of E. coli by the E. coli
Ccm apparatus. (C) Absorption spectra of concentrated cytoplasmic
extracts from E. coli coexpressing heme lyase and cytochrome c.
Wild-type T. brucei cytochrome c (black line) and the CXXCH variant
(gray line). A few grains of disodium dithionite were added to the
samples to reduce the cytochromes. As no extinction coefficients
are available, the spectra were normalized by intensity of the Soret
band. The spectra were also corrected for light scattering by sub-
traction of a wavelength to the power four curve.
Structure of Crithidia fasciculata cytochrome c V. Fu
¨
lo
¨
p et al.
2826 FEBS Journal 276 (2009) 2822–2832 ª 2009 The Authors Journal compilation ª 2009 FEBS
(Fig. 5B). Yields of heme lyase-matured T. brucei
holocytochrome were calculated from absorption
spectra of the concentrated cytochromes (Fig. 5C) as
3.1 lg (wild-type) and 3.2 lg (CXXCH) of holocyto-
chrome c per gram of wet cells, assuming in each
case a reduced Soret band extinction coefficient of
130 000 m

heme-binding motif, rather than through two thioether
bonds to CXXCH, as in all other eukaryotes. No
apparatus for the post-translational attachment of
heme to apocytochrome c has yet been identified in
any euglenozoan, and, in contrast to all other eukary-
otes possessing mitochondrial cytochromes c, no appa-
ratus is evident from the analysis of multiple
completely sequenced trypanosomatid genomes
[9,10,12]. Identification of the novel biogenesis system
for cytochrome c in trypanosomes is a demanding
task. Remarkably, despite these fundamental differ-
ences in heme attachment and cytochrome biogenesis,
the structure of C. fasciculata cytochrome c is very
similar to the structures of typical mitochondrial cyto-
chromes c, e.g. from S. cerevisiae (Fig. 3). This simi-
larity was also observed, other than the missing
thioether bond, in the details of the heme attachment
and around the heme-binding site (Figs 3 and 4) [21].
Different c-type cytochromes have very different folds
[1], but the structural arrangement of the heme-bind-
ing motif around the thioether linkages is absolutely
conserved. The present work extends this observation
to a cytochrome with a natural single cysteine heme-
binding motif. Moreover, as illustrated with E. gracilis
cytochrome c
558
, single thioether attachment of heme
does not significantly affect the reaction between the
cytochrome c and (mammalian) cytochrome bc
1

chromes c be explained? Considerable evidence points
to catalyzed formation and subsequent reduction of
an intramolecular disulfide bond in the CXXCH
motif during cytochrome c biogenesis in bacteria [3];
this may also happen in yeast [27]. It therefore seems
plausible that evolution of the euglenozoan single cys-
teine heme-binding motif, while the protein structure
was otherwise retained (Fig. 3), relates to the redox
environment of the euglenozoan mitochondrial inter-
membrane space (IMS) (the location of the cyto-
chrome c). Loss of one cysteine from the
cytochrome c heme-binding motif could: (a) signifi-
cantly affect the interactions between the apocyto-
chrome and other thiol proteins in the IMS; and ⁄ or
(b) prevent the formation of an undesirable intramo-
lecular disulfide bond in the apocytochrome for
which no suitable reductant would be available in the
IMS; and ⁄ or (c) provide a selective advantage by alle-
viating a constraint on other IMS redox proteins.
V. Fu
¨
lo
¨
p et al. Structure of Crithidia fasciculata cytochrome c
FEBS Journal 276 (2009) 2822–2832 ª 2009 The Authors Journal compilation ª 2009 FEBS 2827
Our structure thus adds to other recent evidence [28]
hinting that the redox environment of the mitochon-
drial IMS in trypanosomatids may be different from
that in animals and yeast.
Notably, the stereochemistry of heme attachment

refined with a nonredundant first part of the dataset,
which showed similar bond lengths, as did refinement
against a lower-resolution dataset collected at a much
less intense beamline (ESRF, BM16). Therefore, this
observation cannot be interpreted as a result of
X-ray-induced radiation damage; rather, it is an
intrinsic feature of the structure. In single thioether
cytochrome c, the heme is less constrained than in a
normal (CXXCH) c-type cytochrome, because it is
covalently anchored to the protein only once rather
than twice. This leads to greater conformational flexi-
bility of the heme, which is reflected, for example, in
broadening of the peaks in the absorption spectrum
[9,24]. Moreover, when heme is attached to a
CXXCH motif, the (quite significant) strain of con-
straining the heme position is spread over two thioe-
ther bonds plus the histidine ligand to the iron,
whereas in the euglenozoan mitochondrial cyto-
chromes, the load must be borne by only one thioe-
ther bond plus the histidine. Together, these factors
presumably lead to a weaker, and hence longer, thioe-
ther bond.
Cytochrome c maturation by other biogenesis
systems
The structure reported here is also informative in the
context of the failure of the E. coli Ccm apparatus to
effectively mature wild-type (AXXCH) T. brucei cyto-
chrome c [9]; the system can mature both a CXXCH
variant [9] and the structurally very similar (Fig. 3)
yeast (CXXCH) mitochondrial cytochrome c [33].

analogy with the Ccm system [35], is that it recognizes
little more than the CXXCH heme-binding motif. The
second is that it recognizes as yet undefined features of
the apoprotein, leading to a productive complex within
which heme is attached. Our results here suggest the
second possibility, and that the recognition features in
the apocytochrome are not related to the overall struc-
ture of the cytochrome. This complements the previous
observation that heme lyase is unable to mature a
bacterial class I c-type cytochrome, Paracoccus denitrif-
icans cytochrome c
550
[33]. Moreover, many taxa that
have heme lyase apparently have separate heme lyases
for the maturation of cytochromes c and c
1
[10]; this
has been demonstrated biochemically for S. cerevisiae,
where only very limited overlap of substrate specificity
was observed [36]. Again, ‘simple’ interaction between
Structure of Crithidia fasciculata cytochrome c V. Fu
¨
lo
¨
p et al.
2828 FEBS Journal 276 (2009) 2822–2832 ª 2009 The Authors Journal compilation ª 2009 FEBS
heme lyase and the apocytochrome CXXCH motif
would appear to be unlikely if separate heme lyases are
required to mature cytochromes c and c
1

)1
NaCl, 0.2 gÆL
)1
KCl, 1.42 gÆL
)1
Na
2
HPO
4
and 0.27 gÆL
)1
KH
2
PO
4
, pH 7.2) and stored at
)80 °C until required. Growth assays for C. fasciculata
were conducted by the addition of respiratory inhibitors as
described in the text; growth was assessed either by counts
using a hemocytometer, or by measurement of D
600 nm
val-
ues. Respiration of C. fasciculata was also investigated
using an oxygen electrode (Rank Brothers, Bottisham,
UK), which was calibrated and used according to the man-
ufacturer’s directions. Cells were placed in the electrode
chamber in their growth medium, and respiratory inhibitors
were added as required.
Purification of cytochrome c
Extracts of C. fasciculata were prepared by disrupting the

3
Fe(CN)
6
dissolved in the buffer to ensure that the C. fasciculata
cytochrome c (sometimes called cytochrome c
555
[5]) was all
oxidized. The protein was eluted from the column with a
500 mL gradient of 0–500 mm NaCl in 50 mm Tris ⁄ HCl
buffer (pH 8.0), with a flow rate of 10 mLÆmin
)1
;8mL
fractions were collected. Fractions were assessed by their
red color, and those with maximum Soret band absorbance
more than one-third that of the best fraction were retained.
The pooled fractions were diluted five-fold in 50 m m
Tris ⁄ HCl (pH 8.0), and applied to an XK26 ⁄ 20 column
containing CM-Sepharose fast-flow resin (GE Healthcare)
at room temperature. The protein was eluted as described
above. Retained fractions containing the purest cytochrome
were concentrated to a volume of $ 1.5 mL, and applied to
a Sephacryl S-200 column (2.6 cm diameter, 1 m length),
pre-equilibrated with 50 mm potassium phosphate buffer
(pH 7.0). This chromatography step was conducted at 4 °C.
The cytochrome was eluted in the same buffer at a flow
rate of 15 mLÆh
)1
; 5 mL fractions were collected and
assessed for purity by absorption spectroscopy and
SDS ⁄ PAGE. Those with A

4
)
2
SO
4
and 0.1 m Hepes
(pH 6.5); these drops were seeded with microcrystals after
equilibration for $ 48 h. Crystals grew, and were harvested,
within 1 week. Crystals were then picked up from the
mother liquor containing 15% glycerol using a cryoloop,
placed in a nitrogen stream at 100 K, and stored in liquid
nitrogen until data collection. Initial diffraction data were
collected at beamline BM16 (European Synchrotron Radia-
tion Facility), but the final dataset used for structure deter-
mination and refinement was collected at the Diamond
Light Source, UK. Integration and scaling were performed
using denzo and scalepack [37]. Subsequent data handling
was carried out using the ccp4 software package [38].
Molecular replacement was carried out using the coordi-
nates of S. cerevisiae iso-1-cytochrome c (Protein Data
Bank code: 1YCC) as a search model with the phaser pro-
gram [39]. Refinement of the structure was carried out by
alternate cycles of refmac [40], using noncrystallographic
symmetry restraints and manual rebuilding in o [41]. Water
molecules were added to the atomic model automatically
by arp ⁄ warp [42], and in the last steps of refinement all
the noncrystallographic symmetry restraints were released.
V. Fu
¨
lo

combination of heme lyase and cytochrome. The E. coli
periplasmic fraction was prepared as previously described
[46], and discarded. The spheroplast pellet was resuspended
by vigorous vortexing in 50 mm Tris ⁄ HCl plus 150 mm
NaCl (pH 7.3), and broken by six freeze–thaw cycles (at
)78 and 37 °C); this was followed by centrifugation at
25 000 g for 1 h to remove the cell debris. The soluble cyto-
plasmic fraction was initially assayed by running the pro-
teins on SDS ⁄ PAGE gels that were stained for proteins
containing covalently bound heme [47]. Subsequently, the
extracts from multiple cultures were pooled and applied to a
5 mL Hi-Trap column containing SP-Sepharose (GE
Healthcare). The bound protein was batch eluted using
500 mm NaCl, concentrated, and then assessed using
absorption spectroscopy and heme-stained SDS ⁄ PAGE gels.
Western blotting was performed using a polyclonal
primary antibody raised against purified, recombinant, Ccm
system-matured CXXCH variant T. brucei holocyto-
chrome c (protein as described in [9]; antibody raised by
Covalab, Villeurbanne, France). Unconcentrated E. coli
soluble cytoplasmic extracts were resolved by SDS ⁄ PAGE
and blotted onto Hybond-C Extra nitrocellulose membrane
(GE Healthcare). The membrane was blocked for 1 h in 5%
(w ⁄ v) milk powder dissolved in NaCl ⁄ Tris [50 mm Tris ⁄ HCl,
pH 7.5, 120 mm NaCl, 1% (v ⁄ v) Tween-20]. It was then
incubated for 1 h with primary antibody diluted 200-fold in
10 mL of 5% milk ⁄ NaCl ⁄ Tris solution; the primary anti-
body was used as crude (unpurified) serum. The membrane
was washed four times (1 · 15 min, 3 · 5 min) in 10 mL of
NaCl ⁄ Tris, and then incubated with the secondary antibody

ij


.
P
j
P
h
hI
h
i, where
I
h,j
is the jth observation of reflection h, and <I
h
> is the mean inten-
sity of that reflection. R
cryst
¼
P
F
obs
jj
À F
calc
jjjj=
P
F
obs
j

Unique reflections 60 049
I ⁄ r(I) 14.8 (2.0)
R
sym
0.118 (0.693)
Completeness (%) 97.6 (100.0)
Refinement
Nonhydrogen atoms 3199 (including three c-type
hemes, seven sulfates and
563 waters)
R
cryst
0.210 (0.291)
Reflections used 57 638 (4079)
R
free
0.247 (0.320)
Reflections used 2411 (171)
R
cryst
(all data) 0.212
Average temperature
factor (A
˚
2
)
24.4
Protein 21.3
Hemes 15.4
Solvent 39.4

2 Stevens JM, Daltrop O, Allen JWA & Ferguson SJ
(2004) C-type cytochrome formation; chemical and
biological enigmas. Acc Chem Res 37, 999–1007.
3 Allen JWA, Daltrop O, Stevens JM & Ferguson SJ
(2003) C-type cytochromes: diverse structures and bio-
genesis systems pose evolutionary problems. Philos
Trans R Soc Lond B Biol Sci 358, 255–266.
4 Kranz R, Lill R, Goldman B, Bonnard G & Merchant S
(1998) Molecular mechanisms of cytochrome c biogenesis:
three distinct systems. Mol Microbiol 29, 383–396.
5 Hill GC & Pettigrew GW (1975) Evidence for the
amino-acid sequence of Crithidia fasciculata cytochrome
c
555
. Eur J Biochem 57, 265–271.
6 Priest JW & Hajduk SL (1992) Cytochrome c reductase
purified from Crithidia fasciculata contains an atypical
cytochrome c
1
. J Biol Chem 267, 20188–20195.
7 Pettigrew GW, Leaver JL, Meyer TE & Ryle AP
(1975) Purification, properties and amino acid
sequence of atypical cytochrome c from two protozoa,
Euglena gracilis and Crithidia oncopelti. Biochem J
147, 291–302.
8 Pettigrew GW, Aviram I & Schejter A (1975) Physico-
chemical properties of two atypical cytochromes c,
Crithidia cytochrome c
557
and Euglena cytochrome c

Eukaryot Cell 6, 1693–1696.
15 Guerra DG, Decottignies A, Bakker BM & Michels PA
(2006) The mitochondrial FAD-dependent glycerol-3-
phosphate dehydrogenase of Trypanosomatidae and the
glycosomal redox balance of insect stages of Trypanoso-
ma brucei and Leishmania spp. Mol Biochem Parasitol
149, 155–169.
16 van Hellemond JJ, Simons B, Millenaar FF &
Tielens AG (1998) A gene encoding the plant-like
alternative oxidase is present in Phytomonas but
absent in Leishmania spp. J Eukaryot Microbiol 45,
426–430.
17 Chaudhuri M, Ord RD & Hill GC (2006) Trypanosome
alternative oxidase: from molecule to function. Trends
Parasitol 22, 484–491.
18 Lamour N, Riviere L, Coustou V, Coombs GH, Barrett
MP et al. (2005) Proline metabolism in procyclic Try-
panosoma brucei is down-regulated in the presence of
glucose. J Biol Chem 280, 11902–11910.
19 Helfert S, Estevez AM, Bakker B, Michels P & Clayton
C (2001) Roles of triosephosphate isomerase and aero-
bic metabolism in Trypanosoma brucei. Biochem J 357,
117–125.
20 Moore GR & Pettigrew GW (1990) Cytochromes c:
Evolutionary, Structural, and Physicochemical Aspects .
Springer-Verlag, New York, NY.
21 Keller RM, Picot D & Wuthrich K (1979) Individual
assignments of the heme resonances in the 360 MHz
1
H

B & Hamel PP (2005) Cyc2p, a membrane-bound
flavoprotein involved in the maturation of mitochon-
drial c-type cytochromes. J Biol Chem 280, 39852–
39859.
28 Allen JWA, Ferguson SJ & Ginger ML (2008) Distinc-
tive biochemistry in the trypanosome mitochondrial
intermembrane space suggests a model for stepwise
evolution of the MIA pathway for import of cysteine-
rich proteins. FEBS Lett 582, 2817–2825.
29 Yamamoto Y & La Mar GN (1986)
1
H NMR study of
dynamics and thermodynamics of heme rotational
disorder in native and reconstituted hemoglobin A.
Biochemistry 25, 5288–5297.
30 La Mar GN, Toi H & Krishnamoorthi R (1984) Proton
NMR investigation of the rate and mechanism of heme
rotation in sperm whale myoglobin: evidence for intra-
molecular reorientation about a heme two-fold axis.
J Am Chem Soc 106, 6395–6401.
31 McLachlan SJ, La Mar GN, Burns PD, Smith KM &
Langry KC (1986)
1
H-NMR assignments and the
dynamics of interconversion of the isomeric forms of
cytochrome b
5
in solution. Biochim Biophys Acta 874,
274–284.
32 Schulz H, Hennecke H & Tho

40 Murshudov GN, Vagin AA, Lebedev A, Wilson KS &
Dodson EJ (1999) Efficient anisotropic refinement of
macromolecular structures using FFT. Acta Crystallogr
55, 247–255.
41 Jones TA, Zou JY, Cowan SW & Kjeldgaard M (1991)
Improved methods for building protein models in elec-
tron density maps and the location of errors in these
models. Acta Crystallogr 47, 110–119.
42 Perrakis A, Morris R & Lamzin VS (1999) Automated
protein model building combined with iterative struc-
ture refinement. Nat Struct Biol 6, 458–463.
43 Kraulis PJ (1991) MolScript: a program to produce
both detailed and schematic plots of protein structures.
J Appl Crystallogr 24 , 946–950.
44 Esnouf RM (1997) An extensively modified version of
MolScript that includes greatly enhanced coloring capa-
bilities. J Mol Graph 15, 132–134.
45 Merritt EA & Murphy ME (1994) Raster3D Version
2.0. A program for photorealistic molecular graphics.
Acta Crystallogr 50, 869–873.
46 Allen JWA, Tomlinson EJ, Hong L & Ferguson SJ
(2002) The Escherichia coli cytochrome c maturation
(Ccm) system does not detectably attach heme to single
cysteine variants of an apocytochrome c. J Biol Chem
277, 33559–33563.
47 Goodhew CF, Brown KR & Pettigrew GW (1986) Haem
staining in gels, a useful tool in the study of bacterial
c-type cytochromes. Biochim Biophys Acta 852, 288–294.
48 Bru
¨