Microbiology (2005), 151, 233–242
DOI 10.1099/mic.0.27573-0
A sulphite respiration system in the
chemoheterotrophic human pathogen
Campylobacter jejuni
Jonathan D. Myers and David J. Kelly
Department of Molecular Biology and Biotechnology, University of Sheffield, Firth Court,
Western Bank, Sheffield S10 2TN, UK
Correspondence
David J. Kelly
[email protected]
Received 19 August 2004
Revised
7 October 2004
Accepted 11 September 2004
The ability to use sulphite as a respiratory electron donor is usually associated with
free-living chemolithotrophic sulphur-oxidizing bacteria. However, this paper shows that the
chemoheterotrophic human pathogen Campylobacter jejuni has the ability to respire sulphite,
with oxygen uptake rates of 23±8 and 28±15 nmol O2 min”1 (mg cell protein)”1 after the
addition of 0?5 mM sodium sulphite or metabisulphite, respectively, to intact cells. The
C. jejuni NCTC 11168 Cj0004c and Cj0005c genes encode a monohaem cytochrome c
and molybdopterin oxidoreductase, respectively, homologous to the sulphite : cytochrome c
oxidoreductase (SOR) of Starkeya novella. Western blots of C. jejuni periplasm probed with
a SorA antibody demonstrated cross-reaction of a 45 kDa band, consistent with the size of
Cj0005. The Cj0004c gene was inactivated by insertion of a kanamycin-resistance cassette.
The resulting mutant showed wild-type rates of formate-dependent respiration but was unable to
respire with sulphite or metabisulphite as electron donors. 2-Heptyl-4-hydroxyquinoline-N-oxide
(HQNO), a cytochrome bc1 complex inhibitor, did not affect sulphite respiration at concentrations
up to 25 mM, whereas formate respiration (which occurs partly via a bc1 dependent route) was
inhibited 50 %, thus suggesting that electrons from sulphite enter the respiratory chain after
the bc1 complex at the level of cytochrome c. Periplasmic extracts of wild-type C. jejuni 11168
showed a symmetrical absorption peak at 552 nm after the addition of sulphite, demonstrating
the reduction of cytochrome c. No cytochrome c reduction was observed after addition of
sulphite to periplasmic extracts of the Cj0004c mutant. A fractionation study confirmed that the
majority of the SOR activity is located in the periplasm in C. jejuni, and this activity was partially
purified by ion-exchange chromatography. The presence of a sulphite respiration system in
C. jejuni is another example of the surprising diversity of the electron-transport chain in this
small-genome pathogen. Sulphite respiration may be of importance for survival in environmental
microaerobic niches and some foods, and may also provide a detoxification mechanism for this
normally growth-inhibitory compound.
INTRODUCTION
Campylobacter jejuni is a Gram-negative, spiral-shaped
bacterium that causes acute diarrhoeal illness, and is now
established as the leading cause of acute bacterial gastroenteritis in both industrialised and developing countries
(Friedman et al., 2000). The most common route of
infection is by ingestion of contaminated food, milk or
water, with poultry being a particularly common source
of contamination (Friedman et al., 2000), as C. jejuni is a
commensal of the gastrointestinal tract of many species of
birds. Acute symptoms of C. jejuni infection in humans
include diarrhoea, fever and abdominal pain. Complications arising after infection are rare although there is
Abbreviation: SOR, sulphite : cytochrome c oxidoreductase.
0002-7573 G 2005 SGM
association with reactive arthritis and certain neurological
disorders such as Miller–Fisher and Guillain–Barré syndromes (Nachamkin et al., 2000; Skirrow & Blaser, 2000).
Several factors contribute to the pathogenicity of C. jejuni,
including environmental survival, adhesion to host mucosa,
the ability to invade host cells and the production of several
toxins (Pickett, 2000). Until recently, many aspects of the
physiology of this food-borne pathogen were poorly defined.
The publication of the genome sequence (Parkhill et al.,
2000) has stimulated a renewed interest in studying the
physiological characteristics of C. jejuni, which are crucial
to understanding its mechanisms of pathogenicity and
ability to survive in the environment.
C. jejuni is an obligate microaerophile which appears to
have a surprisingly complex and highly branched respiratory
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 00:45:15
Printed in Great Britain
233
J. D. Myers and D. J. Kelly
chain for a relatively small-genome pathogen, allowing the
use of a wide variety of electron donors such as formate,
hydrogen, D-lactate, succinate, malate and NAD(P)H, and
alternative electron acceptors to oxygen, including fumarate, nitrate, nitrite, N- or S- oxides and hydrogen peroxide
(Kelly, 2001; Sellars et al., 2002; Myers & Kelly, 2005). The
addition of sulphite to intact cells of C. jejuni was found by
Hoffman & Goodman (1982) to produce a proton pulse
(H+/O ratio=1?58), and sulphite was thus proposed to
serve as a potential electron donor. However, the nature of
the sulphite oxidation system was not elucidated and no
further studies have been carried out. The potential presence of a sulphite respiration system in C. jejuni may be
of importance in survival of this pathogenic organism in
microaerobic aquatic niches, rich in sulphite, and in some
foods, as sulphite is commonly added to processed foods
as a preservative. Interestingly, sulphite is also released in
the human body by neutrophils in response to stimulation
with lipopolysaccharide as part of the host defence, as
sulphite is a well-known antimicrobial (Mitsuhashi et al.,
1998).
Sulphite oxidation can support chemolithotrophic and
phototrophic growth in a diverse range of bacteria and
archaea (Wood, 1988; Sorokin, 1995; Friedrich, 1998)
and is known to occur either by direct oxidation, usually
utilizing a molybdenum-containing sulphite : cytochrome
c oxidoreductase (SOR) or by indirect, AMP-dependent
oxidation via the intermediate adenylylsulphate (Wood,
1988; Kappler & Dahl, 2001). In this study, we demonstrate
the ability of the chemoheterotrophic pathogen C. jejuni
to utilize sulphite and metabisulphite as respiratory electron donors, identify the proteins constituting the sulphite
oxidase and elucidate the pathway for electron transport
from sulphite to oxygen.
METHODS
Bacterial strains, media and culture conditions. Cultures of C.
jejuni strain NCTC 11168 were routinely grown at 37 uC in microaerobic conditions [5 % (v/v) O2, 10 % (v/v) CO2 and 85 % (v/v) N2]
in a MACS-VA500 Incubator (Don Whitley Scientific) on Columbia
agar containing 5 % (v/v) lysed horse blood and 10 mg ml21 each of
amphotericin B and vancomycin. Cells were subcultured onto fresh
medium every 2–3 days to maintain actively dividing cells. Liquid
cultures were grown in brain heart infusion (BHI) broth supplemented with 5 % (v/v) fetal calf serum (BHI-FCS) and the above
antibiotics at 37 uC. Cultures were maintained in the MACS-VA500
Incubator with continuous orbital shaking at 200 r.p.m.
Insertional inactivation of Cj0004c. PCR primers (forward
primer 59-TGAAGGTATAGGAATGAATG-39, reverse primer 59GGAGTTCCTATTTCTAAAGC-39) were designed to amplify a
1?6 kb product containing the Cj0004c gene, which was cloned into
pGEM T-easy (Promega) and insertionally inactivated at a unique
EcoNI site using the aphAIII (kanamycin resistance) gene containing
its own promoter, derived from pJMK30 (van Vliet et al., 1998),
producing plasmid pJDM4. As the downstream gene (gyrB) is transcribed in the opposite orientation to Cj0004c, there is no polarity
effect of this insertion. Plasmid pJDM4 was then electroporated into
C. jejuni 11168, with selection on blood agar plates containing
234
30 mg kanamycin ml21. The above primers were used in a PCR
using genomic DNA from the wild-type 11168 strain or from a
kanamycin-resistant transformant, in order to verify the correct
mutant construction.
Measurement of respiration rates by oxygen uptake. Sub-
strate oxidation was determined as described previously (Hughes
et al., 1998), by measuring the change in dissolved oxygen concentration of cell suspensions in a Clark-type oxygen electrode
linked to a chart recorder and calibrated using air-saturated 25 mM
phosphate buffer (pH 7?5) (220 nmol dissolved O2 ml21 at 37 uC).
A zero-oxygen baseline was determined by the addition of sodium
dithionite. The cell suspension was maintained at 37 uC and stirred
at a constant rate. Substrates and inhibitors were added by injection
through a fine central pore in the airtight plug. Rates were expressed
in nmol O2 utilized min21 (mg cell protein)21.
Analysis of cytochrome spectra, and cytochrome c quantification. Cytochrome spectra were obtained at room temperature
using a Shimadzu UV-2101PC double-beam scanning spectrophotometer. Spectra were scanned from 400 nm to 700 nm. Reduced
minus oxidized spectra were obtained by adding a few grains of
sodium dithionite or ammonium persulphate as reductant or oxidant, respectively (Jones & Poole, 1985). Reduction of cytochromes
by physiological substrates was performed in a sealed cuvette with
addition of substrate by injection through a seal in the lid. These
spectra were scanned against an ammonium-persulphate-oxidized
baseline. The amount of cytochrome c in cell fractions was quantified by the increase in absorbance at 550 nm after reduction of the
sample by excess sodium dithionite, using an absorption coefficient
of 20 mM21 cm21.
Preparation of periplasmic proteins from C. jejuni. Cells were
grown in the microaerobic cabinet at 37 uC in 200 ml BHI-FCS
broth overnight. The cell suspension was centrifuged (15 000 g,
20 min at 4 uC) and the resulting pellet was resuspended in 10 ml
20 % (w/v) sucrose, 30 mM Tris/HCl pH 8 at room temperature.
EDTA was added to a final concentration of 1 mM and the suspension was poured into a 100 ml conical flask and stirred at 180 r.p.m.
in a 25 uC constant-temperature room for 10 min. The suspension
was then centrifuged (10 000 g, 10 min at 4 uC) and the pellet was
resuspended in ice-cold 10 mM Tris/HCl pH 8 to a volume of
10 ml and stirred at 180 r.p.m. in a 4 uC constant-temperature room
for 10 min. The suspension was then centrifuged again (18 000 g,
15 min at 4 uC) and the supernatant collected as the periplasmic
fraction. The pellet was also used for further fractionation of cytoplasm and membrane as described below.
Preparation of C. jejuni cell-free extract. A 200 ml C. jejuni
culture in BHI-FCS was grown for 16 h at 37 uC with shaking in a
microarobic atmosphere. The cells were harvested by centrifugation
(10 000 g, 15 min, 4 uC) and resuspended in 5 ml 10 mM Tris/HCl
pH 8. The cell suspension was sonicated (20 kHz, 6 mm amplitude)
and centrifuged (20 000 g, 20 min, 4 uC) to remove cell debris.
Localization of SOR activity. A 1-litre culture of C. jejuni was
grown overnight in BHI-FCS and harvested by centrifugation
(10 000 g, 15 min, 4 uC). The whole cell pellet was resuspended in
60 ml 20 % (w/v) sucrose, 30 mM Tris/HCl pH 8 and subjected to
the osmotic-shock procedure described above, but with a resuspension volume of 60 ml. After decanting the periplasmic fraction, the
pellet was resuspended in 5 ml 10 mM Tris/HCl pH 8 and sonicated
(20 kHz, 6 mm amplitude) to release the cell contents. This suspension was subjected to low-speed centrifugation (20 000 g, 20 min,
4 uC) to remove cell wall debris and then the supernatant was
ultracentrifuged (100 000 g, 80 min, 4 uC) to separate membrane
and cytoplasmic fractions. The membrane pellet was resuspended in
3 ml 10 mM Tris/HCl pH 8.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 00:45:15
Microbiology 151
Sulphite respiration in Campylobacter jejuni
Enzyme assays. SOR activity was assayed by a method based on
that described by Kappler et al. (2000), using a Shimadzu UV2101PC double-beam scanning spectrophotometer at a wavelength
of 550 nm. Rates were obtained by measuring the increase in absorbance of cytochrome c after adding 0?5 mM sodium sulphite to a
cuvette containing the following: 900 ml 10 mM Tris/HCl pH 8,
100 ml 10 mg horse heart cytochrome c ml21, 10–100 ml cell fraction. An absorption coefficient of 20 mM21 cm21 at 550 nm was
used.
Isocitrate dehydrogenase activity was assayed according to Leyland &
Kelly (1991) in a Shimadzu UV-2101PC spectrophotometer at a
wavelength of 340 nm. Rates were obtained by adding 100 ml 50 mM
sodium isocitrate to a cuvette containing the following: 100 ml 50 mM
Tris/HCl pH 8; 100 ml 10 mM NADP; 100 ml 10 mM MgCl2;
500–590 ml distilled H2O; 10–100 ml cell fraction. The specific activity
of isocitrate dehydrogenase was calculated using an absorption
coefficient of NADPH of 6?22 mM21 cm21 at 340 nm.
Partial purification of SOR activity. Ion-exchange chromatogra-
phy was utilized for partial purification of the periplasm. A column
packed with 30 ml DEAE-Sepharose (Pharmacia) was connected to
a Bio-Rad Biologic HP system. A 50 ml sample of periplasm was
loaded onto the anion-exchange column and equilibrated with
50 mM Tris/HCl pH 8?0. Proteins were eluted from the column by
a linear gradient of 0 to 0?5 M NaCl. Samples of the protein fractions obtained were assayed for SOR activity using the horse heart
cytochrome c assay described above. The SOR-containing fractions
were electrophoresed using SDS-PAGE and stained for protein with
Coomassie blue or blotted onto PVDF membrane for Western blot
analysis as described below.
SDS-PAGE and Western blotting. SDS-PAGE was carried out
according to the method of Laemmli (1970) and gels were stained
for protein using Coomassie brilliant blue R. For Western blots,
periplasmic extracts were prepared as described above and denatured
by heating (100 uC, 5 min) in the presence of SDS sample buffer.
The protein was loaded and separated by SDS-PAGE (12 %, w/v,
acrylamide), then electroblotted onto a PVDF membrane (Bio-Rad)
in a Bio-Rad mini PROTEAN II cell at 11 mA for 16 h. Immunodetection was performed using an ECL detection kit (Amersham
Biosciences) according to the instructions of the manufacturer. The
primary antiserum for detection of Cj0005 was kindly donated by
Christiane Dahl (Institut fur Mikrobiologie und Biotechnologie,
Bonn, Germany). The antiserum was prepared in rabbit and was
raised against SorA of Starkeya novella, as described by Kappler et al.
(2000). The working concentration for this serum was a 1 : 10 000
dilution. An anti-rabbit-horseradish peroxidase conjugate (Amersham
Biosciences) was used as a secondary antibody (1 : 2000 dilution).
Sequence alignment and phylogenetic analysis. Multiple pro-
tein sequence alignments and phylogenetic analyses were performed
using CLUSTAL X (Thompson et al., 1997) with output generated by
ESPript (Gouet et al., 1999). Phyologenetic trees were viewed in
TreeView (Page, 1996).
Determination of protein concentration. This was done by the
Lowry method.
RESULTS
Oxygen-linked respiration of formate, sulphite
and metabisulphite by C. jejuni
C. jejuni has been predicted to utilize several types of
electron donor to support respiration (Myers & Kelly,
2005). Fig. 1(a) shows the ability of C. jejuni to use formate
http://mic.sgmjournals.org
as a respiratory electron donor. Addition of 5 mM sodium
formate to late-exponential-phase wild-type 11168 cells produced a mean oxygen respiration rate of 478±118 nmol
O2 min21 mg21 (mean±SD, n=3 biological replicates).
Fig. 1(b) shows the ability of C. jejuni to respire using
sulphite as an electron donor. Addition of 0?5 mM sodium
sulphite to C. jejuni cells produced a mean oxygen respiration rate of 23±8 nmol O2 min21 mg21 (n=3). Concentrations above 0?5 mM sulphite were found to produce a
background rate of oxygen consumption in the absence
of cells due to chemical oxidation of sulphite; 0?5 mM
sulphite was tested without cells and shown to produce no
residual oxygen consumption. Fig. 1(c) shows the ability of
C. jejuni to respire using 0?5 mM sodium metabisulphite
(Na2S2O5), which produced an oxygen respiration rate of
28±15 nmol O2 min21 mg21 (n=3). The actual electron
donor here is likely to be the bisulphite anion HSO{
3 , which
is produced by reaction of metabisulphite with water.
Identification of a putative SOR in C. jejuni
Examination of the C. jejuni NCTC 11168 genome sequence
(Parkhill et al., 2000) for genes encoding potential components of a sulphite respiration system revealed a putative
periplasmic molybdoenzyme encoded by Cj0005c with a
molecular mass of 46 kDa and an associated 13 kDa periplasmic protein containing a c-type haem-binding motif
(CXXCH) encoded by Cj0004c, as the most likely candidates. Cj0005 shows 32 % sequence identity to the wellcharacterized SorA subunit found in the chemolithotroph
Starkeya novella (formerly Thiobacillus novellus). The sorAB
genes of St. novella encode a periplasmic ab heterodimeric
SOR, consisting of a 40?6 kDa subunit containing a
molybdopterin cofactor (SorA), where sulphite oxidation
takes place, and an 8?8 kDa mono-haem cytochrome c552
(SorB) subunit which acts as a specific electron acceptor
(Kappler et al., 2000, 2001). A relationship between Cj0005
and SorA was previously noted by Kappler & Dahl (2001).
However, Cj0005 also shows significant amino-acid
identity to several uncharacterized molybdoproteins in
chemoheterotrophs, including 46 % amino-acid identity
to SO0715, a putative 45 kDa molybdoprotein containing
a Mo-Co dimerization domain encoded in Shewanella
oneidensis, and 45 % amino-acid identity to VPA1143, a
putative 48 kDa molybdoprotein encoded in the pathogen
Vibrio parahaemolyticus. The close phylogenetic relationship of C. jejuni Cj0005 to SorA of St. novella and other
homologues in Sh. oneidensis and V. parahaemolyticus is
shown in the phylogenetic tree in Fig. 2(a), which demonstrates the relatedness between a broad range of known
molybdoenzymes including examples from the nitrate and
DMSO reductase families from a number of different
prokaryotes. The putative SOR proteins, including Cj0005,
form a well-defined group, which clusters with St. novella
SorA. Interestingly, C. jejuni also encodes an additional
molybdoenzyme (Cj0379) which is distantly related to this
group. Fig. 2(b) shows the gene arrangement around the
putative sorA genes. Each of the genes encoding the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 00:45:15
235
J. D. Myers and D. J. Kelly
Fig. 1. (a–c) Oxygen consumption catalysed
by wild-type 11168 and Cj0004c mutant
cells. Typical oxygen consumption rates at
37 6C of late-exponential-phase BHI-brothgrown wild-type 11168 (WT) and Cj0004c
mutant (JDM4) (each 1 mg cell protein),
after the addition of the following electron
donors: (a) 5 mM sodium formate; (b)
0?5 mM sodium sulphite, (c) 0?5 mM sodium
metabisulphite. (d) Inhibition of wild-type
11168 sulphite- and formate-stimulated
oxygen respiration rates at 37 6C (2 mg cell
protein), after the addition of 2-heptyl-4hydroxyquinoline N-oxide (HQNO), a bc1
complex inhibitor. n, Percentage of oxygen
respiration when stimulated by 5 mM formate; &, percentage of oxygen respiration
when stimulated by 0?5 mM sodium sulphite.
For this experiment, cells were obtained
from blood agar plates and resuspended in
25 mM phosphate buffer pH 8. Respiration
rates with formate and sulphite were lower
than with broth-grown cells; 100 % values
were 13?4 and 5 nmol O2 min”1 mg”1
respectively.
homologues appears to be part of an operon. A single
11?5 kDa (pre Sec-cleavage) cytochrome c (SorB) or
putative 13 kDa cytochrome c (Cj0004) subunit is encoded
downstream of the SorA/Cj0005-molybdoproteins of St.
novella and C. jejuni respectively, while a putative 24 kDa
dihaem cytochrome c (VPA1144) is encoded downstream
of the V. parahaemolyticus SOR-molybdoprotein, and two
separate 14 and 11 kDa cytochrome c subunits (SO0714
and S00716) are encoded upstream and downstream of
the SOR-molybdoprotein of Sh. oneidensis. Fig. 2(c) shows
a multiple sequence alignment of the SorA-downstream
associated cytochrome subunits in these bacteria. As
expected there is a high degree of amino acid conservation
in the CXXCH motif region of these cytochromes, but
while SO0716 and the C-terminal half of VPA1144 show
34 % amino-acid sequence identity to Cj0004 over 92 and
79 residues, respectively, Cj0004 is only 23 % identical to
SorB of St. novella. These comparisons indicate that the
c-type cytochromes that act as the specific electron acceptor
for SorA are much less conserved than is SorA itself.
A Cj0004c mutant is specifically deficient in
sulphite and metabisulphite respiration
In order to determine if the molybdoprotein encoded by
Cj0005c was responsible for the SOR activity in C. jejuni, an
attempt was made to construct a mutant strain. A plasmid
236
vector containing a kanamycin-resistance cassette flanked
by DNA from the Cj0005c ORF was constructed, transformed by electroporation into C. jejuni, and kanamycinresistant transformants were isolated. However, PCR
analysis showed that, for unknown reasons, the transformants were wild-type for the Cj0005c gene, showing the
mutagenesis to be unsuccessful. The Cj0004c gene was
therefore cloned and inactivated by the insertion of the
kanamycin-resistance cassette at a unique EcoNI site
(Fig. 3a). Electroporation of this plasmid into C. jejuni
resulted in kanamycin-resistant colonies, which were confirmed by PCR to be the result of a successful allelic
exchange, verified by the 1?5 kb increase in band size of
the PCR product when using the same primers as for the
cloning of the Cj0004c gene (Fig. 3b). One colony was
picked and the strain designated JDM4. SDS-PAGE and
Western blotting of periplasmic fractions using antibody
raised against the SorA subunit of St. novella demonstrated
the presence of a cross-reacting protein in the region of
the 46 kDa molecular mass of the putative oxidoreductase,
Cj0005 (Fig. 3c). The Cj0004c mutant was shown to be still
expressing this protein, as would be expected (Fig. 3c).
However, SDS-PAGE followed by a haem stain to detect ctype cytochromes showed the absence of a low-abundance
13 kDa haem-staining band in the periplasm of the mutant
strain that was present in the wild-type parent (Fig. 3d),
consistent with the expected loss of the Cj0004 protein.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 00:45:15
Microbiology 151
Sulphite respiration in Campylobacter jejuni
Fig. 2. Sequence similarities between SOR proteins. (a) Phylogenetic tree of molybdoenzymes. Those shown outside boxes
or ellipses are a selection of characterized molybdoenzymes. Rectangular boxes contain C. jejuni molybdoproteins. Cj0005 is
closely related to SorA, SO0715 and VPA1143 subunits (in ellipses) of St. novella, Sh. oneidensis and V. parahaemolyticus,
respectively. Cj0379 is a more distantly related protein to this group. Abbreviations: DorA, dimethyl sulphoxide reductase;
TorA, trimethylamine-N-oxide reductase; BisC, biotin sulphoxide reductase; NasA, assimilatory nitrate reductase; NapA,
periplasmic nitrate reductase; FdhA, formate dehydrogenase; PhsA, thiosulphate reductase; PsrA, polysulphide reductase;
DmsA, dimethyl sulphoxide reductase; NarG, membrane-bound respiratory nitrate reductase; EbdA, ethylbenzene
dehydrogenase; SerA, selenate reductase; DdhA, dimethyl sulphide dehydrogenase; SorA, sulphite oxidase. SN, Starkeya
novella; WS, Wollinella succinogenes; PP, Pseudomonas putida; EC, Escherichia coli; HI, Haemophilus influenzae; BS,
Bacillus subtilis; TS, Thauera selenatis; RS, Rhodovulum sulfidophilum; AS, Azoarcus sp. EB1; RC, Rhodobacter capsulatus;
ST, Salmonella typhimurium; Cj, Campylobacter jejuni; SO, Shewanella oneidensis; VP, Vibrio parahaemolyticus. The scale
bar represents 0?1 changes per amino acid. (b) Genome arrangements of SOR genes in C. jejuni, V. parahaemolyticus, Sh.
oneidensis and St. novella. Black genes encode SorA homologues. Striped genes encode cytochrome c subunits. (c) Protein
sequence alignment of the cytochromes c encoded downstream of the SorA homologues. A black box indicates identical
residue conservation; a clear box identifies conservative substitution. Multiple protein sequence alignments and phylogenetic
analyses were performed using CLUSTAL X and output generated by ESPript.
Fig. 1(a) shows that the Cj0004c mutant (JDM4) was
unaffected in its ability to respire formate, with a similar
rate to wild-type of 498±70 nmol O2 min21 mg21 (n=3).
However, the addition of sulphite (Fig. 1b) or metabisulphite (Fig. 1c) to Cj0004c mutant cells resulted in no
oxygen uptake, showing that Cj0004 is essential for sulphiteand metabisulphite-dependent respiration.
http://mic.sgmjournals.org
The electron transport pathway from sulphite to
oxygen
Fig. 1(d) shows the rate of oxygen respiration in intact
cells of 11168 with either 5 mM formate or 0?5 mM
sodium sulphite as the electron donor, after the addition
of increasing concentrations of the cytochrome bc1 complex
inhibitor 2-heptyl-4-hydroxyquinoline-N-oxide (HQNO).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 00:45:15
237
J. D. Myers and D. J. Kelly
Fig. 3. Inactivation of Cj0004c. (a) Strategy for insertional inactivation of Cj0004c. Primers were used to PCR-amplify a
1?6 kb product containing the gene, which was cloned into pGEM T-easy and insertionally inactivated at a unique EcoNI site,
using an aphAIII cassette containing its own promoter. Plasmid pJDM4 was then electroporated into C. jejuni 11168, with
selection for kanamycin resistance. (b) Verification of mutant construction. Cj0004c specific primers were used in a PCR
using genomic DNA templates from the wild-type 11168 strain (WT) or from a kanamycin resistant transformant (M). Agarose
gel electrophoresis showed a single product, which was 1?7 kb larger in the mutant strain due to the aphAIII insertion.
(c) Coomassie stain and Western blot of a 12?5 % SDS-PAGE gel containing periplasm of wild-type (WT) and Mutant JDM4
(M) using antibody raised against SorA of St. novella. The arrow indicates the 46 kDa large subunit, Cj0005, which crossreacts with the SorA antibody. (d) Coomassie and haem stain of a 15 % SDS-PAGE gel to show the 13 kDa haem-binding
subunit, Cj0004, in the periplasm of wild-type 11168 (WT) and its absence in mutant JDM4 (M). The haem-stained band of
the 13 kDa small subunit is indicated by the arrow.
Concentrations of HQNO of 20–25 mM resulted in ¢50 %
inhibition of oxygen respiration with formate as the electron donor. At the same concentrations of HQNO, the
oxygen respiration rate was not inhibited with sulphite as
the electron donor. The more specific cytochrome bc1
complex inhibitors antimycin A and myxothiazol could
not be used because even at high concentration they did
not inhibit the formate respiration of C. jejuni. This is
probably due to a problem with membrane permeability.
Fig. 4(a) shows the change in absorbance in the 550 nm
region of the visible spectrum after the addition of 0?5 mM
sodium sulphite to a periplasmic extract of wild-type cells.
A symmetrical peak at 552 nm was observed, indicating
the reduction of cytochrome c in the extract. Full reduction by dithionite resulted in a larger symmetrical peak.
Fig. 4(b) shows the same experiment with the periplasm of
the Cj0004c mutant. No change in absorbance was found in
238
the 550 nm region after the addition of 0?5 mM sulphite,
indicating that cytochrome c was not reduced by sulphite
in this strain. The data in Figs 1 and 4 suggest that electrons from sulphite feed directly into the respiratory chain
from Cj0004 at the level of cytochrome c, bypassing the bc1
complex.
Induction, localization and partial purification of
SOR activity
Growth of C. jejuni 11168 in BHI broth containing
5 mM sodium sulphite increased the rate of SOR activity
(measured as sulphite : horse heart cytochrome c reductase
activity) in late-exponential-phase cell-free extracts approximately twofold, from 142±5 nmol cyt c min21 mg21
(mean±SD, n=3) to 290±31 nmol cyt c min21 mg21
(n=3), indicating partial inducibility of the enzyme. The
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 00:45:15
Microbiology 151
Sulphite respiration in Campylobacter jejuni
Fig. 4. Cytochrome spectra of a periplasmic extract (500 mg
protein) from wild-type 11168 (a) and mutant JDM4 (b). Solid
lines, persulphate oxidized minus 0?5 mM sodium sulphite
reduced. Dashed lines, after full reduction of the periplasm by
further addition of sodium dithionite.
fractionation showed 51 % activity of SOR in the periplasm,
22 % in the cytoplasm and 28 % in the membrane fractions. The periplasm was contaminated with 23 % of the
cytoplasmic marker, and the cytoplasmic fraction was
contaminated with 7 % of the periplasmic marker in this
experiment. The fractionation was repeated three times on
separate cultures with similar results. Partial purification
of SOR activity from the periplasm by anion-exchange
chromatography is shown in Fig. 6(a). Fractions 13–18
were found to contain the highest level of activity, with
fraction 16 (95 mM NaCl) being the peak fraction. These
fractions were subjected to SDS-PAGE and then stained
with Coomassie blue, which revealed the 46 kDa Cj0005
protein (Fig. 6b). Fractions 13–18 were also run on SDSPAGE, blotted onto PVDF membrane and the membrane
reacted with an antibody raised against SorA of St. novella.
The antibody cross-reacted with a 46 kDa band in fractions
16, 17 and 18 (Fig. 6b), the same size as Cj0005 observed
in blots of whole periplasm (Fig. 3). Fraction 17 produced
the strongest cross-reaction on the immunoblot.
DISCUSSION
deduced sequence of the Cj0005 protein is predicted to
contain a twin-arginine motif at the N-terminus, suggesting
export to the periplasm via the TAT system. The smaller
Cj0004 protein is predicted to contain a typical Sec signal
peptide, and this is also expected to be periplasmic. A cell
fractionation was carried out to determine the cellular
location of the SOR activity (Fig. 5). The distribution of
activity was compared to that of markers for the periplasm
(dithionite-reducible total cytochrome c) and cytoplasm
(isocitrate dehydrogenase activity). The results of a typical
This study is believed to be the first characterization of a
sulphite respiration system from a chemoheterotrophic
pathogenic bacterium, rather than the typical chemolithotrophic species with which this ability is more commonly
associated (Wood, 1988). The results obtained show unequivocally that C. jejuni has the ability to respire sulphite
and imply that the proton pulse detected by Hoffman &
Goodman (1982) after the addition of sulphite to intact cells
was the result of proton translocation accompanying the
transfer of electrons from sulphite to oxygen catalysed by a
specific sulphite oxidase.
Fig. 5. Localization of SOR activity in C. jejuni cells.
Periplasm, cytoplasm and membrane fractions were assayed for
SOR activity (black bars), isocitrate dehydrogenase activity
(clear bars) and total cytochrome c concentration (striped bars)
as described in Methods. The values corresponding to 100 %
were 26 nmol (mg cell protein)”1 for cytochrome c concentration, 2124 nmol min”1 mg”1 for isocitrate dehydrogenase activity and 807 nmol min”1 mg”1 for SOR activity. Data from a
typical fractionation are shown. The experiment was repeated
three times with similar results.
Sulphite is an extremely widespread inorganic anion in
many environments, particularly low-oxygen niches in
soil or water, where it is more stable than in aerobic conditions, and is thus available for microbial oxidation or
reduction by a variety of pathways (Wood, 1988). The
possession of a sulphite respiration system may thus contribute to the survival of C. jejuni in such environments,
particularly as this bacterium has a very high-affinity cbtype cytochrome c oxidase which could allow sulphite respiration at extremely low oxygen concentrations. However,
for many chemoheterotrophic bacteria sulphite, and particularly metabisulphite (which in aqueous solutions is
rapidly converted to the bisulphite anion, HSO{
3 ), are
well-known growth-inhibitory compounds. Nevertheless,
Bolton et al. (1984) showed enhanced growth and aerotolerance of C. jejuni by addition of a mixture of ferrous
sulphate, sodium metabisulphite and sodium pyruvate
(FBP) as a growth supplement in laboratory media. FBP
is thought to function as a system for reducing oxidative stress by removing free radicals from the medium.
Our demonstration that C. jejuni can respire sulphite,
and metabisulphite provides an explanation for their
relative lack of toxicity for this bacterium and suggests a
http://mic.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 00:45:15
239
J. D. Myers and D. J. Kelly
Fig. 6. Partial purification of SOR from C.
jejuni periplasm. (a) SOR activity (n; arbitrary units) assayed by horse heart cytochrome c reduction at 550 nm, and the
relative protein concentration of fractions
measured by UV absorbance at 280 nm
(&), in fractions (6 ml) eluted by an increasing linear NaCl gradient (dashed line)
applied to a DEAE-Sepharose column
loaded with C. jejuni periplasm. (b) SDSPAGE of fractions where activity was highest, stained with Coomassie blue (left panel)
and subject to Western blot analysis with
antibody raised against SorA of St. novella
(right panel). The arrow indicates the location of the 46 kDa Cj0005.
growth-enhancing role for sulphite/bisulphite as an electron donor. Interestingly, growth of the closely related
gastric pathogen Helicobacter pylori was found to be inhibited by the addition of sulphite or metabisulphite to the
media (Jiang & Doyle, 2000) but FP growth supplement,
which did not contain metabisulphite, did enhance the
growth of H. pylori. Unlike C. jejuni, the genome of H. pylori
does not encode a homologue of the SOR system. Thus,
sulphite detoxification could be an additional function of
this type of sulphite respiration system. This might be
important in vivo, where, for example, there is evidence that
sulphite can be produced by neutrophils in response to
stimulation by lipopolysaccharide (Mitsuhashi et al., 1998).
Analysis of the Cj0004c/Cj0005c operon has demonstrated
that C. jejuni encodes a SOR enzyme for sulphite oxidation,
which is very similar to that of St. novella. The latter has
been studied in detail and shown to be heterodimeric, and
clearly distinct from the eukaryotic SOR enzymes, which
are homodimeric proteins containing haem b and with a
molybdopterin cofactor residing in each subunit (Kappler
et al., 2000). The MPT cofactor is unusual for a prokaryote,
with the majority of known bacterial molybdoenzymes
containing the dinucleotide form of the pterin cofactor
(Hille, 1996; Kisker et al., 1997, 1999). The St. novella SOR
Km value for sulphite was 27 mM and a periplasmically
located cytochrome c550 has been identified as the natural
electron acceptor, with a low Km value of 2?5 mM (Kappler
240
et al., 2000). While the corresponding SorA molybdoproteins of C. jejuni and St. novella are clearly homologous,
the SorB monohaem cytochrome subunits of the two
bacteria are not closely related, and it seems that SorA
homologues in other bacteria may also be associated with
distinct cytochrome subunits serving the same function.
Although we were unable to mutate Cj0005c, several lines of
evidence showed that the cytochrome encoded by Cj0004c
was essential for sulphite respiration. The Cj0004c mutant
possessed no detectable rate of oxygen respiration after the
addition of sulphite or metabisulphite as an electron
donor. Additionally, no cytochrome c reduction was
detected upon addition of sulphite to periplasmic extracts.
The mutation did not affect the expression of the large
molybdopterin-containing subunit, Cj0005, which is
thought to contain the active site responsible for the
oxidation of sulphite. A more complete analysis involving
attempts to remutagenize Cj0005c and complement both
genes should be carried out, although genetic tools for C.
jejuni are still somewhat limited.
SOR activity is located mainly in the periplasm of C. jejuni.
Cell fractionation of C. jejuni is not as reliable as with
‘model’ bacteria like Escherichia coli, and the smaller percentages of activity detected in other cell compartments
are probably due to an unavoidable lysis of some cells
during the osmotic shock procedure and the enzyme nonspecifically binding to the membrane fraction. A periplasmic
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 00:45:15
Microbiology 151
Sulphite respiration in Campylobacter jejuni
location would be in agreement with the predicted topology
of the subunits and the signal sequences contained on the
N-terminus of each subunit. The TAT system has been
shown to export folded or partially folded proteins with the
cofactors previously inserted in the cytoplasm (Sargent et al.,
1998; Berks et al., 2003). The smaller cytochrome c subunit
would be exported via the Sec pathway as an unfolded
polypeptide with the haem group inserted in the periplasm. SOR activity was linked to the Cj0005 protein by the
partial purification of the enzyme using anion-exchange
chromotography. The fractions containing the highest SOR
activity correlated with the presence of Cj0005, which was
detected using the St. novella SorA antibody. However, the
correlation was not exact, with the fraction of highest
activity not being the same as the fraction with the highest
concentration of Cj0005. A possible explanation would be
that a functioning Cj0004 protein must be present for the
detection of the SOR activity using the horse heart cytochrome c reduction method. The Cj0004 protein could be
eluted at a different time to the Cj0005 protein and would
result in a fraction which has the highest concentration of
Cj0005 but does not have the full SOR activity. Unfortunately, the Cj0004 protein could not be positively identified
on haem stains of column fractions.
This study provides more evidence for the highly diverse
and branched structure of the respiratory chains of this
pathogen, and it is now clear that many of the alternative
electron transport pathways employ periplasmic molybdoenzyme dehydrogenases or reductases. These include
formate dehydrogenase, nitrate reductase and DMSO/
TMAO reductase, in addition to the sulphite oxidase
(Kelly 2001; Myers & Kelly, 2005; Sellars et al., 2002). The
one remaining uncharacterized periplasmic molybdoenzyme encoded by Cj0379 is related to YedY in E. coli
(Loschi et al., 2004) and is distantly related to Cj0005,
but is not a sulphite oxidase (J.D. Myers & D. J. Kelly,
unpublished). Fig. 7(a) shows how the SOR system connects
with these other electron-transport chains in C. jejuni, and
Fig. 7(b) shows how electrons from sulphite are transferred
to oxygen via the periplasmic SOR and the membrane
bound cb-type cytochrome c oxidase. It is clear that the
ability to use such a wide range of electron donors and
electron acceptors allows considerable respiratory chain
flexibility, particularly under conditions of severe oxygen
limitation, and this could be a key factor in the organism
being able to survive and grow in a number of different
niches, possibly including human and animal hosts.
ACKNOWLEDGEMENTS
Fig. 7. Proposed mechanism of sulphite respiration in C. jejuni.
(a) Electron-transport chain of C. jejuni (modified from Sellars
et al., 2002). A range of primary dehydrogenases is present. A
cb-type cytochrome c oxidase is present in addition to a
CydAB-like quinol oxidase. There are also two putative cytochrome c peroxidases. Several alternative reductases are present, including nitrate, nitrite and a TMAO/DMSO reductase,
TorA. Solid lines indicate experimentally established or highly
likely routes of electron transport, while dotted lines indicate
uncertainty as to the exact route, possibly with the participation
of unidentified additional redox proteins. The proposed SOR
respiration system is indicated. (b) Proposed topological organization and electron-transport pathway through the SOR system
in C. jejuni NCTC 11168. The formate respiration pathway via
the cytochrome bc1 complex route is shown for comparison.
MPT, molybdopterin cofactor.
http://mic.sgmjournals.org
This work was supported by a studentship from the UK Biotechnology
and Biological Sciences Research Council to J. D. M. Simon Fraser and
Theresa Mascarenhas are acknowledged for help with characterization
of JDM4. We are grateful to Dr Christiane Dahl for supplying antibody
to SorA.
REFERENCES
Berks, B. C., Palmer, T. & Sargent, F. (2003). The Tat protein
translocation pathway and its role in microbial physiology. Adv
Microb Physiol 47, 187–254.
Bolton, F. J., Coates, D. & Hutchinson, D. N. (1984). The ability of
Campylobacter media supplements to neutralize photochemically
induced toxicity and hydrogen peroxide. J Appl Bacteriol 56,
151–157.
Friedman, C. R., Neiman, J., Wegener, H. C. & Tauxe, R. V. (2000).
Epidemiology of Campylobacter jejuni in the United States and other
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 00:45:15
241
J. D. Myers and D. J. Kelly
industrialized nations. In Campylobacter, 2nd edn, pp. 121–138.
Edited by I. Nachamkin & M. J. Blaser. Washington DC: American
Society for Microbiology.
Mitsuhashi, H., Nojima, Y., Tanaka, T., Ueki, K., Maezawa, A.,
Yano, S. & Naruse, T. (1998). Sulfite is released by human
Friedrich, C. G. (1998). Physiology and genetics of sulfur-oxidizing
neutrophils in response to stimulation with lipopolysaccharide.
J Leukoc Biol 64, 595–599.
bacteria. Adv Microb Physiol 39, 235–289.
Myers, J. D. & Kelly, D. J. (2005). Respiratory electron transport in
Gouet, P., Courcelle, E., Stuart, D. I. & Metoz, F. (1999). ESPript:
Helicobacter and Campylobacter. In Respiration in Archaea and
Bacteria, vol 2: Diversity of Prokaryotic Respiratory Systems. Edited by
D. Zannoni. Dordrecht: Kluwer (in press).
multiple sequence alignments in PostScript. Bioinformatics 15,
305–308.
Hille, R. (1996). The mononuclear molybdenum enzymes. Chem Rev
96, 2757–2816.
Hoffman, P. S. & Goodman, T. G. (1982). Respiratory physiology
and energy conservation efficiency of Campylobacter jejuni. J Bacteriol
150, 319–326.
Hughes, N., Chalk, P., Clayton, C. L. & Kelly, D. J. (1998).
Helicobacter pylori porCDAB and oorDABC genes encode distinct
pyruvate and 2-oxoglutarate : acceptor oxidoreductases which
mediate electron transport to NADP. J Bacteriol 180, 1119–1128.
Jiang, X. & Doyle, M. P. (2000). Growth supplements for Helicobacter
pylori. J Clin Microbiol 38, 1984–1987.
Jones, C. W. & Poole, R. K. (1985). The analysis of cytochromes.
Methods Microbiol 18, 285–323.
Kappler, U. & Dahl, C. (2001). Enzymology and molecular biology of
Nachamkin, I., Allos, B. M. & Ho, T. W. (2000). Campylobacter jejuni
infection and the association with Guillain–Barré syndrome. In
Campylobacter, 2nd edn, pp. 155–175. Edited by I. Nachamkin &
M. J. Blaser. Washington, DC: American Society for Microbiology.
Page, R. D. M. (1996). Treeview: an application to display phylo-
genetic trees on personal computers. Comput Appl Biosci 12,
357–358.
Parkhill, J., Wren, B. W., Mungall, K. & 17 other authors (2000). The
genome sequence of the food-borne pathogen Campylobacter jejuni
reveals hypervariable sequences. Nature 43, 665–668.
Pickett, C. L. (2000). Campylobacter toxins and their role in
pathogenesis. In Campylobacter, 2nd edn, pp. 179–190. Edited by
I. Nachamkin & M. J. Blaser. Washington, DC: American Society for
Microbiology.
prokaryotic sulfite oxidation. FEMS Microbiol Lett 203, 1–9.
Sargent, F., Bogsch, E. G., Stanley, N. R., Wexler, M., Robinson, C.,
Berks, B. C. & Palmer, T. (1998). Overlapping functions of
Kappler, U., Bennett, B., Rethmeier, J., Schwarz, G., Deutzmann, R.,
McEwan, A. G. & Dahl, C. (2000). Sulfite : cytochrome c oxido-
components of a bacterial Sec-independent protein export pathway.
EMBO J 17, 3640–3650.
reductase from Thiobacillus novellus. J Biol Chem 275, 13202–13212.
Kappler, U., Friedrich, C. G., Trüper, H. G. & Dahl, C. (2001).
Sellars, M. J., Hall, S. J. & Kelly, D. J. (2002). Growth of
Evidence for two pathways of thiosulfate oxidation in Starkeya
novella (formerly Thiobacillus novellus). Arch Microbiol 175, 102–111.
Campylobacter jejuni supported by respiration of fumarate, nitrate,
nitrite, trimethylamine-N-oxide or dimethylsulfoxide requires
oxygen. J Bacteriol 184, 4187–4196.
Kelly, D. J. (2001). The physiology and metabolism of Campylobacter
jejuni and Helicobacter pylori. J Appl Microbiol 90, 16S–24S.
Skirrow, M. B. & Blaser, M. J. (2000). Clinical aspects of Campylo-
Kisker, C., Schindelin, H. & Rees, D. C. (1997). Molybdenum-
cofactor-containing enzymes: structure and mechanism. Annu Rev
Biochem 66, 233–267.
Kisker, C., Schindelin, H., Baas, D., Rétey, J., Meckenstock, R. U. &
Kroneck, P. M. H. (1999). A structural comparison of molybdenum
cofactor-containing enzymes. FEMS Microbiol Rev 22, 503–521.
Laemmli, U. K. (1970). Cleavage of structural protein during the
assembly of the head of bacteriophage T4. Nature 227, 680–685.
Leyland, M. L. & Kelly, D. J. (1991). Purification and characterisation
of a monomeric isocitrate dehydrogenase with dual coenzyme
specificity from the photosynthetic bacterium Rhodomicrobium
vannielii. Eur J Biochem 202, 85–93.
bacter infection. In Campylobacter, 2nd edn, pp. 69–88. Edited by
I. Nachamkin & M. J. Blaser. Washington, DC: American Society for
Microbiology.
Sorokin, D. Y. (1995). Sulfitobacter pontaiacus gen. nov., sp. nov. – a
new heterotrophic bacterium from the Black Sea specialized on
sulfite oxidation. Microbiology (English translation of Mikrobiologiya)
64, 295–305.
Thompson, J. D., Gibson, T. J., Plewiniak, F., Jeanmougin, F. &
Higgins, D. A. (1997). The CLUSTAL_X Windows interface: flexible
strategies for multiple sequence alignment aided by quality analysis
tools. Nucleic Acids Res 25, 4876–4882.
van Vliet, A. H., Wooldridge, K. G. & Ketley, J. M. (1998). Iron-
responsive gene regulation in a Campylobacter jejuni fur mutant.
J Bacteriol 180, 5291–5298.
Loschi, L., Brokx, S. J., Hills, T. L., Zhang, G., Bertero, M. G.,
Lovering, A. L., Weiner, J. H. & Strynadka, N. C. J. (2004). Structural
Wood, P. M. (1988). Chemolithotrophy. In Bacterial Energy Trans-
and biochemical identification of a novel bacterial oxidoreductase.
J Biol Chem (in press). doi:10.1074/jbc.M408876200
duction, pp. 183–230. Edited by C. Anthony. London: Academic
Press.
242
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 00:45:15
Microbiology 151