Microbiology (2003), 149, 419–430
DOI 10.1099/mic.0.25978-0
Thioredoxin 2 is involved in oxidative stress
defence and redox-dependent expression of
photosynthesis genes in Rhodobacter capsulatus
Kuanyu Li, Elisabeth Härtig3 and Gabriele Klug
Institut für Mikrobiologie und Molekularbiologie, University of Giessen, Heinrich-Buff-Ring
26-32, D-35392 Giessen, Germany
Correspondence
Gabriele Klug
gabriele.klug@
mikro.bio.uni-giessen.de
Received 3 September 2002
Revised
4 November 2002
Accepted 8 November 2002
Thioredoxins are small ubiquitous proteins that display different functions mainly via redox-mediated
processes. The facultatively photosynthetic bacterium Rhodobacter capsulatus harbours at least
two genes for thioredoxin 1 and 2, trxA and trxC. It is demonstrated that thioredoxin 2 of R.
capsulatus can partially replace the thioredoxin 1 function as a hydrogen donor for methionine
sulfoxide reductase but cannot replace thioredoxin 1 as a subunit of phage T7 DNA polymerase.
By inactivating the trxC gene in R. capsulatus, it is shown that thioredoxin 2 is involved in resistance
against oxidative stress. As thioredoxin 1 of Rhodobacter sphaeroides, R. capsulatus thioredoxin
2 affects the oxygen-dependent expression of photosynthesis genes, albeit in an opposite way.
The trxC mutant of R. capsulatus shows a stronger increase in photosynthesis gene expression after
a decrease in oxygen tension than the isogenic wild-type strain. The expression of the trxC gene is
downregulated by oxygen.
INTRODUCTION
Thioredoxins are found in all living organisms. Together
with glutaredoxins, they are the major enzymes responsible
for maintaining disulfide bonds in cytoplasmic proteins
in a reduced state (Ritz & Beckwith, 2001). These small
proteins are capable of catalysing thiol-disulfide redox
reactions by a common active site sequence: Cys-X1-X2Cys. The oxidized thioredoxins are reduced by thioredoxin
reductase under consumption of NADPH.
The electron transfer through disulfide bond exchange
reactions in the cytoplasm recycles essential enzymes such
as ribonucleotide reductase (Orr & Vitols, 1966), which
provides deoxyribonucleotides for DNA synthesis. Other
metabolic enzymes that are recycled by thiol-disulfide
reactions are phosphoadenosine-phosphosulfate reductase
(Lillig et al., 1999), methionine sulfoxide reductase (BoschiMuller et al., 2000, 2001) and arsenate reductase (Shi
et al., 1999).
Furthermore, thioredoxin is an essential subunit of the
DNA polymerase of bacteriophage T7 (Huber et al., 1987;
Mark & Richardson, 1976) and is essential for the assembly
of several filamentous phages (Russel & Model, 1985). In
eukaryotic organisms, many additional roles of thioredoxin
3Present address: Institut für Mikrobiologie, TU Braunschweig,
Spielmannstr. 7, 38106 Braunschweig, Germany.
Abbreviation: t-BOOH, tertiary-butyl hydroperoxide.
0002-5978 G 2003 SGM
have been reported, among which are the regulation of
transcription factors such as NF-kB (Schulze-Osthoff et al.,
1995) and the regulation of apoptosis (Saitoh et al., 1998).
Thioredoxin is also involved in the regulation by light of
photosynthetic enzymes in plant chloroplasts (Buchanan,
1984; Buchanan et al., 1994).
Because of its extremely low redox potential and free thiol
in its reduced form, which can readily form a disulfide
bridge, thioredoxin is considered to be involved in defence
against oxidative stress not only by regeneration of oxidatively damaged proteins (Fernando et al., 1992; Natsuyama
et al., 1992), but also by its ability to reduce hydrogen
peroxide (Mitsui et al., 1992; Nakamura et al., 1994; Spector
et al., 1988; Tomimoto et al., 1993) or by acting as a
hydrogen donor for peroxidase (Chae et al., 1994). Thioredoxin also functions as a singlet oxygen quencher and
hydroxyl radical scavenger independently of its redox state
(Das & Das, 2000).
Two thioredoxin genes, trxA and trxC, encoding thioredoxin 1 and 2, respectively, were identified in Escherichia
coli (Laurent et al., 1964; Miranda-Vizuete et al., 1997).
Thioredoxin 1 and thioredoxin 2 have 29 % sequence
identity, with the greatest difference being a 32 aa extension
of thioredoxin 2 at its N terminus. Thioredoxin 2 possesses
additional cysteine thiols apart from those of the active
site, which, when oxidized, downregulate its activity in
the reduction in insulin disulfides. Thioredoxin 2 is also
less heat-stable than thioredoxin 1 and does not reduce
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 10:35:39
Printed in Great Britain
419
K. Li, E. Härtig and G. Klug
ribonucleotide reductase as efficiently as thioredoxin 1.
Thioredoxin 2 participates in the OxyR-orchestrated
antioxidant response (Ritz et al., 2000). Thioredoxin 1
expression is not controlled by OxyR but in the stationary
phase is controlled by ppGpp (Lim et al., 2000).
The facultatively phototrophic purple bacteria of the genus
Rhodobacter can adapt rapidly to changes in their environment. As long as oxygen is available, they perform oxidative
respiration. If the oxygen tension drops below a threshold
value, the formation of pigment protein complexes is
induced. When oxygen is no longer available, Rhodobacter
gains energy by anoxygenic photosynthesis in the presence
of light. Many different proteins are involved in the
oxygen-dependent regulation of photosynthesis genes in
Rhodobacter (e.g. Gregor & Klug, 2002). A central
component of the redox control system of Rhodobacter
capsulatus is the two-component system RegB/RegA (PrrB/
PrrA in Rhodobacter sphaeroides) (Sganga & Bauer, 1992).
The phosphorylated response regulator RegA activates
the expression of photosynthesis genes at a low oxygen
tension (Masuda et al., 1999; Sganga & Bauer, 1992). Some
photosynthesis genes are repressed at a high oxygen tension
by the CrtJ protein (PpsR in R. sphaeroides). CrtJ binds to
its target DNA sequences in a redox-dependent manner
(Ponnampalam & Bauer, 1997). It was shown recently that
the CrtJ protein forms an intramolecular disulfide bond
when exposed to oxygen, which is critical for binding
to its target promoters (Masuda et al., 2002). In addition to
these DNA-binding proteins, thioredoxin 1 was shown to
be involved in the redox-dependent expression of photosynthesis genes in R. sphaeroides by a yet unidentified
mechanism. A mutant strain, which harbours reduced
levels of thioredoxin 1, shows a lower degree of induction
of the puf and puc genes encoding pigment-binding proteins after a decrease in oxygen tension than the wild-type
strain (Pasternak et al., 1999). In R. sphaeroides, the single
thioredoxin 1 is essential for growth (Pasternak et al., 1997).
The expression of trxA increases during an increase in
oxygen tension (Pasternak et al., 1999).
The genome of R. capsulatus, a close relative of R.
sphaeroides, harbours at least two genes encoding thioredoxins, trxA and trxC. Since thioredoxin 2 resembles
the thioredoxin 1, it is possible that thioredoxin 2 of
R. capsulatus is also involved in the redox-dependent
regulation of photosynthesis genes. To learn more about
the function of thioredoxin 2 in R. capsulatus, we constructed a trxC mutant of this strain and constructed
plasmids with wild-type and mutated thioredoxin genes,
which were expressed in E. coli. We demonstrate (1) that
it can partially replace the thioredoxin 1 function as
hydrogen donor for methionine sulfoxide reductase, but (2)
that it cannot replace thioredoxin 1 as a subunit of phage
T7 DNA polymerase independent on its redox potential,
(3) that R. capsulatus thioredoxin 2 is involved in
defence against oxidative stress and (4) that R. capsulatus
thioredoxin 2 affects the oxygen-dependent expression of
420
photosynthesis genes as thioredoxin 1 from R. sphaeroides,
albeit in an opposite way. Thus, the trxA gene from
R. sphaeroides and the trxC gene from R. capsulatus are
reversely affected by oxygen, and the corresponding
thioredoxins exert an opposite effect on the expression of
photosynthesis genes.
METHODS
Strains, plasmids and growth conditions. The strains and plas-
mids used in this study are listed in Table 1. Thioredoxin sequences
from R. capsulatus and R. sphaeroides genomes were used to design
the primers KH8 (59-GGCCACCATGGCGGAATC-39) and EH26
(59-TCAAGCTTCCCAAACCCC-39) for trxC (RRC00979), encoding
the 147 aa thioredoxin 2 protein of R. capsulatus, RctrxAstart
(59-GAGAATCCCATGGCTACCGTCG-39) and RctrxAendHind
(59-GGTAAGCTTGCCCTCCGTTTGC-39) for trxA (RRC003345),
encoding the 106 aa thioredoxin 1 protein of R. capsulatus, and
EH7 (59-GTCGGAAGCTTGCTCAAGGAA-39) and EH12 (59AATTGATCATGAGCACCGTTCCC-39) for the 106 aa thioredoxin
1 protein of R. sphaeroides. The PCR products were restricted with
NcoI and HindIII, ligated to appropriately digested pUTC51 and
transformed into E. coli strain BH216 (Table 1).
E. coli cultures were grown in Luria–Bertani (LB) or M9 minimal
medium as indicated, R. capsulatus was cultivated at 32 ˚C in a malate
minimal salt medium (Drews, 1983). Cells were grown aerobically by
shaking 100 ml of culture at 140 r.p.m. in 1 l baffled flasks (about 20 %
dissolved oxygen, as determined using an Ag/Pt oxygen electrode).
For semi-aerobic incubation, 40 ml of culture was shaken in 50 ml
Erlenmeyer flasks at 140 r.p.m. (1–2 % dissolved oxygen, as determined using an Ag/Pt oxygen electrode). For oxygen shift experiments,
cells were grown aerobically to an OD660 of 0?5–0?6 and then
transferred to the 50 ml flasks or vice versa. Incubation was in the dark.
Determination of survival rates. Cultures were grown under
semi-aerobic conditions until the optical density reached 0?4–0?5;
then, oxidative stress agents were added at indicated concentrations
for 1 h and dilutions were plated. Survival of 100 % corresponds to
the viable cell number determined immediately before the addition
of the oxidative agents. The percentage of colonies grown from the
treated cultures is given as the percentage survival in Table 2. The
values are the mean of three experiments. Hydrogen peroxide,
tertiary-butyl hydroperoxide (t-BOOH), diamide and methyl viologen (Paraquat) were purchased from Sigma.
Zone of inhibition assays. For zone of inhibition assays, cells
were grown semi-aerobically overnight at 32 ˚C and then diluted in
the same medium to an OD660 of 0?2. Cells were then grown to an
OD660 of 0?5 (mid-exponential phase samples) and 100 ml of culture
was added to 2 ml prewarmed (47 ˚C) top agar (0?7 % agar) and
layered onto malate minimal salt medium plates. After the top agar
had hardened, a 5?5 mm filter paper disk, containing 5 ml of oxidative agent as indicated, was placed on the plate. The plates were then
incubated overnight at 32 ˚C and the radio of the zone of growth
inhibition was examined to determine the sensitivity of the cell
against the agents. All assays were performed three or four times in
duplicate.
Phage assay. To determine whether TrxC of R. capsulatus can
substitute for E. coli TrxA as an essential subunit of T7 DNA polymerase, T7 infection tests were performed as described by Sambrook
et al. (1989). Bacteria were grown overnight in LB broth supplemented with the corresponding antibiotics. Appropriate culture dilutions
were mixed with corresponding-titre T7, incubated for 20 min at
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 10:35:39
Microbiology 149
Thioredoxin 2 function in Rhodobacter
Table 1. Bacterial strains and plasmids
Strain/plasmid
E. coli
JM109
BH216
SM10
M15 (pREP4)
R. capsulatus
SB1003
SB1003trxC2
Plasmids
pRK415
pUTC51
pUTRctrxC
pUTRctrxA
pUTEcTX
pURRctrxC70A
pURRctrxC29A
pURRstrxC29A
pQE32
pQERctrxC
pUC4KSAC
pPHU281
pPHDRctrxC
pPHDRctrxC : : Km
pBBR1MCS-5
pILA
pBBR502
pBBR502lux
Relevant characteristic(s)*
Source or reference
Stratagene
recA1 supE44 endA1 hsdR17 gyrA96 relA1 thi (lac–proAB)
F9[traD36 proAB+ lacIqlacZ M15]
2
F araD139 galU galK hsr strA metE46 argH1 ilvC : : Tn5 trxA2
recA thi thr leu; chromosomal RP4-2 (Tc : : Mu)
Host strain for protein overexpression
Pille et al. (1990)
Simon (1983)
Qiagen
Wild-type
SB1003 DtrxC : : Kmr
Yen (1976)
This study
Mob+ Tcr lacZa IncP
1?9 kb PstI fragment containing trxA gene of R. sphaeroides
under the control of ptrc promoter and lacIq
trxC gene of R. capsulatus replaced trxA gene of R. sphaeroides in pUTC51
trxA gene of R. capsulatus replaced trxA gene of R. sphaeroides in pUTC51
trxA gene of E. coli replaced trxA gene of R. sphaeroides in pUTC51
pUTRctrxC with mutation of R. capsulatus thxC gene to give alanine
in place of 70 cysteine-70 (C70A)
pUTRctrxA with mutation of R. capsulatus thxA gene to give alanine
in place of cysteine-29 (C29A)
pUTC51 with mutation of R. sphaeroides thxA gene to give alanine
in place of cysteine-29 (C29A)
His-tag vector for cloning thioredoxin
pQE32 with insert thioredoxin C coding region
Apr, source of the Kmr cassette
Tcr lacZ9 mob(RP4)
1?4 kb fragment containing the upstream and downstream of the
R. capsulatus trxC stretch cloned into pPHU281
1?3 kb Km cassette cloned into pPHDRctrxC
A broad-host-range cloning vector with getamycin resistence
A promoter probe vector with luxAB
PBBR1MCS with upstream fragment of trxC gene
PBBR502 with 2?45 kb luxAB fragment
Keen & Tamaki (1988)
Assemat et al. (1995)
This
This
This
This
study
study
study
study
This study
This study
Qiagen
This study
Barany (1985)
Hübner et al. (1993)
This study
This study
Kovach et al. (1994)
Kunert et al. (2000)
This study
This study
*Ap, ampicillin; Km, kanamycin; Tc, tetracycline.
room temperature for attachment then added to 0?7 % agar LB
broth and spread on to 1 % LB plates, which were incubated at
37 ˚C for 4 h to monitor T7 development.
Cloning of R. capsulatus and R. sphaeroides trx genes. The
trxC sequence (RRC00979) was used to design the specific mutagenic primers RctrxCexp>A, 59-GGATCCCGGAATCGCTGCGACTGA-39 (forward, introduction of BamHI site), and RctrxCexp>B,
59-AAGCTTCTTGGGCGCGGGTTCTC-39 (reverse, introduction of
HindIII site). The PCR product produced with the above primers
was cloned into pGEM T-vector (Promega) and recloned into
pQE32 His-tag vector (Qiagen) digested with HindIII and BamHI.
The recombinant plasmid designated pQERctrxC was transformed
into E. coli strain JM109. The correct construct as confirmed by
sequencing was transformed into E. coli strain M15 (pREP4) for
overexpression of thioredoxin 2 induced by the addition of 1 mM
IPTG at 32 ˚C for 4 h.
For mutagenesis of the first cysteine of thioredoxin 1 and/or 2
active site -C-X-X-C-, the following primers were designed:
RsC29A>A (forward), 59-TGGGCCGGCCCCTGCCGGCAGAT-39,
http://mic.sgmjournals.org
and RsC29A>B (reverse, complementary to RsC29A>A), 59ATCTGCCGGCAGGGGCCGGCCCA-39, for trxA of R. sphaeroides
(RstrxA) and trxC of R. capsulatus (RctrxC); RcC29A>A (forward),
59-GAATGGGCTGGCCCCTGCAAGATG-39, and RcC29A>B
(reverse, complementary to RcC29A>A), 59-CATCTTGCAGGGGCCAGCCCATTC-39, for trxA of R. capsulatus (RctrxA). The upstream
fragments of RstrxA from primers EH12 and RsC29A>B, RctrxA
from RctrxAstart and RcC29A>B, RctrxC from KH8 and RsC29A>B
and the downstream fragments of RstrxA from primers EH7 and
RsC29A>A, RctrxA from RctrxAendHind and RcC29A>A, RctrxC
from EH26 and RsC29A>A were amplified by PCR. The diluted
overlapped PCR products of upstream and downstream fragments,
which acted as primers and templates, were amplified to produce the
full-length thioredoxin fragments with the mutation of cysteine into
alanine. The full-length trx fragments were cloned into plasmid pUTC5
(Assemat et al., 1995) replacing wild-type trxA of R. sphaeroides.
Mutant construction. R. capsulatus strain SB1003trxC2 was gener-
ated by transferring the suicide plasmid pPHRctrxC : : Km into wildtype SB1003 and screening for insertion of the kanamycin cassette
into the chromosome by double crossover. To this end, parts of the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 10:35:39
421
422
131±19
130±20
87±21
53±12
100±17
95±10
120±11
95±15
93±11
70±9
0?1
0?3
0?5
1
0?6
t -BOOH (mM)
Diamide (mM)
62±7
46±13
100±20
96±11
70±10
20±5
136±20
112±13
141±23
140±21
90±10
70±10
1?5
0?1
0?5
1
0?1
0?5
was carried out in a final volume of 25 ml containing 0?125 mg
random primer (Promega), 5 mg total RNA, 1 mM each of four
deoxyribonucleoside triphosphates and 1?5 units reverse transcriptase (Promega) at 42 ˚C for 1 h and 99 ˚C for 2 min to inactivate
the reverse transcriptase. PCR was carried out in a final volume of
50 ml containing 5 ml reverse-transcribed cDNA solution, 1 unit Taq
polymerase (Qiagen) and 2 pmol each of the oligonucleotide primers 59-GCGCCGCAGTTCCAAGCC-39 and 59-CTTGCCGCGGACGAAGCC-39 for trxC, and 59-GAAGTGCGCCAATCCGAC-39 and
59-AGAGCGAGGCTTCGATCC-39 for trxA. Amplification was carried out by an initial denaturation step at 96 ˚C for 3 min followed
by 38 cycles for trxC or 23 cycles for trxA at 96 ˚C for 1 min, 62 ˚C
(for trxC) or 57 ˚C (for trxA) for 40 s and 72 ˚C for 30 s. A sample
lacking reverse transcriptase was included for each reaction as a control for DNA contamination. Reaction products were subjected to
electrophoresis on 4 % agarose gels (Biozym).
Southern, Northern and colony hybridization analysis. South-
ern and Northern blot hybridizations were performed as described
previously (Engler-Blum et al., 1993; Pasternak et al., 1997). Colony
hybridization was carried out according to the manufacturer’s
recommendation (PALL).
136±22
100±7
157±30
144±25
SB1003
SB1003trxC2
100
100
SA–A
1
Production of antibodies and Western blot analysis. For
134±17
73±10
Paraquat (mM)
H2O2 (mM)
trxC gene together with upstream and downstream sequences were
amplified by PCR using the primers P1 and P2 and primers P3 and
P4, respectively. The primer sequences were as follows: P1, 59GGGGTACCGCAGGCCTCCTACA-39; P2, 59-CGGGATCCGACGGCACCTTGTT-39; P3, 59-CGGGATCCTTGRCGGCTTCGTC-39; P4,
59-CCCAAGCTTATCGGCCATGCTCAG-39. The PCR products
were digested with appropriate restriction enzymes (KpnI and
BamHI for the upstream fragment, BamHI and HindIII for the
downstream fragment) and cloned into pPHU281 (Hübner et al.,
1993) to give plasmid pPHRCtrxC and generating a BamHI site at
the junction. Then, a 1?3 kb BamHI fragment containing the kanamycin cassette from pUC4KSAC (Barany, 1985) was inserted into
the BamHI site of pPHRctrxC to generate pPHRctrxC : : Km, which
was transformed into E. coli strain SM10. For the purpose of generating strain SB1003trxC2, diparental conjugation was carried out
with R. capsulatus SB1003 as a recipient strain. A Southern hybridization was performed to confirm the correct insertion of the kanamycin cassette into the chromosome.
RT-PCR. For RT-PCR analysis, the reverse transcription reaction
SA (t=1 h)
SA (t=0)
Strains
The number of viable cells before change in oxygen tension or addition of oxidative-stress-generating agents equals 100 % (column 2). All other columns refer to the number of viable
cells 1 h after the change in culture conditions. The values give the mean of three experiments with maximal deviation. SA, Semi-aerobic conditions; SA–A, shift from semi-aerobic to
aerobic growth conditions.
Table 2. Survival rates of wild-type SB1003 and a trxC mutant from R. capsulatus after the addition of substances which generate oxidative stress
K. Li, E. Härtig and G. Klug
immunological detection of thioredoxin 1 or 2, 20 mg crude cell
extracts as determined by Bradford (1976) were separated by SDS–
PAGE (Laemmli, 1970) on 15 % polyacrylamide gels and transferred
to Immobilon-P membrane (Millipore). The R. sphaeroides thioredoxin 1 and R. capsulatus thioredoxin 2 proteins were expressed as
His-tag fusion proteins from pQE32-derived plasmids (Qiagen) in
E. coli M15(pREP4) and the antibodies against the gel-purified proteins were raised in rabbits (Clontech). The antibodies were purified
from the serum by protein A Sepharose. Western blotting was performed according to the Western Exposure Chemiluminescent
Detection system (Clontech).
Bacteriochlorophyll measurements. A sample (0?5 ml) of culture
was sedimented and resuspended in 0?5 ml of acetone–methanol
(7 : 2, v/v). The absorbance of the supernatant at 770 nm was determined after spinning in a microcentrifuge for 3 min. The relative
bacteriochlorophyll content of the cells is given as the absorbance at
770 nm divided by the optical density at 660 nm.
Construction of the reporter genes for luciferase assays. The
luciferase reporter plasmids were constructed by cloning the KpnI–
StuI PCR fragments coding the 59-flanking region of the trxC
gene of R. capsulatus into restriction sites KpnI and SmaI of
vector pBBR1MCS-5 (Kovach et al., 1994) to generate pBBR502. To
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 10:35:39
Microbiology 149
Thioredoxin 2 function in Rhodobacter
generate pBBR502lux, the luxAB XbaI fragment from plasmid
pILA (Kunert et al., 2000) was cloned into the appropriate restriction sites of pBBR502. Plasmid pBBR502lux was transferred into
R. capsulatus SB1003 by diparental conjugation to obtain SB1003
(pBBR502lux) (Table 1).
Luciferase assays were performed as described by Kunert et al. (2000) at
room temperature with some modifications. In brief, luminescence of
the luciferase reaction was induced by the addition of decanal (Sigma)
to 1 ml culture (final concentration of decanal 0?5 mM). Light
emission was monitored in a photomultiplier-based luminometer
(BioOrbit; Labsystems). The mean value of four to five data near the
maximum of the peak was used as the luminescence output. All the
readings were related to the optical density of the culture at 660 nm, as
determined with a UV/visible spectrophotometer. All measurements
were made in duplicate and experiments were performed at least twice
using independent cultures.
RESULTS
Thioredoxin 2 from R. capsulatus can replace
the thioredoxin 1 from E. coli as an electron
donor for methionine sulfoxide reductase, but
not as a functional subunit of phage T7
Before analysing the effect of a trxC mutation in
R. capsulatus, we wanted to confirm that thioredoxin 2
from this organism is functionally similar to thioredoxin 2
from E. coli. The E. coli metE strain BH216 is a methionine
auxotroph and lacks a functional trxA gene (Pille et al.,
1990). The parental metE strain can grow on minimal
medium supplemented with methionine sulfoxide, since it is
able to convert methionine sulfoxide to methionine. Since
thioredoxin 1 is required as a hydrogen donor for methionine sulfoxide reductase, BH216 can grow on minimal
medium only if methionine is added. Different plasmids, in
which overexpression of thioredoxin was regulated by an
IPTG-inducible promoter, were constructed and transformed into BH216 to complement trx deficiency, as
evidenced by growth on minimal medium with methionine
sulfoxide (Table 1).
Affinity-purified polyclonal antibodies were used to
determine the presence of thioredoxin of R. sphaeroides or
R. capsulatus in E. coli strains, which harbour the plasmids
with trx genes of R. sphaeroides or R. capsulatus. Fig. 1 shows
that the anti-thioredoxin 1 or anti-thioredoxin 2 antibodies
reacted with one band of the expected size of thioredoxin 1
(11–12 kDa) or thioredoxin 2 (15?3 kDa) in a total crude
extract of the E. coli transformants. Thus, thioredoxin 1 of
R. sphaeroides or R. capsulatus and thioredoxin 2 of
R. capsulatus are expressed in E. coli strain BH216.
Thioredoxin 1 proteins of E. coli, R. sphaeroides and
R. capsulatus were not detectable with anti-thioredoxin 2
antibody of R. capsulatus: only one clear band was observed
in BH216(pUTRctrxC) cell extracts (Fig. 1b). In contrast,
antibodies raised against E. coli thioredoxin 2 could react
with E. coli thioredoxin 1, while E. coli thioredoxin 1
antibodies did not cross-react with thioredoxin 2 (MirandaVizuete et al., 1997).
http://mic.sgmjournals.org
Fig. 1. Test for thioredoxin expression of R. sphaeroides or R.
capsulatus in E. coli using a Western blot. Samples were
probed with the anti-TrxA antibody of R. sphaeroides (a), then
the blot was stripped and reprobed with anti-TrxC antibody of
R. capsulatus (b). (a) Identification of TrxA in cell extracts of
E. coli. Lanes: 1, BH216 control, no TrxA expression; 2,
BH216(pUTC51) expressing RsTrxA (11?3 kDa); 3, BH216
(pUTEcTX) expressing EcTrxA (11?8 kDa); 4, BH216(pUTRctrxA)
expressing RcTrxA (11?2 kDa); 5, BH216(pTURctrxC) expressing RcTrxC, no signal with anti-RsTrxA. Note that the relative
migration of the different thioredoxin proteins in 15 % SDS–
PAGE is not in accordance with the calculated molecular
masses. (b) Identification of TrxC in cell extracts of E. coli
BH216(pUTRctrxC) expressing RcTrxC (15?3 kDa). No TrxC
signal was observed in other BH216 derivative strains.
Since thioredoxin 1 is required as hydrogen donor for
methionine sulfoxide reductase, BH216 can form colonies
overnight on minimal medium only if methionine is added.
Strain BH216, which contains a plasmid harbouring the
E. coli trxA gene, can grow on minimal medium containing
methionine sulfoxide but no methionine. Strain BH216
containing either a plasmid harbouring the trxA gene from
R. sphaeroides or from R. capsulatus can also grow under
these conditions, indicating that the thioredoxin 1 protein
from R. sphaeroides or R. capsulatus can act as a hydrogen
donor for methionine sulfoxide reductase in E. coli (Fig. 2).
Fig. 2. Test for thioredoxin function as an electron donor of
methionine sulfoxide reductase. Strain BH216 with plasmids
harbouring trxC (b), trxA (c) of R. capsulatus, or trxA of
R. sphaeroides (d) or E. coli (e) exhibited growth on minimal
medium overnight with 30 mg methionine sulfoxide ml21 as the
methionine source. Parental strain BH216 used as a control
did not show any growth on the same medium (a).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 10:35:39
423
K. Li, E. Härtig and G. Klug
Strain BH216 containing a plasmid that allows overexpression of the R. capsulatus trxC gene shows only very
weak growth on minimal medium with methionine
sulfoxide, indicating that thioredoxin 2 cannot fully replace
this thioredoxin 1 function.
To see whether the function of thioredoxin of R. sphaeroides
or R. capsulatus as hydrogen donor for methionine sulfoxide
reductase is dependent on the cysteine at the active site, the
first cysteine of the active site was mutated into alanine
in the three thioredoxins (C29A for thioredoxin 1 of
R. sphaeroides and R. capsulatus, C70A for thioredoxin 2
of R. capsulatus). As expected, the mutant thioredoxins
were no longer able to function as a hydrogen donor for
methionine sulfoxide reductase. BH216 derivatives harbouring the plasmid, which enables overexpression of
the C29A or C70A mutant thioredoxin, could not form
colonies on M9 minimal medium with the addition of
methionine sulfoxide after overnight incubation (data not
shown). But with the addition of methionine in M9
medium, colonies of a normal size were formed. These
results confirmed that this function of thioredoxin as a
hydrogen donor is dependent on the cysteine of the active
site and that the two additional pairs of cysteines at the
N terminus of thioredoxin 2 from R. capsulatus cannot
donate hydrogen to methionine sulfoxide reductase.
Reduced thioredoxin complexes with bacteriophage T7
DNA polymerase, resulting in an enzyme with a high
processivity (Huber et al., 1987). T7 phage cannot be
propagated in E. coli strain BH216, which is unable to
express thioredoxin 1. T7 phage can be propagated in
E. coli strain BH216 containing a plasmid habouring the
E. coli trxA gene, the R. sphaeroides trxA gene or the
R. capsulatus trxA gene (data not shown). Thus, thioredoxins 1 from R. sphaeroides and R. capsulatus can function
as subunits of T7 DNA polymerase in E. coli. However,
we observed no propagation of phage T7 in BH216
containing a plasmid, which allows expression of the
R. capsulatus thioredoxin 2 (data not shown). We conclude
that R. capsulatus thioredoxin 2 cannot function as a
subunit of T7 DNA polymerase.
The ability of T7 to grow on the E. coli trxA mutant strains
in which one or both active site cysteine residues of
thioredoxin had been changed to serine or alanine definitively demonstrated that the redox capacity of thioredoxin
is not required for stimulation of DNA polymerase activity
(Huber et al., 1986). In contrast, neither thioredoxin-S2 nor
modified thioredoxin in which a single cysteine had been
methylated formed a complex with T7 DNA polymerase
(Adler & Modrich, 1983), suggesting that the conformation
of the active site is very important for this interaction. To
test the effect of thioredoxin 1 mutation C29A, the identical
constructs as described above for the in vivo methionine
sulfoxide assay were used for T7 infection. Interestingly,
clear plaques were observed in the plate with strain
BH216(pUTRctrxC29A), but not in the plate with strain
BH216(pUTRstrxC29A). Thus, the mutated thioredoxin 1
424
(C29A) from R. capsulatus still acts as a subunit of T7
DNA polymerase to allow T7 phage to amplify, while
mutated thioredoxin 1 of R. sphaeroides does not. Both
thioredoxins 1 of the closely related species, R. sphaeroides
and R. capsulatus, share an amino acid sequence identity
of 82 %. However, the two active sites have different
amino acids surrounding the cysteines. The mutation in
the active site may lead to different conformational changes
in the two thioredoxins (see Discussion).
Construction of a trxC mutant of R. capsulatus
To learn more about the role of thioredoxin 2 in
R. capsulatus and to test for a function in redox-dependent
regulation of photosynthesis genes, we inactivated the trxC
gene of R. capsulatus. The trxC gene is positioned between
the genes for thymidine kinase and Ala-tRNA on the
R. capsulatus chromosome (Fig. 3). All these genes are
oriented in the same direction as the trxC gene. To
inactivate the trxC gene in R. capsulatus, we cloned the
trxC upstream and downstream regions including the
N-terminal and C-terminal segments of the coding regions
into plasmid pPHU281 (Hübner et al., 1993), which is
unable to replicate in Rhodobacter. We then inserted the
kanamycin cassette gene lacking transcriptional terminators
between the trx upstream and downstream segments. After
conjugational transfer into R. capsulatus, about 400 colonies
were screened for insertion of the resistance gene into the
chromosome via a double crossover by testing for resistance
to kanamycin but a lack of the plasmid encoded tetracycline
resistance. The fact that all of the kanamycin-resistant
colonies were tetracycline-sensitive indicated that insertion
of the tetracycline resistance gene into the chromosome by a
single crossover may not lead to expression of tetracycline
resistance. We therefore analysed those kanamycin-resistant
clones by colony hybridization. Twenty-nine clones did not
show hybridization to plasmid pPHU281. Isolated chromosomal DNA from these clones was used for Southern
hybridization to confirm the presence of the resistance
cassette and to test for the interruption of trxC gene. Two
trxC mutant clones were identified, which had the
kanamycin cassette integrated into the chromosomal trxC
gene by a double crossover.
Fig. 3. Schematic representation of the thioredoxin 2 gene,
trxC, and its neighbouring genes in strain SB1003 and the
trxC replacement by a kanamycin cassette. The arrows show
the orientation of the genes on the chromosome. The size of
the hypothetical protein is not to scale, as indicated by slashes.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 10:35:39
Microbiology 149
Thioredoxin 2 function in Rhodobacter
Doubling times and survival rates of the trxC
mutant under different growth conditions
To see whether the trxC mutation affects the doubling time
of R. capsulatus cultures, we determined the growth rates of
the mutant and the isogenic wild-type strain during growth
under a high oxygen tension (about 20 % dissolved oxygen),
low oxygen tension (1–2 % dissolved oxygen), anaerobic
incubation in the light and after a shift from a high to
low oxygen tension or vice versa. Under all conditions
tested, the maximal doubling times of the wild-type and
the trxC mutant showed no significant differences, while
some difference was observed during the sudden change
in oxygen tension. When exponential-growth cultures
(OD660=0?4–0?5) were shifted from low to high oxygen
tension, in the first hour the trxC mutant showed a transient
decrease in optical density, which did not occur in the
wild-type strain (Fig. 4a). Later on, the mutant and the
wild-type grew with identical rates. The number of viable
cells determined in the culture of the trxC mutant 1 h after
the increase in oxygen tension was similar to that directly
before the transition (Table 2). In contrast, the number of
viable cells in the wild-type culture increased to 136 %
during the same time (Table 2). These results indicate
that thioredoxin 2 of R. capsulatus is required for a fast
adaptation to increasing oxygen tension in the environment.
To test whether thioredoxin 2 is involved in the oxidative stress response of R. capsulatus, we determined the
survival rates after treatment with compounds that can
induce oxidative stress artificially. We used Paraquat as a
superoxide-generating compound, H2O2 as a direct oxidant
and compounds leading to glutathione depletion: the
membrane-permeable thiol-specific oxidizing agent diamide
and the organic hydroperoxide t-BOOH. t-BOOH serves
also as a non-physiological model alkylhydroperoxide. The
results of these experiments are summarized in Table 2.
The addition of 1 mM H2O2 inhibited growth of the trxC
mutant to a larger extent than growth of the wild-type
strain. One hour after addition of 1 mM H2O2 to wild-type
cells, the number of viable cells was increased to a mean of
134 % of the number before the treatment, while the number
of viable cells increased to a mean of 157 % in the untreated
control. In the culture of the trxC mutant, the number
of viable cells was decreased to 73 % after 1 h of H2O2
treatment. There were no significant differences in the
survival rates of wild-type and mutant strain after treatment
with lower concentrations of H2O2. The trxC mutant was
very sensitive against treatment with 1 mM Paraquat. The
survival rate after 1 h of treatment with Paraquat was
Fig. 4. Effect of the trxC mutation on growth of R. capsulatus
after (a) an increase in oxygen tension and (b) in the presence
of agents generating oxidative stress. (a) Cultures of the wildtype SB1003 and the trxC mutant were shifted from semiaerobic growth (1–2 % dissolved oxygen) to aerobic growth
(about 20 % dissolved oxygen) at time point zero or were
further incubated without any transition. Filled diamonds, semiaerobic conditions for SB1003; filled squares, shift from
semi-aerobic to aerobic conditions for SB1003; filled triangles,
semi-aerobic conditions for trxC mutant; open circle, shift from
semi-aerobic to aerobic conditions for trxC mutant. The growth
of both strains was identical under semi-aerobic conditions. (b)
Zone of growth inhibition assay. SB1003 or the trxC mutant
cells were plated in top agar. A disk (5?5 mm) containing
t-BOOH (5 ml of 100, 50, 20 and 10 mM stocks from left to
right), H2O2 (5 ml of 100, 50, 20 and 10 mM stocks) or diamide (5 ml of 50, 20, 10 and 5 mM stocks) or Paraquat (5 ml
of 10, 5, 2 and 1 mM stocks) was added to the plates. The
zone of growth inhibition is shown in the plates after overnight
growth. For Paraquat, no clear inhibition zones were formed on
the plates. Similar results were obtained from three independent
experiments in duplicate. Mean values (mm) of the radius of the
inhibition zone are displayed in the table. t, Turbid zone, some
cells grown in the zone; c, clear zone, no cells grown in the
zone; –, no obvious inhibition zone.
http://mic.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 10:35:39
425
K. Li, E. Härtig and G. Klug
70 % in a wild-type culture, but only 20 % in a culture of
the trxC mutant. The trxC mutant also showed a higher
sensitivity against 0?5 mM Paraquat, but survival of both
the mutant and the wild-type strain was affected to the
same extent by 0?1 mM Paraquat. One hour after the
addition of 1?5 mM diamide, 62 % of the cells of a wild-type
culture were viable, but only 46 % of the cells of the trxC
mutant were viable. The trxC mutant also showed lower
survival rates after treatment with 1 or 0?5 mM diamide.
One hour after the addition of 0?6 mM t-BOOH, the
wild-type and the mutant strain showed survival rates of
about 87 and 53 %, respectively, while the survival rates of
the mutant and wild-type were similar at lower concentrations of t-BOOH. These results show that trxC mutant cells
are more sensitive to the agents generating oxidative stress
than the wild-type SB1003.
The observation from the determination of survival rate
was verified by the zone of growth inhibition (Fig. 4b).
The trxC mutant showed a decreased resistance to all
oxidative agents applied. While clear zones of growth
inhibition were observed when the disks were treated with
H2O2 or diamide, turbid inhibition zones formed around
the disks treated with t-BOOH or Paraquat.
Our results imply a function of thioredoxin 2 of R. capsulatus
in the defence against oxidative stress.
Role of thioredoxin 2 in expression of genes for
pigment-binding proteins and formation of photosynthetic complexes in R. capsulatus
We have previously characterized the trxA mutant TK1 of
R. sphaeroides, which produces less thioredoxin than a
wild-type strain. Strain TK1 accumulates less bacteriochlorophyll and less puf and puc mRNAs encoding pigmentbinding proteins after a transition from growth under high
oxygen tension to growth under low oxygen tension
compared with its parental wild-type strain (Pasternak
et al., 1999). When the trxC mutant of R. capsulatus was
subjected to the same change in growth conditions, it
accumulated similar levels of bacteriochlorophyll to that of
the parental wild-type strain, indicating that the same
amounts of photosynthetic complexes are formed in both
strains. However, the bacteriochlorophyll content increased
more rapidly in the trxC mutant than in the isogenic wildtype strain (Fig. 5). We also analysed the levels of puf and
puc mRNA in the trxC mutant. puf and puc are both
polycistronic operons, which harbour genes required for the
formation of pigment protein complexes (Alberti et al.,
1995; Choudhary & Kaplan, 2000; Zsebo & Hearst, 1984).
Using Northern blot analysis, we determined the level of
the 0?5 kb pucBA mRNA that encodes the two pigmentbinding proteins of the light-harvesting II complex. This
mRNA species is a stable processing product of the 2?3 kb
primary puc transcript. We also determined the level of
the 0?5 kb pufBA mRNA, which is a processing product
of the pufQBALMX primary transcript and encodes the
proteins of the light-harvesting I complex.
426
Fig. 5. Relative bacteriochlorophyll content of wild-type (filled
circle) and trxC mutant (open circle) during the shift from aerobic (time point zero) to semi-aerobic conditions. The bacteriochlorophyll specific absorbance of methanol/acetone extracts at
770 nm was normalized to the optical density of the culture at
660 nm. All values are means from at least three measurements.
The expression of puf and puc genes after a decrease in
oxygen tension differed in the trxC mutant and the wildtype (Fig. 6). The amount of the 0?5 kb pucBA mRNA
increased by a factor of 25 in the wild-type and a factor
of about 33 in the trxC mutant. The total pucBA mRNA
amount was about twice as high in the mutant, before
and after the transition to low oxygen tension. The
amount of the 0?5 kb pufBA mRNA increased by a factor
of approximately 6 in the wild-type but by a factor of
approximately 9 in the trxC mutant. The puf mRNA levels
during aerobic growth were about 1?5-fold higher in the
trxC mutant than in the wild-type strain (Fig. 6).
Our results reveal that the effect of the deletion of trxC in
R. capsulatus on the expression of puf and puc genes is
opposite to the effect of the trxA mutation in R. sphaeroides
(Pasternak et al., 1997).
Effects of oxygen on trxC expression
Our data show that thioredoxin 2 of R. capsulatus affects
the oxygen-dependent expression of photosynthesis genes.
To test how trxC expression reacts to changes in oxygen
tension in the environment, we monitored trxC expression under different growth conditions. Since the trxC
mRNA does not show up on Northern blots as a distinct
band (unpublished), we used a semiquantitative RT-PCR
approach for directly quantifying trxC transcript levels.
In addition, we used a trxC9–luxAB fusion harbouring
450 nt of sequence upstream of trxC for following the
expression of trxC.
As shown in Fig. 7(b), the trxC mRNA level strongly
increased after a shift of the cultures from high to low
oxygen tension. It has been reported that the trxA mRNA
level in R. sphaeroides shows a slight decrease under these
conditions (Pasternak et al., 1996). Likewise, we observed a
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 10:35:39
Microbiology 149
Thioredoxin 2 function in Rhodobacter
Fig. 6. Northern blot analysis of puf and puc mRNA levels in the wild-type and the trxC mutant after reducing oxygen tension
at time point 0. Total RNA (10 mg) isolated at the indicated time points was electrophoresed, blotted and hybridized to a pufor puc-specific DNA probe labelled with a-32P. One representative result is shown. RNA isolations and Northern analyses
were repeated at least three times, and the mean numbers of the relative increase in the puf or puc mRNA levels are given.
significant decrease for trxA expression in R. capsulatus in
the Northern blot analysis (Fig. 7a) and by semiquantitative RT-PCR (Fig. 7b). Thus, trxA and trxC expression are
regulated by oxygen tension in an opposite way. When we
expressed the trxC9–luxAB fusion in the wild-type strain,
we were able to confirm this effect of oxygen on trxC
expression. The luciferase activity increased by a factor of
about eight to nine 1?5 h after the transition and then
dropped again. When the cultures were shifted from low
to high oxygen tension, we observed a strong decrease in
luciferase activity (Fig. 7c).
It has been reported that reduced levels of R. sphaeroides
thioredoxin 1 correlate with lower puf and puc mRNA
levels at low oxygen tension (Pasternak et al., 1999). The
R. capsulatus trxC mutant, however, showed higher
amounts of puf and puc mRNA. Thus, the opposite effects
of thioredoxin 1 of R. sphaeroides and thioredoxin 2 of
R. capsulatus on oxygen-dependent expression of photosynthesis genes correlate with an opposite effect of oxygen
on expression of the corresponding genes.
DISCUSSION
Thioredoxins are present in all living organisms, and have
been isolated and characterized from a variety of prokaryotic
and eukaryotic cells. R. sphaeroides is a facultatively
phototropic bacterium that can rapidly adapt to changing
environmental conditions. It contains a single gene for
thioredoxin, trxA, which is essential (Pasternak et al., 1997)
and affects the oxygen-dependent expression of photosynthesis genes (Pasternak et al., 1999). The molecular
mechanisms by which thioredoxin affects the transcription
of photosynthesis genes remain to be elucidated.
http://mic.sgmjournals.org
R. capsulatus, a close relative of R. sphaeroides, contains at
least two thioredoxin genes, trxA and trxC. To determine
whether thioredoxin 2 from R. capsulatus functionally
resembles thioredoxin 2 of E. coli and to test its involvement
in the oxygen-dependent expression of photosynthesis
genes, we inactivated the trxC gene of R. capsulatus and
expressed it in the E. coli strain BH216. Since strain BH216,
which lacks a functional trxA gene but harbours the
intact trxC gene, cannot grow on minimal medium with
methionine sulfoxide and cannot propagate phage T7, E. coli
TrxC is not able to replace TrxA for its functions as
hydrogen donor for methionine sulfoxide reductase and as
a subunit of T7 DNA polymerase when expressed at a
normal cellular level. When the trxC gene is overexpressed
in an E. coli metEtrxA background, it confers weak growth
to this strain on minimal medium with methionine
sulfoxide (Ritz et al., 2000; Stewart et al., 1998), indicating
that it can complement the defect poorly. We made the
identical observation after expressing the trxC gene from
R. capsulatus in an E. coli metEtrxA background. This
strain was unable to allow propagation of phage T7,
indicating that thioredoxin 2 is unable to function as a
subunit for T7 DNA polymerase, even when the trxC gene
is overexpressed.
The inactivation of the trxC gene had little effect on the
growth of R. capsulatus, even under a high oxygen tension.
An extended lag phase of the mutant strain after increasing
the oxygen tension in the culture indicates the involvement
of thioredoxin 2 in adaptation to high oxygen tension. This
assumption is supported by the fact that no growth of
the mutant was observed 1 h after a transition from low
to high oxygen tension. More pronounced differences
between the wild-type strain and the trxC mutant of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 10:35:39
427
K. Li, E. Härtig and G. Klug
Fig. 7. Effect of oxygen tension on trx expression in R. capsulatus. The oxygen tension was altered at time point zero. (a)
Northern blot analysis for decreasing trxA mRNA level of R.
capsulatus after a shift from high to low oxygen tension. (b)
RT-PCR analysis for trxC and trxA mRNA levels of R. capsulatus after a shift from high to low oxygen tension. Total RNA
was isolated at the time points indicated. Northern hybridization
was performed with a a-32P-labelled trxA specific fragment of
R. capsulatus. The identical RNA samples were used for RTPCR (see Methods) with specific primers designed for trxA and
trxC genes. (c) Luciferase activity assay for trxC expression in
R. capsulatus SB1003. Squares, shift from low to high oxygen
tension; diamonds, shift from high to low oxygen tension. All
experiments were repeated at least three times, and one typical
result was represented.
R. capsulatus were observed in the presence of oxidativestress-generating agents (Table 2). The mutant strain
showed a higher sensitivity against diamide, Paraquat,
t-BOOH and H2O2 than the parental wild-type strain. An
E. coli trxC mutant showed a sensitivity similar to that of
the isogenic wild-type after exposure to high concentrations of H2O2 (5 mM) or 0?8 mM diamide (Ritz et al.,
2000). Our data suggest that thioredoxin 2 of R. capsulatus,
like its E. coli counterpart (Ritz et al., 2000), is involved in
the oxidative stress response but is more important in
the resistance against diamide and H2O2 than its E. coli
counterpart. The way in which thioredoxin 2 is regulated
by oxidative stress and the molecular mechanisms of
oxidative stress defence in R. sphaeroides or R. capsulatus
are under investigation.
E. coli thioredoxin 1 is active in reconstituting T7 DNA
polymerase activity by forming a complex with the gene
5 protein in its reduced form; thioredoxin-S2 shows no
428
activity or competition (Adler & Modrich, 1983). The trxA
mutant strains in which one or both active-site cysteine
residues of E. coli thioredoxin had been changed to serine
or alanine can be lysed by T7 phage (Huber et al., 1986),
indicating that conformation of the reduced thioredoxin
at the active site is more important than whether or not
the active site can provide the redox potential. Interestingly,
the C29A mutant of R. sphaeroides thioredoxin 1 could not
act as a subunit of T7 DNA polymerase, whereas the C29A
mutant of R. capsulatus thioredoxin 1 could. R. capsulatus
thioredoxin 1 has the same amino acids KM directly behind
the second cysteine of the active site as E. coli thioredoxin 1.
However, R. sphaeroides thioredoxin 1 has the amino acids
RQ following the second cysteine of the active site, as
R. capsulatus and E. coli thioredoxin 2, which were found
not to act as a subunit of T7 DNA polymerase. These two
residues KM are important in protein interactions and
may stabilize thiolate in the active site (Eklund et al., 1991).
This suggests that the thioredoxin 1 mutation C29A of
R. capsulatus might not change the conformation at the
active site, probably partly contributed by the following
residues KM. The R. sphaeroides thioredoxin 1 C29A
mutation might change it. It is conceivable that the
flexibility of the active site region to allow the reduced
form of the protein to take up functionally significant
conformations of a slightly higher energy than the oxidized
form is more subtle in R. sphaeroides thioredoxin 1 than
in R. capsulatus thioredoxin 1.
Although thioredoxin 1 and 2 have similar functions in
some regards, the two genes respond to changes in oxygen
tension in an opposite way. While a reduction in oxygen
tension in cultures of R. capsulatus resulted in a significant
increase in trxC mRNA level, the trxA mRNA level
decreased. Interestingly, the opposite response of the two
genes to oxygen tension correlates with an opposite effect
on the oxygen-dependent expression of photosynthesis
genes. A trxA mutant of R. sphaeroides harbouring lower
levels of thioredoxin 1 than the isogenic wild-type showed
a diminished increase in puf and puc mRNA levels after
such a transition when compared with the parental wildtype strain (Pasternak et al., 1999). Until now we were
unable to construct a trxA mutant of R. capsulatus even
by applying exactly the same strategy as that used for
constructing the R. sphaeroides strain with an altered
thioredoxin 1 level. The trxC mutant of R. capsulatus
showed a faster increase and a higher accumulation of
puf and puc mRNA levels after a reduction in oxygen
tension than the isogenic wild-type strain. We conclude
that the two thioredoxins have opposite effects on some
cellular functions. At present, it is unknown how the effect
of thioredoxin on the expression of photosynthesis genes
is exerted. Thioredoxins may influence the activity of
some of the proteins (RegB/RegA, CrtJ; see introduction),
which have been identified as regulators of photosynthesis
gene expression in R. capsulatus. We are in the process of
unravelling the molecular basis for the effect of thioredoxin
on the expression of photosynthesis genes to learn more
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 10:35:39
Microbiology 149
Thioredoxin 2 function in Rhodobacter
about such differing functions of thioredoxin 1 and
thioredoxin 2, which may also be of importance for other
cellular functions.
Fernando, M. R., Nanri, H., Yoshitake, S., Nagata-Kuno, K. &
Minakami, S. (1992). Thioredoxin regenerates proteins inactivated
by oxidative stress in endothelial cells. Eur J Biochem 209, 917–922.
Gregor, J. & Klug, G. (2002). Oxygen-regulated expression of genes
for pigment binding proteins in Rhodobacter capsulatus. J Mol
Microbiol Biotechnol 4, 249–253.
ACKNOWLEDGEMENTS
Huber, H. E., Russel, M., Model, P. & Richardson, C. C. (1986).
Kuanyu Li was a recipient of a fellowship from DAAD. This work
has been supported by Deutsche Forschungsgemeinschaft (Kl563/7-4,
Kl563/16-1) and Fonds der Chemischen Industrie.
Interaction of mutant thioredoxins of Escherichia coli with the gene 5
protein of phage T7. The redox capacity of thioredoxin is not
required for stimulation of DNA polymerase activity. J Biol Chem
261, 15006–15012.
Huber, H. E., Tabor, S. & Richardson, C. C. (1987). Escherichia coli
REFERENCES
thioredoxin stabilizes complexes of bacteriophage T7 DNA polymerase and primed templates. J Biol Chem 262, 16224–16232.
Adler, S. & Modrich, P. (1983). T7-induced DNA polymerase.
Hübner, P., Masepohl, B., Klipp, W. & Bickle, T. A. (1993). nif gene
Requirement for thioredoxin sulfhydryl groups. J Biol Chem 258,
6956–6962.
expression studies in Rhodobacter capsulatus: ntrC-independent
repression by high ammonium concentrations. Mol Microbiol 10,
123–132.
Alberti, M., Burke, D. H. & Hearst, J. E. (1995). Structure and
sequence of the photosynthesis gene cluster. In Anoxygenic
Photosynthetic Bacteria, pp. 1083–1106. Edited by R. E. Blankenship, M. T. Madigan & C. E. Bauer. Dordrecht: Kluwer.
Assemat, K., Alzari, P. M. & Clement-Metral, J. D. (1995).
Conservative substitutions in the hydrophobic core of Rhodobacter
sphaeroides thioredoxin produce distinct functional effects. Protein
Sci 4, 2510–2516.
Barany, F. (1985). Two-codon insertion mutagenesis of plasmid
genes by using single-stranded hexameric oligonucleotides. Proc Natl
Acad Sci U S A 82, 4202–4206.
Boschi-Muller, S., Azza, S. & Branlant, G. (2001). E. coli methionine
sulfoxide reductase with a truncated N terminus or C terminus, or
both, retains the ability to reduce methionine sulfoxide. Protein Sci
10, 2272–2279.
Boschi-Muller, S., Azza, S., Sanglier-Cianferani, S., Talfournier, F.,
Van Dorsselear, A. & Branlant, G. (2000). A sulfenic acid enzyme
intermediate is involved in the catalytic mechanism of peptide
methionine sulfoxide reductase from Escherichia coli. J Biol Chem
275, 35908–35913.
Bradford, M. M. (1976). A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the
principle of protein–dye binding. Anal Biochem 72, 248–254.
Buchanan, B. B. (1984). The ferredoxin/thioredoxin system: a key
element in the regulatory function of light in photosynthesis.
Bioscience 34, 378–383.
Buchanan, B. B., Schurmann, P. & Jacquot, J. P. (1994).
Thioredoxin and metabolic regulation. Semin Cell Biol 5, 285–293.
Chae, H. Z., Chung, S. J. & Rhee, S. G. (1994). Thioredoxin-
dependent peroxide reductase from yeast. J Biol Chem 269, 27670–
27678.
Choudhary, M. & Kaplan, S. (2000). DNA sequence analysis of the
photosynthesis region of Rhodobacter sphaeroides 2.4.1. Nucleic Acids
Res 28, 862–867.
Keen, N. T. & Tamaki, S. (1988). Improved broad-host-range
plasmids for DNA cloning in Gram-negative bacteria. Gene 70, 191–197.
Kovach, M. E., Phillips, R. W., Elzer, P. H., Roop, R. M. & Peterson,
K. M. (1994). pBBR1MCS: a broad-host-range cloning vector.
Biotechniques 16, 800–802.
Kunert, A., Hagemann, M. & Erdmann, N. (2000). Construction of
promoter probe vectors for Synechocystis sp. PCC 6803 using the
light-emitting reporter systems Gfp and LuxAB. J Microbiol Methods
41, 185–194.
Laemmli, U. K. (1970). Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227, 680–685.
Laurent, T. C., Moore, E. C. & Reichard, P. (1964). Enzymatic
synthesis of deoxyribonucleotides. IV. Isolation and characterization
of thioredoxin, the hydrogen donor from Escherichia coli. J Biol
Chem 239, 3436–3444.
Lillig, C. H., Prior, A., Schwenn, J. D., Aslund, F., Ritz, D., VlamisGardikas, A. & Holmgren, A. (1999). New thioredoxins and
glutaredoxins as electron donors of 39-phosphoadenylylsulfate
reductase. J Biol Chem 274, 7695–7698.
Lim, C. J., Daws, T., Gerami-Nejad, M. & Fuchs, J. A. (2000).
Growth-phase regulation of the Escherichia coli thioredoxin gene.
Biochim Biophys Acta 1491, 1–6.
Mark, D. & Richardson, C. C. (1976). Escherichia coli thioredoxin: a
subunit of bacteriophage T7 DNA polymerase. Proc Natl Acad Sci
U S A 73, 780–784.
Masuda, S., Dong, C., Swem, D., Setterdahl, A. T., Knaff, D. B. &
Bauer, C. E. (2002). Repression of photosynthesis gene expression by
formation of a disulfide bond in CrtJ. Proc Natl Acad Sci U S A 99,
7078–7083.
Masuda, S., Matsumoto, Y., Nagashima, K. V., Shimada, K., Inoue,
K., Bauer, C. E. & Matsuura, K. (1999). Structural and functional
Das, K. C. & Das, C. K. (2000). Thioredoxin, a singlet oxygen
analyses of photosynthetic regulatory genes regA and regB from
Rhodovulum sulfidophilum, Roseobacter denitrificans, and Rhodobacter
capsulatus. J Bacteriol 181, 4205–4215.
quencher and hydroxyl radical scavenger: redox independent
functions. Biochem Biophys Res Commun 277, 443–447.
Miranda-Vizuete, A., Damdimopoulos, A. E., Gustafsson, J.-A. &
Spyrou, G. (1997). Cloning, expression, and characterization of a
Drews, G. (1983). Mikrobiologisches Praktikum. Berlin: Springer.
novel Escherichia coli thioredoxin. J Biol Chem 272, 30841–30847.
Eklund, H., Gleason, F. K. & Holmgren, A. (1991). Structural and
Mitsui, A., Hirakawa, T. & Yodoi, J. (1992). Reactive oxygen-reducing
functional relations among thioredoxins of different species. Proteins
11, 13–28.
and protein-refolding activities of adult T cell leukemia-derived factor/
human thioredoxin. Biochem Biophys Res Commun 186, 1220–1226.
Engler-Blum, G., Meier, M., Frank, J. & Müller, G. A. (1993).
Nakamura, H., Matsuda, M., Furuke, K. & 7 other authors (1994).
Reduction of background problems in nonradioactive Northern and
Southern Blot analyses enables higher sensitivity than 32P-based
hybridizations. Anal Biochem 210, 235–244.
Adult T cell leukemia-derived factor/human thioredoxin protects
endothelial F-2 cell injury caused by activated neutrophils or
hydrogen peroxide. Immunol Lett 42, 75–80.
http://mic.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 10:35:39
429
K. Li, E. Härtig and G. Klug
Natsuyama, S., Noda, Y., Narimoto, K., Umaoka, Y. & Mori, T.
(1992). Release of two-cell block by reduction of protein disulfide
Saitoh, M., Nishitoh, H., Fujii, M., Takeda, K., Tobiume, K., Sawada,
Y., Kawabata, M., Miyazono, K. & Ichijo, H. (1998). Mammalian
with thioredoxin from Escherichia coli in mice. J Reprod Fertil 95,
649–656.
thioredoxin is a direct inhibitor of apoptosis signal-regulating kinase
(ASK) 1. EMBO J 17, 2596–2606.
Orr, M. D. & Vitols, E. (1966). Thioredoxin from Lactobacillus
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning:
leichmannii and its role as hydrogen donor for ribonucleoside
triphosphate reductase. Biochem Biophys Res Commun 25, 109–115.
a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold
Spring Harbor Laboratory.
Pasternak, C., Assemat, K., Breton, A. M., Clement-Metral, J. D. &
Klug, G. (1996). Expression of the thioredoxin gene (trxA) in
Schulze-Osthoff, K., Schenk, H. & Droge, W. (1995). Effects of
Rhodobacter sphaeroides Y is regulated by oxygen. Mol Gen Genet
250, 189–196.
Pasternak, C., Assemat, K., Clement-Metral, J. D. & Klug, G. (1997).
Thioredoxin is essential for Rhodobacter sphaeroides growth by
aerobic and anaerobic respiration. Microbiology 143, 83–91.
Pasternak, C., Haberzettl, K. & Klug, G. (1999). Thioredoxin is
involved in oxygen-regulated formation of the photosynthetic
apparatus of Rhodobacter sphaeroides. J Bacteriol 181, 100–106.
Pille, S., Chuat, J.-C., Breton, A. M., Clement-Metral, J. D. & Galibert,
F. (1990). Cloning, nucleotide sequence, and expression of the
Rhodobacter sphaeroides Y thioredoxin gene. J Bacteriol 172,
1556–1561.
Ponnampalam, S. N. & Bauer, C. E. (1997). DNA binding
characteristics of CrtJ. A redox-responding repressor of bacteriochlorophyll, carotenoid, and light harvesting-II gene expression in
Rhodobacter capsulatus. J Biol Chem 272, 18391–18396.
Ritz, D. & Beckwith, J. (2001). Roles of thiol-redox pathways in
bacteria. Annu Rev Microbiol 55, 21–48.
Ritz, D., Patel, H., Doan, B., Zheng, M., Aslund, F., Storz, G. &
Beckwith, J. (2000). Thioredoxin 2 is involved in the oxidative stress
response in Escherichia coli. J Biol Chem 275, 2505–2512.
Russel, M. & Model, P. (1985). Thioredoxin is required for
filamentous phage assembly. Proc Natl Acad Sci U S A 82, 29–33.
430
thioredoxin on activation of transcription factor NF-kappa B.
Methods Enzymol 252, 253–264.
Sganga, M. W. & Bauer, C. E. (1992). Regulatory factors controlling
photosynthetic reaction center and light-harvesting gene expression
in Rhodobacter capsulatus. Cell 68, 945–954.
Shi, J., Vlamis-Gardikas, A., Aslund, F., Holmgren, A. & Rosen, B. P.
(1999). Reactivity of glutaredoxins 1, 2, and 3 from Escherichia coli
shows that glutaredoxin 2 is the primary hydrogen donor to ArsCcatalyzed arsenate reduction. J Biol Chem 274, 36039–36042.
Spector, A., Yan, G. Z., Huang, R. R., McDermott, M. J., Gascoyne, P. R.
& Pigiet, V. (1988). The effect of H2O2 upon thioredoxin-enriched
lens epithelial cells. J Biol Chem 263, 4984–4990.
Stewart, E. J., Aslund, F. & Beckwith, J. (1998). Disulfide bond
formation in the Escherichia coli cytoplasm: an in vivo role reversal
for the thioredoxins. EMBO J 17, 5543–5550.
Tomimoto, H., Akiguchi, I., Wakita, H., Kimura, J., Hori, K. & Yodoi, J.
(1993). Astroglial expression of ATL-derived factor, a human
thioredoxin homologue, in the gerbil brain after transient global
ischemia. Brain Res 625, 1–8.
Yen, H. C. & Marrs, B. (1976). Map of genes for carotenoid and
bacteriochlorophyll biosynthesis in Rhodopseudomonas capsulata.
J Bacteriol 126, 619–929.
Zsebo, K. M. & Hearst, J. E. (1984). Genetic-physical mapping of a
photosynthetic gene cluster from Rhodobacter capsulata. Cell 37,
937–947.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 10:35:39
Microbiology 149