Download CitA (citrate) and DcuS (C 4 -dicarboxylate) sensor kinases in

1
1
CitA (citrate) and DcuS (C4-dicarboxylate) sensor kinases in thermophilic
2
Geobacillus kaustophilus and G. thermodenitrificans
3
4
Running title: CitA sensor kinases of Geobacillus
5
6
7
Sabrina Graf, Constanze Broll, Juliane Wissig, Alexander Strecker, Maria Parowatkin,
8
Gottfried Unden*
9
10
Institute for Microbiology and Wine Research, Johannes Gutenberg University Mainz, 55099
11
Mainz, Germany
12
13
14
15
16
*For correspondence:
17
Dr. G. Unden, University of Mainz, Institute for Microbiology and Wine Research
18
Becherweg 15, 55099 Mainz, Germany
19
Phone: +49-6131-3923550
20
[email protected]
21
2
1
The thermophilic Geobacillus thermodenitrificans and Geobacillus kaustophilus are able to
2
use citrate or C4-dicarboxylates like fumarate or succinate as the substrates for growth. The
3
genomes of the sequenced Geobacillus strains (9 strains) each encoded a two-component
4
system of the CitA family. The sensor kinase of G. thermodenitrificans (termed CitAGt) was
5
able to replace CitA of E. coli (CitAEc) in a heterologous complementation assay restoring
6
expression of the CitAEc dependent citC-lacZ reporter gene and anaerobic growth on citrate.
7
Complementation was specific for citrate. The sensor kinase of G. kaustophilus (termed
8
DcuSGk) was able to replace DcuSEc of E. coli. It responded in the heterologous expression
9
system to C4-dicarboxylates and to citrate, suggesting that DcuSGk is like DcuSEc a C4-
10
dicarboxylate sensor with a side-activity for citrate. DcuSGk required unlike the homologous
11
DctS from B. subtilis no binding protein for function in the complementation assay. Thus the
12
thermophilic G. thermodenitrificans and G. kaustophilus contain citrate and C4-dicarboxylate
13
sensor kinases of the CitA and DcuS-type, respectively, and retain function and substrate
14
specificity under mesophilic growth conditions in E. coli.
15
16
17
3
1
Introduction
2
Bacteria of various taxa are able to grow at the expense of C4-dicarboxylates or citrate
3
(Scheu et al., 2010; Unden & Kleefeld, 2004; Kröger 1980; Kröger et al., 1992; Bott, 1997).
4
The metabolic pathways for aerobic and anaerobic growth on C4-dicarboxylates and citrate
5
are well characterized whereas induction of the corresponding pathways has been studied
6
only in a limited number of bacteria. Induction of C4-dicarboxylate metabolism is
7
accomplished by three types of two-component sensor systems, DcuS-DcuR, DctSRc-DctR and
8
DctBsm-DctR. Citrate catabolism is induced by the citrate responsive two-component system
9
CitA-CitB. DcuS-DcuR of enteric bacteria with sensor kinase DcuS represents the prototype of
10
a C4-dicarboxylate responsive two components system (Janausch et al., 2002b; Scheu et al.,
11
2010; Zientz et al., 1998; Golby et al. 1999). The DcuS sensor kinases constitute together
12
with the citrate sensor CitA the CitA family of histidine kinases. The C4-dicarboxylate sensor
13
kinases DctSRc from Rhodobacter capsulatus and DctBSm from Sinorhizobium meliloti, on the
14
other hand, are members of the FixL and NtrB families of sensor kinases, respectively
15
(Hamblin et al., 1993; Reid & Poole, 1998; Valentini & Lapouge, 2013; Janausch et al., 2002b;
16
Scheu et al., 2010).
17
Sensors of the CitA family of γ-proteobacteria are membrane integral and have a common
18
domain structure (Bott, 1997; Scheu et al., 2010). Sensing of the substrates is achieved by an
19
extracytoplasmic PASP type sensor domain that is flanked by transmembrane helices TM1
20
and TM2 (Kaspar et al., 1999; Pappalardo et al., 2003; Kneuper et al., 2005). On the
21
cytoplasmic side, TM2 is followed by a second PAS domain (PASC) that transmits the signal to
22
the C-terminal kinase domain. DcuS of E. coli (DcuSEc) responds to C4-dicarboxylates like
23
fumarate, succinate or L-malate, and with lower sensitivity to citrate (Zientz et al., 1998;
24
Golby et al., 1999; Kneuper et al., 2005). DcuS requires the tranporters DctA under aerobic
4
1
or DcuB under anaerobic conditions as co-regulators (Davies et al., 1999; Steinmetz et al.,
2
2014; Witan et al., 2012; Kleefeld et al., 2009). B. subtilis contains the sensor kinase DctSBs
3
resembling DcuSEc in domain composition and function (Asai et al., 2000; Graf et al., 2014).
4
DctSBs requires like DcuSEc the transporters DctABs, and an extra-cytoplasmic substrate
5
binding protein for function and response to the C4-dicarboxylates (Graf et al., 2014).
6
The citrate sensor CitA is found in bacteria that are able to use citrate as the C- and energy
7
source. CitA is highly specific for citrate and defined by the prototypic CitA sensor kinases of
8
Klebsiella and E. coli (Bott et al., 1995; Bott, 1997; Kaspar et al., 1999). CitA of the
9
proteobacteria, Corynebacterium and the homologous CitS of B. subtilis share the domain
10
composition with the DcuS sensor kinases (Bott et al., 1995; Brocker et al., 2009; Yamamoto
11
et al., 2000). The CitA protein of E. coli functions unlike DcuSEc as a stand-alone sensor
12
without the need for accessory proteins (Scheu et al., 2012).
13
Screening the genome sequences of thermophilic bacteria revealed that various Geobacillus
14
and Deinococcus strains encode sensor kinases of the CitA family with features indicating
15
that C4-dicarboxylate (DcuS-type) and citrate (CitA-type) sensor kinases are present in
16
different strains. Thus sensor kinases from G. thermodenitrificans and G. kaustophilus
17
representing CitA- and DcuS-type sensor kinases, respectively, were selected for
18
characterization and comparison to the corresponding sensor kinases from proteobacteria.
19
G. thermodenitrificans and G. kaustophilus are thermophilic with temperature optima at 60
20
and 55°C, respectively (Narzina et al., 2001). The bacteria that are known to grow on citrate,
21
were tested for growth on C4-dicarboxylates. For identification as CitA or DcuS-type systems,
22
heterologous complementation was used in dcuS or citA deficient strains of E. coli. The E. coli
23
strains were tested for gain of growth on citrate and C4-dicarboxylates, and for the
24
expression of DcuS and CitA dependent reporter genes in response to C4-dicarboxylates or
5
1
citrate.
2
3
Methods
4
Bacteria and molecular genetics methods. Geobacillus strains, derivates of E. coli and
5
plasmids used are listed in Table 1. The molecular methods were performed according to
6
standard procedures (Sambrook & Russel, 2001). All plasmids were isolated via GeneJETTM
7
Plasmid Miniprep Kit and PCR products were purified using the GeneJETTM PCR Purification
8
Kit (Fermentas). Oligonucleotides were synthesized by Eurofins MWG. For transformation of
9
E. coli electroporation (Dower et al., 1988) or heat shock were applied. Antibiotics were used
10
at the following concentrations: 100 µg/mL ampicillin, 20 µg/mL chloramphenicol, 50 µg/mL
11
kanamycin and 15 µg/mL tetracycline. When two or more antibiotics were used the
12
concentrations were halved.
13
Bioinformatics. Blastp and DELTA blast were used for screening all non-redundant GenBank
14
sequences (including cDNA sequence translations, PDB, SwissProt, PIR and PRF data bases,
15
but excluding environmental samples from whole genome shotgun sequencing projects,
16
with 69 159 658 sequences (version 2015/07/14).
17
Overproduction and isolation of His6-DcuSGk and His6-CitAGt, and autophosphorylation.
18
DcuSGk and His6-CitAGt were overproduced from pMW817 or pMW960, respectively, in E. coli
19
C43(DE3) as N-terminal His6 fusion proteins after cloning of dcuSGk and citAGt in pET28a
20
(Table 1). The proteins were isolated from the membrane fraction of the bacteria by
21
extraction with 2 % Empigen in purified by Ni-NTA chromatography in 0.04% LDAO
22
containing buffer by the procedure described by Janausch et al. (2002a) for DcuSEc. 13.5 mg
23
per liter of DcuSGk and 4 mg per liter of CitAGt were purified. For the autophosphorylation
6
1
assay, the proteins were reconstituted in liposomes (protein:lipid ration 1:20, w/w)
2
produced from E. coli phospholipids (E. coli Polar Lipid Extract, Avanti Polar Lipids) as
3
described by Janausch et al. (2002a). Prior to reconstitution, the liposomes were frozen in
4
liquid N2 and thawed at room temperature three times. Then the liposomes were
5
destabilized by Triton X-100 (detergent:lipid ratio of 2.5 (Rigaud et al., 1988, 1995), mixed
6
with the protein solution and the detergent was removed by Bio-Beads SM-2 (Bio-Rad)
7
(Holloway, 1973; Janausch et al., 2002a). The proteoliposomes were sedimented by
8
ultracentrifugation, dissolved in buffer and stored in liquid N 2 (Janausch et al., 2002a).
9
Autophosphorylation of the proteins was tested with different concentrations of [γ33P]-ATP
10
(see Fig. 5) and varying times of incubation. The liposomes were then dissolved in SDS-
11
containing buffer, subjected to SDS-PAGE. Radioactivity associated with the bands of DcuS or
12
CitA was determined by exposure of the gels to a Phosphor imager plate (BAS-MP2040,
13
Fujifilm) and evaluated by Fluorescent Imagereader FLA7000 (Fujifilm). For quantitative
14
evaluation slices of the SDS PAGE with the labelled proteins were digested and counted for
15
radioactivity by scintillation counting. From the radioactivity, the specific radioactivity and
16
the amount of DcuS or CitA used in the experiment, the labeling was calculated as described
17
by Janausch et al., (2002a).
18
ß-galactosidase assays. Expression of the dcuB-lacZ and citC-lacZ reporter gene fusions was
19
measured as the ß-galactosidase activity of exponential growing E. coli (ΔOD578 nm 0.5 to 0.8).
20
Cells were cultivated in 96-deep-well plates, anaerobically at 37 °C under an atmosphere of
21
N2 in enriched mineral (eM9) medium supplemented with acid-hydrolyzed Casamino acids
22
(0.1 %), L-tryptophan (0.005 %), and glycerol (50 mmol/L) plus dimethyl sulfoxide (DMSO)
23
(20 mmol/L) as the substrates. The effectors (citrate, glucose, fumarate, L-malate and
24
succinate (20 mmol/L)) were included as indicated. β-Galactosidase activity (Miller, 1992)
7
1
was quantified in 96-well-microtiter plates (Monzel et al., 2013). Optical density (570 nm)
2
and β-galactosidase activity (415 nm) were measured in a volume of 250 µl per well. For cell
3
permeabilization, 200 µl cell culture were mixed with 800 µl buffer (0.1 mol/L potassium
4
phosphate, 10 mmol/L potassium chloride, 1 mmol/L magnesium chloride, 0.005% (w/v)
5
cetyltrimethylammonium bromide, 0.0025 % (w/v) sodium deoxycholate and 0.027 % (v/v)
6
2-mercaptoethanol). For the assay, 150 µl of the permeabilized cells were incubated with 30
7
µl of ortho-nitrophenyl-ß-galactoside solution (0.4 % w/v) at 30 °C. After 20 min the reaction
8
was stopped with 70 µl sodium carbonate (1 mol/L). The β-galactosidase activity was
9
determined in triplicate in three induction experiments.
10
Growth experiments. Growth experiments with G. kaustophilus and G. thermodenitrificans
11
were performed in supplemented White minimal medium (White, 1972). For G. kaustophilus
12
the White-medium (WM-Gk) was enriched with acid-hydrolyzed Casamino acids (0.1 %), L-
13
tryptophan (0,01 %) and 0,1 % yeast extract. G. thermodenitrificans was cultivated in WM-
14
medium enriched with acid-hydrolyzed Casamino acids (0.1 %), L-tryptophan (0.01 %), 0.1 %
15
meat extract and 0.5 % sodium chloride (WM-Gt). For growth the media were supplemented
16
with 20 mM of the substrates as indicated. The bacteria were grown at 60 °C (G.
17
thermodenitrificans) or 55 °C (G. kaustophilus). For anaerobic growth the media were
18
supplemented with 50 mM glucose with or without the indicated electron acceptor (50 mM
19
or as indicated). Growth experiments with E. coli were performed anaerobically at 37°C in
20
enriched M9 medium (Kim et al. 2007) containing gluconate (3 mM), glycerol (50 mM) and
21
dimethyl sulfoxide as growth substrates plus effector (20 mM) as indicated.
22
23
24
Results
8
1
Genes encoding two-component systems of the CitA family in Geobacillus
2
The genomes of the Geobacillus strains that are available in databases were blasted for
3
genes encoding members of the CitA family of sensor kinases with the sequences of DcuS
4
and CitA of E. coli as the query. All strains (9) encoded homologs of the CitA family
5
(similarities >51 %; identity >29 %) with the respective domain composition. The sensor
6
kinases were members either of the DcuS (G. kaustophilus, G. subterraneus, and G.
7
thermoglucosidasius), CitA (G. thermodenitrificans, G. vulcani, G. stearothermophilus, G.
8
caldoxylosilyticus, G. thermoleovorans) sensor kinases according to sequence similarity. The
9
corresponding genes of G. thermodenitrificans and G. kaustophilus encoding citA and dcuS
10
like homologs, respectively, were selected for analysis. The G. thermodenitrificans genes
11
GTNG_1840 and GTNG_1839 (Fig. 1A) are arranged in a predicted operon and code for
12
proteins with similarity to CitA and CitB from E. coli. The genes are located downstream of
13
gene cluster tctCBA coding for proteins similar to a tripartite tricarboxylate transport system.
14
In G. kaustophilus a gene cluster of dcuS dcuR like genes was located upstream of genes for a
15
dicarboxylate binding protein similar to the binding protein DctB of B. subtilis and a C4-
16
dicarboxylate transporter DctA (Fig. 1A). The effector binding regions of the PASP domains of
17
DcuSEc and CitAKp are known (Gerharz et al., 2003; Reinelt et al., 2003; Kneuper et al., 2005;
18
Cheung & Hendrickson, 2008). The corresponding domains in the proteins of G. kaustophilus
19
and G. thermodenitrificans show 67% and 65% similarity, respectively to the domains of
20
DcuSEc and CitAEc. The domains reveal signature sequences specific for C4-dicarboxylate
21
(DcuSEc or DctSBs) or citrate (CitA) binding (Fig. 1B). The signature sequence is composed of
22
residues that are common for for CitA- and DcuS-type sensors, and additional residues (M122,
23
S144, K152 and S167, numbering of K. pneumonia CitA) that are specific for CitA-like proteins.
24
The conserved residues are mostly ligands for citrate or C4-dicarboxylate binding in the
9
1
citrate or L-malate co-crystals of the CitA and DcuS PASP domains (Reinelt et al., 2003;
2
Cheung & Hendrickson, 2008). The sensor kinase from G. thermodenitrificans contained nine
3
of the CitA specific residues, whereas that of G. kaustophilus carried the C4-dicarboxylate
4
motif and lacked the CitA specific residues. The similarities suggest that G.
5
thermodenitrificans encodes a citrate sensor of the CitA-type (CitAGt) and G. kaustophilus a
6
C4-dicarboxylate sensor of the DcuS-type (DcuSGk).
7
8
Growth on C4-dicarboxylates and citrate
9
Citrate, various sugars and polyols have been shown among others to support growth of G.
10
thermodenitrificans (strain DSM 466) and G. kaustophilus (Nazina et al., 2004; Manachini et
11
al., 2000). In the experiment of Fig. 2 the bacteria were tested for growth on C4-
12
dicarboxylates that had not been used as substrates before. Under aerobic conditions
13
succinate or fumarate supported growth of G. thermodenitrificans but the final cell densities
14
were significantly lower than on glucose. Anaerobic conditions allowed only poor growth on
15
glucose which was neither stimulated by fumarate nor nitrate. G. kaustophilus grew under
16
aerobic conditions with C4-dicarboxylates including L-malate, and the cell densities reached
17
54 % or more of growth on glucose (Fig. 2B). Under anaerobic conditions, glucose enabled
18
significant growth, provided that nitrate was included as an electron acceptor. Overall, G.
19
kaustophilus and G. thermodenitrificans are able to use C4-dicarboxylates for aerobic growth
20
in addition to the citrate that was demonstrated earlier as a substrate. The lack of growth on
21
L-malate by G. thermodenitrificans might be caused by the absence of specific L-malate
22
transporters of the MaeN and YflS type. The transporters are essential in some bacilli for L-
23
malate utilization since L-malate is not transported by the general C4-dicarboxylate
24
transporter DctA in the bacteria (Tanaka et al. 2003).
10
1
2
Citrate specific complementation of E. coli citA mutants by the citAGt gene of Geobacillus
3
The supposed citAGt and dcuSGk genes (or CitAGt and DcuSGk proteins) of G.
4
thermodenitrificans and G. kaustophilus, respectively, were tested for their ability to
5
complement citA or dcuS deficient strains of E. coli. The supposed CitAGt was tested in the E.
6
coli citA mutant for its capacity to restore expression of the CitA-CitB dependent citC-lacZ
7
reporter gene fusion (Fig. 3A) and anaerobic growth on citrate (Fig. 3B). The coding regions
8
of citA and citB of E. coli overlap by 32 bp, and therefore inactivation of citA causes lack of
9
citB gene translation. For complementation by citAGt, the bacteria were therefore supplied
10
by citAGt and by full length citBEc on plasmid, and both genes were under the control of
11
inducible promoters.
12
In the citA citB negative strain the expression of the citC-lacZ reporter drops to background
13
levels (Fig. 3A). When citAGt was supplied on plasmid (together with citBEc), citrate caused
14
induction of citC-lacZ expression that exceeded the induction by complementation with
15
citAEc. Complementation by both citA variants required the presence of citBEc (not shown)
16
and for both variants the induction was specific for citrate whereas fumarate or L-malate
17
produced only a low response. Thus the citAGt gene encodes a sensor kinase which is able to
18
substitute CitAEc, and CitAGt is citrate specific in the heterologous system. The E. coli citA
19
mutant has lost anaerobic growth on citrate whereas anaerobic growth by fumarate
20
respiration is retained as expected (Fig. 3B). CitAGt restored the growth on citrate with an
21
efficiency similar to CitA of E. coli (Fig. 3B) indicating that citrate transport and citrate lyase
22
are produced in the presence of CitAGt. The heterologous complementation of CitAEc by
23
CitAGt suggests that CitAGt functionally interacts with CitBEc. Overall, the data shows that
24
CitAGt is a citrate specific sensor that is able to replace CitAEC in the heterologous system and
11
1
interacts with CitBEc. This finding is in agreement with the citrate dependent growth of the
2
bacteria, but the missing response of CitAGt to fumarate suggests that the growth of G.
3
thermodenitrificans on C4-dicarboxylates is constitutive or independent from CitAGt.
4
The citAGt gene was also used to test complementation of a dcuS deficient strain of E. coli
5
(Fig. 3C). Plasmid encoded citAGt allowed induction of the DcuS-DcuR dependent dcuB-lacZ
6
reporter gene at high levels which even exceeded that by plasmid encoded dcuS of E. coli.
7
However, restoration of dcuB-lacZ expression was maximal with citrate when CitAGt was
8
present, whereas restoration was maximal with fumarate in the presence of DcuSEc.
9
Therefore CitAGt is apparently able to interact with and phosphorylate DcuR. Remarkably, in
10
the CitAGt-DcuREc containing bacteria the specificity for the stimulus (citrate) is that of the
11
senor CitAGt. DcuR retains the specificity for its target (dcuB promoter) demonstrating that
12
the heterologous complementation involves a cross-talk between a CitA sensor kinase and a
13
DcuR response regulator. There is, however, no cross-talk between the E.coli CitAEc and
14
DcuREc (Fig. 3C) since CitAEc provided on plasmid is not able to complement DcuS deficiency.
15
16
Complementation of dcuS of E. coli by dcuS of G. kaustophilus
17
In the same way complementation of an E.coli dcuS mutant was tested by a plasmid
18
encoding the supposed DcuSGk protein (Fig. 4A). The test strain (E. coli IMW260) is deficient
19
of dcuS but proficient for chromosomally encoded dcuR, and contains a (DcuS-DcuR
20
dependent) dcuB-lacZ reporter gene fusion. The strain lacks dcuB-lacZ expression (Fig. 4A).
21
Complementation with plasmid encoded dcuSEc restored expression of dcuB-lacZ by C4-
22
dicarboxylates like fumarate and L-malate, and in agreement with earlier reports to lower
23
extents by citrate (Zientz et al., 1998; Krämer et al., 2007). Plasmid encoded dcuSGk was able
24
to restore expression of dcuB-lacZ, but the effector specificity was significantly different.
12
1
Thus L-malate and citrate stimulated the expression most efficiently, followed by fumarate.
2
Therefore DcuSGk is similar to DcuSEc (and different from the CitA proteins) by the broad
3
specificity, but is has a high sensitivity to citrate as well, exceeding that for fumarate.
4
DcuSEc shows a high fumarate independent background activity in the expression of dcuB-
5
lacZ when the transporters DctA or DcuB are missing. The transporters function as co-
6
regulators of DcuS and infer responsiveness for C4-dicarboxylates to DcuS (Davies et al.,
7
1999; Kleefeld et al., 2009; Witan et al., 2012; Steinmetz et al., 2014). Thus in
8
complementation by DcuSGk, a high background activity of induction of dcuB-lacZ was
9
observed in the absence of inducer (C4-dicarboxylates) (Fig. 4A), resembling the situation
10
when in DcuS+ strains transporters DcuB or DctA are deleted. This observation might be an
11
indication that DcuSGk requires like DcuSEc a co-regulator for adopting the ground-state and
12
for full C4-dicarboxylate responsiveness, and that DctAEc cannot serve this function entirely.
13
Deletion of DcuS causes diminished aerobic growth of E. coli on fumarate or L-malate by
14
decreased expression of dctA encoding the aerobic C4-dicarboxylate transporter DctA
15
(Davies et al., 1999). Thus the decreased growth of the dcuS mutant on fumarate or L-malate
16
was restored by plasmid encoded dcuSEc (Fig. 4B) and to nearly the same extent by dcuSGk.
17
Therefore DcuSGk is able to restore the aerobic growth deficiency of a dcuS mutant by
18
activating dctA expression in E. coli.
19
The data altogether indicate that DcuSGk is a typical DcuS-type sensor kinase. Some sensor
20
kinases of this type, exemplified by DctSBs of B. subtilis require the function of an
21
extracytoplasmic binding protein in addition (Asai et al., 2000; Graf et al., 2014). Thus in
22
agreement with earlier data (Graf et al., 2014) DctSBs alone is not able to complement for
23
DcuSEc deficiency and to restore dcuB-lacZ expression in the presence of fumarate or L-
24
malate (Fig. 4A). This finding is in contrast to the efficient complementation by DcuS Gk
13
1
indicating that the latter functions independent of a binding protein.
2
3
DcuSGk phosphorylation
4
For an initial characterization of one of the thermophilic sensor kinases,
5
autophosphorylation of DcuSGk was studied after purification and reconstitution of the
6
protein in liposomes. DcuSGk was overproduced heterologously in E. coli as a His6-DcuSGk
7
fusion protein. The protein was solubilized from the membrane fraction with detergent
8
LDAO and purified to near homogeneity by Ni-NTA-chromatography (Fig. 5A). The purified
9
DcuSGk showed only very weak autophosphorylation in the presence of [γ33P]ATP. After
10
incorporation into membranes produced from E. coli phospholipids, the protein was
11
autophosphorylated efficiently (Fig. 5B). The degree of autophosphorylation exceeded that
12
of DcuSEc when treated under comparable conditions at 37°C. For a more quantitative study,
13
autophosphorylation of DcuSGk and of DcuSEc was performed in the presence of increasing
14
concentrations of [γ33P]ATP (Fig. 5C). The degree of phosphorylation was determined using
15
the specific radioactivity of the ATP mixture, and after separating the protein by SDS PAGE
16
from non-bound [33P]. The radioactivity and phosphorylation in the bands corresponding to
17
DcuSEc or DcuSGk was calculated from radioactivity incorporated and the specific radioactivity
18
as described earlier for DcuSEc (Janausch et al., 2002a). Phosphorylation reached saturation
19
at high concentrations of [γ33P]ATP. With 10 mM [γ33P]ATP about 18% of the DcuSGk was
20
phosphorylated whereas only 2.2% of the DcuSEc were phosphorylated after reaching
21
maximal phosphorylation. The concentrations for half-maximal phosphorylation were
22
approx. 43 µM and 420 µM ATP for DcuSGk and DcuSEc, indicating that DcuSGk when
23
produced in E. coli and tested at 37°C is active and exceeds DcuSEc in activity and affinity.
24
14
1
Relation of CitAGt and DcuSGk to sensor kinases of the CitA-, DctSRc- and DctBSm-families.
2
The three major classes of C4-dicarboxylate sensor kinases are represented by the CitA/DcuS,
3
DctSRc and the DctBSm sensor kinases. The CitA family is characterized by sequence similarity
4
to CitAKp and the domain composition with a PASP, two TM helices, PASC and the kinase
5
domain (Bott et al., 1995; Zientz et al., 1998). DcuSGk and CitAGt are by domain composition
6
and sequence similarity members of the CitA family. The DctSRc type sensor kinases are
7
defined by the Rhodobacter capsulatus DctSRc that belongs to the FixL family of sensor
8
kinases. DctSRc type sensor kinases are found in R. capsulatus (α-proteobacteria) and
9
Aromatoleum aromaticum (β-proteobacteria) (Hamblin et al., 1993; Trautwein et al. 2012;
10
Scheu et al. 2010). DctSRc has a predicted domain composition similar to CitA or DcuS with
11
two transmembrane helices, a periplasmic and a cytoplasmic PAS domain, and the C-
12
terminal kinase (UniProt (Magrane & Consortium, 2011)). The periplasmic PAS domains of
13
DcuSEc and DctSRc, however, are only distantly related (Golby et al., 1999; Krämer et al.,
14
2007).
15
The γ-proteobacteria Vibrio cholerae and Pseudomonas aeruginosa, and the α-
16
proteobacterium Sinorhizobium meliloti contain C4-dicarboxylate sensor kinases of the
17
DctBSm type (Reid & Poole, 1998; Valentini et al., 2011). DctBSm belong to the NtrB family of
18
sensor kinases (Janausch et al., 2002b; Scheu et al., 2010; Valentini et al., 2011) and contains
19
tandem extracytoplasmic PASP domains with low similarity to DcuSEc and CitAKp. The
20
cytoplasmic part is composed of a coiled coil CC domain and the kinase.
21
For a more detailed analysis the sequences of the extracytoplasmic PASP domains of the C4-
22
dicarboxylate or tricarboxylate sensor kinases were clustered by CLANS (Frickey & Lupas,
23
2004) using the sequences of the PASP domains of DcuSEc, DctSRc, DctSBs and DctBSm in the
24
PSI-Blast. Sequences with an E-value cutoff of 10 (default) and 10 iterations were used
15
1
resulting in the CitA/DcuS, DctSRc, DctB clusters of sensor kinases and two clusters of
2
guanylate cyclases (Fig. 6). CitA/DcuS represents the largest group and contains the
3
prototypic DcuS and CitA sensor kinases from enteric bacteria. DcuSGk of G. kaustophilus and
4
CitAGt of G. thermodenitrificans and the sensor kinases termed DctS (including DctSBs) and
5
CitS of the Gram-positive Bacilli (Yamamoto et al., 2000), Lactobacilli and Clostridia are
6
members of the CitA/DcuS cluster. The MalK malate sensor kinases from Bacilli (Tanaka et al.
7
2003), Streptomyces and Clostridia, and MaeK of Lactobacillus casei (Landete et al., 2010)
8
are part of the CitA/DcuS cluster as well (Fig. 6). Additionally, four strains of the
9
Deinococcus/Thermus group and two Meiothermus strains contained CitA homologs.
10
Remarkably, all of the sensor kinases of the CitA family that were characterized so far cluster
11
in one subgroups within this family (Fig. 6). The bacteria of the second subgroup (left-hand
12
side within the CitA/DcuS cluster) without characterized CitA or DcuS proteins are mostly
13
from the Actinomycetales group of grampositive bacteria. The proteins of the DctSRc cluster
14
represent the smallest group and comprise the DctS proteins of the proteobacteria.
15
The sequence similarity of the PASP domains of DcuSEc and other members within the CitA
16
family is typically higher than 50% (e.g. CitAEc:DcuSEc 62%, CitSBc:DcuSEc 59% similarity),
17
whereas that of DcuSEc with DctSRc is as low as 35%. In agreement with their separate
18
clustering, the PASP domain DctBSm (distal PASP of the PASP tandem structure) shares only 48
19
and 35% similarity with the domains of DcuSEc and DctSRc, respectively
20
21
22
Discussion
23
In Geobacilli two-component systems of the CitA/DcuS family are wide-spread and are in
24
domain composition and sequence similar to the DcuS and CitA proteins from
25
proteobacteria. The cluster analysis shows that the DcuS/CitA group which was originally
16
1
defined by the proteins of proteobacteria, includes many two-component systems of Gram-
2
positive bacteria including those from lactobacilli, bacilli, clostridia, corynebacteria,
3
geobacilli, deinococcus and the actinomycetales. Indeed, a large number of the known C4-
4
dicarboxylate/citrate sensor kinases in this group are from grampositive bacteria, whereas
5
the C4-dicarboxylate sensor kinases from the proteobacteria are found in the DctSRc and
6
DctBSm clusters as well.
7
Heterologous complementation of a dcuS deficient mutant of E. coli by dcuS of G.
8
kaustophilus confirmed the functional similarity of DcuSGk with DcuSEc. In the heterologous
9
complementation DcuSGk showed functional interaction with DcuREc. Interestingly, the
10
mesophilic growth conditions in E. coli and replacement of the Geobacillus lipid composition
11
by the E. coli lipids obviously had no severe effects on the function of the Geobacillus DcuSGk
12
and CitAGt in E. coli.
13
The substrate specificities of DcuSGk was in agreement with that expected by the presence of
14
the C4-dicarboxylate (‘DcuS’) signature in the binding sites as defined in Fig. 1. Thus DcuSGk is
15
a typical DcuS sensor with broad specificity to C4-dicarboxylates and to citrate. G.
16
kaustophilus is able to grow aerobically on C4-dicarboxylates and on citrate. The lack of an
17
additional CitA-CitB system, suggests that DcuSGk-DcuRGk is responsible for induction of both
18
metabolic systems, or that citrate metabolism is constitutively induced. Genes encoding a
19
fumarase (FumC) and DctA next to the genes for the two-component system support the
20
role of DcuSGk-DcuRGk in the control of C4-dicarboxylate metabolism.
21
DcuS-type sensor kinases require transporters like DctA or DcuB as co-regulators, and DcuSEc
22
and DctSBs are in the permanent ON state when the transporters are missing (Davies et al.
23
1999; Kleefeld et al. 2009; Witan et al. 2012; Steinmetz et al. 2014). The high background
24
activity of dcuB-lacZ expression after complementation by DcuSGk in the absence of
17
1
effectors, indicates that DcuSGk requires also a transporter for adjusting the OFF or C4-
2
dicarboxylate responsive state. The E. coli transporters presumably cannot fully replace the
3
G. kaustophilus transporters in this respect. Sensor kinase DctSBs of B. subtilis requires an
4
additional extracytoplasmic binding protein for function (Asai et al., 2000; Graf et al., 2014).
5
The high activity of DcuSGk in the absence of a binding protein indicates that DcuSGk functions
6
independent of an extracytoplasmic binding protein. Overall, it appears that the DcuS-like
7
protein in G. kaustophilus shares many properties with the corresponding sensor kinases of
8
E. coli and that it is independent of an extracytoplasmic binding protein known from bacilli.
9
CitAGt, on the other hand, has the typical properties of a CitA type sensor kinase and the
10
signature of a citrate binding site. Remarkably, cross-talk between non-cognate systems
11
(CitAGt and DcuREc) was observed whereas in the homologous system (CitAEc with DcuREc)
12
cross-talk was lacking in agreement with earlier suggestions (Scheu et al., 2012). Therefore in
13
the heterologous system obviously some specificity in the sensor kinase/response regulator
14
interaction is lost.
15
The data suggests that the CitAGt-CitBGt two-component system is suitable for inducing
16
metabolism by citrate and growth on citrate which is supported by the gene cluster tctABC
17
that is located adjacent to the citAB genes and encodes a citrate transporter. Genome
18
analysis shows that only the CitA type sensor is present in G. thermodenitrificans but no
19
DcuS type sensor kinase. Therefore the relatively weak growth on C4-dicarboxylates depends
20
on constitutive expression of the corresponding metabolism, or the function of an additional
21
unknown system.
22
23
24
Acknowledgements. Financial support by a grant of Deutsche Forschungsgemeinschaft to
18
1
GU (UN49/17-1) is gratefully acknowledged. We are grateful to Drs. J. Schultz, A. Lupas and J.
2
Baßler (Tübingen) for introduction to and help with CLANS.
3
4
5
References
6
7
Asai, K., Baik, S. H., Kasahara, Y., Moriya, S. & Ogasawara, N. (2000). Regulation of the
transport system for C4-dicarboxylic acids in Bacillus subtilis. Microbiology 146, 263–271.
8
9
Bott, M. (1997). Anaerobic citrate metabolism and its regulation in enterobacteria. Arch
Microbiol 167, 78-88.
10
11
Bott, M., Meyer, M. & Dimroth, P. (1995). Regulation of anaerobic citrate metabolism of
Klebsiella pneumoniae. Mol Microbiol 18, 533-546.
12
13
14
Brocker, M., Schaffer, S., Mack, C. & Bott, M. (2009). Citrate utilization by Corynebacterium
glutamicum is controlled by the CitAB two-component system through positive
regulation of the citrate transport genes citH and tctCBA. J Bacterial 191, 3869-3880.
15
16
17
Cheung, J. & Hendrickson, W. A. (2008). Crystal structures of C4-dicarboxylate ligand
complexes with sensor domains of histidine kinases DcuS and DctB. J Biol Chem 283,
30256–30265.
18
19
20
Davies, S. J., Golby, P., Omrani, D., Broad, S. A., Harrington, V.L., Guest, J. R., Kelly, D. J. &
Andrews, S. C. (1999). Inactivation and regulation of the aerobic C4-dicarboxylate
transport dctA gene of Escherichia coli. J Bacteriol 181, 5624–5635.
21
22
Dower, W. J., Miller, J.F. & Ragsdale, C. W. (1988). High efficiency transformation of E. coli
by high voltage electroporation. Nucleic Acids Res 16, 6127–6145.
23
24
Frickey, T. & Lupas, A. (2004) CLANS: A Java application for visualizing protein families based
on pairwise similarity. Bioinf 21, 3702-3704.
25
26
27
Gerharz, T., Reinelt, S., Kaspar, S., Scapozza, L. & Bott, M. (2003). Identification of basic
amino acid residues important for citrate binding by the periplasmic receptor domain of
the sensor kinase CitA. Biochemistry, 42, 5917–5924.
28
29
30
31
Golby, P., Davies, S., Kelly, D. J., Guest, J. R., Andrews, S. C. (1999). Identification and
characterization of a two-component sensor-kinase and response-regulator system
(DcuS-DcuR) controlling gene expression in response to C4-dicarboxylates in Escherichia
coli. J Bacteriol 181, 1238–1248.
32
33
34
Graf, S., Schmieden, D., Tschauner, K., Hunke, S., Unden, G. (2014). The sensor kinase DctS
forms a tripartite sensor unit with DctB and DctA for sensing C4-dicarboxylates in
Bacillus subtilis. J Bacteriol 196, 1084-1093.
35
36
37
Guzman, L. M., Belin, D., Carson, M.J. & Beckwith, J. (1995). Tight regulation, modulation,
and high-level expression by vectors containing the arabinose pBAD promoter. J
Bacteriol 177, 4121–4130.
38
Hamblin, M. J., Shaw, J. G. & Kelly, D. J. (1993). Sequence analysius and interposon
19
1
2
3
mutagenesis of a sensor-kinase (DctS) and response-regulator (DctR) controlling
synthesis of the high-affinity C4-dicarboxylate transport system in Rhodobacter
capsulatus. Mol Gen Genet 237, 215-224.
4
5
6
Heeb, S., Itoh, Y., Nishijyo, T., Schnider, U., Keel, C., Wade, J., Walsh, U., O’Gara, F. Haas, D.
(2000). Small, stable shuttle vectors based on the minimal pSV1 replicon for use in
gram-negative, plant-associated bacteria. Mol Plant Microbe Interact 13, 232-237.
7
8
Holloway, P. W. (1973). A simple procedure for removal of Triton X-100 from protein
samples. Anal Biochem 53, 933-939.
9
10
11
Janausch, I. G., Garcia-Moreno, I. & Unden, G. (2002a). Function of DcuS from Escherichia
coli as a fumarate-stimulated histidine protein kinase in vitro. J Biol Chem 277, 3980939814.
12
13
14
15
16
17
Janausch, I.G., Zientz, E., Tran, Q.H., Kröger, A. & Unden, G. (2002b). C4-dicarboxylate
carriers and sensors in bacteria. Biochim Biophys Acta (Reviews in Bioenergetics). 1553,
39-56.
Kaspar, S., Perozzo, R., Reinelt, S., Meyer, M., Pfister, K., Scapozza, L. & Bott, M. (1999).
The periplasmic domain of the histidine autokinase CitA functions as a highly specific
citrate receptor. Mol Microbiol 33, 858-872.
18
19
20
21
22
23
Kim, O. B., Lux, S. & Unden, G. (2007) Anaerobic growth of Escherichia coli on D-tartrate is
independent of D- or L-tartrate specific transporters and enzymes. Arch Microbiol 188,
583-589.
Kleefeld, A., Ackermann, B., Bauer, J., Krämer, J. & Unden, G. (2009). The
fumarate/succinate antiporter DcuB of Escherichia coli is a bifunctional protein with
sites for regulation of DcuS-dependent gene expression. J Biol Chem 284, 265–275.
24
25
26
Kneuper, H., Janausch, I. G., Vijayan, V., Zweckstetter, M., Bock, V., Griesinger, C. & Unden,
G. (2005). The nature of the stimulus and of the fumarate binding site of the fumarate
sensor DcuS of Escherichia coli. J Biol Chem 280, 20596–20603.
27
28
29
Kneuper, H., Scheu, P. D., Etzkorn, M., Sevvana, M., Dünnwald, P., Becker, S., Baldus, M.,
Griesinger, C. & Unden, G. (2010). Sensing ligands by periplasmic sensing histidine
kinases with sensory PAS domains. In Sensory mechanisms in bacteria, pp. 39–59.
30
31
32
33
Krämer, J., Fischer, J. D., Zientz, E., Vijayan, V., Griesinger, C., Lupas, A. & Unden, G. (2007).
Citrate sensing by the C4-dicarboxylate/citrate sensor kinase DcuS of Escherichia coli:
binding site and conversion of DcuS to a C4-dicarboxylate- or citrate-specific sensor. 189,
4290–4298.
34
35
Kröger, A. (1980). Bacterial electrn transport to fumarate. In: Knowles, C. J. (ed) Diversity of
bacterial respiratory system. CRC Press, Boca Raton, Florida, pp. 1-17.
36
37
Kröger, A., Geisler, V., Lemma, E., Theis, F. & Lenger, R. (1992). Bacterial fumarate
respiration. Arch Microbiol 158, 311-314.
38
39
40
41
Landete, J. M., Garcia-Haro, L., Blasco, A., Manzanares, P., Berbegal, C., Monedero, V. &
Zúniga, M. (2010) Requirement of the Lactobacillus casei MaeKR two-component
system for L-malic acid utilization via a malic enzyme pathway. Appl Environ Microbiol
76, 84-95.
42
43
Magrane, M. & Consortium, U. (2011). UniProt Knowledgebase: a hub of integrated protein
data. Data (Oxford) bar009
44
Manachini, P. L., Mora, D., Nicastro, G., Parini, C., Stackebrandt, E., Pukall, R. & Fortina, M.
20
1
2
G. (2000). Bacillus thermodenitrificans sp. nov., nom. rev. Int J Syst Evolut Microbiol 50,
1331-1337.
3
4
Miller, J. H. (1992). A short course in bacterial genetics. Cold Spring Harbor Laboratory, Cold
spring Harbor, N.Y.
5
6
7
Miroux, B. & Walker, J. E. (1996). Over-production of proteins in Escherichia coli: mutant
hosts that allow synthesis of some membrane proteins and globular proteins at high
levels. J Mol Biol 260, 289-298.
8
9
10
11
Monzel, C., Degreif-Dünnwald, P., Gröpper, C., Griesinger, C. & Unden, G. (2013). The
cytoplasmic PASC domain of the sensor kinase DcuS of Escherichia coli: role in signal
transduction, dimer formation, and DctA interaction. MicrobiologyOpen 2, 912-927.
doi:10.1002/mbo3.127
12
13
14
15
16
17
18
19
20
Nazina, T. N., Tourova, T. P., Poltaraus, A. B., Novikova, E. V., Grigoryan, A. A., Ivanova, A.
E., Lysenko, A. M., Petrunyaka, V. V., Osipov, G. A., Belyaev, S. S. & Ivanov, M. V.
(2001). Taxonomic study of aerobic thermophilc bacilli: descriptions of Geobacillus
subterraneus gen. nov., sp. nov. and Geobacillus uzenensis sp. nov. from petroleum
reservoirs and transfer of Bacillus stearothermophilus, Bacillus thermocatenulatus,
Bacillus thermoleovorans, Bacillus kaustophilus, Bacillus thermoglucosidasius and
Bacillus thermodenitrificans to Geobacillus as the new combinations G.
stearothermophilus, G. thermocatenulatus, G. thermoleovorans, G. kaustophilus, G.
thermoglucosidasius and G. thermodenitrificans. Int J Syst Evol Microbiol 51, 433-446.
21
22
23
24
Nazina, T. M., Lebedeva, E. V., Poltaraus, A. B., Tourova, T. P., Grigoryan, A. A., Sokolova,
D. S., Lysenko, A. M. & Osipov, G. A. (2004). Geobacillus gargensis sp. nov., a novel
thermophile from a hot spring, and the reclassification of Bacillus vulcani as Geobacillus
vulcani comb. nov. Int J Syst Evolut Microbiol 54, 2019-2024.
25
26
27
28
Pappalardo, L., Janausch, I. G., Vijayan, V., Zientz, E., Junker, J., Peti, W., Zweckstetter, M.,
Unden, G. & Griesinger, C. (2003). The NMR structure of the sensory domain oft he
membraneous two-component fumarate sensor (histidine protein kinase) DcuS of
Escherichia coli. J Biol Chem 278, 39185-39188.
29
30
31
Reid, C. J. & Poole, P. S. (1998). Roles of DctA and DctB in signal detection by the
dicarboxylic acid transport system of Rhizobium leguninosarum. J Bacteriol 180, 26602669.
32
33
34
Reinelt, S., Hofmann, E., Gerharz, T., Bott, M., and Madden, D.R. (2003). The structure of
the periplasmic ligand-binding domain of the sensor kinase CitA reveals the first
extracellular PAS domain. J Biol Chem 278, 39189–39196.
35
36
37
38
Rigaud, J. L., Paternostre, M. T. & Bluzat, A. (1988). Mechanisms of membrane protein
insertion into liposomes during reconstitution procedures involving the use of
detergents. 2. Incorporation of the light-driven proton pump bacteriorhodopsin.
Biochemistry 27, 2677-2688.
39
40
41
Rigaud, J.L., Pitard, B. & Levy, D. (1995). Reconstitution of membrane proteins into
liposomes: application to energy transducing membrane proteins. Biochim Biophys Acta
1231, 223-246.
42
43
Sambrook, J. & Russell, D. W. (2001). Molecular Cloning: A Laboratory Manual, 3rd edn.
New York: Cold Spring Harbor Laboratory Press.
21
1
2
3
Scheu, P. D., Kim, O. B., Griesinger, C. & Unden, G. (2010). Sensing by the membrane-bound
sensor kinase DcuS: exogenous versus endogenous sensing of C4-dicarboxylates in
bacteria. Future Microbiol 5, 1383–1402.
4
5
6
7
Scheu, P. D., Witan, J., Rauschmeier, M., Graf, S., Liao, Y. F., Ebert-Jung, A., Basché, T.,
Erker, W. & Unden, G. (2012). CitA/CitB two-component system regulating citrate
fermentation in Escherichia coli and its relation to the DcuS/DcuR system in vivo. J
Bacteriol 194, 636–645.
8
9
10
Sevvana, M., Vijayan, V., Zweckstetter, M., Reinelt, S., Madden, D., Herbst-Irmer, R.
Sheldrick, G.M., Bott, M., Griesinger, C. & Becker, S. (2008). A ligand-induced switch in
the periplasmic domain of sensor histidine kinase CitA. J Mol Biol 377, 512-523.
11
12
13
14
Sievers, F., Wilm, A., Dinee, D., Gibson, T. J., Karplus, K., Li, W., Lopez, R., McWilliam, H.,
Remmert, M., Söding, J., Thompson, J.D. & Higgins, D. G. (2011). Fast, scalable
generation of high-quality protein multiple sequence alignments using Clustal Omega.
Mol Syst Biol 7, 539: doi: 10.1038/msb.2011.75.
15
16
17
Steinmetz, P. A., Wörner, S. & Unden, G. (2014). Differentiation of DctA and DcuS function
in the DctA/DcuS sensor complex of Escherichia coli: function of DctA as an activity
switch and of DcuS as the C4-dicarboxylate sensor. Mol Microbiol 94, 218-229.
18
19
Studier, F.W. & Moffatt, B.A. (1986). Use of bacteriophage T7 RNA polymerase to direct
selective high-level expression of cloned genes. J Mol Biol 189, 113–130.
20
21
22
23
Tanaka, K., Kobayashi, K. & Ogasawara, N. (2003). The Bacillus subtilis YufLM twocomponent system regulates the expression of the malate transporters MaeN (YufR)
and YflS, and is essential for utilization of malate in minimal medium. Microbiology 149,
2317–2329.
24
25
26
Trautwein, K., Grundmann, O., Wöhlbrand, L., Eberlein, C., Boll, M. & Rabus R. (2012).
Benzoate mediates repression of C4-dicarboxylate utilization in Aromatoleum
aromaticum EbN1. J Bacteriol 194, 518-528.
27
28
29
30
Unden, G. & Kleefeld, A. (2004). Chapter 3.4.5. C4-dicarboxylate degradation in aerobic and
anaerobic growth. In R. Curtiss III (Editor in Chief), EcoSal - Escherichia coli and
Salmonella: Cellular and Molecular Biology. [Online] http://www.ecosal.org. ASM Press,
Washington, D.C.
31
32
Valentini, M., Storelli, N. & Lapouge, K. (2013) Identification of C4-dicarboxylate transport
systems in Pseudomonas aeruginosa PAO1. J Bacteriol 193, 4307-4316.
33
34
White, P. J. (1972). The nutrition of Bacillus megaterium and Bacillus cereus. J Gen Microbiol
71, 505–514.
35
36
37
Witan. J., Bauer, J., Wittig, I., Steinmetz, P. A., Erker, W. & Unden, G. (2012). Interaction of
the Escherichia coli transporter DctA with the sensor kinase DcuS: presence of functional
DctA/DcuS sensor units. Mol Microbiol 85, 846–861.
38
39
40
Yamamoto, H., Murata, M. & Sekiguchi, J. (2000). The CitST two-component system
regulates the expression of the Mg-citrate transporter in Bacillus subtilis. Mol Microbiol
37, 898–912.
41
42
43
Zientz, E., Bongaerts, J. & Unden, G. 1998. Fumarate regulation of gene expression in
Escherichia coli by the DcuSR (dcuSR genes) two-component regulatory system. J
Bacteriol 180, 5421–5425.
22
1
Figure legends
2
Figure 1. (A) The dcuS-dcuR and citA-citB gene clusters in G. kaustophilus and G.
3
thermodenitrificans and (B) sequence comparison of C4-dicarboxylate/citrate binding sites
4
in the extracytoplasmic PASP domains of CitA and DcuS proteins. (A) Genes and (putative)
5
encoded proteins: citT, citrate/succinate antiporter; citCDEFXG, citrate-lyase; citA, citrate
6
sensor kinase CitA; citB, citrate response regulator CitB; ydbD, manganese catalase; dctA, C4-
7
dicarboxylate transporter DctA; dctB, C4-dicarboxylate-binding protein DctB; dctR, C4-
8
dicarboxylate response regulator DctR, dctS, C4-dicarboxylate sensor kinase DctS; fumC,
9
fumarate hydratase FumC; GTNG_1838 probable transporter (40 % identitiy to transporter
10
MleP of Oenococcus oeni); tctABC, tripartite tricarboxylate transport system. (B) Signature
11
sequences in the ligand binding sites in the periplasmic PASP domains of DcuS and CitA
12
proteins. Residues involved in L-malate/fumarate or citrate binding (Cheung & Hendrickson,
13
2008; Sevvana et al., 2008; Gerharz et al., 2003; Kneuper et al., 2005) are highlighted:
14
Orange, L-malate/fumarate binding sites; blue, citrate binding sites; grey, residues
15
equivalent for fumarate and the citrate binding in DcuSEc and CitAKp/Ec. Sequence alignments
16
were performed with DcuSEc (P0AEC8), DcuSGk (Q5L2I6), DcuSGtg (A0A0F4E3S3), DctSBs
17
(P96601), CitAEc (P77510), CitAKp (P52687),CitAGt (A4IPE6), and CitAGst (A0A087LI38) using
18
Clustal Omega (Sievers et al., 2011). Ec, Escherichia coli; Gk, G. kaustophilus; Gtg, G.
19
thermoglucosidasius; Bs, B. subtilis; Kp, K. pneumoniae; Gt, G. thermodenitrificans; Gst, G.
20
stearothermophilus.
21
22
Figure 2. Growth of G. thermodenitrificans (A) and G. kaustophilus (B) on glucose and
23
carboxylates. The bacteria were grown at 60 °C (G. thermodenitrificans) or 55 °C (G.
24
kaustophilus) in enriched White minimal medium. The medium was enriched with 0.1% AHC,
23
1
0.01% L-tryptophane, 0.5% NaCl and 0.1% meat extract (G. thermodenitrificans), or with
2
0.1% yeast extract (G. kaustophilus). Growth substrates were added at 20 mM (citrate 5
3
mM) for aerobic growth as indicated, or 50 mM (citrate 5 mM) for anaerobic growth.
4
Growth was measured as the OD578nm after incubation for 7 hours (G. thermodenitrificans) or
5
16 hours (G. kaustophilus). Cit* indicates positive growth reaction on citrate agar.
6
7
Figure 3. Complementation of citAEc (A and B) and dcuSEc (C) by citAGt. For
8
complementation of citAEc (A+B), strain E. coli IMW549 (citA and citC-lacZ) was transformed
9
with a plasmid encoding CitAGt (pMW1652) and CitBEc (pMW1653) or CitA/BEc (pMW1599).
10
The citC-lacZ reporter gene expression was measured as indicated in the methods section.
11
For complementation of dcuSEc (C), strain E. coli IMW260 (dcuS and ΦdcuB-lacZ) was
12
transformed with a plasmid encoding DcuSEc (pMW151), CitAGt (pMW1652), or CitAEc
13
(pMW1651), and dcuB-lacZ reporter gene expression was measured. Reporter gene activities
14
(mean ± standard deviation) are the average of three biological replicates, and of at least
15
four independent measurements. In the complementation studies the strains were grown
16
anaerobically in eM9 medium containing gluconate (3 mM), glycerol (50 mM) and dimethyl
17
sulfoxide as growth substrates plus the effector (20 mM) citrate (black bars), fumarate (grey
18
bars), or without effector (white bar).
19
20
Figure 4. Complementation of dcuSEc by dcuSGk as tested by gene expression (A) and
21
growth (B). For testing complementation of dcuS dependent gene expression (A), E. coli
22
IMW260 (dcuS and ΦdcuB-lacZ) was grown anaerobically in enriched M9 medium with
23
glycerol/gluconate/DMSO as the basic substrates plus one of the inducers (20 mM each)
24
citrate (black bar), fumarate (light grey bar), malate (grey bars), or without (white bar). For
24
1
complementation, strain E. coli IMW260 was transformed with plasmids encoding DcuSEc
2
(pMW151), or DcuSGk (pMW817), or DctSBs (pMW1558). Reporter gene activities (mean ±
3
standard deviation) are the average of three biological replicates, and of at least four
4
independent measurements. For growth complementation (B), the same strain IMW260
5
strain was transformed plasmids encoding either DcuSEc (pMW151), or DcuSGk (pMW817).
6
Bacterial growth is given as the final OD578nm after growth on fumarate or L-malate for the
7
DcuS-negative strain and after complementation with DcuSEc or DcuSGk.
8
9
Figure 5. Purification (A), reconstitution in liposomes and autophosphorylation of His6-
10
DcuSGk and CitAGt (B, C). (A) DcuSGk and CitAGt solubilized and purified from E. coli
11
BL21DE3pMW817 and BL21DE3pMW960, respectively, were separated by SDS-PAGE (12%
12
acrylamide) and stained with Coomassie Blue (A). The SDS-PAGE shows protein marker, His6-
13
DcuSGk (15 µg) His6-CitAGt (10 µg) as eluted from the Ni-NTA-agarose column. (B)
14
Autophosphorylation of purified (Sol) and reconstituted (Rec) DcuSGk and DcuSEc. His6-DcuSGk
15
and His6-DcuSEc were reconstituted in E. coli liposomes. Samples of purified (Sol) and
16
reconstituted (Rec) protein (5 µg each) were incubated for 30 min with 0.1 µM [γ33P]ATP in
17
the presence of fumarate (20 mM). The samples were subject to SDS-PAGE (protein stain,
18
lower part of (B)) and autoradiography (upper part of (B)). (C) Phosphorylation of DcuSGk and
19
DcuSEc as a function of ATP concentration. DcuSGk or DcuSEc in proteoliposomes was
20
incubated with [γ33P]ATP (0.1 µM to 10 mM) for 30 min. The proteoliposomes were
21
dissolved in SDS sample buffer and subject to SDS-PAGE. After slicing the gel, the
22
radioactivity in the protein bands was determined by scintillation counting, and the level of
23
DcuS phosphorylation was calculated from radioactivity and protein content (Janausch et al.,
24
2002a).
25
1
2
Figure 6. Cluster analysis of C4-dicarboxylate sensor kinases based on the sequences of
3
DcuSEc, DcuSRc and DctBSm. The sequences of the binding (PASP) domains of the proteins
4
were used for the PSI-Blast analysis. Sequences of the PSI-Blast searches with an E-value
5
cutoff of 10 were combined and used for cluster analysis by CLANS (Frickey & Lupas, 2004).
6
The CitA cluster is presented in an extended version, and specific members are shown: 1
7
(red), MaeK of Lactobacillus casei (Landete et al., 2010); 2, (green) CitAGt of G.
8
thermodenitrificans; 3 (magenta), CitS of Clostridium ultunense; 4 (brown), CitAKp of K.
9
pneumonia; 5 (light green) DcuSEc; 6 (blue), MalK of C. intestinale; 7 (yellow) DctSBs; 8
10
(orange), DcuSGk of G. kaustophilus; 9 (light blue), DcuS of Deinococcus geothermalis, 10
11
(purple) Corynebacterium variabile.
12
26
1
2
Table 1: Strains and plasmids used in this study
Strain or plasmid
Strains
Geobacillus kaustophilus
DSM7263
Geobacillus
thermodenitrificans
DSM466
Escherichia coli K-12
BL21(DE3)
C43(DE3)
IMW260
IMW549
Plasmids
pET28a
pBAD33
pME6010
pMW151
pMW817
pMW960
pMW1558
pMW1599
pMW1601
pMW1651
pMW1652
pMW1653
3
4
5
Genotype or characteristic(s)
Ref or source
Wildtype
Nazina
2001
Nazina
2001
Wildtype
et
al.,
et
al.,
F-, ompT, hsdSBgal1, dcmλDE3, IPTGinducible chromosomal T7-RNA pol
Mutant of BL21(DE3), for expression of
membrane proteins
MC4100, λ(Φ(dcuB'-'lacZ)hyb, bla+, Δ
lacZ, dcuS::camR
IMW279 (citA::kanR), λ(Φ(citC'-'lacZ)
hyb, ampR
Studier & Moffat,
1986
Miroux & Walker,
1996
Zientz et al., 1998
Overexpression plasmid, His6-tag, kanR
Expression plasmid with pBR322 ori,
arabinose induction, camR
Low-copy plasmid, 8.3 kb, tetR
Novagen
Guzman et al.,
1995
Heeb et al., 2000
pET28a with dcuSEc-his6 kanR
pET28a with dcuSGk-his6 kanR
pET28a with citAGt-his6 kanR
pHT304 with dctSBs with own promotor
and ribosome binding site, ampR (E.
coli) eryR (B. subtilis)
pME6010, with citA citB-his6 behind
citAEc promoter, tetR
pME6010 with intergenic region
upstream of citAEc
pMW1601, with his6-citAEc behind citAEc
promoter, tetR
pMW1601, with his6-citAGt behind citAEc
promoter, tetR
pBAD33, with citBEc-his6 and N-terminal
ribosome-binding-site, camR
This study
This study
This study
This study
Scheu et al., 2012
This study
This study
This study
This study
This study