• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jbacterPermissionsJournals.ASM.orgJournalJB ArticleJournal InfoAuthorsReviewers
J Bacteriol. May 2009; 191(9): 2944–2952.
Published online Mar 6, 2009. doi:  10.1128/JB.00074-09
PMCID: PMC2681794

Potassium Transport in Corynebacterium glutamicum Is Facilitated by the Putative Channel Protein CglK, Which Is Essential for pH Homeostasis and Growth at Acidic pH[down-pointing small open triangle]


We studied the requirement for potassium and for potassium transport activity for the biotechnologically important bacterium Corynebacterium glutamicum, which is used for large-scale production of amino acids. Different from many other bacteria, at alkaline or neutral pH, C. glutamicum is able to grow without the addition of potassium, resulting in very low cytoplasmic potassium concentrations. In contrast, at acidic pH, the ability for growth was found to depend on the presence of K+. For the first time, we provide experimental evidence that a potential potassium channel (CglK) acts as the major potassium uptake system in a bacterium and proved CglK's function directly in its natural membrane environment. A full-length CglK protein and a separate soluble protein harboring the RCK domain can be translated from the cglK gene, and both are essential for full CglK functionality. As a reason for potassium-dependent growth limitation at acidic pH, we identified the impaired capacity for internal pH homeostasis, which depends on the availability and internal accumulation of potassium. Potassium uptake via CglK was found to be relevant for major physiological processes, like the activity of the respiratory chain, and to be crucial for maintenance of the internal pH, as well as for the adjustment of the membrane potential in C. glutamicum.

Potassium plays an essential role for bacterial cells and is the most abundant ion in the cytoplasm. Maintenance of high internal potassium concentration and, consequently, potassium uptake are of crucial physiological significance in bacteria and are key determinants for survival (5). Whereas in the cytoplasm of the gram-negative bacterium Escherichia coli about 400 mM potassium were detected, the gram-positive Bacillus subtilis and Corynebacterium glutamicum contain between 300 and 800 mM potassium, respectively (9, 24, 38). The high potassium content is in general accompanied by high concentrations of glutamate, which acts as counter ion (13, 24).

In general, potassium accumulation was described as a first response of bacterial cells toward hyperosmotic stress (6, 29), and a regulatory role in transcription and translation or a role as secondary messenger for activation of proteins was found (13). Potassium contributes to the global transcription regulation under osmotic stress conditions by interaction with the RNA polymerase and affects protein translation in E. coli and Streptococcus faecalis (12, 15, 17). In B. subtilis, potassium uptake and accumulation were found to be essential for osmotic stress tolerance because of potassium's impact on de novo protein synthesis (18, 37). For C. glutamicum, it was demonstrated that potassium specifically activates the glycine betaine carrier BetP under in vitro conditions (31).

Besides potassium's role during osmotic stress response in E. coli, potassium uptake and accumulation were found to be essential for the maintenance of internal pH as well (5). Under acidic conditions, a neutral pH in the cytoplasm can be maintained only if potassium is available. In the absence of potassium, the internal pH decreases concomitantly with a decreased external pH and pH homeostasis fails. The mechanism of potassium-dependent pH regulation and the impact of potassium on pH homeostasis in C. glutamicum is unknown.

For these purposes, bacteria are equipped with potassium uptake systems. The kind of potassium transport system found in a particular organism was proposed to be related to the availability of potassium in its natural habitat (34). In E. coli, three active potassium transport systems (Trk, Kdp, and Kup), as well as potassium channels (Kch and Kef), are found (13). Whereas the Trk transporter is a high-capacity, constitutively expressed potassium uptake system, Kup and Kdp represent inducible, relatively low-capacity uptake systems. The physiological role of potassium channels in E. coli is largely unknown (13). In B. subtilis, two Trk-related Ktr-type transporter systems are the major uptake carriers (18). The contribution of channels in potassium ion transport is unknown. For C. glutamicum, the requirement for potassium and the presence and properties of potassium carriers have not been investigated so far.

In this study, we provide experimental evidence that the potential potassium channel CglK is the only functional potassium uptake system in C. glutamicum. We thus prove the physiological significance of potassium transport by a bacterial channel protein in its natural membrane environment for the first time. In addition, we show that under acidic stress, uptake of potassium by CglK is essential for pH homeostasis in C. glutamicum cells.


Bacterial strains, growth, and construction of mutants.

E. coli DH5αMCR cells were grown in Luria-Bertani (LB) medium at 37°C and used for molecular cloning procedures. Strain ATCC 13032 served as the wild-type (WT) C. glutamicum. C. glutamicum cells were grown either in brain heart infusion (BHI) medium (Becton-Dickenson, Heidelberg, Germany) or in minimal medium MMI (21) at 30°C. Plates were prepared by the addition of 15 g liter−1 agar to the medium. For all experiments, C. glutamicum cells were precultivated in 5 ml BHI medium for approximately 8 h and subsequently used for inoculation of 20 ml minimal medium with different potassium concentrations. After approximately 20 h, the culture was used to inoculate fresh MMI medium, with the potassium concentrations indicated below, to an optical density at 600 nm (OD600) of 1 to 2, and experiments were started after cultures entered the exponential growth phase. Whereas BHI contains 10 mM potassium, MMI contains 37 mM potassium. The lowest potassium concentration applied in liquid minimal medium was 10 μM and on agar plates 50 μM, due to contamination by other chemicals. All strains were cultivated either in Erlenmeyer flasks with shaking at 130 rpm or in microtiter plates sealed with a gas-permeable membrane in a volume of 200 μl with shaking at 1,200 rpm. The medium contained the buffer substances (250 mM) indicated below in order to maintain the desired pH. If necessary, the medium was supplemented with kanamycin (25 μg ml−1). Growth was followed by measuring the OD600.

C. glutamicum deletion mutants were constructed as described previously (32). Standard molecular cloning techniques were used, and for amplification of flanking regions of the genes kup and cglK, the primers listed in Table Table11 were applied. The mutant lacking the kup gene (cg0187) is referred to as the Δkup mutant, the mutant lacking the cglK gene (cg0887) as the ΔcglK mutant, and the mutant lacking both kup and cglK as the Δkup ΔcglK mutant. The PCR fragments were cloned into the plasmid pDRIVE (Qiagen, Hilden, Germany), and the correct sequence was confirmed by sequencing (GATC, Konstanz, Germany). After ligation of the PCR fragments into vector pK18mobSacB, C. glutamicum cells were transformed by electroporation. After two rounds of selection, transformants were obtained and deletion of the particular gene was proven by PCR. For complementation, the cglK gene was amplified by PCR, using the primers indicated in Table Table1,1, and cloned into the vector pEKEX, resulting in plasmid pEKEXcglK. CglK variants were constructed by replacing the internal start codon at position 137 with a stop codon [cglK(M137Stop)] or a codon for the amino acid isoleucine [cglK(M137I)] (Table (Table1)1) by application of a Stratagene QuikChange site-directed mutagenesis kit. They were confirmed by sequencing, and after electroporation of C. glutamicum Δkup ΔcglK cells, the presence of the plasmid was proven by cultivation on kanamycin-containing plates, as well as by PCR. The resulting strains are listed in Table Table11.

Strains, plasmids, and primers used in this study

Measurement of cell volume, internal pH, membrane potential, and respiratory activity.

During the exponential phase of growth, cells were harvested, washed twice, and resuspended in MMI medium buffered at pH 7 (100 mM morpholinepropanesulfonic acid [MOPS]) or pH 6 (100 mM morpholineethanesulfonic acid [MES]). Cell volumes were determined by the distribution of 3H-labeled H2O (0.55 mCi/liter) and 14C-labeled inulin (0.14 mCi/liter) between the cell pellet and the supernatant (30). The membrane potential was determined by measuring the distribution of tetraphenylphosphonium (TPP) (final concentration, 5 μM, and specific radioactivity, 0.995 Ci/mol) as described previously (30). Processing of samples for rapid separation of extra- and intracellular fluids was performed by using silicone oil centrifugation with perchloric acid in the bottom layer (30). Internal pH was determined by measuring the distribution of 14C-labeled benzoic acid (final concentration, 15 μM, and specific radioactivity, 3.12 Ci/mol). Alternatively, the internal pH was determined by using the pH-sensitive fluorescent probe 2′,7′-bis-(2-carboxyethyl)-5-6-carboxyfluorescein (BCECF) as described previously (27). In short, 10 ml of a potassium-free C. glutamicum suspension (pH 7.5, OD600 of 5) was prepared and incubated with BCECF-acetoxymethylester (AM) (final concentration, 1.2 μM; Sigma-Aldrich, Germany) for 30 min at 30°C in the dark. BCECF-AM is membrane permeable and can be converted into the membrane-impermeable BCECF in the cytoplasm. After BCECF-AM was removed by washing, BCECF fluorescence was measured in 1-s intervals using a Aminco-Bowmann series 2 spectrometer (Spectronic Instruments, Urbana, IL) at 535 nm after excitation at 450 nm (pH insensitive) or 490 nm (pH sensitive). Calibration was performed by incubation of cells at external pH values between pH 5.5 and pH 7.5 in the presence of a mixture of carbonyl cyanide-3-chlorophenylhydrazone (CCCP), valinomycin, and nigericin (final concentrations of 50, 20, and 5 μM, respectively) in order to equilibrate internal and external pH values. The change of the cytoplasmic pH was followed after addition of cells to a potassium-free medium at pH 6 and subsequent addition of potassium.

Rates of oxygen consumption by C. glutamicum cells were measured with a Clark-type electrode (Oxygraph; Hansatech, Reutlingen, Germany) at 30°C in a total volume of 1 ml minimal medium, pH 6 (250 mM MES). After precultivation in potassium-free medium, exponentially grown cells (minimal medium, pH 7) were suspended at an OD600 of 0.3 to 0.5. After incubation for 5 min, constant rates of oxygen consumption were observed. Subsequently, potassium chloride (50 mM final concentration) was added, and 5 min later, valinomycin (20 μM final concentration). All measurements were performed at least in triplicate, and standard deviations were calculated.

Potassium uptake measurements.

Potassium uptake was quantified by monitoring both the external and the internal concentration by flame photometry (ELEX 6361; Eppendorf) or by measuring the initial uptake rates using radioactively labeled rubidium (86Rb) as a tracer. After precultivation in BHI and MMI medium, cells were washed three times with MMI medium containing 1 mM K+ and 250 mM N-(2-hydroxyethyl)piperazine-N′-(3-propanesulfonic) (HEPPS) buffer, pH 8.5, and subsequently inoculated at an OD600 of 2 in 100 ml of the same medium. For measurement of the external potassium concentration, cells were removed by centrifugation (20,000 × g at 30°C for 30 s) and the supernatant was analyzed by flame photometry. For measuring the internal potassium concentration, cells were collected on filter disks, washed with 5 ml medium, and soaked dry by use of a vacuum pump. Resuspension and cell disruption were carried out by adding 2 ml cetyl-trimethyl-ammonium bromide (CTAB; 0.1%), followed by shaking at room temperature for 30 min. After removal of cell debris via centrifugation, the potassium concentration in the supernatant was measured. For the cytoplasmic volume, 1.8 μl mg−1 cell dry matter (CDM) was determined as described previously (30).

For determination of initial potassium uptake rates, C. glutamicum strain DHPF (33) harboring or lacking the cglK and/or kup gene(s) was used (Table (Table1).1). The cells were precultivated, washed three times, and inoculated in potassium-free (strains DHPF, DHPFΔcglK, and DHPFΔkupΔcglK) and potassium-containing (100 mM; strain DHPF) MMI medium at an OD600 of 1.3. After 12 h at 30°C, cells were harvested by centrifugation, washed two times in cold buffer (25 mM NaPi, pH 7.5, 100 mM NaCl at 4°C) and suspended in the same buffer at an OD600 of 6. Transport assays were performed at 30°C at an OD600 of 3 in the presence of 20 mM glucose and 2.5 mM 86RbCl (0.045 mCi/liter). The internal accumulation of radioactive label was followed by fast filtration of 200-μl samples through glass fiber filters (product no. APFF02500; Millipore, Schwalbach, Germany) using a manifold filtration device (product no. FH225V; Hoefer, Holliston, MA). The samples were washed twice with 2.5 ml 0.1 M KCl and counted in a liquid scintillation counter (product no. LS-6500; Beckmann Coulter) using Rotiszinth-Plus (Carl Roth, Germany).

CglK expression in E. coli and purification by Ni-NTA affinity chromatography.

The cglK gene was amplified by PCR (see Table Table11 for primer sequences), cloned into the vector pET52b, and expressed in E. coli BL21 cells by the addition of 1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) (Table (Table1).1). After 5 h, cells were harvested, suspended in buffer A (100 mM NaCl, 100 mM KCl, 5 μg/ml DNase I, 15 mM imidazole, 50 mM Tris-HCl, pH 7.8), and broken by French press treatment, and the total protein fraction obtained by centrifugation (12,500 rpm at 4°C for 60 min) was subjected to batch purification using Ni-nitrilotriacetic acid (NTA)-agarose beads (1 ml; Qiagen, Hilden, Germany) according to the supplier's protocol. The imidazole concentration of the washing buffer was 30 mM and of the elution buffer was 250, 500, 750, or 1,000 mM. Equal volumes (15 μl) were used for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and the proteins were transferred onto a membrane and subjected to immunoblot analysis using an anti-His antibody (Qiagen).


Growth of C. glutamicum is impaired at low pH in a potassium-dependent manner.

We first addressed the minimal potassium requirement of C. glutamicum. We grew WT cells in minimal medium at potassium concentrations from 50 mM down to 10 μM, which represents the lowest potassium concentration resulting from contaminations under standard conditions (MMI medium, pH 7). Surprisingly, the growth rate of C. glutamicum was not affected by decreasing potassium concentration even after four rounds of inoculation of potassium-depleted cells into new medium in the absence of added potassium. The growth rates and final OD values were not different from those obtained for cells supplemented with 50 mM potassium (data not shown). We concluded that, in contrast to other bacteria, like Bacillus subtilis, C. glutamicum tolerates very low potassium concentrations in the medium (9, 18).

A particular requirement for potassium at low pH values was observed for several bacteria (5). Therefore, we inoculated C. glutamicum cells onto agar plates at different pH values containing 50, 1, or 0.05 mM potassium. On plates at pH 8.5, growth was independent of added potassium, and cells exposed to pH 7 were only slightly affected in the presence of the lowest potassium concentration of 0.05 mM. At pH 6, growth was affected even in the presence of 50 mM potassium, and at 0.05 mM potassium, no growth was observed at all (Fig. (Fig.1).1). These results demonstrate that C. glutamicum does not require external potassium under alkaline or neutral pH conditions, in contrast to the situation in an acidic environment.

FIG. 1.
Potassium-dependent growth of C. glutamicum WT and ΔcglK mutant at different pH values. Cells were depleted of potassium by precultivation in minimal medium without potassium supplementation and subsequently diluted to the OD values indicated ...

Subsequently, we addressed the transporter(s) which is (are) responsible for potassium uptake. Inspection of the C. glutamicum genome revealed the absence of well-known high-capacity systems like Trk or Ktr and of a high-affinity Kdp system. Instead, genes encoding a Kup-type transporter (cg0817) and a putative potassium channel (cg0887) were found (20). The putative Kup protein in C. glutamicum consists of 624 amino acids and harbors 12 predicted transmembrane domains and a C-terminal cytoplasmic domain of 181 amino acids. Its sequence is highly similar to that of the Kup transporter of E. coli (E-value, e−133), which probably functions as proton symporter and was found to be important during osmotic stress at low pH (36). The putative potassium channel, which we named CglK (C. glutamicum K+ channel), is a protein of 353 amino acids and is highly similar to the MthK channel of Methanobacterium thermoautotrophicum for which the three-dimensional structure was solved (10). Like the MthK protein, CglK harbors two transmembrane domains, and the P-loop region contains the canonical selectivity filter sequence TVGYGD. The C-terminal region of the CglK protein contains an RCK domain, which was proposed to be involved in regulation of channel opening and closing (26). For the MthK channel, a homotetramer of the full-length protein together with an additional ring of four separately expressed RCK domains was observed in the crystal structure (19). In order to characterize the contribution of the putative transport systems Kup and CglK in potassium uptake of C. glutamicum, we constructed the Δkup, ΔcglK, and ΔcglK Δkup mutants (Table (Table11).

Loss of the cglK gene causes increased pH sensitivity.

The growth of the Δkup and ΔcglK mutants was studied on agar plates at different pH values. For cells lacking the kup gene, no difference from the growth of WT cells was found (data not shown). Growth of the ΔcglK mutant was also comparable to that of WT cells at pH 8.5 and high potassium concentration; however, at pH 7, cells of the ΔcglK mutant required the presence of 50 mM potassium for growth. At 1 mM and without the addition of potassium, growth was hardly detectable. At pH 6, cells were not able to grow even in the presence of 50 mM potassium (Fig. (Fig.1).1). These results indicate that CglK but not Kup is of major importance for potassium-dependent growth of C. glutamicum at low pH values. The mutant lacking both the cglK and the kup gene was identical to cells of the ΔcglK mutant with respect to its potassium-dependent pH sensitivity, indicating that the two transport systems cannot replace each other and that the two mutations were not additive in terms of potassium-dependent pH sensitivity (data not shown). These growth experiments were repeated in liquid medium, and comparable results were obtained (data not shown). At a pH of 5.5, the growth rate of the ΔcglK strain was reduced by 50% in comparison to that of the WT and the Δkup mutant, respectively, even in the presence of 50 mM potassium.

The potential potassium channel CglK is the major uptake system for potassium in C. glutamicum.

In order to study the function of the Kup and CglK proteins as potassium uptake systems under physiological conditions, we set up a potassium transport assay in C. glutamicum cultures. After depletion of potassium at pH 8.5, cells were transferred into medium containing 1 mM potassium; both internal and external potassium concentrations were determined by flame photometry, and growth was also monitored. As expected, at alkaline pH, no differences between the WT and mutants were observed (Fig. (Fig.2).2). During growth (8 h), cells containing a functional cglK gene (WT and Δkup mutant) exhausted nearly all potassium from the medium, whereas the potassium concentration was found to stay unchanged in cultures of mutants lacking cglKcglK and Δkup ΔcglK mutants) (Fig. (Fig.2).2). This observation proves the function of CglK in potassium transport. The ΔcglK and Δkup ΔcglK cells contained lower initial amounts of potassium (200 mM) than WT and Δkup cells (270 mM). In all strains, the internal potassium concentration was found to decrease in the course of the experiment, indicating that potassium uptake was not sufficient to maintain the initial concentration and that the remaining potassium content was lowered due to cell proliferation. The decrease of the internal potassium concentration was faster in ΔcglK mutants, resulting in a potassium concentration of about 20 mM at the end of the growth phase (8 h). The internal potassium content of WT and Δkup cells, harboring the functional potential potassium channel CglK, decreased more slowly, and consequently, the final potassium concentration was about 100 mM (Fig. (Fig.2).2). These results again show that CglK is required for potassium uptake in C. glutamicum. To confirm that inactivation of the cglK gene is responsible for the observed phenotype, we constructed a plasmid for the constitutive expression of the cglK gene in C. glutamicum. After transformation of the Δkup ΔcglK mutant with the plasmid pEKEXcglK, these cells (Δkup ΔcglK/pEKEXcglK strain) were able to accumulate potassium like WT cells and consequently maintained a higher internal potassium concentration than the parental Δkup ΔcglK strain (Fig. (Fig.22).

FIG. 2.
Growth and potassium uptake by C. glutamicum WT (○), Δkup ([down-pointing small open triangle]), ΔcglK (□), Δkup ΔcglK ([open diamond]), and Δkup ΔcglK/pEKEXcglK cells ([open triangle]). Cells were washed and suspended in MMI ...

In order to prove the direct participation of CglK in potassium transport, we followed the uptake of 86Rb, which can substitute for potassium in transport assays. We determined Rb+ uptake for cells grown at 50 mM potassium or depleted for potassium, respectively. For potassium-depleted cells, the uptake rate and accumulation of imported Rb+ were found to be higher than in cells precultivated at a high potassium concentration (Fig. (Fig.3).3). The double mutant lacking cglK and kup was not able to take up Rb+ after depletion of potassium during precultivation (Fig. (Fig.3).3). Deletion of cglK alone had the same effect on Rb+ uptake activity after depletion of potassium, again indicating that the Kup-type transport system does not contribute to potassium uptake under these conditions (Fig. (Fig.33).

FIG. 3.
Initial rates of potassium transport in C. glutamicum DHPF (circles), DHPFΔcglK (squares), and DHPFΔkupΔcglK (triangles). Cells were depleted of potassium during precultivation (open symbols) or grown at standard potassium concentration ...

Potassium uptake by CglK is important for energy homeostasis and cell physiology in C. glutamicum.

We were interested in the specific requirement of C. glutamicum for potassium at acidic pH. A decrease of the external pH affects the electrochemical proton potential at the cytoplasmic membrane, the proton motive force (PMF). The PMF consists of the chemical gradient for protons, ΔpH, and the electric potential, ΔΨ (inside negative), according to the equation PMF = ΔΨ −Z ΔpH (Z = 2.3 RT/F). A decrease of the external pH causes an increase of the pH gradient, and proton influx is accelerated and could impact the PMF. In order to maintain constant PMF values, either the membrane potential or the internal pH value must be lowered. However, since the cell aims to maintain a neutral cytoplasmic pH (pH homeostasis), ΔΨ must be lowered, and this may be achieved by potassium influx via CglK in C. glutamicum.

In order to further elucidate the impact of potassium on membrane potential and internal pH regulation, we performed pH shifts from pH 7 to pH 6 either with (50 mM) or without the addition of potassium and determined the components of the PMF in C. glutamicum cells. In WT cells at pH 7, we measured under our experimental conditions an internal pH of 7.4 ± 0.02 (mean ± standard deviation) and a membrane potential of 170 mV, resulting in a PMF of about 200 mV. Upon a shift to pH 6 in the absence of added K+, the membrane potential was found to increase to values of 180 to 190 mV at a significantly decreased internal pH of 6.1 (Fig. (Fig.4A).4A). The same was found for mutants lacking the kup and/or cglK gene (data not shown). In the presence of 50 mM potassium, the membrane potential was significantly lowered, to 130 mV, and the internal pH was maintained at pH 7.0 ± 0.08 in the WT. In all mutant cell lines harboring a cglK gene (Δkup and ΔcglK/pEKEXcglK mutants) the same adjustment of ΔΨ and ΔpH was found. However, in cells lacking cglKcglK and Δkup ΔcglK mutants), a lower internal pH of 6.53 ± 0.03 and an almost unchanged membrane potential of 168 to 173 mV were determined in the presence of 50 mM potassium (Fig. (Fig.4A).4A). Interestingly, the resulting PMF values were maintained at approximately 200 ± 8 mV in all cases, irrespective of the cell type or the experimental condition. These results indicate that, upon a decreasing external pH, the adjustment toward a decreasing membrane potential, as well as the maintenance of a neutral internal pH, depends on the availability of potassium and on a functional CglK protein in C. glutamicum.

FIG. 4.
Impact of potassium on energetic parameters in C. glutamicum. (A) Steady-state values of the PMF (black bars) were determined after a shift from pH 7 to pH 6 by measuring membrane potential ΔΨ (light-gray bars) and ΔpH (dark-gray ...

For validation and a direct (real-time) quantification of the impact of potassium on internal pH regulation, we followed the change of the internal pH by using the fluorescent pH indicator BCECF after a pH shift from 7 to 6 in the absence of potassium and after subsequent addition of potassium, respectively (Fig. (Fig.4B).4B). For WT cells, we observed an immediate increase of the internal pH upon the addition of potassium, whereas for ΔcglK mutant cells, no change in pH was observed within 5 min. This indicates that ΔcglK cells were not able to regulate the internal pH in a potassium-dependent manner. The experiments were also performed with the Δkup and ΔcglK/pEKEXcglK mutants, resulting in the same increase of the internal pH as observed for WT cells (data not shown). In order to prove that adjustment of the internal pH is in fact closely related to potassium uptake by CglK, we performed the same experiment as that whose results are shown in Fig. Fig.4A4A and simultaneously followed potassium uptake. Cells were subjected to pH 6 in the presence of 1 mM potassium, the cytoplasmic pH was measured by radioactive probes, and potassium uptake was quantified. Within 5 min, an increase in the internal pH by 0.22 ± 0.02 units was observed in WT cells, whereas in ΔcglK cells, no increase in the internal pH was found, as was shown before using the BCECF probe (Fig. (Fig.4B).4B). The concomitant potassium accumulation in WT cells was 225 ± 21 nmol K+ mg−1 CDM, whereas ΔcglK cells accumulated negligible amounts of potassium (0.05 ± 0.02 nmol K+ mg−1 CDM) within 5 min (Fig. (Fig.4C).4C). These results again indicate that potassium uptake by CglK is crucial for maintenance of the internal pH in C. glutamicum.

Furthermore, we analyzed the impact of potassium uptake at low pH values on other processes. The respiratory chain is the major energy conversion process in bacterial membranes and was supposed to be sensitive to changes of membrane potential and pH (23). We measured the oxygen consumption in WT and ΔcglK cells upon a sudden shift of the external pH. In the absence of potassium, the activity of the respiratory chain was strongly reduced in both types of cells (Fig. (Fig.5).5). After the addition of 50 mM potassium, the oxygen consumption of WT cells increased and reached a level of 94 nmol O2 min−1 ml−1 OD600 unit−1, whereas for ΔcglK cells, the value did not change and 31 nmol O2 min−1 ml−1 OD600 unit−1 was measured. When the membrane potential was decreased by the potassium ionophore valinomycin in the presence of 50 mM potassium, the oxygen consumption rates of the WT and the ΔcglK mutant were found to be 83 and 76 nmol O2 min−1 ml−1 OD600 unit−1, respectively. These results indicate that the activity of the respiratory chain during response of C. glutamicum cells to acidic stress conditions depends on potassium and a functional CglK protein.

FIG. 5.
Dependence of respiration in C. glutamicum on potassium. Potassium-depleted cells of the WT (white bars) or the ΔcglK mutant (gray bars) were subjected to a shift of the external pH from 7 to 6 in the absence (−) or presence (+) ...

Lack of the separate RCK protein causes growth deficiency at low pH.

The protein sequence and domain structure of C. glutamicum CglK were found to be similar to those of the potassium channel MthK from Methanobacterium thermoautotrophicum (24% identity, 41% similarity, and E value of 7e−12). The mthK mRNA is translated from an internal start codon (amino acid position 107) into a full-length protein and a separate soluble RCK protein, resulting in the formation of an octameric ring structure at the cytoplasmic side of the membrane (19). In the C. glutamicum CglK protein, a methionine which could represent an alternative start position is present at position 137. For a protein consisting of amino acids 137 to 353, we calculated a theoretical molecular mass of 23 kDa, in comparison to 38 kDa for the full-length CglK protein. After cloning of the full-length cglK gene into the pET52b vector and its expression in E. coli, protein bands of 40 kDa and 25 kDa were observed in SDS-PAGE and Western blot analyses after purification (Fig. (Fig.6).6). Identification of proteins by peptide mass fingerprinting revealed a peptide pattern covering solely the RCK domain of CglK for the 25-kDa protein, whereas for the 40-kDa protein, peptides covering the complete CglK protein, including the membrane part, were found (data not shown). This indicates that the cglK gene of C. glutamicum can be translated as a full-length protein and a cytoplasmic protein harboring the RCK domain only.

FIG. 6.
Copurification of the full-length CglK protein and the RCK subunit by Ni-NTA affinity chromatography. The cglK gene was cloned into the vector pET52b mediating the expression of C-terminal penta-His-tagged proteins in E. coli BL21 cells. During batch ...

In order to further elucidate the function of the RCK domain in C. glutamicum, we constructed different variants of CglK for expression in the Δkup ΔcglK background. Besides the full-length cglK, a gene was cloned that encoded the membrane part of the channel only by changing Met137 into a stop codon. Furthermore, Met137 was changed into Ile in order to prevent the formation of the additional cytoplasmic RCK protein (Table (Table1).1). The resulting C. glutamicum mutants were grown in liquid medium at pH 6.5, i.e., under conditions where potassium uptake was found to be essential. At potassium concentrations of 10 and 5 mM, the WT and the mutant expressing the full-length cglK gene were able to grow (Fig. (Fig.7).7). Mutants lacking CglK or harboring the plasmid encoding the membrane part only were either hardly able to grow (10 mM potassium) or could not grow at all (5 mM potassium). For the mutant expressing the cglK(M137I) variant, we found an intermediate phenotype. The growth rates were lower than for WT cells but significantly higher than for cells lacking cglK. The difference was more evident at 5 mM potassium because mutants lacking cglK were not able to grow at this potassium concentration, whereas mutants harboring CglK_M137I did so. At very low potassium concentrations, all strains were impaired in growth and only for WT cells was significant growth observed. These results indicate that the loss of cglK can be complemented fully by the expression of plasmid-encoded full-length cglK and partially by cglK_M137I, demonstrating the requirement of the cytoplasmic RCK protein for functionality of the potential channel CglK in C. glutamicum.

FIG. 7.
Dependence of CglK function on RCK. (A) Schematic representation of CglK protein variants by two subunits of the full-length protein harboring the membrane-anchored pore region (black stick), the RCK domain (gray ball), and the separate RCK domain (light-gray ...


In contrast to many other bacteria, C. glutamicum tolerates potassium deficiency and low internal potassium concentrations.

Surprisingly, we observed growth of C. glutamicum at neutral or alkaline pH values in the absence of added potassium, even after several rounds of inoculation in nearly potassium-free medium. In WT cells grown with potassium supplementation, cytoplasmic potassium decreased from 300 to 100 mM upon growth in the presence of 1 mM external potassium, which was almost completely exhausted by the cells. Upon growth in the presence of 10 μM potassium, the internal potassium concentration dropped to about 20 mM. Consequently, internal potassium can vary in C. glutamicum from 800 mM (24) down to 20 mM or even lower values during continuing growth under potassium limitation. This represents a reduction to 2.5% or even smaller amounts in comparison to the usual potassium content in C. glutamicum. E. coli cells contain up to 450 mM potassium but are not able to grow in the presence of internal potassium concentrations lower than 150 mM (9). In B. subtilis, 300 mM potassium was found to be the cytoplasmic concentration (38). However, in spite of the presence of highly effective Ktr-type potassium uptake systems, at external potassium concentrations below 0.4 mM, no growth was observed (18). We conclude that, in contrast to E. coli and B. subtilis, C. glutamicum tolerates potassium limitation because it can cope with very low internal potassium content at neutral or alkaline pH. In Klebsiella pneumoniae and Bacillus stearothermophilus, a comparable phenotype was observed (7). In both strains, the potassium content decreased drastically at higher pH values, and B. stearothermophilus was able to grow in the absence of added potassium (10 μM) at a pH of 8.5. Under these conditions, the internal potassium content was decreased to 1.4% of that observed in cells growing at 0.5 mM external potassium. However, the presence of ammonium was required, indicating that potassium can be replaced by ammonium in a pH-dependent manner (7). This could be the case for C. glutamicum as well. As a matter of fact, we found that growth of the ΔcglK mutant on agar plates or in liquid medium in the absence of ammonium chloride was much more potassium dependent (data not shown).

Regulation of CglK depends on the RCK domain.

Based on the high sequence similarity of the CglK and the MthK proteins and on the fact that we detected two different proteins after expression of the cglK gene in E. coli, we assume that the functional potential potassium channel CglK consists of four full-length proteins and four additional soluble proteins harboring the RCK domain. The selectivity filter sequence of CglK resembles that of MthK (TVGYGD), indicating its function as a specific potassium pore (19). For regulation of MthK activity, binding of Ca2+ ions at Glu210, Glu212, and Asp184 of the RCK domain was described and the dynamic oligomerization of the soluble RCK proteins in dependence on Ca2+ binding was proposed (19, 26). In E. coli cells, however, MthK function was shown to be Ca2+ independent, indicating that other parameters may be important under in vivo conditions (28). The eukaryotic SloI potassium channel contains an RCK domain as well and is activated by low internal pH values in the virtual absence of Ca2+, and histidine residues were proposed to participate in this process (3). In spite of the high sequence similarity of MthK and CglK, the residues responsible for Ca2+ binding in MthK are not conserved in the CglK sequence. However, three histidine residues are present in the RCK domain of CglK, at positions 140, 246, and 308, and might be involved in internal pH sensing.

Upon a decrease of the external pH, protons enter the cell along their electrochemical gradient, and this is likely to activate CglK via the RCK domain. The impact of the RCK domain for CglK function at low pH in C. glutamicum was proven (Fig. (Fig.7)7) and resembles observations made for the Kch potassium channel of H. pylori (HpKch). An HpKch variant lacking the separate RCK protein was found to be, on the one hand, more sensitive toward potassium limitation than the WT but, on the other, less sensitive than a mutant lacking the complete HpKch channel (34). Further studies are required in order to unravel the mechanism of signal perception by the C. glutamicum RCK domain and the signal transduction to the membrane-anchored pore.

Potassium transport via CglK is essential for C. glutamicum acid stress response.

It has been shown that at acidic pH values, the presence of K+ is required for maintenance of a neutral internal pH for a number of bacteria from different habitats, among others, Lactococcus lactis (22), E. coli (25), Enterococcus faecalis (4), Bradyrhizobium sp. strain 32H1 (14), Rhodobacter sphaeroides (2), and Streptococcus mutans (8). This also holds true for C. glutamicum. At pH 6 in the medium, the internal pH of WT cells was found to be pH 7 in the presence of K+ and was strongly decreased to pH 6.2 in the absence of K+. Growth experiments at different pH values in the absence of potassium confirmed its importance under acidic conditions. At pH 6, the addition of at least 1 mM K+ was necessary to support growth. In the presence of sufficient K+ (50 mM), the ΔcglK mutant grows similarly to the WT at alkaline and neutral pH, but growth was significantly impaired at acidic pH. We demonstrated that the loss of K+ transport impaired the capacity for pH homeostasis. This proves that the K+-dependent pH homeostasis in C. glutamicum is mediated by CglK, which is therefore essential for growth at acidic conditions.

Measurement of the components of the PMF revealed a direct correlation between the adjustment of the internal pH value and the membrane potential in C. glutamicum. A decreased pH gradient at pH 6 during potassium limitation or in mutants lacking the cglK gene was correlated with an increased membrane potential. After the addition of potassium, a significant pH gradient could be established by cells harboring the cglK gene and the membrane potential was found to be decreased concomitantly. The extent of ΔpH and ΔΨ adaptation was complementary, resulting in constant PMF values of approximately 200 mV. Due to this correlation between adjustment of ΔΨ and increase of cytoplasmic pH, it is likely that lowering of the membrane potential may be required for effective pH homeostasis, as suggested previously for E. coli (25). A sudden decrease of the external pH causes an increase of ΔpH and, thereby, an increase of the electrochemical proton potential. Consequently, the driving force for the influx of protons is increased, which results in a decreased internal pH. WT cells are able to respond by decreasing the membrane potential by the influx of positively charged potassium ions via the potential channel CglK, thus keeping the electrochemical proton potential constant. Regulation of the ΔΨ and maintenance of an internal pH close to neutral are mandatory for many physiological processes, like the activity of the respiratory chain. On the other hand, in the absence of potassium or in mutants lacking CglK, ΔΨ cannot be decreased, the inwardly directed electrochemical proton potential is increased, and as a consequence, the internal pH decreases and causes limitation of growth or cell death.

Potassium transporters differ in apathogenic and pathogenic representatives of the Actinomyces.

C. glutamicum possesses the potential MthK-type channel CglK as the sole functional potassium uptake system. At least under the conditions tested, the Kup-type transporter is not functional. In other Actinomyces, e.g., Corynebacterium jeikeium, Corynebacterium diphtheriae, Corynebacterium efficiens, and Mycobacterium tuberculosis, homologs of CglK are present. Besides a potential channel, C. efficiens possesses a Kup-type protein like that of C. glutamicum. In the genome sequences of C. jeikeium and C. diphtheriae, additional genes encoding Ktr-type transporters are present, and in the genomes of C. jeikeium and M. tuberculosis, Kdp-type transporter subunits are present as well. In conclusion, the apathogenic strains C. efficiens and C. glutamicum harbor a potential potassium channel, as well as a gene encoding a Kup-type transporter, whereas pathogenic strains possess various active carriers in addition. A correlation of the bacterial life style and the occurrence of different potassium transport systems was proposed previously (34). The multiplicity of active potassium transport systems could represent a prerequisite for pathogenic Actinomyces to survive under potassium-limiting and/or acidic conditions during host infection, an ability not necessary for the soil bacterium C. glutamicum.


We thank Anja Wittmann for excellent technical assistance and Henrik Strahl and Evert Bakker for helpful discussions.

We thank the Bundesministerium für Bildung und Forschung (BMBF) and Evonik-Degussa (SysMap program) for financial support.


[down-pointing small open triangle]Published ahead of print on 6 March 2009.


1. Abe, S., K. Takayama, and S. Kinoshita. 1967. Taxonomical studies on glutamic acid-producing bacteria. J. Gen. Microbiol. 13279-301.
2. Abee, T., K. J. Hellingwerf, and W. N. Konings. 1988. Effects of potassium ions on proton motive force in Rhodobacter sphaeroides. J. Bacteriol. 1705647-5653. [PMC free article] [PubMed]
3. Avdonin, V., X. D. Tang, and T. Hoshi. 2003. Stimulatory action of internal protons on Slo1 BK channels. Biophys. J. 842969-2980. [PMC free article] [PubMed]
4. Bakker, E. P., and W. E. Mangerich. 1981. Interconversion of components of the bacterial proton motive force by electrogenic potassium transport. J. Bacteriol. 147820-826. [PMC free article] [PubMed]
5. Booth, I. R. 1985. Regulation of cytoplasmic pH in bacteria. Microbiol. Rev. 49359-378. [PMC free article] [PubMed]
6. Bremer, E., and R. Krämer. 2000. Coping with osmotic challenges, p. 79-97. In G. Storz and R. Hengge-Aronis (ed.), Bacterial stress responses. ASM Press, Washington, DC.
7. Buurman, E. T., J. Pennock, D. W. Tempest, M. J. Teixeira de Mattos, and O. M. Neijssel. 1989. Replacement of potassium ions by ammonium ions in different micro-organisms grown in potassium-limited chemostat culture. Arch. Microbiol. 15258-63. [PubMed]
8. Dashper, S. G., and E. C. Reynolds. 1992. pH regulation by Streptococcus mutans. J. Dent. Res. 711159-1165. [PubMed]
9. Dinnbier, U., E. Limpinsel, R. Schmid, and E. P. Bakker. 1988. Transient accumulation of potassium glutamate and its replacement by trehalose during adaptation of growing cells of Escherichia coli K-12 to elevated sodium chloride concentrations. Arch. Microbiol. 150348-357. [PubMed]
10. Doyle, D. A., J. Morais Cabral, R. A. Pfuetzner, A. Kuo, J. M. Gulbis, S. L. Cohen, B. T. Chait, and R. MacKinnon. 1998. The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 28069-77. [PubMed]
11. Eikmanns, B. J., E. Kleinertz, W. Liebl, and H. Sahm. 1991. A family of Corynebacterium glutamicum/Escherichia coli shuttle vectors for cloning, controlled gene expression, and promoter probing. Gene 10293-98. [PubMed]
12. Ennis, H. L., and M. Lubin. 1961. Dissociation of ribonucleic acid and protein synthesis in bacteria deprived of potassium. Biochim. Biophys. Acta 50399-402. [PubMed]
13. Epstein, W. 2003. The roles and regulation of potassium in bacteria. Prog. Nucleic Acid Res. Mol. Biol. 75293-320. [PubMed]
14. Gober, J. W., and E. R. Kashket. 1986. Effects of K+ on the proton motive force of Bradyrhizobium sp. strain 32H1. J. Bacteriol. 166618-622. [PMC free article] [PubMed]
15. Gralla, J. D., and D. R. Vargas. 2006. Potassium glutamate as a transcriptional inhibitor during bacterial osmoregulation. EMBO J. 251515-1521. [PMC free article] [PubMed]
16. Grant, S. G., J. Jessee, F. R. Bloom, and D. Hanahan. 1990. Differential plasmid rescue from transgenic mouse DNAs into Escherichia coli methylation-restriction mutants. Proc. Natl. Acad. Sci. USA 874645-4649. [PMC free article] [PubMed]
17. Harold, F. M., and J. R. Baarda. 1968. Effects of nigericin and monactin on cation permeability of Streptococcus faecalis and metabolic capacities of potassium-depleted cells. J. Bacteriol. 95816-823. [PMC free article] [PubMed]
18. Holtmann, G., E. P. Bakker, N. Uozumi, and E. Bremer. 2003. KtrAB and KtrCD: two K+ uptake systems in Bacillus subtilis and their role in adaptation to hypertonicity. J. Bacteriol. 1851289-1298. [PMC free article] [PubMed]
19. Jiang, Y., A. Lee, J. Chen, M. Cadene, B. T. Chait, and R. MacKinnon. 2002. The open pore conformation of potassium channels. Nature 417523-526. [PubMed]
20. Kalinowski, J., B. Bathe, D. Bartels, N. Bischoff, M. Bott, A. Burkovski, N. Dusch, L. Eggeling, B. J. Eikmanns, L. Gaigalat, A. Goesmann, M. Hartmann, K. Huthmacher, R. Kramer, B. Linke, A. C. McHardy, F. Meyer, B. Mockel, W. Pfefferle, A. Puhler, D. A. Rey, C. Ruckert, O. Rupp, H. Sahm, V. F. Wendisch, I. Wiegrabe, and A. Tauch. 2003. The complete Corynebacterium glutamicum ATCC 13032 genome sequence and its impact on the production of L-aspartate-derived amino acids and vitamins. J. Biotechnol. 1045-25. [PubMed]
21. Kase, H., and K. Nakayama. 1972. Production of L-threonin by analog-resistant mutants. Agric. Biol. Chem. 361611-1621.
22. Kashket, E. R., and S. L. Barker. 1977. Effects of potassium ions on the electrical and pH gradients across the membrane of Streptococcus lactis cells. J. Bacteriol. 1301017-1023. [PMC free article] [PubMed]
23. Kim, Y. J. 2002. Respiratory chain-linked components of the marine bacterium Vibrio alginolyticus affect each other. J. Microbiol. 40125-128.
24. Krämer, R., C. Lambert, C. Hoischen, and H. Ebbighausen. 1990. Uptake of glutamate in Corynebacterium glutamicum. 1. Kinetic properties and regulation by internal pH and potassium. Eur. J. Biochem. 194929-935. [PubMed]
25. Kroll, R. G., and I. R. Booth. 1981. The role of potassium transport in the generation of a pH gradient in Escherichia coli. Biochem. J. 198691-698. [PMC free article] [PubMed]
26. Kuo, M. M., K. A. Baker, L. Wong, and S. Choe. 2007. Dynamic oligomeric conversions of the cytoplasmic RCK domains mediate MthK potassium channel activity. Proc. Natl. Acad. Sci. USA 1042151-2156. [PMC free article] [PubMed]
27. Negulescu, P. A., and T. E. Machen. 1990. Intracellular ion activities and membrane transport in parietal cells measured with fluorescent dyes. Methods Enzymol. 19238-81. [PubMed]
28. Parfenova, L. V., B. M. Crane, and B. S. Rothberg. 2006. Modulation of MthK potassium channel activity at the intracellular entrance to the pore. J. Biol. Chem. 28121131-21138. [PubMed]
29. Reed, R. H., S. R. C. Warr, D. L. Richardson, and W. D. P. Stewart. 1985. Multiphasic osmotic adjustment in a euryhaline cyanobacterium. FEMS Microbiol. Let. 28225-229.
30. Rottenberg, H. 1979. The measurement of membrane potential and deltapH in cells, organelles, and vesicles. Methods Enzymol. 55547-569. [PubMed]
31. Rübenhagen, R., S. Morbach, and R. Krämer. 2001. The osmoreactive betaine carrier BetP from Corynebacterium glutamicum is a sensor for cytoplasmic K+. EMBO J. 205412-5420. [PMC free article] [PubMed]
32. Schäfer, A., A. Tauch, W. Jäger, J. Kalinowski, G. Thierbach, and A. Pühler. 1994. Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: selection of defined deletions in the chromosome of Corynebacterium glutamicum. Gene 14569-73. [PubMed]
33. Steger, R., M. Weinand, R. Krämer, and S. Morbach. 2004. LcoP, an osmoregulated betaine/ectoine uptake system from Corynebacterium glutamicum. FEBS Lett. 573155-160. [PubMed]
34. Stingl, K., S. Brandt, E. M. Uhlemann, R. Schmid, K. Altendorf, C. Zeilinger, C. Ecobichon, A. Labigne, E. P. Bakker, and H. de Reuse. 2007. Channel-mediated potassium uptake in Helicobacter pylori is essential for gastric colonization. EMBO J. 26232-241. [PMC free article] [PubMed]
35. Studier, F. W., and B. A. Moffatt. 1986. Use of bacteriophage T7 RNA polymerase to direct selective high-level expression of cloned genes. J. Mol. Biol. 189113-130. [PubMed]
36. Trchounian, A., and H. Kobayashi. 1999. Kup is the major K+ uptake system in Escherichia coli upon hyper-osmotic stress at a low pH. FEBS Lett. 447144-148. [PubMed]
37. Whatmore, A. M., J. A. Chudek, and R. H. Reed. 1990. The effects of osmotic upshock on the intracellular solute pools of Bacillus subtilis. J. Gen. Microbiol. 1362527-2535. [PubMed]
38. Whatmore, A. M., and R. H. Reed. 1990. Determination of turgor pressure in Bacillus subtilis: a possible role for K+ in turgor regulation. J. Gen. Microbiol. 1362521-2526. [PubMed]

Articles from Journal of Bacteriology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...