The N-terminal input domain of the sensor kinase KdpD of Escherichia coli stabilizes the interaction between the cognate response regulator KdpE and the corresponding DNA-binding site

The sensor kinase/response regulator system KdpD/KdpE of Escherichia coli regulates expression of the kdpFABC operon, which encodes the high affinity K+ transport system KdpFABC. The membrane-bound sensor kinase KdpD consists of an N-terminal input domain (comprising a large cytoplasmic domain and four transmembrane domains) and a cytoplasmic C-terminal transmitter domain. Here we show that the cytoplasmic N-terminal domain of KdpD (KdpD/1-395) alone supports semi-constitutive kdpFABC expression, which becomes dependent on the extracellular K+ concentration under K+-limiting growth conditions. However, it should be noted that the non-phosphorylatable derivative KdpD/H673Q or the absence of KdpD abolishes kdpFABC expression completely. KdpD/1-395 mediated kdpFABC expression requires the corresponding response regulator KdpE with an intact phosphorylation site. Experiments with an Escherichia coli mutant unable to synthesize acetyl phosphate as well as transposon mutagenesis suggest that KdpE is phosphorylated in vivo by low molecular weight phosphodonors in the absence of the full-length sensor kinase. Various biochemical approaches provide first evidence that kdpFABC expression mediated by KdpD/1-395 is due to a stabilizing effect of this domain on the binding of KdpE approximately P to its corresponding DNA-binding site. Such a stabilizing effect of a sensor kinase domain on the DNA-protein interaction of the cognate response regulator has never been observed before for any other sensor kinase. It describes a new mechanism in bacterial two-component signal transduction.     

The transmembrane domains of the sensor kinase KdpD of Escherichia coli are not essential for sensing K+ limitation

The sensor kinase/response regulator system KdpD/KdpE of Escherichia coli regulates the expression of the kdpFABC operon, which encodes the high affinity K+ transport system KdpFABC. The membrane-bound sensor kinase KdpD consists of four transmembrane domains, a large cytoplasmic N-terminal domain and a cytoplasmic C-terminal transmitter domain. To elucidate the role of the four transmembrane domains, various deletions were introduced in kdpD and the activities of the resulting truncated derivatives of KdpD were determined. A KdpD protein lacking all four transmembrane domains was able to sense low K+ concentrations, whereas at higher K+ concentrations kdpFABC expression was constitutive. These and further results with various truncated KdpD proteins lacking distinct parts of the transmembrane domains or derivatives in which a linker peptide or two transmembrane domains of PutP, the Na+/proline transporter of Escherichia coli, replaced the missing part indicated that the transmembrane domains are not essential for sensing of K+ limitation, but may be important for the correct positioning of the large N- and C-terminal cytoplasmic domains to each other.     

Analysis of the properties of the N-terminal nucleotide-binding domain of human P-glycoprotein

Human P-glycoprotein, the MDR1 gene product, requires both Mg(2+)-ATP binding and hydrolysis to function as a drug transporter; however, the mechanism(s) defining these events is not understood. In the present study, we explored the nature of Mg(2+)-ATP binding in the N-terminal nucleotide-binding domain of human P-glycoprotein and identified the minimal functional unit required for specific ATP binding. Recombinant proteins encompassing amino acids within the region beginning at 348 and ending at 707 were expressed in Escherichia coli, purified from inclusion bodies under denaturing conditions, and renatured by rapid dilution. The ability of ATP to interact with these proteins was examined by use of the photoactive ATP analogue [alpha-(32)P]-8-azido-ATP. Photoaffinity labeling of recombinant proteins identified the region between amino acids 375 and 635 as the region necessary to obtain specific ATP-binding properties. Specific protein labeling was saturable, enhanced by Mg(2+), and inhibited by ATP. Recombinant proteins confined within the region beginning at amino acid 392 and ending at amino acid 590 demonstrated nonspecific [alpha-(32)P]-8-azido-ATP labeling. Nonspecific labeling was not enhanced by Mg(2+) and was inhibited only by high concentrations of ATP. Using a D555N mutated protein, we found that the conserved aspartate residue in the Walker B motif plays a role in magnesium-enhanced ATP-binding. Taken together, these data define the region of the N-terminal nucleotide-binding domain of P-glycoprotein that is required for specific ATP binding and suggest that magnesium may play a role in stabilizing the ATP-binding site.     

The extension of the fourth transmembrane helix of the sensor kinase KdpD of Escherichia coli is involved in sensing

The KdpD sensor kinase and the KdpE response regulator control expression of the kdpFABC operon coding for the KdpFABC high-affinity K+ transport system of Escherichia coli. In search of a distinct part of the input domain of KdpD which is solely responsible for K+ sensing, sequences of kdpD encoding the transmembrane region and adjacent N-terminal and C-terminal extensions were subjected to random mutagenesis. Nine KdpD derivatives were identified that had lost tight regulation of kdpFABC expression. They all carried single amino acid replacements located in a region encompassing the fourth transmembrane helix and the adjacent arginine cluster of KdpD. All mutants exhibited high levels of kdpFABC expression regardless of the external K+ concentration. However, 3- to 14-fold induction was observed under extreme K+-limiting conditions and in response to an osmotic upshift when sucrose was used as an osmolyte. These KdpD derivatives were characterized by a reduced phosphatase activity in comparison to the autokinase activity in vitro, which explains constitutive expression. Whereas for wild-type KdpD the autokinase activity and also, in turn, the phosphotransfer activity to KdpE were inhibited by increasing concentrations of K+, both activities were unaffected in the KdpD derivatives. These data clearly show that the extension of the fourth transmembrane helix encompassing the arginine cluster is mainly involved in sensing both K+ limitation and osmotic upshift, which may not be separated mechanistically.     

Modulation of KdpD phosphatase implicated in the physiological expression of the kdp ATPase of Escherichia coli

The KdpD sensor kinase and the KdpE response regulator control the expression of the kdpFABC operon, encoding the KdpFABC high-affinity K+ transport system of Escherichia coli. Low turgor pressure has been postulated to be the environmental stimulus to express KdpFABC. KdpD has autokinase, phosphotransferase and, like many sensor kinases, response regulator (phospho-KdpE) specific phosphatase activity. To determine which of these activities are altered in response to the environmental stimulus, we isolated and analysed six kdpD mutants that cause constitutive expression of KdpFABC. In three of the mutants, phosphatase activity was undetectable and, in two, phosphatase was reduced. Kinase activity was unaffected in four of the mutants, but elevated in one. In one mutant, a pseudorevertant of a kdpD null mutation, kinase and phosphatase were both reduced to 20% of the wild-type level. These findings suggest that initiation of signal transduction by KdpD is mediated by the inhibition of the phospho-KdpE-specific phosphatase activity of KdpD, leading to an accumulation of phospho-KdpE, which in turn activates the expression of the KdpFABC system. The data also suggest that levels of activity in vitro may differ from what occurs in vivo, because in vitro conditions cannot replicate those in vivo.     

An amphiphilic region in the cytoplasmic domain of KdpD is recognized by the signal recognition particle and targeted to the Escherichia coli membrane

The sensor protein KdpD of Escherichia coli is composed of a large N-terminal hydrophilic region (aa 1–400), four transmembrane regions (aa 401–498) and a large hydrophilic region (aa 499–894) at the C-terminus. KdpD requires the signal recognition particle (SRP) for its targeting to the membrane. Deletions within KdpD show that the first 50 residues are required for SRP-driven membrane insertion. A fusion protein of the green fluorescent protein (GFP) with KdpD is found localized at the membrane only when SRP is present. The membrane targeting of GFP was not observed when the first 50 KdpD residues were deleted. A truncated mutant of KdpD containing only the first 25 amino acids fused to GFP lost its ability to specifically interact with SRP, whereas a specific interaction between SRP and the first 48 amino acids of KdpD fused to GFP was confirmed by pull-down experiments. Conclusively, a small amphiphilic region of 27 residues within the amino-terminal domain of KdpD (aa 22–48) is recognized by SRP and targets the protein to the membrane. This shows that membrane proteins with a large N-terminal region in the cytoplasm can be membrane-targeted early on to allow co-translational membrane insertion of their distant transmembrane regions.     

A chimeric Anabaena/ Escherichia coli KdpD protein (Anacoli KdpD) functionally interacts with E. coli KdpE and activates kdp expression in E. coli

The kdpFABC operon, coding for a high-affinity K(+)-translocating P-type ATPase, is expressed in Escherichia coli as a backup system during K(+) starvation or an increase in medium osmolality. Expression of the operon is regulated by the membrane-bound sensor kinase KdpD and the cytosolic response regulator KdpE. From a nitrogen-fixing cyanobacterium, Anabaena sp. strain L-31, a kdpDgene was cloned (GenBank accession no. AF213466) which codes for a KdpD protein (365 amino acids) that lacks both the transmembrane segments and C-terminal transmitter domain and thus is shorter than E. coli KdpD. A chimeric kdpD gene was constructed and expressed in E. coli coding for a protein (Anacoli KdpD), in which the first 365 amino acids of E. coli KdpD were replaced by those from Anabaena KdpD. In everted membrane vesicles, this chimeric Anacoli KdpD protein exhibited activities, such as autophosphorylation, transphosphorylation and ATP-dependent dephosphorylation of E. coli KdpE, which closely resemble those of the E. coli wild-type KdpD. Cells of E. coli synthesizing Anacoli KdpD expressed kdpFABC in response to K(+) limitation and osmotic upshock. The data demonstrate that Anabaena KdpD can interact with the E. coliKdpD C-terminal domain resulting in a protein that is functional in vitro as well as in vivo.     

Characterization of the N-terminal domain of NrtC, the ATP-binding subunit of ABC-type nitrate transporter of the cyanobacterium Phormidium laminosum

The N-terminal domain of NrtC, the ATP-binding subunit of nitrate/nitrite ABC-transporter in the cyanobacterium Phormidium laminosum, has been expressed in Escherichia coli as a histidine-tagged fusion protein (His(6)NrtC1). Binding of ATP to the pure His(6)NrtC1 was characterized using the nucleotide analogue TNP-ATP [2'(3')-O-(2,4,6-trinitrophenyl) adenosine 5'-triphosphate]. Fluorescence assays showed that His(6)NrtC1 specifically binds Mg(2+) TNP-ATP with high affinity, binding being dependent on protein concentration. The presence of ATP prevents the covalent modification of His(6)NrtC1 by fluorescein 5'-isothiocyanate (FITC), suggesting that this probe reacts at the nucleotide-binding site of NrtC. The active form of the truncated NrtC is a dimer that shows high affinity for TNP-ATP (K(d)=0.76+/-0.1 microM). Evidence for the presence of two nucleotide-binding sites per dimer protein is given. Our results indicate that nucleotide binding is strongly dependent on the dimerization of NrtC and that the N-terminal domain of the protein contains the binding site for ATP. No ATPase activity catalyzed in vitro by the truncated subunit was detected.     

P-glycoprotein. ATP hydrolysis by the N-terminal nucleotide-binding domain.

Two ATP-binding domains are found in members of the family of ATP-dependent transport proteins, which includes P-glycoprotein and cystic fibrosis transmembrane conductance regulator. To investigate the involvement of the two ATP-binding domains in the ATPase activity of P-glycoprotein, full-length and the 5'-half of human MDR1 cDNA, which encodes P-glycoprotein, were fused with the Escherichia coli lacZ gene and expressed in NIH3T3 cells. Immunoprecipitated full-length P-glycoprotein beta-galactosidase showed ATPase activity with apparent specific activity of 180 nmol/mg/min, a value higher than previously reported, in the presence of phospholipids, suggesting that stabilization of the transmembrane domains is necessary for ATP hydrolysis. N-terminal half P-glycoprotein-beta-galactosidase also showed ability to hydrolyze ATP but with slightly lower specific activity. Both ATPase activities showed similar characteristics when the effect of several inhibitors was analyzed, indicating that the N-terminal ATP-binding domain contains all residues necessary to hydrolyze ATP without interacting with the C-terminal ATP-binding domain.     

ATP binding/hydrolysis by and phosphorylation of peroxisomal ATP-binding cassette proteins PMP70 (ABCD3) and adrenoleukodystrophy protein (ABCD1)

The 70-kDa peroxisomal membrane protein (PMP70) and adrenoleukodystrophy protein (ALDP), half-size ATP-binding cassette transporters, are involved in metabolic transport of long and very long chain fatty acids into peroxisomes. We examined the interaction of peroxisomal ATP-binding cassette transporters with ATP using rat liver peroxisomes. PMP70 was photoaffinity-labeled at similar efficiencies with 8-azido-[alpha-32P]ATP and 8-azido-[gamma-32P]ATP when peroxisomes were incubated with these nucleotides at 37 degrees C in the absence Mg2+ and exposed to UV light without removing unbound nucleotides. The photoaffinity-labeled PMP70 and ALDP were co-immunoprecipitated together with other peroxisomal proteins, which also showed tight ATP binding properties. Addition of Mg2+ reduced the photoaffinity labeling of PMP70 with 8-azido-[gamma-32P]ATP by 70%, whereas it reduced photoaffinity labeling with 8-azido-[alpha-32P]ATP by only 20%. However, two-thirds of nucleotide (probably ADP) was dissociated during removal of unbound nucleotides. These results suggest that ATP binds to PMP70 tightly in the absence of Mg2+, the bound ATP is hydrolyzed to ADP in the presence of Mg2+, and the produced ADP is dissociated from PMP70, which allows ATP hydrolysis turnover. Properties of photoaffinity labeling of ALDP were essentially similar to those of PMP70. Vanadate-induced nucleotide trapping in PMP70 and ALDP was not observed. PMP70 and ALDP were also phosphorylated at a tyrosine residue(s). ATP binding/hydrolysis by and phosphorylation of PMP70 and ALDP are involved in the regulation of fatty acid transport into peroxisomes.     

Photoaffinity labelling of the ATP-binding site of the epidermal growth factor-dependent protein kinase.

Epidermal growth factor (EGF), after binding to its receptor, activates a tyrosine-specific protein kinase which phosphorylates several substrates, including the EGF receptor itself. The effects of a photoaffinity analogue of ATP, 3'-O-(3-[N-(4-azido-2-nitrophenyl)amino]propionyl)adenosine 5'-triphosphate (arylazido-beta-alanyl-ATP) on the EGF-dependent protein kinase in A431 human tumour cell plasma membrane vesicles was investigated. This analogue was capable of inactivating the EGF-receptor kinase in a photodependent manner. Partial inactivation occurred at an analogue concentration of 1 microM and complete inactivation occurred at 10 microM when a 2 min light exposure was used. Arylazido-beta-alanine at 100 microM and ATP at 100 microM were incapable of inactivating the enzyme with 2 min of light exposure. The photodependent inactivation of the enzyme by the analogue could be partially blocked by 20 mM-ATP and more effectively blocked by either 20 mM-adenosine 5'-[beta gamma-imido]triphosphate or 20 mM-guanosine 5'-[beta gamma-imido]triphosphate, indicating nucleotide-binding site specificity. Arylazido-beta-alanyl-[alpha-32P]ATP was capable of labelling membrane proteins in a photodependent manner. Numerous proteins were labelled, the most prominent of which ran with an apparent Mr of 53000 on polyacrylamide-gel electrophoresis. A band of minor intensity was seen of Mr corresponding to the EGF receptor (170000). Immunoprecipitation of affinity-labelled and solubilized membranes with an anti-(EGF receptor) monoclonal antibody demonstrated that the Mr 170000 receptor protein was photoaffinity labelled by the analogue. The Mr 53000 peptide was not specifically bound by the anti-receptor antibody. The affinity labelling of the receptor was not enhanced by EGF, suggesting that EGF stimulation of the kinase activity does not result from changes in the affinity of the kinase for ATP. These studies demonstrate that arylazido-beta-alanyl-ATP interacts with the ATP-binding site of the EGF-receptor kinase with apparent high affinity and that this analogue is an effective photoaffinity label for the kinase. Furthermore, these studies demonstrate that the EGF receptor, identified by using monoclonal antibodies, contains an ATP-binding site, providing further confirmation that the EGF receptor and EGF-dependent protein kinase are domains of the Mr 170000 protein.     

NEM modification prevents high-affinity ATP binding to the first nucleotide binding fold of the sulphonylurea receptor, SUR1

Pancreatic beta-cell ATP-sensitive potassium channels, composed of SUR1 and Kir6.2 subunits, serve as a sensor for intracellular nucleotides and regulate glucose-induced insulin secretion. To learn more about the interaction of SUR1 with nucleotides, we examined the effect of N-ethylmaleimide (NEM) modification. Photoaffinity labeling of SUR1 with 5 microM 8-azido-[alpha-32P]ATP or 8-azido-[gamma-32P]ATP was inhibited by NEM with Ki of 1.8 microM and 2.4 microM, and Hill coefficients of 0.94 and 1.1, respectively. However, when the cysteine residue in the Walker A motif of the first nucleotide binding fold (NBF1) of SUR1 was replaced with serine (C717S), photoaffinity labeling was not inhibited by 100 microM NEM. These results suggest that NBF1 of SUR1 has a NEM-sensitive structure similar to that of NBF1 of MDR1, a multidrug transporter, and confirm NBF1 as the high-affinity ATP binding site on SUR1.     

Sequence requirements of the ATP-binding site within the C-terminal nucleotide-binding domain of mouse P-glycoprotein: structure-activity relationships for flavonoid binding

Sequence requirements of the ATP-binding site within the C-terminal nucleotide-binding domain (NBD2) of mouse P-glycoprotein were investigated by using two recombinantly expressed soluble proteins of different lengths and photoactive ATP analogues, 8-azidoadenosine triphosphate (8N(3)-ATP) and 2',3',4'-O-(2,4,6-trinitrophenyl)-8-azidoadenosine triphosphate (TNP-8N(3)-ATP). The two proteins, Thr(1044)-Thr(1224) (NBD2(short)) and Lys(1025)-Ser(1276) (NBD2(long)), both incorporated the four consensus sequences of ABC (ATP-binding cassette) transporters, Walker A and B motifs, the Q-loop, and the ABC signature, while differing in N-terminal and C-terminal extensions. Radioactive photolabeling of both proteins was characterized by hyperbolic dependence on nucleotide concentration and high-affinity binding with K(0.5)(8N(3)-ATP) = 36-37 microM and K(0.5)(TNP-8N(3)-ATP) = 0.8-2.6 microM and was maximal at acidic pH. Photolabeling was strongly inhibited by TNP-ATP (K(D) = 0.1-5 microM) and ATP (K(D) = 0.5-2.7 mM). Since flavonoids display bifunctional interactions at the ATP-binding site and a vicinal steroid-interacting hydrophobic sequence [Conseil, G., Baubichon-Cortay, H., Dayan, G., Jault, J.-M., Barron, D., and Di Pietro, A. (1998) Proc. Natl. Acad. Sci. U.S.A. 95, 9831-9836], a series of 30 flavonoids from different classes were investigated for structure-activity relationships toward binding to the ATP site, monitored by protection against photolabeling. The 3-OH and aromaticity of conjugated rings A and C appeared important, whereas opening of ring C abolished the binding in all but one case. It can be concluded that the benzopyrone portion of the flavonoids binds at the adenyl site and the phenyl ring B at the ribosyl site. The Walker A and B motifs, intervening sequences, and small segments on both sides are sufficient to constitute the ATP site.     

The Universal Stress Protein UspC Scaffolds the KdpD/KdpE Signaling Cascade of Escherichia coli under Salt Stress

The sensor kinase KdpD and the response regulator KdpE control induction of the kdpFABC operon encoding the high-affinity K⁺-transport system KdpFABC in response to K⁺ limitation or salt stress. Under K⁺ limiting conditions the Kdp system restores the intracellular K⁺ concentration, while in response to salt stress K⁺ is accumulated far above the normal content. The kinase activity of KdpD is inhibited at high concentrations of K⁺, so it has been puzzling how the sensor can be activated in response to salt stress. Here, we demonstrate that the universal stress protein UspC acts as a scaffolding protein of the KdpD/KdpE signaling cascade by interacting with a Usp domain in KdpD of the UspA subfamily under salt stress. Escherichia coli encodes three single domain proteins of this subfamily, UspA, UspC, and UspD, whose expression is up-regulated under various stress conditions. Among these proteins only UspC stimulated the in vitro reconstructed signaling cascade (KdpD→KdpE→DNA) resulting in phosphorylation of KdpE at a K⁺ concentration that would otherwise almost prevent phosphorylation. In agreement, in a ΔuspC mutant KdpFABC production was down-regulated significantly when cells were exposed to salt stress, but unchanged under K⁺ limitation. Biochemical studies revealed that UspC interacts specifically with the Usp domain in the stimulus perceiving N-terminal domain of KdpD. Furthermore, UspC stabilized the KdpD/KdpE∼P/DNA complex and is therefore believed to act as a scaffolding protein. This study describes the stimulation of a bacterial two-component system under distinct stress conditions by a scaffolding protein, and highlights a new role of the universal stress proteins.     

Functional asymmetry of the two nucleotide binding domains in the ABC transporter Ste6

The yeast a-factor transporter Ste6 is a member of the ABC transporter family and is closely related to human MDR1. We constructed a set of 26 Ste6 mutants using a random mutagenesis approach. Cell fractionation experiments demonstrated that most of the mutants, with the notable exception of those with alterations in TM1, are transported to the plasma membrane, the presumptive site of action of Ste6. Trafficking, therefore, does not seem to be affected in most of the mutants. To identify regions in Ste6 that interact with the ABC transporter "signature motif" (LSGGQ) we screened for intragenic revertants of the LSGGQ mutant M68 (S507N). Suppressor mutations were identified in TM12 and upstream of TM6. Surprisingly, these mutations also suppressed the Walker A mutation G397D, which should be defective in ATP-binding and hydrolysis at NBD1. Photoaffinity labeling experiments with 8-azido-[alpha-32P]ATP showed that ATP binding at NBD2 is reduced by the suppressor mutation in TM12. The experiments further suggest that the two NBDs of Ste6 are not equivalent and affect each other's ability to bind and hydrolyze ATP.     

X-ray structure of HPr kinase: a bacterial protein kinase with a P-loop nucleotide-binding domain

HPr kinase/phosphatase (HprK/P) is a key regulatory enzyme controlling carbon metabolism in Gram- positive bacteria. It catalyses the ATP-dependent phosphorylation of Ser46 in HPr, a protein of the phosphotransferase system, and also its dephosphorylation. HprK/P is unrelated to eukaryotic protein kinases, but contains the Walker motif A characteristic of nucleotide-binding proteins. We report here the X-ray structure of an active fragment of Lactobacillus casei HprK/P at 2.8 A resolution, solved by the multiwavelength anomalous dispersion method on a seleniated protein (PDB code 1jb1). The protein is a hexamer, with each subunit containing an ATP-binding domain similar to nucleoside/nucleotide kinases, and a putative HPr-binding domain unrelated to the substrate-binding domains of other kinases. The Walker motif A forms a typical P-loop which binds inorganic phosphate in the crystal. We modelled ATP binding by comparison with adenylate kinase, and designed a tentative model of the complex with HPr based on a docking simulation. The results confirm that HprK/P represents a new family of protein kinases, first identified in bacteria, but which may also have members in eukaryotes.     

Homology model of RSK2 N-terminal kinase domain, structure-based identification of novel RSK2 inhibitors, and preliminary common pharmacophore

Ribosomal S6 kinase 2 (RSK2) is a serine/threonine kinase that plays a role in human cancer and Coffin-Lowry syndrome and is comprised of two nonidentical kinase domains, each domain with its own ATP-binding site. RSK2 can be inactivated by different types of small organic molecules. Potent RSK2 inhibitors include the two classic bisindole maleimide PKC inhibitors, Ro31-8220 and GF109203X, and the natural product SL0101 that was shown to bind specifically to the ATP pocket of the N-terminal domain (NTD). In this paper, we present an atomic model of the RSK2 NTD (residues 68-323), which was built to simultaneously bind the distinctive molecular scaffolds of SL0101, Ro31-8220, and GF109203X. The RSK2 NTD model was used to identify two novel RSK2 inhibitors from the National Cancer Institute open chemical repository and to develop a preliminary structure-based pharmacophore model.