Abstract
Treatment of bacterial infections is a great challenge of our era due to the various resistance mechanisms against antibiotics. Antimicrobial peptides are considered to be potential novel compound as antibiotic treatment. However, some bacteria, especially many human pathogens, are inherently resistant to these compounds, due to the expression of BceAB-type ABC transporters. This rather new transporter family is not very well studied. Here, we report the first full characterization of the nucleotide binding domain of a BceAB type transporter from Streptococcus agalactiae, namely SaNsrF of the transporter SaNsrFP, which confers resistance against nisin and gallidermin. We determined the NTP hydrolysis kinetics and used molecular modeling and simulations in combination with small angle X-ray scattering to obtain structural models of the SaNsrF monomer and dimer. The fact that the SaNsrFH202A variant displayed no ATPase activity was rationalized in terms of changes of the structural dynamics of the dimeric interface. Kinetic data show a clear preference for ATP as a substrate, and the prediction of binding modes allowed us to explain this selectivity over other NTPs.
Similar content being viewed by others
Introduction
Therapeutic compounds against bacterial infections are currently one of the biggest needs worldwide. Among antibiotics, antimicrobial peptides (AMP) offer promising potential for the treatment of bacterial infections, alone or in combination with already known molecules1,2. An alarming number of pathogenic multidrug resistant strains have evolved under the selective pressure caused by decades of incorrect antibiotic usage. Among them, methicillin-resistant Staphylococcus aureus (MRSA) or vancomycin-resistant Enterococcus (VRE) pose a high risk to therapeutic regimens3. To include new classes of antibiotics in therapy, studies were performed with lantibiotics, a class of AMPs. These ribosomally-synthesized peptides exhibit high potency against several human pathogenic bacterial strains2,3,4 and show high stability to chemical and enzymatic degradation due to multiple intramolecular thioether rings and unsaturated amino acids4,5,6,7,8.
Most known lantibiotics act similar in that they inhibit cell wall synthesis9. A common target for AMPs is the peptidoglycan layer, which exists in Gram-positive as well as Gram-negative bacteria. It is built up by altering amino sugars such as N-acetylglucosamine (GlcNAc) and N-acetylmuramic acid (MurNAc) and stabilized by a cross-linkage of those polymer chains. The inhibition of the cell wall synthesis results in reduced cell growth and subsequent cell death. The well-known lantibiotic nisin contains five lanthionine rings and primarily targets the cell wall precursor Lipid II. The initial binding of the first two N-terminal lanthionine rings (A and B) of the lantibiotic to Lipid II is followed by a reorientation of the C-terminus into the membrane, resulting in pore formation and subsequently cell lysis10,11. Even though lantibiotics are effective in the nanomolar range, their application is hampered by resistance-conferring mechanisms found in human pathogenic bacteria7,12,13. The resistance is mediated by a newly discovered class of ATP binding cassette transporters, called Bacitracin efflux ABC transporters (BceAB), named after their first discovery in the bacitracin resistant strain of Bacillus subtilis14,15. In Streptococcus agalactiae such a BceAB-type ABC transporter is also present, as part of an operon that confers resistance against the lantibiotic nisin16. This operon consists of the membrane-associated protease SaNsr17, the ABC transporter SaNsrFP8, and the two-component system comprising the response regulator SaNsrR and the histidine kinase SaNsrK18. So far, structural information is known only for SaNsr17 and SaNsrR18.
Like all ABC transporters, BceAB-type transporters are composed of a nucleotide-binding domain (NBD) and a transmembrane domain (TMD). The NBD hydrolyses ATP, which drives conformational changes in the TMD, leading to substrate translocation. The TMD of BceAB-type ABC transporters are characterized by ten predicted transmembrane helices and a large extracellular domain (ECDL) of ~ 220 amino acids that is the hallmark of this transporter family4,8,16.
Sequences of the TMD domains from various BceAB-type ABC transporters are not very similar, which explains the large variety of substances they are able to translocate16. In contrast, NBDs share sequence and distinct motifs which are highly conserved throughout the ABC transporter superfamily19,20,21,22. NBDs are mainly L-shaped and comprise a helical signaling domain and a catalytic domain built of α-helices and β-strands23,24,25. The catalytic domain contains the Walker A motif that forms the nucleotide-binding site. A glutamate residue in the Walker B motif takes part in proper nucleotide binding; the γ-phosphate of the ATP molecule is sensed by a conserved histidine (H-loop) which when mutated results in an inactive variant22,23,26. Signaling and catalytic domains are connected by the Q- and the P-loop. Within the signaling domain the C-loop is located, which is the signature motif of an ABC transporter (for an alignment see Fig. S6 and Table S2)22,27,28.
Dimerization of two NBD monomers in a head-to-tail conformation, is needed to enable ATP hydrolysis with the nucleotide binding sites located in the dimer interface. Each ATP molecule is sandwiched between the Walker A motif of one monomer and the C-loop of the second one, which results in a closed, stable complex24,29,30,31. An interaction between the NBD and the nucleotide is supposed to occur by π–π-stacking between the aromatic ring system of the nucleotide and an aromatic residue of the protein (F or Y). Hence, no preference towards any nucleotide-triphosphate (NTP) has been assumed24, as also observed for example for yeast PDR532. The hydrolysis of ATP is coupled to the presence of a cofactor, almost exclusively Mg2+, which is coordinated by the Walker B motif. The divalent cation participates in the hydrolytic attack on the γ-phosphate of the nucleotide26,28,31.
Here, we report for the first time biochemical and structural characteristics of the BceA nucleotide binding domain SaNsrF, through NTP hydrolysis assays, molecular modeling and simulations. SaNsrF is part of the BceAB-type ABC transporter NsrFP from Streptococcus agalactiae16. We show that the NBD SaNsrFWT and its hydrolysis-deficient variant SaNsrFH202A are monomeric in solution. Broad-ranging in vitro ATPase screenings delivered detailed information about the protein’s properties with regard to its structure and physiology. We show that the preferred substrate of SaNsrF is ATP as demonstrated by its kinetic parameters. Moreover, we built a structural model of the ATP/Mg2+-bound SaNsrF protein in its monomeric and dimeric form by comparative modeling and molecular dynamics simulations. In all, this constitutes the first biochemical characterization of a BceAB-type NBD.
Results
Cloning, expression and purification
For substrate transport BceAB-type ABC transporters depend on energy supply generated by ATP hydrolysis, which is mediated by the NBD. Here, we characterized the NBD NsrF of the BceAB-type ABC transporter NsrFP from Streptococcus agalactiae. To heterologously express SaNsrFWT and SaNsrFH202A, we constructed expression vectors using a codon-optimized version of SaNsrF for the heterologous expression in E. coli (Gen Bank accession number: WP_000923537). These constructs expressed a SaNsrF protein with an N-terminal His10-tag attached for purification using Metal Ion Affinity Chromatography. The corresponding SaNsrF constructs were expressed under the control of the plasmid-based T7-promoter via induction with Isopropyl-β-D-thiogalactopyranoside (IPTG). SaNsrFWT was purified to high homogeneity (Fig. 1A), and was examined by Size Exclusion Chromatography coupled to Multiangle Light Scattering (SEC-MALS)33, which revealed a molecular mass of 31.9 ± 0.4 kDa for the SaNsrFWT protein (Fig. 1B). This corresponds nicely with the calculated theoretical molecular mass of the recombinant monomer of 30.9 kDa including the His10-tag. Thus, the conducted SEC-MALS analysis revealed that SaNsrFWT exists as a stable monomer in solution, which is in line with previous observations of other NBDs from different ABC transporter families34,35,36.
By sequence alignments, His202 was identified to be the essential residue of the H-loop37,38,39. As shown for other NBDs, a point mutation to alanine results in a loss of the ATPase activity of the NBD. We generated this variant of SaNsrF (SaNsrFH202A), which indeed displayed no NTP hydrolysis (see below). This variant served as a negative control in all our experiments. The lack of NTP hydrolysis for SaNsrFH202A is in line with in vivo studies that show that this variant abolishes the activity of SaNsrFP8,40.
Activity of SaNsrFWT
After successful purification, we functionally characterized SaNsrFWT. To do so, we screened the following parameters for their influence on the ATP hydrolysis velocity: (I) pH, (II) salt concentration, (III) nature of the divalent ion and (IV) temperature (see Supporting Information and Fig. S1). As a result, the optimized conditions were found to be 100 mM HEPES at pH 7 with 0 mM NaCl as an assay buffer. The buffer included 10 mM Mg2+ and the reaction was finally performed at 30 °C, with an incubation time of 18 min (Fig S1). These optimized conditions were applied in all following experiments.
Velocity of NTP hydrolysis by SaNsrFWT and SaNsrFH202A
Kinetic measurements were performed by quantifying the NTP hydrolysis under increasing concentrations of the respective nucleotide. We determined the NTP hydrolysis behaviour of SaNsrFWT and SaNsrFH202A using increasing amounts of ATP, GTP, CTP or UTP.
As depicted in Fig. 2A, the SaNsrFWT protein demonstrated a nonlinear dependency of ATPase activity over a range of 0–5 mM ATP. The maximal reaction velocity was calculated to be 190.9 ± 10.0 nmol min−1 mg−1 when using ATP. Moreover, the calculation of the kinetic parameters resulted in a kinetic constant of khalf = 0.41 ± 0.05 mM and a Hill coefficient of h = 1.72 ± 0.27 (Fig. 2A and Table 1). A Hill coefficient > 1 demonstrates a cooperative behaviour, and suggests that SaNsrFWT needs to dimerize to hydrolyze ATP, which is in line with other previously characterized NBDs41,42,43. For GTP, the maximal reaction velocity was 221.6 ± 11.1 nmol min−1 mg−1 with a Hill coefficient of h = 1.82 ± 0.27 and a khalf value of 0.69 ± 0.07 mM (Fig. 2B and Table 1). Interestingly, the highest reaction velocity with a value of 339.0 ± 30.4 nmol min−1 mg−1 was reached using CTP as a substrate with the highest measured khalf value of 1.23 ± 0.20 mM and a Hill coefficient of 1.63 ± 0.53 (Fig. 2C and Table 1). The kinetic parameters using UTP as a substrate resulted in comparably high values of vmax = 314.8 ± 23.4 nmol min−1 mg−1, khalf = 0.90 ± 0.13 mM and h = 1.55 ± 0.25 (Fig. 2D and Table 1). The variant SaNsrFH202A displayed no hydrolytic activity for any of the four used NTPs (Fig. 2, dashed lines).
Structural models of SaNsrF monomer and dimer
Since no experimental structure of SaNsrF is available, we generated a structural model of the NBD by comparative modeling. NBDs are the most conserved parts of ABC transporters and in the case of SaNsrF, the templates used for modeling show a sequence identity of ~ 30–40% and a sequence similarity of 84–89% (Table S1). Of these X-ray structures (resolution between 1.7 and 3.4 Å), two constitute NBDs in the functionally active assembly; they were crystallized with the TMD of the macrolide exporter MacAB from Acinetobacter baumannii (PDB ID 5GKO44) and MacAB-like from Streptococcus pneumoniae (PDB ID 5XU145).
The homology model of SaNsrFWT in the monomeric form is of high quality, given the low overall TopScore46 (TS) value of 0.24 (Fig. 3A). This superimposition-free score evaluates local distance differences47 of all atoms in a model, and a value closer to zero indicates higher quality. The regions modeled with lower reliability (TS > 0.5), accounting only for ~ 6% of the total sequence, are located at the β-hairpin (residues 15–18) and the two C-terminal helices (residues 229–232, 235–236, 246–250). Both substructures can be found in other NDBs, however, indicating the plausibility of the model. For example, when compared to the structure of ComA from Streptococcus mutans (PDB ID 3VX448), the C-terminal helices have a virtually identical fold, with an RMSD of 0.6 Å for the last 50 residues, based on sequence alignment followed by structural superimposition.
The dimeric SaNsrFWT model is structurally similar to other known structures, given RMSD values of ~ 5 Å or lower (RMSD of 3.5 Å, 4.5 Å and 5.2 Å for PDB IDs 1L2T, 5GKO, and 5XU1, respectively), indicating the suitability of the performed protein–protein docking. The reliability of the model is additionally verified by the presence of conserved motifs (Fig. 2B and Table S2), such as the phosphate-binding loop (P-loop or Walker A motif), the cofactor-chelating region (Walker B motif), and a short consensus sequence “LSGGQ” (C-loop or ABC signature motif), which signify ABC transporter family membership at the sequence level. Moreover, the α-helical and RecA-like domains are in the canonical head-to-tail arrangement (Fig. 3C). Interestingly, the calculated electrostatic potential shows a clear polarization (Fig. 3D) with positively charged residues (such as R and K) prevalent on the dimer’s side oriented towards the membrane (named “top”) and negatively charged residues (such as D and E) on the opposite side (named “bottom”) in agreement with the expected topology.
Structural dynamics at the NBD–NBD interface and impact of the SaNsrFH202A substitution
The SaNsrF models were subjected to all-atom MD simulations of in total 10 μs length to investigate the structural dynamics at the NDB-NDB interface and to highlight the impact of the H202A substitution on ATP/Mg2+ binding. The RMSD profiles for SaNsrFWT and SaNsrFH202A monomers (Fig. S2) reach almost immediately a plateau at ~ 4 Å, indicating that the overall structure is mostly invariant over simulation times of 0.5 μs for each replica. Additionally, the low variability of ATP/Mg2+ coordinates (Fig. S3A,B) suggests that the SaNsrFH202A substitution does not impact ATP/Mg2+ binding, at least on the timescale of our simulations.
The RMSD profile for the SaNsrFWT and SaNsrFH202A dimers is mostly invariant (Fig. S4A) when the structures are superimposed onto the two subunits separately (red and blue lines). However, when the superimposition is done with respect to the least mobile regions in the whole dimer (black line), RMSD values reach ~ 6–9 Å in three out of five replicas for SaNsrFWT, indicating that the arrangement of the two subunits changes during the simulations. In particular, the interface between the subunits partially opens (Fig. S4B) up to ~ 25 Å (Fig. S5). The change of ATP molecule and Mg2+ ion positions relative to the protein is more marked for SaNsrFWT dimer (Fig. S3). Interestingly, this is not happening in the SaNsrFH202A variant, where the interface seems to be more stable.
In terms of structural mobility, the central region of SaNsrFWT and SaNsrFH202A (residues ~ 50–150) shows a different profile in monomers and dimers (Fig. 4). In monomers (Fig. 4A,B), this region is less mobile than in dimers (Fig. 4C,D), with RMSF values lower than 2 Å and up to 4 Å, respectively. Moreover, in the dimeric SaNsrFH202A variant, this region is slightly less mobile than in SaNsrFWT. The residues of the central region are oriented towards the TM region of the transporter (Fig. 4E,D). In addition, after the alignment of SaNsrF with NBDs of structures containing the TMD (PDB ID 5XU1, Fig. 4G), most of the residues of this central region are located at < 5 Å distance from the coupling helices (CH1, between TM2 and TM3, and C-terminal CH2) of the transporter, suggesting that this central region is involved in NBD-TMD communication (Fig. 4H). A similar result was found for the HlyB transporter50, where the X-loop motif (corresponding to residues 137–142 in SaNsrF, located in the central region) has been proposed to be an important part of the NBD-TMD communication. Even though we are considering an ATP-bound pre-hydrolysis state, SaNsrF in the dimer seems to be generally more mobile than in the monomer, in agreement with the idea that a dimeric assembly is needed in order to perform its function.
H-bond analysis in SaNsrFWT and SaNsrFH202A dimers reveals that the number of H-bond interactions between SaNsrF and the ligands (ATP molecules and magnesium ions) is on average higher in the case of the SaNsrFH202A variant (Fig. 5A). This is due to the higher structural stability compared to SaNsrFWT. Besides the three residues used as restraints for protein–protein docking (S43–R152–D176), other residues contribute to the stability of the dimer with H-bond occupancies up to 70%, such as R13, T14, R15, E42, E144, and R178 (Fig. 5B,C). Surprisingly, the residue-wise H-bond occupancy in SaNsrFWT is significantly higher (p < 0.01) for two specific H-bonds involving both side chains and backbone atoms (D136–R15 and R133–R15), although the interface of the SaNsrFWT dimer is less structurally stable (see above). Indeed, in the initial dimeric model, these interactions are not present, but require the movement of one monomer to the other for them to form.
To conclude, the generated models show a high structural stability over the simulation lengths. In the dimers, the central region is more mobile than in the monomers; in SaNsrFWT, the interface between subunits is structurally less stable than in the SaNsrFH202A variant. Since a shift of one monomer to the other is necessary for NDBs to perform their function, these results together suggest that the mutation SaNsrFH202A impacts the structural dynamics at the SaNsrF interface and not only the catalytic mechanism.
Small angle X-ray scattering
Unfortunately, we were not able to crystalize the SaNsrF protein, although extensively tried. In order to experimentally validate this new model, we choose Small Angle X-Ray Scattering (SAXS) to compare the theoretical model with the experimental scattering (Fig. 6A) measured with the Xenocs Xeuss 2. Based on the experimental data, we calculated an ab initio model for SaNsrFWT with the program GASBOR51 and obtained a χ2 value of 0.97. Superimposing the ab initio and the TopModel model reveals that the structure and the envelope obtained by the SAXS experiment overlap, but also a density tail at the C-terminus of SaNsrFWT (Fig. 6B) that is not occupied by the model. Scrutinizing the templates used by TopModel52 shows that this helical part (Fig. 6B, orange helix) is rather unstructured or even missing. This finding indicated that this region might be highly flexible in solution, thereby covering the available free space in the SAXS envelope (Fig. 6B, red helix). With the program CRYSOL53 we compared the theoretical scattering curve obtained from the TopModel model against the experimental data. The resulting χ2 value of 1.16 indicates a good agreement between the prediction and the experiment. We uploaded the SAXS data and the corresponding model of SaNsrF to the Small Angle Scattering Biological Data Bank (SASBDB)54,55 with the accession code SASDJR3.
Molecular docking of other NTPs
In order to rationalize the hydrolysis preference for ATP over other NTPs, we predicted the binding mode of these molecules in complex with the SaNsrFWT dimer. Ten different pocket conformations, obtained from five equilibrated structures used also for MD simulations times two pockets each, were considered. When focusing on the configurations with lowest Coulomb (ecoul) and van der Waals (evdw) energies, ATP is slightly enriched compared to the other NTP (3 × ATP, 2 × UTP, 1 × CTP and 1 × GTP), suggesting that ATP binding is preferred due to enthalpic contributions to binding (Fig. 7A). Residues giving rise to this preference are those interacting with the nucleobase, namely F12, T49, A23 of one subunit and F143′ and E144′ of the other (Fig. 7B). In particular, the phenylalanines are interacting with the nucleobase by π-π stacking interactions, and the amino groups of CTP and GTP form H-bonds with the backbone oxygen of F143′ and the carboxylate group of E144′, respectively. Since in ATP the amino group has the same orientation as in CTP, a similar kind of H-bond pattern can be expected.
Over respective pockets 1 or 2, which are not symmetric as described above, ATP shows the largest sums of Coulomb and van der Waals energies compared to the other NTPs (Fig. 7C), indicating strongest binding based on enthalpic components, which is in line with the biochemical data where ATP shows the lowest khalf value (Fig. 2 and Table 1).
Discussion
A rather novel family of ABC-transporters, the Bacitracin efflux (Bce) type transporters, have been identified to confer high-level resistance against bacitracin as well as against lantibiotics such as nisin and gallidermin in Bacillus subtilis, Staphylococcus aureus, and Streptococcus agalactiae8,14,16,57,58,59,60. These transporters have been rudimentarily characterized in vitro. We set out to characterize the NBD of the transporter SaNsrFP; this transporter has been shown to be involved in lantibiotic resistance8.
We have purified and characterized the SaNsrFWT and SaNsrFH202A proteins regarding their ability of ATP hydrolysis. The results revealed that inorganic phosphate is only released in a pH range of 6–8, where an HEPES buffer at pH 7 was found to yield maximal ATPase activity. Interestingly, 20% difference could be found in a TRIS buffer system at the same pH (Fig S1A). Similar results were obtained by Zaitseva et al. examining the HlyB-NBD36. In that study, a correlation between the pH of 6 and the pKa values of the glutamate residue and/or the γ-phosphate of the nucleotide and between the pH of 8 and the pKa value of the conserved histidine bound in a salt bridge with the γ-phosphate was made. On that basis, the nucleophilic attack on the γ-phosphate is preceded, originating from a hydrolytic water molecule, which results in the cleavage of the γ-phosphate moiety26,36,61. Moreover, the importance of the conserved histidine could be confirmed since the SaNsrFH202A variant was shown to be incapable of hydrolysing ATP. Here, the ‘linchpin’-role during ATP-hydrolysis is conducted by the H-loop22,36,38,62. Also, this allows a possible explanation for the observed decrease of activity with increasing concentrations of NaCl (Fig. S1B). Since the conserved histidine is in contact with the γ-phosphate of the nucleotide by forming a salt bridge, rising salt concentration could disrupt this existing interaction. In contrast, a buffer system containing 300 mM of NaCl was used for protein storage, which indicates an inverse correlation between protein stability and activity at rising NaCl concentrations63. The incapability of SaNsrFH202A to hydrolyse ATP supports in vivo studies where a loss of resistance against the lantibiotic nisin was observed when expressed in L. lactis bacterial cells8.
Like many other NBDs, SaNsrF was observed to be strictly dependent on its cofactor Mg2+39,64,65, because this is required as a Lewis acid in the catalytic cycle. Mg2+ is involved in proton abstraction from the nucleotide and the nucleophilic attack of the catalytic water, which results in the hydrolytic cleavage of its γ-phosphate36.
Finally, we conducted kinetic measurements including all optimized parameters and the preference of SaNsrFWT and SaNsrFH202A for hydrolysing different NTPs. We propose that the main interaction of the nucleoside triphosphate and the protein occurs by π–π-stacking between the adenine moiety and F12 downstream of the Walker A motif (Fig. 3B,C) as also observed for other NBD’s22,24,25,30. Also, Mg2+, anchored to the protein through Asp and Glu residues of the Walker B motif, interacts with the phosphate region of ATP. The Walker A motif binds to the other side of the phosphate region (Fig. 3B).
Based on a comparison of docked binding poses of other NTPs, additional interacting residues were predicted (Fig. 7B). Amino group-containing NTPs (ATP, CTP and GTP) can form H-bonds with the backbone oxygen of F143′ and the carboxylate group of E144′, whereas purines in ATP and GTP form more extended π–π stacking interactions with F12 and F143′. ATP shows the largest sums of Coulomb and van der Waals energies compared to the other NTPs in either pocket of the NBD, in line with the biochemical data where ATP displayed the lowest khalf value (Fig. 2 and Table 1).
By comparing the measured kinetic parameters of each examined NTP, it becomes obvious that the reactions including UTP or CTP resulted in a significantly higher reaction velocity, respectively, when compared to ATP. Nevertheless, the CTPase and UTPase activities revealed noticeably high kinetic constants (khalf) as well. With regards to the substrate affinity represented by the khalf value, a minimum of 0.41 ± 0.05 mM was reached using ATP as a substrate, which signifies ATP as the most favoured of all four tested NTPs for SaNsrFWT. Hence, ATP has the highest affinity to SaNsrFWT compared to the other examined NTPs, which corresponds with the physiological appearance in vivo of each NTP ([ATP] > [GTP] > [UTP] > [CTP]), which underlines that ATP is the preferred substrate for the protein32,66,67,68. Considering the physiology of purine (ATP, GTP) and pyrimidine (UTP, CTP) nucleotides, we concluded that the involved aromatic ring systems play a major role concerning the substrate affinity and stability of the protein-substrate-complex. Here, pyrimidine bases exhibit a smaller electron density that can be involved in π–π-stacking. Thus, dissociation of pyrimidine nucleotides from the enzyme occurs faster than purine nucleotides. By contrast, the stabilized protein–purine-complex is less liable to dissociation. Together, this may explain the small khalf values found for ATP and GTP and the high reaction velocities caused by a high turnover of CTP and UTP.
NBDs are assumed to share a large number of properties due to highly conserved sequences and specific motifs (see Fig. 3B,C and Table S2)22,23,24,25,26,30. The presence of a certain substrate such as ATP is supposed to induce a dimerization of the two NBD monomers in a typical head-to-tail formation, resulting in two ATP molecules in the dimer interface, sandwiched by the Walker A motif of one monomer and the signature motif of the other one as a cooperative process22,24,25.
NBDs hydrolyse ATP, which drives substrate translocation by conformational changes of the TMD. In the case of the BceAB-type ABC transporter SaNsrFP, the energy supply is provided by the BceA-domain SaNsrF16. By employing SEC-MALS-coupled analysis we were able to confirm a monomeric state of SaNsrFWT and its variant SaNsrFH202A in solution since the measured molecular masses corresponded with the calculated values for each monomer. This agrees with the oligomeric state of other NBDs from other ABC transporter families in the absence of nucleotide34,35,36.
Furthermore, this is in line with our SAXS data that allowed the construction of a volumetric envelope of the SaNsrFWT monomer. The experimental structure of SaNsrF has not been published yet. Here, we generated a structural model using TopModel52 based on five main templates 1F3O_A, 5XU1_B, 2PCL_A, 5GKO_A, 2OLJ_A (Fig. 3A, 6B). We compared this model with the volumetric envelope obtained from SAXS data, showing high reliability and agreement with experimental data. It is striking that the density of the protein model is partly not occupied. A flexible C-terminus could be the reason, which would make a temporary fit of the versatile C-terminal helix to the proposed model possible. As for well-studied NBDs such as HisP, the modeled SaNsrF dimer exhibits the typical head-to-tail formation including two sandwiched ATP molecules in the dimer interface between the Walker A motif of the first monomer and the C-loop of the second one22,24,25,30. Therefore, the SaNsrF protein shares many structural similarities with other known NBDs. As the γ-phosphate moiety of ATP was predicted to be in close proximity of the conserved histidine (H-loop) and the cofactor Mg2+, one can deduce a consensus with the hypothesis of the H-loop acting as a sensor, whereas the cofactor is involved in hydrolytic cleavage while being coordinated by the Walker B motif (Fig. 3B,C)22,23,26,28. Furthermore, in SaNsrFWT, the interface between subunits is structurally less stable than in SaNsrFH202A. Since a shift of one monomer to the other is necessary for NDBs to perform their function, these results suggest that the substitution SaNsrFH202A impacts the structural dynamics at the SaNsrF interface and not only the catalytic mechanism.
Clearly, the SaNsrF protein represent an isolated NBD and we do not know if the kinetic correspond to the ATP hydrolysis that will occur in the presence of the transmembrane protein SaNsrP. However, when comparing the data with known NBDs which has been described before in the presence and absence of the transmembrane segment it can be observed that vmax might be changed, the km values however remains very similar. For example the ATP hydrolysis kinetics have been described for the HlyB NBD as well as for the purified full length transporter in detergent solution26,36,38,43,69. Here the NBD showed a vmax of 200 nmol min−1 mg−1 with a km value of 0.31 where as the full length transporter displayed a lower vmax of 8.1 nmol min−1 mg−1 with a km value of 0.36. This reduction is likely due to the detergent, which is present to keep the HlyB transporter in solution. Important, however is that in both cases the kinetic displayed cooperativity (Hill coefficient > 1) as in the case of SaNsrF and the corresponding histidine mutation also resulted in an inactive protein. This shows that our NTP analysis of the SaNsrF will likely be similar even when the TMD SaNsrP is present. The same observations were found for the nisin transporter NisT from L. lactis70 and the nukacin ISK-1 transporter NukT from Staphylococcus arneri ISK-171 albeit in detergent solution.
In summary, the experiments revealed the first detailed insights into biochemical properties of the BceA domain of the BceAB-type ABC transporter SaNsrFP. We showed that SaNsrFWT and its variant SaNsrFH202A exist as monomers in solution and determined several physiological and structural properties of the protein by evaluating its ATPase activity in comprehensive in vitro studies and molecular modelling and simulations. Hence, this study contributes to the mechanistic and structural understanding of the BceAB-type ABC transporter family, which opens up the possibility to pharmacologically target this family in order to combat multidrug-resistant species in the long run. It further confirms in vivo data where the H202A variant of SaNsrF displayed a loss in the activity, which now can be pinpointed to a lack of ATP hydrolysis, and shows that this variant can well serve as a negative control in studies concerning BceAB type transporters since the histidine is conserved throughout the sequence of this family.
Materials and methods
Expression of SaNsrFWT and SaNsrFH202A
E. coli BL21 (DE3) strains were transformed via heat shock method72 with pET-16b-NHis10-SaNsrFWT or pET-16b-NHis10-SaNsrFH202A, respectively. Precultures were selectively grown with 20 µg mL−1 ampicillin at 37 °C and 180 rpm overnight. Lysogeny Broth (LB) medium was pre-incubated with 20 µg mL−1 ampicillin and inoculated with the respective preculture to an OD600 of 0.1. The cultures were grown to an OD600 of 0.4 at 37 °C and 180 rpm whereupon the temperature was reduced to 18 °C. Protein expression was induced by the addition of 1 mM IPTG at an OD600 of 0.8 and the cultures were further grown overnight.
Protein purification
SaNsrFWT and SaNsrFH202A were purified using Immobilized Metal Ion Chromatography (IMAC). Therefore, a 5 mL HiTrap Chelating HP column, loaded with Zn2+, was equilibrated with low IMAC-buffer (100 mM HEPES at pH 8, 300 mM NaCl, 20% glycerol). Protein elution was undertaken with the high IMAC-buffer (low IMAC-buffer plus 125 mM histidine). A washing step of 40-percent high IMAC-buffer was introduced before. The concentrated eluted proteins were then injected onto a Superdex 75 16/60 size exclusion column at a flow rate of 0.5 mL min−1, pre-equilibrated with SEC buffer (100 mM HEPES at pH 8, 300 mM NaCl, 20% glycerol). Protein eluates were collected and stored at 4 °C.
ATPase activity assay
The ATPase activity of SaNsrFWT and SaNsrFH202A (diluted in 100 mM HEPES at pH 8, 100 mM NaCl) was examined by the Malachite Green Phosphate Assay at a protein concentration of 0.1 mg mL−1 that was initially undertaken at room temperature (20 °C). Several parameters were screened to determine the optimal buffer and temperature conditions for the protein activity (see Supplementary Information).
Kinetic measurements for SaNsrFWT and SaNsrFH202A were performed under the influence of NTP (ATP, GTP, CTP, UTP) with concentrations ranging from 0 to 5 mM.
Therefore, the kinetics were fitted using the Hill equation:
Y: ATPase activity [nmol min−1 mg−1], X: substrate concentration [mM], khalf: substrate concentration at half-maximal reaction velocity [mM], h: Hill coefficient.
All shown data are representing the average of a triple evaluation at least, with the standard deviation reported as errors.
Small angle X-ray scattering (SAXS)
We collected all SAXS data on our Xeuss 2.0 Q-Xoom system from Xenocs, equipped with a PILATUS 3 R 300 K detector (Dectris) and a GENIX 3D CU Ultra Low Divergence x-ray beam delivery system (Xenocs). The chosen sample to detector distance for the experiment was 0.55 m, results in an achievable q-range of 0.18–6 nm−1. All measurements were performed at 15 °C with protein concentrations between 0.5 and 4.2 mg mL−1. Samples were injected in the Low Noise Flow Cell (Xenocs) via autosampler. For each sample, twelve frames with an exposer time of ten minutes were collected. By comparing these frames, we excluded the possibility of aggregation and radiation damage during the measurement. Data were scaled to absolute intensity against water. All used programs for data processing were part of the ATSAS Software package (Version 3.0.1), available on the EMBL website73. Primary data reduction was performed with the program PRIMUS74. With the Guinier approximation we determined the forward scattering I(0) and the radius of gyration (Rg)75. The program GNOM was used to estimate the maximum particle dimension (Dmax) with the pair distribution function p(r)76. Low resolution ab initio models were calculated using GASBOR51. The superposition of a predicted SaNsrF model (see below) was done using the program SUPCOMB56.
Structural models of SaNsrF complexes
As an experimental SaNsrF structure is not available, a homology model was constructed using the template-based protein structure prediction program TopModel52 and the SaNsrFWT sequence as input (NCBI Reference Sequence: WP_000923535.1). In order to build a SaNsrF model arranged in a dimeric assembly with substrate (ATP) and cofactor (Mg2+) bound, starting from the SaNsrFWT monomer in the apo state, a search for sequence similarity and structural properties was performed on the Protein Data Bank. The results were filtered according to the following criteria: sequence identity ≥ 33% and E-value cutoff 0.001 as determined by BLAST77; oligomeric state equals 2; sequence length of 250 ± 50 residues; resolution ≤ 2 Å. Out of six results, only one (PDB ID: 1L2T28) is crystallized as a functionally active “ATP sandwich” symmetrical dimer and was therefore used as a reference. Since ATP is bound at the interface of the dimer and its binding is influenced by both protein subunits, both protein–ligand and protein–protein docking would be particularly challenging in this case. Hence, we constructed first the SaNsrFWT dimer in the apo form and the ATP/Mg2+-bound form subsequently.
To do so, protein–protein docking was performed with the program HADDOCK78,79, using distances between respective three residues that bridge the two subunits together with H-bond interactions as restraints (S40/S43, R153/R152 and D177/D176, for PDB ID 1L2T/SaNsrFWT sequences, respectively). The most similar docking solution to the reference PDB ID 1L2T was used for further modeling steps.
Both, SaNsrFWT monomer and dimer structures were preprocessed with the Protein Preparation Wizard80 of Schrödinger’s Maestro Suite. Since the residues at the binding sites are highly conserved, ATP and Mg2+ are considered to bind in a very similar way as in PDB ID 1L2T. Thus, their coordinates were copied from the reference into the SaNsrFWT model after alignment to one protein subunit. Residues located ≤ 5 Å away from the ATP molecules were energy-minimized using the OPLS 2005 force field81 with standard cutoff values for van der Waals, electrostatic, and H-bond interactions, until the average RMSD of non-hydrogen atoms reaches 0.30 Å. Bond orders as well as missing hydrogen atoms were assigned, and the H-bond network was optimized. Finally, residue 202 was substituted to construct the SaNsrFH202A variant of the monomer and dimer.
Molecular dynamics simulations
In order to validate the modeled protein–protein interface and the ATP binding mode, and to investigate the impact of the SaNsrFH202A substitution on structural dynamics, a set of MD simulations was performed using Amber 201982. Four different ATP/Mg2+-bound SaNsrF systems were prepared for this with the LEaP program83: monomer and dimer, both for SaNsrFWT and SaNsrFH202A.
After establishing charge neutrality by adding sodium counter ions, each system was placed in a truncated octahedral box of TIP3P84 water with a distance of the nearest atom to the border of the box of ≥ 11 Å. Structural relaxation, thermalization, and production runs of MD simulations were conducted with pmemd.cuda85 using the ff14SB force field86 for the protein, Joung-Cheatham parameters87 for ions, and available ATP parameters88. For each starting complex, five independent replicas of 500 ns length each were performed, resulting in a cumulative simulation time of 10 µs. In order to set up independent replicas and obtain slightly different starting structures, the target temperature was set to different values during thermalization (299.8 K, 299.9 K, 300.0 K, 300.1 K, 300.2 K and 300.3 K). A detailed description of the thermalization protocol can be found elsewhere89. The analysis of the MD trajectories was carried out with cpptraj90 on snapshots extracted every 1 ns. All the MD-generated conformations were clustered applying a hierarchical agglomerative approach and an RMSD cutoff value of 4 Å. The representative structure of the SaNsrFWT monomer was compared to the experimentally determined SAXS density.
The representative structure of the most populated cluster for the SaNsrFWT dimer was used to calculate the electrostatic potential with the Adaptive Poisson-Boltzmann Solver (APBS) software package49 as implemented in PyMOL91. Dielectric constants (ε) of 2.0 and 78.0 were used, respectively, for the protein and for water, and the concentration of monovalent cations and anions was set to 0.15 M.
To measure structural mobility, we computed the residue-wise root-mean-square fluctuation (RMSF) of backbone atoms. Structural changes over time, both for the apo SaNsrF proteins and the ATP/Mg2+-bound form, were detected calculating the root-mean-square deviation of atomic positions (RMSD) compared to the initial structure. To describe the changes occurring at the level of the interface, we performed two analyses: (I) measurement of the distance between the center of mass of two residues located in opposite subunits at the center of the interface (S43 and S146); (II) H-bond analysis (i) in terms of the total number of interactions between two subunits (SaNsrFA–SaNsrFB) and between protein and ligands (SaNsrF-(ATP-Mg2+)) and ii) residue-wise H-bound occupancy between residues of the two subunits (SaNsrFA–SaNsrFB), allowing to identify which residues perform more frequent H-bonds throughout the simulations. For this analysis, only H-bonds with the following criteria were considered: occupancy between specific donor and acceptor > 1%; H-bond present in at least two replicas of the same system; H-bonds between two residues with residue-wise occupancy > 10% in at least one system.
Molecular docking of other NTPs
To predict the binding mode of other NTPs in complex with the SaNsrFWT dimer, molecular docking was performed. The starting points for these calculations were the five structures resulting from thermalization and equilibration steps, then used also for independent MD simulations replicas (production).
First, for each binding site a cubic grid of 20 Å length centered on the respective ATP molecule was built in the Maestro platform92, for a total of 10 different grids. Then, starting from the ATP structures, other NTPs were built (GTP, CTP and UTP) by modifying the nucleobase. The generated conformations were refined and scored with the Glide-Extra precision (XP) mode of Glide93. Only the best solution for each NTP in each grid was considered. The Coulomb interaction energy (ecoul) and the van der Waals energy (evdw), components of the XP GlideScore scoring function, were computed, and used to describe the enthalpic contribution of binding.
References
Gould, K. Antibiotics: from prehistory to the present day. J. Antimicrob. Chemother. 71, 572–575. https://doi.org/10.1093/jac/dkv484 (2016).
Meade, E., Slattery, M. A. & Garvey, M. Bacteriocins, potent antimicrobial peptides and the fight against multi drug resistant species: resistance is futile?. Antibiotics https://doi.org/10.3390/antibiotics9010032 (2020).
Cotter, P. D., Hill, C. & Ross, R. P. Bacterial lantibiotics: strategies to improve therapeutic potential. Curr. Protein Pept. Sci. 6, 61–75. https://doi.org/10.2174/1389203053027584 (2005).
Clemens, R., Zaschke-Kriesche, J., Khosa, S. & Smits, S. H. J. Insight into two ABC transporter Families involved in lantibiotic resistance. Front. Mol. Biosci. 4, 91. https://doi.org/10.3389/fmolb.2017.00091 (2017).
Bierbaum, G. & Sahl, H. G. Lantibiotics: mode of action, biosynthesis and bioengineering. Curr. Pharm. Biotechnol. 10, 2–18. https://doi.org/10.2174/138920109787048616 (2009).
Chatterjee, C., Paul, M., Xie, L. & van der Donk, W. A. Biosynthesis and mode of action of lantibiotics. Chem. Rev. 105, 633–684. https://doi.org/10.1021/cr030105v (2005).
Jabes, D. et al. Efficacy of the new lantibiotic NAI-107 in experimental infections induced by multidrug-resistant Gram-positive pathogens. Antimicrob. Agents Chemother. 55, 1671–1676. https://doi.org/10.1128/AAC.01288-10 (2011).
Reiners, J. et al. The N-terminal region of nisin is important for the BceAB-type ABC transporter NsrFP from Streptococcus agalactiae COH1. Front. Microbiol. 8, 1643. https://doi.org/10.3389/fmicb.2017.01643 (2017).
Malin, J. J. & de Leeuw, E. Therapeutic compounds targeting Lipid II for antibacterial purposes. Infect. Drug Resist. 12, 2613–2625. https://doi.org/10.2147/IDR.S215070 (2019).
Breukink, E. et al. Use of the cell wall precursor lipid II by a pore-forming peptide antibiotic. Science 286, 2361–2364. https://doi.org/10.1126/science.286.5448.2361 (1999).
Hasper, H. E., de Kruijff, B. & Breukink, E. Assembly and stability of nisin-lipid II pores. Biochemistry 43, 11567–11575. https://doi.org/10.1021/bi049476b (2004).
AlKhatib, Z. et al. The C-terminus of nisin is important for the ABC transporter NisFEG to confer immunity in Lactococcus lactis. Microbiologyopen 3, 752–763. https://doi.org/10.1002/mbo3.205 (2014).
Draper, L. A., Cotter, P. D., Hill, C. & Ross, R. P. Lantibiotic resistance. Microbiol. Mol. Biol. Rev. 79, 171–191. https://doi.org/10.1128/MMBR.00051-14 (2015).
Ohki, R. et al. The BceRS two-component regulatory system induces expression of the bacitracin transporter, BceAB, Bacillus subtilis. Mol. Microbiol. 49, 1135–1144. https://doi.org/10.1046/j.1365-2958.2003.03653.x (2003).
Podlesek, Z. et al. Bacillus licheniformis bacitracin-resistance ABC transporter: relationship to mammalian multidrug resistance. Mol. Microbiol. 16, 969–976. https://doi.org/10.1111/j.1365-2958.1995.tb02322.x (1995).
Khosa, S., AlKhatib, Z. & Smits, S. H. NSR from Streptococcus agalactiae confers resistance against nisin and is encoded by a conserved nsr operon. Biol. Chem. 394, 1543–1549. https://doi.org/10.1515/hsz-2013-0167 (2013).
Khosa, S. et al. Structural basis of lantibiotic recognition by the nisin resistance protein from Streptococcus agalactiae. Sci. Rep. 6, 18679. https://doi.org/10.1038/srep18679 (2016).
Khosa, S., Hoeppner, A., Gohlke, H., Schmitt, L. & Smits, S. H. Structure of the response regulator NsrR from Streptococcus agalactiae, which is involved in lantibiotic resistance. PLoS ONE 11, e0149903. https://doi.org/10.1371/journal.pone.0149903 (2016).
Kerr, I. D. Structure and association of ATP-binding cassette transporter nucleotide-binding domains. Biochim. Biophys. Acta 1561, 47–64. https://doi.org/10.1016/s0304-4157(01)00008-9 (2002).
Lawson, J., O’Mara, M. L. & Kerr, I. D. Structure-based interpretation of the mutagenesis database for the nucleotide binding domains of P-glycoprotein. Biochim. Biophys. Acta 1778, 376–391. https://doi.org/10.1016/j.bbamem.2007.10.021 (2008).
Walker, J. E., Saraste, M., Runswick, M. J. & Gay, N. J. Distantly related sequences in the alpha- and beta-subunits of ATP synthase, myosin, kinases and other ATP-requiring enzymes and a common nucleotide binding fold. EMBO J. 1, 945–951 (1982).
Szollosi, D., Rose-Sperling, D., Hellmich, U. A. & Stockner, T. Comparison of mechanistic transport cycle models of ABC exporters. Biochim. Biophys. Acta Biomembr. 818–832, 2018. https://doi.org/10.1016/j.bbamem.2017.10.028 (1860).
Schmitt, L., Benabdelhak, H., Blight, M. A., Holland, I. B. & Stubbs, M. T. Crystal structure of the nucleotide-binding domain of the ABC-transporter haemolysin B: identification of a variable region within ABC helical domains. J. Mol. Biol. 330, 333–342. https://doi.org/10.1016/s0022-2836(03)00592-8 (2003).
Schmitt, L. & Tampe, R. Structure and mechanism of ABC transporters. Curr. Opin. Struct. Biol. 12, 754–760. https://doi.org/10.1016/s0959-440x(02)00399-8 (2002).
Wilkens, S. Structure and mechanism of ABC transporters. F1000Prime Rep 7, 14. https://doi.org/10.12703/P7-14 (2015).
Hanekop, N., Zaitseva, J., Jenewein, S., Holland, I. B. & Schmitt, L. Molecular insights into the mechanism of ATP-hydrolysis by the NBD of the ABC-transporter HlyB. FEBS Lett. 580, 1036–1041. https://doi.org/10.1016/j.febslet.2005.11.012 (2006).
Schmees, G., Stein, A., Hunke, S., Landmesser, H. & Schneider, E. Functional consequences of mutations in the conserved “signature sequence” of the ATP-binding-cassette protein MalK. Eur J. Biochem. 266, 420–430. https://doi.org/10.1046/j.1432-1327.1999.00871.x (1999).
Smith, P. C. et al. ATP binding to the motor domain from an ABC transporter drives formation of a nucleotide sandwich dimer. Mol. Cell. 10, 139–149 (2002).
Chen, J., Lu, G., Lin, J., Davidson, A. L. & Quiocho, F. A. A tweezers-like motion of the ATP-binding cassette dimer in an ABC transport cycle. Mol. Cell. 12, 651–661. https://doi.org/10.1016/j.molcel.2003.08.004 (2003).
Locher, K. P. Structure and mechanism of ABC transporters. Curr. Opin. Struct. Biol. 14, 426–431. https://doi.org/10.1016/j.sbi.2004.06.005 (2004).
Szollosi, D., Szakacs, G., Chiba, P. & Stockner, T. Dissecting the forces that dominate dimerization of the nucleotide binding domains of ABCB1. Biophys. J. 114, 331–342. https://doi.org/10.1016/j.bpj.2017.11.022 (2018).
Wagner, M., Smits, S. H. J. & Schmitt, L. In vitro NTPase activity of highly purified Pdr5, a major yeast ABC multidrug transporter. Sci. Rep. 9, 7761. https://doi.org/10.1038/s41598-019-44327-8 (2019).
Sahin, E. & Roberts, C. J. Size-exclusion chromatography with multi-angle light scattering for elucidating protein aggregation mechanisms. Methods Mol. Biol. 899, 403–423. https://doi.org/10.1007/978-1-61779-921-1_25 (2012).
Greller, G., Horlacher, R., DiRuggiero, J. & Boos, W. Molecular and biochemical analysis of MalK, the ATP-hydrolyzing subunit of the trehalose/maltose transport system of the hyperthermophilic archaeon Thermococcus litoralis. J. Biol. Chem. 274, 20259–20264. https://doi.org/10.1074/jbc.274.29.20259 (1999).
Nikaido, K., Liu, P. Q. & Ames, G. F. Purification and characterization of HisP, the ATP-binding subunit of a traffic ATPase (ABC transporter), the histidine permease of Salmonella typhimurium. Solubility, dimerization, and ATPase activity. J. Biol. Chem. 272, 27745–27752. https://doi.org/10.1074/jbc.272.44.27745 (1997).
Zaitseva, J. et al. Functional characterization and ATP-induced dimerization of the isolated ABC-domain of the haemolysin B transporter. Biochemistry 44, 9680–9690. https://doi.org/10.1021/bi0506122 (2005).
Kolaczkowski, M., Sroda-Pomianek, K., Kolaczkowska, A. & Michalak, K. A conserved interdomain communication pathway of pseudosymmetrically distributed residues affects substrate specificity of the fungal multidrug transporter Cdr1p. Biochim Biophys Acta 479–490, 2013. https://doi.org/10.1016/j.bbamem.2012.10.024 (1828).
Zaitseva, J., Jenewein, S., Jumpertz, T., Holland, I. B. & Schmitt, L. H662 is the linchpin of ATP hydrolysis in the nucleotide-binding domain of the ABC transporter HlyB. EMBO J. 24, 1901–1910. https://doi.org/10.1038/sj.emboj.7600657 (2005).
Zhou, Y., Ojeda-May, P. & Pu, J. H-loop histidine catalyzes ATP hydrolysis in the E. coli ABC-transporter HlyB. Phys Chem Chem Phys 15, 15811–15815. https://doi.org/10.1039/c3cp50965f (2013).
Lagedroste, M., Reiners, J., Smits, S. H. J. & Schmitt, L. Systematic characterization of position one variants within the lantibiotic nisin. Sci. Rep. 9, 935. https://doi.org/10.1038/s41598-018-37532-4 (2019).
Davidson, A. L., Laghaeian, S. S. & Mannering, D. E. The maltose transport system of Escherichia coli displays positive cooperativity in ATP hydrolysis. J. Biol. Chem. 271, 4858–4863 (1996).
Senior, A. E. & Bhagat, S. P-glycoprotein shows strong catalytic cooperativity between the two nucleotide sites. Biochemistry 37, 831–836. https://doi.org/10.1021/bi9719962 (1998).
Zaitseva, J. et al. A structural analysis of asymmetry required for catalytic activity of an ABC-ATPase domain dimer. EMBO J. 25, 3432–3443. https://doi.org/10.1038/sj.emboj.7601208 (2006).
Okada, U. et al. Crystal structure of tripartite-type ABC transporter MacB from Acinetobacter baumannii. Nat. Commun. 8, 1336. https://doi.org/10.1038/s41467-017-01399-2 (2017).
Yang, H. B. et al. Structure of a MacAB-like efflux pump from Streptococcus pneumoniae. Nat. Commun. 9, 196. https://doi.org/10.1038/s41467-017-02741-4 (2018).
Mulnaes, D. & Gohlke, H. TopScore: using deep neural networks and large diverse data sets for accurate protein model quality assessment. J. Chem. Theory. Comput. 14, 6117–6126. https://doi.org/10.1021/acs.jctc.8b00690 (2018).
Mariani, V., Biasini, M., Barbato, A. & Schwede, T. lDDT: a local superposition-free score for comparing protein structures and models using distance difference tests. Bioinformatics 29, 2722–2728. https://doi.org/10.1093/bioinformatics/btt473 (2013).
Ishii, S., Yano, T., Okamoto, A., Murakawa, T. & Hayashi, H. Boundary of the nucleotide-binding domain of Streptococcus ComA based on functional and structural analysis. Biochemistry 52, 2545–2555. https://doi.org/10.1021/bi3017069 (2013).
Baker, N. A., Sept, D., Joseph, S., Holst, M. J. & McCammon, J. A. Electrostatics of nanosystems: application to microtubules and the ribosome. Proc. Natl. Acad. Sci. USA 98, 10037–10041. https://doi.org/10.1073/pnas.181342398 (2001).
Damas, J. M., Oliveira, A. S., Baptista, A. M. & Soares, C. M. Structural consequences of ATP hydrolysis on the ABC transporter NBD dimer: molecular dynamics studies of HlyB. Protein Sci. 20, 1220–1230. https://doi.org/10.1002/pro.650 (2011).
Svergun, D. I., Petoukhov, M. V. & Koch, M. H. Determination of domain structure of proteins from X-ray solution scattering. Biophys J 80, 2946–2953. https://doi.org/10.1016/S0006-3495(01)76260-1 (2001).
Mulnaes, D. et al. TopModel: template-based protein structure prediction at low sequence identity using top-down consensus and deep neural networks. J. Chem. Theory Comput. 16, 1953–1967. https://doi.org/10.1021/acs.jctc.9b00825 (2020).
Svergun, D., Barberato, C. & Koch, M. H. J. CRYSOL: a program to evaluate x-ray solution scattering of biological macromolecules from atomic coordinates. J. Appl. Crystallogr. 28, 768–773. https://doi.org/10.1107/S0021889895007047 (1995).
Valentini, E., Kikhney, A. G., Previtali, G., Jeffries, C. M. & Svergun, D. I. SASBDB, a repository for biological small-angle scattering data. Nucleic Acids Res. 43, D357-363. https://doi.org/10.1093/nar/gku1047 (2015).
Kikhney, A. G., Borges, C. R., Molodenskiy, D. S., Jeffries, C. M. & Svergun, D. I. SASBDB: towards an automatically curated and validated repository for biological scattering data. Protein Sci. 29, 66–75. https://doi.org/10.1002/pro.3731 (2020).
Kozin, M. B. & Svergun, D. I. Automated matching of high- and low-resolution structural models. J. Appl. Crystallogr. 34, 33–41. https://doi.org/10.1107/S0021889800014126 (2001).
Collins, B., Curtis, N., Cotter, P. D., Hill, C. & Ross, R. P. The ABC transporter AnrAB contributes to the innate resistance of Listeria monocytogenes to nisin, bacitracin, and various beta-lactam antibiotics. Antimicrob. Agents Chemother. 54, 4416–4423. https://doi.org/10.1128/AAC.00503-10 (2010).
Hiron, A., Falord, M., Valle, J., Debarbouille, M. & Msadek, T. Bacitracin and nisin resistance in Staphylococcus aureus: a novel pathway involving the BraS/BraR two-component system (SA2417/SA2418) and both the BraD/BraE and VraD/VraE ABC transporters. Mol. Microbiol. 81, 602–622. https://doi.org/10.1111/j.1365-2958.2011.07735.x (2011).
Popella, P. et al. VraH is the third component of the Staphylococcus aureus VraDEH system involved in gallidermin and daptomycin resistance and pathogenicity. Antimicrob. Agents Chemother. 60, 2391–2401. https://doi.org/10.1128/AAC.02865-15 (2016).
Staron, A., Finkeisen, D. E. & Mascher, T. Peptide antibiotic sensing and detoxification modules of Bacillus subtilis. Antimicrob. Agents Chemother. 55, 515–525. https://doi.org/10.1128/AAC.00352-10 (2011).
Yang, J., Yu, M., Jan, Y. N. & Jan, L. Y. Stabilization of ion selectivity filter by pore loop ion pairs in an inwardly rectifying potassium channel. Proc. Natl. Acad. Sci. USA 94, 1568–1572. https://doi.org/10.1073/pnas.94.4.1568 (1997).
Ramaen, O. et al. Structure of the human multidrug resistance protein 1 nucleotide binding domain 1 bound to Mg2+/ATP reveals a non-productive catalytic site. J. Mol. Biol. 359, 940–949. https://doi.org/10.1016/j.jmb.2006.04.005 (2006).
Benabdelhak, H. et al. Positive co-operative activity and dimerization of the isolated ABC ATPase domain of HlyB from Escherichia coli. Biochem. J. 386, 489–495. https://doi.org/10.1042/BJ20041282 (2005).
Ramaen, O., Sizun, C., Pamlard, O., Jacquet, E. & Lallemand, J. Y. Attempts to characterize the NBD heterodimer of MRP1: transient complex formation involves Gly771 of the ABC signature sequence but does not enhance the intrinsic ATPase activity. Biochem. J. 391, 481–490. https://doi.org/10.1042/BJ20050897 (2005).
Schultz, K. M., Merten, J. A. & Klug, C. S. Characterization of the E506Q and H537A dysfunctional mutants in the E. coli ABC transporter MsbA. Biochemistry 50, 3599–3608. https://doi.org/10.1021/bi101666p (2011).
Osorio, H. et al. H2O2, but not menadione, provokes a decrease in the ATP and an increase in the inosine levels in Saccharomyces cerevisiae: an experimental and theoretical approach. Eur. J. Biochem. 270, 1578–1589. https://doi.org/10.1046/j.1432-1033.2003.03529.x (2003).
Buckstein, M. H., He, J. & Rubin, H. Characterization of nucleotide pools as a function of physiological state in Escherichia coli. J. Bacteriol. 190, 718–726. https://doi.org/10.1128/JB.01020-07 (2008).
Bochner, B. R. & Ames, B. N. Complete analysis of cellular nucleotides by two-dimensional thin layer chromatography. J. Biol. Chem. 257, 9759–9769 (1982).
Reimann, S. et al. Interdomain regulation of the ATPase activity of the ABC transporter haemolysin B from Escherichia coli. Biochem. J. 473, 2471–2483. https://doi.org/10.1042/BCJ20160154 (2016).
Lagedroste, M., Reiners, J., Smits, S. H. J. & Schmitt, L. Impact of the nisin modification machinery on the transport kinetics of NisT. Sci. Rep. 10, 12295. https://doi.org/10.1038/s41598-020-69225-2 (2020).
Zheng, S., Nagao, J. I., Nishie, M., Zendo, T. & Sonomoto, K. ATPase activity regulation by leader peptide processing of ABC transporter maturation and secretion protein, NukT, for lantibiotic nukacin ISK-1. Appl. Microbiol. Biotechnol. 102, 763–772. https://doi.org/10.1007/s00253-017-8645-2 (2018).
Hanahan, D. Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166, 557–580. https://doi.org/10.1016/s0022-2836(83)80284-8 (1983).
Franke, D. et al. ATSAS 2.8: a comprehensive data analysis suite for small-angle scattering from macromolecular solutions. J. Appl. Crystallogr. 50, 1212–1225. https://doi.org/10.1107/S1600576717007786 (2017).
Konarev, P. V., Volkov, V. V., Sokolova, A. V., Koch, M. H. J. & Svergun, D. I. PRIMUS: a Windows PC-based system for small-angle scattering data analysis. J. Appl. Crystallogr. 36, 1277–1282. https://doi.org/10.1107/S0021889803012779 (2003).
Guinier, A. Diffraction of x-rays of very small angles-application to the study of ultramicroscopic phenomenon. Ann. Phys. 12, 161–237 (1939).
Svergun, D. I. Determination of the regularization parameter in indirect-transform methods using perceptual criteria. J. Appl. Crystallogr. 25, 495–503. https://doi.org/10.1107/S0021889892001663 (1992).
Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. Basic local alignment search tool. J. Mol. Biol. 215, 403–410. https://doi.org/10.1016/S0022-2836(05)80360-2 (1990).
Dominguez, C., Boelens, R. & Bonvin, A. M. HADDOCK: a protein-protein docking approach based on biochemical or biophysical information. J. Am. Chem. Soc. 125, 1731–1737. https://doi.org/10.1021/ja026939x (2003).
van Zundert, G. C. P. et al. The HADDOCK2.2 web server: user-friendly integrative modeling of biomolecular complexes. J. Mol. Biol. 428, 720–725. https://doi.org/10.1016/j.jmb.2015.09.014 (2016).
Schrödinger Release 2017-1: Schrödinger Suite 2017-1 Protein Preparation Wizard, Schrödinger, LLC, New York, NY, 2017.
Banks, J. L. et al. Integrated modeling program, applied chemical theory (IMPACT). J. Comput. Chem. 26, 1752–1780. https://doi.org/10.1002/jcc.20292 (2005).
Case, D. A. et al. Amber (University of California, San Francisco, 2019).
Schafmeister Ceaf, R. W. & Romanovski, V. LEaP (University of California, San Francisco, 1995).
Jorgensen, W. L., Chandrasekhar, J., Madura, J. D., Impey, R. W. & Klein, M. L. Comparison of simple potential functions for simulating liquid water. J. Chem. Phys. 79, 926–935. https://doi.org/10.1063/1.445869 (1983).
Salomon-Ferrer, R., Gotz, A. W., Poole, D., Le Grand, S. & Walker, R. C. Routine microsecond molecular dynamics simulations with AMBER on GPUs 2 explicit solvent particle mesh ewald. J. Chem. Theory Comput. 9, 3878–3888. https://doi.org/10.1021/ct400314y (2013).
Maier, J. A. et al. ff14SB: improving the accuracy of protein side chain and backbone parameters from ff99SB. J. Chem. Theory. Comput. 11, 3696–3713. https://doi.org/10.1021/acs.jctc.5b00255 (2015).
Joung, I. S. & Cheatham, T. E. 3rd. Determination of alkali and halide monovalent ion parameters for use in explicitly solvated biomolecular simulations. J. Phys. Chem. B 112, 9020–9041. https://doi.org/10.1021/jp8001614 (2008).
Meagher, K. L., Redman, L. T. & Carlson, H. A. Development of polyphosphate parameters for use with the AMBER force field. J. Comput. Chem. 24, 1016–1025. https://doi.org/10.1002/jcc.10262 (2003).
Frieg, B. et al. Molecular mechanisms of glutamine synthetase mutations that lead to clinically relevant pathologies. PLoS Comput. Biol. 12, e1004693. https://doi.org/10.1371/journal.pcbi.1004693 (2016).
Roe, D. R. & Cheatham, T. E. 3rd. PTRAJ and CPPTRAJ: SOFTWARE for processing and analysis of molecular dynamics trajectory data. J. Chem. Theory Comput. 9, 3084–3095. https://doi.org/10.1021/ct400341p (2013).
The PyMOL Molecular Graphics System, Version 2.0 Schrödinger, LLC.
Schrödinger Release 2020-2: Maestro, Schrödinger, LLC, New York (2020).
Schrödinger Release 2020-2: Schrödinger Suite 2020-2 Glide, Schrödinger, LLC, New York (2020).
Acknowledgements
We thank Lutz Schmitt for his continuous support and the members of the Institute of Biochemistry for valuable discussions. This work has been funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)—Project number 270650915 (Research Training Group GRK2158, TP4a to H.G. and TP4b to S.S.). The Center for Structural studies is funded by the DFG (Grant Number 417919780 and INST 208/761-1 FUGG to S.S.). H.G. is grateful for computational support by the “Zentrum für Informations und Medientechnologie” at the Heinrich-Heine-Universität Düsseldorf and the computing time provided by the John von Neumann Institute for Computing (NIC) to H.G. on the supercomputer JUWELS at Jülich Supercomputing Centre (JSC) (user IDs: HKF7).
Funding
This work has been funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)—Project Number 270650915 (Research Training Group GRK2158, TP4a to H.G. and TP4b to S.S.). The Center for Structural studies is funded by the DFG (Grant Number 417919780 and INST 208/761-1 FUGG to S.S.). H.G. is grateful for computational support by the “Zentrum für Informations und Medientechnologie” at the Heinrich-Heine-Universität Düsseldorf and the computing time provided by the John von Neumann Institute for Computing (NIC) to H.G. on the supercomputer JUWELS at Jülich Supercomputing Centre (JSC) (user IDs: HKF7). Open Access funding provided by Projekt DEAL.
Author information
Authors and Affiliations
Contributions
S.H.J.S. conceived and coordinated the study and evaluated all data; F.F performed overexpression, purification and functional characterization of the SaNsrF enzymes and contributed to the writing; J.R. performed the SAXS and data analysis, and contributed to the writing; D.T.H., J.S. and J.G. did the cloning and established the purification protocol; H.G. conceptualized and supervised molecular modeling and simulation, analyzed the data, and contributed to the writing; N.P. performed molecular modeling and simulation, analyzed the data, and contributed to the writing.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Furtmann, F., Porta, N., Hoang, D.T. et al. Characterization of the nucleotide-binding domain NsrF from the BceAB-type ABC-transporter NsrFP from the human pathogen Streptococcus agalactiae. Sci Rep 10, 15208 (2020). https://doi.org/10.1038/s41598-020-72237-7
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-020-72237-7
This article is cited by
-
New insights into the resistance mechanism for the BceAB-type transporter SaNsrFP
Scientific Reports (2022)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.