Abstract

Antibiotic-resistant bacteria pose the greatest threat to human health. Among the list of such bacteria released by WHO, carbapenem-resistant Acinetobacter baumannii, for which almost no treatment exists, tops the list. A. baumannii is one of the most troublesome ESKAPE pathogens and mechanisms that have facilitated its rise as a successful pathogen are not well studied. Efforts in this direction have resulted in the identification of Hpa2-Ab, an uncharacterized histone acetyltransferase enzyme of GNAT superfamily. Here, we show that Hpa2-Ab confers resistance against aminoglycoside antibiotics using Escherichia coli DH5α strains in which Hpa2 gene is expressed. Resistivity for aminoglycoside antibiotics is demonstrated with the help of CLSI-2010 and KB tests. Isothermal titration calorimetry, MALDI and acetylation assays indicate that conferred resistance is an outcome of evolved antibiotic acetylation capacity in this. Hpa2 is known to acetylate nuclear molecules; however, here it is found to cross its boundary and participate in other functions. An array of biochemical and biophysical techniques were also used to study this protein, which demonstrates that Hpa2-Ab is intrinsically oligomeric in nature, exists primarily as a dimer and its interface is mainly stabilized by hydrophobic interactions. Our work demonstrates an evolved survival strategy by A. baumannii and provides insights into the mechanism that facilitates it to rise as a successful pathogen.

Introduction

Adaptation to the changing environment is essential for survival of living organisms and at molecular level, adaptation is an outcome of the ability of many enzymes to evolve new functions necessary for survival, in environment of changing chemical conditions [1,2]. This results in a conspicuous feature, i.e. substrate ambiguity. Evolutionary exploitation of substrate ambiguity in a variety of organisms is well known; capability of bacteria to acquire resistance to drugs and pesticides is one such example. Recently, Anderson and co-workers discussed the influence of various factors on evolution of drug-resistant clones [3]. Factors, like adverse environmental conditions, mutation rate and fitness capacity for drug resistance, speed up the emergence of resistivity [4,5]. MDR (multiple drug resistance) A. baumannii exhibits enormous capacity to evolve resistance against most of the clinical antibiotics. Studies done to test aminoglycoside susceptibility pattern in A. baumannii strains demonstrate an alarming increase in resistance against this class of antibiotics. Since last few decades, A. baumannii has emerged as a troublesome nosocomial infectious agent, and it is known to cause secondary infections like pneumonia, skin infection, meningitis and endocarditis in intensive care units (ICUs) [6]. A. baumannii is known to manifest resistance against most trusted therapeutics, i.e. carbapenem, this leads to complications in treatment regime and necessitate development of new drugs to cure A. baumannii infections.

A. baumannii is emerged as a serious threat for the health care systems, still mechanisms responsible for its rise as a successful pathogen are not well known. In MDR pathogens, many mechanisms have evolved to escape the effect of antimicrobials. Inactivation of antibiotics by enzymatic transfer of a chemical group is one of three major mechanisms [7]. Although, to date a large number of aminoglycoside-modifying enzymes have been studied, such studies on the enzymes from A. baumannii have been rather limited [8].

Here, we have selected histone acetyltransferase (Hpa2-Ab) enzyme, a member of the Gcn5-related N-acetyl transferases (GNATs) superfamily, as a possible potential drug target. GNATs have been known to transfer an acetyl group to the amine group in a wide range of substrates and participate in numerous aspects of prokaryotic physiology; however, exact molecular functions of many of them are still not understood. Literature study and results obtained from our in silico work [9] reveal the following interesting properties of Hpa2-Ab enzyme. Firstly, it has the capacity to acetylate molecules like histone, nuclear HMG proteins and polyamines [1013]; however, the actual molecular function of this enzyme is not known. Secondly, it has structural and catalytic site identity with AAC(6_)-Ii enzyme which does acetylation of aminoglycoside class of antibiotics [14]. Also, strong in silico binding affinity with these antibiotics [9] indicates that Hpa2-Ab plausibly binds and acetylates these antibiotics.

To unravel the functional secrets of promiscuous Hpa2-Ab, we performed biophysical and biochemical activity analysis in native, acetyl-CoA and antibiotics-bound forms. The results indicate that Hpa2-Ab exists as a dimer under physiological conditions; however, in the presence of acetyl-CoA it dissociates into a monomer. The plausible role of Hpa2-Ab in antibiotic acetylation and drug resistance is explored using ITC (isothermal titration calorimetry) and in vitro assays. Significant ITC-binding affinity, acetylation activity and resistance activity are seen for kanamycin and streptomycin.

Resistance for aminoglycoside antibiotics emerged many decades ago and several mechanisms responsible for resistance, including chemical modification of aminoglycosides by aminoglycoside-modifying enzymes, have been identified [8,14]. Acetylation is a well-known mechanism of chemical modification [7], and Hpa2-Ab is known to acetylate a variety of nuclear molecules including small molecules, peptides and proteins [1013]. However, that Hpa2-Ab has the capacity to acetylate aminoglycoside antibiotics was not known so far. In the present study, we found that Hpa2-Ab acetylates kanamycin and streptomycin, two of the common aminoglycoside antibiotics.

Materials and methods

Details of experimental parameters and steps performed during cloning, recombinant expression, purification, cross-linking assay, native PAGE, DLS and analytical size exclusion chromatography of Hpa2-Ab are given in the Supplementary File.

ITC assay

ITC assay was performed using an ITC-200 Microcal calorimeter (GE Healthcare) [15]. Reaction cell was filled with a 200 µl protein in 25 mM Tris buffer, and a syringe was loaded with 40 µl of each substrate (2.0, 2.5 mM of acetyl-CoA and aminoglycoside antibiotics) as binding partner. Change in molar enthalpy ΔH, molar Gibbs free energy ΔG, stoichiometry (n) of reaction and binding constants were calculated from the fitted curve. ΔG values were calculated from association constants (KA) derived from fitted titration curves using the equation ΔG = −RTln(KA). ΔH was derived from fitted titration curves, TΔS values were then determined from the relationship ΔG = ΔH − TΔS, where T = 288 K.

NMR and MALDI analysis of apo and bound form of Hpa2-Ab

For NMR experiments, 15N labeled Hpa2-Ab protein was overexpressed in minimal medium having 15NH4Cl as nitrogen source and was purified using the same protocol as given in the Supplementary file. 1H–15N HSQC experiment was performed under physiological conditions. HSQC experiment produces one backbone 1H–15N correlation peak for each amino-acid residue and also peaks from side chains of Trp, Gln and Asn residues. 1H–15N HSQC spectra were also recorded in the presence of acetyl-CoA and kanamycin. Molecular mass of apo and bound forms of Hpa2-Ab were identified by MALDI. Detailed methodologies of MALDI and NMR experiments are explained in supporting information.

Quantitative measurement of in vitro acetyltransferase activity of Hpa2-Ab

5,5′-Dithiobis-(2-nitrobenzoic acid) (DTNB) reagent was used to quantitatively measure in vitro acetylation activity of Hpa2-Ab [16,17]. Standard reaction mixture consists of buffer [25 mM Tris–HCl pH 7.8, 150 mM NaCl, 2 mM EDTA, 5% (v/v) glycerol], 250 µM of acetyl-CoA, and 200 µM of each substrate in each eppendorf and final volume was adjusted to 900 µl. Different reaction mixtures were prepared by adding similar concentration of different substrates and reaction was initiated by the addition of Hpa2-Ab. Reaction mixture was incubated for ∼1 h after that 100 µl of DTNB (8 mg/ml) reagent was added. Formation of yellow color product enables spectrophotometric measurement in cuvettes at wavelength 412 nm and ε = 14 150 M–1 cm–1 using an Analytik Jena, UV–VIS spectrophotometer SPECORD-205 spectrometer. Absorbance data were analyzed to calculate kinetic parameters of acetylation reaction. Detailed methodology is given in the Supplementary file.

Measurement of resistance conferred by Hpa2-Ab using micro-dilution and KB test

Antibiotic susceptibility test was performed by micro-dilution method as recommended by CLSI-2006 method [18]. Normal Escherichia coli DH5α cells were used as control and transformed E. coli DH5α cells having pET28a vector with gene encoding for enzyme Hpa2-Ab, i.e. E. coli DH5α/Hpa2-Ab transformants were used to test resistance against all aminoglycoside antibiotics. Resistance was also tested using a modified Kirby–Bauer disk-susceptibility method [19,20], for KB test paper disk containing various concentrations of kanamycin (50, 100 and 200 µg/ml) and streptomycin (50 and 100 µg/ml) were used. Detailed methodology of micro-dilution and KB test is given in Supplementary Information.

FoldX algorithm was used to calculate molecular interactions contributing to dimer stability. Multiple sequence alignment and phylogenetic tree analysis was performed to get insights about new functional (aminoglycoside acetylation) adaptation in Hpa2-Ab. All these have been detailed in Supplementary Information.

Results and discussion

Recombinant expression and purification of pET28a-Hpa2-Ab construct

In A. baumannii ACICU genome, Hpa2 protein is encoded by the ACICU_RS15490 gene. Gene encoding for protein Hpa2, was cloned into pET28a with N-terminal 6xHis-tag. Analysis of PCR product on 1% agarose gel electrophoresis showed a single amplification band of ∼500 bp size, corresponding to gene size (417 bp) of Hpa2-Ab (Supplementary Figure S1). PCR product was ligated into pET28a vector and the ligated product was transformed into E. coli DH5α cells and E. coli BL21 (DE3) cells. Soluble protein was obtained by induction with 0.7 mM IPTG and growing transformed E. coli at 18°C for ∼16 h (Supplementary Figure S2). Purification of Hpa2-Ab protein was done using affinity and size exclusion chromatography. Fractions obtained from size exclusion column were analyzed using 12% SDS–PAGE gel, which indicated that protein was pure up to 95% and its MW is ∼16 kDa (Supplementary Figure S3).

Hpa2-Ab in the native state is intrinsically a stable dimer

Functional homolog of Hpa2-Ab is Hpa2-Sc (from yeast) and exists as dimer in solution. Hpa2-Sc self-associates to build a tetramer in the presence of acetyl-CoA [12]. Here, oligomeric state of Hpa2-Ab was explored using native PAGE, cross-linking assay, DLS and size exclusion chromatography.

From the column, Hpa2-Ab eluted as a single peak at a volume of 94.08 ml. MW of Hpa2-Ab was estimated by interpolation using globular protein calibration curve (Supplementary Figure S4). Elution volume plotted on standard graph indicates a molecular mass of ∼32 ± 3 kDa, which is twice to the mass of Hpa2-Ab, calculated on the basis of primary sequence. This indicates that Hpa2-Ab exists as a dimer in solution (Figure 1A).

Purification and secondary structure analysis of Hpa2-Ab.

Figure 1.
Purification and secondary structure analysis of Hpa2-Ab.

(A) Analytical size exclusion chromatography of the Hpa2-Ab. Chromatogram obtained after loading the Hpa2-Ab protein on calibrated column Superdex-200 16/600, protein peak appeared at 94.08 ml. Collected protein fractions were checked on the 12% SDS–PAGE and depicted with chromatogram. (B) Far–UVCD spectrum of recombinant Hpa2-Ab protein (0.5 mg/ml in 50 mM phosphate buffer) at pH 7.6 and temperature 20°C ( represents the Far–UVCD spectrum normalized by five window function and ⋄ represent the raw data).

Figure 1.
Purification and secondary structure analysis of Hpa2-Ab.

(A) Analytical size exclusion chromatography of the Hpa2-Ab. Chromatogram obtained after loading the Hpa2-Ab protein on calibrated column Superdex-200 16/600, protein peak appeared at 94.08 ml. Collected protein fractions were checked on the 12% SDS–PAGE and depicted with chromatogram. (B) Far–UVCD spectrum of recombinant Hpa2-Ab protein (0.5 mg/ml in 50 mM phosphate buffer) at pH 7.6 and temperature 20°C ( represents the Far–UVCD spectrum normalized by five window function and ⋄ represent the raw data).

Native PAGE separates protein according to their size and their charge/mass ratio in acrylamide gradient gels [21]. During a run, protein migration gradually decelerates with running distance due to the decreasing pore size of gradient gel, and movement of the protein halts when it approaches its size-dependent specific pore-size limit. Freshly purified Hpa2-Ab was subjected to native PAGE electrophoresis, and a single band is observed on gel at ∼32 kDa which corresponds to dimeric Hpa2-Ab (Figure 2A). Concentration-dependent native PAGE analysis of Hpa2-Ab at concentrations 20–75 µM (Supplementary Figure S5A) and 0.12–0.5 mM (Supplementary Figure S5B) depicts similar band position corresponding to dimeric Hpa2-Ab.

Analysis of Hpa2-Ab in native state.

Figure 2.
Analysis of Hpa2-Ab in native state.

(A) Native PAGE analysis of Hpa2-Ab. Pure protein fractions obtained from column were loaded on well. Lanes, W2 = Wash2 and E1–E5 = Elution fractions. (B,C) Chemical cross-linking analysis of Hpa2-Ab, Hpa2-Ab treated with dimethyl suberimidate at pH 7.8, cross-linked product was analyzed using SDS–PAGE. (C) Glutaraldehyde cross-linking assay of Hpa2-Ab analyzed on 15% SDS–PAGE, Hpa2-Ab treated with different glutaraldehyde conc (0.03, 0.04, 0.05, 0.07, 1, 1.5, 2 mM; Lanes 1 to 7, respectively) for 15 min at 30°C and analyzed using SDS–PAGE. (D) The distribution of hydrodynamic radii obtained using DLS experiments for Hpa2-Ab (red), plotted with hydrodynamic radii values of standard proteins (black).

Figure 2.
Analysis of Hpa2-Ab in native state.

(A) Native PAGE analysis of Hpa2-Ab. Pure protein fractions obtained from column were loaded on well. Lanes, W2 = Wash2 and E1–E5 = Elution fractions. (B,C) Chemical cross-linking analysis of Hpa2-Ab, Hpa2-Ab treated with dimethyl suberimidate at pH 7.8, cross-linked product was analyzed using SDS–PAGE. (C) Glutaraldehyde cross-linking assay of Hpa2-Ab analyzed on 15% SDS–PAGE, Hpa2-Ab treated with different glutaraldehyde conc (0.03, 0.04, 0.05, 0.07, 1, 1.5, 2 mM; Lanes 1 to 7, respectively) for 15 min at 30°C and analyzed using SDS–PAGE. (D) The distribution of hydrodynamic radii obtained using DLS experiments for Hpa2-Ab (red), plotted with hydrodynamic radii values of standard proteins (black).

Chemical cross-linking assay is a powerful technique that provides information about subunit interactions within static protein assemblies. Here, dimethyl suberimidate and glutaraldehyde cross-linking agents [22] were used for cross-linking reaction. Dimethyl suberimidate is quite reactive and its derivatives exhibit unaltered charge, and hence, good solubility. Analysis of dimethyl suberimidate cross-linked Hpa2-Ab showed band at ∼32 kDa (Figure 2B). Concentration-dependent (20–80 µM) cross-linking assays (Supplementary Figure S6A,B) showed no concentration dependence. However, glutaraldehyde cross-linked Hpa2-Ab showed presence of higher molecular mass species (at ∼50 and 65 kDa; Figure 2C), though in minor amounts; higher molecular mass species produced were multiples of the monomer molecular mass. Formation of oligomeric species might be due to non-specific reaction between the protein and glutaraldehyde. At lower concentrations of glutaraldehyde, the protein does not show any oligomeric species. In the absence of cross-linking agent, Hpa2-Ab migrates as a single band at ∼16 kDa (Supplementary Figure S3).

Analysis of dimer interface interactions using FoldX algorithm resulted in favorable free energy of association ΔGassociation, this is contributed by the van der Waals interactions (ΔGvdw = −8.62 kcal mol−1) and hydrophobic term (ΔGHydrophobic = −12.29 kcal mol−1). The analysis attests to the stability of dimeric Hpa2-Ab and suggests that interface is predominantly stabilized by hydrophobic interactions.

Secondary structure of Hpa2-Ab

Circular dichroism spectrum of Hpa2-Ab protein is dominated by helical signature, with minima at 222 and 208 nm (Figure 1B). Far–UV CD spectrum obtained at 20°C (190–260 nm) was analyzed by Dichroweb program using the CDSSTR method [23]. The analysis showed ∼40% α-helix and 29% β-sheets at physiological conditions. CD spectra recorded at protein concentrations of 0.5 and 3.0 mM indicated that the secondary structure and spectral properties (data not shown) are the same, suggesting no change in Hpa2-Ab structure at higher concentrations.

Acetyl-CoA converts dimeric Hpa2-Ab into monomeric form

Oligomeric state of Hpa2-Ab is also studied in the presence of acetyl-CoA. For native PAGE experiment freshly purified Hpa2-Ab was incubated with 750 µM of acetyl-CoA and loaded on the gel. After running the experiment (Figure 3A), two bands were observed at molecular mass ∼16 and ∼32 kDa, corresponding to monomeric and dimeric forms of Hpa2-Ab. Cross-linking assay performed with similar concentrations as utilized in native PAGE exhibits a major band at ∼16 kDa and a minor band at 32 kDa (Figure 3B). Gel band at ∼16 kDa indicates that acetyl-CoA-Hpa2-Ab complex exists in monomeric form. Size exclusion chromatogram depicts two peaks at elution volumes of ∼94 and ∼102 ml for acetyl-CoA-Hpa2-Ab complex (Figure 3C and Supplementary Figure S7A–D). Plotting these values on standard Kav versus log MW graph yields molecular mass of ∼32 and 16 kDa, respectively. Effect of acetyl-CoA concentrations (1 : 1, 1 : 5, 1 : 10 and 1 : 20 ratios) was also monitored here. Twenty times higher concentration of acetyl-CoA was used, but complete conversion into monomer form could not be achieved. DLS experiment performed to derive hydrodynamic radii (RH) indicates that Hpa2-Ab exists as dimeric species with RH equals to 2.69 nm (Figure 2D). However, in the presence of acetyl-CoA, two species with RH equal to 2.18 and 2.69 nm were observed (Figure 3D). Putting observed RH values into a standard graph depicts a linear relationship between RH and MW [24], which correspond to monomeric and dimeric forms, respectively.

Analysis of Hpa2-Ab-acetyl-CoA complex.

Figure 3.
Analysis of Hpa2-Ab-acetyl-CoA complex.

(A) For native PAGE, freshly purified protein was incubated with acetyl-CoA. After electrophoresis, two bands appeared on gel corresponding to monomer and dimer form of Hpa2-Ab. (B) Chemical cross-linking by dimethyl suberimidate, Hpa2-Ab-acetyl-CoA complex in ratio 1 : 1, 1 : 5, and 1 : 10 was cross-linked with DMS and loaded in lanes 1, 2 and 4, respectively. (C) Analytical size exclusion chromatogram of Hpa2-Ab-acety-CoA complex is depicted here. Elution profile of Hpa2-Ab (red line), Hpa2-acetyl-CoA complex (black line) and lysozyme (blue line) is overlaid. (D) The distribution of hydrodynamic radii obtained using DLS experiments for Hpa2-Ab bound with acetyl-CoA (red), plotted with hydrodynamic radii values of standard proteins (black).

Figure 3.
Analysis of Hpa2-Ab-acetyl-CoA complex.

(A) For native PAGE, freshly purified protein was incubated with acetyl-CoA. After electrophoresis, two bands appeared on gel corresponding to monomer and dimer form of Hpa2-Ab. (B) Chemical cross-linking by dimethyl suberimidate, Hpa2-Ab-acetyl-CoA complex in ratio 1 : 1, 1 : 5, and 1 : 10 was cross-linked with DMS and loaded in lanes 1, 2 and 4, respectively. (C) Analytical size exclusion chromatogram of Hpa2-Ab-acety-CoA complex is depicted here. Elution profile of Hpa2-Ab (red line), Hpa2-acetyl-CoA complex (black line) and lysozyme (blue line) is overlaid. (D) The distribution of hydrodynamic radii obtained using DLS experiments for Hpa2-Ab bound with acetyl-CoA (red), plotted with hydrodynamic radii values of standard proteins (black).

ITC experiments for acetyl-CoA and antibiotic binding to Hpa2-Ab

ITC experiments were performed with acetyl-CoA and antibiotic substrates. Binding isotherms acquired by integration of raw data are depicted in Supplementary Figure S8 and calculated binding parameters are given in Table 1. Binding curve for acetyl-CoA was best fitted to one-site model. It turns out that binding of acetyl-CoA is an exothermic process, with change in enthalpy, ΔH = −8.96 ± 0.19 kcal mol−1, change in entropy, ΔS = 3.715 ± 0.73 cal mol−1 deg−1. Net free energy value ΔG (−7.89 ± 0.15 kcal mol–1) demonstrates that acetyl-CoA binding is a thermodynamically favorable process (Supplementary Figure S8A). Acetyl-CoA and Hpa2-Ab-binding constants Kd = 0.98 ± 0.21 µM, Ka = 1.02 × 106 M–1 values are in good correlation with binding constant values obtained for Pat-protein of GNAT superfamily [25].

Table 1
Thermodynamic binding parameters for acetyl-CoA, streptomycin, streptomycin-acetyl-CoA, kanamycin and gentamicin with Hpa2-Ab

Given error was calculated as the standard error of the mean of three trials, where N is the number of sites, KD is the dissociation constant and ΔG is the Gibbs free energy change.

Substrate N KD (µM) KA (M–1ΔG (kcal mol–1
Acetyl-CoA 0.94 0.97 ± 0.6 10.3 × 105 −7.89 ± 0.5 
Streptomycin 1.20 78.6 ± 0.7 12.7 × 103 −5.37 ± 0.3 
Kanamycin 0.95 7.73 ± 0.9 12.9 × 104 −6.71 ± 0.4 
Streptomycin-acetyl-CoA 0.93 17.1 ± 0.9 58.4 × 103 −6.25 ± 0.3 
Gentamicin 3.90 96.2 ± 0.7 10.3 × 103 −5.26 ± 0.7 
Substrate N KD (µM) KA (M–1ΔG (kcal mol–1
Acetyl-CoA 0.94 0.97 ± 0.6 10.3 × 105 −7.89 ± 0.5 
Streptomycin 1.20 78.6 ± 0.7 12.7 × 103 −5.37 ± 0.3 
Kanamycin 0.95 7.73 ± 0.9 12.9 × 104 −6.71 ± 0.4 
Streptomycin-acetyl-CoA 0.93 17.1 ± 0.9 58.4 × 103 −6.25 ± 0.3 
Gentamicin 3.90 96.2 ± 0.7 10.3 × 103 −5.26 ± 0.7 

Titrations of all aminoglycoside antibiotics were performed here; however, most of them did not produce measurable heat exchange, only kanamycin and streptomycin produce a significant amount of heat. ΔG and Kd values of streptomycin (Supplementary Figure S8C) and kanamycin (Supplementary Figure S8D) (ΔG = –5.37 ± 0.3 kcal/mol, Kd = 78.6 ± 0.7 µM and ΔG = −6.71 ± 0.4 kcal/mol, Kd = 7.73 ± 0.9 µM, respectively) are comparable to those of AAC(6_)-Ii enzyme-aminoglycoside antibiotic binding [26]. Higher ΔH and TΔS values for kanamycin binding indicate that after kanamycin binding, Hpa2-Ab might get converted to higher oligomeric form or undergo conformational changes (Figure 4A,B) to drive binding process. Hegde and co-workers studied ternary complex using ITC experiment; titration of aminoglycoside on binary complex (AAC(6′)-Ii-acetyl-CoA) resulted in enhancement of Ka value; increase in Ka value was contributed by increased ΔH value [27]. Similarly, titration of streptomycin and kanamycin on binary complex (Hpa2-Ab-acetyl-CoA) was also performed, but kanamycin did not exhibit any considerable change. For streptomycin, ΔΔG values increased from −5.37 to −6.25 kcal/mol (Supplementary Figure S8B). This increment in ΔΔG value is contributed by favorable ΔS term; however, ΔH value remains almost the same indicating unaltered binding interactions. Binding sites of antibiotics and acetyl-CoA are independent and adjacent on enzyme (Figure 4A and Supplementary Figure S9A), binding affinity of streptomycin is more with binary complex in comparison with free enzyme (Figure 4C). Higher binding affinity of streptomycin with Hpa2-Ab-acetyl-CoA suggests that formation of ternary complex is a synergistic phenomenon.

Characterization of ternary complex.

Figure 4.
Characterization of ternary complex.

(A) Hpa2-Ab_acetyl-CoA_antibiotic ternary complex model. Hpa2-Ab model (interacting part is colored blue), acetyl-CoA (colored according to atoms) and antibiotic (colored green) ternary complex. (B) Conformational changes (in the loop preceding β6) obtained on substrates binding are depicted. (C) ITC profile of streptomycin titration on acetyl-CoA_Hpa2-Ab, Raw data from titration were represented as the heat change (µcal/s) upon injection over time (Top), binding isotherm obtained by integration of the raw data (reported as kcal/mol). Solid line represents the best-fit curve generated from a one-site-binding model (Bottom). (D) MALDI spectrum of Hpa2-Ab-acetylCoA-antibiotic complex, M + H+ species bound to acetyl-CoA (809 Da) and streptomycin (728 Da).

Figure 4.
Characterization of ternary complex.

(A) Hpa2-Ab_acetyl-CoA_antibiotic ternary complex model. Hpa2-Ab model (interacting part is colored blue), acetyl-CoA (colored according to atoms) and antibiotic (colored green) ternary complex. (B) Conformational changes (in the loop preceding β6) obtained on substrates binding are depicted. (C) ITC profile of streptomycin titration on acetyl-CoA_Hpa2-Ab, Raw data from titration were represented as the heat change (µcal/s) upon injection over time (Top), binding isotherm obtained by integration of the raw data (reported as kcal/mol). Solid line represents the best-fit curve generated from a one-site-binding model (Bottom). (D) MALDI spectrum of Hpa2-Ab-acetylCoA-antibiotic complex, M + H+ species bound to acetyl-CoA (809 Da) and streptomycin (728 Da).

DLS experiments were performed to explore the oligomeric state of Hpa2-Ab-kanamycin, Hpa2-Ab-streptomycin complexes. Calculated RH values, 3.6 and 2.7 nm, correspond to molecular masses of 64 ± 62 and 32 ± 2 kDa, respectively. It indicates that while Hpa2-Ab-kanamycin complex exists as a tetramer, streptomycin complex remains in dimeric state (Supplementary Figure S9B). Similar oligomeric state is observed for Hpa2-Ab–actyl-CoA-kanamycin and Hpa2-Ab–actyl-CoA-streptomycin ternary complexes too.

Characterization of Hpa2-Ab in apo and complex forms using MALDI and NMR experiments

MALDI experiments were performed in native, acetyl-CoA and antibiotic-bound form. MALDI data showed peaks at m/z values, 8057, 8085 and16 119.23 which represent M + 2H+, M + 2H++Na+ and M + H+ mass species of Hpa2-Ab, respectively (Supplementary Figure S10).

Calculated average mass of Hpa2-Ab is16 117.55 Da. Analysis of acetyl-CoA and kanamycin-bound forms using MALDI remained unsuccessful. However, Hpa2-Ab-acetyl-CoA-streptomycin ternary complex is captured by the MALDI (Figure 4D); for ternary complex, MALDI peak at m/z value,17 656.08 Da is observed. MALDI peak also obtained at m/z value,16 861.05 Da represents Hpa2-Ab-acetyl-CoA binary complex.

1H–15N HSQC spectrum of Hpa2-Ab recorded at physiological conditions (pH 7.6 and 150 mM NaCl) is depicted in Figure 5. An HSQC spectrum is considered as a fingerprint of the protein, and chemical shift dispersion is an indicator of Hpa2-Ab foldedness. Dispersion of Hpa2-Ab peaks is good (Figure 5A), which indicates that the protein is well folded. Peak count matches the expected number, one peak per residue, and suggests that Hpa2-Ab dimer is symmetric (note that the protein is a dimer under the present conditions as discussed in the previous sections). 1H–15N HSQC spectrum of Hpa2-Ab bound to acetyl-CoA is depicted in Supplementary Figure S11A, its comparison with apo protein spectrum reveals appearance of several extra peaks which is due to coexistence of free dimer and monomer-acetyl-CoA complex. Titration of Hpa2-Ab was performed with acetyl-CoA and kanamycin, and the effect of varying concentrations on tryptophan side chain peaks is depicted in Figure 5B and Supplementary Figure S11B. Titration of kanamycin, monitored on tryptophan (W9, W36) peaks (Figure 5B) suggests that binding of kanamycin induces conformational change in the protein. 15N–1H HSQC spectrum of Hpa2 was recorded at different concentrations of acetyl-CoA and which are depicted in Figure 5C. It shows changes in overall 3D structure of protein on binding.

1H–15N HSQC spectra of Hpa2-Ab in apo and complex form.

Figure 5.
1H–15N HSQC spectra of Hpa2-Ab in apo and complex form.

(A) 1H–15N HSQC spectrum of apo-Hpa2-Ab in phosphate buffer, at pH 7.6 and 15°C. (B) Kanamycin titration on Hpa2-Ab using 1H–15N HSQC, an expanded region of the 1H–15N HSQC demonstrates different behavior of tryptophan peak in response to kanamycin binding. Kanamycin bindings are shown in overlaid spectra (red). (C) 1H–15N HSQC spectra of apo-Hpa2-Ab (blue) and 1H–15N HSQC spectra of Hpa2-Ab bound to different concentrations of acetyl-CoA (1 : 1 — red, 1 : 5 — magenta, 1 : 10 — green), square box represents tryptophan side chain peaks (black).

Figure 5.
1H–15N HSQC spectra of Hpa2-Ab in apo and complex form.

(A) 1H–15N HSQC spectrum of apo-Hpa2-Ab in phosphate buffer, at pH 7.6 and 15°C. (B) Kanamycin titration on Hpa2-Ab using 1H–15N HSQC, an expanded region of the 1H–15N HSQC demonstrates different behavior of tryptophan peak in response to kanamycin binding. Kanamycin bindings are shown in overlaid spectra (red). (C) 1H–15N HSQC spectra of apo-Hpa2-Ab (blue) and 1H–15N HSQC spectra of Hpa2-Ab bound to different concentrations of acetyl-CoA (1 : 1 — red, 1 : 5 — magenta, 1 : 10 — green), square box represents tryptophan side chain peaks (black).

Acetyl-CoA binding motif (RXXGXG) is conserved among all members of GNAT superfamily [28,29]. To get more insight into motif residues directly involved in acetyl-CoA binding, mutational study was performed. Change in free energy values for mutations of residues Arg76, Gly79 and Gly81 is high (∼2 kcal/mol) (Supplementary Table S2), indicating that mutations of Gly81, Gly79 and Arg76 residues decrease the stability of active site. Initial addition of acetyl-CoA to protein affects a few peaks (∼10); upon further addition, several different peaks appear in the spectrum. The spectral changes indicate appearance of additional species; more peaks appear and many peaks decrease in intensity.

Our efforts to obtain specific assignments were not successful. Even so, without explicit assignment, we measured transverse relaxation rates (line widths at half heights) of many peaks which were common in both free and bound proteins, these are listed in Supplementary Table S1. Interestingly, we noted that for many residues the line widths for free protein are twice those of complex. This is consistent with the fact that the free protein is a dimer, and the protein dissociates into monomer in the bound form. For many other residues, the line widths in the complex are more than half of those in the free protein, perhaps these residues might be involved in some sort of a chemical exchange. A detailed analysis of this process can only be done after complete assignment of the protein.

Chemical determination of the Hpa2-Ab acetylation mechanism

Aminoglycoside antibiotic acetylation capacity of Hpa2-Ab is monitored using a DTNB reagent. Acetylation reaction involves transfer of acetyl group from acetyl-CoA to antibiotic, this leads to conversion of acetyl-CoA to CoA-SH. Reaction between DTNB and CoA-SH produces light yellow colored 2-nitro-5-thiobenzoate ion, which was detected by colorimetric assays. Data were acquired as absorbance/min and absorbance data were subsequently analyzed in Excel. Product concentration was determined using a molar extinction coefficient of 14 150 M–1 cm–1 at wavelength 412 nm, Figure 6A depicts the concentration of acetylated substrates.

Hpa2-Ab acetylates aminoglycoside antibiotics.

Figure 6.
Hpa2-Ab acetylates aminoglycoside antibiotics.

In DNTB acetylation, acetyl-CoA donates its acetyl group in the acetylation reaction and (CoA-SH) thiols get exposed. Thiol reacts with DTNB and generates a light yellow color. Change in intensity of color was used as an index of acetylation. (A) Absorbance was recorded as a function of time and used to calculate the concentration of acetylated product. Product concentration is depicted in the graph. (B) Acetylation rate calculated (using equation, ln[P] = K.T + ln[P0]) for kanamycin and streptomycin is depicted here. Acetylation reaction executed in the absence of substrate was used as control and subtracted prior to calculation.

Figure 6.
Hpa2-Ab acetylates aminoglycoside antibiotics.

In DNTB acetylation, acetyl-CoA donates its acetyl group in the acetylation reaction and (CoA-SH) thiols get exposed. Thiol reacts with DTNB and generates a light yellow color. Change in intensity of color was used as an index of acetylation. (A) Absorbance was recorded as a function of time and used to calculate the concentration of acetylated product. Product concentration is depicted in the graph. (B) Acetylation rate calculated (using equation, ln[P] = K.T + ln[P0]) for kanamycin and streptomycin is depicted here. Acetylation reaction executed in the absence of substrate was used as control and subtracted prior to calculation.

We tried to calculate the steady-state kinetic parameters (kcat and Km) for Hpa2-Ab; however, observed strong substrate inhibition hinders the process [17,30]. Here, rate constant for acetylation at a single concentration of each substrate was measured. Acetylation follows first-order rate law and obtained rate constant of each substrate exhibits good correlation with the ITC-binding affinity (Figure 6B and Supplementary Figure S12A). Rate of acetylation and concentration of acetylated product is higher for the kanamycin followed by streptomycin. Concentrations of obtained acetylated gentamicin and amikacin are negligible. Differences in acetylation rate and product concentration indicate differential substrate preference of Hpa2-Ab.

Measurement of resistance conferred by Hpa2-Ab against kanamycin and streptomycin

For the assay, E. coli DH5α susceptible to all aminoglycosides antibiotics are taken as control. Agar dilution experiments performed on E. coli DH5/Hpa2-Ab transformants and on normal E. coli DH5α cells demonstrate that former cells are resistant to aminoglycoside kanamycin and streptomycin with MIC values, ∼256 and ∼128 mg/L, respectively. All aminoglycoside antibiotics were also tested, but no positive results were obtained for them. KB test is performed by using paper disks containing varying concentrations of antibiotics, paper disks were placed on the agar plate having E. coli DH5/Hpa2-Ab transformants cells. After ∼14 h of incubation, well-defined zone of clear inhibition (Figure 7A) and zone of resistance were observed (Supplementary Figure S12B). The diameter of zone of clear inhibition and zone of resistance showed good correlation with the taken concentrations of antibiotics. Higher resistance to kanamycin in comparison with streptomycin is depicted by larger clear zone of growth inhibition in Supplementary Figure S12B.

Hpa2-Ab depicts resistance against aminoglycoside antibiotics.

Figure 7.
Hpa2-Ab depicts resistance against aminoglycoside antibiotics.

(A) Kirby–Bauer disk diffusion susceptibility method. Paper disk containing various concentrations of kanamycin (50–200 µg/ml) and streptomycin (50–100 µg/ml) was kept in the middle of agar plate, and after inoculation plates were incubated for ∼14 h. The plate having zero concentration of both antibiotics was used as control. (B) Multiple sequence alignment of Hpa2 with aminoglycoside acetylating homologous proteins was done using DALI server, residues interacting at dimer interface are marked by the yellow star, acetyl-CoA interacting residues are marked by red star and purple star, representing substrate-binding site. (C) Aligned sequences obtained from DALI server were used for phylogenetic tree construction using NJ method. (D) Structures of Hpa2-Ab model (I) and aminoglycoside N-acetyltransferase aac(6′)-ly(II) were superimposed according to the common fold of the catalytic domain but shown side-by-side for easier comparison.

Figure 7.
Hpa2-Ab depicts resistance against aminoglycoside antibiotics.

(A) Kirby–Bauer disk diffusion susceptibility method. Paper disk containing various concentrations of kanamycin (50–200 µg/ml) and streptomycin (50–100 µg/ml) was kept in the middle of agar plate, and after inoculation plates were incubated for ∼14 h. The plate having zero concentration of both antibiotics was used as control. (B) Multiple sequence alignment of Hpa2 with aminoglycoside acetylating homologous proteins was done using DALI server, residues interacting at dimer interface are marked by the yellow star, acetyl-CoA interacting residues are marked by red star and purple star, representing substrate-binding site. (C) Aligned sequences obtained from DALI server were used for phylogenetic tree construction using NJ method. (D) Structures of Hpa2-Ab model (I) and aminoglycoside N-acetyltransferase aac(6′)-ly(II) were superimposed according to the common fold of the catalytic domain but shown side-by-side for easier comparison.

Relationship of Hpa2-Ab to antibiotic acetylating GNAT family proteins

Successful acquisition of kanamycin and streptomycin acetylating capacity in Hpa2-Ab, which is in turn responsible for resistance against these antibiotics, is explored using sequence/structure similarity and phylogenetic analysis. It is widely accepted that phylogenetic tree analysis facilitates the functional assignment and establishes the correct evolutionary relation between the proteins. For phylogenetic tree construction, blast search was performed against the PDB data bank. Homologous proteins exhibiting polyamine and antibiotic acetylating property were chosen for MSA using DALI server and phylogenetic tree was constructed using NJ method [3133]. Results of multiple sequence alignment are depicted in Figure 7B. Residues involved in acetyl-CoA and substrate binding (marked by red and yellow stars, respectively) exhibit significant conservation. Structural comparison of aac(6′)ly and Hpa2-Ab using DALI server and pymol software (Figure 7C) demonstrates similar topology. Antibiotic binding site is lined by a loop preceding β6 sheet, it adds to the conformational flexibility of Hpa2-Ab and helps in accommodating a variety of molecules for catalysis. Observed structural difference is depicted as blue color. Residues involved in substrate binding also participate in the dimer formation (marked by purple star), this might be responsible for substrate-induced oligomeric state conversion in the Hpa2-Ab. NJ method constructs the tree according to evolutionary distance which reflects the expected mean number of changes per site (Supplementary Figure S13). Phylogenetic tree analysis (Figure 7D) demonstrates that Hpa2-Ab exhibits close phylogeny with aminoglycoside acetyltransferase enzymes. Hpa2-Ab and Hpa2-Sc are functional homologs, but they have distant phylogenetic relationship. Hpa2-Sc acetylates the nuclear molecules such as histone proteins, HMG proteins and polyamines. However, the absence of nucleus in prokaryotes provides opportunity to evolve the acetylation capability for other cellular molecules. Close phylogeny of Hpa2-Ab with polyamine and antibiotic acetylating acetyltransferase attests to the theory that substrate ambiguity and close phylogeny are keys to evolvability [34,35].

Discussion

Awareness of virulence factors responsible for resistivity in A. baumannii is a keystone for producing new antibiotics. Research interest about A. baumannii is mainly due to two reasons: firstly, due to its enormous capacity to acquire antibiotic resistance and secondly, it is relatively an unexplored ESKAPE pathogen [36,37]. Despite the significant role of acetyltransferases in resistance, very little is known about their structure and function in the case of MDR A. baumannii. We successfully expressed Hpa2-Ab, an acetyltransferase and purified up to 98% homogeneity by a two-step purification protocol. Biophysical studies indicated that Hpa2-Ab is intrinsically oligomeric in nature, it exists primarily as a dimer under physiological conditions and dimer is independent of concentration. Oligomeric state of Hpa2-Ab is also explored in the presence of acetyl-CoA, and our results suggest that unlike its functional homolog, Hpa2-Ab dissociates to monomeric form. This difference in the oligomeric states of Hpa2-Ab and Hpa2-Sc can be explained as follows: dimeric structure of yeast Hpa2 is formed by domain swapping and interface is extensively interdigitated and ∼33% of residues of each monomer contribute in dimerization due to which conversion in to monomeric form becomes unlikely. However, in Hpa2-Ab the dimer interface involves <15% residues of each monomer. Also each monomer contains individual active site, so we believe that oligomerization in the case of yeast is related to the increase in the number of binding sites to hasten the acetylation process.

NMR data indicate that binding of kanamycin induces structural changes in the protein. ITC experiments depict 1 : 1 stoichiometry of the interactions. Favorable thermodynamic parameters establish specific binding interactions between acetyl-CoA/antibiotics and protein. Effect of antibiotic binding was established using E. coli DH5α strains in which the Hpa2 gene is expressed. E. coli DH5/Hpa2-Ab transformant cells shows tolerance for high concentrations of kanamycin and streptomycin. This confirms that Hpa2 gene expression is responsible for the antibiotic resistance. To determine the mechanism by which Hpa2 gene product confers resistance, we tested acetyltransferase activity of Hpa2-Ab for kanamycin and streptomycin. Positive acetylation assay results attested to evolve antibiotic acetylation capacity of Hpa2-Ab

Functional characterization of unexplored and unconventional drug targets helps in understanding how resistance develops at molecular level. Functional characterization of unconventional target Hpa2-Ab suggested its contribution in drug resistance. NJ tree analysis establishes the close phylogeny between Hpa2-Ab and aminoglycoside antibiotic acetylating enzyme aac(6′)ly. [1,38]. Data from our studies reveal that antibiotic resistance shown by Hpa2-Ab is an outcome of its substrate ambiguous nature and close phylogenetic relation with aminoglycoside acetyltransferase enzymes. Exploration of all pathways or enzymes associated to a particular resistance mechanism is crucial for understanding and slowing down the resistance development. This, in turn, is crucial for effective control of infections caused by multiple drug-resistant A. baumannii.

Abbreviations

     
  • DTNB

    5,5′-dithiobis-(2-nitrobenzoic acid)

  •  
  • ESKAPE

    Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species

  •  
  • GNATs

    Gcn5-related N-acetyl transferases

  •  
  • Hpa2-Ab

    histone and other protein acetyltransferase-Acinetobacter baumannii

  •  
  • ITC

    isothermal titration calorimetry

  •  
  • MDR

    multiple drug resistance

Author Contribution

Experiments were performed by Dr Jyoti Singh Tomar, data analysis was done by Dr Tomar and Prof Hosur, and the manuscript was prepared by Dr Tomar with guidance from Prof Hosur and Dr Peddinti.

Funding

Funding for this research was provided by the Department of Atomic Energy, Government of India and Tata Institute of Fundamental Research, Mumbai.

Acknowledgements

Jyoti Singh Tomar thanks MHRD, Department of Atomic Energy for Research Fellowships and National Facility of NMR at TIFR Mumbai.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Davies
,
J.
and
Davies
,
D.
(
2010
)
Origins and evolution of antibiotic resistance
.
Microbiol. Mol. Biol. Rev.
74
,
417
433
2
Crofts
,
T.S.
,
Gasparrini
,
A.J.
and
Dantas
,
G.
(
2017
)
Next-generation approaches to understand and combat the antibiotic resistome
.
Nat. Rev. Microbiol.
15
,
422
434
3
Hughes
,
D.
and
Anderson
,
D.I.
(
2015
)
Evolutionary consequences of drug resistance: shared principles across diverse targets and organisms
.
Nat. Rev. Genet.
16
,
459
471
4
Sommer
,
M.O.A.
,
Munck
,
C.
,
Toft-Kehler
,
R.V.
and
Andersson
,
D.I.
(
2017
)
Prediction of antibiotic resistance: time for a new paradigm?
Nat. Rev. Microbiol.
15
,
689
696
5
Spellberg
,
B.
and
Bonomo
,
R.A.
(
2015
)
Combination therapy for extreme drug resistant (XDR) Acinetobacter baumannii: ready for Prime-Time?
Crit. Care Med.
43
,
1332
1334
6
Fishbain
,
J.
and
Peleg
,
A.Y.
(
2010
)
Treatment of Acinetobacter infections
.
Clin. Infect. Dis.
51
,
79
84
7
Shaw
,
K.J.
,
Rather
,
P.N.
,
Hare
,
R.S.
and
Miller
,
G.H.
(
1993
)
Molecular genetics of aminoglycoside resistance genes and familial relationships of the aminoglycoside-modifying enzymes
.
Microbiol. Rev.
57
,
138
163
PMID:
[PubMed]
8
Ghada
,
E.-S.M.
(
2017
)
Aminoglycosides resistance among Acinetobacter baumannii complex isolated from hospital acquired blood stream infections
.
Int. J. Curr. Microbiol Appl. Sci.
6
,
1103
1112
9
Tomar
,
J.S.
and
Peddinti
,
R.K.
(
2017
)
A. baumannii histone acetyl-transferases Hpa2: optimization of homology modeling, analysis of protein–protein interaction and virtual screening
.
J. Biomol. Struct. Dyn.
35
,
1115
1126
10
Salah
,
A.L.
,
Tikhomirova
,
A.
and
Roujeinikova
,
A.
(
2016
)
Structure and functional diversity of GCN5-related N-acetyltransferases (GNAT)
.
Int. J. Mol. Sci.
17
,
1018
1063
11
Sampath
,
V.
,
Liu
,
S.
,
Tafrov
,
S.
,
Srinivasan
,
M.
,
Rieger
,
R.
,
Chen
,
E.I.
et al.  (
2013
)
Acetyltransferases from S. cerevisiae Hpa3-two small closely related biochemical characterization of Hpa2 and enzymology
.
J. Biol. Chem.
288
,
21506
21513
12
Angus-Hill
,
M.L.
,
Dutnall
,
R.N.
,
Tafrov
,
S.T.
,
Sternglanz
,
S.
and
Ramakishnan
,
V.
(
1999
)
Crystal structure of the histone acetyltransferase Hpa2: a tetrameric member of the Gcn5-related N-acetyltransferase superfamily
.
J. Mol. Biol.
294
,
1311
1325
13
Thao
,
S.
and
Escalante-Semerena
,
J.C.
(
2011
)
Biochemical and thermodynamic analyses of Salmonella enterica Pat, a multidomain, multimeric N-Lysine acetyltransferase involved in carbon and energy metabolism
.
mBio
2
,
1
11
14
Burk
,
D.L.
,
Ghuman
,
N.
,
Wybenga-Groot
,
L.E.
and
Berghuis
,
A.M.
(
2003
)
X-ray structure of AAC(6_)-Ii antibiotic resistance enzyme at 1.8 Å resolution; examination of oligomeric arrangements in GNAT superfamily members
.
Protein Sci.
12
,
426
437
15
Tomar
,
J.S.
,
Narwal
,
M.
,
Kumar
,
P.
and
Peddinti
,
R.K.
(
2016
)
Characterization of substrate binding and enzymatic removal of a 3-methyladenine lesion from genomic DNA with TAG of MDR A. baumannii
.
Mol. BioSyst.
12
,
3259
3265
16
Foyn
,
H.
,
Thompson
,
P.R.
and
Arnesen
,
T.
(
2017
)
DTNB-Based quantification of in vitro enzymatic N-terminal acetyltransferase activity
.
Methods Mol. Biol.
1574
,
9
15
17
Robicsek
,
A.
,
Strahilevitz
,
J.
,
Jacoby
,
G.A.
,
Macielag
,
M.
,
Abbanat
,
D.
,
Park
,
C.H.
et al.  (
2006
)
Fluoroquinolone-modifying enzyme: a new adaptation of a common aminoglycoside acetyltransferase
.
Nat. Med.
12
,
83
88
18
Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically, Approved Standard 10th Edition
.
CLSI document M07-A10
.
Clinical and laboratory standards institute
,
Wayne, PA
,
2015
19
Bauer
,
A.W.
,
Kirby
,
W.M.
,
Sherris
,
J.C.
and
Turck
,
M.
(
1966
)
Antibiotic susceptibility testing by a standardized single disk method
.
Am. J. Clin. Pathol.
45
,
493
496
20
Jorgensen
,
J.
and
Turnidge
,
J.
(
2015
)
Susceptibility Test Methods: Dilution and Disk Diffusion Methods: Manual of Clinical Microbiology
,
11th edn
, pp.
1253
1273
,
ASM Press
,
Washington
21
Wittig
,
I.
,
Braun
,
H.-P.
and
Schägger
,
H.
(
2006
)
Blue native PAGE
.
Nat. Protoc.
1
,
418
428
22
Fadouloglou
,
V.E.
,
Kokkinidis
,
M.
and
Glykos
,
N.M.
(
2008
)
Determination of protein oligomerization state: two approaches based on glutaraldehyde crosslinking
.
Anal. Biochem.
373
,
404
406
23
Johnson
, Jr,
W.C.
(
1988
)
Secondary structure of proteins through circular dichroism spectroscopy
.
Annu. Rev. Biophys. Biophys. Chem.
17
,
145
166
24
Claes
,
P.D.M.
, and
Vardy
,
P.
(
1992
) An on-line dynamic light scattering instrument for macromolecular characterisation. In
Laser Light Scattering in Biochemistry
(
Harding
,
S. E.
,
Sattell
,
D. B.e.
and
Bloomfield
,
V. A.
, eds), pp.
66
76
,
The Royal Society of Chemistry
,
Cambridge
25
Adrianne
,
L.
,
Norris
,
C.Ö
and
Engin
,
H.S.
(
2000
)
Thermodynamics and kinetics of association of antibiotics with the aminoglycoside acetyltransferase (3)-IIIb, a resistance-causing enzyme
.
Biochemistry
49
,
4027
4035
26
Wright
,
E.
and
Serpersu
,
E.H.
(
2006
)
Molecular determinants of affinity for aminoglycoside binding to the aminoglycoside nucleotidyl-transferase(200)-Ia
.
Biochemistry
45
,
10243
10250
27
Hegde
,
S.S.
,
Dam
,
T.K.
,
Brewer
,
C.F.
and
Blanchard
,
J.S.
(
2001
)
Thermodynamics of aminoglycoside and acyl-coenzyme a binding to the Salmonella enterica AAC(6′)-Iy aminoglycoside N-acetyltransferase
.
Biochemistry
41
,
7519
7527
28
Neuwald
,
A.F.
and
Landsman
,
D.
(
1997
)
GCN5-related histone N-acetyltransferases belong to a diverse superfamily that includes the yeast SPT10 protein
.
Trends Biochem. Sci.
22
,
154
155
29
Xie
,
L.
,
Zeng
,
J.
,
Luo
,
H.
,
Pan
,
W.
and
Xie
,
J.
(
2014
)
The roles of bacterial GCN5-related N-acetyltransferases
.
Crit. Rev. Eukaryot. Gene Expr.
24
,
77
87
30
Smith
,
C.A.
,
Bhattacharya
,
M.
,
Toth
,
M.
,
Stewart
,
N.K.
and
Vakulenko
,
S.B.
(
2017
)
Aminoglycoside resistance profile and structural architecture of the aminoglycoside acetyltransferase AAC(6′)-Im
.
Microb. Cell
4
,
402
410
31
Eisen
,
J.A.
and
Martin
,
W.
(
2002
)
Phylogenetic analysis and gene functional predictions: phylogenomics in action
.
Theor. Popul. Biol.
61
,
481
487
32
Holm
,
L.
and
Rosenstrom
,
P.
(
2010
)
Dali server: conservation mapping in 3D
.
Nucleic Acids Res.
38
,
W545
W549
33
Nemec
,
A.
,
Dolzani
,
L.
,
Brisse
,
S.
,
van den Broek
,
P.
and
Dijkshoorn
,
L.
(
2004
)
Diversity of aminoglycoside-resistance genes and their association with class1 integrons among strains of pan-European A. baumannii clone
.
J. Med. Microbiol.
53
,
1233
1240
34
Mak
,
J.K.
,
Kim
,
M.J.
,
Pham
,
J.
,
Tapsall
,
J.
and
White
,
P.A.
(
2009
)
Antibiotic resistance determinants in nosocomial strains of multidrug-resistant Acinetobacter baumannii
.
J. Antimicrob. Chemother.
63
,
47
54
35
Barbe
,
V.D.
,
Vallenet
,
N.
,
Fonknechten
,
A.
,
Kreimeyer
,
S.
,
Oztas
,
L.
,
Labarre
,
S.
et al.  (
2004
)
Unique features revealed by the genome sequence of Acinetobacter sp. ADP1, a versatile and naturally transformation competent bacterium
.
Nucleic Acids Res.
32
,
5766
5779
36
Weber
,
B.S.
,
Ly
,
P.M.
,
Irwin
,
J.N.
,
Pukatzki
,
S.
and
Feldman
,
M.F.
(
2015
)
A multidrug resistance plasmid contains the molecular switch for type VI secretion in Acinetobacter baumannii
.
Proc. Natl Acad. Sci. U.S.A.
30
,
9442
9447
37
Aminov
,
R.I.
(
2009
)
The role of antibiotics and antibiotic resistance in nature
.
Environ. Microbiol.
11
,
2970
2988
38
Wang
,
X.
,
Minasov
,
G.
and
Shoichet
,
B.K.
(
2002
)
Evolution of an antibiotic resistance enzyme constrained by stability and activity trade-offs
.
J. Mol. Biol.
320
,
85
95