DOI QR코드

DOI QR Code

Rv3168 Phosphotransferase Activity Mediates Kanamycin Resistance in Mycobacterium tuberculosis

  • Ahn, Jae-Woo (Structural and Molecular Biology Laboratory, School of Life Sciences and Biotechnology, Kyungpook National University) ;
  • Kim, Kyung-Jin (Structural and Molecular Biology Laboratory, School of Life Sciences and Biotechnology, Kyungpook National University)
  • Received : 2013.06.20
  • Accepted : 2013.07.29
  • Published : 2013.11.28

Abstract

Tuberculosis is a worldwide epidemic disease caused by Mycobacterium tuberculosis, with an estimated one-third of the human population currently affected. Treatment of this disease with aminoglycoside antibiotics has become less effective owing to antibiotic resistance. Recent determination of the crystal structure of the M. tuberculosis Rv3168 protein suggests a structure similar to that of Enterococcus faecalis APH(3')-IIIa, and that this protein may be an aminoglycoside phosphotransferase. To determine whether Rv3168 confers antibiotic resistance against kanamycin, we performed dose-response antibiotic resistance experiments using kanamycin. Expression of the Rv3168 protein in Escherichia coli conferred antibiotic resistance against $100{\mu}M$ kanamycin, a concentration that effected cell growth arrest in the parental E. coli strain and an E. coli strain expressing the $Rv3168^{D249A}$ mutant, in which the catalytic Asp249 residue was mutated to alanine. Furthermore, we detected phosphotransferase activity of Rv3168 against kanamycin as a substrate. Moreover, docking simulation of kanamycin into the Rv3168 structure suggests that kanamycin fits well into the substrate binding pocket of the protein, and that the phosphorylation-hydroxyl-group of kanamycin was located at a position similar to that in E. faecalis APH(3')-IIIa. On the basis of these results, we suggest that the Rv3168 mediates kanamycin resistance in M. tuberculosis, likely through phosphotransferase targeting of kanamycin.

Keywords

Introduction

Tuberculosis is a worldwide epidemic disease, and it has been estimated that up to one-third of the human population harbors Mycobacterium tuberculosis, the causative agent of the disease [13]. Although a combination of several antibiotics is normally used for the treatment of tuberculosis, complete cure of tuberculosis is made difficult by the existence of mutant strains and the acquisition of antibiotic resistance. Among the antibiotics used for treating tuberculosis, aminoglycosides are a class of broad-spectrum antimicrobial compounds that contain an amino-modified sugar [17].

Aminoglycosides have an affinity for decoding the aminoacyl site of 16S rRNA in the bacterial 30S ribosomal subunit, thereby effecting the dissociation of aminoacyltRNA from the 30S subunit, resulting in protein miscoding and loss of bacterial cell wall integrity [6,9]. It has been shown that the major mechanism of bacterial resistance to aminoglycosides is based on the chemical modification of the antibiotic by bacterial enzymes [16]. Enzymes that modify aminoglycosides form a large and diverse group that is classified into three general classes, depending on the chemical group of the antibiotic that is modified; namely, ATP-dependent phosphotransferases (APH), ATPdependent adenylyltransferases (ANT), and acetyl CoAdependent acetyltransferases (AAC). Studies have shown several structures of the APH enzymes, including APH (3’)-IIIa [1,3,5], APH (3’)-IIa [10], APH (2’’)-IIa [11], APH (2’’)-IVa [14], and APH (9’)-Ia [4].

Recent studies on the genome sequence of M. tuberculosis have suggested candidate genes responsible for the acquisition of aminoglycoside resistance [12,16], and we have previously reported the crystal structure of Rv3168, a putative aminoglycoside phosphotransferase [7]. Although lacking significant amino acid sequence similarity, the overall structure of Rv3168 was similar to that of E. faecalis APH(3’)-IIIa, which is a characterized aminoglycoside phosphotransferase. Moreover, the structure of the ATPbound form of Rv3168 implied an ATP-binding mode similar to that of E. faecalis APH(3’)-IIIa. Together with the existence of a large negatively charged substrate-binding pocket located near the ATP-binding pocket of Rv3168, these data collectively suggest that Rv3168 is a candidate phosphotransferase that confers aminoglycoside antibiotic resistance in M. tuberculosis [7].

Here, we report that an E. coli strain in which the Rv3168 expression is induced exhibits resistance to a concentration of kanamycin that is lethal to strains in which the Rv3168 is not induced. Moreover, Rv3168 protein has phosphotransferase activity against kanamycin as a substrate. Finally, docking simulation of kanamycin into the Rv3168 structure suggests a possible binding of the kanamycin substrate and the enzyme.

 

Materials and Methods

Antibiotic Resistance Test

The Rv3168 coding gene (Gene ID: 888778) was cloned into pPROEX HTa (Life Technology), and the resulting plasmid pPROEX HTa:Rv3168 was transformed into the E. coli BL21(DE3) strain. The strain was grown overnight and the culture, to the final concentration of 1%, was inoculated into 1 L of LB broth liquid medium containing ampicillin at 37℃. When the OD600 reached at 0.4, the Rv3186 protein was induced by adding 1 mM IPTG. At 30 min after the induction, various concentrations of kanamycin (0, 50, 100, and 200 μM) were added to the culture, and the cell growth was measured spectrophotometrically at A600. For control experiments, cells without IPTG induction were grown, and their growth rates were measured. The viable cell growth was measured by plating the aliquots of the cell culture as well. The cell growth of the E. coli strains transformed with the pPROEX HTa empty vector and the pPROEX HTa:Rv3168D249A were measured with the same method as described above, under 100 μM kanamycin.

Preparation of Rv3168 Proteins

For the preparation of the Rv3168D249A mutant protein, a sitedirected mutagenesis method was applied using the pPROEX HTa:Rv3168 plasmid as a template. The primers 5’-GTTGCTGTGGGGGGCCGCGCGGGTGGGCA-3’ and 5’-TGCCCACCCGCGCGGCCCCCCACAGCAAC-3’ were used for the polymerase chain reaction. The recombinant wild-type Rv3168 and Rv3168D249A mutant proteins were prepared by following a previously reported procedure [8]. Briefly, the IPTG-induced cell culture was harvested by centrifugation at 5,000 ×g at 277 K. The cell pellet was resuspended in ice-cold buffer A (50 mM Tris-HCl, pH 8.0, 5 mM β-mercaptoethanol) and disrupted by ultrasonication. The cell debris was removed by centrifugation at 11,000 ×g for 1 h, and lysate was bound to Ni-NTA agarose (Qiagen). After washing with buffer A containing 10 mM imidazole, the bound proteins were eluted with 300 mM imidazole in buffer A. The 6× His-tag was released from the Rv3168 proteins by incubating with rTEV protease (Gibco). A trace amount of contamination was removed by applying HiTrap Q ion exchange and Superdex75 size exclusion chromatography.

Phosphotransferase Activity Assay

For the phosphotransferase activity assay, the reaction mixture, containing 10 μl of 40 mM Tris-HCl, pH 8.0, 2.5 mM MgCl2, 5 mM ATP, 25 mM kanamycin, and 50 μM Rv3168 protein, was incubated overnight at room temperature, and 1 μl aliquot of the reaction mixture was spotted onto a cellulose-F TLC plate (Merck). For the controls, the ATP and ADP molecules and the reaction mixture without the protein were spotted as well. Ascending TLC was performed with a buffer containing saturated ammonium sulfate, 3M sodium acetate, and isopropanol (80/6/2) in the closed chamber for 3 h. The hydrolysis of ATP was monitored by visualizing nucleotides under UV light.

Docking Simulation

The docking simulations of a kanamycin molecule to the Rv3168 structure was performed using the Autodock Vina program [15]. As a template, the crystal structure of Rv3168 in complex with Mg2+ and ATP was used (PDB code 3ATT). The water, acetate, calcium, and glycerol molecules were removed from the crystal structure of Rv3168 and hydrogen atoms were added in accordance with only polar atoms. The structure of kanamycin was obtained from PDBeChem [2] and was prepared as pdbqt files. For the precise docking simulation, the substrate binding site of the Rv3168 structure was defined using Autodocking tools software. The simulation results were checked using the PyMOL software.

 

Results and Dicussion

Kanamycin Resistance of Rv3168-Expressing E. coli

We previously reported the crystal structure of Rv3168 protein, a putative aminoglycoside phosphotransferase in M. tuberculosis. Based on the structural comparison of M. tuberculosis Rv3168 with Enterococcus faecalis APH(3’)-IIIa, a characterized aminoglycoside phosphotransferase that confers aminoglycoside resistance in this strain, we speculated that Rv3168 was a candidate phosphotransferase family enzyme conferring aminoglycoside resistance to M. tuberculosis. We first determined whether Rv3168 confers the antibiotic resistance effect when expressed in E. coli. A pPROEX HTa vector harboring an inducible Rv3168 gene was transformed into an E. coli strain, which was then treated using various concentrations of kanamycin. The Rv3168-expressing E. coli exhibited resistance to 100 μM of kanamycin (Figs. 1A- 1D), a concentration which effected growth arrest in the E. coli strains in which Rv3168 was not expressed and in strains harboring an empty pPROEX HTa vector (Figs. 1A- 1D). The growth rate of the Rv3168-expressing E. coli was dramatically decreased at kanamycin concentrations in excess of 200 μM (Fig. 1E).

Fig. 1.Kanamycin resistance of the E. coli strain expressing the Rv3168 protein. For the measurement of the antibiotic resistance activity of Rv3168 against kanamycin, the Rv3168 coding gene was cloned into the pPROEX HTa vector, and the resulting plasmid was transformed into E. coli strain BL21(DE3). The cell growth under various concentrations of kanamycin was measured spectrophotometrically at A600. Each experiment was performed with and without IPTG induction, and presented with open and closed circles, respectively. pEX is an abbreviation of a pPROEX HTa vector. (A) The cell growth of the E. coli strain harboring pEX:Rv3168 without kanamycin. (B) The cell growth of the E. coli strain harboring an empty pEX vector under 100 μM kanamycin. (C-E) The cell growth of the E. coli strain harboring pEX:Rv3168 under 50, 100, and 200 μM kanamycin, respectively. (F) The cell growth of the E. coli strain harboring pEX:Rv3168D249A under 100 μM kanamycin.

Fig. 2.Viable cell measurements of the kanamycin resistance effect of Rv3168. The aliquots of each cell culture presented in Fig. 1 were collected after 90 min of the induction, and spread on the LBAMP plates. The plates were incubated at 37℃ for overnight. (A) The cell growth of the E. coli strain harboring pEX:Rv3168 without kanamycin. (B) The cell growth of the E. coli strain harboring an empty pEX vector under 100 μM kanamycin. (C-E) The cell growth of the E. coli strain harboring pEX:Rv3168 under 50, 100, and 200 μM kanamycin, respectively. (F) The cell growth of the E. coli strain harboring pEX:Rv3168D249A under 100 μM kanamycin.

Structural comparison of M. tuberculosis Rv3168 with E. faecalis APH(3’)-IIIa suggested that both enzyme catalysis is mediated by aspartic acid residues in M. tuberculosis Rv3168 (Asp249) and E. faecalis APH(3’)-IIIa (Asp190). We next performed the kanamycin resistance test by using an E. coli strain harboring an Rv3168D249A mutant, in which the catalytic aspartic acid residue was mutated to alanine. Loss of Asp249 resulted in a cell growth rate, in 100 μM of kanamycin, comparable to an E. coli strain harboring an empty vector, confirming that Rv3168 mediates kanamycin resistance (Fig. 1F).

To confirm the kanamycin resistance activity of Rv3168, we collected aliquots of each of the above cultures, and spread them on LB solid medium with ampicillin. The viable cell growth showed marked results with the same tendency as those observed using spectrophotometric cell growth measurement (Fig. 2). The E. coli strain expressing Rv3168 showed high cell growth in the presence of 100 μM kanamycin, whereas no significant cell growth was observed for strains in which Rv3168 expression was not induced (Figs. 2A-2D). Moreover, the viable cell count of the E. coli strain expressing Rv3168 was markedly decreased by the addition of 200 μM kanamycin (Fig. 2E). As anticipated, the E. coli strain expressing the Rv3168D249A mutant had low viability in 100 μM kanamycin (Fig. 2F). Based on these spectrophotometric and viable cell growth measurements, we concluded that Rv3168 confers antibiotic resistance at low concentrations of kanamycin.

Fig. 3.Phosphotransferase activity of Rv3168. As standard nucleotides, the ATP and ADP molecules are spotted, and labeled on the left side of the figure. The contents of each reaction mixture are shown at the bottom of the figure with“+” and “-” for representing addition and no addition, respectively. W/T and D249A indicate wild-type and D249A mutant proteins of Rv3168, respectively. Time indicates reaction time, and O/N is an abbreviation of overnight.

Phosphotransferase Activity of Rv3168

To determine whether the kanamycin resistance was mediated by possible phosphotransferase activity of the Rv3168 protein, we performed a phosphotransferase activity assay by using the recombinant Rv3168 protein. When the reaction mixture containing 5 mM ATP, 50 mM kanamycin, and 50 μM of Rv3168 was incubated for 20 min, and spotted on a TLC plate, no significant phosphotransferase activity was detected (Fig. 3). Increasing the time to overnight, however, resulted in detectable ATP hydrolysis, indicating that the recombinant Rv3168 protein has very low phosphotransferase activity (Fig. 3). Moreover, ATP hydrolysis was undetectable in a reaction mixture containing the Rv3168D249A mutant protein instead of the wild-type Rv3168 protein (Fig. 3). These results indicate that the Rv3168 protein has kanamycin phosphotransferase activity, which may confer kanamycin resistance to M. tuberculosis. The low kanamycin phosphotransferase activity of the Rv3168 protein is consistent with the results showing that the Rv3168 protein conferred antibiotic resistance only in low concentrations of kanamycin. The low phosphotransferase activity and mild antibiotic resistance effect of Rv3168 in E. coli are conceivably attributable to the fact that the protein originates from the wild-type M. tuberculosis H37Rv strain, which does not show a strong antibiotic resistance to aminoglycosides. In general, mutations in the target proteins of antibiotics confer antibiotic resistance to the organism, as previously shown in the wild-type M. tuberculosis strain H37Rv, which acquired aminoglycoside resistance through mutations in the 30S ribosomal subunit. We speculated that mutations in the Rv3168 protein might increase its aminoglycoside phosphotransferase activity, thereby conferring greater aminoglycoside resistance to M. tuberculosis. With this in mind, the examination of Rv3168-coding sequences in aminoglycoside-resistant M. tuberculosis strains may shed light on whether specific mutations, if any, play a role in conferring resistance.

Kanamycin Binding Simulation

The previously reported crystal structure of Rv3168 showed that the highly charged large substrate binding pocket can accommodate a charged aminoglycoside substrate, and that Gly248, Asp249, Asn254, and Asp267 form an xxDxxxxNx kinase motif that is located in the tunnel connecting the ATP and substrate-binding pockets. We next modeled the complex structure of Rv3168 bound to kanamycin to identify the substrate binding mode of the protein. Unfortunately, we were unable to obtain the complex structure owing to the blocking of the substratebinding entrance by a neighboring molecule of the P212121 crystal packing, thereby preventing the entry of a kanamycin molecule. We then performed autodocking simulation of a kanamycin molecule by using the ATP-bound form of the Rv3168 structure (PDB code 3ATT). In the simulated Rv3168- kanamycin complex structure, a kanamycin molecule was observed to fit well in the substrate-binding pocket (Fig. 4A). The charged kanamycin substrate appeared to be stabilized by the charged residues, including Asp50, Thr52, Glu57, Asp249, Arg251, Glu269, Thr344, Arg347, Arg348, and Glu353. The overall orientation of bound kanamycin in Rv3168 was distinct from that in E. faecalis APH(3’)-IIIa, whereas the orientation and the binding mode of ATP were almost identical (Fig. 4A). Interestingly, when the simulated structure of the Rv3168-kanamycin complex was superimposed onto that of the kanamycin-bound E. faecalis APH(3’)-IIIa, the phosphorylation-hydroxyl groups of kanamycin of the two proteins were observed to be located at the same positions (Fig. 4B). Moreover, the phosphorylation-hydroxyl group of kanamycin in Rv3168 is proximal to the catalytic Asp249 residue ~3.2 Å, which is comparable to that observed in the kanamycin-bound form of E. faecalis APH(3’)-IIIa (Fig. 4B). These observations suggest that kanamycin is a natural substrate for Rv3168, which functions as a kanamycin phosphotransferase with a reaction mechanism similar to that of E. faecalis APH(3’)-IIIa.

Fig. 4.Docking simulation of a kanamycin molecule to the Rv3168 structure. Docking simulation of a kanamycin molecule into the ATP-bound form of the Rv3168 structure was performed using the Autodock Vina program. (A) Comparison of kanamycin binding to Rv3168 (top) and E. faecalis APH(3’)-IIIa (bottom). The proteins are presented as electrostatic potential models. The bound kanamycin and ATP molecules are shown as stick models with cyan and yellow colors, respectively, and labeled. (B) Positions of bound kanamycin molecules. The simulated kanamycin-bound form of Rv3168 was superposed with the kanamycin-bound form of E. faecalis APH(3’)-IIIa. The catalytic Asp residues, the bound magnesium atoms, and kanamycin molecules of M. tuberculosis Rv3168 are shown with cyan color and indicated as “Rv”. Those of E. faecalis APH(3’)-IIIa are shown with green color and indicated as “Ef”. The phosphorylationhydroxyl atoms of Rv3168 and E. faecalis APH(3’)-IIIa are indicated by dotted circles with cyan and green colors, respectively. The distances between the catalytic Asp residue and the phosphorylation-hydroxyl atom of Rv3168 and E. faecalis APH(3’)-IIIa are shown with cyan and green colors, respectively.

We also performed the antibiotic resistance test of Rv3168 against other aminoglycoside antibiotics such as neomycin and streptomycin, resulting in no significant resistance effect, and the protein exhibited no detectable phosphotransferase activity with these antibiotics as well (data not shown). Taken together, we suggest that Rv3168 does not have broad aminoglycoside substrate specificity, but rather the resistance is limited to kanamycin. In future studies, we recommend the detailed examination of this protein.

References

  1. Burk DL, Hon WC, Leung AK, Berghuis AM. 2001. Structural analyses of nucleotide binding to an aminoglycoside phosphotransferase. Biochemistry 40: 8756-8764. https://doi.org/10.1021/bi010504p
  2. de Matos P, Alcantara R, Dekker A, Ennis M, Hastings J, Haug K, et al. 2010. Chemical entities of biological interest: an update. Nucleic Acids Res. 38: D249-D254. https://doi.org/10.1093/nar/gkp886
  3. Fong DH, Berghuis AM. 2002. Substrate promiscuity of an aminoglycoside antibiotic resistance enzyme via target mimicry. EMBO J. 21: 2323-2331. https://doi.org/10.1093/emboj/21.10.2323
  4. Fong DH, Lemke CT, Hwang J, Xiong B, Berghuis AM. 2010. Structure of the antibiotic resistance factor spectinomycin phosphotransferase from Legionella pneumophila. J. Biol. Chem. 285: 9545-9555. https://doi.org/10.1074/jbc.M109.038364
  5. Hon WC, McKay GA, Thompson PR, Sweet RM, Yang DS, Wright GD, et al. 1997. Structure of an enzyme required for aminoglycoside antibiotic resistance reveals homology to eukaryotic protein kinases. Cell 89: 887-895. https://doi.org/10.1016/S0092-8674(00)80274-3
  6. Karimi R, Ehrenberg M. 1994. Dissociation rate of cognate peptidyl-tRNA from the A-site of hyper-accurate and errorprone ribosomes. Eur. J. Biochem. 226: 355-360. https://doi.org/10.1111/j.1432-1033.1994.tb20059.x
  7. Kim S, Nguyen CM, Kim EJ, Kim KJ. 2011. Crystal structure of Mycobacterium tuberculosis Rv3168: a putative aminoglycoside antibiotics resistance enzyme. Proteins 79: 2983-2987. https://doi.org/10.1002/prot.23119
  8. Kim S, Nguyen CM, Yeo SJ, Ahn JW, Kim EJ, Kim KJ. 2011. Cloning, expression, purification, crystallization and X-ray crystallographic analysis of Rv3168 from Mycobacterium tuberculosis H37Rv. Acta Crystallogr. Sect. F Struct. Biol. Cryst. Commun. 67: 627-629. https://doi.org/10.1107/S1744309111010487
  9. Kotra LP, Haddad J, Mobashery S. 2000. Aminoglycosides: perspectives on mechanisms of action and resistance and strategies to counter resistance. Antimicrob. Agents Chemother. 44: 3249-3256. https://doi.org/10.1128/AAC.44.12.3249-3256.2000
  10. Nurizzo D, Shewry SC, Perlin MH, Brown SA, Dholakia JN, Fuchs RL, et al. 2003. The crystal structure of aminoglycoside- 3'-phosphotransferase-IIa, an enzyme responsible for antibiotic resistance. J. Mol. Biol. 327: 491-506. https://doi.org/10.1016/S0022-2836(03)00121-9
  11. Paul DJ, Seedhouse SJ, Disney MD. 2009. Two-dimensional combinatorial screening and the RNA Privileged Space Predictor program efficiently identify aminoglycoside-RNA hairpin loop interactions. Nucleic Acids Res. 37: 5894-5907. https://doi.org/10.1093/nar/gkp594
  12. Philipp WJ, Poulet S, Eiglmeier K, Pascopella L, Balasubramanian V, Heym B, et al. 1996. An integrated map of the genome of the tubercle bacillus, Mycobacterium tuberculosis H37Rv, and comparison with Mycobacterium leprae. Proc. Natl. Acad. Sci. USA 93: 3132-3137. https://doi.org/10.1073/pnas.93.7.3132
  13. Ryan KJ, Ray CG, Sherris JC. 2004. Sherris Medical Microbiology: An Introduction to Infectious Diseases, 4th Ed. McGraw-Hill, New York.
  14. Toth M, Vakulenko S, Smith CA. 2010. Purification, crystallization and preliminary X-ray analysis of Enterococcus casseliflavus aminoglycoside-2''-phosphotransferase-IVa. Acta Crystallogr. Sect. F Struct. Biol. Cryst. Commun. 66: 81-84. https://doi.org/10.1107/S1744309109050039
  15. Trott O, Olson AJ. 2010. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J. Comput. Chem. 31: 455-461.
  16. Wright GD. 1999. Aminoglycoside-modifying enzymes. Curr. Opin. Microbiol. 2: 499-503. https://doi.org/10.1016/S1369-5274(99)00007-7
  17. Wright GD, Berghuis AM, Mobashery S. 1998. Aminoglycoside antibiotics. Structures, functions, and resistance. Adv. Exp. Med. Biol. 456: 27-69. https://doi.org/10.1007/978-1-4615-4897-3_4

Cited by

  1. New Insights in to the Intrinsic and Acquired Drug Resistance Mechanisms in Mycobacteria vol.8, pp.None, 2013, https://doi.org/10.3389/fmicb.2017.00681
  2. Mycobacterial Aminoglycoside Acetyltransferases: A Little of Drug Resistance, and a Lot of Other Roles vol.10, pp.None, 2013, https://doi.org/10.3389/fmicb.2019.00046