Integron gene cassettes harboring novel variants of d-alanine-d-alanine ligase confer high-level resistance to d-cycloserine

Rahman, Md. Ajijur; Kaiser, Frank; Jamshidi, Shirin; Freitas Monteiro, Marta; Rahman, Khondaker Miraz; Mullany, Peter; Roberts, Adam P.

doi:10.1038/s41598-020-77377-4

Download PDF

Article
Open Access
Published: 26 November 2020

Integron gene cassettes harboring novel variants of d-alanine-d-alanine ligase confer high-level resistance to d-cycloserine

Md. Ajijur Rahman^1,2,
Frank Kaiser¹,
Shirin Jamshidi³,
Marta Freitas Monteiro¹^nAff6,
Khondaker Miraz Rahman³,
Peter Mullany¹ &
…
Adam P. Roberts^4,5

Scientific Reports volume 10, Article number: 20709 (2020) Cite this article

1921 Accesses
3 Citations
15 Altmetric
Metrics details

Subjects

Abstract

Antibiotic resistance poses an increasing threat to global health. To tackle this problem, the identification of principal reservoirs of antibiotic resistance genes (ARGs) plus an understanding of drivers for their evolutionary selection are important. During a PCR-based screen of ARGs associated with integrons in saliva-derived metagenomic DNA of healthy human volunteers, two novel variants of genes encoding a d-alanine-d-alanine ligase (ddl6 and ddl7) located within gene cassettes in the first position of a reverse integron were identified. Treponema denticola was identified as the likely host of the ddl cassettes. Both ddl6 and ddl7 conferred high level resistance to d-cycloserine when expressed in Escherichia coli with ddl7 conferring four-fold higher resistance to D-cycloserine compared to ddl6. A SNP was found to be responsible for this difference in resistance phenotype between the two ddl variants. Molecular dynamics simulations were used to explain the mechanism of this phenotypic change at the atomic scale. A hypothesis for the evolutionary selection of ddl containing integron gene cassettes is proposed, based on molecular docking of plant metabolites within the ATP and d-cycloserine binding pockets of Ddl.

Introduction

Globally, the emergence of antibiotic resistance in clinically important pathogens pose a significant threat to human health. Antibiotic resistance genes (ARGs) carried on various mobile genetic elements (MGEs), including plasmids, transposons and integron gene cassettes (GCs) are responsible for the emergence, and worldwide dissemination, of multidrug-resistance. Identifying the principal reservoirs of ARGs^1,2 as well as understanding the drivers for their evolutionary selection and spread³ are fundamental to tackling the problem of antibiotic resistance.

Integrons are genetic elements capable of capturing and expressing open reading frames (ORFs) embedded in GCs. They are well known for their role in dissemination of ARGs particularly in gram-negative pathogens. The basic structure of an integron is composed of a gene (intI) encoding a DNA integrase, an integron associated recombination site; attI and a gene cassette promoter (Pc), which expresses the inserted cassettes within the GC array. Usually the integron integrase gene and the genes in the GC array are divergently transcribed, however where the integrase and the GC array are transcribed in the same direction the integron is known as a “reverse integron”^4,5. The integron integrase (IntI) catalyses recombination between the cassette associated recombination site; attC and the attachment site; attI, within the integron platform^6,7,8. More than 130 MGE associated ARGs have been detected in integrons, conferring resistance to most classes of antibiotics including aminoglycosides, chloramphenicol, beta-lactams, trimethoprim and streptothricin^9,10. Integrons carry ARGs, plus other genes encoding proteins with more diverse functions including adaptation to stress, virulence and resistance to xenobiotics, and many, particularly those integrons located on chromosomes, encode proteins of unknown functions⁵. For instance, among the 70 ORFs detected within 47 gene cassettes in the chromosomal integron of Treponema denticola¹¹, an oral pathogen associated with periodontitis, only five ORFs matched with proteins of known functions (last confirmed August 2019).

This article reports the detection of two novel reverse integron-located genes encoding D-alanine-D-alanine ligases (Ddls). Ddls are ATP-dependent enzymes that play a critical role in the synthesis of peptidoglycan during the cytoplasmic stage by catalysing the formation of the d-ala-d-ala dipeptide required for cross-linking peptidoglycan strands during cell wall synthesis¹². The gene encoding Ddl is a housekeeping gene and has always been found in the chromosome, never in a MGE. The formation d-Ala-d-Ala dipeptide catalysed by Ddl is critical for bacterial growth^13,14. As there is no human homologue, Ddl is considered an excellent target for designing new antibiotics¹⁵. d-Cycloserine is the only inhibitor of Ddl used in the clinic¹³, however, due to the severe neurologic side effects, its use is currently reserved for the treatment of multidrug resistant (MDR) and extensively-drug resistant (XDR) tuberculosis^16,17.

We demonstrate d-cycloserine resistance associated with these integron located ddls and present an experimentally supported hypothesis as to why ddl, a housekeeping gene, has evolved to be on an integron gene cassette.

Results

Two novel variants of ddl were discovered from the human oral cavity located immediately downstream of the core structure of a reverse integron

During a PCR screen for the presence of integron GCs carrying ARGs in the saliva-derived metagenomic DNA of healthy volunteers from the UK (n = 11) and Bangladesh (n = 12), two natural variants of ddl located within the first GC of a reverse integron were discovered. In the libraries of PCR amplicons of integrons and associated GCs, a total of five clones (three from the UK and two from Bangladesh) were recovered carrying ddl predicted to encode a d-alanine-d-alanine ligase. Among the three ddl-carrying clones obtained from UK metagenomic libraries, one clone carried ddl6 and two clones carried ddl7 and among the two clones obtained from BD library, one clone carried ddl6 and another clone ddl7.The genetic organisation of the ddl inserts in the clones are shown in Fig. 1A. The intI and the ddl gene cassettes are oriented in the same direction in reverse integrons (where the integrase is orientated in the opposite direction compared to usual integrons) of T. denticola¹¹. Two putative integrase binding sites [L or S2 and R or S1], two direct repeat sequences (DR1 and DR2) and a putative attI/attC recombination site were detected (Fig. 1B). The stop codon of the ddl ORF was located on the 7 bp attC core site, R″ (Fig. 1C). The R″ site is followed by the L″ site-which covers the priming site of the reverse primer MARS2.

Two variants of ddl (named as ddl6 and ddl7) were different from each other by two SNPs at c.490 and c.777 positions of the coding sequences of the genes. The CTT (Leu164) and TGG (Trp259) codons of ddl6 were substituted with TTT (Phe164) and TGT (Cys259) in ddl7, respectively.

The closest homologue of ddl6 and ddl7 is a ddl located within the accessory genome of T. pedis B683

The closest homologue of the cassette-located ddls having 98% nucleotide sequence identity (97% identity at the amino acid level) was located on a 5699 bp contig of T. pedis B683 genome (GenBank accession: NZ_AOTN01000179) containing a total of six ORFs (Fig. 1D). This homologous ddl of T. pedis, named ddl(GC), was flanked by two attC recombination sites. The sequences of the core sites (R″ and L″) of attC located downstream of ddl(GC), were 100% identical with the corresponding sites of attC located downstream of ddl6 and ddl7 (Fig. 1D). Another attC was found upstream of ddl(GC) flanking two ORFs encoding hypothetical proteins (Fig. 1D). However, as there was no integrase gene, the association of this contig with an integron, although likely, could not be confirmed. The next closest homologue was Ddl of Syntrophobotulus glycolicus (ADY55260) with 55% amino acid identity, followed by Clostridium sp. (54%; WP_033164556), Lachnoclostridium phytofermentans (53%, WP_029502590) and Paenibacillus pini (52%; WP_036650467). Ddl proteins with altered specificity that confer resistance to vancomycin such as VanA (AAA65956), VanB (YP_009076352) and VanC (P29753) also shared low sequence identity (27 to 30%) with Ddl6 and Ddl7.

T. pedis B683 carries two ddl genes, one of which (GenBank accession: AOTN01000124: 8689-9912 bp) is most likely the house-keeping ddl as it is found in other strains of T. pedis including TA4 and TM1 and its G+C content (38.0%) is similar to the genome of T. pedis B683 (36.9%). In contrast, the G+C content of the GC located ddl is 30.6% suggesting that this second, cassette-located ddl may have been acquired by horizontal gene transfer by T. pedis B683.

Comparative genomics shows that a likely horizontal gene transfer (HGT) occurred between the T. pedis B683 and the host of ddl6 and ddl7

The following three observations suggested HGT between T. pedis B683 and the host of ddl6/ddl7: (i) The identical 40 bp sequences upstream of ddl6/ddl7 and their homologue ddl(GC); (ii) the nearly identical (90%) 29 bp downstream sequences containing R″ and L″ sites of attC (Fig. 1D); (iii) the phylogenetic tree constructed using the ddls of different species of Treponema showed that the integron-located ddl6 and ddl7 and ddl(GC) formed a separate clade within the tree (Fig. 1E).

Sequence of the upstream region of the integron carrying ddl7 suggests that the putative host of the integron carrying ddl6 and ddl7 is a strain of T. denticola

To confirm if the initially recovered 2024 bp PCR amplicons were obtained from the T. denticola genome, sequence of the intI that was further upstream was amplified by PCR. Analysis of a 4421 bp PCR product showed that it contained a partial ORF encoding a hypothetical protein (1–1362 bp), a complete ORF encoding IntI (1266 bp: 1925–3191 bp), a putative attI, the Pc promoter followed by a GC carrying ddl7 (3389–4421 bp) (Fig. 2). The complete intI had 96.0% nucleotide identity with intI of T. denticola H-22 (GenBank accession no: AGDV01000005:19792-21072). At the amino acid level, the identity ranges between 81 and 97% with IntI of different Treponema sp. Maximum identity was found with the IntI of T. denticola AL-2 (97%, EMB42956).

The genetic organisation of the 4421 bp amplicon was different from the closely related hits in the NCBI databases (Fig. 2). The non-protein-coding 556 bp sequence located between the partial ORF and the intI harbors the putative promoter for the intI (P_int) with a sequence of 5′-ATGAAT |19 bp| TAAACT-3′. The upstream sequence of the intI was analyzed in silico for potential LexA binding sites similar to that found in the upstream of intI of mobile and chromosomal integrons²⁰. However, no LexA binding motif could be detected. Additionally, this region contains several inverted repeat and direct repeat sequences as well as a putative transcriptional terminator sequence.

A phylogenetic tree was constructed to compare the evolutionary relationship of intIs associated with the integron carrying ddl with the intIs of different species of Treponema. The tree showed that the intI associated with ddl7 is closely related with other intIs of T. denticola and distantly related to T. pedis (Supplementary Fig. S1). The results obtained from the intI tree as well as the identity of the upstream sequence intI with the other strain of T. denticola suggest that the likely host of the integron associated with ddl7 is a species of the genus Treponema and most likely a strain of T. denticola.

Expression of ddl6 and ddl7 confers resistance to D-cycloserine only, not to vancomycin and beta-lactams

As the overexpression of ddl of M. tuberculosis and M. smegmatis was found to confer D-cycloserine resistance²¹, we hypothesised that the expression ddl6 or ddl7 in an integron array will result in d-cycloserine resistance, especially as they are detected in the first position due to the efficient expression by the Pc. When the minimum inhibitory concentrations (MIC) of d-cycloserine against the surrogate E. coli hosts EC126 and EC127 carrying pGEM-T Easy::intI-attI-ddl6 and pGEM-T Easy::intI-attI-ddl7, respectively were tested, a two to four-fold increase of MIC of d-cycloserine was observed (Table 1). As the upstream region of ddl6 and ddl7 cassettes carrying the attI and Pc were 100% identical, we concluded that the fourfold change in the MIC was due to either the SNP c.490 or c.777 or both.

Table 1 MIC of D-cycloserine against different strains of E. coli and B. subtilis expressing ddl.

Full size table

To determine if the overexpression of ddl in gram-positive bacteria can also confer resistance to d-cycloserine, vancomycin and β-lactam antibiotics, the 1032 bp coding sequence of ddl6 and ddl7 was cloned into a medium-copy expression vector, pHCMC05 and transformed into B. subtilis 168. The MIC of D-cycloserine against the strains BS706 and BS707 which carry pHCMC05::ddl6 and pHCMC05::ddl7 vectors, respectively was found to increase by eight-fold (64 µg/mL) compared to the BS700 strain that carries the empty plasmid (8 µg/mL) (Table 1). However, no difference in the MIC of D-cycloserine was observed between BS706 and BS707 carrying ddl6 and ddl7, respectively, when the expression of the genes was induced by IPTG. No resistance to vancomycin was observed (Table 1). Nor was there any resistance to penicillin G, amoxicillin, oxacillin and ampicillin against both E. coli and B. subtilis strains. This suggests that overproduction of Ddl does not confer cross-resistance to antibiotics targeting cell-wall biosynthesis.

Site-directed mutagenesis confirmed that the SNP at c.777 was responsible for the alteration of MIC of d-cycloserine

To identify which SNP of ddl6 and ddl7 was responsible for altering the susceptibility of E. coli to d-cycloserine, the nucleotides at position c.490 and c.777 of ddl6 were changed to the corresponding nucleotides of ddl7 by site-directed mutagenesis. MICs of d-cycloserine against E. coli carrying the constructs were determined and no change in the susceptibility was observed by the c.490 C>T substitution in ddl6, however, the G>T substitution which alters Trp259 to Cys259 (W259C) increased the MIC fourfold which is similar to the MIC of the strain carrying ddl7 (Table 1).

The integron-encoded Ddls are functional

The biological activity of Ddl6 and Ddl7 was confirmed using purified proteins (Fig. 3A). The release of the inorganic phosphate (Pi) in the Ddl-catalyzed reaction was assayed and both variants of Ddl catalyse the release of a significant amount of Pi into the reaction compared to the control (P > 0.05) (Fig. 3B). This indicates that ATP is consumed in the reaction and Pi is released. Furthermore, paper chromatography analysis of the Ddl-catalysed reaction product showed that both proteins catalyze the ligation of d-ala resulting in the synthesis of d-ala-d-ala dipeptide (Fig. 3C). The substrate specificity assay indicated that they could only catalyse the synthesis of d-ala-d-ala dipeptide, and not d-ala-d-ser or d-ala-d-lac.

The predicted 3D structures of Ddl6 and Ddl7 provided information on alteration of resistance phenotype and substrate specificity

To determine if the alteration of Trp259 of Ddl6 to Cys259 of Ddl7 caused a conformational change, 3D models of both Ddl6 and Ddl7 were constructed using I-TASSER²² (Supplementary Fig. S2). When the structures were superimposed using TM-align²³, they superimpose near perfectly with a TM score of 0.97. However, a notable predicted change in the conformation of the omega-loop region was observed likely resulting from the substitutions at 164 and 259 positions of Ddl6 and Ddl7 (Supplementary Fig. S3).

Three putative domains of Ddl6 and Ddl7 were identified by comparing predicted 3D structures of Ddl6 and Ddl7 with the crystal structure of D-ala-D-ser ligase (VanG, PDB code: 4FU0)²⁴, DdlB of E. coli²⁵, VanA of E. faecium²⁶ and Ddl of S. aureus²⁷. The predicted N-terminal domain of Ddl6 and Ddl7 runs from the N-terminus to Gly125, central domain from Ser126 to Gly219 and C-terminal domain from Arg220-Ile340. The putative omega-loop, which plays an important role in catalytic activity and substrate specificity of Ddl proteins, was located on the C-terminal domain of Ddl6 and Ddl7 and runs from Lys240-Ser255. The Trp259 of Ddl6 whose substitution with Cys259 causes a fourfold increase in D-cycloserine resistance was found to be located at the end of the 4th β-sheet (residues Asn256-Ile258) of the C-terminal domain and outside of the omega-loop (Supplementary Fig. S2). The amino acid residues (Ser150 and Tyr216) of E. coli DdlB which are conserved in the Ddls that can only produce D-ala-D-ala residues were also conserved in Ddl6 and Ddl7 with the corresponding residues of Ser184 and Tyr250 (Supplementary Fig. S4).This supports the results of the in-vitro assays and explains why it could produce only the d-ala-d-ala dipeptide, not d-ala-d-ser or d-ala-d-lac.

Although the integron-encoded Ddls were not found to confer vancomycin resistance and did not carry the amino acid residues for altered substrate specificity (d-Ser or d-Lac), among the closest homologues in PDB that matched with the predicted 3D structures of Ddl6 and Ddl7, four were related to vancomycin resistance including VanG (PDB code: 4FU0) and VanA (PDB code: 1E4E) with high TM scores (Supplementary Table S1).

Molecular dynamics simulations post-analysis provides insights into the mechanism of alteration of MIC of d-cycloserine due to W259C mutation in Ddl6

The results of 50 ns molecular dynamics (MD) simulations of wild-type Ddl6 and its W259C mutant derivative in complex with d-cycloserine (Fig. 4) showed that the ligand is released from the recombinant protein after approximately 20 ns, however, the complex with the wild-type protein remains stable in the binding site during the course of the MD simulations.

The movement of the ligand from the protein was studied by monitoring the distances (Supplementary Fig. S4) between the centre of d-cycloserine and Trp259 of the wild-type Ddl6 and Cys259 of the Ddl6 W259C mutant during the course of the simulation. The distance between the centre of d-cycloserine and Cys259 of the mutant protein progressively increased suggesting a movement away from the binding site, whereas the distance between the centre of d-cycloserine and Trp259 of the wild-type protein remained relatively stable throughout the simulation (Supplementary Fig. S5).

The relative binding free energy values calculated from the MD simulations trajectories further supported this observation as the complex of wild-type Ddl6 and d-cycloserine was more favorable (− 14.47 kcal/mol) than the complex of d-cycloserine and Ddl6 W259C mutant complex (− 12.15 kcal/mol) (Supplementary Table S2). Videos of the MD simulations showing the movement of the ligands are available in Supplementary Video S1 and Video S2.

Kinetic analysis of pure proteins showed that Ddl7 is more active than Ddl6

The integron-encoded Ddl6 and Ddl7 were characterised kinetically and the kinetic parameters were compared with DdlTd (the Ddl encoded by housekeeping ddl of T. denticola) and DdlAEc as positive control. The K_m,d-ala2 of Ddl6 and Ddl7 was determined as 0.39 and 0.54 mM, respectively (Table 2), thus, the values of K_m,d-ala2 could not explain the alteration of D-cycloserine phenotype between these two variants. However, the turnover number (K_cat) of Ddl7 (3.85 s⁻¹) was approximately three-fold higher than Ddl6 (1.20 s⁻¹), which nearly matches the fourfold increase of MIC of D-cycloserine for Ddl7 as compared to Ddl6.

Table 2 Steady-state kinetic parameters of Ddl6, Ddl7, DdlTd and DdlEc.

Full size table

Molecular docking analysis showed that flavonoids could be responsible for selection of the ddl6/ddl7 within the first position of the integron

We also tried to identify the likely selection pressure that could drive the evolution of ddls within integrons and maintain them as the first GC. As the use of D-cycloserine is restricted only for the MDR and XDR Mycobacterium tuberculosis, it is very unlikely to be responsible for this. We assumed that this selective pressure is common for both countries from which ddl containing strains were isolated (Bangladesh and the UK), as the ddls were ubiquitous within samples from both geographic locations. Candidates for providing this selective pressure are two plant flavonoids, quercetin and apigenin, which are abundantly present in many teas, fruits and vegetables and inhibit Ddl²⁹. Molecular docking was used to see if these flavonoids can bind to Ddl within the predicted D-alanine and/or ATP binding sites of Ddl6 and Ddl7. Salvicine (a terpenoid isolated from a Chinese medicinal plant, Salvia prionitis) was used as a positive control for the docking because this terpenoid was previously identified as a potent inhibitor of Ddl of T. pallidum using molecular docking³⁰. GOLD molecular docking showed that quercetin and apigenin can dock to both d-alanine (Fig. 5; Supplementary Table S3) and ATP binding sites (Supplementary Fig. S6; Supplementary Table S3) of Ddl6 and its mutants therefore, the integron-encoded Ddls may play a role in both the sequestration of these inhibitors, lowering the intracellular concentration and allowing the house-keeping Ddl to perform its physiological functions and also take on the role of the house-keeping Ddl during cell wall development.

Discussion

This study reports for the first time the discovery of integron GCs encoding two novel isoforms of Ddl (Ddl6 and Ddl7) derived from the oral cavity microbiota of healthy humans in the UK and Bangladesh. The ORFs encoding Ddl were located immediately downstream of intI on reverse integrons. All of the genetic characteristics of an integron and its associated GCs were found within the amplicon, including the presence of a putative Pc, an attI region harbouring putative integrase binding sites (S1 and S2), a putative attI/attC recombination site, a putative RBS located 6-bp upstream of the start codon of the ddls as well attC core sites (R″ and L″)^6,7,10,31.

The closest homologue of ddl6 and ddl7 with approximately 98.0% nucleotide identity was located on a whole genome sequence (WGS) contig of T. pedis B683 and is one of two copies of ddl within this genome and is not present in the genome of other T. pedis strains, thus it is likely part of the accessory genome. T. pedis is an animal pathogen that causes digital dermatitis in cattle and necrotic ulcers in pigs³². B683 (GenBank (NZ_AOTN00000000.1) was isolated from a shoulder ulcer of a pig. The evidence showed that there is a likely genetic exchange between T. pedis and the host of ddl6/ddl7, likely to be T. denticola, which may involve one or more transfer intermediates.

Analysis of the upstream sequence of the integron carrying ddl6/ddl7 suggests that the likely host of the integron is a strain of T. denticola which carries two copies of ddl in its genome: one as an integron GC and another as house-keeping gene (this house-keeping ddl of T. denticola ATCC35404 showed only 46.0% nucleotide identity to ddl6 and ddl7). T. denticola is one of the most comprehensively characterised species among the 49 species of Treponema in the human oral cavity³³. It is one of the members of the ‘red complex’ associated with clinical progression of periodontitis³⁴ and is capable of transferring genes to different genera of bacteria in dental plaque³⁵. It is the only bacterial species in the oral cavity that has been found to carry integrons¹¹ although we have recently demonstrated other bacterial species are also likely to contain integrons³⁶. From the in silico analysis of the metagenomic datasets of the NIH Human microbiome project, a total of 826 gene cassettes related to Treponema sp. were detected in the oral cavity³⁷. The closely related strains of T. denticola shared a high nucleotide identity (> 80%) of their integron gene cassettes³⁸.

Our results provide novel insights into the evolutionary mechanisms that have facilitated the acquisition of a second copy of ddl on an integron in the original host of ddl6 and ddl7. Zawadzke et al.²⁸ proposed multiple explanations for the duplication of ddl seen in E. coli however these are likely not applicable here because, unlike the duplicated ddlA and ddlB genes of E. coli, the ddl cassettes described here are likely to have been acquired by HGT and thus, they may be related to gaining a new function (resistance to antimicrobial compounds) in addition to, or rather than, providing increased amount of the gene product^28,39. The kinetic analysis of the purified proteins (Ddl6, Ddl7 and DdlTd) also indicates that the ddls are interchangeable with respect to their core function of cell wall assembly because, the K_cat (s⁻¹) of DdlTd was very close to Ddl7 thus, the acquisition of a second copy of ddl within integrons provide resistance by titrating out intracellular antimicrobials plus complementation of the original Ddl function.

We found that the ectopic expression of ddl6 and ddl7 confers resistance to d-cycloserine. As Ddl is a target of d-cycloserine, overproduction of Ddl is likely to facilitate resistance by mechanisms similar to D-cycloserine resistance in Mycobacterium which is mediated by high expression of both Ddl and Alr^21,40. The other known mechanisms of D-cycloserine resistance in Mycobacterium were the non-synonymous mutations in Alr including S22L, K133E, L89R, M319T, Y364D and R373G^41,42 as well as loss of function mutation in alr⁴². In E.coli the mutation in cycA⁴³, dadA and pnp⁴⁴ confer resistance to d-cycloserine. The mutations in a penicillin binding protein (PBP4) homologue in M. smegmatis confer resistance to d-cycloserine and vancomycin⁴⁵. Although several mutations in alr have been implicated in d-cycloserine resistance, so far no mutations in ddl were found to be associated with alteration of d-cycloserine resistance by the host. Moreover, several functional metagenomic analyses reported that ectopic expression of various ddl also conferred resistance to d-cycloserine^46,47,48,49.

While the expression of both integron encoded ddls conferred resistance to d-cycloserine, the expression of ddl7 had a four-fold higher MIC compared to ddl6. Site-directed mutagenesis confirmed that a single substitution at c.777, which alters Trp259 of Ddl6 to Cys259, was responsible for this. Using MD simulations, we have shown how this mutation is predicted to alter the binding affinity of the ligand d-cycloserine to the active sites of the proteins. The wild-type Ddl6 with a tryptophan residue at its 259-position had more stable binding to d-cycloserine, therefore less d-cycloserine would be required to occupy active sites which in turn will decrease the MIC. When the aromatic, non-polar, hydrophobic tryptophan residue was changed to polar, thiol reactive⁵⁰ and disulphide bridge forming⁵¹ cysteine (the residue at 259 for Ddl7), binding between the mutant Ddl6 and d-cycloserine complex was comparatively less stable and thus, more antibiotic will be needed to occupy the active sites which causes an increase in the MIC.

Four of the closest structural homologues of Ddl6/Ddl7 were related to vancomycin resistance including VanA of E. faecium, and VanG of E. faecalis. It is known that the function of a protein is more related to the structure rather than the sequence⁵². The inclination of the predicted 3D structures of Ddl6 and Ddl7 towards vancomycin resistant determinants suggests that if the corresponding residues of VanG or VanA, which are responsible for binding and positioning d-ser (VanG-type) or d-lac (VanA-type) instead of d-ala2, are substituted to the corresponding residue of Ddl6 and Ddl7 (Tyr250), they could gain d-ala-d-ser dipeptide or d-ala-d-lac depsipeptide ligase activity. The E. coli DdlB gained a weak d-ala-d-lac depsipeptide activity following Tyr216 and Ser150 substitutions with phenylalanine and alanine of LmDdl2⁵³, respectively supporting our hypothesis.

We also suggest a hypothesis for evolution of ddl within an integron cassette at the first position. d-Cycloserine is currently only used in the treatment of MDR and XDR-tuberculosis⁵⁴ and there is no history of environmental use of d-cycloserine in agriculture. Additionally, none of the volunteers received this drug (or any other antibiotic), at least in the three months prior to saliva collection. Therefore, it is very unlikely that d-cycloserine represents the driving force for this event. We hypothesize that dietary flavonoids or other inhibitory small molecules present in our diet that can bind Ddl drive the acquisition and selection of ddl within integrons. Some flavonoids including apigenin, quercetin and catechin are known to have antibacterial activity^55,56 and are found in fruits and vegetables as well as widely consumed beverages such as tea. Apigenin and quercetin has been shown to inhibit the activity of Ddl of H. pylori by competitively inhibiting the binding of ATP to ATP-binding sites^29,57. MD analysis performed in this study supported the findings and demonstrated that these flavonoids are likely to bind to both d-ala and ATP binding sites of Ddl6 and Ddl7. Thus, the excess Ddl produced by the efficient expression under the control of Pc could confer a selective advantage on the host titrating out inhibitory flavonoids (Fig. 6). This hypothesis requires testing experimentally which is now under way.

The results show that ddl, which is generally considered a housekeeping gene has moved onto an integron and can be found in the first position, with the most likely selection pressure being dietary compounds with antimicrobial activity. Because these genes also confer resistance to d-cycloserine and are potentially one mutation away from vancomycin resistance, the study also sheds light on potential environmental drivers of clinically relevant antimicrobial resistance.

Materials and methods

Source of chemicals, reagents and antibiotics

Chemicals, solvents, culture media and antibiotics were purchased from Sigma-Aldrich Ltd (Dorset, UK), BDH (UK) and Life Technologies (USA) unless stated otherwise. Plasmid preparation kits were obtained from Qiagen (UK). Restriction enzymes were obtained either from New England Biolabs (UK) Ltd or from Promega (UK). The EnzChek Phosphate assay kit and Pierce BCA Protein assay kits were purchased from Life Technologies (USA). E. coli α-Select competent cells were obtained from Bioline Reagents Ltd (UK). The primers were obtained from Sigma-Aldrich (UK).

Bacterial strains and plasmids

The pGEM-TEasy vector (Promega) was used for constructing the metagenomic library of PCR amplicons of integrons and gene cassettes. E. coli α-select silver efficiency (Bioline) was used to manipulate the pGEM-TEasy library of gene cassettes. pET28a (Novagen) and E. coli BL21 (DE3) (New England Biolabs) were used for cloning and expression of ddls and purification of corresponding His-tagged proteins, respectively. Bacillus subtilis 168 and pHCMC05 (Bacillus Genetic Stock Centre) were used for cloning and expression of ddl6 and ddl7 to determine the susceptibility of D-cycloserine and vancomycin in Gram-positive bacteria. E. coli strains were cultured in LB medium at 37 °C. Bacillus subtilis 168 was cultured in Brain–heart infusion (BHI) medium unless stated otherwise. The bacterial strains were preserved at − 80 °C in 30% (v/v) sterile glycerol.

Detection of ddl gene cassettes in the metagenomic DNA obtained from saliva

The collection of saliva samples, preparation of metagenomic DNA from pooled saliva from UK and Bangladesh and detection of integrons and gene cassettes in the library PCR amplicons have been described previously³⁶. Primers for reverse integrons were utilized as PCRs using primers to detect integrons failed to yield amplicons from metagenomic DNA³⁶. Ethical approval was obtained from University College London (UCL) Ethics Committee (project number 5017/001) and the Institutional Animal, Medical Ethics, Biosafety and Biosecurity Committee (IAMEBBC) for Experimentations on Animal, Human, Microbes and Living Natural Sources, University of Rajshahi (project number 54/320/IAMEBBC/IBSC). All volunteers provided informed consent and all procedures were carried out according to relevant guidelines and regulations. Briefly, the PCR products produced using the forward integrase primer (TDIF) and reverse attC-based primer (MARS-2) were cloned into pGEM-TEasy and transformed into E. coli alpha select cells. Transformants were randomly picked, plasmids were prepared from 5 mL overnight culture using Miniprep kits (Qiagen) and the plasmids were sequenced using M13F-M13R primers (Supplementary Table S4).

Determination of the upstream region of integron carrying ddl6 and ddl7

A total of four different forward primers (Upint_5100F, Upint_3685F, Upint_721F, Upint_122F) (Supplementary Table S4) were designed randomly at different locations upstream of the intI gene of the chromosome of T. denticola ATCC35405 (NC_002967.9). PCR was performed by combining these primers with the reverse primer, ddlR designed from the 3′-end of ddl6/ddl7. It was assumed that if the integrons carrying ddl GCs are located on the chromosome of T. denticola, the PCR products will yield a different length of upstream region of intI.Q5 High-Fidelity 2X Master Mix (NEB, UK) was used for PCR using the pooled metagenomic DNA of UK saliva samples as template. The reaction condition for PCR was as follows: an initial denaturing temperature of 98 °C for 30 s followed by 30 cycles at 98 °C for 10 s (denaturing), 50–58 °C for 30 s (annealing), and 72 °C for 200 s (extension). The PCR run ended with a final 72 °C elongation step for 10 min before holding the reaction at 4 °C. The expected size PCR products were purified by gel extraction using QIAquick Gel Extraction Kit (Qiagen, Germany) and A-tailed following the protocol for cloning blunt-ended PCR products described in pGEM-T Easy Vector Systems Technical Manual (Promega, UK). A 30 µL reaction mixture for A-tailing contained about 30 ng purified PCR product, 3 µL of 10 × standard Taq reaction buffer, 1.5 µL 100 mM dATP, 2 µL Taq DNA polymerase (NEB, UK) and molecular grade water up to 30 µL. The reaction mixture was then incubated for 30 min at 72 °C in a block of PCR machine. An aliquot of A-tailed PCR products was cloned into pGEM-T Easy and then transformed into E. coli alpha select cells. The transformants were selected on LB agar plates supplemented with ampicillin, IPTG and X-gal. Plasmids were prepared and the inserts were sequenced.

Site-directed mutagenesis

The Phusion Site-Directed Mutagenesis Kit (Thermo Scientific, USA) was used to construct two different mutants of ddl6: ddl6 c.490 C>T and ddl6 c.777 G>T using the pGEM-TEasy::intI-attI-ddl6 plasmid as template. The phosphorylated mutagenic primers used for constructing the mutants are listed in the Supplementary Table S4. After confirming the success of the mutagenesis reactions, the mutant plasmids were transformed to E. coli α-select cells and MIC of the transformants were determined using the agar dilution method.

Sub-cloning of ddl6 and ddl7 into pHCMC05 and transformation into B. subtilis

The genes for Ddl6 and Ddl7 were amplified by PCR using the pGEM-TEasy vectors with the 2024-bp insert containing ddl6 and ddl7 as template (see Supplementary Table S4 for primers used). The genes were cloned into BamHI and XbaI sites of pHCMC05 (E.coli-B. subtilis shuttle vector) using the primers TddlF and TddlR (Supplementary Table S4). Chemically competent B. subtilis 168 cells were prepared and transformed with the recombinant pHCMC05 vectors using the method described by Hardy⁵⁸. Transformants were spread onto BHI plates supplemented with 10 µg/mL of chloramphenicol and incubated at 37 °C for 16–18 h.

Cloning and expression of ddl6 and ddl7

The coding sequence of the genes for Ddl6 and Ddl7 (1032-bp) were amplified and cloned into the BamHI and XhoI sites of pET28a vector (Novagen) using the primers TddlF and Tddl28aR (Supplementary Table S4). The recombinant plasmids were transformed into E. coli BL21 (DE3) cells (New England Biolabs) and transformants were selected on LB agar supplemented with kanamycin (30 µg/mL). 1L LB was inoculated with 10 ml overnight culture, and expression was induced in exponential growth phase by addition of 0.5 mM isopropyl β-d-1-thiogalactopyranoside for 3 h at 37 °C. Cells were harvested by centrifugation at 15,000×g at 4 °C and stored at − 20 °C until protein purification.

Purification of N- and C-terminal His-tagged Ddl6 and Ddl7 proteins by FPLC

Preserved cells were thawed on ice for 30 min and resuspended in buffer A (50 mM NaH₂PO₄, 300 mM NaCl, 10 mM imidazole, pH 7.4) containing cOmplete EDTA-free protein inhibitor cocktail (Roche, 1 tablet/10 mL), lysozyme (1.0 mg/mL) and DNAse (1 U/µL). The suspension was incubated on ice for 30 min and sonicated (6 pulse, each 10 s). Cellular debris was removed by centrifugation at 15,000×g for 30 min at 4 °C. Recombinant, N-and C-terminally His-tagged proteins were purified using a Ni–NTA Superflow cartridge (1 mL, Qiagen) and the Fast Protein Liquid Chromatography (FPLC) system (BioRad). The proteins were eluted with filter sterilized buffer B (50 mM NaH₂PO₄, 300 mM NaCl, 250 mM imidazole, pH 7.4). Protein concentration was determined using Pierce BCA Protein Assay Kit (ThermoFisher Scientific) and purity was checked by sodium-dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE). The gels were stained with Coomassie Blue (BioRad) and destained with destaining reagent (10% acetic acid and 20% methanol).

Enzymatic activity assay

The d-ala-d-ala ligase activity of purified Ddl enzymes was assayed as described previously with some modifications^59,60: (1) Formation of d-ala-d-ala dipeptide: The reaction mixture containing the following components were incubated for 4 h at 37 °C: 50 mM Tris–HCl, 10 mM MgCl2, 10 mM KCl, 20 mM d-alanine, 2 mM ATP, 2.5 mM Glutathione and 20 µg/mL of purified Ddl. To stop the reaction, the mixtures were boiled for 5 min at 95 °C. To determine the substrate specificity, an equimolar amount of D-Ser or d-Lac was substituted with d-Ala. Next, 4 µL of the reactions were loaded onto cellulose chromatography paper (Whatman) and ascending chromatography was developed for 3 h in butanol-acetic acid–water (12:3:5 vol/vol/vol) containing 1.0% ninhydrin. The paper was dried for 5 min at 95 °C. d-ala-d-ala dipeptide formation in the reaction catalysed by Ddl was identified by comparison with commercially available d-ala-d-ala (Sigma). (2) Measurement of inorganic phosphate in the Ddl-catalysed reaction: The enzymatic activity of Ddl was also assayed by measuring the release of inorganic phosphate in the reaction (similar to the d-ala-d-ala dipeptide formation assay) with EnzChek phosphate assay kit (Life Technologies). 10 µL of the reaction product was added to 790 µL of standard reaction mixture containing 230 µL dH₂O, 50 µL 20X reaction buffer, 200 µL 2-amino-6-mercapto-7-methylpurine riboside (MESG) substrate solution and 10 µL purine nucleoside phosphorylase and incubated at 22 °C for 30 min. After incubation absorbance of the reactions was measured at 360 nm with a spectrophotometer (Ultrospec 2000, Pharmacia Biotech). The amount of inorganic phosphate released into the reactions was calculated from the standard curve of inorganic phosphate.

Kinetic analysis of Ddl6 and Ddl7, DdlAEc and DdlTd

The V_max and K_{m, d-ala2} of Ddl6, Ddl7, DdlTd and DdlAEc were determined by using pyruvate kinase (PK)/lactate dehydrogenase (LDH)-coupled spectrophotometric system^60,61.

All assays were conducted at 37 °C in 96-well plate system using CLARIOstar (BMG Labtech, Germany) in 200 µL reaction buffer. The assay mixture (200µL) contained 100 mM HEPES (pH 7.5), 10 mM MgCl₂, 10 mM KCl, 0.2 mM NADH, 6–10 U/mL PK, 9–14 U/mL LDH, 2 mM phosphoenolpyruvate (PEP) and variable concentrations of ATP and D-alanine (saturating concentrations were 500 µM and 100 mM for ATP and d-alanine, respectively). The reactions were initiated by adding Ddl enzymes at a final concentration of 0.0025 µg/µL. The initial velocity (v₀) of the reactions was determined by measuring the changes in the absorbance at 340 nm for 500 s (8.3 min). The slope of the reaction progression curves (time (s) vs change in absorbance at 340 nm) is equivalent to the v₀ (v₀ = slope = ∆Y/∆X)⁶². The slope was calculated by the MARS data analysis software (BMG Labtech, Germany). After the v₀ values for each substrate concentration (ATP and D-alanine) were obtained, GraphPad Prism version 7 (GraphPad Software, Inc., USA) was used to determine the V_max and K_m values by fitting the curves for Michaelis–Menten model.

The values for K_m,ATP of different Ddls were determined at a fixed saturating concentration of d-alanine (100 mM) and varying concentrations of ATP starting from 5 µM to 320 µM. The K_m values for second d-alanine binding (K_m,d-ala2) was determined by using a fixed saturating concentration of ATP (500 µM; approximately 10 times of K_m,ATP). The values of K_cat (s⁻¹) for all enzymes were determined by dividing the Vmax (µM/s) with the concentration of enzymes, [E] (µM).

Phylogenetic analysis

The phylogenetic tree was constructed using MEGA X software¹⁹. The sequences of Ddl6 and Ddl7 and their homologs were aligned with Clustal Omega. The default settings for Neighbour-Joining method in MEGA X were applied to construct the phylogenetic tree.

Homology modelling of Ddl6 and Ddl7

The homology models of Ddl6 and Ddl7 were constructed using I-Tasser server (https://zhanglab.ccmb.med.umich.edu/I-TASSER)²². The amino acid sequences were submitted to the server and among the top five models built by the server, model 1 was selected based on the highest C-score and TM-score. A higher value of C-score signifies a model with a high confidence. A TM-score > 0.5 indicates a model of correct topology and a TM-score < 0.17 means a random similarity⁶³. Models were visualised with PyMOL (The PyMOL Molecular Graphics System, Version 1.8 Schrödinger, LLC). The homology models of Ddl6 and Ddl7 were matched with each other with the TM-align structural alignment program (https://zhanglab.ccmb.med.umich.edu/TM-align) and the overlapped structures were evaluated with PyMOL.

Determination of minimum inhibitory concentrations (MIC):

The MIC of d-cycloserine, vancomycin and beta-lactams were determined by agar dilution method following the guidelines of Clinical and Laboratory Standards Institute (CLSI) published in 2012. Three individual colonies of E. coli harbouring pET-ddl6 or pET-ddl7 and B. subtilis 168 strains harbouring pHCMC05-ddl6 or pHCMC05-ddl7 were inoculated in Mueller–Hinton (MH) broth supplemented with appropriate antibiotics and cultured at 37 °C for 4–5 h and the OD₆₀₀ was adjusted to 0.08 to 0.1 using sterile saline. MH agar plates supplemented with various concentration of antibiotics starting from 0.125 to 256 µg/mL were prepared and inoculated with the adjusted bacterial suspensions with a multipoint inoculator. The spots were allowed to dry for 30 min and the plates were then incubated aerobically at 37 °C for 16–20 h. The MIC value was defined as the lowest concentration of the inhibitory compounds giving rise to no visible growth. MIC determinations were performed at least three times in triplicates. As a control, E. coli or B. subtilis 168 strains transformed with empty vectors were used.

Molecular docking

The predicted 3D-structures of Ddl6 and its two single mutants, L164F and W259C, as well as Ddl encoded by house-keeping gene of T. denticola (DdlTd) were used as the targets for molecular docking analyses. The 3D structure of Ddl6 was applied in PyMol program to generate L164F and W259C mutants. These protein targets were used for molecular docking of six ligands including d-alanine, d-cycloserine, ATP, apigenin, quercetin and salvicine, after minimisation and equilibration using AMBER program. AutoDock SMINA⁶⁴ was used to determine the best binding pocket by exploring all probable binding cavities in the proteins. Then GOLD (Genetic Optimization for Ligand Docking)^65,66 was used for docking the ligands into the SMINA-located binding site to perform flexible molecular docking analyses.

Genetic algorithm (GA) was used in GOLD ligand docking to thoroughly examine the ligand conformational flexibility along with partial flexibility of the protein⁶⁷. The maximum number of runs for the ligand was set to 20 and in each run a population size of 100 with 100,000 operations was employed. The number of islands was 5, and the niche size of 2 was considered. The default cut-off values of hydrogen bonds was set to 2.5 Å (dH-X), and for the van-der-Waals distance it was 4.0 Å. The GA docking was terminated when the top solutions attained the root mean square deviation (RMSD) values within 1.5 Å.

Molecular dynamics (MD) simulations

After molecular docking, the best poses of d-cycloserine in the d-Ala binding site of wild-type and W259C mutant of Ddl6 were selected as the starting structures to run MD simulations for 50 ns. The MD simulations were carried out using the AMBER 16.0 software package⁶⁸. The force fields parameters for D-cycloserine were generated using the ANTECHAMBER module of the AMBER programme. Each system was solvated using an octahedral box of TIP3P water molecules. Periodic boundary conditions and particle-mesh Ewald (PME) method were employed in all the simulations⁶⁹. Particle-mesh Ewald method enabled us to calculate the 'infinite' electrostatics without truncating the parameters. During each simulation, all bonds in which the hydrogen atom was present were considered fixed, and all other bonds were constrained to their equilibrium values by applying the SHAKE algorithm⁷⁰. A cutoff radius of 12 Å was used for the systems. Minimization was performed in two phases, and each phase was performed in two stages. In the first phase, ions and all water molecules were minimized for 1500 cycles of steepest descent followed by 1500 cycles of conjugate gradient minimization. Afterwards, the systems were minimized for a total of 5000 cycles without restraint wherein 2500 cycles of steepest descent were followed by 2500 cycles of conjugate gradient minimization. After minimizations, the systems were heated for 1000 ps while the temperature was raised from 0 to 300 K, and then equilibration was performed without a restraint for 1000 ps while the temperature was kept at 300 K. Sampling of reasonable configurations was conducted by running a 50 ns simulation with a 2 fs time step at 300 K and 1 atm pressure. A constant temperature was maintained by applying the Langevin algorithm while the pressure was controlled by the isotropic position scaling protocol used in AMBER 16 package program⁶⁸.

Statistical analysis

Statistical analysis was performed by using GraphPad Prism (GraphPad Software, Inc., USA; version 7.0). Statistical significance was calculated using one-way ANOVA. P values of ≤ 0.05 were considered statistically significant.

GenBank accession numbers

The sequences of the 2024 bp PCR amplicons containing ddl6 and ddl7 were submitted separately to GenBank with the accession number of KU886208 and KU886209, respectively. The 4421 bp sequence of PCR product containing the upstream sequence of intI along with the downstream ddl7 was submitted to GenBank with the accession number KY039278.

References

Bush, K. et al. Tackling antibiotic resistance. Nat. Rev. Micro. 9, 894–896 (2011).
Article CAS Google Scholar
D’Costa, V. M., McGrann, K. M., Hughes, D. W. & Wright, G. D. Sampling the antibiotic resistome. Science 311, 374–377. https://doi.org/10.1126/science.1120800 (2006).
Article ADS PubMed Google Scholar
Holmes, A. H. et al. Understanding the mechanisms and drivers of antimicrobial resistance. Lancet 387, 176–187. https://doi.org/10.1016/S0140-6736(15)00473-0 (2016).
Article CAS PubMed Google Scholar
Labbate, M. et al. Use of chromosomal integron arrays as a phylogenetic typing system for Vibrio cholerae pandemic strains. Microbiology 153, 1488–1498. https://doi.org/10.1099/mic.0.2006/001065-0 (2007).
Article CAS PubMed Google Scholar
Gillings, M. R. Integrons: past, present, and future. Microbiol. Mol. Biol. Rev. 78, 257–277. https://doi.org/10.1128/MMBR.00056-13 (2014).
Article CAS PubMed PubMed Central Google Scholar
Collis, C. M. & Hall, R. M. Expression of antibiotic resistance genes in the integrated cassettes of integrons. Antimicrob. Agents Chemother. 39, 155–162 (1995).
Article CAS PubMed PubMed Central Google Scholar
Collis, C. M., Grammaticopoulos, G., Briton, J., Stokes, H. W. & Hall, R. M. Site-specific insertion of gene cassettes into integrons. Mol. Microbiol. 9, 41–52 (1993).
Article CAS PubMed Google Scholar
Recchia, G. D. & Hall, R. M. Gene cassettes: a new class of mobile element. Microbiology 141(Pt 12), 3015–3027. https://doi.org/10.1099/13500872-141-12-3015 (1995).
Article CAS PubMed Google Scholar
Cambray, G., Guerout, A. M. & Mazel, D. Integrons. Annu. Rev. Genet. 44, 141–166. https://doi.org/10.1146/annurev-genet-102209-163504 (2010).
Article CAS PubMed Google Scholar
Partridge, S. R., Tsafnat, G., Coiera, E. & Iredell, J. R. Gene cassettes and cassette arrays in mobile resistance integrons. FEMS Microbiol. Rev. 33, 757–784. https://doi.org/10.1111/j.1574-6976.2009.00175.x (2009).
Article CAS PubMed Google Scholar
Coleman, N., Tetu, S., Wilson, N. & Holmes, A. An unusual integron in Treponema denticola. Microbiology 150, 3524–3526. https://doi.org/10.1099/mic.0.27569-0 (2004).
Article CAS PubMed Google Scholar
Neuhaus, F. C. The enzymatic synthesis of D-alanyl-D-alanine. Biochem. Biophys. Res. Commun. 3, 401–405. https://doi.org/10.1016/0006-291X(60)90053-X (1960).
Article CAS PubMed Google Scholar
Škedelj, V. et al. 6-Arylpyrido[2,3-d]pyrimidines as novel ATP-competitive inhibitors of bacterial D-alanine:D-alanine ligase. PLoS ONE 7, e39922. https://doi.org/10.1371/journal.pone.0039922 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Halouska, S. et al. Metabolomics analysis identifies d-alanine-d-alanine ligase as the primary lethal target of d-cycloserine in mycobacteria. J. Proteome Res. 13, 1065–1076. https://doi.org/10.1021/pr4010579 (2014).
Article CAS PubMed Google Scholar
Tytgat, I. et al. DD-ligases as a potential target for antibiotics: past, present and future. Curr. Med. Chem. 16, 2566–2580 (2009).
Article CAS PubMed Google Scholar
Caminero, J. A., Sotgiu, G., Zumla, A. & Migliori, G. B. Best drug treatment for multidrug-resistant and extensively drug-resistant tuberculosis. Lancet. Infect. Dis. 10, 621–629. https://doi.org/10.1016/S1473-3099(10)70139-0 (2010).
Article CAS PubMed Google Scholar
Anderson, R. J., Groundwater, P. W., Todd, A. & Worsley, A. J. Antibacterial Agents 319–326 (Wiley, Hoboken, 2012).
Book Google Scholar
Saitou, N. & Nei, M. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406–425 (1987).
CAS PubMed Google Scholar
Kumar, S., Stecher, G., Li, M., Knyaz, C. & Tamura, K. MEGA X: molecular evolutionary genetics analysis across computing platforms. Mol. Biol. Evol. 35, 1547–1549. https://doi.org/10.1093/molbev/msy096 (2018).
Article CAS PubMed PubMed Central Google Scholar
Guerin, É. et al. The SOS response controls integron recombination. Science 324, 1034–1034. https://doi.org/10.1126/science.1172914 (2009).
Article ADS CAS PubMed Google Scholar
Feng, Z. & Barletta, R. G. Roles of Mycobacterium smegmatis D-alanine:D-alanine ligase and D-alanine racemase in the mechanisms of action of and resistance to the peptidoglycan inhibitor D-cycloserine. Antimicrob. Agents Chemother. 47, 283–291 (2003).
Article CAS PubMed PubMed Central Google Scholar
Yang, J. & Zhang, Y. I-TASSER server: new development for protein structure and function predictions. Nucleic Acids Res. 43, W174-181. https://doi.org/10.1093/nar/gkv342 (2015).
Article CAS PubMed PubMed Central Google Scholar
Zhang, Y. & Skolnick, J. TM-align: a protein structure alignment algorithm based on the TM-score. Nucleic Acids Res. 33, 2302–2309. https://doi.org/10.1093/nar/gki524 (2005).
Article CAS PubMed PubMed Central Google Scholar
Meziane-Cherif, D., Saul, F. A., Haouz, A. & Courvalin, P. Structural and functional characterization of VanG D-Ala:D-Ser ligase associated with vancomycin resistance in Enterococcus faecalis. J. Biol. Chem. 287, 37583–37592. https://doi.org/10.1074/jbc.M112.405522 (2012).
Article CAS PubMed PubMed Central Google Scholar
Fan, C., Moews, P. C., Walsh, C. T. & Knox, J. R. Vancomycin resistance: structure of D-alanine:D-alanine ligase at 2.3 A resolution. Science 266, 439–443 (1994).
Article ADS CAS PubMed Google Scholar
Roper, D. I., Huyton, T., Vagin, A. & Dodson, G. The molecular basis of vancomycin resistance in clinically relevant Enterococci: crystal structure of d-alanyl-d-lactate ligase (VanA). Proc. Natl. Acad. Sci. USA. 97, 8921–8925 (2000).
Article ADS CAS PubMed PubMed Central Google Scholar
Liu, S. et al. Allosteric inhibition of Staphylococcus aureus d-alanine:d-alanine ligase revealed by crystallographic studies. Proc. Natl. Acad. Sci. USA. 103, 15178–15183. https://doi.org/10.1073/pnas.0604905103 (2006).
Article ADS CAS PubMed PubMed Central Google Scholar
Zawadzke, L. E., Bugg, T. D. & Walsh, C. T. Existence of two D-alanine:D-alanine ligases in Escherichia coli: cloning and sequencing of the ddlA gene and purification and characterization of the DdlA and DdlB enzymes. Biochemistry 30, 1673–1682 (1991).
Article CAS PubMed Google Scholar
Wu, D. et al. D-Alanine:D-alanine ligase as a new target for the flavonoids quercetin and apigenin. Int. J. Antimicrob. Agents 32, 421–426. https://doi.org/10.1016/j.ijantimicag.2008.06.010 (2008).
Article CAS PubMed Google Scholar
Dwivedi, U. N. et al. Treponema pallidum putative novel drug target identification and validation: rethinking syphilis therapeutics with plant-derived terpenoids. OMICS 19, 104–114. https://doi.org/10.1089/omi.2014.0154 (2015).
Article CAS PubMed PubMed Central Google Scholar
Partridge, S. R. et al. Definition of the attI1 site of class 1 integrons. Microbiology 146(Pt 11), 2855–2864 (2000).
Article CAS PubMed Google Scholar
Svartstrom, O., Mushtaq, M., Pringle, M. & Segerman, B. Genome-wide relatedness of Treponema pedis, from gingiva and necrotic skin lesions of pigs, with the human oral pathogen Treponema denticola. PLoS ONE 8, e71281. https://doi.org/10.1371/journal.pone.0071281 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Dashper, S. G., Seers, C. A., Tan, K. H. & Reynolds, E. C. Virulence factors of the oral spirochete Treponema denticola. J. Dent. Res. 90, 691–703. https://doi.org/10.1177/0022034510385242 (2011).
Article CAS PubMed PubMed Central Google Scholar
Holt, S. C. & Ebersole, J. L. Porphyromonas gingivalis, Treponema denticola, and Tannerella forsythia: the ‘red complex’, a prototype polybacterial pathogenic consortium in periodontitis. Periodontology 2000(38), 72–122. https://doi.org/10.1111/j.1600-0757.2005.00113.x (2005).
Article Google Scholar
Wang, B. Y., Chi, B. & Kuramitsu, H. K. Genetic exchange between Treponema denticola and Streptococcus gordonii in biofilms. Oral Microbiol. Immunol. 17, 108–112. https://doi.org/10.1046/j.0902-0055.2001.00001.x (2002).
Article CAS PubMed Google Scholar
Tansirichaiya, S., Rahman, M. A., Antepowicz, A., Mullany, P. & Roberts, A. P. Detection of novel integrons in the metagenome of human saliva. PLoS ONE 11, e0157605. https://doi.org/10.1371/journal.pone.0157605 (2016).
Article CAS PubMed PubMed Central Google Scholar
Wu, Y. W., Rho, M., Doak, T. G. & Ye, Y. Oral spirochetes implicated in dental diseases are widespread in normal human subjects and carry extremely diverse integron gene cassettes. Appl. Environ. Microbiol. 78, 5288–5296. https://doi.org/10.1128/AEM.00564-12 (2012).
Article CAS PubMed PubMed Central Google Scholar
Wu, Y. W., Doak, T. G. & Ye, Y. The gain and loss of chromosomal integron systems in the Treponema species. BMC Evol. Biol. 13, 16. https://doi.org/10.1186/1471-2148-13-16 (2013).
Article CAS PubMed PubMed Central Google Scholar
Treangen, T. J. & Rocha, E. P. C. Horizontal transfer, not duplication, drives the expansion of protein families in prokaryotes. PLoS Genet. 7, e1001284. https://doi.org/10.1371/journal.pgen.1001284 (2011).
Article CAS PubMed PubMed Central Google Scholar
Caceres, N. E. et al. Overexpression of the D-alanine racemase gene confers resistance to D-cycloserine in Mycobacterium smegmatis. J. Bacteriol. 179, 5046–5055 (1997).
Article CAS PubMed PubMed Central Google Scholar
Merker, M. et al. Whole genome sequencing reveals complex evolution patterns of multidrug-resistant Mycobacterium tuberculosis Beijing strains in patients. PLoS ONE 8, e82551. https://doi.org/10.1371/journal.pone.0082551 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Desjardins, C. A. et al. Genomic and functional analyses of Mycobacterium tuberculosis strains implicate ald in D-cycloserine resistance. Nat Genet 48, 544–551. https://doi.org/10.1038/ng.3548 (2016).
Article CAS PubMed PubMed Central Google Scholar
Curtiss, R., Charamella, L. J., Berg, C. M. & Harris, P. E. Kinetic and genetic analyses of d-cycloserine inhibition and resistance in Escherichia coli. J. Bacteriol. 90, 1238–1250 (1965).
Article CAS PubMed PubMed Central Google Scholar
Baisa, G., Stabo, N. J. & Welch, R. A. Characterization of Escherichia coli d-Cycloserine transport and resistant mutants. J. Bacteriol. 195, 1389–1399. https://doi.org/10.1128/jb.01598-12 (2013).
Article CAS PubMed PubMed Central Google Scholar
Karakousis, P. C. Antimicrobial Drug Resistance 271–283 (Humana Press, New York, 2009).
Book Google Scholar
Forsberg, K. J. et al. The shared antibiotic resistome of soil bacteria and human pathogens. Science 337, 1107–1111. https://doi.org/10.1126/science.1220761 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Sommer, M. O., Dantas, G. & Church, G. M. Functional characterization of the antibiotic resistance reservoir in the human microflora. Science 325, 1128–1131. https://doi.org/10.1126/science.1176950 (2009).
Article ADS CAS PubMed PubMed Central Google Scholar
Forsberg, K. J. et al. Bacterial phylogeny structures soil resistomes across habitats. Nature 509, 612–616. https://doi.org/10.1038/nature13377 (2014).
Article ADS CAS PubMed PubMed Central Google Scholar
Moore, A. M. et al. Pediatric fecal microbiota harbor diverse and novel antibiotic resistance genes. PLoS ONE 8, e78822. https://doi.org/10.1371/journal.pone.0078822 (2013).
Article ADS PubMed PubMed Central Google Scholar
Poole, L. B. The basics of thiols and cysteines in redox biology and chemistry. Free Radic. Biol. Med. 80, 148–157. https://doi.org/10.1016/j.freeradbiomed.2014.11.013 (2015).
Article CAS PubMed Google Scholar
Sevier, C. S. & Kaiser, C. A. Formation and transfer of disulphide bonds in living cells. Nat. Rev. Mol. Cell Biol. 3, 836–847 (2002).
Article CAS PubMed Google Scholar
Sanchez, R. et al. Protein structure modeling for structural genomics. Nat. Struct. Biol. 7(Suppl), 986–990. https://doi.org/10.1038/80776 (2000).
Article CAS PubMed Google Scholar
Park, I. S. & Walsh, C. T. D-Alanyl-D-lactate and D-alanyl-D-alanine synthesis by D-alanyl-D-alanine ligase from vancomycin-resistant Leuconostoc mesenteroides. Effects of a phenylalanine 261 to tyrosine mutation. J. Biol. Chem. 272, 9210–9214 (1997).
Article CAS PubMed Google Scholar
Evangelopoulos, D. et al. Comparative fitness analysis of D-cycloserine resistant mutants reveals both fitness-neutral and high-fitness cost genotypes. Nat. Commun. 10, 4177. https://doi.org/10.1038/s41467-019-12074-z (2019).
Article ADS CAS PubMed PubMed Central Google Scholar
Cowan, M. M. Plant products as antimicrobial agents. Clin. Microbiol. Rev. 12, 564–582 (1999).
Article CAS PubMed PubMed Central Google Scholar
Osonga, F. J. et al. Antimicrobial activity of a new class of phosphorylated and modified flavonoids. ACS Omega 4, 12865–12871. https://doi.org/10.1021/acsomega.9b00077 (2019).
Article CAS PubMed PubMed Central Google Scholar
Salam Pradeep, S. et al. Competitive inhibition of quercetin and apigenin at the ATP binding site of D-Alanine-D-Alanine Ligase of Helicobacter pylori: a molecular modeling approach. Curr. Biotechnol. 7, 340–348. https://doi.org/10.2174/2211550107666180612100441 (2018).
Article CAS Google Scholar
Hardy, K. G. Bacillus Cloning Methods In DNA Cloning 1–11 (IRL Press, New York, 1985).
Google Scholar
Bruning, J. B., Murillo, A. C., Chacon, O., Barletta, R. G. & Sacchettini, J. C. Structure of the Mycobacterium tuberculosis D-alanine:D-alanine ligase, a target of the antituberculosis drug D-cycloserine. Antimicrob. Agents Chemother. 55, 291–301. https://doi.org/10.1128/AAC.00558-10 (2011).
Article CAS PubMed Google Scholar
Daub, E., Zawadzke, L. E., Botstein, D. & Walsh, C. T. Isolation, cloning, and sequencing of the Salmonella typhimurium ddlA gene with purification and characterization of its product, D-alanine:D-alanine ligase (ADP forming). Biochemistry 27, 3701–3708 (1988).
Article CAS PubMed Google Scholar
Neuhaus, F. C. The enzymatic synthesis of D-alanyl-D-alanine. I. Purification and properties of D-alanyl-D-alanine synthetase. J. Biol. Chem. 237, 778–786 (1962).
CAS PubMed Google Scholar
Brooks, H. B. et al. in Assay Guidance Manual (eds G. S. Sittampalam et al.) (Eli Lilly & Company and the National Center for Advancing Translational Sciences, 2004).
Zhang, Y. & Skolnick, J. Scoring function for automated assessment of protein structure template quality. Proteins 57, 702–710. https://doi.org/10.1002/prot.20264 (2004).
Article CAS PubMed Google Scholar
Koes, D. R., Baumgartner, M. P. & Camacho, C. J. Lessons learned in empirical scoring with smina from the CSAR 2011 benchmarking exercise. J. Chem. Inf. Model. 53, 1893–1904. https://doi.org/10.1021/ci300604z (2013).
Article CAS PubMed PubMed Central Google Scholar
Jones, G., Willett, P., Glen, R. C., Leach, A. R. & Taylor, R. Development and validation of a genetic algorithm for flexible docking1. J. Mol. Biol. 267, 727–748. https://doi.org/10.1006/jmbi.1996.0897 (1997).
Article CAS PubMed Google Scholar
Jones, G., Willett, P. & Glen, R. C. Molecular recognition of receptor sites using a genetic algorithm with a description of desolvation. J. Mol. Biol. 245, 43–53. https://doi.org/10.1016/S0022-2836(95)80037-9 (1995).
Article CAS PubMed Google Scholar
Nissink, J. W. et al. A new test set for validating predictions of protein-ligand interaction. Proteins 49, 457–471. https://doi.org/10.1002/prot.10232 (2002).
Article CAS PubMed Google Scholar
Case, D. A. et al. The Amber biomolecular simulation programs. J. Comput. Chem. 26, 1668–1688. https://doi.org/10.1002/jcc.20290 (2005).
Article CAS PubMed PubMed Central Google Scholar
Darden, T., York, D. & Pedersen, L. Particle mesh Ewald: an N⋅log(N) method for Ewald sums in large systems. J. Chem. Phys. https://doi.org/10.1063/1.464397 (1998).
Article Google Scholar
Ryckaert, J.-P., Ciccotti, G. & Berendsen, H. J. C. Numerical integration of the Cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes. J. Comput. Phys. https://doi.org/10.1016/0021-9991(77)90098-5 (1977).
Article Google Scholar

Download references

Acknowledgements

We gratefully acknowledge the Commonwealth Scholarship Commission in the UK for supporting MAR (BDCA 2013-4), Public Health England for supporting SJ through a research grant to KMR (award code JGALASR) and the Erasmus programme for supporting MFM. We are thankful to Dr Haitham Hussain (UCL) for providing us with the pHCMC05 vector.

Author information

Marta Freitas Monteiro
Present address: Division of Immunology and Allergology, University Hospitals and Medical Faculty, Geneva, Switzerland

Authors and Affiliations

Department of Microbial Diseases, University College London, 256 Gray’s Inn Road, London, WC1X 8LD, UK
Md. Ajijur Rahman, Frank Kaiser, Marta Freitas Monteiro & Peter Mullany
Department of Pharmacy, University of Rajshahi, Rajshahi, 6205, Bangladesh
Md. Ajijur Rahman
School of Cancer and Pharmaceutical Science, King’s College London, London, SE1 9NH, UK
Shirin Jamshidi & Khondaker Miraz Rahman
Department of Tropical Disease Biology, Liverpool School of Tropical Medicine, Pembroke Place, Liverpool, L3 5QA, UK
Adam P. Roberts
Centre for Drugs and Diagnostics, Liverpool School of Tropical Medicine, Liverpool, UK
Adam P. Roberts

Authors

Md. Ajijur Rahman
View author publications
You can also search for this author in PubMed Google Scholar
Frank Kaiser
View author publications
You can also search for this author in PubMed Google Scholar
Shirin Jamshidi
View author publications
You can also search for this author in PubMed Google Scholar
Marta Freitas Monteiro
View author publications
You can also search for this author in PubMed Google Scholar
Khondaker Miraz Rahman
View author publications
You can also search for this author in PubMed Google Scholar
Peter Mullany
View author publications
You can also search for this author in PubMed Google Scholar
Adam P. Roberts
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

M.A.R.; Conceptualization, experimental, analysis, manuscript writing and editing. F.K.; experimental, analysis, manuscript editing. S.J.; experimental, analysis, manuscript editing. M.F.M.; experimental, analysis, manuscript editing. K.M.R.; experimental, analysis, manuscript editing. P.M.; analysis, manuscript editing. A.P.R.; conceptualization, analysis, manuscript writing and editing.

Corresponding author

Correspondence to Adam P. Roberts.

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

Supplementary Information 1.

Supplementary Video Legends.

Supplementary Video 1.

Supplementary Video 2.

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/.

Reprints and Permissions

About this article

Cite this article

Rahman, M.A., Kaiser, F., Jamshidi, S. et al. Integron gene cassettes harboring novel variants of d-alanine-d-alanine ligase confer high-level resistance to d-cycloserine. Sci Rep 10, 20709 (2020). https://doi.org/10.1038/s41598-020-77377-4

Download citation

Received: 28 May 2020
Accepted: 10 November 2020
Published: 26 November 2020
DOI: https://doi.org/10.1038/s41598-020-77377-4

This article is cited by

Inverse PCR-based detection reveal novel mobile genetic elements and their associated genes in the human oral metagenome
- Supathep Tansirichaiya
- Endre Winje
- Mohammed Al-Haroni
BMC Oral Health (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.