Analysis of antibiotics resistant genes in different strains of Staphylococcus aureus

The control of Staphylococcus aureus infection is being hampered by methicillin and other resistant strains. The identification of the unique antibiotic resistant genes from the genomes of various strains of S. aureus is of interest. We analyzed 11 S. aureus genomes sequences for Antibiotics Resistance Genes (ARGs) using CARD 2017 platform. We identified 32 ARGs across 11 S. aureus strains. Tet(38), norB, lmrB, mepA and mepR were present across genomes except for S. aureus strain UTSW MRSA 55. The mepA and mepR were found across 11 different genomes. However, FosB3, vgaALC, mphC and SAT-4 were found in UTSW MRSA 55, S.a. strain ISU935 and S.a. strain FDAARGOS_159. The prevalent mode of mechanism of antibiotics resistant was efflux pump complex or subunit conferring antibiotic resistance as well as protein(s). Analysis of norB, ImrB, norA, ImrB, tet (38), sav1866 and mecA have 12 to 14 TMHs. The results help in the understanding of Staphylococcus aureus pathogenesis in the context of antibiotic resistance.

One of the challenges confronting the treatment of S. aureus infection is resistance to many commonly used antimicrobial drugs [10,11]. When S. aureus was first discovered, it was easy to treat using available antibiotics. Some years after the introduction of penicillin in 1940 to combat S. aureus, there were strains of the pathogen that were resistant to these antibiotics. Methicillin was developed and introduced to treat penicillin-resistant S. aureus strains in 1961 [13]. The antibiotics, penicillin and methicillin mode of action is very similar and it involves inhibiting the synthesis of cell wall through the stoppage of peptidoglycan formation by the pathogen and finally lysis of the bacterium. In less than a year after the introduction of methicillin, strains of S. aureus were reported to be methicillin resistant and gradually these strains spread globally [14,15]. MRSA became a deadlier strain, which has become resistant to most β-lactam antibiotics [14]. It is known that certain genes are involved in the resistance to antibiotic drugs [14, 16,17], which have been transferred from a different bacterium, S. sciuri. It has been suggested that there may be unknown Antibiotic Resistance Genes (ARGs), which involves S. aureus resistance to antibiotics [4,18]. The recent sequencing of different S. aureus strains genomes as well as development of bioinformatics tools holds great promise for more efficient and high throughput in the identification and characterization of target genes [4,19]. The mechanisms by which these genes are involved in resistant to antibiotics could also be deduced. These genes could also provide insight into the pathogenesis and biology of the pathogen [19]. Therefore, the identification of the unique antibiotic resistant genes from the genomes of various S. aureus strains is of interest to deduce its mechanism.

Methodology: Genome Retrieval and Identification Analyses:
The complete genome of S. aureus sequences was downloaded from The National Center for Biotechnology Information (NCBI) Genome Database (https://www.ncbi.nlm.nih.gov/genome). The fasta file format of the genome sequence of 11 strains of bacteria were thoroughly analysed for Antibiotics Resistance Genes (ARGs) on the bulk analysis Resistance Gene Identifier (RGI) or CARD 2017 Platform (https://card.mcmaster.ca/ analyze/rgi) [20]. Default select criteria, which identified gene base on strict or perfect only was used. On the RGI platform each genome sequence file was uploaded and all settings were left at default. To have an inter-relation as well as qualitative and quantitative pattern of these ARGs in the various S. aureus strain, a heatmap chart was constructed using Microsoft Excel 2016 version for Mac. The methodology workflow is presented in Figure 1.

Physiochemical Properties:
Prediction of lipoprotein and secretory signal peptides in grampositive bacteria was done for each sequences using Pred-LiPo (http://bioinformatics.biol.uoa.gr/PRED-LIPO/), a web server that used Hidden Markov Models (HMM). This was validated using CW-Pred (http://bioinformatics.biol.uoa.gr/CW-PRED/), a tool that is also HMM based for the classification of cell wallanchored proteins of Gram-positive bacteria. LipoP (http://www.cbs.dtu.dk/services/LipoP/), SignalP (http://www.cbs.dtu.dk/services/SignalP/) and TargetP (http://www.cbs.dtu.dk/services/TargetP/) (all tools belong to the CBS prediction server) platform were also employed to validate the signaling properties of these proteins and subcellular location of proteins was done. SecretomeP was employed to perform non-classical and leaderless secretion of proteins. Serine and threonine phosphorylation sites in all the obtained antibiotic resistance genes were predicted using NetPhosBac 1.0 platform (http://www.cbs.dtu.dk/services/NetPhosBac/). The Serine and threonine phosphorylation sites were validated using GPS 3.0 Mac version downloaded from (http://gps.biocuckoo.org/online_full.php).

Phylogenetic analysis:
The antibiotic resistance genes in one Fasta file format were edited further using textEdit mac version prior to phylogenetics and evolutionary analyses on Molecular Evolutionary Genetics Analysis (MEGA) platform version 7.0 for Mac, obtained from http://www.megasoftware.net [21]. The 237 sequences were aligned using muscle tools with large alignment (Max iterations = 2) selected while other settings were left at the defaults. Evolutionary history was inferred using the Maximum Likelihood method. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (500 replicates) was also analysed. The tree was drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. All positions containing gaps and missing data were eliminated. The phylogenetics and evolutionary analyses were confirmed and validated using Phylogeny

Protein-Protein Interaction Networks:
Protein-protein interaction network was predicted for each of the ARGs on StringDB Version 10.5 (https://string-db.org) [23]. The sequences for each protein were used in the analyses and S. aureus was selected. The hit with the highest E value and bit score was selected for the final analysis. For each result, molecular action was selected under the setting. Proteins with weak interaction were excluded for further analyses.

Results:
Genomes of 11 strains of S. aureus that were retrieved from the NCBI are presented in Table 1. Most of the identified ARGs have high expectation (E) value and bit score, with mostly strict cut-off while few were perfect. However, 6 out of the 8 ARGs identified in the S. aureus strain UTSW MRSA 55 were perfect, while the other two were strict. The V521 and USA300_TCH959 SCAFFOLD2 strain of S. aureus with the accession number CP013957.1 and GG697986 had the highest and lowest number of base pairs. After the identification of acquired ARG by various bacteria genomes on the RGI platform, it was noticed that S. aureus strain V521 and V605 had 32 ARGs, which is the highest in the different strains genomes as presented in Figure 2. The S. aureus strain`s genome with the lowest ARGs are USA300_TCH959 SCAFFOLD2 and UTSW MRSA 55 that had 7 and 8 respectively. The heatmap chart in Figure 3 presents the relationship of the different S. aureus strains genomes studied in this work and the ARGs identified in each genome. We noticed that two ARGs that include norB and ImrB were identified in triplicate in 9 and 3 S. aureus strain`genomes respectively ( Figure 3). Hence, the norB gene was the most dominant gene in the different genomes studied. It was also noticed that five ARGs that include AN (9)-la, ErmA, ImrB, mecl and norB were identified in duplicate in 5, 5, 7, 2 and 1 S. aureus strain`genomes respectively ( Figure 3). The other ARGs identified were all single in each genome. While mphC, SAT-4 and vgaALC were the least common ARO genes identified across the eleven different S. aureus strains. The mphC and SAT-4 genes were identified only in S. a. strain FDAARGOS_159 while the vgaALC and FosB3 were identified only in S. a. ISU935 and UTSW MRSA 55 strains, respectively ( Figure 3 and Table 2).
The different novel Antibiotic Resistance Ontology based on the gene names are presented in Figure 3 and 4. All the genomes studied in this work had tet (38), norB, lmrB, mepA and mepR in common except the S aureus strain UTSW MRSA 55. It was noticed also that mepA and mepR were spread across the 11 different genomes. All the ARGs, except for FosB3, identified in the S. aureus strain UTSW MRSA 55 were shared by more than one other strains of S. aureus. Other shared ARGs are presented in Table 2. However, some ARO were unique to just one strain genome. The mphC and SAT-4 genes are unique to S. a. strain FDAARGOS_159 while vgaALC gene is unique to S. a. strain ISU935. The transmembrane helix (TMH) prediction shows that norB, ImrB, norA, ImrB, tet (38), sav1866 and mecA that have 12 and 14 TMHs (Figure 4). While mecRi, arls and mepA have fewer than 5 TMH. The ARGs are mostly identified as perfect in S. aureus strain UTSW MRSA 55. Almost all the ARGS were identified with protein homolog model. While only antibiotic resistant fabI and S. aureus gyrA conferring resistance to fluoroquinolones were identified with protein variant model. The PC1 beta-lactamase (blaZ) is the only gene that was predicted to have signal peptide cleavage sites in the different identified ARO genes from the various genomes. Localization prediction shows that most of the identified genes in this work are either membrane, cytoplasmic or lipoprotein. We also noticed that most of these genes have more than one phosporylation sites.
The norB, vgaALC, mepA, lmrB and other genes have a single antibiotics resistance mechanism. While mecA, mecR1, arlS, arlR, PC1 beta-lactamase (blaZ), mecI and others have more than one antibiotic resistance mechanisms. The most prevalence mode of mechanism of antibiotic resistance are efflux pump complex or subunit conferring antibiotic resistance as well as protein(s) (norA, norB, sav1866, ImrB, arlB, mgrA, TaeA, tet 38, mepR, arls, vgaALC, bcrA and tet K) and two-component regulatory system modulating antibiotic efflux (mepR, arlR, mgrA and arls). Therefore we noticed that ARGs that have two-component regulatory system modulating antibiotic efflux also carry out the antibiotic resistance are efflux pump complex or subunit conferring antibiotic resistance. Other mechanisms include; antibiotic resistance gene cluster, cassette, or operon [MecA, mecl, mecR1 and PC1 beta-lactamase (blaZ)] and antibiotic inactivation enzyme (ANT(9)-Ia, AAC(6')-Ie-APH(2'')-Ia, SAT-4 and mphC). ARG variant or mutant were also obtained for antibiotic resistant fabI and S. aureus gyrA conferring resistance to fluoroquinolones genes alone. These ARGs that were identified to have variant or mutant were also the ones that possess Single Nucleotide Polymorphisms (SNP); G93A and S85P for antibiotic resistant fabI and gyrA. After thorough analyses and extruding proteins with no hit and poor interaction network on the stringDB, the following ARGs; arlR, arlS, gyrA and Tet M were selected for further discussion. The protein-protein interaction networks are presented in Figure  6a-c. The arlS and arlR gene dependently regulate and modulate other genes as presented in Figure 6a to carry out its resistance action on antibiotic compounds. The gyrA demonstrated unspecific reaction and binding on other genes such as gyrB, dnaN and pare as well as positive activation of the gyrB to carry out antibiotic resistance (Figure 6b). Tet (M) also demonstrated unspecific reaction and binding on a wide range of genes presented in Figure 6c to be involved in antibiotic resistance.    The major mechanisms of antibiotic resistance identified in this work are the efflux pump complex or subunit conferring antibiotic resistance. This mechanism enhances efflux through overexpressed pumps is for bacterial pathogens such as S. aureus by which efficiently extrude antimicrobial drugs outside the cell [31,32]. These transporters can extrude a wide range of unrelated compounds, which can lead to multidrug-resistant (MDR) [33]. This efflux of drugs that are shown by S. aureus was first discovered in Escherichia coli [33]. Some bacterial pathogens such as E. coli and other gram-negative bacteria also employed this mechanism to play other roles in pathogenicity of bacteria such as colonization, infection, and the persistence of microorganisms in the host. The antibiotic resistance gene cluster mechanism we found in S. aureus strains has been investigated in Vibrio cholerae Our findings also identified other ARGs that are proposed to be associated with multidrug efflux pump in S. aureus efflux systems. Some of these ARGs are NorA, NorB, MepA. NorA and NorB are chromosomal genes that belong to the MFS and demonstrated some genetic diversity [33,41]. The NorA and NorB is multidrug efflux pump, which belongs to the major facilitator transporter. NorA has a higher percent identity with other resistance genes in other bacterial pathogens such as Bmr from Bacillus subtilis and Tet (A) from Escherichia coli [42]. However, Bmr was not identified in the work. NorA is mostly expressed on the membrane that has an active efflux pump of a hydrophilic molecule such as quinolones [31]. NorB acts irrespective of NorA to carry out resistance against a wide range of quinolone compounds and other antibiotic agents [43]. The norB gene is also one of the best-studied multidrug efflux pumps that play a vital role in fluoroquinolone resistant in diverse strains S. aureus [33,44]. Strains  These two genes are fully interconnected and regulate the expression of another AGR such as NorA, (which was also identified and discussed in this work) [54]. Their interconnection has been shown to negatively and positively regulate the Agr; virulence accessory gene, and SarA; staphylococcal accessory regulator, respectively. This regulation has been investigated to modulate several virulence factors such as serine protease, surface protein and alpha-hemolysin [55]. Research has shown that ArlS may probably act as a sensor protein at a histidine residue and transfers its phosphate group to ArlR [56].
One of the identified AGRs across the diverse strains studied and also in MRSA is the MepA gene. It is an efflux pump gene that belongs to the multidrug and toxin extrusion (MATE) transport protein family [33,57]. Our finding shows that MepA have 12 TMHs and this is in accord with other researchers. The sav1866 ARGs was also identified in MRSA strains and is a multidrug transporter belonging to the ABC family. Our study shows that sav1866 have 5 TMHs. These TMHs could be part of the pore that triggers by ATP binding that is presumably the drug translocation pathway [49,58]. The last ARG identified in this work that will be discussed is the gyrA, which is, involved in S. aureus antibiotic resistance to fluoroquinolones. S. aureus strains such as UCI 28, NCCP14562 and NCCP14558 were identified to express the gyrA, hence these strains will have resistance to fluoroquinolones antibiotics [59]. However, the strains could be susceptible to antibiotics such as Besifloxacin, tosufloxacin and structurally other similar compounds [45].

Conclusion:
We have identified about 32 genes that could have serious implication in antibiotic resistance. Phylogenetics analyses show the relationship of these ARGs. S. aureus are interconnected in function when one or more other genes are expressed. The strain with serious clinical implication on human and animal health is the S. aureus strain UTSW MRSA 55 expressing eight different ARGs. Seven of these genes are also expressed by other S. aureus strains. However, FosB3 is unique in MRSA strains. ARG are resistant to certain antibiotics. They are susceptible to several antibiotics in some strains. The results are helpful in S. aureus clinical surveillance in the context of antibiotic resistance.