Emergence of Raoultella ornithinolytica in human infections from different hospitals in Ecuador with OXA-48-producing resistance

Abstract

Purpose:

The purpose of this study was to highlight the clinical and molecular features of 13 Raoultella ornithinolytica strains isolated from clinical environments in Ecuador, and to perform comparative genomics with previously published genomes of Raoultella spp. As Raoultella is primarily found in environmental, clinical settings, we focused our work on identifying mechanisms of resistance that can provide this bacterium an advantage to establish and persist in hospital environments.

Methods:

We analyzed 13 strains of Raoultella ornithinolytica isolated from patients with healthcare associated infections (HAI) in three hospitals in Quito and one in Santo Domingo de Los Tsáchilas, Ecuador, between November 2017 and April 2018. These isolates were subjected to phenotypic antimicrobial susceptibility testing, end-point polymerase chain reaction (PCR) to detect the presence of carbapenemases and whole-genome sequencing.

Results:

Polymerase chain reaction revealed that seven isolates were positive isolates for bla_OXA–48 and one for bla_KPC–2 gene. Of the seven strains that presented the bla_OXA–48 gene, six harbored it on an IncFII plasmid, one was inserted into the bacterial chromosome. The bla_KPC gene was detected in an IncM2/IncR plasmid. From the bioinformatics analysis, nine genomes had the gene bla_OXA–48, originating from Ecuador. Moreover, all R. ornithinolytica strains contained the ORN-1 gene, which confers resistance for β-lactams, such as penicillins and cephalosporins. Comparative genome analysis of the strains showed that the pangenome of R. ornithinolytica is considered an open pangenome, with 27.77% of core genes, which could be explained by the fact that the antibiotic resistance genes in the ancestral reconstruction are relatively new, suggesting that this genome is constantly incorporating new genes.

Conclusion:

These results reveal the genome plasticity of R. ornithinolytica, particularly in acquiring antibiotic-resistance genes. The genomic surveillance and infectious control of these uncommon species are important since they may contribute to the burden of antimicrobial resistance and human health.

1. Introduction

Raoultella species are gram-negative encapsulated bacilli belonging to Enterobacteriaceae (Drancourt et al., 2001). Until 2001, Raoultella species were considered part of the genus Klebsiella; however, with the current advances in molecular analysis based on rpoB sequencing, Raoultella was classified as a distinct and unique genus (Boye and Hansen, 2003; Martínez et al., 2004). Species such as R. terrigena, R. planticola, R. electrica, R. trevisani, and R. ornithinolytica belong to this genus. R. ornithinolytica is the most important because it has been associated with symptomatic cases of bacteremia (Kanki et al., 2002; Mau and Ross, 2010; Haruki et al., 2014; Tafoukt et al., 2017; Avellaneda et al., 2020; Wyres et al., 2020), urinary tract infections (García-Lozano et al., 2013; Haruki et al., 2014), joint infections (Beye et al., 2018), and biliary tract infections (Cleveland et al., 2014).

Raoultella ornithinolytica, is considered an unusual microorganism in health settings (Reyes et al., 2020). These bacteria are environmental microorganisms found in water, soil, and plants. For many years, this species have not been considered harmful to humans (Kanki et al., 2002). However, it has been found that some R. ornithinolytica strains may harbor antibiotic-resistance genes, such as bla_NDM–1, bla_OXA–48, and bla_OXA–232 in the environment (Iovleva et al., 2019), constituting a possible route of transmission and spread of antimicrobial resistance genes (ARGs) through mobile elements (horizontal gene transfer) (Tafoukt et al., 2017; Weng et al., 2020; Zou et al., 2020). A recent report from Croatia, described a case of septicemia in a 64-year-old male patient, caused by R. ornithinolytica and Klebsiella pneumoniae, both associated with antibiotic resistance and presence of bla_OXA–48 gene, which contributed to severity of infection and course of antibiotic treatment.

The correct identification of R. ornithinolytica is one of the main challenges in clinical settings. R. planticola and R. ornithinolytica share from of 98.3 to 99.5% of their genome content, which leads to form a tight phylogenetic cluster (Bravo, 2006; Walckenaer et al., 2008). Since most clinical laboratories relie on routine automated systems, high rates of misidentification have been reported, (Park et al., 2011; Sȩkowska et al., 2018). These systems are not sensitive enough to discriminate one species from the other through the ornithine decarboxylase (ODC) assay (Alves et al., 2006). In the absence of a robust biochemical assay and specific genetic markers allowing to detect of differences between these two species, the application of proteomic and genomic tools with whole-genome sequencing (WGS) have aided to accurately and rapidly discriminate at the species level. WGS allows us to have a greater number of genetic elements to differentiate them, but also to assess or infer functionality in terms of antimicrobial resistance (Galata et al., 2018).

In this study, we performed a comparative genomic analysis of 13 R. ornithinolytica strains isolated from different Ecuadorian hospitals identified by Whole Genome Sequencing (WGS). This comparative analysis revealed that the 13 bacterial strains correspond to different upsurge.

2. Materials and methods

2.1. Bacterial strains

We received 13 Raoultella spp. strains from Hospital Eugenio Espejo–HEE, Hospital Militar–HMI, Hospital Carlos Andrade Marín–HCA, and Hospital Gustavo Dominguez—HGD. These samples were received at the National Reference Center for Antimicrobial Resistance (CRN-RAM) Dr. Leopoldo Izquieta Pérez in Quito. The origin of the samples is shown in Supplementary Table 1. These samples have a similar phenotype to those at the Antimicrobial Resistance Reference Center as part of the national surveillance program which were isolated in Ecuador between November 2017 and April 2018.

The Antimicrobial Resistance (AMR) surveillance, the National Antimicrobial Resistance Reference Center (CRN-RAM) has defined a list of microorganisms with antimicrobial susceptibility patterns that are included in epidemiological surveillance actions in different hospitals in Ecuador. In this case, the hospitals reported the presence of resistance to carbapenems and susceptibility to third-generation cephalosporins, suggesting that this observation could be attributed to resistance mediated by bla_OXA–48 carbapenemase. Consequently, epidemiological alert control measures were implemented, indicating the possible presence of an outbreak with a common source not identified, which was subsequently investigated through phenotypic observation in the laboratory followed by PCR and genomic sequencing. Following the last reported case of Raoultella ornithinolytica, hospitals conducted active surveillance for 6 months, during and no further cases were detected.

Isolates were recovered from BD CultureSwab MAXV smears with Stuart liquid medium by plating them on BD Difco nutrient agar and MacConkey agar. Plates were incubated at 37°C for 24 h. Pure colonies were selected for bacterial identification and antimicrobial susceptibility using VITEK^®2 GN cards and the VITEK^®2 Compact Microbial Detection System (BioMerieux Inc., France) according to the manufacturer’s recommendations.

2.2. Genome sequencing and assembly

Genomic DNA was extracted and purified using the Wizard Genomic Purification Kit (Promega). DNA extracts were used to perform WGS at the Malbran Institute in Argentina by the Illumina Miseq platform using the Nextera XT DNA library preparation kit (Illumina^®, San Diego, CA, USA). Libraries were generated with a length of 300 bp paired-end reads. The quality control analysis of reads was performed using FastQC software v.011.9 (“Babraham Bioinformatics, 2022–FastQC A Quality Control Tool for High Throughput Sequence Data” n.d.). Subsequently, reads were trimmed using TrimGalore v.0.6.7 (Krueger et al., 2021) with Q>20. The reads were assembled into contigs using SPAdes v.3.13.1 (Bankevich et al., 2012) with the default settings, and the assembly statistics were performed with QUAST v5.0.2 (Gurevich et al., 2013).

2.3. Genome identification, annotation, and phylogenomic tree construction

Bacterial identification was performed using the criterion Average Nucleotide Identity (ANI), using PyANI v.0.2.12 with default parameters (Pritchard et al., 2016). Briefly, ANI is defined as the mean nucleotide identity of orthologous gene pairs shared by two microbial genomes considered one genome >95% ANI as the same species (Barco et al., 2020). We retrieved 41 external complete genomes of the genus Raoultella from the NCBI portal (retrieved October 9, 2022), as well as 5 genomes of Klebsiella as controls for this analysis. A total of 59 genomes were included in the ANI analysis. The assembled genomes were annotated using Prokka v.1.14.6 (Seemann, 2014) and then compared for size and similarity to the reference sequence using BLAST BRIG ring imager v.0.95 (Alikhan et al., 2011).

For phylogenomic reconstruction, a set of 92 bacterial core genes was retrieved from the 59 genomes and aligned using Up-to-date Bacterial Core Gene Software (UBCG) v.3.0 (Na et al., 2018). A maximum-likelihood (ML) tree was estimated with IQ-TREE v2.0.3 (Nguyen et al., 2015) based on 1,000 bootstrap replicates. The five genomes of Klebsiella were used as the phylogenetic tree root.

2.4. Core and pan-genome analysis

Core and pan-genome analysis were performed using only R. ornithinolytica genomes, including our 13 Ecuadorian genomes and 22 complete genomes from NCBI (Supplementary Table 1). The 35 genomes were analyzed using Roary (Page et al., 2015), with 95% identity for blastp and a strict definition of nuclear genome (i.e., 100% of isolates with core genes) (Page et al., 2015).

2.5. Antimicrobial susceptibility test, identification, and search for antimicrobial resistance genes and plasmids using bioinformatics

Susceptibility testing was performed using a VITEK-2^® AST N272 card (BioMerieux Inc., France). The modified carbapenem inactivation method (mCIM) was performed according to the Clinical and Laboratory Standards Institute (CLSI) to evaluate the presence and expression of enzymatic antimicrobial resistance mechanisms. Molecular analysis by PCR was performed to identify the following ARGs: bla_KPC, bla_NDM, bla_VIM, bla_IMP, and bla_OXA–48, according to previously published protocols described by Poirel et al. (2011) (Table 1). We used the Comprehensive Antibiotic Resistance Database (CARD) in Resistance Gene Identifier (RGI) v.5.1.0 to search ARGs (Alcock et al., 2020), and PLACNETw (Vielva et al., 2017) to search for plasmids, and circular plasmids were visualized through BRING v.0.95, in previously assembled genomes.

Incompatibility groups (Inc.) and pMLST subtypes were determined in silico using PlasmidFinder and pMLST (Carattoli and Hasman, 2020), and ccfind (Nishimura et al., 2017) was used to determine the circulated genome.

2.6. Gain and loss genes from ancestral reconstruction in R. ornithinolytica

We performed an ancestral reconstruction analysis to estimate genetic gain and loss events. We used the Wagner parsimony model with flexible gain-loss ratios for all lineages and a Poisson distribution at the root of the tree in the program COUNT (Csurös, 2010). The number of orthologs in each genome was determined using Proteinortho, with a threshold of 50% for identity and coverage (Lechner et al., 2011).

3. Results

3.1. Bacterial strains

From November 2017 to April 2018, a proportional increase in Raoultella spp. infections were reported in tertiary hospitals in Quito and Santo Domingo de los Tsachilas, Ecuador; The strains were identified in NRLAR using the VITEK^®2 GN card compact system (BioMerieux Inc., France) and confirmed as R. ornithinolytica by sequencing of the rpoB gene.

3.2. Genomic features of the Raoultella ornithinolytica isolated

Whole-genome sequencing sequences of the isolated Raoultella strains were sequenced individually and then merged. The general characteristics of the 13 assembled genomes are summarized in Table 2. The size of the genomes ranged from 5.5 Mb to 5.9 Mb. The average GC content of each genome was approximately 55.6% (Figure 1). The average number of coding sequences (CDS) in each genome was 5,485 pb, as shown in Table 2. Taxonomic identification was based on average nucleotide identity (ANI), and whole genome comparison results showed that all 13 genomes were nearly 99% concordant with the 22 R. ornithinolytica genome references (Figure 2). These results indicate that the isolates from different hospitals in Ecuador corresponded to R. ornithinolytica.

FIGURE 1

FIGURE 2

A phylogenetic tree was constructed using a set of 92 bacterial conserved genes to infer the phylogenetic relationship between our 13 genomes of R. ornithinolytica, and other 46 complete genomes of the genera Raoultella and Klebsiella retrieved from the NCBI portal. There are five main clades, as shown by the different colors in Figure 3. Clade I for R. ornithinolytica, Clade II for R. planticola, Clade III for Raoultella spp., Clade IV for R. terrigena, and Clade V for Klebsiella spp. In Clade I, nine out of thirteen Ecuadorian genomes of R. ornithinolytica were grouped together in a well-supported subclade, suggesting the idea of a common origin. The remaining four genomes, came from a single healthcare setting and seem to be dispersed throughout different subclades within Clade I, which suggests independent introduction events. In contrast, strains scattered in the tree may not belong to the same outbreak despite the geographical distance between the hospitals.

FIGURE 3

3.3. Pan-genome comparative analyses of R. ornithinolytica genomes

The pan-genome of the 35 genomes analyzed is shown in Figure 4. A total of 16,661 gene clusters (orthologs) were found, of which 3,467 genes (27.77%) were assigned to the core genome, 548 to the soft-core genome (3.29%), and 10,975 to the unique genome (65.87%) (Figure 4A). To investigate if the R. ornithinolytica pan-genomic is open, which it will suggest and increase in genome size due to the addition of new genomes, Heap’s law was applied (Figure 4B). The trend of Heap’s law diagram for the pangenome shows gradual expansion due to the addition of new genomes, with the slope continuing to increase. The curve shows the relationship between the core genome and the number of genomes, which decreased and did not exceed 5,000 genes. Thus, the pangenome of R. ornithinolytica is reflected in the openness of the pan-genome, which could indicate high genomic plasticity of this species and, thus, the greater potential of adaptability.

FIGURE 4

3.4. Antimicrobial resistance genes and plasmids

The results of antimicrobial susceptibility testing showed three different resistance phenotypic patterns (Figure 5 and Supplementary Table 2). The first includes resistance to ampicillin/sulbactam (SAM), piperacillin/tazobactam (TPZ), ertapenem (ETP), imipenem (IMP), and intermediate resistance to meropenem (MEM). The second pattern shows, resistance to SAM, TPZ, and ETP, and intermediate resistance to IMP. And the third one shows the resistance to SAM and TPZ exclusively. No specific antimicrobial resistance patterns were observed within hospitals. Therefore, the phenotypic antimicrobial resistance patterns observed in the first, second, and third groups have no consistent association with any particular hospital, as described above.

FIGURE 5

Seven R. ornithinolytica strains showed interesting patterns of at least two out of three carbapenems testing resistance, but appearing susceptible to third and fourth-generation cephalosporins (ceftazidime, ceftriaxone, and cefepime), which is a consistent phenotypic profile, indicating the possible presence of a bla_OXA–48 carbapenemase, which was confirmed by them CIM assay and PCR (Figure 5).

In addition, bioinformatic analysis of antimicrobial resistance was performed using all the genomes collected using RGI software (Alcock et al., 2020; Figure 3). The heatmap results showed bit scores using BLAST with RGI values Perfect (green), Strict (blue), and Losse (gray) (Figure 3B). Among the genes predicted with a perfect bitscore with RGI analysis was the ORN-1 gene, which encoded for an Ambler class A beta-lactamase conferring resistance to penicillins and cephalosporins (Walckenaer et al., 2008). ORN-1 gene was found in all of our isolates and in 14 of the R. ornithinolytica genomes retrieved from NCBI for comparison purposes. The Ecuadorian genomes grouped in Clade I (Figure 3A) positive for ORN-1, also carried the mphA gene, which confers resistance to macrolides (Li et al., 2022) and the bla_OXA–48 gene, which confers resistance to carbapenems; however, only seven of these isolates expressed a phenotypic resistance to carbapenems.

In addition, genes conferring resistance to aminoglycosides, beta-lactams, sulfates, tetracyclines, and dihydrofolate, such as acrB, acrD, mdtC, marA, mdtB, and adeF were present in all the Ecuadorian isolates. Nevertheless, the diversity of ARGs showed among the Ecuadorian isolates was lower when compared with the ARGs diversity from other R. ornithinolytica strains (Figure 3). Only one Ecuadorian strain (HEE0406) harbored additional antimicrobial resistance genes (SHV1.2, tetA, and fosA).

We have identified genes associated with antibiotic resistance in 58 isolates, and the gene fosA was also found in 56 isolates.

Finally, plasmid analyses revealed that the Ecuadorian R. ornithinolytica strains contained segments of plasmids. The partial plasmid containing the bla_OXA–48 gene (22 Kb), is found in HEE0315, HEE0686, HEE0463, HEE0314, and HCA1847 strains (Figure 6). Unfortunately, the complete plasmid could not be obtained, but based on previous literature and based on our knowledge, we believe that this gene could be responsible for the resistance in R. ornithinolytica.

FIGURE 6

3.5. Gain and loss of genes from ancestral reconstruction in R. ornithinolytica

Since our results revealed that the pangenome of R. ornithinolytica is open, the gain or loss of genes during their evolutionary history is expected. To this end, we performed an ancestral reconstruction analysis using WGS (Figure 7). The genomes examined shared over 4,000 protein families with their common ancestor Klebsiella/Raoultella (KRCA), and the reconstruction analysis predicted more gain than loss events.

FIGURE 7

Our analysis revealed that six protein families associated with antibiotic resistance were acquired from five predicted common ancestors (Figure 7). We hypothesize that the most remote ancestor, KRCA, harbor genes that encode an efflux pump providing resistance to fluoroquinolones (Poole, 2000), and that these genes could have been transmitted as orthologs to the analyzed genomes, except for the K. huaxiensis WCHKl090001 which lost that protein family. Our second hypothesis is that the common ancestor of Raoultella (RCA) acquired genes associated with β-lactamase resistance ORN-1/PLA-1, which were not subsequently lost. The common ancestor of R. terrigena (RTCA) acquired genes conferring β-lactamase resistance TER −2/DHA −1 and passed vertically to strains in this group.

In addition, the common ancestor of R. ornithinolytica WP8-W19-CRE-01 and R. ornithinolytica Ro24724 acquired genes for a pentapeptide repeat protein and aminoglycoside acetyltransferases related to fluoroquinolone and aminoglycoside resistance, respectively (Vetting et al., 2006; Reeves et al., 2020). Even though they have been isolated from different sources in China. Strain WP8-W19-CRE-01 was obtained from wastewater, whereas strain Ro24724 was obtained from necrotic tissue.

The last common ancestor with resistance protein families was the ancestor of clinical strains HEE0686, HGD0386, HCA1847, HEE0314, HMI0367, HCA1697, HEE0315, HMI0464, and HEE0463. This ancestor acquired two protein families related to the multidrug efflux pump Tap, and a hypothetical protein related to macrolide resistance. This ancestor may also have gained OXA-48, which was only present in this clade (Figure 3). In general, there are other gains/losses occurring in the genomes, but these do not present traceable ancestors in the tree because changes may have occurred recently.

4. Discussion

Raoultella ornithinolytica is a gram-negative bacterium isolated from various environments, including water and soil, as well as from various animals, including birds, humans, and other mammals (Seng et al., 2016). R. ornithinolytica has been shown to cause human diseases; however, its incidence may be underestimated because of the high misidentification rate of automated systems based on phenotypic characteristics (Sȩkowska et al., 2018). The Raoultella genus has been identified as Klebsiella oxytoca (Walckenaer et al., 2008; Park et al., 2011; Sȩkowska et al., 2018).

To overcome these limitations, Whole Genome Sequencing (WGS) has been used to identify R. ornithinolytica with greater accuracy, allowing the identification of genes that may be important for bacterial metabolism and pathogenicity (Matar et al., 2020). In Ecuador, WGS was used to identify Raoultella isolates, and the results showed that 13 bacteria isolated were R. ornithinolytica. The ANI, and taxonomic position resulted in a phylogenetic tree analysis with 92 marker genes, which showed that the strains belonged to R. ornithinolytica (Matar et al., 2020).

In the phylogenetic tree, nine strains from Ecuador formed a cluster, suggesting that these strains had greater genomic and phenotypic similarities than the other four strains scattered independently in the tree. This suggests that the first nine strains belonged to a single outbreak. The genomic characteristics of the 13 isolated genomes found in Ecuadorian hospitals are summarized in Table 2. The genomes reported in this study had a 1% size difference average variation among themselves, suggesting lost and gain events.

The phylogenetic clustering of strains within the observed phylogenetic tree indicates a potential localized transmission event or a shared exposure to a common source of infection, despite being identified in different hospitals. The clustering pattern strongly suggests the likelihood of an outbreak, which appears to have occurred within a specific time frame spanning from November 2017 to April 2018. Following this period, an active surveillance program was implemented for an additional 6 months, during which screening efforts were undertaken to identify strains exhibiting the same phenotypic resistance pattern for subsequent genomic sequencing. However, no strains with the identified resistance pattern were detected beyond this surveillance period.

On the other hand, the core genome was found to include 33% of the pangenome of all the R. ornithinolytica assemblies, using a high identity value (≥95% identity). In other species, small pan-genome size indicates a process of speciation/adaptation (Zhang et al., 2018). As more genomes were examined, the total number of genes observed in the analysis also increased, as shown in Figure 4B. However, this pan-genomic feature allows different strains to survive in various environmental niches (Medini et al., 2005; McInerney et al., 2017). The core genome of R. ornithinolytica is small compared to close bacteria, such as Streptococcus. pneumoniae (55.06%), Staphylococcus aureus subsp. aureus (75.23%), Escherichia coli (53.34%), Salmonella enterica subsp. enterica (71.23%), and Acinetobacter baumannii (61.36%) (Park et al., 2019).

Furthermore, results of antibiotic resistance detected by PCR showed seven positive isolates for bla_OXA–48 and one for bla_KPC–2, while bioinformatic analysis revealed nine genomes from Ecuador. Carbapenem resistance in aquatic environments is caused by bla_NDM, bla_OXA–48, and bla_KPC–2 in R. ornithinolytica (Hajjar et al., 2020). The emergence of carbapenem-resistant K. pneumoniae strains in Ecuador have increased recently. Indeed, OXA-48 carbapenemase has been detected in K. pneumoniae isolates from clinical samples (Reyes et al., 2020). In another case, a genome of R. ornithinolytica from Ecuador was reported to harbor the bla_{OXA–48–like} gene (Reyes et al., 2020). Given the potential for multidrug resistance and the ability of R. ornithinolytica to survive in different environmental niches, it is important to continue surveillance to better understand its potential to spread antibiotic resistance.

We conducted ancestral reconstruction on all the genes in the Raoultella genome and identified genes that are unique to this species. In particular, we paid close attention to resistance genes to determine if they were present in the ancestral lineage. However, our analysis provides evidence to suggest that some resistance genes were ancestral, but many others are relatively recent in origin. An instance of this is the OXA-48-β-lactamase family, ORN-1/PLA-1, and fluoroquinolone resistance genes found in Raoultella genomes, which are likely acquired through horizontal gene transfer and not ancestral. This is consistent with the open pangenome of R. ornithinolytica. Ancestral reconstruction analysis reveals that each species in the genus has gained genes recently, potentially via gene transfer, indicating a worrying trend of rapid acquisition and dissemination of antibiotic resistance genes.

A gene harbored by the common ancestor of the R. ornithinolytica hospital clade, important for antibiotic resistance, was the IS5 family transposase IS4811, also found in some Klebsiella and R. terrigena assemblies. This transposase is important for the transfer of genetic material between bacteria, and it can contribute to the spread of antibiotic resistance genes among bacterial populations (Vandecraen et al., 2017).

Also, this ancestor contains the Multidrug efflux pump Tap, and a gene encoding for a hypothetical protein related to resistance to macrolides, suggesting that these genes are important for antibiotic resistance in this clade of bacteria. Multidrug efflux pump Tap and the hypothetical protein genes were also present in other genomes, such as R. ornithinolytica STEFF_15, R. planticola JBIWA001, and R. planticola Rp_CZ180511, suggesting that they may have been horizontally transferred between different Raoultella species.

The presence of the β-lactamase ORN-1/PLA-1 protein family since the Raoultella common ancestor, and its acquisition by K. huaxiensis WCHKl090001, suggests that this gene family is important for the survival of these bacteria in the presence of β-lactam antibiotics.

The β-lactamase TER-2, a β-lactamase DHA-1, is common to all R. terrigena, and according to Figure 7, it has orthologs in Raoultella sp. XY-1, R. planticola Rp_CZ180511 and K. aerogenes Ka37751. These genes likely provide these bacteria with resistance to β-lactam antibiotics, which are commonly used to treat bacterial infections. The acquisition of these resistance genes may have facilitated the spread of antibiotic resistance among different bacterial species, which is a major public health concern (Pérez-Llarena et al., 2014).

The only common ancestor with an aminoglycoside resistance family was found between the R. ornithinolytica WP8-W19-CRE-01 and R. ornithinolytica Ro24724 genomes, including the gene for Aminoglycoside N(6′)-acetyltransferase type 1. The other aminoglycoside resistance gene families (aac, aad, arm) were gained by the strains.

The common ancestor for Raoultella and Klebsiella has one fluoroquinolone resistance protein family, encoding for Efflux pump periplasmic linker BepF, related to this activity (Li and Nikaido, 2009). All the reported Raoultella genomes from Ecuador keep these genes. The Raoultella sp. XY-1 gained two copies of this gene and, like some other later characters, acquired the repeat protein Qnr (Figure 3).

In conclusion, continued monitoring and research of R. ornithinolytica are necessary to better understand its potential for multidrug resistance. This study highlights the importance of using Whole Genome Sequencing as a valuable tool in identifying R. ornithinolytica with greater accuracy and overcoming the limitations of automated systems based on phenotypic characteristics. Finally, the study emphasizes the emergence of multidrug-resistant enterobacteria, including R. ornithinolytica, which raises concerns regarding the spread of antibiotic resistance and the potential for horizontal gene transfer.

References

• 1

  AlcockB.RaphenyaA.LauT.TsangK.BouchardM.EdalatmandA.et al (2020). CARD 2020: Antibiotic resistome surveillance with the comprehensive antibiotic resistance database.Nucleic Acids Res.48D517–D525. 10.1093/nar/gkz935

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 2

  AlikhanN.PettyN.Ben ZakourN.BeatsonS. (2011). BLAST ring image generator (BRIG): Simple prokaryote genome comparisons.BMC Genom.12:402. 10.1186/1471-2164-12-402

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 3

  AlvesM.DiasR.de CastroA.RileyL.MoreiraB. (2006). Identification of clinical isolates of indole-positive and indole-negative Klebsiella spp.J. Clin. Microbiol.443640–3646. 10.1128/JCM.00940-06

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 4

  AvellanedaH.ArbeliZ.TeranW.RoldanF. (2020). Transformation of TNT, 2,4-DNT, and PETN by Raoultella planticola M30b and Rhizobium radiobacter M109 and exploration of the associated enzymes.World J. Microbiol. Biotechnol.36:190. 10.1007/s11274-020-02962-8

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 5

  Babraham Bioinformatics (2022). FastQC A quality control tool for high throughput sequence data. Available online at: http://www.bioinformatics.babraham.ac.uk/projects/fastqc/(accessed Oct 13, 2022).

  • Google Scholar

• 6

  BankevichA.NurkS.AntipovD.GurevichA.DvorkinM.KulikovA.et al (2012). SPAdes: A new genome assembly algorithm and its applications to single-cell sequencing.J. Comput. Biol.19455–477. 10.1089/cmb.2012.0021

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 7

  BarcoR.GarrityG.ScottJ.AmendJ.NealsonK.EmersonD. A. (2020). Genus definition for bacteria and archaea based on a standard genome relatedness index.mBio11:e02475–19. 10.1128/mBio.02475-19

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 8

  BeyeM.HasniI.SengP.MichelleC.La ScolaB.RaoultD.et al (2018). Genomic analysis of a Raoultella ornithinolytica strain causing prosthetic joint infection in an immunocompetent patient.Sci. Rep.8:9462. 10.1038/s41598-018-27833-z

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 9

  BoyeK.HansenD. (2003). Sequencing of 16S rDNA of Klebsiella: Taxonomic relations within the genus and to other Enterobacteriaceae.Int. J. Med. Microbiol.292495–503. 10.1078/1438-4221-00228

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 10

  BravoF. (2006). “Faculty opinions recommendation of sequencing of 16S rDNA of klebsiella: Taxonomic relations within the genus and to other Enterobacteriaceae,” in Faculty opinions – post-publication peer review of the biomedical literature.10.3410/f.1044055.493961

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 11

  CarattoliA.HasmanH. (2020). PlasmidFinder and in silico pMLST: Identification and typing of plasmid replicons in whole-genome sequencing (WGS).Methods Mol. Biol.2075285–294. 10.1007/978-1-4939-9877-7_20

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 12

  ClevelandK.MazumderS.GelfandM. (2014). Association of Raoultella bacteremia with diseases of the biliary tract.Scand. J. Infect. Dis.46541–542. 10.3109/00365548.2014.896032

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 13

  CsurösM. (2010). Count: Evolutionary analysis of phylogenetic profiles with parsimony and likelihood.Bioinformatics261910–1912. 10.1093/bioinformatics/btq315

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 14

  DrancourtM.BolletC.CartaA.RousselierP. (2001). Phylogenetic analyses of Klebsiella species delineate Klebsiella and Raoultella gen. nov., with description of Raoultella ornithinolytica comb. nov., Raoultella terrigena comb. nov. and Raoultella planticola comb. nov.Int. J. Syst. Evol. Microbiol.51925–932. 10.1099/00207713-51-3-925

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 15

  GalataV.BackesC.LacznyC.Hemmrich-StanisakG.LiH.SmootL.et al (2018). Comparing genome versus proteome-based identification of clinical bacterial isolates.Brief Bioinform.19495–505. 10.1093/bib/bbw122

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16

  García-LozanoT.Pascual PláF.Aznar OrovalE. (2013). Raoultella ornithinolytica in urinary tract infections. Clinical and microbiological study of a series of 4 oncologic patients.Med. Clin.141138–139. 10.1016/j.medcli.2012.11.021

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17

  GilchristC.ChooiY. (2021). Clinker & clustermap.js: Automatic generation of gene cluster comparison figures.Bioinformatics372473–2475. 10.1093/bioinformatics/btab007

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 18

  GurevichA.SavelievV.VyahhiN.TeslerG. (2013). QUAST: Quality assessment tool for genome assemblies.Bioinformatics291072–1075. 10.1093/bioinformatics/btt086

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 19

  HajjarR.AmbaraghassiG.SebajangH.SchwenterF.SuS. (2020). Raoultella ornithinolytica: Emergence and resistance.Infect Drug Resist.131091–1104. 10.2147/IDR.S191387

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 20

  HarukiY.HagiyaH.SakumaA.MuraseT.SugiyamaT.KondoS. (2014). Clinical characteristics of Raoultella ornithinolytica bacteremia: A case series and literature review.J. Infect Chemother.20589–591. 10.1016/j.jiac.2014.05.005

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 21

  IovlevaA.MettusR. T.McElhenyC. L.GriffithM. P.MustaphaM. M.PasculleA. W.et al (2019). High-level carbapenem resistance in OXA-232-producing Raoultella ornithinolytica triggered by ertapenem therapy.Antimicrob. Agents Chemother.64:e1335-19. 10.1128/AAC.01335-19

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 22

  KankiM.YodaT.TsukamotoT.ShibataT. (2002). Klebsiella pneumoniae produces no histamine: Raoultella planticola and Raoultella ornithinolytica strains are histamine producers.Appl. Environ. Microbiol.683462–3466. 10.1128/AEM.68.7.3462-3466.2002

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 23

  KruegerF.JamesF.EwelsP.AfyounianE.Schuster-BoecklerB. (2021). FelixKrueger/TrimGalore: V0.6.7.Hawaii: Zenodo, 10.5281/zenodo.5127899

  • CrossRef
  • Google Scholar

• 24

  LechnerM.FindeissS.SteinerL.MarzM.StadlerP.ProhaskaS. (2011). Proteinortho: Detection of (co-)orthologs in large-scale analysis.BMC Bioinform.12:124. 10.1186/1471-2105-12-124

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25

  LiX.-Z.NikaidoH. (2009). Efflux-mediated drug resistance in bacteria.Drugs691555–1623.

  • Google Scholar

• 26

  LiY.QiuY.GaoY.ChenW.LiC.DaiX.et al (2022). Genetic and virulence characteristics of a Raoultella planticola isolate resistant to carbapenem and tigecycline.Sci. Rep.12:3858. 10.1038/s41598-022-07778-0

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 27

  MartínezJ.MartínezL.RosenbluethM.SilvaJ.Martínez-RomeroE. (2004). How are gene sequence analyses modifying bacterial taxonomy? The case of Klebsiella.Int. Microbiol.7261–268.

  • Google Scholar

• 28

  MatarG. M.AndremontA.BazziW. (2020). Editorial: Combating antimicrobial resistance – a one health approach.Front. Cell. Infect. Microbiol.9:458. 10.3389/fcimb.2019.00458

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 29

  MauN.RossL. (2010). Raoultella ornithinolytica bacteremia in an infant with visceral heterotaxy.Pediatr. Infect. Dis. J.29477–478. 10.1097/INF.0b013e3181ce9227

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 30

  McInerneyJ. O.McNallyA.O’ConnellM. J. (2017). Why prokaryotes have pangenomes.Nat. Microbiol.2:17040.

  • Google Scholar

• 31

  MediniD.DonatiC.TettelinH.MasignaniV.RappuoliR. (2005). The microbial pan-genome.Curr. Opin. Genet. Dev.15589–594.

  • Google Scholar

• 32

  NaS.KimY.YoonS.HaS.BaekI.ChunJ. (2018). UBCG: Up-to-date bacterial core gene set and pipeline for phylogenomic tree reconstruction.J. Microbiol.56280–285. 10.1007/s12275-018-8014-6

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 33

  NguyenL.SchmidtH.von HaeselerA.MinhB. Q. (2015). IQ-TREE: A fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies.Mol. Biol. Evol.32268–274. 10.1093/molbev/msu300

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 34

  NishimuraY.WataiH.HondaT.MiharaT.OmaeK.RouxS.et al (2017). Environmental viral genomes shed new light on virus-host interactions in the ocean.mSphere2:e359-16. 10.1128/mSphere.00359-16

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 35

  PageA.CumminsC.HuntM.WongV.ReuterS.HoldenM.et al (2015). Roary: Rapid large-scale prokaryote pan genome analysis.Bioinformatics313691–3693. 10.1093/bioinformatics/btv421

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 36

  ParkJ.HongK.LeeH.ChoiS.SongS.SongK.et al (2011). Evaluation of three phenotypic identification systems for clinical isolates of Raoultella ornithinolytica.J. Med. Microbiol.60492–499. 10.1099/jmm.0.020768-0

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 37

  ParkS.-C.LeeK.KimY. O.WonS.ChunJ. (2019). Large-scale genomics reveals the genetic characteristics of seven species and importance of phylogenetic distance for estimating pan-genome size.Front. Microbiol.10:834. 10.3389/fmicb.2019.00834

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 38

  Pérez-LlarenaF. J.ZamoranoL.KerffF.BeceiroA.GarcíaP.MiróE.et al (2014). Genetic and kinetic characterization of the novel AmpCβ-lactamases DHA-6 and DHA-7.Antimicrob. Agents Chemother.586544–6549. 10.1128/AAC.03144-14

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 39

  PoirelL.WalshT.CuvillierV.NordmannP. (2011). Multiplex PCR for detection of acquired carbapenemase genes.Diagn. Microbiol. Infect Dis.70119–123. 10.1016/j.diagmicrobio.2010.12.002

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 40

  PooleK. (2000). Efflux-mediated resistance to fluoroquinolones in gram-negative bacteria.Antimicrob. Agents Chemother.442233–2241.

  • Google Scholar

• 41

  PritchardL.GloverR. H.HumphrisS.ElphinstoneJ. G.TothI. K. (2016). Genomics and taxonomy in diagnostics for food security: Soft-rotting enterobacterial plant pathogens.Analyt. Methods812–24.

  • Google Scholar

• 42

  ReevesC. M.MagallonJ.RochaK.TranT.PhanK.VuP.et al (2020). Aminoglycoside 6’-N-acetyltransferase type Ib [AAC(6’)-Ib]-mediated aminoglycoside resistance: Phenotypic conversion to susceptibility by silver ions.Antibiotics (Basel)10:29. 10.3390/antibiotics10010029

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43

  ReyesJ.VillavicencioF.VillacísJ.PavónE.CampoverdeN.EspinelM.et al (2020). First report of a clinical isolate of blaOXA-48- carbapenemase producing Raoultella ornithinolytica in South America.Rev. Argent Microbiol.5282–83. 10.1016/j.ram.2019.02.002

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 44

  SeemannT. (2014). Prokka: Rapid prokaryotic genome annotation.Bioinformatics302068–2069. 10.1093/bioinformatics/btu153

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 45

  SȩkowskaA.MikuckaA.Gospodarek-KomkowskaE. (2018). Identification of Raoultella spp.: Comparison of three methods.Indian J. Med. Microbiol.36197–200. 10.4103/ijmm.IJMM_17_99

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 46

  SengP.BoushabB.RomainF.GourietF.BruderN.MartinC.et al (2016). Emerging role of Raoultella ornithinolytica in human infections: A series of cases and review of the literature.Int. J. Infect. Dis.4565–71. 10.1016/j.ijid.2016.02.014

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 47

  TafouktR.TouatiA.LeangapichartT.BakourS.RolainJ. (2017). Characterization of OXA-48-like-producing Enterobacteriaceae isolated from river water in Algeria.Water Res.120185–189. 10.1016/j.watres.2017.04.073

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48

  VandecraenJ.ChandlerM.AertsenA.Van HoudtR. (2017). The impact of insertion sequences on bacterial genome plasticity and adaptability.Crit. Rev. Microbiol.43709–730.

  • Google Scholar

• 49

  VettingM. W.HegdeS. S.FajardoJ. E.FiserA.RoderickS. L.TakiffH. E.et al (2006). Pentapeptide repeat proteins.Biochemistry451–10.

  • Google Scholar

• 50

  VielvaL.de ToroM.LanzaV.de la CruzF. (2017). PLACNETw: A web-based tool for plasmid reconstruction from bacterial genomes.Bioinformatics333796–3798. 10.1093/bioinformatics/btx462

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 51

  WalckenaerE.Leflon-GuiboutV.Nicolas-ChanoineM. (2008). How to identify Raoultella spp. including R. ornithinolytica isolates negative for ornithine decarboxylase? The reliability of the chromosomal bla gene.J. Microbiol. Methods75405–410. 10.1016/j.mimet.2008.07.011

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 52

  WengX.ShiQ.WangS.ShiY.SunD.YuY. (2020). The characterization of OXA-232 carbapenemase-producing ST437 Klebsiella pneumoniae in China.Can. J. Infect. Dis. Med. Microbiol.2020:5626503. 10.1155/2020/5626503

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 53

  WyresK.LamM.HoltK. (2020). Population genomics of Klebsiella pneumoniae.Nat. Rev. Microbiol.18344–359. 10.1038/s41579-019-0315-1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 54

  ZhangX.LiuX.YangF.ChenL. (2018). Pan-genome analysis links the hereditary variation of Leptospirillum ferriphilum with its evolutionary adaptation.Front. Microbiol.9:577. 10.3389/fmicb.2018.00577

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 55

  ZouH.BerglundB.XuH.ChiX.ZhaoQ.ZhouZ.et al (2020). Genetic characterization and virulence of a carbapenem-resistant Raoultella ornithinolytica isolated from well water carrying a novel megaplasmid containing blaNDM-1.Environ. Pollut.260: 114041. 10.1016/j.envpol.2020.114041

  • Pubmed Abstract
  • CrossRef
  • Google Scholar