*Article* **Antibiotic Resistance Characteristics of** *Pseudomonas aeruginosa* **Isolated from Keratitis in Australia and India**

**Mahjabeen Khan 1 , Fiona Stapleton 1 , Stephen Summers 2 , Scott A. Rice 2,3 and Mark D. P. Willcox 1, \***


Received: 2 July 2020; Accepted: 9 September 2020; Published: 14 September 2020

**Abstract:** This study investigated genomic differences in Australian and Indian *Pseudomonas aeruginosa* isolates from keratitis (infection of the cornea). Overall, the Indian isolates were resistant to more antibiotics, with some of those isolates being multi-drug resistant. Acquired genes were related to resistance to fluoroquinolones, aminoglycosides, beta-lactams, macrolides, sulphonamides, and tetracycline and were more frequent in Indian (96%) than in Australian (35%) isolates (*p* = 0.02). Indian isolates had large numbers of gene variations (median 50,006, IQR = 26,967–50,600) compared to Australian isolates (median 26,317, IQR = 25,681–33,780). There were a larger number of mutations in the *mutL* and *uvrD* genes associated with the mismatch repair (MMR) system in Indian isolates, which may result in strains losing their efficacy for DNA repair. The number of gene variations were greater in isolates carrying MMR system genes or*exoU*. In the phylogenetic division, the number of core genes were similar in both groups, but Indian isolates had larger numbers of pan genes (median 6518, IQR = 6040–6935). Clones related to three different sequence types—ST308, ST316, and ST491—were found among Indian isolates. Only one clone, ST233, containing two strains was present in Australian isolates. The most striking differences between Australian and Indian isolates were carriage of *exoU* (that encodes a cytolytic phospholipase) in Indian isolates and *exoS* (that encodes for GTPase activator activity) in Australian isolates, large number of acquired resistance genes, greater changes to MMR genes, and a larger pan genome as well as increased overall genetic variation in the Indian isolates.

**Keywords:** antibiotic susceptibility; WGS; phylogenetic analysis; DNA mismatch repair system

#### **1. Introduction**

*Pseudomonas aeruginosa* is a ubiquitous bacterium which can cause opportunistic or nosocomial infections in immuno-compromised patients [1]. *P. aeruginosa* commonly causes corneal (keratitis) [2], respiratory, burn and wound infections, and infections related to medical or surgical devices including ventilator-associated pneumonia [3,4]. *P. aeruginosa* corneal infections are usually related to contact lens wear, but other risk factors for keratitis in non-contact lens wearers include ocular trauma, ocular surgery, and prior ocular surface disease [5–8].

The prevalence of multi-drug resistant (MDR) or extensively drug resistant strains of *P. aeruginosa* reduces treatment options, significantly increasing morbidity rates [9]. *P. aeruginosa* is naturally resistant to some antibiotics due to the possession of specific resistance genes such as *catB* that confers

chloramphenicol resistance and an inducible *ampC* which encodes for a β-lactamase that hydrolyses cephalothin and ampicillin, conferring resistance to β-lactams [10]. Additionally, the regulation of efflux pumps also contributes towards an elevated resistance to antibiotics [11]. For example, expression of the efflux pump MexAB-OprM contributes towards intrinsic resistance to a broad spectrum of antibiotics [12], whereas the efflux pump MexXY-OprM is involved in the adaptive resistance to aminoglycosides [13]. Other resistance mechanisms in *P. aeruginosa* include the acquisition of transferrable resistance determinants, including those associated with transposons and integrons [14]. Antibiotic resistance of *P. aeruginosa* varies according to the region where the strains have been isolated [15,16] presumably due to the prescribing practices, availability of antibiotics, and perhaps their use in animal husbandry. Various epidemiological studies have identified MDR *P. aeruginosa* from different infections and these isolates have acquired different resistance characteristics. For example, aminoglycoside resistance [17] and ciprofloxacin persistence [18] are found in cystic fibrosis isolates of *P. aeruginosa.* Some of these MDR strains are clonal and such clonal strains are often the predominant global clinical MDR isolates [19] which spread resistance characteristics into the wider population which enables clonal lineages to expand with time.

*ExoU* has been associated with virulence of *P. aeruginosa* at the ocular surface. *ExoU* is a phospholipase that causes mammalian cell death [20] and *exoU* possession is common in strains isolated from ocular infections [21]. There is a correlation between carriage of *exoU* and elevated resistance to fluoroquinolones and aminoglycosides [22]. *ExoU* is carried by strains on a genomic island that also contains resistance genes for a range of antibiotics [23].

In addition to the acquisition of resistance genes, bacteria can develop resistance through mutation of genes so that antibiotic targets are modified. Mutation rates are elevated in strains that carry mutations in DNA mismatch repair (MMR) systems and hence such mutator strains will normally carry more mutations than non-mutator strains [24]. In *P. aeruginosa*, the MMR system is composed of *mutS*, *mutL*, and *UvrD* genes [25]. Strains of *P. aeruginosa* isolated from the lungs of cystic fibrosis patients have alterations in the DNA MMR system and this has been correlated with multiple antimicrobial resistance [23].

In Australia, there is a tight regulation of prescribing antibiotics, and antibiotics can only be obtained legally with a prescription from a qualified healthcare professional according to the Therapeutic Goods Act 1989. In India, on the other hand, whilst branded antibiotics exist, other forms such as counterfeit, substandard, and 'spurious' antibiotics have been reported [26], making surveillance and regulation difficult [27]. While the antibiotic consumption per person in Australia and India in 2010 was approximately similar, there was a more rapid increase between 2000 and 2010 in India [28]. These differences may affect antibiotic resistance development.

The aim of the current study was to compare the phenotypic resistance and genetic characteristics associated with resistance between strains isolated from Australia and India to better understand the underlying factors that may lead to an increased resistance in *P. aeruginosa* strains associated with ocular infection.

#### **2. Results**

#### *2.1. Antibiotic Susceptibility*

The minimum inhibitory concentrations (MICs) and minimum bactericidal concentrations (MBCs) of the *P. aeruginosa* isolates were determined (Table 1). Strains showing intermediate resistance (I) as well as full resistance to antibiotics were categorized as resistant (R) for subsequent analyses. Based on the Centers for Disease Control and Prevention's (CDC, Atlanta, GA, USA) definition of multi-drug resistance as "an isolate that is resistant to at least one antibiotic in three or more drug classes", isolates 198, 202, 216, 217, 218, 219, 220, and 221 were deemed to be multi-drug resistant. Australian isolates 223, 224, 225, 227, 233, and 235 were also resistant to three antibiotics but these antibiotics were not of different classes. Isolates 176, 193, and 206 were sensitive to all antibiotics, but all

other isolates were resistant to at least one antibiotic. Overall, Indian isolates were more resistant to antibiotics compared to Australian isolates. Among Australian isolates (*n* = 14), resistance was 78% for imipenem, 57% for ceftazidime, 50% for ciprofloxacin, 21% for piperacillin, 14% for levofloxacin, 7% for tobramycin, and no isolates were resistant to gentamicin or polymyxin. In contrast, resistance in Indian isolates (*n* = 12) was 75% for ciprofloxacin, 58% for imipenem, 50% for levofloxacin, tobramycin, and ceftazidime, 41% for piperacillin, 40% for gentamicin, and 25% for polymyxin.


**Table 1.** Minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of antibiotics to *Pseudomonas aeruginosa* keratitis isolates.

Data for Australian isolates (shaded in gray). Data for 123–182 is from a previously published study [29]. Strains 188–221 were Indian keratitis isolates. R = resistant, I = intermediate resistance. \* Cipro = Ciprofloxacin, Levo = Levofloxacin, Genta = Gentamicin, Tobra = Tobramycin, Pipera = Piperacillin, Imi = Imipenem, Ceftaz = Ceftazidime, PMB = Polymyxin B; # = Antibiotic breakpoints for sensitive, intermediate, resistant classifications.

#### *2.2. General Features of the Genomes*

The isolates after de novo assembly consisted of different numbers of contigs ranging from 50 for isolate 169 to 1917 for isolate 216. The average number of coding sequences was 6162 ± 359.2 for the Australian isolates and 6544 ± 889 for the Indian isolates. Isolates had an average of 66.1% G + C content. The tRNA copy number for the isolates ranged from 57 to 86 (which may vary between studies that use different assembly methods). The general features of the isolates are provided in Supplementary Table S1.

#### *2.3. Acquired Resistance Genes*

*P. aeruginosa* isolates were examined for horizontally acquired antibiotic resistance genes (Table 2) using the Resfinder database. Altogether, 33 different acquired antibiotic resistance genes for various classes of antibiotics including aminoglycosides, fluoroquinolones, beta-lactams were found in these isolates (Table 2).


**Table 2.** Acquired resistance genes in *P. aeruginosa* isolates from India and Australia.

\**\* msr(E) encodes macrolide and lincosamide resistance msr(E) encodes macrolide and lincosamide resistance*. Isolates shaded in grey indicate Australian strains. . Isolates shaded in grey indicate Australian strains. *Black color represents gene presence*.

An aminoglycoside resistance gene (*aph(3')-IIb*), a beta-lactam resistance gene (*blaPAO*), a fosfomycin resistance gene (*fosA*), and a chloramphenicol resistance gene (*catB7*) were common to all isolates. The Australian isolates (123–182) had acquired only eight resistance genes, while the Indian isolates (188–221) had acquired 26 different resistance genes (Table 2). Five Indian isolates (198, 202, 217, 219, and 221, with large pan genomes) acquired the largest number of resistance genes. Of these five isolates, the pairs 198/219 and 202/221 had the most similar resistance gene profiles and each member of the pair were of the same sequence type, ST308 and ST316 respectively. As acquired resistance genes may be carried on integrons, the genomes of the *P. aeruginosa* isolates were analyzed for integrons using Integron Finder version 1.5.1. *qnrCV1* was associated with a class 1 integron in isolates 202 and 221 and a Tn3 transposon in isolates 198 and 219.

Several types of non-synonymous variations were found in the core genome of the keratitis *P. aeruginosa* isolates when compared with the reference genome of PAO1 (Table 3). These non-synonymous mutations included single nucleotide polymorphisms (SNPs), multi-nucleotide polymorphisms (MNPs), deletions, insertions, and complex variations (where more than one change occurred at one specific location compared to the reference strain). The total variations in the isolates ranged from 76,080 in isolate 206 to 22,536 in isolate 181. There was a median of 26,317 (IQR = 25,681–33,780) variations in the genomes of Australian isolates and a median of 50,006 (IQR = 26,967–50,600) in the Indian isolates (*p* = 0.09). Based on the grouping of core genome phylogeny, isolates within group 2 (198, 202, 219, 220, 221, 233) had the most variations. Isolate 206, which had a unique sequence type and was placed in a separate group by pan genome analysis, had an exceptionally high number of variations (76,080) and SNPs (67,271).


**Table 3.** Frequency of different types of variation in the genes of *P. aeruginosa* isolates.

SNP = single nucleotide polymorphism; MNP = multi-nucleotide polymorphism. Isolate numbers highlighted in gray are from Australia.

Non-synonymous mutations were assessed in resistance genes of the *P. aeruginosa* isolates (Supplementary Table S2). There were no large differences in the mutations in resistance genes of any of the isolates except the antibiotic efflux-related genes *opmH* and *rosC*. *opmH* had ≥9 mutations in all isolates except 127, 162,169, 202, 218, 220, and 221 (mostly isolates of group 2 of core and pan genome phylogenies except 127 and 218). *rosC* had 20 non-synonymous mutations including insertions/deletions in isolate 206, 11 in 233, and ≥5 mutations in isolates 162, 169, 176, 202, 216, 217, 219, 220, and 221 (mostly isolates of group 2 of core and pan genome phylogenies except 176, 216), but ≤3 mutations in

isolates 123, 126, 123, 181, 182, 188, 189, 193, 198, and 218 (mostly isolates of group 1 of core and pan genome phylogenies except 198). Mutations in efflux genes encoding efflux pumps were also found, including *mexX*, *mexT*, *mexD*, *mexM*, and *mexY*, although there was no significant difference between two groups in the possession of mutations in these genes. All other mutations in the genes were random without any association to sequence type, phylogeny, or susceptibility to antibiotics.

#### *2.4. Possession of exoU and Mutations in the DNA Mismatch Repair System*

*ExoU* was present in the genomes of all isolates in group 2 (core and pan genome phylogenetic group) as well as isolates 123 and 127. All other isolates possessed *exoS* with the exception of isolate 126 which possessed both *exoU* and *exoS*. To address differences in the numbers of sequence variants between the isolates, the genes involved in the DNA mismatch repair (MMR) system *mutS* (that encodes a protein which binds to errors in DNA), *mutL* (that encodes a protein that works in synergy with MutS and activates UvrD), and *uvrD* (a DNA helicase active in DNA replication) were examined. The mutations in the MMR system included SNPs, indels, and complex variants. The number of mutations in *mutL* ranged from 1 to 2 and mutations in *mutS* (which ranged between 0 and 2) were found in seven isolates (Table 4). In *uvrD*, the number of mutations ranged between 0 and 5 (Table 4). *exoU* containing isolates possessed a median of two (IQR = 1–3) mutations in *mutL*, zero (IQR = 0–2) mutations in *mutS*, and four (IQR = 2–5) mutations in *uvrD*, whereas *exoS* containing isolates possessed a median of zero (IQR = 0–1) mutations in *mutL*, zero (IQR = 0–1) mutations in *mutS*, and two (IQR = 0–2) median mutations in *uvrD*. There were significant differences in the number of *mutL* (*p* = 0.0021) and *uvrD* (*p* = 0.02) mutations in *exoS* and *exoU* isolates but not with *mutS* (*p* = 0.3). Isolate 206, an *exoS* strain and an outlier in the core genome analysis, had one mutation in *mutL*. Details of mutations occurring in nucleotide and respective proteins are provided in Supplementary Table S3.


**Table 4.** Possession of *exoU* and *exoS* and number and type of non-synonymous mutations in the mismatch repair system genes in *P. aeruginosa* isolates.


#### **Table 4.** *Cont*.

SNP = single nucleotide polymorphism, MNP = multinucelotide polymorphism. Isolates shaded in grey indicate Australian strains.

#### *2.5. Sequence Type Analysis and Phylogenetics*

All Australian isolates were of different sequence types (ST), except 225 and 227 which belonged to ST233. Among the 12 Indian isolates, one isolate was designated as belonging to a new sequence type, two isolates (198 and 219) belonged to ST308, two others (188 and 189) belonged to ST491, and three isolates (202, 220, and 221) belonged to ST316 (Table 5).


**Table 5.** Sequence types of *P. aeruginosa* isolates.

Gray shading denotes Australian isolates. \*, †, **#**, ¤ indicates strains belong to the same sequence types (STs).

The number of core and total or pan (or total) genes were reported from the statistical summary of Roary v3.11.2. The core genomes of the isolates were aligned using PA7 (Accession number NC\_009656.1), PA14 (Accession number NC\_004863.1), and PAO1 (Accession number NC\_002516.1) as

reference strains. The eight published genomes of *P. aeruginosa* isolates from eye as well as strains from other sources were also included. The core genes of published isolates are provided in Supplementary Table S4. The isolates were sub-grouped based on the number of core genes; isolates with a similar number of core genes were closely aligned and isolates with the same sequence type were grouped together. The core genomes formed two groups in the phylogenetic tree (Figure 1). Isolates in group 1 tended to have a larger number of core genes than isolates in group 2. Isolate 206, PA57, and PA7 were outliers based on core genome phylogeny. The Australian and Indian isolates had a similar number of core genes, whereas the Indian isolates had a larger number of pan genes (10,889) due to the acquisition of shell genes (genes present in two or more strains) (Table 4).

**Figure 1.** Core genome phylogeny of *P. aeruginosa* isolates using Parsnp. PAO1 was used as reference. PA7 and PA14 were also included.

The phylogenetic relationships of these *P. aeruginosa* isolates were assessed by aligning their pan genome against PAO1 as a reference. The output generated using Roary showing the gene presence or absence in all isolates is provided in Supplementary Figure S1. This again divided the *P. aeruginosa* isolates into two major groups. Six multi-drug resistant Indian isolates (198, 202, 217, 219, 220, 221) and the VRFPA04 isolate (isolated from the cornea) were clustered in one group, which also contained the two Australian isolates 162 and 169. The Indian isolate 216 was categorized in a separate sub-group due to the large number of shell genes and possession of *exoS*.

The second group (group 2 of the pan genome analysis) included most of the Australian (123, 126, 127, 162, 176, 181, 182, 223, 224, 225, 227, 235) and Indian (188, 189, 193, 216 218) isolates along with reference strain PAO1 (Figure 2). Overall, the multi-drug resistant Indian isolates had a large pan genome (total of 10,889 genes obtained from the statistical summary in Roary v3.11.2). The pan genome grouping of isolates was broadly based on the number of pan (or total) genes and possession of either

*exoU* or *exoS* in each group, except two Australian isolates 123 and 127 which were in group 2 but possessed *exoU*. The other exception to this grouping pattern was for isolates 181 and 182 which had large pan genomes and were clustered into group 1 but carried *exoS*.

**Figure 2.** Pan genome phylogeny of *P. aeruginosa* isolates. Branches with no color representation indicate non-ocular isolates. Red color indicates Australian, blue color represents Indian, and orange color represents published eye isolates. Green color represents reference strains. Purple color represents reference strains.

The isolates of group 2 usually had a large number of pan genes and were *exoU*+. Isolates having similar numbers of pan genes were sub-grouped together. For example, isolate 193 (pan genes = 6084) and 218 (pan genes = 6001) were sub-grouped together. Isolate 218 had a similar number of pan genes to isolate 123 (pan genes = 6001), but isolate 218 possessed *exoS*, while 123 possessed *exoU*, and thus these were not grouped together. Isolates belonging to the same sequence type were also grouped together. The MDR isolates, the isolates with same STs, and isolates with large gene variations were clustered in one pan-group. The previously published isolates PA\_D1, PA\_D2, PA\_D9, and PA\_D16 with the same ST and those with large shell genes were grouped with the MDR isolates of the current study.

#### **3. Discussion**

This study investigated genomic differences in Australian and Indian *P. aeruginosa* isolates from keratitis. Phenotypically, more resistance was found in Indian isolates compared to Australian isolates as has been shown in previous studies [30,31]. Unregulated antibiotic use in India has been linked to increased antibiotic resistance [32]. Resistance to antibiotics is problematic even in the treatment of keratitis, where a topical application of antibiotics is used. Infection with antibiotic resistant strains results in prolonged infection [33], more severe keratitis [5], and an increase in the cost of treatment [34,35].

Indian *P. aeruginosa* strains harbored more resistance genes compared to Australian isolates, although *aph(3')-IIb, blaPAO1 (fosA)*, and *catB7* were found in all isolates, which was consistent with previous studies [31,36]. *qnrVC1* was found in four Indian isolates but no Australian isolates. This fluoroquinolone resistance gene has not been previously reported in *P. aeruginosa* ocular isolates [31], but it has been reported in burns isolates and has been identified as being carried on an integron [37]. Similarly, in the current study *qnrVC1* was carried on a class 1 integron in isolates 202 and 221, but integrated into a Tn3 transposon in isolates 198 and 219. This gene has also been isolated from the high risk ST773 clone of *P. aeruginosa* from urine in Hungary [38]. High risk clones are isolates with high mutational rates in resistance genes and those that have acquired a large number of resistance genes. As previously described, resistance to fluoroquinolones in keratitis *P. aeruginosa* isolates was also due to mutations in the quinolone resistance determining regions of *gyrA* and *parC* [15]. Possession of *qnrVC1* and mutations in *gyrA* and *parC* were associated with high levels of fluroquinolone resistance. The possession of large numbers of acquired resistance genes by Indian isolates likely contributed to the higher rates of resistance of these isolates. The Indian isolates 198, 202, 217, 219, 220, and 221 also had a high number of gene variations which is an independent mechanism of resistance.

The aminoglycoside resistance gene *aph(6)-Id* which encodes for streptomycin resistance was found in six Indian isolates, including the four that carried *qnrVC1*, but in no Australian isolates. Previously, *aph(6)-Id* was found in only one Indian ocular isolate from 1997 [31], but has been found in cystic fibrosis *P. aeruginosa* isolates [39] and has been associated with the transposon Tn5393 on a plasmid in one strain of *P. aeruginosa* [39]. As streptomycin is no longer used in clinical treatment [40], this resistance may not be clinically relevant but does suggest environmental selection for the persistence of this gene.

The total number of gene variants found in the Indian isolates 198, 202, 219, and 221 were greater than Australian isolates. However, there were a small number of SNPs found in the genes associated with resistance for these isolates. The Indian isolate 206 (NEWST) had a high number of SNPs in antibiotic resistance genes *mexC*, *mexD*, *mexM*, *mexX*, *mexS*, *opmE*, *mexP*, *mexK, oprJ*, *ampC*, *rosC*, and *mprF*. There was no difference in the mutations of other *mex* genes including *mexX*, *mexT*, *mexD*, *mexM*, and *mexY* between Australian and Indian isolates. Given that most isolates from both countries, whether they were sensitive or resistant, had a similar number of mutations in the resistance genes, it is likely that the resistance to antibiotics was related to the possession of acquired resistance genes rather than mutations in chromosomal genes.

In the Australian isolates, four out of the eight isolates (50%) carried *exoU*, while one isolate was both *exoU*+/*exoS*+ and three (38%) were *exoS*+. In Indian isolates, 50% carried *exoU* and 50% carried *exoS*. A previous study has also shown an equal ratio of both genes [41] in keratitis isolates. The possession of the *exoU* genotype in *P. aeruginosa* ocular isolates has been related to elevated resistance to disinfectants [42], fluoroquinolones [43], and multiple antibiotics [41]. Furthermore, one study reported worst clinical outcomes and more resistance by *exoU* carrying isolates [43]. The isolates of this study showed similar findings because the *exoU*+ isolates 198, 202, 217, 219, 220, 221, and 233 were also MDR.

The DNA mismatch repair system (MMR) in *P. aeruginosa* is based on the protein trimer MutS-MutL-UvrD and functions to correct errors and preserve the integrity of the genome [24,44,45]. The *mutH* component of MMR, which is important in other Gram negative bacteria, such as *E. coli* [46], has not been found previously in *P. aeruginosa* [25] and was not present in the isolates of the current study. Mutations in *mutS*, *mutL* and *uvrD* can reduce the ability of the bacterium to repair DNA lesions [46]. Strong mutator strains have defects in their MMR system and mutations in *mutS* predominate [47]. Mutations in the MMR can be a reason for the development of hypermutations in isolates. In cystic fibrosis, hypermutations were found to be a key factor in the development of MDR resistant *P. aeruginosa* strains [23]. Similar findings were found in this study where isolates 198, 202, 206, 219, and 221 had mutations in the MMR genes, and these isolates had an overall larger variation in their genomes. In the current study, isolates had more mutations in *mutL* and *uvrD,* suggesting the strains may not be

strong mutators (which is usually associated with mutation in *mutS*), but nevertheless can undergo uncorrected genetic changes. Indeed, the *P. aeruginosa* isolates in the current study which had mutations in *mutL* and *uvrD* had greater numbers of SNPs, insertions and deletions, acquired genes, and had large pan genomes. Among these isolates, 198, 202, 219, and 221 possessed either the transposon Tn3 or class 1 integrons which carried the acquired genes. This also might be due to mutated MMR, as mutations in MMR genes increase the chances of horizontal gene transfer in mutator isolates [47]. The number of mutations in MMR was greater in *exoU* possessing isolates with large gene variations. *exoU* is carried on genomic islands [48,49] and these *exoU* carrying isolates had larger pan genomes with possession of mobile genetic elements. Therefore, the isolates with the mutated MMR systems may have a greater ability of strains to accumulate gene variations and the acquisition of *exoU*. Isolate 206, on the other hand, possessed *exoS* and was not MDR but possessed a large number of SNPs and a large pan genome with one mutation in each of *mutS* and *mutL*. Further in-depth studies are required to understand the influence of the MMR system on genomic changes in *P. aeruginosa*.

Analysis of the sequence types of the *P. aeruginosa* ocular isolates revealed the presence of three clones, two in the Indian and one in the Australian isolates. The isolates with the same STs had mostly the same phenotypic and genotypic features. The exception to this was isolate 220 that had acquired fewer resistance genes compared to the other two isolates 202 and 221 of ST316. Previously, five ocular *P. aeruginosa* strains from India isolated in 1997 were of sequence type ST308 [31]. The two isolates of ST308 in the current study, isolated in 2017 and 2018, had acquired more resistance genes compared to isolates from 1997 [31]. This indicates that the clonal isolates have continued to evolve over this time period, although the specific selection factors driving those changes are yet to be elucidated. None of the isolates were collected from the same patient. The majority of the isolates with the same STs grouped in the same phylogeny including previously published isolates (PA\_D1, PA\_D2, PA\_D9, P\_D16) with ST1971.

Core and pan genome phylogenies of the isolates produced two almost identical groups, which was in agreement with previously published studies [31,50]. Both phylogenies included isolates from either Australia or India, but those in group 2 tended to be the MDR Indian isolates and possessed higher numbers of antibiotic resistance genes. About 65% of all ocular isolates grouped together which indicated less diversity in the ocular *P. aeruginosa* isolates [31,51,52]. The grouping of MDR strains from this study with PA14 along a MDR ocular isolate VRFPA04 [36] in both core genome and pan genome analysis, and the grouping of the sensitive strains with PAO1 along the commonly studied cystic fibrosis isolates DK2 and LESB58, was similar to a previous study examining older isolates from India and Australia [31]. Isolate 206, which had the smallest number of core genes and was of a new sequence type, was an outlier in the core genome phylogeny similar to the taxonomic outlier PA7 [53]. However, isolate 206 was grouped together with other isolates in the pan genome because it had acquired a large number of genes. Acquired genes are part of the pan rather than the core genome [53] and the presence of larger pan genomes in MDR *P. aeruginosa* isolates points towards the acquisition of new genes [54]. Previously, a smaller core genome size of 4910 genes has been reported in ocular *P. aeruginosa* isolates [31]. However, the current study found a core genome size similar to *P. aeruginosa* from different sources, comprising 5316–5233 genes [55,56]. The core genome (which is almost 90% of total genome) refers to the conserved genes present in a species [57] which might differ in each individual strain within that species. Additionally, SNPs can be a result of poor sequencing quality and hence it is important to have a good sequencing depth at those positions to identify them as a mutation rather than sequencing error [58]. Grouping of all the isolates including ocular and non-ocular remained the same in both core and pan genome phylogeny.

#### **4. Materials and Methods**

#### *4.1. P. aeruginosa Strains and Susceptibility Testing*

Twenty-six *P. aeruginosa* keratitis isolates, eight isolated in Australia from 2004 to 2006, six from 2018 and 2019 (total 14 Australian isolates), and twelve isolated in India between 2017 and 2018, were included in this study. These isolates were selected from a larger collection of strains based on their antibiotic susceptibilities (those phenotypically resistant to multiple antibiotics, some resistant to one or multiple antibiotics, and some which were sensitive to all antibiotics). The susceptibilities of Australian strains (2004–2006) included in this study have been previously published [29]. Strains were selected after comparing their susceptibilities to antibiotics that are used to treat ocular infections. For genetic comparisons, the data of 34 *P. aeruginosa* isolates from eyes and other sources were also included. The general characteristics of these isolates are described in Supplementary Table S4. The genomes of these isolates were downloaded from the NCBI database and reannotated for this study using the same parameters as of the isolates of this study to avoid any bias in results.

The minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of various antibiotics which are commonly used to treat *P. aeruginosa* keratitis [16] were assessed for the isolates using the broth microdilution method in 96-well plates following the Clinical and Laboratory Standard Institute guidelines [59]. The antibiotics tested were ciprofloxacin, levofloxacin, gentamicin, ceftazidime (Sigma-Aldrich, St. Louis, MO, USA), polymyxin B (Sigma-Aldrich, Vandtårnsvej, Søborg, Denmark), tobramycin, piperacillin (Cayman Chemical Company, Ann Arbor, MI, USA), and imipenem (LKT Laboratories Inc., St. Paul, MN, USA). The susceptibility results were interpreted using the EUCAST v9 [60] and CLSI [61] 2017 breakpoints.

#### *4.2. Genomic Sequencing*

DNeasy Blood and Tissue Kits (Qiagen, Hilden, Germany) were used for DNA extraction as per the manufacturer's recommendations. The Nextera XT DNA library preparation kit (Illumina, San Diego, CA, USA) was used to prepare paired-end libraries. All the libraries were multiplexed on one MiSeq run. FastQC version 0.117 (https://www.bioinformatics.babraham.ac.uk/projects/fastqc) was used to assess the quality of sequenced genomes using raw reads. Version 0.38 of Trimmomatic [61] was used for trimming the adapters from the reads following de novo assembly using Spades v3.13.0 [62]. Genomes were annotated using Prokka v1.12 [63].

Sequence types were investigated using PubMLST https://pubmlst.org/. Pan genomes of the *P. aeruginosa* isolates were analyzed using Roary v3.11.2 [64] using PAO1 as a reference, while core genome phylogeny was constructed using Harvest Suite Parsnp v1.2 [65] with strains PAO1, PA7, and PA14 used as reference strains. The output file 'genes\_ presence\_absence' was used to compare the *P. aeruginosa* isolates. Acquired resistance genes were identified using the online database Resfinder v3.1 (Centre for Genomic Epidemiology, DTU, Denmark) [66]. Integron Finder v1.5.1 was used to identify any integrons present in the isolates. Mutations in the genes were detected using Snippy V2 [67]. Isolates with same sequence types were compared for nucleotide similarities using the MUMmer online web tool (http://jspecies.ribohost.com/jspeciesws/#analyse).

Using the Pseudomonas genome database (http://www.pseudomonas.com) and comprehensive antibiotic resistance database (https://card.mcmaster.ca), 76 genes related to *P. aeruginosa* resistance were selected to investigate the presence of single nucleotide polymorphisms. All isolates were analyzed for the presence of the type III secretion system associated virulence factors *exoU* and *exoS* using the BlastN database.

#### *4.3. Statistical Analysis*

The statistical analysis was performed using GraphPad Prism v8. Medians were calculated with the 'descriptive statistics' option during analysis of variance (ANOVA). *P*-values less than 0.05 were considered as significant. Fischer's Exact test was used to find the difference between acquired genes. To analyze the significant difference in the DNA mismatch repair genes between *exoU* and *exoS* isolates and gene variations in the isolates, the Mann–Whitney test was used.

#### **5. Conclusions**

Indian isolates and Australian isolates were clearly distinct in carrying a type III secretion system related to *exoU* and *exoS*. There was an association in the isolates for carrying acquired resistance genes with a large number of pan genes. Indian isolates were more resistant to antibiotics compared to Australian isolates. Additionally, isolates of *P. aeruginosa* from ocular infection had a large number of genetic variations (mutations) and a mutated mismatch repair system. However, the isolates collected from the same region or time will give a clearer idea of these differences.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2079-6382/9/9/600/s1, Table S1: Details of the *Pseudomonas aeruginosa* isolates used in the current study. Table S2: Gene variations of resistance genes in *Pseudomonas aeruginosa* isolates (the gray-shaded strains were isolated from Australia). Nucleotide accession: The nucleotide sequences are available in the GenBank under the Bio project accession number PRJNA590804. Table S3: Types of mutations in the mismatch repair system; Table S4: Genomics features of *P. aeruginosa* isolates; Figure S1: Pan-genome phylogenetic tree. The data on the right of the figure shows the presence and absence of genes. The tree was built using the genome of PAO1 as a reference.

**Author Contributions:** Conceptualization, M.K., F.S., and M.D.P.W.; methodology, M.K., F.S., S.S., S.A.R., and M.D.P.W.; writing—original draft preparation, M.K.; writing—review and editing, F.S., S.A.R., and M.D.P.W.; supervision, F.S. and M.D.P.W.; funding acquisition, F.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** The authors would like to acknowledge the Singapore Centre for Environmental Life Sciences Engineering (SCELSE), whose research is supported by the National Research Foundation Singapore, Ministry of Education, Nanyang Technological University, and National University of Singapore, under its Research Centre of Excellence Programme. We would also like to acknowledge Engineer Ahsan Ullah Khan, Independent Monitoring Unit Haripur KPK, Pakistan, for his help in the computational analysis and Doctor Nicole Carnt, School of Optometry and Vision Science UNSW, Sydney, for her help in the isolates collection. We are also thankful to UNSW high performance computing facility KATANA for providing us with the cluster time for the data analysis.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Clonal Clusters, Molecular Resistance Mechanisms and Virulence Factors of Gram-Negative Bacteria Isolated from Chronic Wounds in Ghana**

**Denise Dekker 1,2, \* , Frederik Pankok 3 , Thorsten Thye 1 , Stefan Taudien 3 , Kwabena Oppong 4 , Charity Wiafe Akenten 4 , Maike Lamshöft 1,2 , Anna Jaeger 1 , Martin Kaase 3 , Simone Scheithauer 3 , Konstantin Tanida 5 , Hagen Frickmann 5,6 , Jürgen May 1,2,7 and Ulrike Loderstädt 3**


**Abstract:** Wound infections are common medical problems in sub-Saharan Africa but data on the molecular epidemiology are rare. Within this study we assessed the clonal lineages, resistance genes and virulence factors of Gram-negative bacteria isolated from Ghanaian patients with chronic wounds. From a previous study, 49 *Pseudomonas aeruginosa*, 21 *Klebsiella pneumoniae* complex members and 12 *Escherichia coli* were subjected to whole genome sequencing. Sequence analysis indicated high clonal diversity with only nine *P. aeruginosa* clusters comprising two strains each and one *E. coli* cluster comprising three strains with high phylogenetic relationship suggesting nosocomial transmission. Acquired beta-lactamase genes were observed in some isolates next to a broad spectrum of additional genetic resistance determinants. Phenotypical expression of extended-spectrum beta-lactamase activity in the Enterobacterales was associated with *blaCTX-M-15* genes, which are frequent in Ghana. Frequently recorded virulence genes comprised genes related to invasion and iron-uptake in *E. coli*, genes related to adherence, iron-uptake, secretion systems and antiphagocytosis in *P. aeruginosa* and genes related to adherence, biofilm formation, immune evasion, iron-uptake and secretion systems in *K. pneumonia* complex. In summary, the study provides a piece in the puzzle of the molecular epidemiology of Gram-negative bacteria in chronic wounds in rural Ghana.

**Keywords:** wounds; Gram-negative bacteria; colonization; infection; clonal lineages; resistance genes; virulence factors

#### **1. Introduction**

The microbiology of chronic infected wounds, also on a molecular level, is poorly understood in sub-Saharan Africa (SSA) [1]. However, studies highlight the importance of antibiotic resistant Gram-negative bacteria [2–6].

**Citation:** Dekker, D.; Pankok, F.; Thye, T.; Taudien, S.; Oppong, K.; Akenten, C.W.; Lamshöft, M.; Jaeger, A.; Kaase, M.; Scheithauer, S.; et al. Clonal Clusters, Molecular Resistance Mechanisms and Virulence Factors of Gram-Negative Bacteria Isolated from Chronic Wounds in Ghana. *Antibiotics* **2021**, *10*, 339. https:// doi.org/10.3390/antibiotics10030339

Academic Editor: Manuela Oliviera

Received: 24 February 2021 Accepted: 19 March 2021 Published: 22 March 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

From other parts in the world, in particular from industrialized countries, information on the microbiology and the role of biofilm-forming microorganisms causing such infections are well established [7–10].

In chronic wounds, *Pseudomonas aeruginosa* is amongst the most frequently isolated Gram-negative bacteria, associated with biofilm formation [11,12]. Tightly adhering biofilms pose a challenge in the diagnosis of *P. aeruginosa* using standard culturing methods [13].

In comparison, the role of Enterobacterales in chronic wounds has been much less characterized [14–17]. Studies have shown that geography seems to play a role in the estimation of their etiological relevance [18]. It was shown that skin colonization with Gram-negative bacteria is frequent in resource-limited (sub)tropical settings [19–21], in contrast to skin colonization of individuals from industrialized countries, where Grampositive bacteria dominate [19]. Temperature and moisture have been discussed as likely reasons for the difference seen [22].

Isolation of potentially pathogenic bacteria from non-sterile sites like wounds does not necessarily indicate clinical relevance, which poses challenge to clinical interpretation.

In a recent study that focused on the overall bacterial composition of chronic wound infections in Ghana, from which the isolates for the present molecular analysis were taken, Enterobacterales and *Pseudomonas aeruginosa* constituted the majority of isolated bacterial strains [23]. A moderate proportion of ESBL-positive Enterobacterales suggests lower frequencies of antibiotic resistance [23] than what was recorded from other Ghanaian hospitals [5,24].

Within this study, we aim at characterizing clonal lineages, resistance-associated genetic elements and virulence genes of *P. aeruginosa*, the *Klebsiella pneumoniae* complex and *Escherichia coli*, which were recently isolated from chronic wounds of Ghanaian adult patients [23]. The molecular epidemiology of dominating clonal lineages and associated resistance genes will be assessed. Further, analysis of highly abundant virulence factors will be conducted.

#### **2. Results**

#### *2.1. Clustering Based on Core Genome Multilocus Sequence Typing (cgMLST) Results*

Of the 49 *P. aeruginosa* analyzed, a total of nine clusters comprising isolates without any recorded differences (*n* = 2) or with one or two alleles difference (*n* = 7) were found, suggesting closely related phylogeny (Figure 1). In addition to the clusters, 31 singletons with differences ranging from 80 to 3584 alleles were observed. MLST sequence types (ST) are indicated in Figure 1 and Tables A1 and A2. Cluster sequence types included the following: ST244, ST245, ST381, ST554, ST856, ST1485, ST2033, ST3227 and ST3590.

No clusters were identified among the 21 assessed *K. pneumonia* complex members, which were all singletons with differences ranging from 647 to 2244 alleles. *K. pneumoniae* complex sequence types are summarized in Figure 2. From the 12 *E. coli* isolates, three isolates in a cluster of close phylogenetic relationship were found (1× no allelic differences, 1 × 1 allele difference) (Figure 3). In addition to the cluster observed, nine singletons with differences ranging from 41 to 2365 alleles were recorded. The sequence type of the cluster was ST132 (Pasteur MLST scheme). Sequence types of all *E. coli* isolates are illustrated in Figure 3.

**Figure 1.** Minimum spanning tree of *P. aeruginosa* based on 3867 targets (core genome). Isolate numbers are found within the nodes, and numbers between nodes indicate the number of different alleles. Isolates within clusters are colored based on MLST sequence type (ST). The ST types of white nodes are indicated in Table A1.

#### *2.2. Identified Molecular Resistance Mechanisms in Correlation to Previous Phenotypic Antibiotic Resistance*

Table 1 summarizes acquired antimicrobial resistance determinants for *E. coli* and acquired genes mediating tolerance to disinfectants. Data for *P. aeruginosa* and *K. pneumoniae* are presented in Tables A1 and A2. Tables A3–A8 summarize the phenotypic resistance results as previously recorded [23].

**Figure 2.** Minimum spanning tree of *K. pneumoniae* complex based on 2358 targets (core genome). Isolate numbers are found within the nodes, and the numbers between the nodes indicate the number of different alleles. Colors demonstrate the MLST sequence type of the isolates.

**Figure 3.** Minimum spanning tree of *E. coli* based on 2513 targets (core genome). Isolate numbers are found within the nodes, and the numbers between the nodes indicate the number of different alleles. Colors demonstrate the Pasteur sequence type of the isolates.

In the present study, phylogenetically identical or almost identical isolates also carried the same resistomes. All *E. coli* strains harbored acquired beta-lactamase genes with the majority coding for small spectrum beta-lactamases such as *bla*TEM-1 or *bla*OXA-1. Only four strains carried the gene for an ESBL, in all cases *bla*CTX-M-15. Among the *K. pneumoniae* complex strains, two belonged to the species *K. variicola*, one to the species *K. quasipneumoniae* and the remaining to the species *K. pneumoniae* sensu stricto as reflected by intrinsic *bla*LEN, *bla*OKP and *bla*SHV-1 like, respectively. Genes coding for ESBL (*bla*CTX-M-15) were found solely in four out of 18 *K. pneumoniae* sensu stricto strains that also displayed resistance to oxyimino cephalosporins. In addition, several *K. pneumoniae* complex strains harbored *bla*TEM-1, single strains also contained *bla*OXA-1 and *bla*SCO-1.

With respect to *P. aeruginosa*, only one strain harbored acquired beta-lactamase genes (*bla*TEM-1 and *bla*SCO-1). Increased minimum inhibitory concentrations (MICs) for carbapenems as observed in some *P. aeruginosa* strains were neither explained by matching acquired carbapenemase genes nor by full sequence analysis of the *oprD* gene. The associated amino acid sequences are shown in Figure A1. As indicated, the complete *oprD* gene was found in all 49 *P. aeruginosa* isolates; there was no evidence of protein truncation by premature stop of translation. The 49 isolates could be divided into 7 subgroups according to the protein sequence of the oprD protein, which differ in a total of 30 individual amino acid exchanges and in a single 12aa/10aa-stretch. Therefore, genotypic assessment could not identify the reason for the single carbapenem-resistant *P. aeuroginosa* isolate 088 (ST 1682).


**Table 1.** Analysis of antimicrobial resistance determinants, ordered by strain and MLST type, of the assessed*E. coli* isolates. ST = Sequence type.

\**sitABCD*= peroxides resistance, *qacE*= quaternary ammonium compounds resistance.

Other frequently detected resistance genes in *P. aeruginosa* were the fosfomycin resistance gene *fosA*, the chloramphenicol resistance gene *catB7*, the aminoglycoside resistance gene *aph(3*′ *)-IIb* and the fluoroquinolone-resistance gene *crpP*. In the *Klebsiella pneumoniae* complex isolates, single amino acid exchanges and the fosmomycin resistance gene *fosA* were frequent. Various fluoroquinolone resistance genes and disinfectant tolerance mediating genes also quantitatively dominated. Finally, a broad spectrum of acquired genes causing resistance to the assessed classes of antimicrobial drugs and tolerance to disinfectants was observed in the *E. coli* strains.

#### *2.3. Identified Molecular Virulence Mechanisms*

Table 2 summarizes the analysis of virulence-related genes in *E. coli* (without genes mediating enteropathogenicity). Data for *P. aeruginosa* and *K. pneumoniae* are presented in Tables A9 and A10.

**Table 2.** Analysis of virulence determinants, ordered by strain and MLST type, of the assessed *E. coli* isolates. ST = Sequence type.



**Table 2.** *Cont.*

The virulence-associated gene *exoU*, which has been described in association with the *P. aeruginosa* high-risk clone ST 135 [25], was recorded three times, associated with ST 135 (sample ID 296), ST 532 (sample ID 310) and ST 2483 (sample ID 22), respectively. Based on a Kleborate assessment, a positive virulence score was calculated for 7 out of 21 *K. pneumoniae* strains, comprising the known high-risk clones ST 17 (sample IDs 177, 199) and ST 152 (sample ID 100) [26], next to the clones ST 4 (sample ID 146), ST 6 (sample ID 214), ST 36 (sample ID 267) and ST 39 (sample ID 73), respectively. With focus on some important virulence associated genes in *Klebsiella* spp., *ybt* genes were detected in the abovementioned 7 samples, *iroE* was recorded in all 21 strains, while *clb* or *rpmA* genes were not detected.

Iron-uptake-related genes were numerous in all analyzed bacterial strains. For *P. aeruginosa* and *K. pneumoniae*, various secretion system-associated genes were found. Immune evasion-related genes were highly abundant in *K. pneumoniae* but not in *E. coli* isolates. Adherence-related genes were numerous in *P. aeruginosa* and in *K. pneumoniae* but not in *E. coli*.

Numerous invasion-associated genes were detected in *E. coli*, antiphagocytosisassociated genes were found in *P. aeruginosa*, and biofilm-associated genes in *K. pneumoniae*.

Less frequently detected were: toxin genes in *E. coli* and *K. pneumoniae*, protease genes in *E. coli* and *P. aeruginosa*, regulation genes in *P. aeruginosa* and *K. pneumoniae*, biosurfactant and pigment genes in *P. aeruginosa* and nutrition factor, efflux pumps and serum resistance genes in *K. pneumoniae*.

#### **3. Discussion**

Within this study, we aimed at filling information gaps on the molecular epidemiology of Gram-negative bacteria from chronic infected wounds in rural Ghana. Phylogenetic analyses based on core genome comparison indicated a high clonal diversity of the woundassociated isolates. Clonal clusters were restricted to nine *P. aeruginosa* clusters and one *E. coli* cluster, most likely indicating nosocomial transmission, which has most likely occurred in the wound dressing room that patients' visit on a weekly basis.

ST 135 and ST 244, which are among the worldwide top 10 *P. aeruginosa* high-risk clones [25], were found among the *P. aeruginosa* wound isolates. In detail, one ST 135 *Pseudomonas aeruginosa* isolate was detected, carrying the beta-lactamase-encoding genes *bla*TEM-1B and *bla*SCO-1 and an *exoU* gene, next to five ST 244 without acquired betalactamases. Focusing on known pathogenic *K. pneumoniae* clones [26], two ST 17 strains, a clone reported to be associated with carbapenem-resistance, and one ST 152 strain, a clone known from the Caribbean as common carrier of multiple resistance genes, were detected. Strains carrying the *ybt* and *iro* genes were also identified as high-risk clones by the Kleborate software. From the observed *E. coli* ST types, none have been previously reported as being associated with pathogenic clones so far [27].

In line with the phenotypical antibiotic resistance results previously published [23], numerous acquired resistance determinants were detected in the bacterial strains under investigation. Focusing on the few observed clusters, comparable resistome compositions point towards recent nosocomial transmission. The gene *bla*CTX-M-15 was identified as the determinant of the detected extended spectrum beta-lactamase (ESBL) expression in ESBL positive Enterobacterales [23]. This is in line with previous reports from both human and livestock-associated ESBL positive Enterobacterales in Ghana [28–34]. In *P. aeruginosa* and *K. pneumoniae*, *bla*SCO-1, which has initially been described from an *Acinetobacter baumannii* isolate from Argentina [35], was observed. Beta-lactamases with high hydrolytic effects on carbapenems were lacking, the same applies to protein truncation by premature stop of translation of the *oprD* gene in *P. aeruginosa*. Accordingly, the genetic background of carbapenem resistance of a single *P. aeruginosa* strain could not be resolved, although downregulation of *oprD* expression due to mutations outside of the gene or *amp*C (class C betalactamase) overexpression could not be excluded as likely reasons.

Substance-specific genes and genes encoding efflux pumps mediating tolerance to disinfectants were observed in Enterobacterales. Therefore, further monitoring of the spread of disinfectant tolerance-associated genes and the effects of their abundance on disinfectant-based skin and wound decolonization strategies [36] seem advisable.

The importance of highly abundant virulence factors like iron-uptake- and secretion system-related genes in *P. aeruginosa* is comprehensively described in the literature [37,38]. Other genes reported in the literature like regulation-associated virulence genes, recently reported, were less frequently observed in our isolates [39,40]. However, due to lacking information on the individual etiological relevance of each isolate, any association with clinical effects remains speculative.

Further limitations of this study include a rather small sample size and the lack of a comparison strain collection containing isolates from other clinical specimens and environmental strains. Accordingly, the interpretation of the etiological relevance of individual strains remains challenging and is clearly beyond the scope of this work.

In summary, a broad spectrum of Gram-negative clones was isolated from the chronic wounds of the Ghanaian patients. Thereby, known high-risk clones [25–27] played only a minor role. Observed resistance patterns and mechanisms were in line with the spectrum expected from previous reports [23,28–34].

#### **4. Materials and Methods**

#### *4.1. Sample Collection, Bacterial Culture and Antibiotic Susceptibility Testing*

Single patient strains of *P. aeruginosa, E. coli* and *K. pneumoniae* complex were isolated from patients ≥15 years with an infected chronic wound at the Outpatient Department (OPD) of the Agogo Presbyterian Hospital, in the Asante Akim North District of rural Ghana. Patients typically visit the wound dressing room of the OPD on a weekly basis. Sampling was performed from January 2016 to November 2016. Sample collection and microbiological investigations were reported previously [23]. Antibiotic susceptibility was tested by the disk diffusion method and interpreted following the European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines v.6.0 (http://www.eucast.org

(accessed on 15 January 2016)). Bacterial strains and antibiotic susceptibility were confirmed using the VITEK2 System. Those data have been published before [23].

#### *4.2. DNA Isolation and Whole Genome Sequencing*

Bacterial DNA was isolated using the MasterPure Complete DNA and RNA Purification Kit (LGC standards GmbH, Wesel, Germany) and sent for whole genome sequencing (WGS) to BGI Europe, Denmark, Copenhagen. A BGISEQ-500 device was used for sequencing, generating 2 × 150 bp paired-end reads with an aimed coverage of 100×. Original raw data were upload for public use to the short-read archive (SRA, NCBI) under the accession number PRJNA699140. Details on the strain-specific SRA accession numbers are provided in Table A11.

#### *4.3. Whole Genome Sequencing and Data Analysis*

All raw data passed quality control using FASTQC v.0.11.4 [41] and were used for further analysis. Taxonomic classification and contamination check of raw-reads was performed using KRAKEN2 v.2.0.8-beta [42]. Phylogenetic analysis based on core genome multi locus sequence typing (cgMLST) analysis was performed using the commercial software SeqSphere+ v. 7.2.0 (Ridom GmbH, Münster, Germany) [43]. The software pipeline included assessment of read data and adapter control using FASTQC followed by genome assembly using the internally provided assembler Velvet, applying default settings. The reference genomes NC\_000913.3 (*E. coli*), NC\_002516.2 (*P. aeruginosa*) and NC\_01273.1 (*K. pneumoniae* species complex) were used for cgMLST analyses. Only samples with a ration of "good cgMLST targets" higher than 90% were included in the phylogenetic analysis. Novel cgMLST-based complex types (CT) were automatically assigned by the SeqSphere software. Unknown alleles and profiles of MLST genes were submitted to pubmlst.org or Institute Pasteur to establish novel sequence types (ST). Isolates were defined to be clonally identical with allele differences less than four. Moreover, raw data were assembled with SPAdes v3.13.11 [44] using the careful option. Scaffolds shorter than 500 bp or with a coverage smaller than ten were sorted out, using an in-house script. Abricate v.0.9.9 [45] was used to screen for resistance and virulence genes in SPAdes assembly files, using NCBI AMRFinderPlus [46] and VFDB [47] as reference databases (both updated 6 November 2020), respectively. Additionally, SPAdes assemblies were uploaded to ResFinder4.1 [48] to obtain WGS predicted phenotypes against different antimicrobials by using default settings (%ID > 90, minimum length > 60%) and to Kleborate to predict virulence genes in *Klebsiella* isolates.

#### *4.4. Ethical Considerations*

The Committee on Human Research, Publications and Ethics, School of Medical Science, Kwame Nkrumah University of Science and Technology in Kumasi, Ghana, approved this study (approval number CHRPE/AP/078/16).

#### **5. Conclusions**

In conclusion, this study provides a molecular insight into the epidemiology of Gramnegative bacteria isolated from chronic wound infections from patients in rural Ghana. Epidemiological data that focus on the distribution and spread of antimicrobial resistance determinants and associated virulence factors in resource-limited settings are scarce. Although the study is a small cross-sectional assessment, which cannot replace continuous surveillance programs, it might provide a glimpse of prevailing Gram-negative bacteria isolated from wound infections in this area of Ghana. Considering the ongoing need for resistance and virulence surveillance in tropical regions, larger future studies are desirable. **Author Contributions:** U.L., D.D. and J.M. designed and coordinated this study. T.T., F.P. and S.T. performed bioinformatic analysis. M.L. supported the management of this study. A.J. managed the data collection. H.F., D.D. and U.L. wrote the first draft of this manuscript. K.O. conducted and supervised fieldwork. C.W.A. and K.T. conducted and supervised lab work. M.K. and S.S. supported the interpretation of the results, writing and editing the manuscript. All authors read and approved the final manuscript.

**Funding:** This study was funded by institutional funds of the Bernhard Nocht Institute for Tropical Medicine (BNITM).

**Institutional Review Board Statement:** The study was conducted according to guidelines of the Declaration of Helsinki. The Committee on Human Research, Publications and Ethics, School of Medical Science, Kwame Nkrumah University of Science and Technology in Kumasi, Ghana, approved this study (approval number CHRPE/AP/078/16).

**Informed Consent Statement:** Informed consent was obtained from all study participants.

**Data Availability Statement:** All relevant data have been provided in the paper and its Appendix A materials. Raw data are available applying the links as indicated in the methods chapter and can also be provided by the authors on reasonable request.

**Acknowledgments:** We thank all patients that participated in this study and the staff at the Agogo Presbyterian Hospital. Without their support, this research study would not have been possible. We thank the team of curators pubmlst.org and the Institute Pasteur MLST and whole genome MLST databases for curating the data and making them publicly available at http://bigsdb.pasteur.fr/ (accessed on 22 March 2021).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Appendix A**


**Figure A1.** Clustal omega multiple alignment of oprD proteins—one example for the 7 detected subgroups.


**Table A1.**Analysis of antimicrobial resistance determinants, ordered by strain and MLST type, of the assessed*P. aeruginosa*isolates. ST = Sequence type.

**Table A1.** *Cont.*


**Table A1.** *Cont.*


Acquired resistance genes for macrolides, rifampicin, resistance-associated point mutations, genes for efflux pumps or genes mediating tolerance against disinfectants were not detected.


**Table A2.** Analysis of antimicrobial resistance determinants, ordered by strain and MLST type, of the assessed*K. pneumoniae*isolates. ST = Sequence type.


**Table A2.** *Cont.*


**Table A2.** *Cont.*

65


**Table A2.** *Cont.*


**Table A2.** *Cont.*


**Table A2.** *Cont.*


**Table A2.** *Cont.*


**Table A2.** *Cont.*


**Table A2.** *Cont.*

71


**Table A2.** *Cont.*

\**qacE*= quaternary ammonium compounds resistance and*oqxB*and *oqxA*= efflux pumps mediating resistance against disinfectants.


**Table A3.** Phenotypic resistance the *P. aeruginosa* strains. Data are missing for strains 198, 218 and 312, due to loss during subcultivation. MIC = minimum inhibitory concentration. N.a. = value missing due to loss of strain or failed reaction.


**Table A3.** *Cont.*


**Table A4.** Phenotypic resistance of *P. aeruginosa* strains. Data are missing for strains 198, 218 and 312 due to loss during subcultivation. MIC = minimum inhibitory concentration. N.a. = value missing due to loss of strain or failed reaction.


**Table A4.** *Cont.*


**Table A5.**Phenotypic resistance of the*Klebsiella*strains. MIC = minimum inhibitory concentration. ESBL = signal in phenotypic testing for extended-spectrum beta-lactamases.


**Table A6.**Phenotypic resistance of the*Klebsiella*strains. MIC = minimum inhibitory concentration. ESBL = signal in phenotypic testing for extended-spectrum beta-lactamases.


**Table A7.** Phenotypic resistance of*Escherichia coli*strains. MIC = minimum inhibitory concentration. ESBL = signal in phenotypic testing for extended-spectrum beta-lactamases.



80


**Table A9.** Analysis of virulence determinants, ordered by strain and MLST type, of the assessed*P. aeruginosa*isolates. ST = Sequence type.

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 032 ST 3587 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 069 ST 360 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 081 ST 244 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA like, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW like, xcpX, xcpY, xcpZ* 082 ST 514 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimV, pilB, pilD, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 088 ST 1682 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, xcpA/pilDalg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 099 ST 244 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 106 ST 1521 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdD, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 114 ST 244 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA like, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW like, xcpX, xcpY, xcpZ* 137 ST 3014 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA like, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 144 ST 245 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 147 ST 245 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilDalg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

**Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 149 ST 381 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimV, pilA, pilB, pilD, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 153 ST 704 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA plcH lasI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 154 ST 244 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA like, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilDalg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW like, xcpX, xcpY, xcpZ*

**Table A9.** *Cont.*

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 157 ST 2616 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimV, pilA, pilB, pilD, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 160 ST 170 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimV, pilA, pilB, pilD, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 162 ST 274 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimV, pilA, pilB, pilD, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilC like, xcpA/pilDalg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 180 ST 856 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 183 ST 244 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA like, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW like, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

89

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 186 ST 3588 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA like, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 190 ST 871 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 195 ST 988 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilDalg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

**Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 196 ST 2475 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 198 ST 2476 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimV, pilB, pilD, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 204 ST 639 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilDalg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.** *Cont.*

91

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 208 ST 132 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA like, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 218 ST 856 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 229 ST 270 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilDalg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdD, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

**Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 233 ST 3227 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimV, pilB, pilD, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 236 ST 266 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 238 ST 3589 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA plcH lasI, rhlI xcpP, xcpQ, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.** *Cont.*

93

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 242 ST 3590 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3 like, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 243 ST 3590 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 272 ST 2033 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilDalg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 274 ST 2033 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 278 ST 988 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilDalg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 282 ST 554 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA like, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 285 ST 554 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilA like, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilDalg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 289 ST 1485 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdD, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 290 ST 1485 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdD, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 296 ST 235 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilW, pilX pilY1, pilY2, pilC, xcpA/pilDalg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW like, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 298 ST 3227 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimV, pilB, pilD, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 301 ST 3593 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 302 ST 1755 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC like, xcpA/pilDalg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*

 **Sample ID ST-Type Pathogenicity Factor Groups Adherence Anti- Phagocytosis Biosurfactant Iron Uptake Pigment Protease Toxin Regulation Secretion System** 309 ST 3592 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimT, fimU, fimV, pilB, pilD, pilE, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilV, pilW, pilX pilY1, pilY2, pilC, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdE phzM, phzS aprA, lasA plcH lasI, rhlI xcpP, xcpQ, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 310 ST 532 *waaA, waaC, waaF, waaG, waaP, chpA, chpB, chpC, chpD, chpE, fimV, pilB, pilD, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilC like, xcpA/pilD alg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, fpvA pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA phzM, phzS aprA lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ* 312 ST 381 *waaA, waaC, waaF, waaG, waaP, wzy, wzz, chpA, chpB, chpC, chpD, chpE, fimV, pilB, pilD, pilF, pilG, pilH, pilI, pilK, pilM, pilN, pilO, pilP, pilQ, pilR, pilS, pilT, pilU, pilC like, xcpA/pilDalg44, alg8, algA, algB, algC, algD, algE, algF, algG, algI, algJ, algK, algL, algP/algR3, algQ, algR, algU, algW, algX, algZ, mucA, mucB, mucC rhlA, rhlB fptA, pchA, pchB, pchC, pchD, pchE, pchF, pchG, pchH, pchI, pchR, pvdA, pvdD, pvdE phzM, phzS aprA, lasA toxA, plcH lasI, rhlI xcpP, xcpQ, xcpR, xcpS, xcpT, xcpU, xcpV, xcpW, xcpX, xcpY, xcpZ*

**Table A9.***Cont.*


**Table A10.** Analysis of virulence determinants, ordered by strain and MLST type, of the assessed*K. pneumoniae*isolates. ST = Sequence type.


**Table A10.** *Cont.*


**Table A10.** *Cont.*


**Table A10.** *Cont.*


**Table A10.** *Cont.*


**Table A11.**Details on the strain-specific short-read archive (SRA) accession numbers.


**Table A11.** *Cont.*


**Table A11.** *Cont.*


**TableA11.***Cont.*

#### **References**


## *Article* **Clonal Lineages, Antimicrobial Resistance, and PVL Carriage of** *Staphylococcus aureus* **Associated to Skin and Soft-Tissue Infections from Ambulatory Patients in Portugal**

**Carolina Ferreira 1 , Sofia Santos Costa 1 , Maria Serrano 1 , Ketlyn Oliveira 1 , Graça Trigueiro 2 , Constança Pomba 3,4 and Isabel Couto 1, \***


**Abstract:** *Staphylococcus aureus* (*S. aureus*) is a leading cause of skin and soft-tissue infections (SSTIs) in the community. In this study, we characterized a collection of 34 *S. aureus* from SSTIs in ambulatory patients in Portugal and analyzed the presence of Panton–Valentine leucocidin (PVL)-encoding genes and antibiotic-resistance profile, which was correlated with genetic determinants, plasmid carriage, and clonal lineage. Nearly half of the isolates (15, 44.1%) were methicillin-resistant *Staphylococcus aureus* (MRSA) and/or multidrug resistant (MDR). We also detected resistance to penicillin (33/34, 97.1%), fluoroquinolones (17/34, 50.0%), macrolides and lincosamides (15/34, 44.1%), aminoglycosides (6/34, 17.6%), and fusidic acid (2/34, 5.9%), associated with several combinations of resistance determinants (*blaZ*, *erm*(A), *erm*(C), *msr*(A), *mph*(C), *aacA-aphD*, *aadD*, *aph*(3 ′ )-*IIIa*, *fusC*), or mutations in target genes (*fusA*, *grlA/gyrA*). The collection presented a high genetic diversity (Simpson's index of 0.92) with prevalence of clonal lineages CC5, CC22, and CC8, which included the MRSA and also most MDR isolates (CC5 and CC22). PVL-encoding genes were found in seven isolates (20.6%), three methicillin-susceptible *Staphylococcus aureus* (MSSA) (ST152-*agr*I and ST30-*agr*III), and four MRSA (ST8-*agr*I). Plasmid profiling revealed seventeen distinct plasmid profiles. This work highlights the high frequency of antimicrobial resistance and PVL carriage in SSTIs-related *S. aureus* outside of the hospital environment.

**Keywords:** *Staphylococcus aureus*; skin and soft-tissue infections; antibiotic resistance; clonal lineages; plasmids; Panton–Valentine leucocidin

#### **1. Introduction**

*Staphylococcus aureus* (*S. aureus*) is a major human pathogen responsible for a wide range of infections both in hospitals and in the community. It is one of the main causes of severe nosocomial infections such as bacteremia and infective endocarditis and in the community is a frequent cause of skin and soft-tissue infections (SSTIs) [1]. Besides their potential severity, infections caused by *S. aureus* are usually difficult to treat due to the frequent acquisition of antimicrobial resistance determinants. In the last decades, there has been an emergence and dissemination of methicillin-resistant *S. aureus* (MRSA) as well as of multidrug-resistant (MDR) strains [2,3]. Consequently, MRSA are now included in the

**Citation:** Ferreira, C.; Costa, S.S.; Serrano, M.; Oliveira, K.; Trigueiro, G.; Pomba, C.; Couto, I. Clonal Lineages, Antimicrobial Resistance, and PVL Carriage of *Staphylococcus aureus* Associated to Skin and Soft-Tissue Infections from Ambulatory Patients in Portugal. *Antibiotics* **2021**, *10*, 345. https:// doi.org/10.3390/antibiotics10040345

Academic Editor: Giovanna Batoni

Received: 19 February 2021 Accepted: 19 March 2021 Published: 24 March 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

World Health Organization (WHO) list of high-priority bacteria for development of new drugs [4].

*S. aureus* is the most frequent pathogen associated with SSTIs, which can range from minor or superficial infections such as impetigo to life-threating infections such as necrotizing fasciitis [5]. Topical antibiotics that are often used for the prevention or treatment of milder infections include mupirocin, fusidic acid, neomycin, and bacitracin [6,7]. The use of some of these topical antibiotics is particularly relevant in the community/ambulatory settings, where they may not require medical prescription. Other antibiotics for systemic use, such as clindamycin, trimethoprim-sulfamethoxazole, tetracyclines, and linezolid, are also indicated for treatment of severe forms of SSTIs caused by *S. aureus* [6,8]. The frequency of antibiotic-resistant *S. aureus* isolates associated with SSTIs is rising worldwide [9–12], particularly to fusidic acid and mupirocin, which is probably linked with the widespread use of these antibiotics [3].

Resistance to antibiotics in *S. aureus* can be mediated by several mechanisms, such as antibiotic modification or degradation, target mutation, or antibiotic efflux. Resistance to penicillins can occur by inactivation of the antibiotic molecule through the action of the β-lactamase BlaZ. The *blaZ* gene occurs frequently in *S. aureus* clinical isolates. Resistance to penicillins and other β-lactams, with the exception of fifth-generation cephalosporins, is mediated by the acquisition of the *mecA* gene, which is part of the mobile genetic element SCC*mec* (staphylococcal cassette chromosome *mec*) and encodes for an additional penicillinbinding protein, PBP2a, with low affinity for the β-lactam antibiotics [3]. Resistance to macrolides and lincosamides can occur through several mechanisms, including the acquisition of rRNA methylases-encoding *erm* genes that methylate the binding site of the antibiotics [3]. Resistance to aminoglycosides is associated with acquisition of several genes, like *aacA-aphD* or *aadD* that encode enzymes that modify the antibiotic molecule rendering it inactive [3]. Resistance to fluoroquinolones is usually linked to the occurrence of mutations in the quinolone-resistant determining region (QRDR) of the *grlA/B* and *gyrA/B* genes that encode the DNA topoisomerase IV and DNA gyrase, respectively. Fluoroquinolone resistance can also be conveyed by overexpression of chromosomally-encoded efflux pump genes such as *norA/B/C* and *mepA* [13]. Resistance to fusidic acid can be achieved by the acquisition of the *fusB/C* genes that encode ribosomal protection proteins or by mutations in the *fusA* gene [3].

*S. aureus* produces several virulence factors, including toxins, proteins associated with immune evasion, and tissue-degrading enzymes [1]. The cytotoxin Panton–Valentine leucocidin (PVL), encoded by the genes *lukF-PV* and *lukS-PV* carried on bacteriophage ϕSa2, is a two-component pore-forming protein that has been strongly associated with *S. aureus* isolates causing skin infections in the community and with necrotizing pneumonia [1]. Nevertheless, the role of PVL in *S. aureus* infection pathogenesis is still not fully elucidated [14]. The *S. aureus* accessory gene regulator *(agr*) locus regulates the expression of several virulence factors like cell-wall-associated and extracellular proteins, contributing to infection severity and persistence. The polymorphism of the *agr* locus allows the classification of *S. aureus* in four predominant *agr* types (I to IV), that may differ in terms of infection type, carriage of virulence factors, and temporal patterns of autoinduction [15].

Most antimicrobial resistance and virulence genes of *S. aureus* are located on mobile genetic elements (MGEs) such as plasmids, bacteriophages, pathogenicity islands, transposons, integrative conjugative elements (ICEs), integrons, and staphylococcal chromosome cassettes (SCCs), which make up to 15–20% of its genome [16]. The acquisition of antimicrobial resistance by *S. aureus* is mostly due to horizontal gene transfer (HGT), and plasmids have been identified as one of the main responsible for the dissemination of resistance genes [17].

Several studies have evaluated the main clones of MRSA circulating both in hospitals and in the community in Portugal, a country with a high prevalence of MRSA [18–22]. However, there have been fewer studies focusing on *S. aureus*, both methicillin-susceptible *Staphylococcus aureus* (MSSA) and MRSA causing SSTIs. The aim of this work was to perform a phenotypic and genotypic characterization of a collection of *S. aureus* isolated from SSTIs in ambulatory patients and to assess their virulence determinants and susceptibility to the main antibiotics used in SSTI therapeutics, correlating their resistance profile to genetic determinants and identifying their main mechanisms of dissemination among *S. aureus* strains.

#### **2. Results**

#### *2.1. Antimicrobial Susceptibility Profile and Correlation with Resistance Determinants*

The antimicrobial susceptibility profile of the 34 isolates is described in Table 1. Resistance to penicillin was detected in 97.1% (33/34) of the isolates, and 44.1% (15/34) were MRSA (*mecA*<sup>+</sup> and cefoxitin resistant). We have also observed resistance to the fluoroquinolones ciprofloxacin and moxifloxacin (50.0%, 17/34), erythromycin (44.1%, 15/34), clindamycin (35.3%, 12/34) either constitutive (2.9%, 1/34) or inducible (32.4%, 11/34), kanamycin (17.6%, 6/34), tobramycin (14.7%, 5/34), amikacin (8.8%, 3/34), gentamycin (2.9%, 1/34), and fusidic acid (5.9%, 2/34). Fifteen isolates (44.1%) were MDR, mainly resistant to β-lactams, fluoroquinolones, macrolides, and lincosamides. All isolates were susceptible to tetracyclines, tigecycline, rifampicin, trimethoprim-sulfamethoxazole, linezolid, chloramphenicol, retapamulin, and quinupristin-dalfopristin. All MRSA isolates were susceptible to ceftaroline. Only one isolate was susceptible to all antibiotics tested. Although no breakpoints or epidemiological cut-off values are established by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) for bacitracin or neomycin (for 30 µg discs), one isolate showed no inhibition zone toward each of these topical antibiotics. The presence of antibiotic-resistance determinants was confirmed for all isolates presenting phenotypic resistance (Table 1). The *blaZ* gene was detected in all isolates resistant to penicillin. All isolates showing resistance to cefoxitin harbored the *mecA* gene. Mutations in QRDR regions of *grlA* and *gyrA* genes were found in different combinations in all the representative fluoroquinolone resistant isolates screened. Resistance to macrolides and lincosamides was associated with *erm*(A), *erm*(C), *msr*(A), and/or *mph*(C). Resistance to aminoglycosides was mainly linked to the *aadD* gene. For the two isolates resistant to fusidic acid, one harbored the *fusC* gene, whereas the other carried three mutations in the *fusA* gene.

**Table 1.** Antimicrobial resistance phenotypes of the 34 *S. aureus* included in this study and correlation with resistance determinants.


PEN: penicillin; CXI: cefoxitin; ERY: erythromycin; CLI: clindamycin; CIP: ciprofloxacin; MOX: moxifloxacin; KAN: kanamycin; GEN: gentamycin; TOB: tobramycin; AMI: amikacin; and FUS: fusidic acid.

#### *2.2. Efflux Activity*

The presence of increased efflux activity in the 34 isolates was assessed by different approaches. The minimum inhibitory concentrations (MICs) of ethidium bromide (EtBr) for the entire collection ranged from 2 to 16 µg/mL with a unimodal distribution (data not shown). Eleven isolates presented an EtBr MIC of 16 µg/mL, suggesting increased efflux activity in those isolates. In addition, these 11 isolates were also resistant to fluoroquinolones, a class of antibiotics that is substrate of the main efflux pumps in *S. aureus* [13]. To verify the presence of an efflux-mediated resistance in these 11 isolates, the EtBr and ciprofloxacin (CIP) MICs were determined in the presence of the known efflux inhibitors (EIs) thioridazine (TZ) and verapamil (VER) and compared to their original values (Table 2). A significant decrease (four- to eight-fold) in EtBr MICs was observed for all isolates but one, confirming the presence of increased efflux activity in these isolates. However, none of the isolates carried the plasmid-encoded *qacA/B* or *smr* genes, which code for the efflux pumps QacA/B and Smr, respectively, responsible for the extrusion of EtBr and several biocides. These results indicate that the increased efflux activity present in these isolates may be driven by chromosomally-encoded efflux pumps, like NorA, which extrudes EtBr and biocides but also several fluoroquinolones like ciprofloxacin and norfloxacin [13]. The effect of EIs on CIP MICs was less significant, with MIC reductions of two-fold for the majority of the isolates. This result does not exclude the presence of increased efflux activity associated with fluoroquinolone resistance, since these isolates harbor mutations in the QRDR of *grlA* and *gyrA* genes, which are responsible for conferring high-level fluoroquinolone resistance and thus may be hindering the screening of efflux activity associated with resistance to these antibiotics [23].


**Table 2.** The effect of the efflux inhibitors thioridazine and verapamil on ethidium bromide and ciprofloxacin MICs for selected *S. aureus* isolates.

MIC: minimum inhibitory concentration; EtBr: ethidium bromide; CIP: ciprofloxacin; TZ: thioridazine; VER: verapamil. Bold-type numbers indicate MIC reductions ≥ four-fold in the presence of EIs when compared to the original MIC values.

#### *2.3. Main Clonal Lineages and Genetic Diversity of the S. aureus Isolates*

Analysis of *Sma*I-macrorestriction profiles revealed the presence of 15 pulsed-field gel electrophoresis (PFGE) types (A to O) and 18 subtypes (Figure 2) among the collection studied. The three most common profiles, PFGE types G, N, and E, are represented by seven, five, and four isolates, respectively. An isolate representative of each PFGE type was selected for typing by multilocus sequence typing (MLST). Fourteen sequence types (STs) were identified belonging to 10 clonal complexes. The clonal complexes identified were CC5 (ST5, ST105, and the newly identified ST6531, which is a single-locus variant (SLV) of ST5), CC8 (ST8, ST72), CC152 (ST152), CC30 (ST30), CC7 (ST7), CC97 (ST97), CC15 (ST15), CC25 (ST25), CC22 (ST22), and CC45 (ST278). We also detected a newly identified singleton, ST6564. In general, each ST identified was associated with a single PFGE type,

except for ST5 (CC5), associated with PFGE types F and D and ST8 (CC8) associated with PFGE types C and E. The most common PFGE types were linked to ST105 (CC5), ST22 (CC22), and ST8 (CC8). The Simpson's index of diversity (SID), calculated based upon the PFGE *Sma*I-macrorestriction profiles, revealed a highly diverse *S. aureus* population (SID = 0.92, CI: 0.87–0.98).

#### *2.4. Correlation of Strain Lineage with agr Typing and PVL Carriage*

The *agr* typing of the *S. aureus* isolates identified *agr*I as the predominant type, which was detected in 20 out of the 34 isolates (58.8%), followed by *agr* type II, identified in 13/34 (38.2%) isolates, and *agr* type III, observed in a single isolate (1/34, 2.9%). No isolate of *agr* type IV was identified. The PVL-encoding genes *lukS-lukF* were detected in seven isolates (20.6%), corresponding to three MSSA (3/19, 15.8%) and four MRSA (4/15, 26.7%), all classified as *agr* type I or III.

As shown in Figure 2, an association was observed between *S. aureus* clonal lineages ST8 (CC8), ST25 (CC25), ST22 (CC22), ST7 (CC7), ST278 (CC45), ST97 (CC97) and ST152 (CC152) and *agr* type I, whereas clonal complexes CC5 and CC15 were linked to *agr* type II, and the singe isolate harboring *agr* type III belonged to ST30 (CC30). The newly identified singleton ST6564 belongs to *agr* type I.

Carriage of PVL was associated with MRSA belonging to ST8 (CC8) and MSSA assigned to ST152 (CC152) or ST30 (CC30).

#### *2.5. Correlation of Strain Lineage with Antimicrobial Resistance and Plasmid Profiles*

Analysis of the methicillin resistance status and clonal lineage showed that the MRSA isolates identified in the collection were restricted to the clonal complexes CC22, CC8 (ST8), and CC5 (ST5 and ST105) (Figure 1). Most isolates from ST22 and clonal lineages of the CC5 presented MDR phenotypes (Figure 1).

ree plasmids. Large plasmids (≥

–

**Figure 1.** Relation of clonal lineages identified amongst the *S. aureus* associated with SSTIs in ambulatory patients determined using PHYLOViZ software and correlation with (**A**) methicillin resistance status, PVL carriage, and *agr* type; and (**B**) MDR phenotypes and plasmid profile. In panel (**A**), MRSA isolates are displayed in orange whereas MSSA isolates are shown in light blue. In panel (**B**), MDR isolates are presented in red, while non-MDR isolates are shown in green.

(10 kb or ≤3


**Figure 2.** *Sma*I-PFGE macrorestriction profile analysis of the *S. aureus* isolates associated with SSTIs in ambulatory patients and corresponding clonal lineages as determined by MLST and their correlation with PVL carriage and *agr* types, plasmid profiles, and phenotypic and genotypic resistance traits. The pairs of isolates recovered from different anatomical sites of the same patient are marked by (A) to (F), where each letter corresponds to a different patient. The dendrogram was built using Bionumerics and the UPGMA algorithm, using Dice coefficient, and an optimization of 0.5% and tolerance of band of 1%. The dashed lines correspond to the similarity criteria for considering isolates belonging to the same PFGE type ( ≥81%) or subtype ( ≥97%). Isolates sharing the same PFGE type or subtype were considered as belonging to the same sequence type (ST). The isolates subjected to MLST are indicated in bold-type. Each plasmid profile corresponds to a unique pattern of undigested and/or *Eco*RI-digested plasmids. CC: clonal complex; ST: sequence type; PFGE: pulsed-field gel electrophoresis; PVL: Panton–Valentine leucocidin; PEN: penicillin; CXI: cefoxitin; ERY: erythromycin; CLI: clindamycin; CIP: ciprofloxacin; MOX: moxifloxacin; KAN: kanamycin; GEN: gentamycin; TOB: tobramycin; AMI: amikacin; and FUS: fusidic acid. Resistance mutations: GrlA1: S80Y; GrlA2: S80F; GrlA3: S80Y, E84G; GyrA1: S84L; GyrA2: E88K; FusA1: A71V, H547Q, G476C.

The majority of the *S. aureus* isolates studied carried plasmids (28/34, 82.4%), with 19 isolates (55.9%) carrying one plasmid, eight isolates (23.5%) carrying two plasmids, and only one isolate (2.9%) carrying three plasmids. Large plasmids (≥23 kb) were present in most isolates (24/34, 70.6%), alone or in combination with medium or smaller plasmids (10 kb or ≤3 kb). Isolates with large plasmids harbor, in general, a higher number of resistance determinants than those carrying small or no plasmids (Figure 2).

Seventeen plasmid profiles were identified, designated P1 to P17 (Figure 2). For strains carrying a single plasmid, these profiles were defined after restriction with *Eco*RI–profiles P1 to P10. The most frequent profile, P1, is represented by a single large plasmid (>23 kb), identified in six isolates, five of which belonging to ST105 (CC5). All isolates with this plasmid profile are MDR and carry several resistance genes. The second most frequent profile, P12, is shared by three isolates belonging to CC5 and CC22. Isolates of the same clonal complex show a high variety of plasmid profiles. For example, isolates of CC5, CC8, and CC22 have four different plasmids profiles each (Figure 2).

Of the six pairs of isolates recovered from two anatomical sites of the same patient, only one pair was assigned to two distinct PFGE types. Four pairs of isolates were indistinguishable by PFGE, while the remaining pair included subtypes of the same PFGE type. However, different phenotypical or genotypical trait(s) were observed within each pair except one (Figure 2). Isolates of three pairs differed in plasmid content, while two pairs of isolates differed in terms of resistance profile and/or resistance determinants. Another pair of isolates displayed different resistance profile and determinants although sharing the same plasmid profile.

#### **3. Discussion**

*S. aureus* is a leading cause of bacterial infections not only in healthcare settings but also in the community, many of which are caused by MRSA and MDR strains [2]. According to the most recent data of EARS-Net (European Antimicrobial Resistance Surveillance Network), in 2019, the prevalence of MRSA in bloodstream infections in Portugal was 34.8%. Even though this value has been decreasing over the last decade, it was still the fifth highest registered in Europe [24].

A high frequency of antibiotic resistance was observed in this collection. All isolates except one were resistant to at least one class of antibiotics, mainly β-lactams, and nearly half (44.1%) were MDR, which was unexpected considering they were not from hospitalized patients. However, these isolates were collected from ambulatory patients who could have been under antibiotic therapy or could have had recent contact with hospitals and that might explain the high rates of resistance observed. The 44.1% rate of MRSA identified is higher than the MRSA rates reported in the community (21.6%) in Portugal [19] and in children affected by SSTIs attending a pediatric emergency in Lisbon area (8.6%) [25] in years close to the year of collection of these isolates and is closer to the values observed in hospitals (47.4%) for 2014 [26]. On the other hand, the fact that these patients used laboratory services suggests that these may reflect more complex infections, which may explain the high frequencies of resistance observed [27]. Of the 15 MRSA, 11 (73.3%) were also MDR and the most common pattern was resistance to β-lactams, fluoroquinolones, macrolides, and lincosamides, which is a profile frequently observed in hospital-acquired MRSA (HA-MRSA) [19,20]. Previous studies have shown that there is a high prevalence of HA-MRSA strains in the community in our country, due to dissemination of these strains from the hospital [19–21]. The molecular analysis revealed that most of the isolates studied (16/34, 47.1%) presented genetic backgrounds related to hospital-associated lineages, such as CC5 and CC22, which were the predominant HA-MRSA lineages in Portugal during this period, identified in nosocomial or community isolates [18,19]. This finding, together with the use of community laboratory services to treat possible resilient and complex infections, may explain the high frequency of resistance observed in this collection. Regarding MSSA strains, only four strains showed an MDR profile (4/19, 21%). This observed rate of MDR

strains is higher than previously reported for other MSSA collected from the community in Portugal [25,28].

With the exception of ST278 and the two new STs, all the other strain lineages identified in this work have been found in other studies in Portugal, with ST8 being the most frequent CA-MRSA clone, while ST30 and ST72 were the most prevalent MSSA clones [18,19,29]. ST278 belongs to CC45 and has been reported in the USA as a MSSA clone [30,31]. Strains of CC45 are prevalent in Portugal [18,19,29], but as far as we know, ST278 has not been yet reported in our country. We have also identified two new STs, ST6531, a SLV of ST5, and the singleton ST6564.

Although the *S. aureus* studied were isolated from patients with SSTIs, a low frequency of resistance was observed toward topical antibiotics, particularly to neomycin and fusidic acid, which are some of the most commonly used for the treatment of SSTIs in the community [7]. Only two isolates (5.9%) were resistant to fusidic acid, and only one isolate (2.9%) did not show inhibition zone to neomycin or bacitracin. The current rates of resistance to fusidic acid reported in the literature for SSTIs-associated *S. aureus* vary geographically, ranging from over 30% in Africa [32] to much lower rates, 2 to 6% in Asia or South America [33,34]. These low levels of resistance to fusidic acid are similar to the ones detected in other contemporary studies in Portugal [19]. The rates of resistance to neomycin and bacitracin in our collection are lower than the ones reported for other CA-MRSA from SSTIs [34]. However, a higher frequency of resistance was detected toward clindamycin (35.3%), an antibiotic also recommended for topical treatment of these infections, in comparison with other CA-MRSA from SSTIs [35,36].

Antibiotic-resistance determinants were identified in all isolates presenting phenotypic resistance (Table 1). The distribution of the fusidic-acid-resistance determinants in *S. aureus* reported in the literature is variable. While some studies report that *fusB* and *fusC* are the most prevalent genes [37], others report *fusA* mutations as the most common mechanism of fusidic-acid resistance [38]. In this study, only two isolates were resistant to fusidic acid. One of these had three mutations in the *fusA* gene, two of which (A71V and H457Q) already associated with resistance to this antibiotic [39–41], while the third mutation found, G476C, was described for the first time in this work and could also be contributing to fusidic-acid resistance. The other isolate resistant to fusidic acid carried the *fusC* gene. This is an MSSA that belongs to ST5. Several studies have shown that *fusC* gene can be located in SSC*mec* cassettes, with or without *mecA* gene [42,43].

Screening for mutations in fluoroquinolone-resistant representative isolates identified several patterns of mutations in the QRDR regions of GrlA and GyrA (namely, GrlA S80Y, GyrA S84L; GrlA S80F, GyrA E88K; GrlA S80Y E84G, GyrA E88K; and GrlA S80F GyrA S84L) already associated with high level resistance to these antibiotics [44,45]. These patterns of QRDR mutations were also detected in an earlier study from *S. aureus* clinical isolates in Lisbon [46]. The GrlA S80F and GyrA S84L mutations are the most commonly described in the literature [47–51] and are characteristic of ST22 and some ST8 lineages [51]. In our study, only one isolate carried both mutations and belonged to ST22. The GyrA S84L mutation was also found in one isolate of ST8. The GrlA E84G and S80Y and GyrA E88K mutations are also described in some studies [50,51] but appear to be less frequent. Besides these mutations, the activity of chromosomally-encoded MDR efflux pumps might also be contributing to fluoroquinolone resistance. A subset of fluoroquinolone-resistant isolates presented increased efflux activity of EtBr, a common substrate of MDR efflux pumps like NorA/B/C and MepA, which also extrude fluoroquinolones. No significant reduction in CIP MICs was observed in the presence of EIs, yet the effect of these compounds may be potentially hindered by the presence of QRDR mutations. The absence in this collection of the plasmid-encoded efflux pump genes *qacA/B* or *smr* indicates that the higher EtBr efflux activity detected is probably due to the overexpression of chromosomal efflux pump genes such as *norA/B/C* or *mepA* [13,46]. In the future this, hypothesis can be confirmed be quantifying the expression levels of these genes by RT-qPCR.

Plasmid profiling revealed a high proportion of plasmid-bearing isolates (82.4%) and a high diversity of plasmids, with 17 different profiles identified distributed amongst 14 clonal lineages. Most isolates carried a large plasmid, potentially associated with determinants for resistance to β-lactams, macrolides, lincosamides, and aminoglycosides (Figure 2). These results are similar to the ones found in a previous study that analyzed the plasmid content of a collection of 53 *S. aureus* isolated from a hospital in Lisbon between 2006 and 2007 [52]. The proportion of plasmid-bearing isolates in that study was 83%, and most isolates carried a large plasmid that was frequently associated with resistance to β-lactams, macrolides, and lincosamides. Other studies have also demonstrated that large plasmids are quite common in *S. aureus* and that they can carry several resistance determinants associated with resistance to the classes of antibiotics mentioned above [16].

The occurrence of PVL is linked to the bacteriophage ϕSa2 and generally associated with community-acquired MRSA (CA-MRSA), being traditionally considered a marker for the identification of CA-MRSA isolates [53], although some CA-MRSA strains do not produce this toxin. Its prevalence in HA-MRSA isolates, albeit lower, has been documented in several countries [54]. PVL is also strongly linked with *S. aureus* isolates collected from skin infections [53,54]. The overall rate of 20.6% of PVL-positive isolates in our set of *S. aureus* associated with SSTIs is lower than the ones reported from children with SSTI attending a pediatric emergency (37%) [25] but higher than the ones reported for other MSSA, CA-MRSA, or HA-MRSA collections in Portugal [19,22,55,56], albeit most of these other collections are not exclusively associated with skin infections. PVL carriage in our set of MRSA isolates appears restricted to the ST8-*agr*I clonal lineage, as found in a previous study by Tavares and colleagues [19]. Interestingly, the single MSSA ST8 isolate of our collection did not harbor PVL. The PVL-positive MSSA detected in our collection belong to the genetic backgrounds ST30-*agr*III and ST152-*agr*I, different from the ones reported in that earlier study [19]. However, PVL-positive ST30 isolates were also detected in children with SSTIs attending a pediatric emergency [25]. The ST8 and ST30 clonal lineages were frequently encountered in isolates from the community and less frequently associated with nosocomial isolates [19,20].

In this study, we performed a phenotypic and genotypic characterization of a collection of *S. aureus* isolated from SSTIs in ambulatory patients. Although this can be considered a relatively small sample, this is a convenience collection that represents the diversity of the population affected by SSTIs in an ambulatory setting over a five-months period where the only condition criteria for inclusion of the *S. aureus* isolates was to be SSTI-related. The genetic diversity of this collection was demonstrated by the high value of the Simpson's index (SID of 0.92).

#### **4. Materials and Methods**

#### *4.1. Bacterial Isolates*

The study comprised a collection of 34 *S. aureus* isolates associated with SSTIs of 28 ambulatory patients. Of the 34 isolates, 31 were collected from wounds (legs, *n* = 17; foot, *n* = 5; armpit, *n* = 3; ear, *n* = 1; and from unidentified sites, *n* = 4), and three were collected from ulcers. Six pairs of isolates (*n* = 12) were collected from different anatomical sites (right/left leg, *n* = 6; right/left armpit, *n* = 2; ear/leg, *n* = 2; and unidentified sites, *n* = 2) of six patients. The isolates were collected between February and June of 2014 at a community clinical diagnostic laboratory in Lisbon, Portugal. All isolates were grown in tryptic soy broth (TSB) (Oxoid™, Hampshire, UK), with shaking or tryptic soy agar (TSA) (Oxoid™) at 37 ◦C. Species identification was confirmed by amplification of the *nuc* gene following the protocol described by Poulsen and colleagues [57], using the primers described in Table S1 of Supplementary Data.

#### *4.2. Antimicrobial Susceptibility Testing*

Antimicrobial susceptibility was determined for a panel of 24 antibiotics by disk diffusion in Mueller-Hinton agar (MHA, Oxoid™), according to the EUCAST guidelines [58]. Antibiotics discs were obtained from Oxoid™. The following antibiotic discs (antibiotic content per disc) were used: penicillin (PEN, 1 U), oxacillin (OXA, 1 µg), cefoxitin (CXI, 30 µg), ceftaroline (CPT, 5 µg), ciprofloxacin (CIP, 5 µg), moxifloxacin (MOX, 5 µg), gentamicin (GEN, 10 µg), kanamycin (KAN, 30 µg), tobramycin (TOB, 10 µg), neomycin (NEO, 30 µg), amikacin (AMI, 30 µg), tetracycline (TET, 30 µg), minocycline (MIN, 30 µg), tigecycline (TIG, 15 µg), chloramphenicol (CHL, 30 µg), erythromycin (ERY, 15 µg), clindamycin (CLI, 2 µg), quinupristin/dalfopristin (QD, 15 µg), linezolid (LIN, 10 µg), trimethoprimsulfamethoxazole (TRS, 25 µg), rifampicin (RIF, 5 µg), bacitracin (BAC, 10 U), fusidic acid (FUS, 10 µg), and mupirocin (MUP, 200 µg). The D-zone test was performed for detection of inducible clindamycin resistance, and the penicillin inhibition zone was examined to detect production of β-lactamases. Susceptibility testing to ceftaroline was performed for MRSA isolates only. Susceptibility to retapamulin (RET) was evaluated by determination of MICs by the two-fold microdilution method with cation-adjusted Mueller-Hinton broth (CAMHB, Oxoid™), according to the Clinical and Laboratory Standards Institute (CLSI) guidelines [59]. Retapamulin was acquired in powder form from Sigma-Aldrich (St. Louis, MO, USA), dissolved in dimethyl sulfoxide, and diluted in water with 10% βcyclodextrin [60]. The reference strain *S. aureus* ATCC®29213™ was used as quality control. Isolates resistant to one antibiotic of at least three classes of antibiotics were considered multidrug resistant [61].

#### *4.3. Detection of Resistance Genes by PCR*

Total DNA was extracted from each isolate by the boiling method as described by Alexopoulou and colleagues [62]. All isolates were screened by PCR for the presence of the resistance genes *mecA* and *blaZ* and plasmid-encoded efflux pump genes *qacA*/*B* and *smr* (reduced susceptibility to biocides and EtBr). Isolates presenting phenotypic resistance to antibiotics were also screened for the presence of the genes *erm*(A), *erm*(B), *erm*(C), *msr*(A), *mph*(C), *vga*(A), *vga*(C) (resistance to macrolides, lincosamides, and streptogramins), *aadD*, *aph*(3′ )-*IIIa*, *aacA-aphD* (resistance to aminoglycosides), *fusB*, and *fusC* (resistance to fusidic acid) using the primers described in Table S1 of Supplementary Data.

#### *4.4. Screening of Mutations in grlA, gyrA, and fusA Genes*

Mutations in the QRDRs of *grlA* and *gyrA* genes associated with fluoroquinolone resistance were screened for representative isolates, chosen according to their PFGE types. Mutations in the *fusA* gene were screened for isolates presenting resistance to fusidic acid. The primers used for amplification and sequencing of *grlA*, *gyrA*, and *fusA* genes are described in Table S1. Amplification products were purified using the kit NZYGelpure (NZYTech, Lisboa, Portugal) and sequenced. Sequences were analyzed using the programs SnapGene Viewer (GSL Biotech; available at snapgene.com) and blastx (NCBI, Bethesda, MD, USA).

#### *4.5. Evaluation of Efflux Activity*

The presence of increased efflux activity was evaluated by (i) determining the EtBr MIC [63] and (ii) determination of EtBr and CIP MICs in the presence of the EIs TZ and VER [23,46]. MICs of EtBr, CIP, TZ, and VER (Sigma-Aldrich) were determined by the twofold broth microdilution method. Briefly, from overnight cultures, a cellular suspension equivalent to McFarland 0.5 was prepared in CAMHB and aliquoted in 96-well plates containing two-fold dilutions of the compound to be tested. Plates were incubated at 37 ◦C for 18 h, and the MIC registered as the lowest concentration of compound that inhibited visible growth. EtBr and CIP MICs were then redetermined in the presence of TZ and VER at 12.5 µg/mL and 400 µg/mL, respectively, corresponding to a subinhibitory concentration (1/2 MIC) [23]. The 96-well plates were prepared as described previously, except for the addition of a 0.01 mL aliquot of TZ or VER to each well prior to inoculation of the plate. Each assay was performed in duplicate. A four-fold, or higher, decrease in MICs values in the presence of EIs is indicative of inhibition of efflux activity [23].

#### *4.6. Plasmid DNA Extraction and Profiling*

Plasmid DNA of each isolate was extracted with the kit NZYMiniprep (NZYTech), adding 35 µg/mL of lysostaphin (Sigma-Aldrich) in the cell lysis step with buffer A1, followed by an incubation at 37 ◦C for 90 min. For isolates carrying a single plasmid, plasmid DNA was digested with 10 U of the enzyme *Eco*RI (NZYTech). The reaction mixture was incubated at 37 ◦C for 90 min and inactivated at 65 ◦C for 20 min. Restriction profiles were analyzed by 1% (*w*/*v*) agarose gel electrophoresis for 90 min.

#### *4.7. Detection of lukSF Genes*

The presence of the determinants *lukF-PV* and *lukS-PV* encoding PVL was screened by PCR, using the primers described in Table S1 of the Supplementary Data.

#### *4.8. Molecular Typing*

All isolates were characterized by PFGE. *Sma*I-PFGE was performed as previously described [64], and macrorestriction profiles were analyzed with the Bionumerics software v 7.6 using the Dice coefficient and dendrograms built based on the UPGMA algorithm, considering a band tolerance of 1% and an optimization of 0.5%. Isolates presenting macrorestriction profiles with a similarity ≥81% or ≥97% were considered as belonging to the same PFGE type or subtype, respectively [65]. The genetic diversity of the collection was calculated, based on PFGE types, by Simpson's index of diversity with a confidence interval of 95% [66].

A subset of isolates representative of each PFGE type was further analyzed by MLST. Isolates sharing the same PFGE type or subtype were considered as belonging to the same ST. Internal fragments of the seven housekeeping genes *arcC*, *aroE*, *glpF*, *gmk*, *pta*, *tpi*, and *yqiL* were amplified by PCR and sequenced using the primers and conditions previously described [67,68]. Allelic profiles and STs were obtained from MLST database (PubMLST.org (accessed on 28 December 2020). New alleles and ST profiles were submitted to PubMLST for validation and allele/ST assignment. The relationship between clonal lineages were inferred with the PHYLOViZ freeware using the goeBurst algorithm [69].

*agr* typing of all isolates was performed according to the protocol described by Lina and colleagues [70]. The set or primers used for *agr* typing is described in Table S1.

#### **5. Conclusions**

This work demonstrates a high prevalence of antibiotic resistance in *S. aureus* of SSTIs from outside the hospital environment, correlating it with the presence of several antibioticresistance determinants and a high prevalence of PVL-positive isolates, assigned to three MSSA (ST152-*agr*I and ST30-*agr*III) and four MRSA (ST8-*agr*I) isolates. This study also highlights the phenotypic and genotypic variability that may be present in *S. aureus* isolates causing infection in distinct anatomical sites of the same patient. The high diversity of plasmids identified in this collection demonstrates the important role these MGEs have in the transmission of antimicrobial resistance in *S. aureus* and the relevance of studying these elements to further prevent the dissemination of MDR strains.

**Supplementary Materials:** The following is available online at https://www.mdpi.com/2079-6382/ 10/4/345/s1, Table S1: Primers used in this study.

**Author Contributions:** Conceptualization, S.S.C. and I.C.; Funding acquisition, S.S.C. and I.C.; Investigation, C.F., S.S.C., M.S. and K.O.; Methodology, S.S.C. and I.C.; Project administration, S.S.C. and I.C.; Resources, G.T., C.P. and I.C.; Supervision, I.C.; Validation, S.S.C. and I.C.; Visualization, C.F. and S.S.C.; Writing—original draft, C.F. and S.S.C.; Writing—review and editing, C.F., S.S.C., M.S., K.O., G.T., C.P. and I.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by Project BIOSAFE funded by FEDER through the Programa Operacional Factores de Competitividade—COMPETE and by Fundação para a Ciência e a Tecnologia (FCT, Portugal)—Grant LISBOA-01-0145-FEDER-030713, PTDC/CAL-EST/30713/2017 and by FCT through funds to GHTM (UID/04413/2020) and CIISA Project (UID/CVT/00276/2020).

**Institutional Review Board Statement:** Ethics approval was not required for this study.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** All relevant data have been provided in the paper. Raw data can be provided by the authors upon reasonable request.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Nisin Influence on the Antimicrobial Resistance Ability of Canine Oral Enterococci**

**Eva Cunha 1 , Rita Janela 1 , Margarida Costa 1 , Luís Tavares 1 , Ana Salomé Veiga <sup>2</sup> and Manuela Oliveira 1, \***


Received: 20 November 2020; Accepted: 8 December 2020; Published: 10 December 2020

**Abstract:** Periodontal disease (PD) is one of the most common diseases in dogs. Although previous studies have shown the potential of the antimicrobial peptide nisin for PD control, there is no information regarding its influence in the development of antimicrobial resistance or horizontal gene transfer (HGT). Nisin's mutant prevention concentration (MPC) and selection window (MSW) were determined for a collection of canine oral enterococci. Isolates recovered after the determination of the MPC values were characterized for their antimicrobial profile and its nisin minimum inhibitory and bactericidal concentrations. The potential of *vanA* HGT between *Enterococcus faecium* CCGU36804 and nine clinical canine staphylococci and enterococci was evaluated. Nisin MPC values ranged from 400 to more than 600 µg/mL. In comparison with the original enterococci collection, the isolates recovered after the determination of the nisin MPC showed increased resistance towards amoxicillin/clavulanate (5%), vancomycin (5%), enrofloxacin (10%), gentamicin (10%) and imipenem (15%). The HGT of *vanA* gene was not observed. This work showed that nisin selective pressure may induce changes in the bacteria's antimicrobial resistance profile but does not influence horizontal transfer of *vanA* gene. To our knowledge, this is the first report of nisin's MPC and MSW determination regarding canine enterococci.

**Keywords:** nisin; mutant prevention concentration; mutant selection window; antimicrobial susceptibility testing; horizontal gene transfer

#### **1. Introduction**

Periodontal disease is one of the most widespread inflammatory diseases in dogs [1,2], that results from the establishment of a polymicrobial biofilm (dental plaque) on the teeth surface and a subsequent local inflammatory response [3]. Recently, we proposed the use of a new nisin biogel as a promising strategy to control this disease [4]. Nisin is an antimicrobial peptide, active mainly against Gram-positive bacteria, including multi-drug-resistant bacteria [4–7], with demonstrated potential for medical application [8]. However, considering that antimicrobial resistance is a major public health problem, any antimicrobial compound under investigation for clinical purposes should be characterized for the mechanisms responsible for resistance development and its environmental persistence [9,10]. Bacteria can become resistant to antimicrobial compounds by acquisition of resistance genes through horizontal gene transfer (HGT), but resistance can also result from the accumulation of mutations that decrease susceptibility [11,12]. Therapeutic protocols that favor mutant subpopulations may

facilitate resistance development when compared with regimens that suppress mutant formation [11]. Thus, it is important to optimize the antimicrobial concentrations needed to prevent the selection and amplification of resistant mutants [13]. In this context, the mutant selection window (MSW) hypothesis, described by Zhao and Drlica, postulates that single-step resistant mutant subpopulations, although naturally present, are selectively enriched and amplified when drug concentrations fall within a specific range [14–16]. The MSW comprises a range of concentrations between the minimal inhibitory concentration (MIC) and the mutant prevention concentration (MPC) [11]. The MIC is the lowest drug concentration that inhibits the multiplication of the majority of susceptible cells, while MPC is the drug concentration that blocks the growth of the least susceptible, first step mutant, when a high inoculum is applied [11,14–19].

As described, resistance dissemination can also occur by HGT, which plays an important role in the emergence of new pathogens [12,20]. The dental plaque biofilm is a perfect environment for the transfer of resistance and virulence genes between bacteria [21]. Present in the oral cavity of dogs with PD, commensal enterococci have a high genome plasticity, being capable of acquiring, conserving, and disseminating genetic determinants, such as resistance genes, easily becoming opportunistic pathogens and being associated with PD-systemic consequences [22,23]. In fact, vancomycin resistance associated with the *vanA* gene is one of the most important antimicrobial resistance determinants associated with vancomycin-resistant *Enterococcus faecium*, considered by WHO as a high priority pathogen [23,24]. This gene is usually present in the Tn1546 transposon, harbored in a plasmid, being transferred by HGT [23,25,26]. Several studies demonstrated that vancomycin resistant enterococci (VRE) can transfer *vanA* to other bacteria, such as staphylococci, which are commensals of the skin and mucosa of animals and humans [23,26]. This transfer ability was associated by some authors to the presence of a pSK41-like plasmids in the recipients [27,28]. Furthermore, a continuous antimicrobial pressure due to the presence of sub inhibitory concentrations of antibiotics in the environment may contribute to the mobilization of acquired resistance genes, being essential to understand how the application of new antimicrobial compounds, such as antimicrobial peptides, can interfere with this phenomenon [26,29,30].

In this work, we determined the MSW of nisin from a previously characterized collection of enterococci obtained from the oral cavity of dogs with PD [31]. The isolates recovered after the determination of the MPC were collected and used to re-evaluate nisin's inhibitory (MIC) and bactericidal (MBC) concentrations, as well as their antimicrobial susceptibility profile against 11 antimicrobials relevant for veterinary medicine and public health. The influence of subinhibitory concentrations of nisin in the horizontal transfer of *vanA* gene from *Enterococcus faecium* to canine staphylococci and enterococci was also evaluated.

#### **2. Results**

#### *2.1. Determination of the Nisin Mutant Prevention Concentration (MPC)*

Nisin MPC values were determined regarding a previously characterized canine enterococci collection, obtained from the oral cavity of dogs with PD [31]. These values, in combination with the previously determined MIC values [1], allowed us to define the MSW of nisin towards the isolates under study. In addition, the relationship between these values, expressed as the ratio MPC/MIC, was determined, which can be used to compare antimicrobial agents for their ability to select resistant mutants [14].

It was possible to determine the MPC values for 85% (*n* = 17) of the strains used, with the exception of strains B28d, B29c and B32a, which presented an MPC higher than 600 µg/mL. The nisin MPC values for the 17 strains ranged from 400 to 600 µg/mL, with an average MPC of 447.06 ± 84.84 µg/mL (Table 1). Considering the nisin MPC/MIC, the resulting MPC values were 15 to 39 times higher than the previously determined MIC values [4].


**Table 1.** Minimum inhibitory concentration (MIC), mutant prevention concentration (MPC), and MIC/MPC ratio of nisin against the enterococci collection obtained from the oral cavity of dogs with periodontal disease (PD).

ID—identification, MIC—minimum inhibitory concentration, MPC—mutant prevention concentration, SD—standard deviation.

#### *2.2. Antimicrobial Susceptibility Testing*

The antimicrobial susceptibility profiles of the original enterococci collection were compared with the ones obtained with the bacterial isolates recovered from the MPC protocol plates with the highest nisin concentrations. It was possible to observe that none of the isolates were susceptible to all the antibiotics tested, being in fact resistant to more than one compound. In the original enterococci collection, resistance levels ranged from 0% (imipenem and amoxicillin/clavulanate) to 100% (cefotaxime and gentamicin-10 µg), while for the isolates recovered from the MPC protocol, the resistance levels varied between 5% (amoxicillin/clavulanate) and 100% (cefotaxime and gentamicin−10 µg). When compared with the original isolates, the MPC recovered isolates presented an increased resistance towards amoxicillin/clavulanate (5%), vancomycin (5%), imipenem (15%), enrofloxacin (10%) and gentamicin-120 µg (10%) (Table 2).

According to the definitions proposed by Magiorakos and collaborators (2012) [32], which indicate that a multidrug-resistant *Enterococcus* spp. is non-susceptible to at least one agent in three or more antimicrobial categories, in our study, 15 isolates (75%) in the original collection and 18 isolates (90%) in the MPC recovered collection exhibited a multidrug-resistance profile.



#### *2.3. Determination of Nisin's Minimum Inhibitory Concentration (MIC) and Minimum Bactericidal Concentration (MBC)*

The isolates recovered after determination of the nisin MPC values were used to evaluate the effect of nisin selective pressure on the nisin 's MIC and MBC values, by comparison with the values from the original collection [1].

MIC values of nisin regarding the isolates of the original collection and the isolates recovered in the MPC protocol are presented in Table 3.

Concerning the isolates recovered after the determination of the MPC, nisin MIC values ranged from 18.75 to 81.25 µg/mL, with an average value of 48.41 ± 21.62 µg/mL. MIC values were higher than 100 µg/mL for three isolates. MBC values ranged from 37.50 to 92.19 µg/mL, with an average value of 60.46 ± 19.40 µg/mL. An MBC value higher than 100 µg/mL was observed for eight isolates.

Nisin MIC results obtained against the MPC recovered isolates were higher and statistically different (*p*-value < 0.05) when compared with the nisin MIC values of the original collection. Concerning the MBC values, no statistical difference was observed between the results regarding the isolates from the two collections; however, most isolates recovered in the MPC protocol showed higher MBC values when compared with the original ones (65%, *n* = 13/20).

**Table 3.** MIC, MBC, and MBC/MIC ratio of nisin regarding the enterococci collection recovered after MPC protocol and the original enterococci collection obtained from the oral cavity of dogs with PD.



**Table 3.** *Cont.*

ID—identification, MIC—minimum inhibitory concentration, MBC—minimum bactericidal concentration, MPC—mutant prevention concentration, SD—standard deviation.

#### *2.4. Nisin's Influence on vanA Horizontal Gene Transfer (HGT)*

Bacterial acquired resistance by HGT is an important form of resistance dissemination [12]. The *vanA* gene is responsible for vancomycin resistance and its transfer between enterococci and staphylococci is well documented [25]. To evaluate nisin's influence in *vanA* transfer between enterococci and staphylococci, first we assessed the presence or absence of *vanA* in the isolates under study. In the case of staphylococci, we also evaluated the presence of *mecA* gene, associated with methicillin resistance, and of the pSK41-like plasmid, that may prompt *vanA* transfer [23,28].

In the initial PCR screening, none of the isolates from the oral enterococci collection (*n* = 20) presented the *vanA* gene. In addition, none of the six staphylococci obtained from canine skin lesions presented the *vanA* and *mecA* genes or the pSK41-like plasmid. Then, three isolates from the enterococci collection, M3b, M23a and M29b, were selected to participate in the HGT protocol, based on their strong capacity of biofilm production [31].

Afterwards, two mating rounds aiming to promote HGT of the *vanA* gene from *E. faecium* CCUG 36804 to nine clinical enterococci and staphylococci were performed. One round was performed in the absence of antimicrobial environmental pressure, and another in the presence of subinhibitory concentrations of nisin. None of the two mating experiments allowed the development of transconjugants in the MSA plates supplemented with rifampicin and vancomycin. All isolates recovered from the SBA and the MSA plates supplemented with rifampicin were submitted to PCR analysis, the results of which confirmed the absence of the *vanA* gene.

#### **3. Discussion**

Antimicrobial resistance is considered one of the major health treats of our time [10,24]. Misuse, overuse, and improper antimicrobial dosage promote a selective environmental pressure to bacteria, favoring resistance development [9,10]. Nisin, commonly used as a preservative in the food industry, is showing relevance in the biomedical field, being a promising antimicrobial agent to be used for the control of canine PD [4,33]. Despite the low resistance rate associated with this antimicrobial peptide, a few cases of nisin resistance have been reported, reinforcing the need to unveil related mechanisms, to evaluate its influence on dental plaque bacterial interaction, and to adopt correct doses to prevent the emergence and amplification of nisin resistant strains [8,34,35].

The MSW hypothesis allows us to define a range of concentrations that can promote mutant's development, which is useful to evaluate dose regimens [14]. In the present work, the MPC values of nisin against a collection of enterococci from the oral cavity of dogs with PD were determined. The values obtained were up to 39 times higher than the previously determined MIC values for the same isolates [4]. Similar results were also obtained for daptomycin against *Enterococcus faecalis*, with MPC values being 2 to 32 times the MIC values [19]. In fact, several authors have used this methodology to evaluate multiple antimicrobials and bacteria [36–43]. All these studies revealed high MPC values in comparison with the MIC values for the same microorganism; however, a high variation was observed between bacteria and drugs. For example, a MPC/MIC ratio of 48 to 72 was obtained for fosfomycin against *Escherichia coli,* while for *Pseudomonas aeruginosa* the ratio was 28 to 57; likewise, orbifloxacin presented a MPC/MIC ratio against *E. coli* of 4 to 32, while for *P. aeruginosa* the ratio was 16 to 64 [41,43]. In fact, Gianvecchio and collaborators (2019) suggested that MPC values present high variability for a given bacterial strain–antimicrobial combination, and should be understood as a range with confidence intervals, contrasting with MIC values [43,44].

To better understand the effect of nisin selective pressure over 72 h, as promoted in the determination of the MPC, antimicrobial susceptibility profiling along with nisin's MIC and MBC determination were performed on the isolates obtained in the MPC protocol and results were compared with the original collection.

Considering isolates' antimicrobial susceptibility profile, differences in the resistance profile of the isolates recovered after MPC protocol were observed when compared with those of the original isolates, specifically concerning amoxicillin/clavulanate, vancomycin, imipenem, enrofloxacin and gentamycin (120 µg). These results suggest a possible influence of nisin in increasing antimicrobial resistance. Cross resistance between nisin and antimicrobials is rare; however, there are some reports describing its occurrence [34,35,45]. Cross resistance may occur regarding antimicrobials that present a similar mode of action, or when the resistance mechanisms are related [35]. Nisin acts by binding to the lipid II, present in the bacterial membrane, which leads to pore formation and inhibition of peptidoglycan synthesis [8]. Considering that, parallel mechanisms may be observed in resistance to vancomycin, an antimicrobial that also acts on lipid II but in a different location, or in resistance to antimicrobials that act on the bacterial wall, such carbapenems or aminopenicillins [5,46]. Resistance to nisin is usually related to proteolytic degradation (by nisinase and nisin resistant protein); however, there are descriptions suggesting that resistance can also arise from mutations that induce changes in the membrane and cell wall composition, such as cell wall thickening, increased positive charges, the presence of penicilin binding proteins and modifications of membrane phospholipid and fatty acid composition [8,34,35,47]. Other nisin resistance mechanisms described so far are related to ABC transporters and multiple regulatory networks [34].

In addition, Drlica (2003) showed that the mutants derived from the MPC protocol are expected to develop mechanisms that inactivate the antimicrobial agent, including efflux or degradation systems [11]. These mechanisms may explain the increased resistance towards enrofloxacin and gentamicin, that act by inhibition of nucleic acid and protein synthesis, respectively [12].

Considering the MIC and MBC determinations, MIC values were higher and statistically different (*p*-value ≤ 0.05) towards all the recovered isolates in comparison with those of the original collection, while MBC values were higher regarding 65% of the recovered isolates in comparison with the originals. These results suggest that incubation in the presence of nisin leads to a reduction in the inhibitory activity, in spite of its bactericidal activity being maintained towards most isolates (60%, Table 3). According to Levinson and collaborators (2009), an MBC/MIC ratio lower than four indicates that the

antimicrobial agent is bactericidal [48]. As such, nisin presented a bactericidal activity towards 42% of the isolates from the original collection [4]. On the other hand, nisin presented a bactericidal activity against all isolates recovered from the MPC protocol, except for isolates with nisin MBC values higher than 100 µg/mL.

Animals' oral cavities present a high bacterial concentration and diversity [21]. Located at the teeth surface, dental plaque is a highly complex polymicrobial biofilm where bacteria easily interact and act as reservoirs of transferable resistance genes [21]. *Enterococcus* spp. are known to be a central hub for resistance gene acquisition, conservation, and dissemination [23]. Classified by the WHO as a high priority pathogen, vancomycin-resistant *Enterococcus faecium*, along with other enterococcal species, are opportunistic pathogens frequently associated with nosocomial infections, and capable of transferring relevant genes to other bacterial species such as *Staphylococcus aureus*, *E. coli* and *Listeria* spp. [23,24,26]. In this work, a protocol aiming to promote the horizontal transfer of *vanA* from *E. faecium* to *Staphylococcus* spp. and *Enterococcus faecalis* clinical isolates was established. This gene is linked to vancomycin and teicoplanin resistance in enterococci, being harbored in a mobile genetic element, allowing its transfer to other bacteria [26]. Mating experiments performed in the absence of nisin selective pressure did not allow the transfer of *vanA* gene. Several studies demonstrated that *vanA* transfer may be facilitated by some molecules, such as pheromone-inducible surface proteins, or be related to the presence of specific plasmids, such as *S. aureus* pSK41 [28,49–51]. Although two of our enterococci recipients (M3b and M23a) were able to express an aggregation substance—more specifically, a pheromone-inducible surface protein that facilitates conjugative exchange [4,31,49,52]—no transfer occurred. None of the staphylococcal recipients presented the pSK41 plasmid, which may have influenced the results. Nevertheless, *vanA* gene horizontal transfer is a complex process which is not yet fully understood.

It is known that the use of antimicrobials can enhance gene transfer between bacteria [26]. In order to evaluate the influence of nisin in HGT, mating experiments were performed in the presence of this antimicrobial peptide at subinhibitory concentration [4]. None of the recipients presented the *vanA* gene after the mating experiments, reinforcing the potential of nisin to be used in the clinical setting, more precisely in veterinary medicine for canine PD control.

To our knowledge, this is the first report of nisin's MPC and MSW determination regarding canine enterococci and of its influence on gene transfer between enterococci and staphylococci. This approach is an important step in the development of new antimicrobial compounds, allowing to understand their potential influence in resistance evolution.

#### **4. Materials and Methods**

#### *4.1. Bacterial Collection*

A collection of 20 oral enterococci obtained from the oral cavity of dogs diagnosed with PD, previously characterized regarding clonality, antimicrobial resistance and virulence profiles, were used as bacterial models [31]. From these 20 isolates, 17 correspond to strains belonging to the species *Enterococcus faecalis*, and the remaining 3 to *Enterococcus faecium* [4].

For the HGT protocol, one *Staphylococcus aureus* and five *Staphylococcus pseudintermedius* obtained from canine skin lesions and an *Enterococcus faecium* reference strain (CCUG 36804, *vanA* positive) were used.

*Enterococcus faecalis* ATCC® 29212, *Staphylococcus aureus* ATCC® 25293, a *Staphylococcus aureus mecA* positive strain kindly provided by Dr. Birgit Strommenger, Robert Koch Institute, Germany, and a *Staphylococcus aureus* RN4220 pGO1 positive strain kindly provided by Dr. Alex O'Neill, University of Leeds, were included as controls.

#### *4.2. Nisin Preparation*

A nisin stock solution (1000 µg/mL) was obtained by dissolving 1 g of nisin powder (2.5% purity, 1000 IU/mg, Sigma-Aldrich, St. Louis, MO, USA) in 25 mL of HCl (0.02 M) (Merck, Darmstadt, Germany) [4]. Then, the stock solution was filtered using a 0.22 µm Millipore filter, and serial dilutions were prepared in distilled sterile water. Solutions were kept at 4 ◦C during the study.

#### *4.3. Determination of the Nisin Mutant Prevention Concentration (MPC)*

To determine the MPC of nisin against the canine oral enterococci collection [31], a modified version of the protocol described by Sinel and collaborators (2016) was performed [19]. Briefly, each isolate was spread onto three brain heart infusion (BHI) agar plates (VWR, Leuven, Belgium) and incubated for 24 h at 37 ◦C. Afterwards, all the bacterial lawn developed in the three BHI plates was resuspended in 450 µL of BHI broth and further incubated at 37 ◦C for 20 min, to achieve a bacterial suspension of 10<sup>10</sup> CFU/mL, which was confirmed by viable cell count. Then, an aliquot of 50 µL of this bacterial suspension was inoculated onto Mueller–Hinton (MH) agar plates (Oxoid, Hampshire, UK), supplemented with two-fold concentration increments of nisin ranging from 6.25 to 40× the MIC value of 14.9 µg/mL [4]. Thus, the MH agar plates series contained 6.25, 15, 25, 50, 100, 200, 400, 600 µg/mL of nisin, previously determined. Plates were incubated at 37 ◦C for 72 h and observed daily for detection of colony growth.

MPC was defined as the lowest concentration of nisin that prevented the growth of any resistant mutant subpopulations after a 72 h incubation period [19]. It was also possible to establish the mutant selection window (MSW) of nisin for the collection of oral enterococci isolates from dogs, a value defined as the antimicrobial concentration ranging between the MIC and MPC values [11,15]. In addition, colonies grown in MH plates with the higher nisin concentration were isolated and kept at −80 ◦C in a solution of buffered peptone water with 20% glycerol. These isolates were classified as MPC recovered isolates and further used for antimicrobial profiling and for the determination of nisin's MIC and MBC.

#### *4.4. Antimicrobial Susceptibility Testing*

Antimicrobial susceptibility profiling was performed regarding the original clinical isolates and those recovered after the MPC protocol to determine if incubation in the presence of nisin interferes with the susceptibility profiles. Using the disk diffusion method, the susceptibility profile regarding a total of 11 different antibiotics, (MASTDISCS® AST, Mast Group, Liverpool, UK), presented in Table 4, was determined in accordance with the Clinical and Laboratory Standards Institute (CLSI) guidelines [53,54]. For that, a 0.5 MacFarland bacterial suspension was prepared for each isolate. Afterwards, the inoculum was evenly spread over the surface of a MH agar plate and the disks impregnated with the antimicrobial agents were placed over the surface of the agar plate. Plates were then incubated at 37 ◦C under aerobic conditions for 18 h or, in the case of vancomycin, 24 h. After incubation, the inhibition zone diameters were measured and compared with the CLSI standard breakpoints established in VETS01-S2 and M100S, allowing to define the antimicrobial profile (resistant, intermediate or susceptible) of each isolate regarding the antimicrobial agents tested [53,54]. Quality control was performed using the reference strain *Staphylococcus aureus* ATCC® 25293.

Antibiotics were selected based on their relevance to veterinary medicine, as well as to public health. Specifically, CN-120 µg and S-300 µg were included to detect high-level aminoglycoside resistance in *Enterococcus* spp., whereas IMI and VA were chosen due to their importance to public health [53–55].


**Table 4.** Antimicrobial agents used in the antimicrobial susceptibility test, grouped by mechanism of action, class and concentration [46].

\* Beta lactamase inhibitor.

#### *4.5. Determination of Nisin's Minimum Inhibitory Concentration (MIC) and Minimum Bactericidal Concentration (MBC)*

Nisin MIC determination was performed regarding the isolates recovered after the MPC protocol using the broth microdilution method, to assess their current susceptibility to nisin, as previously described by Cunha and collaborators (2018) [4]. Briefly, a 96-well microplate was filled with 20 µL of nisin solution at different concentrations (final nisin concentrations ranged from 1.25 to 100 µg/mL) and 180 µL of 10<sup>6</sup> CFU/mL bacterial suspensions of each isolate. A negative control with only tryptic soy broth (TSB) medium (VWR, Leuven, Belgium) and positive controls with bacterial suspensions were also included.

The 96-well microplates were incubated at 37 ◦C for 24 h, after which bacterial growth was visually assessed in order to determine MIC value. This parameter is defined as the lowest nisin concentration capable of preventing bacterial multiplication in vitro, with no visible growth on the well [4].

Subsequently, after MIC assessment, the MBC was determined. Five microliters of the bacterial suspension from each well with no visible growth were plated onto tryptic soy agar (TSA) plates (VWR, Leuven, Belgium), followed by incubation at 37 ◦C for 24 h. MBC was defined as the lowest antimicrobial concentration that inhibits bacterial growth after sub-culture of the suspensions on solid unselective media without any antimicrobial agent [6].

These assays were performed in triplicate, on independent days, and 10% of replicates were tested to assure results representability.

#### *4.6. Nisin's Influence on vanA Horizontal Gene Transfer (HGT)*

#### 4.6.1. DNA Extraction and Isolates PCR Screening

DNA extraction was performed based on the protocol described by Semedo-Lemsaddek et al. (2016) and Mottola et al. (2016) [22,56]. Then, all canine staphylococci and enterococci were evaluated by multiplex PCR, in order to identify the presence of the gene *vanA* and *mecA* [57]. Two pairs of primers synthesized by STABVIDA® (Lisbon, Portugal), targeting *vanA* (5′ GGGAAAACGACAATTGC 3 ′ and 3′ GTACAATGCGGCCGTTA 5′ ) and *mecA* (5′ TCCAGATTACAACTTCACCAGG 3′ and 3 ′CCACTTCATATCTTGTAACG 5′ ) were used [56,57]. The PCR mixture had a final volume of 28.5 µL, with 10 µL of Supreme NZYTaq 2× Green Master Mix (NZYtech®, Lisbon, Portugal), 0.5 uM of the *vanA* primer, 0.4 uM of the *mecA* primer, 16.88 µL of PCR-grade water and 5 µL of DNA template. PCR amplification was completed using the conditions: initial denaturation at 94 ◦C for 4 min; 10 cycles involving denaturation at 94 ◦C for 30 s, annealing at 64 ◦C for 30 s and elongation at 72 ◦C for 45 s; 25 cycles involving denaturation at 94 ◦C for 30 s, annealing at 50 ◦C for 45 s and elongation at 72 ◦C for 2 min, and a final extension step at 72 ◦C for 10 min. Electrophoresis (90 V for 45 min) was performed to evaluate the amplified products, using a 1.5% agarose gel (NZYtech®, Lisbon, Portugal) stained with GreenSafe (NZYtech®, Lisbon, Portugal). A molecular weight marker, NZYDNA ladder VI (NZYtech®, Lisbon, Portugal) was also included. Results were visualized by transillumination.

Two positive control strains, *Staphylococcus aureus* 01-00694 (*mecA* positive) and *Enterococcus faecium* CCUG 36,804 (*vanA* positive), were included [56].

In addition, the presence of the pSK41-like plasmid in the 6 canine staphylococci under study was evaluated by PCR, using a primer targeting the *traE* gene (5′ ACAAATGCGTA CTACAGACCCTAAACGA 3′ and 3′GCCCTGCTGTTGCTGTATCCATATT 5′ ), synthesized by STABVIDA® [28,58].

A PCR mixture composed by 10 µL of Supreme NZYTaq 2× Green Master Mix (NZYtech®, Lisbon, Portugal), 0.4 uM of *traE* primer; 39.2 µL of PCR-grade water and 5 µL of DNA template was used. The PCR amplification was completed using the following conditions: initial denaturation at 94 ◦C for 2 min; 30 cycles involving denaturation at 95 ◦C for 15 s, annealing at 55 ◦C for 90 s and elongation at 72 ◦C for 90 s, and a final extension step at 72 ◦C for 7 min [28].

An electrophoresis (90 V for 45 min) was performed to evaluate the amplified products, using a 1.5% agarose gel (NZYtech®, Lisbon, Portugal) stained with GreenSafe (NZYtech®, Lisbon, Portugal). A molecular weight marker, NZYDNA ladder VII (NZYtech®, Lisbon, Portugal), was also included. Results were visualized by transillumination.

A positive control strain, *Staphylococcus aureus* RN4220 (pGO1 positive), was included in the PCR amplification protocol [59].

#### 4.6.2. HGT Protocol

To test if selective pressure due to the presence of subinhibitory concentrations of nisin induces HGT, a protocol adapted from Niederhäusern and collaborators (2011) was developed [25]. Mating experiments were performed in two rounds, using the VRE rifampicin susceptible (Van<sup>r</sup> Rif<sup>s</sup> ) *Enterococcus faecium* CCUG 36804 strain, as donor of the *vanA* gene, and as recipients the 6 canine staphylococci from canine skin lesions, and 3 canine *Enterococcus faecalis* isolates from our collection of enterococci from the oral cavity of dogs with PD, selected according to their strong biofilm forming ability. All recipients were susceptible to vancomycin, and rifampicin resistance was induced (Van<sup>s</sup> Rif<sup>r</sup> ) [25], as it is associated with a point mutation rather than to an acquired gene [60]. After performing a 0.5 MacFarland suspension in 0.9% NaCl for each isolate, 500 µL of the donor suspension and 500 µL of the suspension of one of the recipients were inoculated into 5 mL of TSB and incubated at 37 ◦C for 18 h. After incubation, 1 mL of the dual bacterial suspension was added to 5 mL of TSB and further incubated for 6 h at 37 ◦C. Afterwards, 2 mL of each dual suspension were inoculated in TSA plates and incubated for 5 h at 37 ◦C in a slight movement on a shaker, to promote mating. Then, plates were incubated at 37 ◦C for 24 h, after which the bacterial suspension that remained at the surface of the agar plates was removed and inoculated in 5 mL of TSB. After an incubation period of 12 h at 37 ◦C, 100 µL of the suspension was inoculated in Mannitol Salt agar (MSA, PanReac AppliChem, Barcelona, Spain) supplemented with rifampicin (64 µg/mL, PanReac AppliChem, Barcelona, Spain) and vancomycin (8 µg/mL, PanReac AppliChem, Barcelona, Spain), to select for transconjugants. If mating occurred, recombinant isolates developed should be resistant to rifampicin and vancomycin. In addition, the suspension was also inoculated in Manitol Salt Agar (MSA, PanReac AppliChem, Barcelona, Spain) and Slanetz and Bartley agar (SBA, PanReac AppliChem, Barcelona, Spain) supplemented only with rifampicin (64 µg/mL).

The second mating round was performed in the presence of nisin, with all the media used being supplemented with nisin at sub-MIC concentration, 7.45 µg/mL for enterococci [4] and 5.63 µg/mL for staphylococci [6]. All recovered isolates and transconjugants recovered from the supplemented media were submitted to a PCR analysis to detect the presence of the *vanA* gene.

#### *4.7. Statistical Analysis*

Data statistical analysis was carried out using Microsoft Excel 2016®. All quantitative data are expressed as means ± standard deviation. Student's *t*-test was used for statistical analysis of the nisin MIC and MBC values regarding the original collection and the collection recovered after MPC protocol. A confidence interval of 95% was considered, with a *p*-value ≤ 0.05 indicating statistical significance.

#### **5. Conclusions**

Periodontal disease is a highly prevalent inflammatory disease in dogs, and nisin might be a promising molecule for its control. The study of nisin influence on mutant's development, antimicrobial signatures and transfer of resistance determinants revealed that this compound can influence isolates antimicrobial profiles. MPC and MSW determinations can be an interesting measure to establish more accurate treatment protocols based on appropriate antimicrobial doses. However, the utility of the MSW in the definition of dose regimens must be demonstrated not only in vitro but also in vivo. In addition, this study showed that nisin did not promote horizontal transfer of the *vanA* gene between the isolates tested, which emphasizes its potential to be used in PD control. To our knowledge, this is the first report of nisin's MPC and MSW determination regarding canine enterococci, being a relevant step towards its application in both human and veterinary medicine.

**Author Contributions:** Conceptualization, M.O., A.S.V. and E.C.; methodology, E.C., M.O., R.J. and M.C.; software, E.C. and R.J.; validation, M.O., L.T. and A.S.V.; formal analysis, E.C.; investigation, E.C. and M.C.; data curation, M.O.; writing—original draft preparation, E.C.; writing—review and editing, M.O., L.T. and A.S.V.; visualization, E.C.; supervision, M.O. and A.S.V.; project administration, M.O.; funding acquisition, M.O. and E.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** Authors would like to acknowledge the Foundation for Science and Technology (Eva Cunha PhD fellowship SFRH/BD/131384/2017) and to CIISA–Centro de Investigação Interdisciplinar em Sanidade Animal, Faculdade de Medicina Veterinária, Universidade de Lisboa, Project UIDB/00276/2020 (Funded by FCT), for financial support of this work.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article Salmonella* **spp. in Pet Reptiles in Portugal: Prevalence and Chlorhexidine Gluconate Antimicrobial Efficacy**

**João B. Cota \* , Ana C. Carvalho, Inês Dias, Ana Reisinho, Fernando Bernardo and Manuela Oliveira**

CIISA–Centro de Investigação Interdisciplinar em Sanidade Animal, Faculdade de Medicina Veterinária, Universidade de Lisboa, Av. da Universidade Técnica, 1300-477 Lisboa, Portugal; accarvalho@fmv.ulisboa.pt (A.C.C.); ines.adpdias@gmail.com (I.D.); anareisinho@fmv.ulisboa.pt (A.R.); fbernardo@fmv.ulisboa.pt (F.B.); moliveira@fmv.ulisboa.pt (M.O.)

**\*** Correspondence: joaobcota@fmv.ulisboa.pt; Tel.: +351-213-652-800

**Abstract:** A fraction of human *Salmonella* infections is associated with direct contact with reptiles, yet the number of reptile-associated Salmonellosis cases are believed to be underestimated. Existing data on *Salmonella* spp. transmission by reptiles in Portugal is extremely scarce. The aim of the present work was to evaluate the prevalence of *Salmonella* spp. in pet reptiles (snakes, turtles, and lizards), as well as evaluate the isolates' antimicrobial resistance and virulence profiles, including their ability to form biofilm in the air-liquid interface. Additionally, the antimicrobial effect of chlorhexidine gluconate on the isolates was tested. *Salmonella* was isolated in 41% of the animals sampled and isolates revealed low levels of antimicrobial resistance. Hemolytic and lypolytic phenotypes were detected in all isolates. The majority (90.63%) of the *Salmonella* isolates were positive for the formation of pellicle in the air-liquid interface. Results indicate chlorhexidine gluconate is an effective antimicrobial agent, against the isolates in both their planktonic and biofilm forms, demonstrating a bactericidal effect in 84.37% of the *Salmonella* isolates. This study highlights the possible role of pet reptiles in the transmission of non-typhoidal *Salmonella* to humans, a serious and increasingly relevant route of exposure in the *Salmonella* public health framework.

**Keywords:** *Salmonella*; reptiles; isolation; antimicrobial resistance; biofilms; chlorhexidine gluconate; public health

#### **1. Introduction**

*Salmonella* is a well-known food-borne illness etiological agent, reported as the second most common zoonotic agent, causing 91,857 confirmed cases of disease in the European Union during 2018 [1] and an estimated number of 93.8 million cases worldwide annually [2]. The clinical manifestations of human salmonellosis are frequently those associated with a self-limited gastroenteritis, namely nausea, vomiting and diarrhea, but can also include severe complications, including bacteremia and extra-intestinal infections [3]. Though most commonly associated with contaminated food, human salmonellosis can also occur through the contact with infected animals, such as farm animals and pets, including reptiles [4].

In the course of the past years, reptiles have been increasingly regarded as household pets, with their estimated numbers ascending up to 8 million only in the Europe Union in 2019 [5]. *Salmonella* not only can be found in the gastrointestinal tract of healthy reptiles, but also in the environments where those animals are kept [6,7]. *Salmonella enterica* subspecies *enterica* is commonly found in warm-blooded animals, while the remainder subspecies, *salamae*, *arizonae*, *diarizonae*, *houtenae*, and *indica*, along with *Salmonella bongori* are frequently isolated either from reptiles or from the environment [8]. Furthermore, among more than 2500 known *Salmonella* serotypes, over 40% are associated with reptiles and are rarely isolated from other animals, including humans [9]. Although infrequent when compared with food-borne cases, accounting for 6% of all human salmonellosis cases both in the

**Citation:** Cota, J.B.; Carvalho, A.C.; Dias, I.; Reisinho, A.; Bernardo, F.; Oliveira, M. *Salmonella* spp. in Pet Reptiles in Portugal: Prevalence and Chlorhexidine Gluconate Antimicrobial Efficacy. *Antibiotics* **2021**, *10*, 324. https://doi.org/ 10.3390/antibiotics10030324

Academic Editor: Jonathan Frye

Received: 27 February 2021 Accepted: 17 March 2021 Published: 19 March 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

USA and in Europe [10], reptile-associated salmonellosis (RAS) seems to be more related with more severe clinical scenarios, such as systemic and severe disease development, especially in children, elderly people, and pregnant women [6]. In fact, RAS is a growing public health concern worldwide, with different reports pointing out for its role in disease outbreaks [11,12]. Despite the several RAS cases that have been reported in different European countries [13], there seems to be no available data regarding Portugal.

As observed for non-typhoidal salmonellae of other sources, there has been an increasing focus on antimicrobial resistance in reptile-associated *Salmonella* [14–16] since this feature can impair the success of treatments of both human and veterinary *Salmonella* infections [17]. Antimicrobial resistance can either arise from mutations in chromosomal genes (intrinsic resistance), which are caused by selective pressure, or through the acquisition of antimicrobial resistance determinants encoded in plasmids (extrinsic resistance), by horizontal transfer [18]. The role of reptiles as disseminators of antimicrobial resistant (AMR) *Salmonella* has been suggested [19,20]. Furthermore, *Salmonella* is known to have the ability of producing biofilms in different biotic and abiotic surfaces [21]. Not only are bacterial cells in biofilms more tolerant to antimicrobials when compared with the corresponding planktonic cells [22] but also more resistant to several chemical disinfectants [21].

Chlorhexidine is a biocide widely included in antiseptic products, especially in handwashing and oral products, due to its broad-spectrum efficacy and low irritability [23]. For surgical skin preparations and hand scrub, chlorhexidine is available in 4% solutions, while for wound cleaning is used as a 0.5% concentrated solution [24]. In veterinary care, chlorhexidine gluconate is a common disinfectant. In reptile treatment, chlorhexidine solutions are frequently used for topical application and preoperative scrubs, in concentrations below 2% [25], but there is a lack of clear guidelines regarding the most appropriate concentration to use.

The aim of the present study was to assess the presence of *Salmonella* spp. among the intestinal microbiota of pet reptiles in the Metropolitan area of Lisbon, Portugal, and to characterize those isolates, regarding antimicrobial susceptibility and virulence traits, bringing more information on the role of reptile-associated *Salmonella* on the public health scenario. Additionally, the antimicrobial efficacy of chlorhexidine gluconate against both planktonic cells and biofilms was also evaluated.

#### **2. Results**

#### *2.1. Salmonella spp. Isolates*

Of the 78 reptiles sampled 32 were identified as *Salmonella* positive (41%), specifically four Ophidians (50%), 14 Saurians (51.9%), and nine Chelonians (20.9%), belonging to 12 different owners (Table 1). Overall, the *Salmonella* recovery rates where higher both in Ophidians and Saurians when comparing with the one recorded in Chelonians (*p* = 0.016). After assessing the biochemical profile using API20E strip tests of the presumptive *Salmonella* isolates, 13 were identified as *Salmonella enterica* subspecies *arizonae* and 19 as *Salmonella* spp. (Table 2).

More than half of all *Salmonella* positive animals (62.5%) were detained by only three owners (E, F, and J). Moreover, owner J alone kept 12 *Salmonella* positive reptiles, more specifically Saurians. *Salmonella* isolates from co-habiting animals belonged to similar species with the exception for the isolates recovered from the animals of owner J, where the majority was identified as *Salmonella enterica* subspecies *arizonae* (10/12) and the remaining as *Salmonella* spp. (2/12) (Table 2). Notably, whenever an owner possessed multiple *Salmonella* positive animals, those animals belonged to the same reptile group.


**Table 1.** *Salmonella* positive animals, divided by category and species.

**Table 2.** Detailed information regarding the *Salmonella* isolates under study.


#### *2.2. Antimicrobial Resistance*

All of the studied isolates were susceptible to gentamicin (CN) and ciprofloxacin (CIP) (Table 3). High levels of susceptibility to amikacin (AK) (96.87%), sulfamethoxazole/trimethoprim (SXT) (96.87%), nalidixic acid (NA) (93.75%), enrofloxacin (ENR) (90.63%), amoxicillin/clavulanic acid (AMC) (90.63%), ampicillin (AMP) (90.63%), cefotaxime (CTX) (87.50%), tetracycline (TE) (87.50%), and to chloramphenicol (C) (81.25%) were also recorded. On the other hand, 31 of the *Salmonella* isolates (96.87%) were resistant to penicillin (P).

**Table 3.** Antimicrobial resistance and virulence phenotype results.


Abbreviations: AMC, amoxicillin/clavulanic acid; AMP, ampicillin; AK, amikacin; C, chloramphenicol; CN, gentamicin; CTX, cefotaxime; ENR, enrofloxacin; NA, nalidixic acid; P, penicillin; CIP, ciprofloxacin; SXT, sulfamethoxazole/trimethoprim; TE, tetracycline; N.S., non-significant.

When comparing groups, resistance to AMC (*p* = 0.0286) and AMP (*p* = 0.0286) were associated with Chelonian *Salmonella* spp. isolates, as resistance to both antimicrobials was only detected, and simultaneously, in isolates 26, 36, and 47, all originating from turtles of different owners (Supplementary Table S1). No other statistically significant differences regarding antimicrobial susceptibility were detected.

Only three isolates (9.37%), all from Chelonians, were resistant to three or more of the antimicrobial compounds tested (isolates 26, 36, and 47) (Supplementary Table S1). The multiple resistance patterns were AMC/AMP/P, observed in isolates 26 and 36, and AMC/AMP/P/TE, revealed by isolate 47. None of the isolates was considered to be multidrug resistant, since the detected resistance patterns included antibiotics from the same class.

#### *2.3. Virulence Phenotype*

Virulence phenotypic testing revealed that all of the isolates studied expressed both hemolytic and lipolytic behaviors (Table 3). Contrarily, gelatinase activity was not detected in any of the *Salmonella* isolates studied. Overall, DNase activity was observed in more than half (59.37%) of the isolates. No statistically significant differences in phenotypical behavior were identified when comparing isolates from different animal groups.

#### *2.4. Minimum Inhibitory Concentration and Minimum Bactericidal Concentration*

The minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) values of chlorhexidine gluconate calculated for each isolate can be found on Supplementary Table S1.

The overall average MIC value was 11.90 mg/L ± 3.68, ranging from 8.16 mg/L (MIC value observed towards a Chelonian isolate), to 23.81 mg/L (MIC value towards

a Chelonian and a Saurian isolates, all from different owners), with a median value of 10.72 mg/L. The majority of the chlorhexidine gluconate MIC values (75%) calculated for each *Salmonella* isolate only ranged between 9.52 mg/L and 14.29 mg/L. When comparing groups, the average MIC values regarding Ophidian, Chelonian, and Saurian isolates were 11.98, 11.25, and 12.19 mg/L, respectively, the differences were not statistically significant (*p* = 0.802) (Table 4).

**Table 4.** Chlorhexidine gluconate minimum inhibitory concentrations, minimum bactericidal concentrations, minimum biofilm inhibitory concentrations, minimum biofilm eradication concentrations and biofilm formation results.


Abbreviations: MIC, minimum inhibitory concentration; MBC, minimum bactericidal concentration; MBIC, minimum biofilm inhibitory concentration; MBEC, minimum biofilm eradication concentration; N.S., nonsignificant. \* Values above 714.29 mg/L were not included.

Regarding MBC, the overall mean value was 38.8 mg/L ± 50.25, with a minimum value of 9.52 mg/L (observed towards a Chelonian isolate), and a maximum value of 247.62 mg/L (regarding a Saurian isolate), with a median value of 23.22 mg/L. Although a high variability in MBC values was found, towards half of the studied isolates those values ranged between 11.91 mg/L and 23.81 mg/L. When comparing groups, the average MBC values obtained regarding the Ophidian isolates, 86.84 mg/L, the Chelonian isolates, 27.87 mg/L and the Saurian isolates, 33.87 mg/L, were not statistically different (*p* = 0.257).

Chlorhexidine gluconate demonstrated to have a bactericidal effect in the majority of the *Salmonella* isolates (84.37%), since only five isolates (15.63%) had MBC/MIC ratio above 4 (Supplementary Table S1).

#### *2.5. Biofilm Formation in the Air-Liquid Interface*

The biofilm formation capability of the *Salmonella* isolates obtained from pet reptiles was studied by observing the development of a pellicle in the air-liquid interface. Of all isolates, only three (9.37%) were not able to form biofilms, thus the vast majority (90.63%) formed a clearly detectable biofilm. The shortest period required for biofilm formation was three days, and the longest was six days. The average number of days until the biofilm was formed was 4.4 days ± 0.90, and the majority of the isolates (75.9%) were able to form the biofilm in five days or less.

The differences on the average number of days until biofilm formation by Ophidian (5.1 days), Chelonian (4.7 days), and Saurian isolates (4.2 days) were considered not to have statistical significance (*p* = 0.211) (Table 4).

#### *2.6. Minimum Biofilm Inhibitory Concentration and Minimum Biofilm Eradication Concentration Determination*

The minimum biofilm inhibitory concentration (MBIC) and minimum biofilm eradication concentration (MBEC) values of chlorhexidine gluconate regarding each isolate can be found on Supplementary Table S1.

The MBIC values ranged from 14.29 mg/L to 232.15 mg/L, with an average value of 68.41 mg/L ± 32.68, and a median value of 71.43 mg/L. Despite the broad range of values, 71.43 mg/L of chlorhexidine gluconate was the MBIC value for more than half (59.4%) of the isolates tested. When comparing groups, the recorded average MBIC values regarding Ophidian, 57.15 mg/L, Chelonian, 64.02 mg/L, and Saurian isolates, 72.87 mg/L, did not statistically differ (*p* = 0.509) (Table 4).

Concerning the MBEC values, the average chlorhexidine gluconate biofilm eradication concentration was 360.08 mg/L ± 235.18, with a minimum of 33.34 mg/L and a maximum of 714.29 mg/L, and a median value of 392.86 mg/L. Regarding six isolates, one Ophidian, one Chelonian, and four Saurian related isolates, the MBEC values were considered to be greater than the highest concentration tested, therefore, the results were expressed as >714.29 mg/L.

#### **3. Discussion**

Several research groups from multiple countries have reported the isolation of *Salmonella* spp. from pet or captive reptiles, including turtles, lizards, and snakes [26–32]. Although this is not a recent issue, to the author's best knowledge, the present report is the first regarding the isolation of *Salmonella* spp. from healthy pet reptiles in Portugal. Our results point out to an overall *Salmonella* spp. prevalence of 41%, which is similar to studies performed with captive or pet reptiles in Australia (47%) [32], Spain (48%) [14], Norway (43%) [33], or Sweden (49%) [7], but higher than reports from smuggled reptiles in Taiwan (30.9%) [15] or captive animals in Croatia (13%) [29] or in New Zealand (11.4%) [31]. Furthermore, in our study, the prevalence of *Salmonella* spp. was higher in both Ophidians (50%) and Saurians (51.9%), when compared with Chelonians (20.9%) (*p* = 0.016). The lower isolation rates in turtles when compared with other reptiles can be associated with seasonal variations, observed when turtles are preparing for hibernation [28], but also with the diet of these animals [12,15,16]. In fact, the sample collection period occurred before the hibernation stage of Chelonians, during the colder months of the year. Nevertheless, the impact of pet turtles in the reptile-associated salmonellosis scenario should not be underestimated, since exposure to *Salmonella* positive turtles has been linked to disease outbreaks [34–36].

High levels of antimicrobial susceptibility to the majority of the antibiotics tested were found in most the *Salmonella* isolates, and only three isolates (9.37%) were resistant to three or more of the compounds tested. Our results differ from those reported in a recent study carried out in Spain, in which 72% of the isolates were considered to be multidrug resistant [14]. *Salmonella* isolates from reptiles are known to be resistant to several antibiotics frequently used in therapy. This not only implies that reptiles can shed multidrug resistant salmonellae to the environment and to other animals, including humans, but also the genes responsible for those antimicrobial resistances could be transferred to other enteric bacteria [17].

All the isolates studied expressed both hemolytic and lipolytic behaviors on plate tests. These two virulence phenotypes should be further investigated. Hemolysis is not associated with human non-typhoidal salmonellosis cases, and it has not been reported as a virulence trait by other authors, though it was shown that the hemolytic activity in *Salmonella enterica* serovar Typhimurium is dependent of the pathogenicity island 1 type III secretion system [37]. Extracellular lipases have been proposed as potential virulence factors in other pathogenic bacteria, such as *Staphylococcus aureus*, *Staphylococcus epidermis*, or *Pseudomonas aeruginosa* [38], though their role in *Salmonella* spp. virulence does not seem to be fully studied [39]. DNase testing pointed out the presence of extracellular desoxiribonucleases in more than half of the isolates. Gelatinase activity was not detected, even though it is a biochemical characteristic of *Salmonella enterica* subsp. *arizonae* [40]. It is possible that the analyzed isolates harbored the gene responsible for gelatin digestion, even though the isolates under the present study conditions did not express that phenotype. Recently, Salmonellae isolated from ready-to-eat shrimps were also found to express hemolytic, lipolytic, DNA degrading activity and also gelatinase production [41]. Additional studies are necessary in order to understand the extent of the possible role of these phenotypes both in animal and in human *Salmonella* infections. Actually, from the obtained data, the possibility of the same bacterial clone infecting different animals and adapting/evolving within the hosts cannot be excluded. Although a molecular based approach would bring valuable information regarding the identity and the possible genetic relationship between

the studied isolates, the present report was designed to clarify the therapeutic potential of chlorhexidine, testing one isolate from each animal. Despite the possible genetic similarities, the foremost important assessed feature of each *Salmonella* isolate was the phenotypical behavior, namely the susceptibility to a commonly used biocide, chlorhexidine gluconate. Thus, the information resulting from this study can be adapted and applied in reptile medicine.

In the present study, the occurrence of both bactericidal and bacteriostatic effects of chlorhexidine gluconate is an example of the duality of the antimicrobial effect that takes place according to the applied concentration. Previous reports revealed chlorhexidine gluconate MIC values ranging from 8 to 16 mg/L when tested towards *Salmonella* Bredeney, Dublin, Gallinarum, Montivideo Virshow and Typhimurium [42]. Another study recorded a range of MIC values for *Salmonella* isolates of animal origin (broilers, cattle and pigs) between 2 and 64 mg/L [43]. More recent studies reported MIC values of 1–8 mg/L in turkey *Salmonella* isolates from commercial processing plants, and MIC values below 4 mg/L to 64 mg/L regarding different *Salmonella* serovars isolated from chicken and in egg production chains [6,44,45]. The overall mean MIC value calculated for the studied *Salmonella* spp. isolates from pet reptiles was 11.90 mg/L, which is coherent with those values. The global mean MBC value is approximately three times the mean MIC. The suggestion that both MIC and MBC values should be included in the monitorization of biocidal susceptibility is consistent with the results obtained in this study considering that both values provide complementary information [46].

Although MIC an MBC values are valuable for evaluating the antimicrobial effect of chlorhexidine gluconate, the previous studies were carried out with planktonic cells. The fact that the *Salmonella* spp. isolates are capable of biofilm formation is worrisome, since *Salmonella* organized in biofilms is less susceptible to disinfectants than planktonic cells, with preliminary studies indicating that disinfectants used at an effective concentration for *Salmonella* biofilm reduction can cause the selection of more virulent cells [47]. The high frequency of the studied reptile *Salmonella* isolates capable of forming biofilms (90.63%) is similar to previously reported data. High frequencies of pellicle formation in the airliquid interface by *Salmonella* Agona (100%), *Salmonella* Montevideo (100%), and *Salmonella* Senftenberg (88%) were already described [48]. However, in the same study, only 55% of the *Salmonella* Typhimurium isolates tested were biofilm producers [48]. On other studies, the expression of biofilm formation by *Salmonella* Typhimurium isolates varied under the same circumstances, with different strains and morphotypes demonstrating different biofilm capabilities [49,50].

Biofilms are common on liquid-hard surfaces interfaces [51], such as in certain type of reptile cages or in aquariums. In order to simulate a more realistic approach to the effects of chlorhexidine gluconate on *Salmonella* cultures, the antimicrobial action of chlorhexidine gluconate activity was tested on the biofilms formed by the reptile *Salmonella* isolates during a 24 h-period. A chlorhexidine gluconate MBIC value within the concentration limits tested was obtained regarding all the *Salmonella* isolates studied. Regarding the *Salmonella* isolates towards which the MBEC values exceeded 714.29 mg/L, chlorhexidine gluconate was simply not effective in terms of eradicating those biofilms. Overall, chlorhexidine gluconate MBIC and MBEC results show that *Salmonella* biofilms are less susceptible to this biocide, what is consistent with a previous report which stated that three-day old *Salmonella* Typhimurium biofilms were less susceptible to chlorhexidine gluconate when compared to the corresponding planktonic cells [52].

#### **4. Materials and Methods**

#### *4.1. Sample Collection and Salmonella spp. Isolation*

A total of 78 cloacal swabs were obtained from pet reptiles, specifically 43 Chelonians (commonly referred as turtles), 27 Saurians (commonly named lizards), and eight Ophidians (usually known as snakes). The cloacal swabs were performed using cotton swabs in AMIES transport media (VWR, Amadora, Portugal) during rou-

tine health check-ups at the house of the owners or at pet shops, all located in the Lisbon Metropolitan Area, Portugal. All animals were cared for according to the rules given by the current EU (Directive 2010/63/EC) and national (DL 113/2013) legislation and by the competent authority (Direção Geral de Alimentação e Veterinária, DGAV, (www.dgv.min-agricultura.pt/portal/page/portal/DGV, accessed on 20 January 2021) in Portugal. Verbal informed consent was obtained from all the owners. Trained veterinarians performed sample collection of all the samples, following standard routine procedures. After collection, swabs were kept under refrigeration conditions (4 ◦C) for no longer than 48 h until processing in the Microbiology Laboratory of the Veterinary Medicine Faculty—University of Lisbon for *Salmonella* spp. isolation.

Briefly, each cloacal swab was homogenized and incubated in 5 mL of buffered peptone water (BPW) (Scharlau, Valencia, Spain) for 18 ± 2 h at 37 ◦C. After the initial incubation, 1 mL of BPW was then added to 10 mL of Muller-Kaufmann Tetrathionate (MKTT) Broth (Oxoid, Hampshire, UK) and incubated for 18–24 h at 37 ◦C. Simultaneously, 0.1 mL of the BPW solution was added to 10 mL Rappaport–Vassiliadis broth (Oxoid, Hampshire, England) and the resulting suspension was incubated for 18–24 h at 41.5 ◦C. Afterwards, suspensions were inoculated in Hektoen Agar (Liofilchem, Teramo, Italy) and xylose lysine deoxicholate agar (Scharlau, Valencia, Spain) plates, by streaking, and incubated at 37 ◦C for 20 ± 2 h. The resulting presumptive *Salmonella* spp. colonies were selected and transferred to triple sugar iron (TSI) Agar (Scharlau, Valencia, Spain) and to urea broth (Oxoid, Dadirlly, France) and incubated for 20 ± 2 h at 37 ◦C. Presumptive *Salmonella* spp. isolates were identified through the growth pattern in TSI agar and in Urea Broth. The method described is an adaptation of a previously described method [53]. *Salmonella* spp. isolates were identified using biochemical profile system API 20E (BioMérieux, Craponne, France). The biochemical identification was later confirmed by agglutination with Antiserum *Salmonella* OMNIVALENT Omni-O (Bio-Rad Laboratories, Inc., Marnes-la-Coquette, France).

#### *4.2. Antimicrobial Susceptibility Testing*

Antimicrobial susceptibility testing was performed by the disk diffusion method, according to Clinical and Laboratory Standards Institute guidelines (CLSI) [54]. The tested antibiotics were amoxicillin/clavulanic acid (AMC, 30 µg), ampicillin (AMP, 10 µg), amikacin (AK, 30 µg), chloramphenicol (C, 30 µg), gentamicin (CN, 10 µg), cefotaxime (CTX, 30 µg), enrofloxacin (ENR, 5 µg), nalidixic acid (NA, 30 µg), penicillin (P, 10 U), ciprofloxacin (CIP, 5 µg), sulfamethoxazole/trimethoprim (SXT, 25 µg), and tetracycline (TE, 30 µg). All antibiotics were purchased from Oxoid, Dadirlly, France. *Escherichia coli* ATCC 25922 was used as the control strain for test performance. Multidrug resistance (MDR) phenotype was considered to be present whenever an isolate revealed resistance to three or more antimicrobial compounds belonging to different classes [55].

#### *4.3. Virulence Phenotype Analysis*

In order to assess the virulence phenotype of the *Salmonella* isolates, plate tests were performed for evaluating their DNase, gelatinase, hemolytic and lipase activities.

DNase activity testing was performed by streaking the bacterial isolates on DNase test Agar plates (Liofilchem, Teramo, Italy) supplemented with 0.01% toluidine blue. The plates were incubated for 48 h at 37 ◦C and positive results showed a transparent halo surrounding the colonies.

Gelatinase activity was tested by streaking the isolates on Gelatinase test Agar plates (Liofilchem, Teramo, Italy), followed by incubation at 37 ◦C for 48 h. Afterwards, plates were flooded with a mercury chloride solution and the gelatinase positive isolates showed a transparent halo around the colonies.

Production of hemolysins was determined by streaking the isolates on Columbia Agar plates supplemented with 5% sheep blood (BioMérieux, Craponne, France) and incubated for 48 h at 37 ◦C. The presence of clear halos surrounding the colonies was interpreted as β-hemolysis.

Lipase activity testing was achieved by culturing the isolates in Spirit Blue Agar plates (Difco, Algés, Portugal) supplemented with Tween 80 (30 g/L) and incubating for 48 h at 37 ◦C. Lipase producing isolates exhibited clear halos around the colonies.

#### *4.4. Chlorhexidine Gluconate Minimum Inhibitory Concentration and Minimum Bactericidal Concentration Determination*

The in vitro susceptibility profile of the *Salmonella* isolates to chlorhexidine gluconate was assessed by an adapted protocol based on the microtiter broth dilution method [56,57]. Isolates were grown in a nonselective brain heart infusion (BHI) agar medium (VWR Chemicals, Leuven, Belgium) at 37 ◦C for 24 h. Bacterial suspensions with 10<sup>8</sup> CFU/mL were prepared directly from plate cultures in sterile normal saline (Merck, Germany) to a 0.5 McFarland suspension. The bacterial suspensions were then diluted in fresh BHI broth (VWR Chemicals, Leuven, Belgium) to a concentration of 10<sup>7</sup> CFU/mL.

Chlorhexidine gluconate dilutions were prepared from a stock solution at a concentration of 4% (*w*/*v*) (AGA, Lisboa, Portugal). A volume of 25 µL of chlorhexidine gluconate at 0.5, 0.1, 0.05, 0.01, 0.005 and 0.001% were distributed in 96-well flat-bottomed polystyrene microtiter plates (Nunc, Thermo Fisher Scientific, Roskilde, Denmark), apart from the negative and positive controls. All the wells were inoculated with 150 µL of the 10<sup>7</sup> CFU/mL bacterial suspensions, with exception of the negative control wells, which contained only broth medium. Therefore, the final concentration of chlorhexidine gluconate in the wells corresponded to 714.28, 142.86, 71.43, 14.29, 7.14, and 1.43 mg/L. Afterwards, microplates were statically incubated for 24 h at 37 ◦C. The minimum inhibitory concentration (MIC) was determined as the lowest concentration of chlorhexidine gluconate that visually inhibited microbial growth.

The minimum bactericidal concentration (MBC) value was assessed by inoculating 3 µL of the suspensions from the wells were no growth was observed on BHI agar plates, which were incubated at 37 ◦C for 24 h. MBC was determined as the lowest chlorhexidine gluconate concentration that did not allow colony development [57,58].

The ratio between MBC and MIC was calculated in order to determine the antimicrobial effect of chlorhexidine gluconate. The effect was considered to be bactericidal when the MBC was no more than fourfold the MIC, or bacteriostatic when the ratio exceeded four [58].

All experiments were conducted in duplicate and independent assays were performed at least three times in different dates.

#### *4.5. Biofilm Formation in the Air-Liquid Interface*

Biofilm forming ability was assessed through a biofilm formation assay in the air– liquid interface, by inoculating 0.5 mL of an overnight BHI broth culture, adjusted to a 0.5 McFarland standard, in a 4.5 mL of Luria broth (LB) without NaCl (1:10), prepared using yeast extract (Oxoid, Hampshire, England) and bacto tryptone (BD, Oeiras, Portugal). Isolates were incubated at 28 ◦C for eight days and each isolate was visually examined for pellicle formation on a daily basis [49]. The isolates capable of forming a pellicle in two distinct occasions were considered to be positive for biofilm formation, and the number of days required until the pellicle was perceivable was used to calculate the mean time for biofilm formation.

All assays were repeated in three independent dates, including 10% replicates.

#### *4.6. Chlorhexidine Gluconate Minimum Biofilm Inhibitory Concentration and Minimum Biofilm Eradication Concentration Determination*

The antimicrobial susceptibility of the *Salmonella* isolates when embedded in a 24 h biofilm was evaluated by a modified version of the Calgary Biofilm Pin Lid Device [57,59]. For minimum biofilm inhibitory concentration (MBIC) and minimum biofilm eradication concentration (MBEC) assays, the bacterial isolates were grown in BHI agar medium (VWR Chemicals, Leuven, Belgium) at 37 ◦C for 24 h. Bacterial suspensions with approximately 10<sup>8</sup> CFU/mL were prepared directly from plate cultures in sterile normal saline

(Merck, Darmstadt, Germany) by comparison with a 0.5 McFarland standard (BioMérieux, Craponne, France). Suspensions were then diluted in fresh BHI broth (VWR Chemicals, Leuven, Belgium) to a concentration of 10<sup>6</sup> CFU/mL. Then, 175 µL of the bacterial suspensions were distributed in 96-well flat-bottomed polystyrene microtiter plates, covered with 96-peg polystyrene lids (Nunc-TSP; Thermo Fisher Scientific, Roskilde, Denmark) and statically incubated for 24 h at 37 ◦C, allowing biofilm formation on the pegs. Peg lids were then rinsed three times in sterile normal saline to remove planktonic bacteria and placed on new microplates containing the set of chlorhexidine gluconate solutions previously described, corresponding to a final concentration by well of 714.28, 142.86, 71.43, 14.29, 7.14, and 1.43 mg/L.

Microplates were again incubated for 24 h at 37 ◦C, without shaking. After incubation, peg lids were removed, and the MBIC value was determined as the lowest chlorhexidine gluconate concentration that visually inhibited microbial growth. Subsequently, in order to determine the MBEC value, peg lids were rinsed three times in sterile normal saline, placed in new microplates containing only 175 µL of fresh BHI medium and incubated in an ultrasound bath (Grant MXB14, Essex, England), at 50 Hz during 15 min in order to disperse the biofilm-based bacteria from the peg surface. Afterwards, peg lids were discarded, and microplates were covered with normal lids and incubated for 24 h at 37 ◦C. The MBEC value was determined through direct observation of bacterial growth in the wells and defined as the lowest chlorhexidine gluconate concentration that visually eliminates the microbial growth [57].

Experiments were conducted in duplicate and independent assays were performed at least two times on different dates.

#### *4.7. Statistical Analysis*

For statistical analysis, the associations between frequency of *Salmonella* isolation and reptile group, AMR *Salmonella* and reptile group and virulence phenotype and reptile group were evaluated using the Fisher exact test. Association between different MIC, MEC and MBIC values of chlorhexidine gluconate on *Salmonella* isolates, the number of days until biofilm formation and the reptiles group was assessed recurring to the Brown–Forsythe robustness test based on a one-way ANOVA test. All statistical tests were performed on IBM SPSS Statistical program version 26 for Windows (SPSS Inc., Chicago, IL, USA). Associations were considered to be significant whenever *P* values were less than 0.05.

#### **5. Conclusions**

The present study reports the isolation of *Salmonella* from healthy pet reptiles and stresses their possible role in human non-typhoidal salmonellosis cases. Although presenting high levels of antimicrobial susceptibility, the expression of phenotypical virulence traits and the ability to form biofilms by these isolates are worrisome. Pet reptile owners should always employ good hygiene practices whenever manipulating the animals, but also when in contact with the environment in which the animals are kept. Overall, the use of chlorhexidine gluconate was considered to be effective, both in planktonic cells and biofilms, pointing out the potential of this biocide's use in reptile clinics.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/2079-638 2/10/3/324/s1, Table S1: Detailed information regarding the studied Salmonella isolates and the results of Antimicrobial Susceptibility analysis, Virulence Phenotype analysis, Minimum Inhibitory Concentration, Minimum Bactericidal Concentration, Minimum Biofilm Inhibitory Concentration and Minimum Biofilm Eradication Concentration Determination and Biofilm Formation.

**Author Contributions:** Conceptualization: F.B. and M.O.; methodology: A.C.C., I.D. and A.R.; software: J.B.C.; formal analysis: A.C.C., I.D. and J.B.C.; writing—original draft preparation: J.B.C.; writing—review and editing: J.B.C., F.B. and M.O.; supervision: M.O.; funding acquisition: F.B. and M.O. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by CIISA—Centro de Investigação Interdisciplinar em Sanidade Animal, Faculdade de Medicina Veterinária, Universidade de Lisboa, Project UIDB/00276/ 2020 (funded by FCT—Fundação para a Ciência e Tecnologia IP).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this study are available in Supplementary Table S1.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


#### *Article*
