Efflux Pumps and Different Genetic Contexts of tet(X4) Contribute to High Tigecycline Resistance in Escherichia fergusonii from Pigs

Tigecycline is a last-resort antibiotic for the treatment of infections caused by multidrug-resistant bacteria. The emergence of plasmid-mediated tigecycline resistance genes is posing a serious threat to food safety and human health and has attracted worldwide attention. In this study, we characterized six tigecycline-resistant Escherichia fergusonii strains from porcine nasal swab samples collected from 50 swine farms in China. All the E. fergusonii isolates were highly resistant to tigecycline with minimal inhibitory concentration (MIC) values of 16–32 mg/L, and all contained the tet(X4) gene. In addition, 13–19 multiple resistance genes were identified in these isolates, revealed by whole-genome sequencing analysis. The tet(X4) gene was identified as being located in two different genetic structures, hp-abh-tet(X4)-ISCR2 in five isolates and hp-abh-tet(X4)-ΔISCR2-ISEc57-IS26 in one isolate. The role of efflux pumps in tigecycline resistance was evaluated by using inhibitor carbonyl cyanide 3-chlorophenylhydrazone (CCCP). The MIC values of tigecycline showed a 2- to 4-fold reduction in the presence of CCCP, indicating the involvement of active efflux pumps in tigecycline resistance in E. fergusonii. The tet(X4) gene was found to be transferable to Escherichia coli J53 by conjugation and resulted in the acquisition of tigcycline resistances in the transconjugants. Whole-genome multilocus sequence typing (wgMLST) and phylogenetic analysis showed a close relationship of five isolates originating from different pig farms, suggesting the transmission of tet(X4)-positive E. fergusonii between farms. In conclusion, our findings suggest that E. fergusonii strains in pigs are reservoirs of a transferable tet(X4) gene and provide insights into the tigecycline resistance mechanism as well as the diversity and complexity of the genetic context of tet(X4) in E. fergusonii.


Introduction
The emergence and spread of bacterial antimicrobial resistance pose a serious threat to food safety as well as human and animal health [1]. Of special concern is the emerging resistance to tigecycline, a tetracycline-class antibacterial agent, which has been regarded as one of the few therapeutic choices left to combat multidrug-resistant (MDR) bacterial infections [2,3].
The tet(X4) gene is a novel plasmid-mediated high-level tigecycline resistance gene discovered in Enterobacteriaceae and Acinetobacter isolates from animals and humans in China in 2019 [2]. It was worth noting that Tet(X4) could degrade all tetracyclines [25], including tigecycline and the USFDA newly approved eravacycline [26], which poses a new threat to public health. So far, tet(X4) has mainly been discovered in E. coli from animal origin and sporadically in several other bacteria species, such as Klebsiella pneumoniae, Shigella flexneri, Shigella boydii, Shigella sonnei, Aeromonas caviae, Acinetobacter sp., and Escherichia fergusonii [2,18,25,27,28]. Understanding the genetic context of tet(X4) in different bacterial species is important to uncovering the resistance mechanism and its potential effects on human health, but this has not yet been thoroughly investigated.
The aim of this study was to characterize six tigecycline-resistant E. fergusonii isolates from porcine nasal samples obtained across 50 farms in Fujian, China, including analyzing the efflux pump activity in relation to tigclycline resistance, genetic context of tet(X4) genes, as well as transferability and phylogenetic relationship of the strains in order to gain insight into the possible public-health impact of tigecycline-resistant E. fergusonii from pigs.

Figure 1.
Phylogenetic analysis based on whole-genome multilocus sequence typing (wgMLS the six Escherichia fergusonii isolates.

Genetic Context of Tigecycline Resistance Gene
All strains contained the tet(X4) gene. In addition to tet(X4), tet(B) without mu was identified in isolate 2022GZP273, and tet(A), as well as tet(M) with the same tions, was found in all the other isolates (Tables 1 and S1). For five of the six isolate tet(X4) gene was located in a classic genetic context, hp-abh-tet(X4)-ISCR2, while a genetic context of tet(X4), hp-abh-tet(X4)-ΔISCR2-ISEc57-IS26, was observed in i 2022GZP273 ( Figure 2).

Genetic Context of Tigecycline Resistance Gene
All strains contained the tet(X4) gene. In addition to tet(X4), tet(B) without mutation was identified in isolate 2022GZP273, and tet(A), as well as tet(M) with the same mutations, was found in all the other isolates (Table 1 and Table S1). For five of the six isolates, the tet(X4) gene was located in a classic genetic context, hp-abh-tet(X4)-ISCR2, while a new genetic context of tet(X4), hp-abh-tet(X4)-∆ISCR2-ISEc57-IS26, was observed in isolate 2022GZP273 ( Figure 2).
The classic genetic context, hp-abh-tet(X4)-ISCR2, has been mainly identified in plasmids in E. coli and less frequently in other bacteria species (Figure 2, Supplementary Table S2). Two E. fergusonii isolates have previously been reported to be hosts of this structure, as indicated in Figure 2 [29,30]. Compared with the reported genetic context in the two E. fergusonii isolates, hp-abh-tet(X4)-ISCR2 share the same downstream structure as those in pHNCF11W-tetX4 (GenBank accession number CP053047), and the same upstream structure as pQZZ116-tetX-190K (GenBank accession number CP095844) ( Figure 2).

Efflux Pumps' Activity
The MIC value of CCCP is 8 mg/L for all isolates. After addition of 2 mg/L CCCP, there was a fourfold decline in the MIC value for the tigecycline of one isolate (2022GZP491) and half a decline in the MIC values for the tigecycline of the remaining five isolates (2022GZP462, 2022GZP331, 2022GZP221, 2022GZP175 and 2022GZP273), indicating the presence of active efflux pumps mediating tigecycline resistance in all isolates ( Table 1).
The known efflux pumps contributing to tigecycline resistance, AcrAB-TolC, AcrZ and NorM, were identified in the genome sequences of all isolates. In addition to the known efflux pumps, several efflux pumps and porins associated with multi-drug resistance, such as OmpF outer membrane porin, MarA and its local repressor MarR, were also identified in all isolates (Table S3).  Table  S2). Two E. fergusonii isolates have previously been reported to be hosts of this structure, as indicated in Figure 2 [29,30]. Compared with the reported genetic context in the two E. fergusonii isolates, hp-abh-tet(X4)-ISCR2 share the same downstream structure as those in pHNCF11W-tetX4 (GenBank accession number CP053047), and the same upstream structure as pQZZ116-tetX-190K (GenBank accession number CP095844) (Figure 2).
In the new genetic context, hp-abh-tet(X4)-ΔISCR2-ISEc57-IS26, ISCR2 was truncated by ISEc57 and linked by a copy of IS26 (Figure 2). In the analysis, two E. coli isolates, obtained from a human gut in Singapore in 2019, were identified as containing hp-abh-

Conjugation
Polymerase Chain Reaction (PCR) results confirmed the successful transfer of the tet(X4) genes from all E. fergusonii strains to a plasmid-free recipient E. coli J53. Antimicrobial susceptibility testing revealed that the acquisition of the tet(X4) genes by E. coli J53 caused at least a 32-fold increase for tigecycline ( Table 2). The conjugation rates ranged from 3.4 × 10 −7 to 2.6 × 10 −6 transconjugant per recipient cell in E. fergusonii strains ( Table 2).

Phylogenetic Analysis
The wgMLST and phylogenetic analysis showed that isolates 2022GZP462, 2022GZP331, 2022GZP221, 2022GZP175, and 2022GZP491 were all highly related, while they were distantly related to 2022GZP273 (Figure 1).

Discussion
E. fergusonii is an opportunistic pathogen infecting humans and animals [31]. It causes a wide range of infections in poultry and has incurred significant economic losses worldwide [32], and it has been reported in several clinical cases in humans, including wound infections, urinary tract infections, bacteremia, and diarrhoea [33]. E. fergusonii from livestock have been reported to be an underrated repository for antimicrobial resistance, especially with regards to mcr-1 gene [34]. Thus, the emergence of tigecycline resistance in E. fergusonii significantly increases its importance to public health [29,34].
Tigecycline resistance has been found in many species. However, tigecycline-resistant E. fergusonii isolates have only been reported in two samples in China, a pig feces sample and a chicken feces sample [34,35]. In these two isolates, the tigecycline resistance was found to be associated with the tet(X4) gene, which was located in the classic genetic structure hp-abh-tet(X4)-ISCR2. Similar to these findings, for five out of six E. fergusonii isolates in the present study, the tet(X4) gene was located in hp-abh-tet(X4)-ISCR2. Interestingly, the genetic structure of tet(X4) in these E. fergusonii isolates shared either the same upstream or downstream sequences, indicating that a recombination process at an earlier stage is likely to have happened at fixed sites in E. fergusonii. Importantly, the wide distribution of the hpabh-tet(X4)-ISCR2 in broad bacterial species indicates its high transferability, which has also been observed in this study and which might cause an expansion of tigecycline resistance.
Notably, a novel genetic structure, hp-abh-tet(X4)-∆ISCR2-ISEc57-IS26, was observed in an E. fergusonii isolate. In this structure, the downstream ISCR2 was truncated and inserted by ISEc57 and was associated with IS26. ISCR2 is likely to have been truncated by different IS elements, such as IS26, IS1D, IS1R, ISEc57, and ISKpn19 [18]. However, the unit hp-abh-tet(X4)-∆ISCR2-ISEc57 has only been observed in two E. coli isolates. Notably, the unit was found to be associated with one more IS26 in the E. fergusonii isolate in this study, suggesting that the emergence of this structure in E. fergusonii was a newer event at a molecular level. IS26 can mediate the formation of a hybrid plasmid between tet(X4)positive and -negative plasmids [36]. Thus, the combined effect of ISCR2, ISEc57 and IS26 might be a major driving force in the rapid expansion of tet(X4) in E. fergusonii.
Besides the tet(X4) gene, efflux pumps were observed to be a contributor to tigecycline resistance in E. fergusonii isolates in the current study. The contribution of efflux pumps to tigecycline resistance in different bacterial species has been investigated in many studies [20,37,38]. However, the activity of efflux pumps on tigecycline resistance in E. fergusonii has not been elucidated. In this study, by searching through genome sequences, we identified various RND-type efflux pumps (AcrAB-TolC, AcrZ and NorM) that existed in all E. fergusonii isolates, which have been confirmed to be associated with tigecycline resistance [14,38,39]. The contribution of efflux pumps to tigecycline resistance in E. fergusonii was further revealed by the efflux pump inhibitor, CCCP. CCCP has been shown to have a good activity against RND-type efflux pumps associated with tigecycline resistance [37]. In this study, the addition of CCCP resulted in a 2-to 4-fold decline in MIC values for tigecycline, indicating that RND-type efflux pumps that exist in E. fergusonii isolates also partly contribute to the high level of tigecycline resistance. However, in addition to tet(X4) and RND-type efflux pumps, which were confirmed to contribute to tigecycline resistance in E. fergusonii, we cannot exclude the fact that other factors may also contribute to phenotypical tigclycline resistance, such as mutations in tet(A) and tet(M) genes, OmpF outer membrane porin, MarA and its local repressor MarR, which have been described as contributing to overexpression of the AcrAB efflux pump and may indirectly lead to phenotypical tigclycline resistance in E. fergusonii [40][41][42][43]. Therefore, more in vitro research is needed to determine how these different mutation types and factors are involved in tigclycline resistance in E. fergusonii.
Unexpectedly, the five E. fergusonii isolates harboring the same genetic structure of tet(X4) were closely related, despite being obtained from different farms. This finding could indicate that E. fergusonii in pigs are reservoirs of the tet(X4) gene and that tet(X4)positive E. fergusonii may have been transmitted between different pig farms in China. Since E. fergusonii isolates in this study were all multi-drug-resistant, the emergence of transferable tet(X4) in E. fergusonii from pigs needs to be monitored and further investigated, as it may spread along the food chain to humans.

Bacterial Isolation and Identification
In August 2019, a total of 250 porcine nasal swab samples were collected from a total of 50 farms in Fujian, China. Five samples were collected from each farm and pooled for strain isolation. A low-temperature box with ice was used to transport samples from the field to the laboratory for further processing. The samples were incubated in buffered peptone water (BPW) broth for 18 to 24 h and then inoculated onto Luria-Bertani (LB; Guangdong Huankai Microbial Sci. & Tech., Guangzhou, China) agar plates with 2 mg/L tigecycline. Colonies were identified using 16S rRNA gene sequencing using universal primers (16S-F, 5 -AGAGTTTGATCCTGG CTCAG-3 ; 16S-R, 5 -GGTTACCTTGTTACGACTT-3 ) [44].

Antimicrobial Susceptibility Testing
MIC of tigecycline (Sigma-Aldrich, St. Louis, MO, USA) was determined by broth microdilution [45]. The resistance breakpoint (>2 mg/L) was interpreted as resistant according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines (http://www.eucast.org/clinical_breakpoints/, (accessed on 5 March 2021). E. coli ATCC25922 was used as a control, and all experiments were performed with three biological replicates.

Effect of CCCP on Tigecycline MIC
The activity of the efflux pump system on tigecycline MIC was tested using efflux pump inhibitor CCCP [37]. The MIC of CCCP for each isolate was tested first, and then a final concentration of 1/4 MIC (2 mg/L) of CCCP (subinhibitory concentrations that did not affect bacterial growth) was added to each well when testing MIC for tigecycline.

Phylogenetic Analysis of the Genomic Sequences
In order to assess the relatedness of the six E. fergusonii isolates, a minimum spanning tree was constructed in Enterobase using the RapidNJ algorithm and the whole genome multilocus sequence typing (wgMLST) (wgMLST scheme available on EnteroBase) scheme [50].

Conjugation Experiments
The transferability of tet(X4) was assessed by performing the conjugation experiment, using solid mating on a filter (Whatman, Maidstone, UK). The sodium azide-resistant E. coli strain J53 was used as a recipient strain [51].
Briefly, recipient and donor strains were cultured overnight in LB broth, and then the cells were harvested, washed with saline, mixed together in a ratio of 1:1, and spotted onto a 0.45 µm pore size filter (Millipore) on LB plates to be cultured for 20 h. They were also spotted individually on LB plates as controls. The transconjugants were selected on LB plates containing 150 mg/L sodium azide and 2, 4, 8, or 16 mg/L of tigecycline after being cultured for 24-48 h. The test was conducted at least three times. Control spots were transferred to the same selective media to make sure that no growth was observed.
The conjugation frequency was calculated as the ratio of transconjugants over the number of recipients. The transfer of the plasmid was confirmed by PCR, with the primers listed in Table S4.

Nucleotide Sequence Accession Numbers
The Illumina sequence data were deposited in the Enterobase database under the barcode numbers ESC_ZA4185AA, ESC_ZA4186AA, ESC_ZA4187AA, ESC_ZA4188AA, ESC_ZA4189AA, and ESC_ZA4190AA.

Conclusions
To summarize, this study for the first time reports the involvement of efflux pumps and transferable diverse genetic structures of tet(X4) in the tigecycline resistance of E. fergusonii. Our study revealed that E. fergusonii in pigs are reservoirs of tet(X4) gene and that they may have been transmitted between different pig farms in China, which poses potential hazards to associated pork products' safety and a public health risk and which thus requires continuous investigations.