Analysis of Extended Spectrum Beta Lactamase (ESBL) Genes of Non-Invasive ESBL Enterobacterales in Southeast Austria in 2017

Extended spectrum beta lactamases producing Enterobacteriaceae are a major player in the antibiotic resistance challenge. In general, the situation regarding antibiotic resistance in Austria is very good compared to many other countries. Perhaps this is why there is a lack of data on the distribution of ESBL genes in the clinical setting. The aim of this study was to collect data on ESBL genes from a larger sample of human non-invasive clinical isolates from one region in Austria. In total, 468 isolates from different sample materials isolated at the Medical University of Graz from 2017 were examined. The most frequent organisms were Escherichia coli and Klebsiella pneumoniae. Among the enzymes produced, CTX-M-15 was clearly dominant, exotic ESBLs were only represented by three Proteus mirabilis isolates harboring genes for VEB-6 and one P. mirabilis for CTX-M-2, respectively. Compared to other countries, the results are in line with the expectations. The data help to better classify the many studies from the non-clinical field in Austria and to shift the focus slightly away from the exotic results and sample sites.


Introduction
Extended spectrum beta lactamases (ESBLs), especially found in Enterobacteriaceae, are one of the big players in the rise in antibiotic resistance in Gram-negative bacilli. ESBLs are roughly defined as enzymes that lead to resistance to penicillins (at least) to third and/or fourth generation cephalosporins and aztreonam, and can be inhibited by betalactamase inhibitors. However, there are also variants that are insensitive to beta-lactamase inhibitors [1][2][3][4]. These were (and still are) evolved as a reaction to the introduction of modern cephalosporins. The first ESBLs could be observed in the enzyme family TEM and SHV, caused by mutations of already known non-ESBL variants in the 1980s [3,5]. In the following decades, new enzyme families with ESBL phenotype members have occurred. Nowadays, the CTX-M family is dominant worldwide with CTX-M-15 being the most often detected ESBL enzyme. While TEM ESBL has mostly disappeared, SHV ESBLs (close to the CTX-M family) are still one of the most common ESBLs found [5]. However, there are now dozens of other beta lactamase families that consist entirely of ESBL members, most of which belong to Ambler class A (like GES, VEB, BEL). However, some enzymes of Ambler class D (OXA-2 and OXA-10 derivates) also belong to the ESBL group of enzymes [1,3,5].
The successful spread of ESBL enzymes is also due to the high gene exchange within the Enterobacteriaceae and related groups like the Pseudomonadales. Many genes coding for ESBLs are found on mobile genetic elements (such as plasmids) and can be expressed in  All isolates were resistant to ampicillin (AM including the natural resistance of most isolated species) as well as to cefuroxime (CXM) and cefotaxime (CTX). Five isolates of E. cloacae, four E. coli, one K. pneumoniae, and one P. mirabilis (in total 1.3% of all isolates) showed resistance to meropenem (MEM), which is therefore the most effective drug available in in vitro testing (Table 2). Highlighting the most important species E. coli and K. pneumoniae in comparison (Figure 1), the resistance patterns showed a high similarity with the exception of the ßlactamase inhibitors and combinations (AMC, TZP) as well as GM. Generally, E. coli showed a lower percentage of resistance. A total of 399 ESBL genes and 148 non-ESBL variants (bla TEM-1 , bla SHV-1 , bla SHV-11 ; Table S1) were detected in the 392 E. coli isolates. The most abundant ESBL genes for E. coli were members of the bla CTX-M-1 group (n = 288, 72.18%) ( Figure 2). From the 54 K. pneumoniae isolates, 59 ESBL genes, and additionally, 87 non-ESBL variants (bla TEM-1 , bla SHV-1 , bla SHV-11 , bla SHV-28 , and bla SHV-76 ; Table S1) were detected. Forty-eight (81.36%) of the 59 ESBL genes belonged to the bla CTX-M-1 group . Not surprisingly, bla CTX-M-15 (a member of the CTX-M-1 group of genes) was proportionately most common in both species with 61.15% (n = 244) of the ESBL positive E. coli and 72.88% (n = 43) of the ESBL positive K. pneumoniae. ESBL bla SHV gene variants were more abundant in the K. pneumoniae isolates (11.86%) than in the E. coli isolates (1.75%). For the bla CTX-M-9 group , the proportional abundance was higher in E. coli (26.07%) than in K. pneumoniae (6.78%). CTX-M-15 was also the most widespread enzyme among the other species. Bla VEB-6 was detected three times (all in P. mirabilis isolates) and only one bla CTX-M-2 also in one P. mirabilis isolate was detected, none of those were carrying another ESBL gene (Table S1). There were eight E. coli, four K. pneumoniae, and one E. cloacae, each carrying two different ESBL genes (Table 3). Table 3. ESBL coding genes detected and the number of isolates with those genes (n). Genes for non-ESBL enzyme TEM-1 were also carried by a large proportion of isolates (across all species), whereas wild-type SHV, apart from two E. coli isolates, was other-wise only found in K. pneumoniae (belongs to the chromosomal wild-type of this species) (Table S1).

Discussion
The presence of ESBL-producing strains was documented in Austria for the first time in the 1990s. After the turn of the millennium, many more studies followed that documented and analyzed the presence of ESBL-carrying bacteria in Austria. Most of them focused on isolates from various livestock and wild animals as well as isolates from the environment. ESBL-producing Enterobacteriaceae could be detected in a wide variety of animals, but also on food and different areas of the water cycle [16,[19][20][21][22][23][24][25].
However, there are few studies on the distribution of the various ESBL enzymes in the clinical setting. In detail, there are no studies dealing with Enterobacteriaceae species from these types of isolates. Epidemiological studies with large numbers of isolates are also available, but these are unfortunately limited to the phenotype [16,21,22,26].
The aim of the work was to capture the totality of the different ESBL genes at the present time for Austria, to document the occurrence of exotic variants, and thereby provide a basis for comparison for other (non-clinical) studies. Therefore, the limitations of the study are also predefined. The aim of this study was testing a high number of isolates for ESBL genes, strain backgrounds (via the species description) and genetic localization were not.
Due to the high occurrence in infections in general, the high proportion of E. coli and Klebsiella in the ESBL isolates is not surprising. The situation regarding the prevalence of ESBL producers, with about 10.3% of E. coli isolates and 10.5% for Klebsiella pneumonia isolates in 2017, is relatively moderate in Styria compared to other regions [1,[27][28][29] The low number of Enterobacter and Citrobacter in the isolates is somewhat critical. Here, the selection by intrinsic AmpC could have led to a false designation as AmpC hyperproducers, even if an ESBL was the cause of the resistance. Furthermore, it needs to be considered that the prevalence of the ESBL of Enterobacter is also low in other parts of Austria [22,30].  [31][32][33]. Exotic ESBL enzymes, on the other hand, were very rare in the current study. Although there have been reports of CTX-M-2 (group) or GES ESBL being found in Austria in the recent past, none of the isolates examined were found to harbor genes encoding GES enzymes, and only one P. mirabilis harbored a CTX-M-2.
Furthermore, three P. mirabilis isolates carried a gene for VEB-6. These isolates have already been described in a case report, and originate from only one patient. This work was primarily concerned with the carbapenemases. However, since ESBL genes were also present in the isolates, these isolates were also included in this study [19,34].
Looking at the co-resistance, as expected, ciprofloxacin is a non-beta lactam-antibiotic with the highest resistance rates. The co-occurrence of ESBL (especially CTX-M-15 mediated) and fluoroquinolone resistance is a well-known phenomenon [5,35,36]. Resistance to the last line antibiotic meropenem is below 2% in E. coli and Klebsiella spp. This approximately corresponds to what is found in neighboring countries [28,29,37].
In most cases, ESBL enzymes are sensitive to beta-lactamase inhibitors. However, a high percentage of the isolates examined are resistant to amoxicillin-clavulanic acid, and the proportion of isolates resistant to piperacillin/tazobactam is also in the mid-range for most species, with the high proportion for K. pneumonia being particularly worrying. However, this is most likely not caused by inhibitors of resistant varieties of ESBL; these isolates carry other mechanisms that are likely to give them this additional resistance [3,38,39].

Conclusions
The clinical ESBL picture in Austria, especially in the federal state of Styria, shows that it is primarily caused by E. coli and Klebsiella; mainly by the CTX-M-15 enzyme. With the help of the non-invasive isolates analyzed, a good picture can be presented that allows for a better classification of the many studies on ESBL from animals, from the environment, or from individual patients.

Collection of Isolates
Between 1 January 2017 and 31 December 2017, all presumptive ESBL Enterobacteriaceae first isolates (n = 1057) from non-invasive infections were collected at the Diagnostic and Research Center for Molecular BioMedicine of the Medical University of Graz and stored at −80 • C in cryotubes (Viabank™, MWE Medical Wire, Corsham, UK). Isolates from every other month (January, March, May, July, September, November), in total 593 isolates, were tested for their ESBL genotype [40]. Different ESBL harboring species from one patient were treated as separated first isolates. From those double (or multiple) isolates from one patient were excluded/included in the statistical analysis following this procedure: All isolates from the same patient belonged to the same species and showed the same antibiotic resistance profile were only listed once (independent if isolated from the same or different body sites). Different ESBL species from the same patient are listed as such. Isolates that did not show any genetic ESBL profile were also not included in the statistical analysis, eight isolates fell into this category. Species and resistance patterns of these isolates are given in Table S2. Following this procedure, a total of 468 different isolates (101 from hospitalized patients and 367 from not hospitalized patients) were finally included in the statistical analysis. These isolates stem from a total of 423 patients as 23 patients were listed twice (14 not hospitalized, five hospitalized, and four were listed once in both groups) three patients three times (all not hospitalized), and two patients (one not hospitalized and one hospitalized) had four infections with different ESBL isolates.

Strain Cultivation
The samples were stored at -80 • C in cryotubes (Viabank™, MWE Medical Wire) until analysis. For MALDI-TOF identification and DNA isolation, strains were recultivated on Luria-Bertani (LB) agar plates (10 g/L tryptone, 10 g/L NaCl, 5 g/L yeast extract, 15 g Agar; all Carl Roth, Karlsruhe, Germany). For each isolate, one beat from the corresponding cryotube was plated on one LB-plate and incubated at 37 • C for 16 h.

MALDI-TOF
The exact identification of the bacteria was carried out using MALDI-TOF microflexTM LT/SH (Bruker AXS GmbH, Billerica, MA, USA) at the Institute for Laboratory Diagnostics and Microbiology of the Klinikum Klagenfurt am Wörthersee, Austria. The matrix used was the IVD matrix HCCA-portioned from Bruker.

DNA Isolation
A distinct colony was stirred into 50 µL of nuclease-free water and heated for 10 min at 100 • C on the heating block. Subsequently, centrifugation was carried out at 21,200 g for 2 min. The supernatant was transferred to a fresh reaction tube and kept in the freezer at −20 • C until further analysis.
Briefly, for PCR, the Taq * 5x Master Mix (NEB, Ipswich, MA, USA) was used and the PCRs employing the different primer sets (Table 4) were performed on a GeneTouch Thermal Cycler (Biozym, Oldendorf, Germany) initializing with 94 • C for 5 min, 35 cycles at 95 • C for 30 s, 52 • C for 45 s, and 72 • C for 60 s, and closing with a final incubation for 10 min at 72 • C. Gel electrophoresis was performed to identify positive PCR results, and positive PCR samples where purified using the Monarch PCR and DNA Cleanup Kit (NEB, Ipswich, MA, USA). The purified samples were sent for Sanger Sequencing (Eurofins, Luxemburg, Luxemburg or Genewiz, Leipzig, Germany). For this purpose, the corresponding reverse primer and the sample were pipetted into a 96-well PCR plate or in reaction tubes in the ratio requested by the respective company, sealed, and shipped.
Identification of the sequencing results was obtained using the BLAST (Basic Local Alignment Search Tool) program "Standard Nucleotide Blast" of the NCBI (National Center for Biotechnology Information, Bethesda, MD, USA).

Data Analysis
All data were analyzed using MS Excel (Office 2019) and the figures were prepared using MS Excel and CorelDraw X5 (Corel Corporation, Ottawa, ON, Canada, 2010, version 15.0.0.486).
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/antibiotics12010001/s1, Table S1: Abundance of non-ESBL phenotype bla genes.; Table S2: Isolates with the ESBL phenotype but no detected ESBL genes. Data Availability Statement: The data can be obtained from the corresponding author upon reasonable request.