In Vitro Activity of Ceftolozane-Tazobactam against Escherichia coli, Klebsiella pneumoniae and Pseudomonas aeruginosa Obtained from Blood Cultures from Sentinel Public Hospitals in South Africa

Multidrug-resistant (MDR) Gram-negative bacteria are responsible for the majority of healthcare-associated infections and pose a serious threat as they complicate and prolong clinical care. A novel cephalosporin-β-lactamase-inhibitor combination, ceftolozane-tazobactam (C/T) was introduced in 2014, which improved the treatment of MDR pathogens. This study aimed to evaluate the activity of C/T against Escherichia coli (n = 100), Klebsiella pneumoniae (n = 100), and Pseudomonas aeruginosa (n = 100) blood culture isolates in South Africa (SA). Isolates were sequentially selected (2010 to 2020) from the Group for Enteric, Respiratory, and Meningeal Diseases Surveillance (GERMS) programme in SA. Organism identification was performed using the matrix-assisted laser desorption/ionisation-time of flight mass spectrometry (MALDI-TOF MS) instrument (Microflex, Bruker Daltonics, Bremen, Germany), and antibiotic susceptibility was performed using the Sensititre instrument (Trek Diagnostic Systems, East Grinstead, UK). C/T resistance was reported in 16 E. coli, 28 K. pneumoniae and 13 P. aeruginosa isolates. Fifty percent of the C/T resistant isolates were subjected to whole-genome sequencing (WGS). According to the whole genome multilocus sequence typing (MLST) analysis, the E. coli isolates (n = 8) belonged to sequence type (ST)10, ST131, ST405, and ST410, the K. pneumoniae isolates (n = 14) belonged to ST1, ST37, ST73, ST101, ST231, ST307, ST336 and ST6065 (novel ST), and the P. aeruginosa isolates (n = 7) belonged to ST111, ST233, ST273, and ST815. The WGS data also showed that all the E. coli isolates harboured aminoglycoside (aph (3′′)-Ib, aph (6)-Id), macrolide (mdfA, mphA), and sulphonamide (sul2) antibiotic resistance genes, all the K. pneumoniae isolates harboured β-lactam (blaCTX-M-15), and sulphonamide (sul2) antibiotic resistance genes, and all the P. aeruginosa isolates harboured aminoglycoside (aph (3′)-IIb), β-lactam (PAO), fosfomycin (fosA), phenicol (catB7), quinolone (crpP), and disinfectant (qacE) antibiotic resistance genes. It is evident that E. coli, K. pneumoniae and P. aeruginosa can adapt pre-existing resistance mechanisms to resist newer β-lactam molecules and inhibitors, since these isolates were not exposed to ceftolozane-tazobactam previously.


Introduction
Escherichia coli, Klebsiella pneumoniae and Pseudomonas aeruginosa are frequently isolated pathogens in the community and healthcare settings, and are the causes of infection-related morbidity and mortality [1,2]. Of concern is that these pathogens are becoming more resistant to available antibiotics [1].
A new cephalosporin-β-lactamase-inhibitor combination, ceftolozane/tazobactam (C/T) (sold under the brand name Zerbaxa, MERCK Connect, United States of America

E. coli Isolates
The average genome size of the sequenced E. coli isolates was 5035 kb with a coverage depth of 68 × to 133 ×. The average G + C content was 50.

K. pneumoniae Isolates
The average genome size of the sequenced K. pneumoniae isolates was 5611 kb with a genome coverage of 32× to 125×. The average G + C content was 57.

P. aeruginosa Isolates
The average genome size of the sequenced P. aeruginosa isolates was 6974 kb with a genome coverage of 57× to 84×. The average G + C content was 50.
All the sequenced E. coli, K. pneumoniae and P. aeruginosa isolates were resistant to cephalosporins (cefotaxime and ceftazidime) ( Table A2). All the E. coli and K. pneumoniae isolates were resistant to aztreonam except for most of the P. aeruginosa isolates, which were shown to be susceptible. All E. coli and P. aeruginosa isolates were resistant to one or more tested carbapenems. However, all the sequenced K. pneumoniae isolates were susceptible to all tested carbapenems.

Discussion
With the continuous rise of multidrug resistance in K. pneumoniae and P. aeruginosa, and to a lesser extent in E. coli there are only a few treatment options left for infected patients [7]. With the introduction of C/T in 2014, it has proven to be highly effective against MDR Gram-negative pathogens. The in vitro activity of C/T, against E. coli, K. pneumoniae, and P. aeruginosa isolates obtained from blood cultures from sentinel public hospitals in SA were investigated. The majority of the isolates were collected from adult patients in the Gauteng province; this province has the highest population in South Africa [10].
This study showed that the majority of the E. coli, and P. aeruginosa isolates are highly susceptible to C/T (82% to 85%). In comparison, K. pneumoniae isolates indicated decreased susceptibility towards C/T (66%). Findings from a previous study showed similar C/T susceptibility rates to E. coli, K. pneumoniae, and P. aeruginosa [7]. In contrast, higher C/T susceptibility rates in K. pneumoniae were reported in Germany and the USA [1,11]. The decreased susceptibility to C/T in the K. pneumoniae isolates reported in this study could potentially be due to increased antibiotic resistance [7]. In this study, we have detected multiple β-lactamase genes in the sequenced K. pneumoniae isolates (1 to 14 β-lactamase genes per isolate).
In this study, multidrug resistance was detected in 30% of the E. coli isolates, 61% of the K. pneumoniae isolates, and 24% of the P. aeruginosa isolates ( Table 2). C/T was not effective against the majority of MDR isolates detected in this study. In contrast, other studies reported that treatment with C/T generally led to favorable clinical outcomes among patients with MDR, extensively drug-resistant (XDR) or pan drug-resistant (PDR) bloodstream infections associated with E. coli, K. pneumoniae, and P. aeruginosa [12][13][14][15].
WGS is a powerful tool that can be used to characterise the genetic diversity of bacterial populations and can also be used for the prediction of bacterial antibiotic resistance profiles [16]. However, the prediction of susceptibility or resistance to antibiotics based only on the presence or absence of previously known genes is still under investigation and discordances are reported in the literature [16]. Half of the C/T resistant isolates were subjected to sequencing for genomic characterisation. AmpC (PDC) and bla GES hyperproduction are factors linked to C/T resistance [3,4,[17][18][19][20]. However, the bla GES and AmpC genes were not detected in our isolate collection that was subjected to WGS. Other antibiotic resistant genes reported in C/T resistant isolates are the bla CTX-M , and bla SHV genes [18]. The bla CTX-M-15 gene was detected in the majority of the E. coli isolates, and in all the K. pneumoniae isolates. The bla SHV genes were only detected in K. pneumoniae isolates. The precise resistance mechanisms leading to C/T resistance could not be determined. Further studies are required to assess the expression and functionality of the detected genes in the studied isolate population to predict the phenotype consequences of the C/T resistant genotype.
Limitations of the study: (i) C/T was only evaluated for blood cultures, (ii) isolates and data originated from sentinel public hospitals in SA, (iii) small sample size, and (iv) not all the C/T resistant isolates could be sequenced due to limited funding.

Study Setting
A total of 100 E. coli, 100 K. pneumoniae, and 100 P. aeruginosa blood culture isolates were chronologically selected from storage (−70 • C). A three-week exclusion period was applied to avoid duplicate isolates of the same organism from the same patient.
The isolates were initially collected for the Group for Enteric, Respiratory, and Meningeal Diseases Surveillance (GERMS) program in SA (2010 to 2020), which receives clinical isolates from sentinel sites in the Free State, Gauteng, KwaZulu-Natal, Eastern-and Western Cape provinces. Demographic and clinical information of patients was collected by surveillance officers through medical record reviews and/or patient interviews using standard case report forms (CRFs). All GERMS-SA isolates were initially processed and stored as followed: the bacterial cultures were grown on blood or chocolate agar plates (Diagnostic Media Products (DMP), National Health Laboratory Service (NHLS), Johannesburg, SA) for no more than 18-27 h. The agar plates were carefully inspected for any contaminating bacterial or fungal colonies. Using a sterile disposable Pasteur pipette, 1 mL of TSB + 10% glycerol (DMP, NHLS, Johannesburg, SA) was dispensed into cryovials. Using a sterile swab or loop, a heavy sweep of growth was taken and emulsified in the cryovial containing the TSB + 10% glycerol. The cryovials were tightly sealed and placed in allocated cryoboxes and immediately stored at −70 • C.

Phenotypic Characterisation
All organisms were processed by the Centre for Healthcare-associated infections, Antimicrobial Resistance and Mycoses (CHARM), National Institute for Communicable Diseases (NICD), a division of the NHLS, Johannesburg, SA. The selected GERMS-SA isolates were retrieved from −70 • C storage. A loop full of the isolate + TSB + 10% glycerol mixture was streaked and grown on blood agar plates (DMP, NHLS, Johannesburg, SA) for no more than 18 h to 24 h. The agar plates were carefully inspected for any contaminating bacterial or fungal colonies. Organism identification was confirmed using matrix-assisted laser desorption/ionisation-time of flight mass spectrometry (MALDI-TOF MS) (Microflex, Bruker Daltonics, Bremen, Germany). All MALDI-TOF MS score values were between 2.00 to 3.00 (high confidence identification) and fell in the consistency category A (high consistency). Antimicrobial susceptibility testing (AST) was performed using the Sensititre instrument (Trek Diagnostic Systems, East Grinstead, UK) with the commercially available Gram-negative DKMGN panel (Separation Scientific, Johannesburg, SA) which included amikacin, amoxicillin/clavulanic acid, aztreonam, cefotaxime, ceftazidime, C/A, C/T, ciprofloxacin, colistin, ertapenem, gentamicin, imipenem, meropenem, tigecycline, tobramycin, and trimethoprim/sulfamethoxazole. The AST results were interpreted using the 2021 Clinical Laboratory Standards Institute (CLSI) guidelines (no interpretation for tigecycline was provided) [21]. MDR isolates were defined as non-susceptibility to one or more antibiotic agents in three or more antimicrobial classes [22]. XDR isolates were defined as non-susceptibility to at least one agent in all but two or fewer antimicrobial categories (i.e., bacterial isolates remain susceptible to only one or two categories), and PDR isolates were defined as non-susceptible to all agents in all antimicrobial categories [22].

Statistical Analysis
Microsoft Excel (version 2016) was used for data entry and basic statistical analysis (medians, interquartile ranges and percentiles). Additional data analyses were carried out using STATA statistical software package (version 14; StataCorp LP, College Station, TX, USA).

Conclusions
Antibiotic resistance is a global health threat that limits the optimal care of patients infected with healthcare-associated pathogens. The increasing antibiotic resistance should be closely monitored. Since these isolates were not exposed to C/T previously, it is evident that E. coli, K. pneumoniae, and P. aeruginosa can adapt pre-existing resistance mechanisms, such as the production of catalytic enzymes (i.e., ESBLs and carbapenemases), altered PBPs, reduction in outer membrane channels resulting in decreased porin activity influx/expression, and an increase in efflux pumps to resist β-lactam molecules and inhibitors. In this study, C/T was not effective against the majority of MDR isolates. Informed Consent Statement: Patient consent was waivered since patient care was not influenced at any stage.

Data Availability Statement:
The datasets presented in this study can be found in the article or in the Appendix A.  Antibiotic resistance genes Antibiotic resistance genes blaSHV-