Molecular characterization of carbapenem resistant Gram-negative rods in Neonatal Intensive Care Unit of Mansoura University Children's Hospital

Carbapenems are group of extended-spectrum β-lactam antimicrobials frequently used for treating multidrug-resistant Gram-negative bacilli (GNB) infections. This study aimed at detecting and characterizing carbapenem resistance (CR) genes among GNB isolated from patients treated in neonatal intensive care unit (NICU) of Mansoura University Children's Hospital (MUCH), Egypt. It is a prospective study conducted from 2015 to 2016. A total of 158 GNB isolates were examined for CR both phenotypically and genotypically. Among 158 Gram negative isolates, there were 58 (36.7%) CR strains. Extended-spectrum β-lactamase (ESBL) production was confirmed in all 58 (100%) isolates. Carbapenemase production was detected in 52 (89.5%) strains while metallo beta-lactamase (MBL) production was found in 33 (56.9%) strains. Molecular characterization of CR strains revealed that 57 (98.3%) isolates were positive for carbapenemase encoding genes. KPC gene was the most frequent detected gene (34/58). VIM, IPM, OXA and NDM genes were also detected in 15, 13, 9 and 1 isolate, respectively. Only one isolate was negative for all encoding resistance genes despite positive for ESBL phenotype. Infection with CR strains has been increasing in clinical settings which limit the use of carbapenems. Programmal Thermal Controller, Peltier-Effect Cycling, MJ. The amplification conditions include: initial denaturation at 94°C for 5 min; 30 cycles of a final elongation at 72°C for 5 min. For multiplex PCR of the carbapenemase genes, the annealing temperature was 55°C for amplification of bla VIM , bla IMP and bla KPC genes, 45°C for bla NDM gene and 57°C for amplification of bla OXA-48 genes (Karuniawati et al. 2013). The amplicons were analyzed by electrophoresis in a 1.5% agarose gel.


INTRODUCTION
Treating neonatal infections is challenging due to spread of multidrug-resistant bacteria (Mzimela et al., 2021). Carbapenems are final antimicrobial therapy for life threatening microbes, nevertheless, the appearance of carbapenemases in Gram negative bacteria has put the clinicians in front of restricted treatment choices (Choudhury et al., 2018). Carbapenem resistance (CR) may be due to either carbapenemase production or other mechanisms, such as alteration of outer membrane permeability together with extended-spectrum βlactamase (ESBL) production, over expression of AmpC type β-lactamases or activation of efflux pumps. The production of carbapenemases is generally a more potent mechanism of CR compared with the other mechanisms (Nordmann et al., 2012;Woodford et al., 2004).
Ambler classification categorizes the β-lactamases into four classes (A to D). Most of carbapenemase-producing bacteria related to A class (bla KPC ), B class (bla VIM , IMP , and NDM ), and D class (bla OXA-48 ) Albiger et al., 2015). The Pseudomonas and Acinetobacter species releasing these enzymes have a wide geographical distribution and have been associated with hospital outbreaks (Woodford et al., 2004). Dissemination of carbapenemases is rapid and widespread in healthcare settings (Elbadawi et al., 2021).
Several methods for detecting carbapenemase production have been used. These methods include phenotypic and molecular methods (Al-Zahrani, 2018). This study aimed at molecular characterization of betalactamases associated with CR Gram negative neonatal infections.

Study design
This prospective study was conducted over 2 years during 2015 and 2016 in the NICU of MUCH, which is a level III unit with 25 incubators in 5 equal-sized rooms, that admits approximately 450 neonates per/year; there are no single rooms for isolation.

Sample collection and processing
A total of 158 GNB were detected from clinical samples (blood, urine, wound, tracheal aspirate, abscess, etc.) of 350 admitted neonates. Collected specimens were sent to Microbiology Diagnostics and Infection Control Unit laboratory in less than 2 h. If delay in transportation is expected, the specimens (except blood culture bottles) were kept at 4°C in the refrigerator. Clinical samples were processed using the standard laboratory techniques. Microscopic examination of Gram-stained films of the different samples was carried out to find any bacterial cells. Followed by culture on Nutrient agar, Blood agar, MacConkey agar and CLED agar (for urine samples) plates (Oxoid, UK) using the streak plate technique. Plates were kept in incubator at 37°C for 24 h.

Phenotypic detection of carbapenemase activity and ESBL production
The production of ESBL, carbapenemase and MBLs were tested using cephalosporin/clavulanic acid (BD Diagnostics, Franklin Lakes, NJ, USA) combination disc, The Modified Hodge test (MHT) and synergy combined disc test (CDT), respectively.
Phenotypic ESBL production was detected with the combination disc diffusion test with clavulanic acid. The inhibition zone surrounding the cephalosporin (Cefotaxime, Ceftazidime and Cefepime) discs combined with clavulanic acid is compared to the zone around the disc with the cephalosporin alone. The reaction was positive if the inhibition zone was 5 mm larger with clavulanic acid than without (Al Naiemi et al., 2012).

The Modified Hodge test
a 1:10 dilution of the Escherichia coli ATCC 25922 (NAMRU-3 Institute, Naval Medical Research Unit Three, Cairo, Egypt) was made by adding 0.5 ml of the 0.5 McFarland to 4.5 ml of saline. This dilution was streaked on the surface of MH agar plates using a swab, and allowed to dry for 3 to 5 min. Thereafter, 10 μg imipenem disc was placed in the center of the test plate. The test organism was streaked in a straight line from the disk to the edge of the plate. Plates were Kept at 35°C for 24 h. Plates were examined for a clover leaf like indentation of the test isolate and the reference strain of E. coli within the zone of inhibition of the imipenem disc (Tamma and Simner, 2018).

Detection of MBL (class B)
CDT was performed according to Joji et al. (2019). Two 10 µg imipenem discs and two 30 µg ceftazidime discs (Becton Dickinson) were put on a plate inoculated with the test bacteria. 10 µL of sterilized 0.5 M EDTA solution (dissolve 186.1 g disodium EDTA in 1000 ml distilled water at pH 8.0) was supplemented to one disc of each antibiotic. After that, inhibition zones of the imipenem and imipenem + EDTA and ceftazidime and ceftazidime + EDTA discs were compared after 18 to 24 h of incubation at 35°C. A zone diameter difference between the imipenem and imipenem + EDTA discs or the ceftazidime and ceftazidime + EDTA discs ≥7 mm was *Corresponding author. E. mail: nawalsalama@gmail.com. Tel: 00966502933179.
reative Commons Attribution C Author(s) agree that this article remain permanently open access under the terms of the License 4.0 International License

DNA template
Genomic DNA was extracted using Gene JET DNA Purification Kit (Thermo Scientific, For 50 preps, Lot. 00189391, The European Union (EU) Lithuania) following the manufacturer instructions.

Multiplex PCR method
The primers used in this study were based on primers published by Poirel et al. (2011) and are listed in Table 1. The amplification of DNA and thermal cycling conditions were done as described by Karuniawati et al. (2013). One multiplex PCR reaction was done detecting blaIMP and blaVIM-2 genes. Second multiplex PCR reaction was done detecting bla KPC and bla OXA-48 genes. Third reaction was done for detecting blaNDM-1 gene. DNA amplification was performed through a 50 μl reaction mix having 10 mM Tris-HCL, 50 mM KCL, 1.5 mM MgCl2, 0.1% Triton X-100, 200 mΜ of each of the deoxynucleoside triphosphate, 1U of DNA polymerase, 5 μl of template DNA, and 0.4 mΜ of each primer.
The thermal cycler program was adjusted by using PTC-100TM Programmal Thermal Controller, Peltier-Effect Cycling, MJ. The amplification conditions include: initial denaturation at 94°C for 5 min; 30 cycles of a final elongation at 72°C for 5 min. For multiplex PCR of the carbapenemase genes, the annealing temperature was 55°C for amplification of bla VIM, bla IMP and bla KPC genes, 45°C for bla NDM gene and 57°C for amplification of bla OXA-48 genes (Karuniawati et al. 2013). The amplicons were analyzed by electrophoresis in a 1.5% agarose gel.

Statistical analysis
Data were statistically analyzed using Microsoft Excel 2010 and Statistical Package of Social Science (SPSS) software version 22 (SPSS Inc., Chicago, IL, USA) and categorical variables were presented as counts and percentages.

Antibiotic resistance pattern among CR Gramnegative isolates
The overall resistance to carbapenems was high (75.9% for imipenem and 57% for meropenem). High rate of CR was detected among K. pneumoniae (86.4% for imipenem and 63.6% for meropenem) and E. coli strains (78.6% for imipenem and 71.4% for meropenem). The best susceptibility for these highly resistant strains was detected with pipracillin/tazobactem combination (K. pneumoniae resistance was 18.2% and E. coli resistance was 14.3%). Amoxicillin/Clavulinate had some activity against few strains (only one E. coli, one Enterobacter cloacae and one Enterobacter aerogenes strains) ( Table  3).

Phenotypic characterization of carbapenem resistance
ESBL production was confirmed in all 58 (100%) isolates by the cephalosporin/clavulanic acid combination disc. Detection of carbapenemase production by MHT ( Figure  1) was found in 52 (89.6%) isolates while MBL production was detected in 33 (56.9%) isolates by synergy CDT (Figure 2 and Table 4).

Genotypic characterization of carbapenem resistance
The molecular characterization of the 58 isolates showed that 57 (98.3%) were positive for carbapenemase encoding genes. bla KPC was the most prevalent gene detected in 34 isolates (58.6%), next to it was the bla VIM gene which was present in 15 (25.8%), bla IMP in 13 (22.4%), bla OXA-48 in 9 (15.5%) and bla NDM-1 in one isolate (1.7%). Five isolates were positive for both bla OXA and bla KPC and 4 isolates were tested positive for both bla IMP and bla VIM. One strain was negative for all the tested genes despite positive ESBL phenotype (Table 4).

DISCUSSION
The emergence and global spread of acquired CR isolates are designated a "global sentinel event" (Woodford et al., 2004). The overall rate of CR Gram negative strains in our NICU during the study period was high (36.7%). This finding agrees with another Egyptian study conducted by Makharita et al. (2020) who detected 36% isolated Enterobacteriacea were carbapenemase producers. Much lower CR isolation rates have been reported in Turkey 2.82% (Baran and Aksu, 2016) and the United States 1.4 to 4.2% (Pollett et al., 2014). Moreover, one Chinese study reported 0.9% CR isolates (28/3286) (Liao et al., 2014). The reason for higher rate of CR detected in our study could be related to the extensive use of carbapenems to treat life threatening infections in neonatal patients and absence of antibiotic stewardship program. In this study all CR strains were retrieved from the NICU in which many CR infection and colonization factors, including impaired immune status, prolonged hospital stay, and frequent use of antibiotics, were present. For this reason, proper hand hygiene and infection control measures must be emphasized. Monitoring of CR must be encouraged to reduce CR infection rate.
K. pneumoniae (37.9%) was the most common CR species isolated followed by E. coli (24.1%). This is in agreement with other CR studies in Egypt, United States, Europe and China (Metwally and Elnagar, 2019;Centers for Disease Control and Prevention, 2013;Akova et al., 2012;Yang et al., 2018) and is consistent with the CDC's recommendations that K. pneumoniae, E. coli, and Enterobacter spp. are the key health care-associated pathogens to focus on in the control of US CR epidemic (Centers for Disease Control and Prevention, 2014). That is why attention should be raised when any of these bacteria is isolated from clinical specimens.
In the present study, the overall resistance to carbapenems was high (75.9% for imipenem and 57% for meropenem). Higher rate of resistance was detected by Pollett et al. (2014); Mohamed et al. (2018), who found low rate of susceptibility to meropenem (4.3%, 3%) and to imipenem (1.7%, 3%). On the other hand, lower rate of resistance was detected by Okoche et al. (2015) who found that only 18.4% of study isolates were resistant to meropenem. The varied range in susceptibility/resistance rate of carbapenems among GNB in different studies may be due to different antibiotic usage patterns in different geographic regions.
It is difficult to compare the present results with those of others taking into consideration variations in study populations either pediatrics or adults, different carbapenem breakpoints, and different definitions of CR (Vading et al., 2011).
In the present study, all CR strains were ESBL producers and carbapenemase production was detected in 52 isolates (89.6%), while MBL production was detected in 33 (56.9%) isolates. Similarly, Fattouh et al. (2015) identified 88.14% of CR isolates as carbapenemase producers but lower MBLs activity was detected in 33.9% of the isolates by CDT. ESBL production was confirmed by Netikul and Kiratisin (2015) in 76.2% of the CR isolates and only 14.4% were positive for MHT. In accordance, Makharita et al. (2020) detected 33.2% MBLs positive isolates while only 33.6% of the CR isolates were positive MHT. On the contrary to the present results but also in Mansoura, carbapenamase activity was detected in 61.9% of CR K. pneumoniae isolated from ICUs of Mansoura University hospitals by MHT method (Moemen and Masallat, 2017).
Molecular characterization of the isolates revealed that 57 (98.3%) were positive for carbapenemase encoding genes. In accordance with the present results, Makharita et al. (2020) found 98.6% (74/75) of CR strains were carrying KPC gene, moreover 97.3% (73/75) were carrying GES gene. Also, similar to the present results, Yang et al. (2018) found that, among all CR strains bla KPC carrying rate was 60.3% and bla IMP were detected in 4.5% of total CR strains. Not so far from the present results. Pollett et al. (2014) stated that, a carbapenemase-encoding gene was found in 81.7% (94/115) of CR with bla KPC the most prevalent (78.3%).
These results were different with Elbadawi et al. (2021) who detected that the most prevalent gene was NDM-1 gene. On the other hand, Baran and Aksu (2016). reported that the OXA-48 gene was the most frequent gene as it was detected in 86/181 (47.5%) strains, NDM-1 gene in 6 (3.3%) strains, and VIM gene in 1 (0.6%) strain. IMP and KPC genes were not identified. Both OXA-48 and NDM-1 were produced by 3 strains and both OXA-48 and VIM were produced by one strain.
It is cleared that carbapenemase genes tend to be frequent in certain countries. It was found that bla KPC genes are dominant in USA, Greece and Egypt, while bla NDM genes are commonly detected in isolates recovered from India, Pakistan and Far East (Elbadawi et al., 2021).
A study in our locality conducted by Moemen and Masallat (2017) detected that 92.9% of the CR isolates were positive for one or more carbapenemase genes. Five of the 39 carbapenemase gene carrying isolates harbored two or more genes. The most prevalent gene was bla KPC 47.8% followed by bla VIM1 21.7%.
One strain that was negative for all the tested genes was detected despite positive ESBL phenotype. Similarly, Moemen and Masallat (2017) found that, none of the tested carbapenemase genes tested (bla KPC , bla NDM , bla VIM-1 , bla IMP , and bla OXA-48 -like) were detected in three isolates. Carbapenem resistance of these isolates is mostly due to a combination of ESBLs and changes in outer membrane proteins (ESBL/Omp) (Endimiani et al., 2010).
This study had several limitations. First, the study did not check tested isolates for sensitivity to last-line antimicrobials, such as colistin, tigecycline, and fosfomycin. Secondly, only isolates that showed imipenem or meropenem in the resistant were tested. Unlikely, carbapenemases have been found in carbapenem-sensitive Enterobacteriaceae, specially bla   (Nordmann et al., 2012). Finally, clinically significant isolates were mainly examined and this undervalues the colonized patients, which may forcefully propagate CR (Viau et al., 2012). For this reason, given that CR are increasing in this region, augmentation of local infection control measures beyond core elements to active surveillance may be useful, as stressed by the CDC (Centers for Disease Control and Prevention, 2014).