Phenotypic and Genotypic Features of Klebsiella pneumoniae Harboring Carbapenemases in Egypt: OXA-48-Like Carbapenemases as an Investigated Model

This study aimed at the characterization of carbapenem-resistant Klebsiella pneumoniae isolates focusing on typing of the blaOXA-48-like genes. Additionally, the correlation between the resistance pattern and biofilm formation capacity of the carbapenem-resistant K. pneumoniae isolates was studied. The collected isolates were assessed for their antimicrobial resistance and carbapenemases production by a modified Hodge test and inhibitor-based tests. The carbapenemases encoding genes (blaKPC, blaNDM, blaVIM, blaIMP, and blaOXA-48-like) were detected by PCR. Isolates harboring blaOXA-48-like genes were genotyped by Enterobacterial Repetitive Intergenic Consensus-Polymerase Chain Reaction (ERIC-PCR) and plasmid profile analysis. The discriminatory power of the three typing methods (antibiogram, ERIC-PCR, and plasmid profile analysis) was compared by calculation of Simpson’s Diversity Index (SDI). The transferability of blaOXA-48 gene was tested by chemical transformation. The biofilm formation capacity and the prevalence of the genes encoding the fimbrial adhesins (fimH-1 and mrkD) were investigated. The isolates showed remarkable resistance to β-lactams and non-β-lactams antimicrobials. The coexistence of the investigated carbapenemases encoding genes was prevalent except for only 15 isolates. The plasmid profile analysis had the highest discriminatory power (SDI = 0.98) in comparison with ERIC-PCR (SDI = 0.89) and antibiogram (SDI = 0.78). The transferability of blaOXA-48 gene was unsuccessful. All isolates were biofilm formers with the absence of a significant correlation between the biofilm formation capacity and resistance profile. The genes fimH-1 and mrkD were prevalent among the isolates. The prevalence of carbapenemases encoding genes, especially blaOXA-48-like genes in Egyptian healthcare settings, is worrisome and necessitates further strict dissemination control measures.

extracellular polymeric substance (EPS) adhere to each other and/or to a surface" [16]. Biofilm formation participates in the aggravation of serious clinical manifestations and leads to reduced susceptibility to antimicrobial agents [17]. Type 1 and 3 fimbriae have a crucial role in adhesion, which is a key factor in K. pneumoniae biofilm formation besides capsule and lipopolysaccharide [18].
In this study, we aimed to characterize carbapenem-resistant K. pneumoniae isolated from patients admitted to an Egyptian tertiary hospital. Additionally, we focused on the typing of the detected bla OXA-48-like genes. Additionally, we shed light on the biofilm formation capacity of the carbapenem-resistant isolates and studied the possible correlation between the two traits

Distribution and Identification of K. pneumoniae Isolates
A total of 117 isolates of carbapenem-non-susceptible K. pneumoniae were collected and recovered from diverse clinical specimens as shown in Figure 1. The confirmation of isolate identification using MALDI-TOF/MS had a high confidence ranging from 99.7 to 99.9%.
Antibiotics 2020, 9, x FOR PEER REVIEW 3 of 16 extracellular polymeric substance (EPS) adhere to each other and/or to a surface" [16]. Biofilm formation participates in the aggravation of serious clinical manifestations and leads to reduced susceptibility to antimicrobial agents [17]. Type 1 and 3 fimbriae have a crucial role in adhesion, which is a key factor in K. pneumoniae biofilm formation besides capsule and lipopolysaccharide [18].
In this study, we aimed to characterize carbapenem-resistant K. pneumoniae isolated from patients admitted to an Egyptian tertiary hospital. Additionally, we focused on the typing of the detected blaOXA-48-like genes. Additionally, we shed light on the biofilm formation capacity of the carbapenem-resistant isolates and studied the possible correlation between the two traits

Distribution and Identification of K. pneumoniae Isolates
A total of 117 isolates of carbapenem-non-susceptible K. pneumoniae were collected and recovered from diverse clinical specimens as shown in Figure 1. The confirmation of isolate identification using MALDI-TOF/MS had a high confidence ranging from 99.7 to 99.9%.

Phenotypic and Genotypic Characterization of Carbapenem-Resistant Isolates
Only 66 isolates (56.41%) were positive for non-specific detection of carbapenemases production by the modified Hodge test (MHT). In contrast, the inhibitor-based tests were more specific by confirmation of carbapenemases production and differentiation between their types. Seventy-nine isolates (67.52%) were positive for MβLs production, and no isolates showed positive results for KPC production or AmpC activity. The remaining isolates (n = 38) did not comply with the criteria of the previously mentioned enzymes but showed resistance to temocillin, giving a presumptive estimation of the presence of the bla OXA-48-like genes. The results of both phenotypic tests in comparison with the further molecular screening of carbapenemases genes are shown in Table S2.
The results of molecular screening of the investigated carbapenemases are shown in Table S2. VIM gene was the most prevalent gene amongst the studied genes as it was detected in 99 isolates (84.62%) followed by 88 isolates (75.21%) for NDM gene, 69 isolates (58.97%) for IMP gene, and 34 isolates (29.06%) for OXA-48-like genes. The presence of KPC gene was very limited as it was detected in only 6 isolates (5.13%).
Furthermore, all isolates showed coexistence of at least two carbapenemases under investigation except 4 isolates, namely, 33, 136, 147, and 396, which had VIM gene only; 2 isolates, namely, 16 and 417 had NDM gene only; and 9 isolates, namely, 8,28,113,127,151,346,356,378, and 399 had OXA-48-like genes only. On the other hand, only 4 isolates, namely, 137, 142, 410, and 419 were carbapenem-resistant and showed positive MHT but did not show augmentation in the inhibitor-based tests or any positive result via PCR, and this gives an indication that carbapenem resistance in those isolates may be for mechanisms other than carbapenemases production.

Genotyping of Isolates Harboring bla OXA-48-like Genes
Genotyping of the 34 isolates harboring bla OXA-48-like genes was performed by two molecular methods: Enterobacterial Repetitive Intergenic Consensus-Polymerase Chain Reaction (ERIC-PCR) and plasmid profile analysis. Figure 4 shows the ERIC-PCR dendrogram that divided the 34 isolates into 16 genotypes. Among them, there were three major groups: each represented genetic relatedness between its isolates despite their collection from different clinics at different periods. Additionally, 10 isolates showed unique patterns and were not related to each other. On the other hand, the plasmid profile analysis showed variation between isolates. Under experimental conditions, out of 34 isolates, 2 isolates (5.88%) harbored one plasmid, and other isolates harbored many plasmids ranging from 2 to 8 plasmids. The highest number of plasmids was 8 plasmids which were found in one isolate followed by 6 plasmids which were found in another one, representing 2.94% for each. Additionally, there were 11 isolates (32.35%) that had 3 plasmids, followed by 9 isolates (26.47%) that had 5 plasmids, 6 isolates (17.65%) that had 4 plasmids, and 4 isolates (11.76%) that had 2 plasmids. The results showed that the isolates belonging to the same cluster were mostly collected from different sources and at different times.
The Simpson's Diversity Index was calculated for antibiogram, ERIC-PCR, and plasmid profile analysis giving convergent values of 0.78, 0.89, and 0.98, respectively as shown in Figure 5. The SDI value of the plasmid profile analysis was very close to 1 indicating the maximum diversity that was achieved by this method. The criteria for SDI calculation of the applied typing methods-antibiogram, ERIC-PCR, and plasmid profile analysis-are shown in Tables S3-S5, respectively. A comparative summary of the SDI calculations of the three methods is shown in Table S6.

Transferability of blaOXA-48 Gene
The plasmids of five selected isolates, namely, 311, 383, 399, 405, and 409, underwent genetic transfer by chemical transformation. PCR results of transformants showed the absence of blaOXA-48 gene. Further investigations were performed for two selected transformants, namely, 383 and 409, by an antimicrobial susceptibility test and PCR. The results showed reduced susceptibility to different β-lactams including carbapenems, sulfamethoxazole/trimethoprim, and tigecycline in comparison with Escherichia coli DH5α as shown in Table 1. Although the isolate 383 had blaNDM,

Transferability of blaOXA-48 Gene
The plasmids of five selected isolates, namely, 311, 383, 399, 405, and 409, underwent genetic transfer by chemical transformation. PCR results of transformants showed the absence of blaOXA-48 gene. Further investigations were performed for two selected transformants, namely, 383 and 409, by an antimicrobial susceptibility test and PCR. The results showed reduced susceptibility to different β-lactams including carbapenems, sulfamethoxazole/trimethoprim, and tigecycline in comparison with Escherichia coli DH5α as shown in Table 1. Although the isolate 383 had blaNDM, Figure 5. A comparative graph of calculated Simpson's Diversity Index (SDI) of the applied typing methods to study the relatedness among the 34 K. pneumoniae isolates harboring bla OXA-48-like genes. SDI values were convergent as they were 0.78, 0.89, and 0.98 for antibiogram, ERIC-PCR, and plasmid profile analysis, respectively.

Transferability of bla OXA-48 Gene
The plasmids of five selected isolates, namely, 311, 383, 399, 405, and 409, underwent genetic transfer by chemical transformation. PCR results of transformants showed the absence of bla OXA-48 gene. Further investigations were performed for two selected transformants, namely, 383 and 409, by an antimicrobial susceptibility test and PCR. The results showed reduced susceptibility to different β-lactams including carbapenems, sulfamethoxazole/trimethoprim, and tigecycline in comparison with Escherichia coli DH5α as shown in Table 1. Although the isolate 383 had bla NDM , bla VIM , and bla IMP encoding genes and the isolate 409 had bla NDM and bla VIM encoding genes, only bla VIM encoding gene was detected by colony PCR for both isolates transformants. This explains the successful transfer of such gene. The quantitative biofilm formation assay resulted in OD 600 values ranging from 0.2 to 1.1. All isolates were classified as biofilm formers but with different capacities. The majority of the isolates were weak biofilm formers and were represented by 82 isolates (70.09%), followed by 34 isolates as moderate formers (29.06%), and only one isolate was classified as a strong biofilm former (0.85%). Furthermore, the PCR screening of the isolates showed the prevalence of the two adhesion encoding genes in most isolates. Out of 117 isolates, fimH-1 gene was found in 108 isolates (92.31%) and mrkD gene was found in 115 isolates (98.29%). Additionally, 106 isolates (90.60%) represented the coexistence of both genes as shown in Table S7.

The Correlation between Non-Susceptibility Pattern and Biofilm Formation Category of Carbapenem-Resistant K. pneumoniae Isolates
As shown in Table 2, there is no significant correlation between non-susceptibility patterns and biofilm formation categories among carbapenem-resistant isolates. All p-values were > 0.05, and this was considered statistically insignificant.

Discussion
The dissemination of carbapenemases encoding genes in Egypt, especially in healthcare settings, is of great concern in several mapping studies [20][21][22]. The prevalence of such genes in Egypt could be attributed to considerable reasons that are well-identified globally. For example, hospital environments are considered as a pool for the dissemination of such genes through the horizontal gene transfer that enables their spread via mobile genetic elements such as plasmids [23,24]. Additionally, plasmids may carry more than one resistance gene, and this leads to resistance to diverse antimicrobial classes and as a result limited therapeutic options [25].
Although the transferability of the OXA-48 gene of our selected isolates failed and this was compliant with Skalova et al. trials [26], the reduced transformants susceptibility to most antimicrobials confirms the genetic transfer risk. Additionally, over-the-counter prescription and the misuse of antimicrobials participate in the exacerbation of the antimicrobial resistance problem, especially in Egypt [27,28].
In parallel, the investigated isolates in our study showed increased resistance to various antimicrobial agents and also high MIC values for imipenem and meropenem. This may be markedly attributed to various resistance mechanisms that led to resistance to most antimicrobial classes. Additionally, only 15 isolates (12.82%) of the total investigated isolates were of sole carbapenemase, indicating remarkable coexistence. This finding is most consistent with studies from Egyptian hospitals and unlike the study by Argente et al. in which bla OXA-48 gene was the only detected carbapenemase among the tested isolates [29]. This reflects the diverse regional distribution of carbapenemases.
In the current study, the detection and typing of carbapenemases depended on the merge between the phenotypic and molecular methods, and this is in consistency with the study by Karampatakis et al. [30]. The application of the molecular methods in the detection of resistance genes and further typing of isolates is considered as a gold standard tool in the field of epidemiology, and this appears on two occasions in this study; Firstly, the disagreement between phenotypic and genotypic methods for the detection of bla KPC encoding gene which was detected in only 6 isolates by PCR. In this case, the genotypic methods act as an alarming tool for the silent dissemination of such gene which is absent phenotypically but likely to be successfully expressed under certain conditions [31]. Secondly, there is no phenotypic test for the direct detection of bla OXA-48-like genes. Resistance to temocillin is considered as a stepwise prediction of them, since other resistance mechanisms may confer temocillin resistance [32].
In our work, we applied ERIC-PCR and plasmid profile analysis, besides the previously determined antibiogram for the characterization of isolates harboring bla OXA-48-like genes. The discriminatory power of the applied typing methods (antibiogram, ERIC-PCR, and plasmid profile analysis) was approximated according to the SDI calculations, while the highest diversity was represented by plasmid profile analysis (SDI = 0.98). The three typing methods are different in their concepts, and this leads to variation in the clonality of the isolates among the three methods.
The interplay between the antimicrobial resistance and biofilm formation was studied from different sides. On the one hand, resistance genes encoded on some plasmids regulate the expression of fimbrial genes, which are among the essentials of biofilm formation. On the other hand, the plasmid horizontal gene transfer is enhanced in biofilms more than in planktonic cells [33]. In our work, all investigated isolates were carbapenem-resistant and showed biofilm formation ability which was various and independent on the resistance profile of those isolates. It is worth mentioning that despite the absence of such correlation, having 100% of isolates as biofilm formers with the high resistance pattern is considered a troublesome feature of the studied isolates.
Our results were in concordance with the study by El Fertas-Aissani et al. on the prevalence of fimH-1 and mrkD adhesion encoding genes, giving an indication of their conservation in K. pneumoniae pathogen [34]. Furthermore, the prevalence of those two adhesion genes and the weak biofilm patterns among the isolates may suggest that the strong biofilm formation may need factors other than fimbrial adhesion genes.
Three isolates in our study, namely, 349, 354, and 375, represented unique findings, as they were resistant to all applied antimicrobials except tigecycline. Additionally, they showed positive results for the MHT, all carbapenemases (bla KPC , bla NDM , bla VIM , bla IMP , and bla OXA-48-like ) encoding genes, and for fimH-1 and mrkD encoding genes. Although our applied typing methods are relatively simple and of low cost, the results of the three isolates encourage the adoption of other typing methods such as whole-genome sequencing (WGS) [35]. Whole-genome sequencing provides massive data for the determination of the sequence type, plasmid replicon, antimicrobial resistance determinants, and virulence genes. Additionally, K. pneumoniae was of special concern for the newly emerged bioinformatics interfaces such as Kaptive Web [36] and KlebNet [37]. Additionally, we are moving globally toward the extended drug resistance with some reported cases of alarmingly pan-drug-resistant isolates [38], research is directed toward finding alternatives for combating antibiotic resistance. The alternative strategies may include antibiotic combinations [39], phage therapy [40], vaccine development [41], natural products [42], revival of classical antibiotics [43], or repurposing non-antimicrobials for targeting pathogens [44].
Finally, the applied infection control measures and awareness campaigns may not be enough to combat the problem of antimicrobial resistance, especially in a developing country such as Egypt. Among the important strategies that are worth emphasizing is launching collaborative platforms between Egyptian healthcare institutes and other international ones for networking and gaining accurate knowledge about the possible patterns of antimicrobial resistance that could be easily disseminated globally [45,46].

Isolates Collection and Confirmation
A total of 117 archival non-duplicate and non-consecutive isolates of carbapenem-non-susceptible (resistant or intermediate) K. pneumoniae were recovered from clinical specimens. The specimens were collected and processed by dedicated healthcare workers. Furthermore, the specimens were collected from patients admitted to different departments of Kasr Al-Ainy Hospital in Cairo, Egypt, from September 2014 to December 2016. The study was approved by the Ethics Committee of the Faculty of Pharmacy, Al-Azhar University (Girls' branch). The ethical approval code is SMR/2015.
The collected isolates were obtained from the diagnostic routine work of the Clinical Pathology Department of Kasr Al-Ainy Hospital without direct contact with the patient or their related personal data. All laboratory techniques and procedures of isolation, identification, and storage of the included isolates in the current study were performed according to the standard microbiological techniques.
The isolates were previously identified using conventional microbiological methods which included Gram staining, cultural characteristics on MacCkonky's agar, and biochemical testing such as growth on Triple Sugar Iron agar (TSI) and Indole, Methyl red, Voges-Proskauer, and Citrate utilization (IMViC) test. The identification of K. pneumoniae isolates was confirmed using Matrix-Assisted Laser Desorption/Ionization Time-Of-Flight Mass Spectrometry (MALDI-TOF/MS) (Vitek MS; BioMérieux, Inc., Marcy-l'Etoile, France).

Phenotypic Confirmation of Carbapenemases Production
Non-specific screening of carbapenemases production was performed by the modified Hodge test (MHT) according to guidelines of CLSI [47]. The formation of cloverleaf-like indentation of grown E. coli ATCC ® 25922 along the tested isolate is an indication of the positive result. K. pneumoniae ATCC ® BAA-1705 was used as a positive control, and K. pneumoniae ATCC ® BAA-1706 was used as a negative control. Moreover, the inhibitor-based tests were performed as another phenotypic approach for further differentiation between carbapenemases producing isolates. The inhibitor-based tests were based on the inactivation of carbapenemases by combining some substrates independently with meropenem disc. For example, EDTA as metallo-β-lactamases (MβLs) inhibitor, boronic acid as KPC inhibitor, and cloxacillin as AmpC inhibitor [48]. The inhibitor-based tests were performed using MASTDISCS TM ID Carbapenemase Detection Disc Set D70C (MAST group, Merseyside, UK). The set consists of four different discs as shown in Table 3. The interpretation of the tests depends on the comparison of the inhibition zone of the meropenem disc with the inhibition zone of the disc containing meropenem combined with the carbapenemases inhibitor. An additional Temocillin disc (30 µg) was applied for obtaining a preliminary indication of the bla OXA-48-like production until performing further genotypic confirmatory tests.

Molecular Detection of Carbapenemases Encoding Genes
Total DNA of the isolates was extracted by the boiling method according to Vaneechoutte et al. [49]. Uniplex PCR reactions were performed for detection of bla KPC , bla NDM , bla VIM , bla IMP , and bla oxa-48-like encoding genes with primers listed in Table S9 [50][51][52]. The primer specificity was checked against sequences retrieved from GenBank sequence databases using the National Center for Biotechnology Information/Basic Local Alignment Search Tool (NCBI/BLAST), allowing the detection of several variants of each carbapenemase. The primers were synthesized by Eurofins MWG Operon (Ebersberg, Germany). The applied profile was according to the criteria that were mentioned in primer references except for adjustment of annealing temperature to be 56 • C. The positive controls were obtained from the Clinical and Chemical Pathology Department laboratory collection (Kasr Al-Ainy hospital, Cairo, Egypt) and represented by positive strains that harbored investigated carbapenemases encoding genes and had been previously sequenced. The negative control was a template-free master mix.
PCR products were submitted for purification and sequencing in both directions via (Solgent Co. Ltd., Daejeon, Korea). The resulted DNA sequences were trimmed using Geneious Software v.8.1.6 (www.geneious.com) and compared to the Genbank sequence database using the BlASTn tool at https://blast.ncbi.nlm.nih.gov/Blast.cgi.

Genotyping of Isolates Harboring bla OXA-48-like Genes
Thirty-four isolates were confirmed either by inhibitor-based tests and/or PCR as harboring bla OXA-48-like genes. These isolates were genotyped by two methods: the first one is the Enterobacterial Repetitive Intergenic Consensus-Polymerase Chain Reaction (ERIC-PCR) which aimed to determine the clonal relatedness between the investigated isolates by detection of specific conserved sequences among them [54]. The second typing method is the plasmid profile analysis which depends on the investigation of the isolate plasmids and characterization of their profile [55].

Enterobacterial Repetitive Intergenic Consensus (ERIC-PCR)
The genomic DNA was extracted using a GeneJet Genomic DNA Purification Kit (K0721, Thermo Scientific, Rockford, IL, USA) according to the manufacturer's protocol. The investigated isolates were genotyped by ERIC-PCR using ERIC-2 primer (5' AAGTAAGTGACTGGGGTGAGCG 3') [56]. The PCR profile was according to Abdulall et al. [57] with slight modifications as follows: initial denaturation at 95 • C for 5 min, followed by 35 cycles of denaturation at 95 • C for 1 min, annealing at 45 • C for 1 min, and extension at 72 • C for 8 min, with a final extension at 72 • C for 10 min. Aliquots of PCR amplicons were electrophorized alongside a 10-kb Ladder. The interpretation of the results depended on binary data by which the positive and negative amplifications were assigned as 1/0, respectively. Cluster analysis was performed using NTSYS 2.01 software and the schematic dendrogram was built using the Unweighted Pair-Group Method using Arithmetic Mean (UPGMA).

Plasmid Profile Analysis
The isolate plasmids were extracted by a QIAprep ® Spin Miniprep Kit according to the manufacturer's protocol. Extracted plasmids were electrophorized, and the generated patterns were judged visually and considered of distinct type if they showed a single band of different size, according to the ESGEM recommendations [15].

Calculation of the Discriminatory Power of Different Applied Typing Methods
The discriminatory power of the applied typing methods (antibiogram, ERIC-PCR, and plasmid profile analysis) among the 34 isolates was assessed using Simpson's Diversity Index (SDI). The concept of discrimination based on if there are two unrelated isolates, which were sampled from a test population, they will be probably found in different typing groups [58]. The SDI calculation depended on the following equation D = 1 − n(n−1) N(N−1) , where D is the diversity index, n is the number of individuals of each species, and N is the total number of individuals of all species. The SDI ranges from 0, which indicates the identical types of the tested isolates, to 1, which indicates the maximum diversity among them.

Genetic Transfer of bla OXA-48 Gene by Transformation
The acquisition of resistance determinants harbored by the isolate plasmids was assessed by transformation. The preparation and transformation of competent E. coli DH5α cells were performed using the CaCl 2 method according to Cohen et al. [59]. The genetic transfer of bla OXA-48 gene was assessed for the plasmids of five selected isolates, namely, 311, 383, 399, 405, and 409. The transformants were selected on LB agar containing ampicillin (50 mg/L) and also on LB agar containing meropenem (0.5 mg/L). The plasmids of transformants were extracted and underwent PCR for the detection of the bla OXA-48 gene.

Quantitative Biofilm Formation Assay for the Carbapenem-Resistant Isolates by Microtitre Plate Method
The quantitative assessment of biofilm formation for the 117 carbapenem-resistant isolates was performed by the crystal violet staining assay according to O'Toole protocol with slight modifications [60]. Briefly, overnight culture of each isolate was grown in Luria-Bertani broth, and then was diluted 1:100 into fresh LB medium. One hundred microliters of the isolate dilution was added per well in a flat-bottomed 96 well microtitre plate. After overnight incubation at 37 • C, the medium was discarded, and the wells were washed twice by adding 100 µL of distilled water in each well for further removal of unattached cells and media. The plate was allowed to dry, stained by adding 125 µL of 0.1% crystal violet to each well, and incubated at room temperature for 15 min. The crystal violet was discarded and the plate was washed four times for removing excess stain and allowed to dry. The stained biofilms were solubilized by adding 125 µL of 30% acetic acid to each well and incubated for 15 min at room temperature. The solubilized stain of each well was transferred to a new microtiter plate. The optical density of each well was measured at 600 nm (OD 600 ) using a microplate reader (Stat Fax ® 2100, Awareness Technology Inc., Palm City, FL, USA). The experiments were carried out three times in triplicates; the readings of each isolate in each plate were averaged and compared for further interpretation. The positive control was K. pneumoniae ATCC ® 700603 and the negative control was un-inoculated LB media. The interpretation of the results was according to the recommendations of Stepanović et al. recommendations [61].

Molecular Detection of Type 1 (fimH-1) and Type 3 (mrkD) Adhesion Encoding Genes
The carbapenem-resistant isolates were screened for the presence of fimH-1 and mrkD encoding genes by uniplex PCR using primers listed in Table S9 [34,62]. The applied profile was as follows: initial denaturation at 94 • C for 4 min followed by 30 cycles of denaturation at 94 • C for 30 s, annealing at 55 • C for 40 s, extension at 72 • C for 1 min, and a final extension at 72 • C for 10 min. The positive controls were obtained from the MTI University Microbiology Laboratory archive, Egypt, and represented by positive strains that harbored investigated adhesion genes and had been previously sequenced. The negative control was a template-free master mix.

Statistical Analysis
Statistical analysis was performed using GraphPad Prism software (San Diego, CA, USA). The description of qualitative data was as a frequency percentage. The chi-square or Fisher's exact test was used for comparing categorical variables. The results were considered statistically significant when p-value < 0.05. Spearman's test was applied for the determination of the correlation between the biofilm formation category and antimicrobial resistance profile of all isolates.

Conclusions
In our work, we studied the prevalence of the "big five" carbapenemases encoding genes, (bla KPC , bla NDM , bla VIM , bla IMP , and bla OXA-48-like ) among carbapenem-resistant K. pneumoniae isolated from a tertiary hospital in Egypt. The tested carbapenem-resistant K. pneumoniae isolates showed a remarkable coexistence of different investigated carbapenemases. Additionally, we shed light on bla OXA-48-like genes which represented about one-third of our isolates. The assessment of the simple epidemiological typing methods-antibiogram, ERIC-PCR, and plasmid profile analysis-showed their convergent discriminatory power. Further investigations showed the biofilm formation ability of all isolates with the prevalence of adhesion genes (fim-H and mrkD). The presence of a high resistance pattern with biofilm formation ability represents a worrisome feature of a superbug such as K. pneumoniae in the Egyptian healthcare setting. Thus, there is an urgent need for further strategies for combating resistance, in-depth studies for the resistance genes and their transfer, and also other characteristics such as biofilm formation which all strongly improve the ability of any pathogen to cause high rates of morbidity and mortality.
Supplementary Materials: The following are available online http://www.mdpi.com/2079-6382/9/12/852/s1. Table S1: Antimicrobial susceptibility patterns of Klebsiella pneumoniae isolates by disc diffusion method; Table S2: Results of modified Hodge test, inhibitor-based tests, and genotypic detection of investigated carbapenemases; Table S3: Simpson's Diversity Index calculations for antibiogram; Table S4: Simpson's Diversity Index calculations for ERIC-PCR; Table S5: Simpson's Diversity Index calculations for plasmid profile analysis; Table S6: A comparative summary of SDI calculations of the three typing methods; Table S7: Distribution of fimH-1 and mrkD adhesion genes among the carbapenem-resistant isolates; Table S8: The applied antimicrobial agents, their classes, and their interpretive categories; Table S9: Primer sequences and the expected amplicon sizes of investigated genes.

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