Antimicrobial Resistance Pattern, Pathogenicity and Molecular Properties of Hypervirulent Klebsiella pneumonia (hvKp) among Hospital-Acquired Infections in the Intensive Care Unit (ICU)

Hypervirulent Klebsiella pneumoniae (hvKp) is a new emerging variant of K. pneumoniae that is increasingly reported worldwide. The variant hvKp is known to cause severe invasive community-acquired infections such as metastatic meningitis, pyogenic liver abscesses (PLA) and endophthalmitis, but its role in hospital-acquired infections (HAIs) is little known. The aim of this study was to evaluate the prevalence of hvKp among hospital-acquired (HA) K. pneumoniae infections in the intensive care unit (ICU) and to compare between hvKp and classical K. pneumoniae (cKP) regarding antimicrobial resistance pattern, virulence and molecular characteristics. The study was cross-sectional and included 120 ICU patients suffering from HA K. pneumoniae infections between January and September 2022. K. pneumoniae isolates were subjected to antimicrobial susceptibility testing and detection of extended-spectrum-β-lactamase (ESBL) production by the Phoenix 100 automated microbiology system, string test, biofilm formation, serum resistance assay, and detection of virulence-associated genes (rmpA, rmpA2, magA, iucA) and capsular serotype-specific genes (K1, K2, K5, K20, K57) by polymerase chain reaction (PCR). Of 120 K. pneumoniae isolates, 19 (15.8%) were hvKp. The hypermucoviscous phenotype was more significantly detected in the hvKp group than in the cKP group (100% vs. 7.9%, p ≤ 0.001). The rate of resistance to different antimicrobial agents was significantly higher in the cKP group than that in the hvKp group. Fifty-three strains were identified as ESBL-producing strains, which was more frequent in the cKP group than in the hvKp group (48/101 [47.5%] vs. 5/19 [26.3%], respectively, p ≤ 0.001). The hvKP isolates were highly associated with moderate and strong biofilm formation than cKP isolates (p = 0.018 and p = 0.043 respectively). Moreover, the hvKP isolates were highly associated with intermediate sensitivity and re sistance to serum in the serum resistance assay (p = 0.043 and p = 0.016 respectively). K1, K2, rmpA, rmpA2, magA and iucA genes were significantly associated with hvKp (p ≤ 0.001, 0.004, <0.001, <0.001, 0.037 and <0.001, respectively). However, K5, K20 and K57 were not associated with hvKp. The hvKp strains have emerged as a new threat to ICU patients because of their ability to cause more severe and life-threatening infections than cKP. The string test alone as a laboratory test for screening of hvKp has become insufficient. Recently, hvKp was defined as hypermucoviscous- and aerobactin-positive. It is important to improve the awareness towards the diagnosis and management of hvKp infections.


Introduction
Klebsiella pneumoniae (Kp) is a clinically important member of the Enterobacteriaceae family that can cause a wide range of infections such as pneumonia, urinary tract, bacteremia and wound infections. K. pneumonae usually cause hospital-acquired infections (HAIs) and occur primarily in patients with impaired immunity. Klebsiella pneumoniae has two different pathotypes: classical (cKp), which is the most common subtype of the K. pneumoniae strains that has received increased notoriety due to its ability to develop multidrug resistance and a new emerging variant termed hypervirulent (hvKp) [1][2][3]. The variant hvKp was first recognized in Taiwan in 1986 [4]. The hvKp strains are able to cause metastatic and life-threatening infections in immunocompetent and young healthy individuals. The variant hvKp is a leading cause of pyogenic liver abscesses (PLA) and it has also been implicated in metastatic meningitis and endophthalmitis. The variant hvKp has a characteristic hypermucoviscous colonies when grown on an agar plate, which can be detected by string test [5]. However, there are some controversies about the definition of hvKp. Many factors should be considered such as host, pathogen and host-pathogen interactions when defining hvKp. Nevertheless, most published papers have concentrated only on the bacteria. A previous study revealed that nutritional status, major histocompatibility complex (MHC) variants, composition of gut microbiota and eating habits are important host factors to investigate in order to improve our understanding of the hypervirulence phenomenon [6]. In addition, some controversies have arisen regarding the relationship between hypermucoviscous phenotype and virulence. Using in vivo and in vitro experiments, several studies revealed that few hypermucoviscous K. pneumonia (hmvKp) isolates are associated with high virulence [7,8]. In animal models, hmvKp did not show more serious infections or a higher mortality rate than non-hmvKp. Therefore, hvKp cannot be defined by string test alone [6,9].
Currently, aerobactin has been considered to be a crucial virulence factor for hvKp, which is often associated with the hypermucoviscous phenotype. Based on this finding, research conducted in China that included multiple centres first described the clinical and molecular characteristics of hvKp (defined as aerobactin-positive) isolates. The results revealed that invasive infections, especially PLA, hypermucoviscous phenotype and most of the virulence factors such as rmpA, rmpA2 (both regulators of the mucoid phenotype), and magA (mucoviscosity-associated gene) and capsular serotype-specific genes (K1, K2), are strongly associated with aerobactin-positive Kp [9][10][11]. Furthermore, some studies have revealed that iron acquisition genes and the genes that encode the hypermucoviscous phenotype are situated on the same virulence plasmid, which is present in most hvKp isolates but rarely present in cKp strains [12][13][14]. Thus, it may be more appropriate to combine aerobactin positivity and hypermucoviscosity when defining hvKp.
This study was conducted to evaluate the prevalence of hvKp among hospital-acquired (HA) K. pneumoniae infections in the intensive care unit (ICU) and to compare hvKp and classical K. pneumoniae (cKp) regarding antimicrobial resistance pattern, virulence and molecular characteristics.

Patients
The study was cross-sectional and included 120 ICU patients suffering from HA K. pneumoniae infections in Prince Mutaeb Bin Abdulaziz Hospital, Aljouf, Kingdom of Saudi Arabia between January and September 2022. Duplicate isolates from the same patient were excluded. Infections were considered as HA when a new infection developed 48 h after patient admission.

Clinical K. pneumoniae Isolates
The strains were isolated from the following clinical specimens: respiratory secretions, urine, blood and wound swabs. The strains were identified by the Phoenix 100 automated microbiology system (BD, Franklin Lakes, NJ, USA). The hvKp strains were defined as hypermucoviscous-and aerobactin-positive, which was confirmed by string test and PCR, respectively. The strains were stored on glycerol at −80 • C until PCR was performed.

String Test
Bacterial colonies grown overnight on a blood agar plate at 37 • C were stretched by a standard bacteriological loop. If a mucoviscous string > 5 mm in length was formed, the string test was considered positive and the isolate was identified as hypermucoviscous [15].

Biofilm Formation
Biofilm formation was examined by using the semi-quantitative assay in 96-well flat bottom plates as previously described [16].

Serum Resistance Assay
The serum resistance assay was performed as previously described [15]. Then, viable counts (VCs) of bacteria were determined for a period of 3 h. Responses were classified into six grades as follows: highly sensitive (grade 1 and 2), intermediately sensitive (grade 3 and 4), or resistant (grade 5 and 6) [17].

Detection of Capsular Serotype-Specific Genes and Virulence-Associated Genes
Detection of capsular serotype-specific genes (K1, K2, K5, K20 and K57) and virulenceassociated genes (rmpA, rmpA2, magA and iucA) were performed by polymerase chain reaction (PCR). The presence of the iucA gene was used to identify hvKp.
All isolated K. pneumoniae strains were subjected to DNA extraction by the boiling method as previously described [18]. The PCR used for amplification of capsular serotypespecific genes (K1, K2, K5, K20 and K57) was conducted as previously described [19]. PCR for detection of virulence-associated genes (rmpA, rmpA2 and magA) was applied as previously described [20,21]. PCR conditions for amplification of iucA (aerobactin) were as follows: initial denaturation at 95 • C for 15 min, followed by denaturation at 95 • C for 15 s, annealing at 49 • C for 15 min and extension at 72 • C for 1 min for 30 cycles followed by a final extension at 72 • C for 10 min. Primers used in this study are listed in Table 1. Then, the PCR products were subjected to electrophoresis at 100 V for 2 h in a 2% agarose gel containing ethidium bromide (0.5 µg/mL). DNA bands were visualized by UV illumination at 302 nm on a UV transilluminator. Table 1. List of primers used in the study.

Gene
Primers Sequences Reference Forward Reverse Forward Reverse Forward Reverse

Statistical Analysis
Statistical Package for Social Sciences (SPSS) version 20 was used for statistical analysis of the data. Qualitative data were described as numbers and percentages. Comparisons between groups were performed by chi-square test or Fisher's exact test. Test results with p < 0.05 were considered significant.

Comparison between hvKp and cKp Isolates Regarding Antimicrobial Susceptibility Testing and ESBL Production
The rate of resistance to antimicrobial agents was significantly higher in cKp than in hvKp strains, except ampicillin (all hvKp strains were resistant to ampicillin). Fifty-five strains were identified as ESBL-producing strains, which was more significantly detected in the cKp strains than in the hvKp strains (51/101 [50.5%] vs. 4/19 [20.05%], respectively, p = 0.018). The results of antibiotic susceptibility testing and ESBL production for hvKp and cKp are shown in Table 3.

Comparison between hvKP and cKP Isolates Regarding Biofilm Formation and Serum
The hypermucoviscous phenotype (based on the string test) was detected in 27 (22.5%) of all K. pneumoniae strains and this phenotype was more frequent in hvKp isolates than in cKp isolates (100% vs. 7.9%, p ≤ 0.001).

Comparison between non ESBL Producing hvKp and non ESBL Producing cKp Isolates Regarding Biofilm Formation and Serum Resistance
The hvKp isolates were highly associated with moderate and strong biofilm formation than the cKp isolates (p = 0.018 and p = 0.043, respectively). Moreover, the hvKp isolates were highly associated with intermediate sensitivity and resistance to serum in the serum resistance assay (p = 0.043 and p = 0.016, respectively). Results are shown in Table 4.

Comparison between non ESBL Producing hvKP and ESBL Producing cKP Isolates Regarding Biofilm Formation
The hvKp isolates were highly associated with moderate biofilm formation and intermediate sensitivity to serum in the serum resistance assay than the ESBL-producing cKp isolates (p = 0.044 and p = 0.031, respectively). There were no significant differences with strong biofilm formation and resistance to serum in serum resistance assay between hvKp and ESBL-producing cKp isolates (p = 0.621 and p = 0.542, respectively). Results are shown in Table 5.

Comparison between ESBL-Producing hvKp and ESBL-Producing cKp Isolates Regarding Biofilm Formation and Serum Resistance
There was no significant difference between ESBL-producing hvKp and ESBL-producing cKp isolates regarding biofilm formation or serum resistance assay (all p > 0.05). Results are shown in Table 6. Table 6. Biofilm formation and serum resistance assay of ESBL-producing hvKp vs. ESBL-producing cKp.

Comparison between hvKp and cKp Isolates Regarding Risk Factors and Associated Invasive Infections
Diabetes mellitus was a statistically significant risk factor associated with hvKp isolates (78.9% in hvKp vs. 35.6% in cKp, p = 0.003). On the other hand, there were no significant differences between hvKp and cKp strains regarding other underlying conditions of patients. Invasive infections were more significantly detected in patients with hvKp infections than those with cKp (42.1% in hvKp vs. 4.9% in cKp, p ≤ 0.001). Differences between hvKp and cKp groups are shown in Table 7.

Discussion
The variant hvKp is an increasingly reported pathotype of K. pneumoniae characterized clinically by its capability to cause metastatic and life-threatening infections in immunocompetent and young healthy individuals. In our study, the hypermucoviscous phenotype (based on the string test) was identified in 22.5% of all K. pneumoniae strains and this phenotype was significantly higher in hvKp strains. Another study performed in Egypt reported that the hypermucoviscous phenotype was detected in about 40% of K. pneumoniae isolates [22]. This discrepancy may be due to the difference in sample size.
In this study, hvKp (identified as hypermucoviscous-and aerobactin-positive) accounted for 15.8% of all HA K. pneumoniae infections in ICUs. This is in line with previous studies that showed that the prevalence of hvKp ranged from 7.8% to 25.4 [23,24].
Our results showed that the mean age of hvKp-infected patients was significantly younger than that of cKp-infected patients and invasive infections were significantly higher in patients with hvKp infections. These results are consistent with previous reports that showed that hvKp frequently causes invasive infections in young people without underlying disease [25][26][27][28].
Regarding antimicrobial resistance and ESBL production, our results showed that the resistance rate to common antibiotics in hvKp strains was significantly lower than in cKp strains and that ESBL production was more frequent in the cKp strains than in the hvKp strains. This was in accordance with another study conducted in China that reported the same findings [29]. In another study, the population genomics data suggested that acquisition of antimicrobial resistance plasmids by hvKp was more difficult than in cKp. The authors postulated that this may be due to the hyper-expression of the capsule, which may provide a physical barrier against the acquisition of antimicrobial resistance plasmids [30]. However, another study conducted in Egypt reported that there was no significant difference between hvKp and cKp strains regarding the antimicrobial resistance pattern [31]. The reason for this discrepancy may be due to the difference in sample size.
Regarding biofilm formation and serum resistance assay, our study showed that the hvKp isolates were highly associated with moderate and strong biofilm formation than the cKp isolates. Moreover, the hvKp isolates were highly associated with intermediate sensitivity and resistance to the serum in serum resistance assay than the cKp isolates. This is in agreement with a previous study that reported the same findings [32]. In contrast, another study conducted in Egypt reported that there was no significant difference between hvKp and cKp regarding biofilm formation and serum resistance assay [31]. This discrepancy may be attributed to the difference in sample size.
Regarding the virulence factors, the capsule is considered the major virulence factor in K. pneumoniae; there are several types of K-antigens [33][34][35]. K1 and K2 are the most important serotypes as they frequently result in severe infections [36,37]. In this study, K1 and K2 were significantly higher in hvKp than in cKp strains. The genes responsible for the hypermucoviscous phenotype (rmpA, mpA2 and MagA) are considered another virulence determinants in addition to K1/K2 [14,15,38,39]. Our results showed that rmpA, rmpA2 and magA were significantly associated with hvKp strains. These results are consistent with a previous study, which reported that there was a significant association between these genes and hvKp but not with cKp [29]. Aerobactin is considered a crucial virulence determinant of hvKp. In agreement with another study, our results showed that aerobactin was significantly associated with hvKp [40]. Therefore, these results showed that most of the virulence factors are more strongly associated with hvKp than with cKp.
Aligning with the results of a previous study, our results revealed that diabetes mellitus was a statistically significant risk factor associated with hvKp infections and that there were no significant differences between hvKp and cKp strains regarding other underlying conditions of patients [31].

Conclusions
The hvKp strains have emerged as a new threat to ICU patients because of their ability to cause more severe and life-threatening infections than cKp. The string test alone as a laboratory test for screening of hvKp has become insufficient. Recently, hvKp has been defined as hypermucoviscous-and aerobactin-positive. It is important to improve the awareness towards the diagnosis and management of hvKp infections. Antibiotic resistance, biofilm formation and biofilm-related genes were studied to evaluate their relationships with one another and with the genetic genotypes and phenotypes of hvKp isolated from clinical settings. Analysis of our data showed that genes have a role in biofilm formation and antibiotic resistance patterns. These methods may help shed light on the connections between hvKp biofilm formation and antibiotic resistance, as well as the transmission pathways of clinical isolates. Further insight into the link between biofilm production and antibiotic resistance might aid in the fight against drug-resistant bacteria.