Virulence Factors, Capsular Serotypes and Antimicrobial Resistance of Hypervirulent Klebsiella pneumoniae and Classical Klebsiella pneumoniae in Southeast Iran

Klebsiella pneumoniae (Kp) is a Gram-negative, encapsulated, non-motile bacterium, that is one of the most common causes of both health care and community-associated infections, including urinary tract infections, bacteremia, and pneumonia [1]. Two different groups of K. pneumoniae have been described: classical (cKP) and hypervirulent (hvKP). Currently, most K. pneumoniae infections, particularly in immunocompromised patients are due to cKp strains [2]. However, between the mid-1980s and 1990s, reports from Taiwan described a new and hypervirulent clinical variant of hvKP [3]. Although hvKP infection appears to occur often in diabetic patients, the importance of this variant is linked to its ability to cause community-acquired, life-threatening infection among young and healthy individuals [4]. HvKP strains often produce colonies with a hypermucoviscous phenotype that can be defined semi-quantitatively by a positive “string test,” a method that has been extensively used for identification of hvKP [4]. This hypermucoviscous phenotype can be related to overexpression of capsule polysaccharides. Most of hvKp isolates belong to capsular serotype K1 and K2 [5]. A capsule with mucoid character protects Klebsiella from phagocytosis and bactericidal effect of serum [6, 7]. A number of potential virulence factors, principally including mucoviscosity-associated gene A (magA) and regulator of mucoid phenotype A (rmpA), have been reported to be associated with hypermucoviscous phenotype [3]. Initially, magA was considered to mediate the hypermucoviscous phenotype. Further studies showed that magA is responsible for capsular serotype K1 of K. pneumoniae [8]. The rmpA gene is localized on a 180-kilobase plasmid and is a positive regulator of extracapsular polysaccharide synthesis [9].

Other virulence genes that are more prevalent in hypervirulent K. pneumoniae than the classical K. pneumoniae include siderophores such as enterobactin (Ent), the prototypical catecholate siderophore; aerobactin, a hydroxamate siderophore whose receptor is encoded by iutA; yersiniabactin (YbtS), a phenolate-type siderophore that is structurally distinct from enterobactin and Kfu, which mediates uptake of ferric [10]. Likewise, allS (a gene associated with allantoin metabolism) have been shown to be present more often in hvKP strains with a K1 serotype [10]. The fimH and mrkD genes which encode type 1 and type 3 fimbriae respectively are responsible for attachment to host cells [11]. These factors are known to contribute to virulence and are responsible for colonization, invasion, and pathogenicity. Alarmingly, some studies reported that multi-drug-resistant, even carbapenem-resistant hvKP isolates have emerged, which is becoming a major public health concern [3, 12]. In the current study, we investigated the distribution of various virulence genes, capsular serotypes and antibiotic resistance of cKP and hypermucoviscous/hvKP isolated from different clinical specimens in Kerman, south-east of Iran.

The present study was conducted on 146 K. pneumoniae isolates collected from patients hospitalized in Kerman hospitals from January 2017 to October 2018. HvKP isolates were defined on the basis of a positive string test. Of all K. pneumoniae isolates tested in this study, 22 isolates (15.1%) were hvKP. The hvKP prevalence of our study was lower than those reported by Li et al. (33%) in china [18], Yu et al. (38%) in Taiwan [19] and a study conducted by Jung et al. (42.4%) in Korea [20], but higher than the reports from a hospital in Spain (5.4%) [21] and a study in Alberta, Canada (8.2%) [22]. These variations in prevalence can be partly attributed to the geographical location. For example, previous studies showed that the prevalence of the hvKP in European countries is lower than East Asian countries [2, 20, 21, 23, 24].

In the current study, we evaluated six serotypes (K1, K2, K5, K20 K54 and K57), which are considered as highly virulent phenotypes associated with severe infections in humans [25]. The majority of hvKP isolates were associated with non-K1/K2/K5/K20/K54/K57 serotypes, and only 27.3% of hvKP isolates had K1/K2 serotypes. Although the K1/K2 capsular serotypes are common among hvKP isolates, some studies have shown that a considerable proportion of the hvKP strains may have a non-K1/K2 serotype [1, 25]. Our results revealed that serotypes other than K1/K2/K5/K20/K54/K57 may also play key roles in the hvKP infections. In this study K20 was the most frequent capsular type, but it was detected almost exclusively in the cKP isolates. This finding is consistent with the results of previous studies in which the K20 has been detected more commonly in non-hvKP strains [2, 18, 20]. Therefore, it seems that the role of K20 capsule in the pathogenesis of hypervirulent K. pneumoniae may not be as important as previously thought.

Besides hypermucoviscosity, a number of virulence factors have been suggested to contribute to the pathogenesis of hvKP strains. To date, rmpA and magA are the most frequently reported factors that have direct correlation with hvKP virulence [3, 26]. Although rmpA and magA were associated with positive string test, we found only 5 rmpA and/or magA positive among the 22 string test positive isolates. These results imply the presence of mucoid factors other than magA and rmpA; for example, some variations in the composition of lipopolysaccharide [21].

Iron acquisition factors including enterobactin (Ent), aerobactin, aerobactin receptor (IutA), and yersiniabactin (YbtS) and Kfu have been described as virulence factors in K. pneumoniae strains [27]. In our study, a majority of both hvKP and cKP isolates were demonstrated to carry ent gene, a finding in line with most previous reports [3, 10, 27]. More than half of the isolates also possessed ybtS and there was no significant difference in the prevalence of this gene between hvKP and cKP isolates. These results are almost in accordance with the findings of Yan et al. and Gao et al. [3, 27]. In agreement with other studies, the prevalence of iutA and kfu in hvKP isolates were significantly higher than those in cKP isolates [11, 27]. Moreover, an association between iutA and rmpA genes was found. This association is not surprising since both genes are located on the same 180-kb plasmid [20].

Fimbrial related genes, such as mrkD and fimH are also involved in the virulence of K. pneumoniae but, no significant differences in the prevalence of these genes were observed between hvKP and cKP strains. Guo et al. reported that only hypermucoviscous/hypervirulent isolates were positive for mrkD [3] but in many other studies, this gene was observed in both cKP and hvKP isolates [14, 27]. Moreover, consistent with most prior studies, we found that almost all of the K. pneumoniae strains carried mrkD [10, 14, 27].

In the present investigation, cKP isolates exhibited significantly higher antimicrobial resistant rate for all antimicrobial agents than hvKP isolates with the exception of ampicillin and trimethoprim/sulfamethoxazole. The same studies reported a high rate of resistance to ampicillin among hvKP isolates which is in agreement with our findings [3, 27, 28]. Although previous studies also indicated that hvKP strains are generally susceptible to antimicrobial agents, the reason for this difference is unclear. However, it can be hypothesized that hvKP strains cannot acquire resistance-related plasmids, or that some drug-resistant genes are lost when they become hypervirulent [18].

In the current study, cKP isolates were resistant to most of the antibiotics tested, including imipenem. Similarly, a high rate of imipenem resistance in K. pneumoniae was reported from Iran [29, 30]. Our previous studies demonstrated that New Delhi Metallo-beta lactamase (NDM) is responsible for carbapenem resistance in K. pneumoniae in Kerman, Iran [31, 32]. Although the prevalence of imipenem-resistant hvKP strains in our study was low, the enhanced virulence of these strains and the possibility of acquiring antimicrobial resistance are a cause for concern.

In conclusion, the prevalence of hvKP was low, but overall, the prevalence of virulence-related genes was higher in hvKP than cKP. HvKP was not related to specific serotypes. Furthermore, although hvKP isolates were more susceptible to clinically common used antimicrobial agents relative to cKP isolates, the emergence of antimicrobial resistance among these strains may become a future threat.