Klebsiella pneumoniae and the pyogenic liver abscess: implications and association of the presence of rpmA genes and expression of hypermucoviscosity.

Klebsiella pneumoniae is a common Gram-negative human opportunistic pathogen that causes both community and nosocomial infections primarily in immunocompromised hosts. This pathogen produces a wide range of infections including, pneumonia, bacteremia, urinary tract infection, meningitis, intra-abdominal infections, and community-acquired pyogenic liver abscess (PLA).[1-5]

Several virulence factors have been characterized clearly demonstrating their contribution to K. pneumoniae pathogenesis. The major virulence factors so far described are: a) the capsular polysaccharide (CPS), which confers the bacteria the ability to evade phagocytosis by immune cells and impedes bacterial killing by bactericidal complement serum;[6] b) the LPS O-antigen, prevents complement protein deposition and complement-associated serum lytic activity;[7] c) several pili or fimbrial adherence factors required for epithelial cell attachment and host colonization such as type 1 and type 3 pili, K. pneumoniae fimbriae 28 (KPF-28), the non-fimbrial adhesin CF29K, and the E. coli common pilus (ECP) have been described;[2,8] d) enterobactin and aerobactin siderophores, are required for a pathogen to establish infection when entering the hosts and enhance iron uptake by K. pneumoniae,[9] e) mucosity, a phenotype associated with a muco-polysaccharide network associated outside the capsule can facilitate mucosal colonization by K. pneumoniae and protects it from the interaction with anti-capsule-specific antibodies to evade phagocytosis.[10] Further, the hypermucoviscosity phenotype (HV) reported for certain strains are believed to increase the virulence of K. pneumoniae. The HV is regulated by the expression of 2 capsular polysaccharide genes: regulator of the mucoid phenotype gene (rmpA), a determinant controlling the CPS biosynthesis located in the chromosome or on the 180–220 Kb plasmid, and the plasmid-encoded transcriptional regulator of mucoid phenotype rmpA2. Iron-acquisition factors encoded on the same plasmid are also involved in regulation of the HV phenotype.[11-13] Studies in animal models have shown that the HV-phenotype of K. pneumoniae isolates is one of the major virulence factors required to produce bacteremia in diabetic mice and that other factors could be involved in the systemic spread of bacteria.[14] Another important point essential to the ability of K. pneumoniae to overcome antimicrobial therapies and to survive in demanding environments is multi-drug resistance, especially of isolates associated with nosocomial infections. One of the most common mechanisms of antimicrobial resistance of K. pneumoniae is the extended-spectrum β-lactamases (ESBL) production. ESBL-producing K. pneumoniae are rarely associated with community-acquired bacteremia and rarely express the HV phenotype.[15] However, a recent report in China showed that antimicrobial resistance of K. pneumoniae HV+ variants has increased significantly.[12]

In 1986 a new invasive syndrome, primer pyogenic liver abscess (PLA), caused by K. pneumoniae was first described in Taiwan.[16] In the last 3 decades, this syndrome has been reported in patients from many Asian countries, South Africa, Australia, Europe, and America.[17,18] This community-acquired primary infection occurred primarily in healthy individuals and was associated with predisposing risk factors such as diabetes, malignancy, renal disease, and pneumonia.[19] PLA was caused by HV+ variants of K. pneumoniae, and thus were designated as hypervirulent.[13] These isolates exhibit the potential to cause nosocomial infections including pneumonia, meningitis, endophtalmitis, and may also cause metastatic infections.[12]

Several controversies have risen regarding the ability of K. pneumoniae to express the HV phenotype and its association with the prevalence of the rmpA/rmpA2 genes. Although some authors have reported that rmpA is present in the vast majority of HV+ isolates, a small proportion of HV+ isolates do not possess these genes.[20] Similarly, most of the HV-negative isolates do not carry rmpA/rmpA2 but a small group of them may carry rmpA and/or rmpA2.[20,21] Another point that remains unclear is whether the HV phenotype correlates with virulence. For instance, while the majority of isolates exhibiting the HV phenotype are considered more virulent, some HV-negative isolates show greater ability to cause bacteremia in the diabetes mouse model.[15] Thus, it is apparent that virulence cannot be defined by the expression of HV alone. It is possible that HV is a key factor induced by specific host signals at different niches and could be important at certain stages during infection, for example during biofilm formation, particularly to evade the immune system and to colonize certain tissues within the host. This would explain, in part, the versatility of this pathogen to cause a myriad of human diseases.

In a recent issue of Virulence, Yu et al.[22] investigated why some rmpA-positive K. pneumoniae were HV-negative. To this aim, they performed nucleotide sequencing analysis of the rmpA and rmpA2 genes in HV− and HV+ isolates. The comparative gene sequence analysis indicated the presence of frameshift mutations in rmpA or rmpA2 genes in all of the HV− isolates, but not in the HV+ isolates. The authors suggested that mutations in rmpA/rmpA2 influenced negatively, to some extent, the level of virulence associated with the HV phenotype of K. pneumoniae and also that these mutations are responsible for the poor association of HV with ESBL-producing K. pneumoniae isolates.

The emergence and spread of ESBL-producing K. pneumoniae strains raise important questions regarding optimal therapies for serious infections caused by these microorganisms.[23] K. pneumoniae causing nosocomial infections are ESBL-producing; however, hypervirulent K. pneumoniae variants causing PLA have been reported as non-ESBL-producing.[15] The possibility that K. pneumoniae isolates causing PLA can increase resistance to antimicrobials is a potential public health problem and a continuous threat. Recent studies indicate that hypervirulent isolates continue to be highly susceptible but have increased their degree of antimicrobial resistance over time.[12] Li et al. speculated why hypervirulent isolates exhibit low antimicrobial resistance, and suggested that these isolates perhaps cannot acquire resistance plasmids or may lose some resistance genes when they become hypervirulent.[12] Yu et al.[22] explained this controversy conducting conjugation experiments with ESBL-producing K. pneumoniae that exhibited the HV phenotype. They showed that the SHV-5 gene, one of the most prevalent ESBL genes and the rmpA gene are located on different plasmids. Based on their data they speculated that the ESBL-producing K. pneumoniae reduce genetic function of the rmpA system when the bacteria need to express ESBL genes during exposure to antibiotics. This is consistent with the general idea of bacterial metabolic economy to regulate gene expression in response to different host environment signals such as one under antibiotic selective pressure.

In sum, the epidemiological and molecular as well as experimental data from mouse models on infection indicate a strong correlation between the ability of PLA-causing K.pneumoniae strains to exhibit HV and to carry the rmpA/rmpA2 genes. The studies discussed here highlight the need to deepen our understanding of the biological significance of HV as it relates to pathogenesis in the host. It is clear that further molecular and epidemiological characterization studies of potential factors that may contribute to virulence of K. pneumoniae causing PLA are needed. Chief to the clinical importance of the HV phenotype of K. pneumoniae strains worldwide is need to expand the number of ESBL- or non-ESBL-producing K. pneumoniae isolates originating from different geographic regions of the world and to determine the distribution of rmpA genes with or without genetic mutations in these strains. This opens an avenue for an interesting line of research on the genetics factors of the bacteria and/or the environmental cues that regulate rmpA gene function and the origin of its frameshift mutations. The investigation of the true role of the rmpA genes in expression of the HV phenotype and the virulence of PLA-causing strains calls for knock-out/complementation studies to fulfill molecular Koch's Postulates. Of vital importance is the need to stand alert and increase the surveillance of antimicrobial resistance of these isolates, particularly in different regions the world where the syndrome is prevalent for earlier detection and optimal treatment of patients.