Relationships between phagocytosis, mucoid phenotype, and genetic characteristics of Klebsiella pneumoniae clinical isolates

Klebsiella pneumoniae is an enterobacterium responsible for urinary tract infections, pneumonia, meningitis, septicemia, and wound infections, especially in hospitalized immunocompromised individuals¹. The bacteria tend to produce extended-spectrum β-lactamases (ESBLs), carbapenemase (K.pneumoniaecarbapenemase, KPC), and efflux pumps^1–3.

The many virulence factors in K. pneumoniae that determine bacterial pathogenicity include the polysaccharide capsule, which constitutes a physical barrier and hinders phagocytosis⁴. Other factors include adherence to host cells via fimbrial and non-fimbrial adhesins, synthesis of siderophores (yersiniabactin, enterobactin, and/or aerobactin)⁴ allowing uptake of iron, the presence of lipopolysaccharide to protect bacteria from the bactericidal effect of antibodies and complement system, and the mucoid phenotype of colonies⁵. The latter pathogenicity factor of K. pneumoniae is associated with the presence of a 180 kb plasmid^6,1.

Immunological studies have demonstrated the role of macrophages in the immune system, emphasizing the importance of analyzing the virulence of bacterial strains and the response developed by the host^8,9. One of the defense mechanisms of the respiratory tract against invading microorganisms and other inhaled particles depends on the phagocytic ability of alveolar macrophages^8,9. These macrophages are involved in all aspects of the immune response, which include the innate and adaptive systems⁹. The failure of recognition of an invading organism resulting in failure of phagocytosis increases the susceptibility of the host to infection¹⁰.

The aim of this study was to evaluate the correlation of macrophage phagocytosis with the mucoid phenotype of K pneumoniae clinical isolates and genetic characteristics, specifically the plasmid profile and presence of the irp2 and bla _CTX-M-2 genes, related to the synthesis of the siderophore yersiniabactin and antimicrobial resistance, respectively, and the phenotypic feature of antimicrobial resistance.

Thirty clinical isolates of K. pneumoniae were obtained from different clinical samples from unrelated patients treated at three public hospitals in Recife-PE, Brazil (Table 1). For further identification, the cultures were biochemically analyzed using the computerized ID Mini API 32 E system (Bio Merieux, France), according to the manufacturer’s instructions. The isolates had been previously characterized regarding their plasmid profile, antimicrobial resistance profile, and presence of the irp2 and bla _CTX-M-2, as well as being analyzed using random amplification of polymorphic DNA-polymerase chain reaction (RAPD-PCR)^2,11. All 30 isolates were analyzed concerning the mucoid phenotype of the colonies. Four of the isolates that did not display a clonal relationship¹¹ by RAPD-PCR were selected according to their genetic features and antimicrobial resistance to determine the rate of phagocytosis by alveolar macrophages. For determination of the mucoid phenotype, the 30 isolates were seeded onto tryptose soy agar (Oxoid, UK), and incubated at 37ºC for 18 hours.

Alveolar macrophages were obtained by bronchoalveolar lavage from rats weighing 300 g and an average age of 90 to 120 days. Anesthesia was induced intraperitoneally with chloralose (0.5%) plus uretama (12.5%), using 1 ml of anesthetic per 100 g body weight. Tracheostomy was performed and the bronchoalveolar lavage fluid was aspirated using a syringe with saline. The material obtained was centrifuged at 252 x g for 10 min and the pellet was resuspended in 1 ml of 0.2% NaCl and centrifuged for 10 min. The pellet was resuspended in 2 ml of RPMI medium (Gibco BRL, NY), pH 7.0¹².

To prepare concentrated bacterial preparations, the four selected K. pneumoniae isolates (Table 2) were used to individually inoculate Brain Heart Infusion broth (Oxoid, UK). After incubation at 37°C for 18 hours, the bacterial cultures were centrifuged at 252 x g for 10 minutes and the supernatant was discarded. The pellet was centrifuged two more times (252 x g, 10 min) in 4 mL of phosphate buffered saline (PBS; 0.1 M, pH 7.2). Each final pellet was resuspended in 1 ml of PBS. The Neubauer chamber was used to enumerate bacteria and alveolar macrophages. Each cell suspension and 0.05% trypan blue dye was added to obtain a 1:10 dilution of the bacteria, as calculated (cells/mm³) according to the formula n × 10 × 10 × 10³, where n represents number of cells. The desired density was 1 × 10⁶ macrophages/ml and 1 × 10⁷ colony forming units (CFU)/ml for bacteria. In the experiments of macrophage phagocytosis of the K. pneumoniae isolates, macrophage suspensions in RPMI 1640 medium and bacterial suspensions in PBS were mixed on a glass slide for optical microscopy. Each slide was then incubated for 1 hour at 37°C in a humid chamber. Eight repetitions were done for each bacterial isolate. After an interval of one hour, the slides were stained using the Fast Panoptic kit based on hematoxylin and eosin. The slides were washed, dried at room temperature, and observed by optical microscopy to determine the rate of phagocytosis of 100 alveolar macrophages. Statistical analysis of the average rate of phagocytosis involved the analysis of variance (ANOVA).

Among the 30 K. pneumoniae isolates tested, nine (30%) displayed the mucoid phenotype (Table 1 and Figure 1A). There was no evidence of a relationship between the mucoid phenotype and the source of the bacterial isolates. Comparison of the mucoid phenotype with the plasmid profile revealed two bacterial strains (6.6%, K6 and K8-C-R) that were mucoid and lacked the 180 kb plasmid, whereas 12 isolates (40%) were non-mucoid and harbored the plasmid (Table 1). Additionally, among three clinical isolates without plasmids, one displayed the mucoid phenotype.

Before analyzing the phagocytic ability of alveolar macrophages, the cell viability based on the exclusion of trypan blue was determined. Viability exceeded 95% in all the samples, which indicating that the bronchoalveolar lavage fluid method was capable of recovering viable cells that were capable of phagocytosis. Alveolar macrophages activated by K. pneumoniae displayed rates of phagocytosis ranging from 21.5% to 43.43% (Table 2). Figure 1B displays a typical image of the phagocytosis of K. pneumoniaeby an activated macrophage. The activated macrophage displayed cytoplasmic vacuoles that were larger and irregular in shape. To compare the average rates of phagocytosis of the four K. pneumoniae isolates tested, we used ANOVA. For the parameters analyzed at a significance level of 5%, the calculated F (0.56) was lower than the F table (2:57), therefore it was not necessary to perform the Tukey test since the means were not significantly different.

FIGURE 1: Mucoid phenotype and phagocytosis of K. pneumoniae. A. K. pneumoniae colonies presenting non-mucoid phenotype (K16-R, left) and mucoid phenotype (K-14R, right). B. Alveolar macrophage phagocytosing a K. pneumoniae K14-R bacterium. 

The phagocytosis rate of the multiresistant isolates K9-C and K16-R were 21.5% and 27.43%, respectively. These rates were lower than those found for K. pneumoniae isolates sensitive to antibiotics (K6-R and K14-R), suggesting a correlation between the phagocytosis rate and antimicrobial susceptibility (Table 2). The highest rates of phagocytosis, 27.43%, 29.43%, and 43.43% (Table 2) were seen in K. pneumoniae isolates with mucoid phenotype, whereas the lowest phagocytosis rate of 21.5% was observed in a non-mucoid strain. The findings indicated a relationship between the mucoid colony phenotype and the phagocytosis rate. The phagocytosis rate was lower in the K16-R resistant isolate that harbored the irp2 and bla _CTX-M-2 genes. There was also no correlation between the plasmid profile and the phagocytosis rate.

Since the mucoid phenotype of colonies is an important virulence factor in K. pneumoniae⁵, this phenotype was investigated in the 30 clinical isolates. Thirty percent of the isolates displayed the mucoid phenotype, which was consistent with the previous description of the low prevalence of mucoidy in K. pneumoniaeclinical isolates⁷. The mucoid phenotype is determined by the regulator of mucoid phenotype (rmpA) gene found in a 1.6 kb DNA segment present in a 180 kb plasmid in Escherichia coli and K. pneumoniaestrains^5,7. The analysis of the 30 K. pneumoniae isolates from patients with nosocomial infections showed no correlation between the mucoid phenotype in colonies with the presence of the 180 kb plasmid, since 6.6% of the mucoid isolates did not harbor the plasmid. These results suggested that the gene encoding the K. pneumoniae mucoid phenotype is not necessarily associated with the presence of the 180 kb plasmid, as has been described in the literature. The mucoid phenotype of the K. pneumoniae colonies did not decrease the level of phagocytosis by alveolar macrophages, since mucoid isolates showed the highest phagocytosis rate. However, the non-mucoid bacterial strain showed the lowest rate. Keisari et al.¹⁰demonstrated that macrophages recognize and phagocytose K. pneumoniae in two steps. Recognition of the capsular structures by macrophage mannose receptors is followed by opsonization by surfactant protein A (SP-A), which binds to the capsular polysaccharide of K. pneumoniae and the macrophage mannose receptors. According to these authors, bacterial surface specific structures may be targets for macrophage recognition by several mechanisms, as exemplified in the case of the polysaccharide capsule of K. pneumoniae. Therefore, in the present study, the absence of mucoid colonies of K. pneumoniae most likely hampered recognition by macrophages, thereby reducing the phagocytosis rate.

We observed that rat alveolar macrophages activated by K. pneumoniae displayed phagocytic rates ranging from 21.5% to 43.43%. This rate was lower than the 89.4 ± 6.2% rate of alveolar macrophage phagocytosis from bronchoalveolar lavage fluid of horses activated by cell wall of nonviable Saccharomyces cerevisiae¹³. However, the rate we observed exceeded the previously detected rates of alveolar macrophage phagocytosis in vitro to methicillin-resistant Staphylococcus aureus (12.1 ± 3.0 %) and methicillin-sensitive Staphylococcus aureus (10.4 ± 3.2 %)¹⁴, and exceeds the rate of phagocytosis detected by Salgado et al.¹⁵, 6.29 ± 0.46 (MSSA) and 5.87 ± 0.34 (MRSA).

The phagocytosis rate was related to antimicrobial resistance, because the K9 and K16-R-C multiresistant strains displayed a lower phagocytosis rate than that in the antimicrobial sensitive strains. These findings may be related to the presence of the 180 kb plasmid that was present in the K9-C and K16-R isolates. This plasmid encodes several virulence factors, including mucoidy, adhesion factor, aerobactin, and ESBL TEM-5, which confers resistance to third-generation cephalosporins^5,7. Therefore, the present results suggest that the low rate of phagocytosis of the multiresistant K. pneumoniae isolates was probably due to the presence of the 180 kb plasmid, which may carry genes for resistance and virulence factors, and which may consequently inhibit phagocytosis. Furthermore, K16-R, which was non-mucoid and virulent, had the lowest rate of phagocytosis by alveolar macrophages and was most resistant to antimicrobials (positive to bla _CTX-M-2 gene which encodes ESBL) and is probably the most virulent (positive to irp2 gene which encodes the yersiniabactin siderophore)⁴.

The collective data from this study indicate that several virulence factors act in parallel in K. pneumoniaeto impair host defense. The results highlight the need for further research aimed at providing data concerning virulence factors of K. pneumoniae, as well as to clarify the relationship of these factors with the host defense response.