Veterinary Microbiology 176 (2015) 134–142
Contents lists available at ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Resistance of Klebsiella pneumoniae to the innate immune system of African green monkeys Brandi L. Cox a, Holly Schiffer a, Gregory Dagget Jr.a, Amy Beierschmitt a,b, Fortune Sithole a, Elise Lee a, Floyd Revan a, Iona Halliday-Simmonds a, Janet Beeler-Marfisi a,c, Roberta Palmour b, Esteban Soto a,* a b c
Department of Biomedical Sciences, Ross University School of Veterinary Medicine, Saint Kitts and Nevis Behavioural Science Foundation, Estridge Estate, Saint Kitts and Nevis Gribbles Veterinary Pathology, Christchurch, New Zealand
A R T I C L E I N F O
A B S T R A C T
Article history: Received 28 July 2014 Received in revised form 1 January 2015 Accepted 3 January 2015
In recent years, an emergent Klebsiella pneumoniae hypermucoviscosity (HMV) phenotype has been associated with increased invasiveness and pathogenicity in primates. In this project, bacteria recovered from infected African green monkeys (AGM) (Chlorocebus aethiops sabaeus) were screened for HMV phenotype, and were compared to non-HMV isolates in in vitro, serum, and oxidative-mediated killing assays. Complementmediated killing was assessed utilizing freshly collected serum from healthy AGM. Oxidative-mediated killing was investigated utilizing sodium hypochlorite and hydrogen peroxide. Compared to non-HMV isolates, HMV isolates were more resistant to serum-mediated and oxidative killing (p < 0.05). Phagocytosis resistance was evaluated using AGM peripheral blood monocytes (PBMC), and results indicated that non-HMV isolates associated with the AGM PBMC to a greater extent than HMV isolates (p < 0.001). Measurement of lactate dehydrogenase release showed that HMV isolates were more cytotoxic to AGM PBMC than non-HMV isolates (p < 0.001). Thus, the hypermucoid phenotype appears to be an important virulence factor that promotes evasion of innate immune defenses. ß 2015 Elsevier B.V. All rights reserved.
Keywords: Hypermucoviscosity Oxidative killing Cytotoxicity Phagocytosis Serum resistance
1. Introduction Klebsiella pneumoniae is a Gram-negative, facultative anaerobic bacterium belonging to the family Enterobacteriaceae. Typically found in the environment, and on mucosal surfaces (Podschun and Ullmann, 1998), it has been associated with a range of infections in humans and animals, including bacterial pneumonia, septicemia, urinary tract infections, meningitis, and soft tissue infections in hospitalized human patients (Podschun and Ullmann,
* Corresponding author. Tel.: +869 465 4161; fax: +869 465 6165. E-mail address: [email protected] (E. Soto). http://dx.doi.org/10.1016/j.vetmic.2015.01.001 0378-1135/ß 2015 Elsevier B.V. All rights reserved.
1998; Doud et al., 2009). Similar infections have been observed in Old and New World monkeys (Gozalo and Montoyo, 1991; Soto et al., 2012). K. pneumoniae is often associated with mastitis in ruminants, urinary tract infections in dogs, and metritis in mares (Kikuchi et al., 1987; Ling et al., 2001; Munoz et al., 2007). In both human and veterinary medicine this bacterium is regarded as an emergent and common nosocomial pathogen (Lederman and Crum, 2005; Jang et al., 2010). A novel, invasive form of K. pneumoniae has emerged over the last two decades. First noted in humans in Taiwan, it has now been reported worldwide (Lederman and Crum, 2005; Nadasy et al., 2007). In humans, infection produces a syndrome characterized by osteomyelitis, endophthalmitis,
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
meningitis, and multisystemic abscesses in the lung, liver, cervix and brain. Invasive K. pneumoniae strains are associated with the hypermucoviscosity (HMV) phenotype (Kawai, 2006). The HMV phenotype is characterized by different capsular serotypes, and is associated with several virulence genes including rmpA (regulator of mucoid phenotype) and magA (mucoviscosity-associated) genes (Kawai, 2006; Turton et al., 2010). The magA gene encodes a 43-kD outer membrane protein, whereas the rmpA gene is a transcriptional activator of colanic acid biosynthesis (Nassif et al., 1989). The best characterized virulence determinants of K. pneumoniae are the capsule, lipopolysaccharides, siderophores, and types 1 and 3 fimbriae (Broberg et al., 2014). There is well documented evidence regarding the importance and role of the capsule in the pathogenesis of K. pneumoniae infection. As with other Gram-negative bacteria, the capsule is associated with attachment to host receptors, protection from phagocytosis, and with barrier function against innate host defense components such as complement, lysozyme, and oxidative-mediated killing (Simoons-Smit et al., 1986). Distinct capsular components and an increased amount of capsular material in HMV K. pneumoniae have been described in hypervirulent K. pneumoniae human isolates (Shon et al., 2013). However, little work elucidating the role of the HMV phenotype in K. pneumoniae pathogenicity exists, and no direct comparison of HMV and non-HMV isolates using components of the innate immune system of naturally susceptible hosts has been performed. In humans, serotypes K1–K6 are associated more frequently with severe respiratory infection and septicemia than the higher numbered serotypes (Simoons-Smit et al., 1986; Fung et al., 2011). In equine hosts, serotypes K2, K5, and K7 are considered the most pathogenic (Crouch et al., 1972). Similarly, K. pneumoniae isolates with the HMV phenotype, that are PCR positive for the rmpA and K2 serotype associated gene (wyz) and negative for the K1 serotype associated gene (magA) have been reported as the etiologic agent of suppurative pneumonia, pleuritis and abscesses in California sea lions (Zalophus californianus) (Jang et al., 2010). Similar outbreaks of acute suppurative infection due to K. pneumoniae have been reported in New Zealand sea lion (Phocarctos hookeri) pups during and after epidemics on Enderby Island (Castinel et al., 2008). While African green monkeys (AGM), commonly used in biomedical research worldwide, have been reported with fatal multisystemic abscesses caused by HMV serotypes of K. pneumoniae (Twenhafel et al., 2008; Hartman et al., 2009; Soto et al., 2012) very little is known about the pathogenesis of HMV K. pneumoniae infection in this host. To gain a better understanding about the pathogenesis of this important emergent disease in primates, we compared specific components of the AGM innate immune response to HMV and non-HMV K. pneumoniae isolates previously recovered from clinically affected AGM. This is particularly important since both HMV and non-HMV isolates have been recovered from clinically ill animals and few chemotherapeutics are available for use in veterinary medicine against these bacteria.
135
2. Materials and methods 2.1. Animal care and use Both the Ross University School of Veterinary Medicine and Behavioural Science Foundation Institutional Animal Care and Use Committees reviewed and approved this study. 2.2. Bacterial strains and culture conditions K. pneumoniae strains isolated from septicemic AGM with single or multifocal abscesses were isolated at the Ross University School of Veterinary Medicine Diagnostic Laboratory from 2010 to 2012. Identification of the isolates was made according to standard clinical microbiologic and molecular methods (Soto et al., 2012; Hartman et al., 2009). All isolates were suspended in brain–heart infusion broth (BHI) (Sigma–Aldrich, St. Louis, MO, USA) with a final concentration of 20% glycerol and stored at 80 8C before use. K. pneumoniae reference (American Type Culture Collection [ATCC] #700603) served as the control in several assays. For general use, K. pneumoniae was grown on 5% sheep blood agar plates or in BHI broth at 37 8C. The presence of genes previously associated with invasive diseases and hypermucoviscosity was determined by real time PCR utilizing previously published protocols (Table 1) (Hartman et al., 2009). 2.3. String test for hypermucoviscosity K. pneumoniae isolates were inoculated on 5% sheep blood agar plates (Sigma–Aldrich, St. Louis, MO, USA) and incubated at 37 8C overnight. A standard bacteriologic loop was used to stretch a mucoviscous string from the colony. Hypermucoviscosity was defined by the formation of viscous strings >5 mm in length when a loop was used to stretch the colony on the agar plate (positive string test) (Fang et al., 2004). 2.4. Serotype specific PCR A multiplex PCR, designed to detect K. pneumoniae capsular types K1, K2, K5, K20, K54, and K57, two putative virulence factors (rmpA and wcaG), and the 16S–23S internal transcribed spacer unit of K. pneumoniae, was utilized to serotype isolates used in this study (Table 1), following published protocols (Turton et al., 2010). 2.5. Animal inclusion criteria by serological and molecular analysis In order to assess prior exposure or current infection (clinical or sub-clinical) of K. pneumoniae, 61 captive AGM from the Behavioural Science Foundation, St. Kitts, were screened to identify non-exposed animals (inclusion criteria). Blood, fecal and oral swabs were collected from animals with no previous history of K. pneumoniae infection, or with no previous history of being in contact with clinically infected animals.
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
136
Table 1 Klebsiella pneumoniae isolates used in this study. Isolate designation
Species identification
Host isolated from
HMV phenotype as determined by string test
Detection of khea
Detection of rmpAa
Detection of magAa
Serotypeb
KS1 KS3 KS4
Klebsiella pneumoniae Klebsiella pneumoniae Klebsiella pneumoniae (ATTC 700603) Klebsiella pneumoniae
AGM AGM Human
Positive Positive Negative
Positive Positive Positive
Positive Positive Negative
Positive Negative Negative
K1 K5 ND
AGM
Negative
Positive
Negative
Negative
ND
KS5
AGM, African green monkey; ND, not determined. a Done as described by Hartman et al. (2009). b Done as described by Turton et al. (2010).
Serum was harvested and assayed to determine antibody concentration by ELISA following published methods (Soto et al., 2011). Briefly, Immulon II 96-well flat-bottom microtitre plates (Thermo Labsystems, Franklin, MA, USA) were coated with 5 106 colony forming units (CFU) per well live K. pneumoniae in carbonate coating buffer, pH 9.6, at 100 mL per well, and incubated overnight at 4 8C. Plates were washed three times in PBS containing 0.05% Tween-20 (PBST), and blocked for 1 h at room temperature (RT) with ELISA Blocking Buffer (Sigma– Aldrich, St. Louis, MO, USA). Serum samples were serially diluted using 2-fold intervals. Negative control wells were incubated with PBST alone. Plates were incubated overnight at 4 8C and washed 5 with PBST. HRP-conjugated goat-anti-Monkey IgG-H&L polyclonal secondary antibody (Abcam, Cambridge, MA, USA) was diluted 1:10,000 in PBST and 100 mL was added to each well. After incubation at 25 8C for 1 h, the plate was washed 5 with PBST before adding 100 mL of ABTS Peroxidase Substrate (KPL, Gaithersburg, MD) to each well. The ELISA reaction was stopped after 30 min with 100 mL 1% sodium dodecyl sulfate, and the optical density (OD) of the reactions was read at 405 nm with a SpectraMax M2/M2e Microplate Reader (Molecular Devices, Sunnyvale, CA, USA). The endpoint titer was defined as the lowest dilution of serum that gave an OD value three standard deviations above that of the negative control wells. Oral and fecal swab samples were suspended in 500 mL PBS and total DNA was extracted utilizing, the DNeasy Kit and the QIAamp DNA Stool Mini Kit (Qiagen, Valencia, CA, USA), respectively. Extracted DNA served as template in a real-time PCR assay for the detection of the K. pneumoniae khe, rpmA, or magA genes following published protocols (Hartman et al., 2009). 2.6. Serum resistance assay The serum resistance of K. pneumoniae strains was determined using previously published methods (Soto et al., 2010). Briefly, pooled AGM serum from healthy donors (meeting inclusion criteria) was mixed at a 1:1 vol/ vol ratio with an inoculum of 5 104 CFU prepared from mid-log phase cultures. A control group was created by heating pooled serum at 56 8C for 30 min and mixed the same way. The final mixtures were incubated at 37 8C. Colony counts, determined by the serial dilution method, were checked at the baseline (0 h), and at 1 h and 2 h post-
inoculation. All experiments were performed in triplicate on a minimum of three separate occasions to confirm repeatability of results. 2.7. Oxidative-mediated killing assay Hydrogen peroxide (H2O2) (Sigma–Aldrich, St. Louis, MO, USA) and sodium hypochlorite (NaOCl) (Sigma– Aldrich, St. Louis, MO, USA) were used to evaluate each isolates resistance to oxidative compounds following published protocols (McKenna and Davies, 1988; Bonvillain et al., 2011). Fresh preparations of H2O2 were made at concentrations of 4 mM, and 8 mM. Similarly, fresh preparations of NaOCl were made at concentrations of 0.04 mM, and 0.08 mM. An inoculum of 105 CFU/ml, prepared from mid-log phase cultures, was mixed at a 1:1 vol/vol ratio with PBS (controls), H2O2 or NaOCl. The final mixtures were composed of 0, 2 or 4 mM H2O2, or 0, 0.02 or 0.04 mM NaOCl by volume. The mixtures were incubated at 37 8C for 1 h then transferred to ice to halt growth. The average survival percentage was plotted against the control. 2.8. Phagocytosis by peripheral blood mononuclear cell assays Blood samples (4–6 ml) were collected from adult AGM confirmed as healthy on physical examination. Briefly, monkeys were isolated by tunneling into a squeeze cage and anesthetized with ketamine (10 mg/kg intramuscular) and brought to a central husbandry area for examination by a veterinarian. Whole blood (4–6 ml/animal) was collected from the femoral vein using EDTA vacutainer tubes (Becton Dickinson and Company, Sparks, MD, USA), and placed on ice. Animals recovered from anesthesia in their home cages or tunnels of their home cages. They were kept sheltered and under observation until completely recovered. Within 1 h of collection, peripheral blood mononuclear cells (PBMC) were purified using Histopaque 1077 (Sigma–Aldrich, St. Louis, MO, USA) following manufacturer’s instructions. Mononuclear cells were plated at a concentration 1 106 per well in 96-well tissue culture plates (Becton Dickinson and Company, Sparks, MD, USA). After a 4 h incubation at 37 8C, wells were washed with PBS to remove non-adherent cells followed by a 1 h incubation at 37 8C with a suspension of 1 106 bacteria in cell culture medium to obtain an
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
multiplicity of infection (MOI) of 1:20 (PBMC to bacteria). Final estimation of attached cells was done by counting the pooled non-adherent cells in culture media and PBS wash. The population of attached cells was >95% monocytes as determined by microscopic examination of Wright’s stained plates. After incubation, the wells were washed three times with PBS, and PBS supplemented with 0.5% Triton X-100 (Sigma–Aldrich, St. Louis, MO, USA) was added to lyse PBMC. Adhered and phagocytized bacteria were quantified by plating appropriate dilutions on agar plates. Similar methods were used in the internalization assays; however, after incubation of PBMC with the bacterial suspension, the wells were washed with PBS and incubated for 1 h with fresh medium containing gentamicin (100 mg/ml) (Sigma–Aldrich, St. Louis, MO, USA) to kill extracellular bacteria. PBMC were lysed, and intracellular bacteria were quantified as described above. 2.9. Detection of K. pneumoniae-mediated cytopathogenicity Cytopathogenicity was assessed by measuring the release of cytosolic lactate dehydrogenase (LDH) into the cell culture medium, which reflects a loss of plasma membrane integrity in infected PBMC. LDH concentration was measured using the colorimetric Cytotox 96 Kit (Promega, Madison, WI, USA) according to the manufacturer’s instructions. The percentage of cytopathogenicity was calculated as 100 [(experimental release spontaspontaneous release)]/[total release spontaneous release)], where spontaneous release is the amount of LDH in the supernatant of uninfected cells and the total release is the activity in cell lysates. 2.10. Statistical analysis The statistical software Stata version 13 (StataCorp LP, College Station, TX, USA) was used for all analyses described below and a Bonferroni correction was applied in evaluating all pairwise comparisons. 2.10.1. Serum resistance Complement-mediated killing (serum killing factor) present in normal or heat-inactivated serum was combined with the isolate factor to create a combined factor. Similarly, serum killing factor was combined with hypermucoid phenotype (present or absent) to create a new combined factor. An ANOVA model was run with the log transformed bacterial counts as the outcome and the following factors: combined serum killing-hypermucoid, time, and the interaction between the two factors. 2.10.2. Oxidative-mediated killing assay Analyses for the impact of the oxidizing agents (hydrogen peroxide and hypochlorous acid) were run separately but in a similar manner. In both instances, an ANOVA model was run with the log transformed bacterial counts as the outcome and the following factors: combined oxidizing agent-isolate, time, and the interaction between the two factors.
137
2.10.3. Attachment and uptake by peripheral blood mononuclear cells An ANOVA model was used to evaluate whether there was a significant difference in the number of bacteria recovered from PBMC, between each of the bacterial isolates. 2.10.4. K. pneumoniae-mediated cytotoxicity An ANOVA model was run to evaluate whether there was a significant difference in cytosolic LDH released by infected AGM PBMC, between each of the bacterial isolates. 3. Results 3.1. Detection of HMV phenotype and HMV associated genes and serotypes The identity of K. pneumoniae isolates was confirmed after amplification of the K. pneumoniae specific khe gene. The HMV phenotype associated genes magA and rmpA were detected in the KS1 isolate, also identified as presenting a K1 serotype (Table 1). In contrast, only the rmpA gene was detected by real-time PCR in the KS3 isolate; later identified as K5 serotype (Table 1). Both KS1 and KS3 isolates displayed the HMV phenotype (positive for the string test) (Fig. 1). The non-HMV K. pneumoniae, ATTC 700603 strain, and KS5 isolate contained neither of the HMV associated genes, and were negative for the string test (data not shown). 3.2. Inclusion criteria Real-time amplification of oral and fecal swab material detected the khe in 27/61 oral swabs and in 35/61 fecal swabs. The khe gene was detected in both the oral and fecal swabs in 18/61 animals. The rmpA gene was detected in an oral swab of only 1/61 monkeys; the magA gene was not detected in any sample. Antibodies against K. pneumoniae isolate KS1, KS3, KS4 and KS5 were detected in 44/61, 40/ 61, 50/61 and 36/61 serum samples, respectively. Only monkeys with no detectable K. pneumoniae antibodies and real-time PCR negative for the detection of the khe, rmpA or magA genes were considered K. pneumoniae negative; a prerequisite for inclusion as blood donors in this project. 3.3. K. pneumoniae isolates displaying the HMV phenotype are significantly more resistant to serum killing than nonHMV isolates After a 1 h incubation, only non-HMV isolates (KS4 and KS5) had significantly reduced CFU counts (p < 0.0001) (Fig. 2A–D). One hour post-incubation in normal serum, HMV isolates (KS1 and KS3) showed significantly greater CFU counts than non-HMV isolates (KS4 or KS5) (p < 0.0001). Similarly, 2 h post-incubation, significant reductions in CFU counts were observed in non-HMV strains, and no viable bacteria were detected in the KS5 isolate. In contrast, 74% of KS1 and 83% of KS3 HMV isolates were still viable at this time point (Figs. 2A–D and 3A). No significant difference was detected in the survival rate of KS1 and KS3 isolates (p > 0.05). Similarly, no significant
138
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
Fig. 1. Example of the string-test in hypermucoid (HMV) K. pneumoniae isolates used in this study. A. Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+). B. Klebsiella pneumoniae KS3 (khe+, rmpA+, magA).
difference was detected between the non-HMV isolates KS4 and KS5, after 1 h (p = 0.28), but KS5 CFU were significantly reduced after 2 h of incubation (p < 0.001). When incubated with heat-inactivated serum, no killing was observed in any isolate, suggesting that the killing observed in non-HMV bacteria was due to the action of complement (Figs. 3B and 4). 3.4. K. pneumoniae isolates displaying the HMV phenotype are significantly more resistant to oxidative-mediated killing than non-HMV isolates When incubated at a concentration of 0.02 mM NaOCl, no significant difference was observed in survival rates among any of the isolates (p > 0.05). However, when incubated at a concentration of 0.04 mM NaOCl, HMV isolates had significantly higher survival rates than nonHMV isolates (p < 0.05) (Fig. 5A and B). Similarly, when incubated with 2 mM H2O2, no significant difference in
survival rates was observed among any of the isolates (p > 0.05). However, when incubated with 4 mM H2O2, HMV isolates KS1 and KS3 were significantly more resistant to oxidative-mediated killing than non-HMV isolates (p < 0.05) (Fig. 5C and D). 3.5. K. pneumoniae isolates displaying the HMV phenotype are protected from phagocytosis Attachment and uptake of K. pneumoniae isolates by AGM PBMC was investigated in vitro. KS1 attached to AGM PBMC significantly less than did the HMV KS3 and nonHMV isolates (Fig. 6). Although few KS1 bacteria attached to the AGM PBMC, no KS1 was taken up by the phagocytes (Fig. 7). The KS3 HMV isolate was taken up at significantly lower rates than the non-hypermucoid KS5 isolate (p < 0.05) (Fig. 7). Uptake of the K. pneumoniae ATTC 700603 (KS4) isolate was not evaluated since it is resistant to gentamicin.
Fig. 2. Serum-mediated killing of K. pneumoniae with defined phenotype. Survival and growth of Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+) (A), Klebsiella pneumoniae KS3 (khe+, rmpA+, magA) (B), Klebsiella pneumoniae KS4 (khe+, rmpA, magA) (C), and Klebsiella pneumoniae KS5 (khe+, rmpA, magA) (D) in normal serum. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
139
Fig. 3. Serum-mediated bacterial killing of hypermucoid (HMV) and non-HMV K. pneumoniae isolates. Survival and growth of HMV and non-HMV Klebsiella pneumoniae in normal serum (A) and normal serum that had been heat inactivated at 55 8C for 30 m (B). The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
3.6. K. pneumoniae isolates displaying the HMV phenotype are significantly more cytotoxic to AGM PBMC than non-HMV isolates After 1 h of incubation, both HMV isolates were found to be significantly more cytotoxic to AGM PBMC than nonHMV isolates (p < 0.05) (Fig. 8). Among non-HMV isolates, KS5, was found to be significantly more cytotoxic than KS4 (p < 0.05). 4. Discussion The emergence of HMV and hypervirulent strains of K. pneumoniae is a concern in human and veterinary medicine. The potential of these strains to acquire multidrug resistance genes, lack of an objective, routine
diagnostic test in the clinical microbiology laboratory, lack of vaccines for humans and animals against these strains, their capacity to persist in the environment and to infect a wide-range of hosts, and, a lack of understanding of pathogenesis of the diseases caused by these strains, creates the potential for serious consequences in humans and animals. Non-human primates (NHPs) are not only extremely important animals in biomedical research, but also are naturally susceptible to K. pneumoniae infections. Due to their close phylogenetic relationship to humans, they are recognized as an intermediate animal model between humans and rodents, and are an indispensable model for human diseases (Yeager et al., 2012; Jasinska et al., 2013). In comparison with other model organisms such as mice, NHPs show remarkable similarities to humans
Fig. 4. Effect of serum on serum-mediated killing on Klebsiella pneumoniae of defined phenotype. Survival and growth of Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+) (A), Klebsiella pneumoniae KS3 (khe+, rmpA+, magA) (B), Klebsiella pneumoniae KS4 (khe+, rmpA, magA) (C), and Klebsiella pneumoniae KS5 (khe+, rmpA, magA) (D) in normal serum that had been heat inactivated at 55 8C for 30 m. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
140
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
Fig. 5. Effect of NaOCl and H2O2 on the survival of Klebsiella pneumoniae of defined phenotype. Survival of Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+), Klebsiella pneumoniae KS3 (khe+, rmpA+, magA), Klebsiella pneumoniae KS4 (khe+, rmpA, magA), and Klebsiella pneumoniae KS5 (khe+, rmpA, magA) 1 h post-incubation in different concentrations of sodium hypochlorite (NaOCl) or hydrogen peroxide (H2O2). Percent survival was calculated using controls suspended in PBS as 100% viable bacteria. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
in behavior, neurologic, reproductive and immunological development, and physiologic functions; providing a more suitable animal model of human infectious disease (Shen et al., 2013). Moreover, similar to human clinical cases, AGM naturally infected with HMV K. pneumoniae presented with abdominal, cerebral and pulmonary abscesses, that were associated with peritonitis, bronchopneumonia, mesenteric lymphadenitis, typhlocolitis, and hydroureter (Kawai, 2006; Twenhafel et al., 2008).
Thus, AGM are an ideal host to investigate the interaction of the immune system with this bacterium. Because of this, our main objective was to investigate and compare the survival and resistance of HMV and non-HMV isolates when challenged with primate innate immune components, and to further investigate the influence of the HMV phenotype on virulence. Two clinically relevant non-HMV isolates were used for comparative analysis.
Fig. 6. Influence of K. pneumoniae phenotype on attachment to primate PBMCs. Adhesion of Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+), Klebsiella pneumoniae KS3 (khe+, rmpA+, magA), Klebsiella pneumoniae KS4 (khe+, rmpA, magA), and Klebsiella pneumoniae KS5 (khe+, rmpA, magA) by African green monkey peripheral blood mononuclear cells was performed as described. The results are expressed as a percentage of inoculated bacteria. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
Fig. 7. Influence of K. pneumoniae phenotype on PBMC uptake. Uptake of Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+), Klebsiella pneumoniae KS3 (khe+, rmpA+, magA), and Klebsiella pneumoniae KS5 (khe+, rmpA, magA) by African green monkey peripheral blood mononuclear cells was determined as described. The results are expressed percentage of inoculated bacteria. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
Fig. 8. Influence of K. pneumoniae phenotype on PBMC cytotoxicity. Cytotoxicity of Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+), Klebsiella pneumoniae KS3 (khe+, rmpA+, magA), Klebsiella pneumoniae KS4 (khe+, rmpA, magA), and Klebsiella pneumoniae KS5 (khe+, rmpA, magA) in African green monkey peripheral blood mononuclear cells was determined at 1 h post inoculation. Cytotoxicity was assayed by release of LDH from infected cells as described in Section 2. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
The HMV phenotype of K. pneumoniae has been associated with the presence of virulence genes, magA and rmpA, and with serotypes K1, K2 and K5. The importance of the K antigen, magA and rmpA and HMV phenotype in K. pneumoniae virulence has been shown in a mouse model (Simoons-Smit et al., 1986; Lin et al., 2011; Fung et al., 2011), but as in vivo challenges have not been conducted in AGM, a direct comparison between species is not possible. In mice, magA+/rmpA+ and magA/rmpA+ K. pneumoniae isolates were found to be significantly more virulent than magA, rmpA, and an unencapsulated isogenic K1 mutant (DK1, magA+, rmpA+) isolate (Fung et al., 2011). magA+/ rmpA+ and magA/rmpA+ K. pneumoniae were significantly more invasive and virulent in vivo, and presented significantly different cytokine and chemokine responses than magA/rmpA isolates (Fung et al., 2011). Cytokines in the liver and serum from mice infected with low-doses (
Contents lists available at ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Resistance of Klebsiella pneumoniae to the innate immune system of African green monkeys Brandi L. Cox a, Holly Schiffer a, Gregory Dagget Jr.a, Amy Beierschmitt a,b, Fortune Sithole a, Elise Lee a, Floyd Revan a, Iona Halliday-Simmonds a, Janet Beeler-Marfisi a,c, Roberta Palmour b, Esteban Soto a,* a b c
Department of Biomedical Sciences, Ross University School of Veterinary Medicine, Saint Kitts and Nevis Behavioural Science Foundation, Estridge Estate, Saint Kitts and Nevis Gribbles Veterinary Pathology, Christchurch, New Zealand
A R T I C L E I N F O
A B S T R A C T
Article history: Received 28 July 2014 Received in revised form 1 January 2015 Accepted 3 January 2015
In recent years, an emergent Klebsiella pneumoniae hypermucoviscosity (HMV) phenotype has been associated with increased invasiveness and pathogenicity in primates. In this project, bacteria recovered from infected African green monkeys (AGM) (Chlorocebus aethiops sabaeus) were screened for HMV phenotype, and were compared to non-HMV isolates in in vitro, serum, and oxidative-mediated killing assays. Complementmediated killing was assessed utilizing freshly collected serum from healthy AGM. Oxidative-mediated killing was investigated utilizing sodium hypochlorite and hydrogen peroxide. Compared to non-HMV isolates, HMV isolates were more resistant to serum-mediated and oxidative killing (p < 0.05). Phagocytosis resistance was evaluated using AGM peripheral blood monocytes (PBMC), and results indicated that non-HMV isolates associated with the AGM PBMC to a greater extent than HMV isolates (p < 0.001). Measurement of lactate dehydrogenase release showed that HMV isolates were more cytotoxic to AGM PBMC than non-HMV isolates (p < 0.001). Thus, the hypermucoid phenotype appears to be an important virulence factor that promotes evasion of innate immune defenses. ß 2015 Elsevier B.V. All rights reserved.
Keywords: Hypermucoviscosity Oxidative killing Cytotoxicity Phagocytosis Serum resistance
1. Introduction Klebsiella pneumoniae is a Gram-negative, facultative anaerobic bacterium belonging to the family Enterobacteriaceae. Typically found in the environment, and on mucosal surfaces (Podschun and Ullmann, 1998), it has been associated with a range of infections in humans and animals, including bacterial pneumonia, septicemia, urinary tract infections, meningitis, and soft tissue infections in hospitalized human patients (Podschun and Ullmann,
* Corresponding author. Tel.: +869 465 4161; fax: +869 465 6165. E-mail address: [email protected] (E. Soto). http://dx.doi.org/10.1016/j.vetmic.2015.01.001 0378-1135/ß 2015 Elsevier B.V. All rights reserved.
1998; Doud et al., 2009). Similar infections have been observed in Old and New World monkeys (Gozalo and Montoyo, 1991; Soto et al., 2012). K. pneumoniae is often associated with mastitis in ruminants, urinary tract infections in dogs, and metritis in mares (Kikuchi et al., 1987; Ling et al., 2001; Munoz et al., 2007). In both human and veterinary medicine this bacterium is regarded as an emergent and common nosocomial pathogen (Lederman and Crum, 2005; Jang et al., 2010). A novel, invasive form of K. pneumoniae has emerged over the last two decades. First noted in humans in Taiwan, it has now been reported worldwide (Lederman and Crum, 2005; Nadasy et al., 2007). In humans, infection produces a syndrome characterized by osteomyelitis, endophthalmitis,
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
meningitis, and multisystemic abscesses in the lung, liver, cervix and brain. Invasive K. pneumoniae strains are associated with the hypermucoviscosity (HMV) phenotype (Kawai, 2006). The HMV phenotype is characterized by different capsular serotypes, and is associated with several virulence genes including rmpA (regulator of mucoid phenotype) and magA (mucoviscosity-associated) genes (Kawai, 2006; Turton et al., 2010). The magA gene encodes a 43-kD outer membrane protein, whereas the rmpA gene is a transcriptional activator of colanic acid biosynthesis (Nassif et al., 1989). The best characterized virulence determinants of K. pneumoniae are the capsule, lipopolysaccharides, siderophores, and types 1 and 3 fimbriae (Broberg et al., 2014). There is well documented evidence regarding the importance and role of the capsule in the pathogenesis of K. pneumoniae infection. As with other Gram-negative bacteria, the capsule is associated with attachment to host receptors, protection from phagocytosis, and with barrier function against innate host defense components such as complement, lysozyme, and oxidative-mediated killing (Simoons-Smit et al., 1986). Distinct capsular components and an increased amount of capsular material in HMV K. pneumoniae have been described in hypervirulent K. pneumoniae human isolates (Shon et al., 2013). However, little work elucidating the role of the HMV phenotype in K. pneumoniae pathogenicity exists, and no direct comparison of HMV and non-HMV isolates using components of the innate immune system of naturally susceptible hosts has been performed. In humans, serotypes K1–K6 are associated more frequently with severe respiratory infection and septicemia than the higher numbered serotypes (Simoons-Smit et al., 1986; Fung et al., 2011). In equine hosts, serotypes K2, K5, and K7 are considered the most pathogenic (Crouch et al., 1972). Similarly, K. pneumoniae isolates with the HMV phenotype, that are PCR positive for the rmpA and K2 serotype associated gene (wyz) and negative for the K1 serotype associated gene (magA) have been reported as the etiologic agent of suppurative pneumonia, pleuritis and abscesses in California sea lions (Zalophus californianus) (Jang et al., 2010). Similar outbreaks of acute suppurative infection due to K. pneumoniae have been reported in New Zealand sea lion (Phocarctos hookeri) pups during and after epidemics on Enderby Island (Castinel et al., 2008). While African green monkeys (AGM), commonly used in biomedical research worldwide, have been reported with fatal multisystemic abscesses caused by HMV serotypes of K. pneumoniae (Twenhafel et al., 2008; Hartman et al., 2009; Soto et al., 2012) very little is known about the pathogenesis of HMV K. pneumoniae infection in this host. To gain a better understanding about the pathogenesis of this important emergent disease in primates, we compared specific components of the AGM innate immune response to HMV and non-HMV K. pneumoniae isolates previously recovered from clinically affected AGM. This is particularly important since both HMV and non-HMV isolates have been recovered from clinically ill animals and few chemotherapeutics are available for use in veterinary medicine against these bacteria.
135
2. Materials and methods 2.1. Animal care and use Both the Ross University School of Veterinary Medicine and Behavioural Science Foundation Institutional Animal Care and Use Committees reviewed and approved this study. 2.2. Bacterial strains and culture conditions K. pneumoniae strains isolated from septicemic AGM with single or multifocal abscesses were isolated at the Ross University School of Veterinary Medicine Diagnostic Laboratory from 2010 to 2012. Identification of the isolates was made according to standard clinical microbiologic and molecular methods (Soto et al., 2012; Hartman et al., 2009). All isolates were suspended in brain–heart infusion broth (BHI) (Sigma–Aldrich, St. Louis, MO, USA) with a final concentration of 20% glycerol and stored at 80 8C before use. K. pneumoniae reference (American Type Culture Collection [ATCC] #700603) served as the control in several assays. For general use, K. pneumoniae was grown on 5% sheep blood agar plates or in BHI broth at 37 8C. The presence of genes previously associated with invasive diseases and hypermucoviscosity was determined by real time PCR utilizing previously published protocols (Table 1) (Hartman et al., 2009). 2.3. String test for hypermucoviscosity K. pneumoniae isolates were inoculated on 5% sheep blood agar plates (Sigma–Aldrich, St. Louis, MO, USA) and incubated at 37 8C overnight. A standard bacteriologic loop was used to stretch a mucoviscous string from the colony. Hypermucoviscosity was defined by the formation of viscous strings >5 mm in length when a loop was used to stretch the colony on the agar plate (positive string test) (Fang et al., 2004). 2.4. Serotype specific PCR A multiplex PCR, designed to detect K. pneumoniae capsular types K1, K2, K5, K20, K54, and K57, two putative virulence factors (rmpA and wcaG), and the 16S–23S internal transcribed spacer unit of K. pneumoniae, was utilized to serotype isolates used in this study (Table 1), following published protocols (Turton et al., 2010). 2.5. Animal inclusion criteria by serological and molecular analysis In order to assess prior exposure or current infection (clinical or sub-clinical) of K. pneumoniae, 61 captive AGM from the Behavioural Science Foundation, St. Kitts, were screened to identify non-exposed animals (inclusion criteria). Blood, fecal and oral swabs were collected from animals with no previous history of K. pneumoniae infection, or with no previous history of being in contact with clinically infected animals.
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
136
Table 1 Klebsiella pneumoniae isolates used in this study. Isolate designation
Species identification
Host isolated from
HMV phenotype as determined by string test
Detection of khea
Detection of rmpAa
Detection of magAa
Serotypeb
KS1 KS3 KS4
Klebsiella pneumoniae Klebsiella pneumoniae Klebsiella pneumoniae (ATTC 700603) Klebsiella pneumoniae
AGM AGM Human
Positive Positive Negative
Positive Positive Positive
Positive Positive Negative
Positive Negative Negative
K1 K5 ND
AGM
Negative
Positive
Negative
Negative
ND
KS5
AGM, African green monkey; ND, not determined. a Done as described by Hartman et al. (2009). b Done as described by Turton et al. (2010).
Serum was harvested and assayed to determine antibody concentration by ELISA following published methods (Soto et al., 2011). Briefly, Immulon II 96-well flat-bottom microtitre plates (Thermo Labsystems, Franklin, MA, USA) were coated with 5 106 colony forming units (CFU) per well live K. pneumoniae in carbonate coating buffer, pH 9.6, at 100 mL per well, and incubated overnight at 4 8C. Plates were washed three times in PBS containing 0.05% Tween-20 (PBST), and blocked for 1 h at room temperature (RT) with ELISA Blocking Buffer (Sigma– Aldrich, St. Louis, MO, USA). Serum samples were serially diluted using 2-fold intervals. Negative control wells were incubated with PBST alone. Plates were incubated overnight at 4 8C and washed 5 with PBST. HRP-conjugated goat-anti-Monkey IgG-H&L polyclonal secondary antibody (Abcam, Cambridge, MA, USA) was diluted 1:10,000 in PBST and 100 mL was added to each well. After incubation at 25 8C for 1 h, the plate was washed 5 with PBST before adding 100 mL of ABTS Peroxidase Substrate (KPL, Gaithersburg, MD) to each well. The ELISA reaction was stopped after 30 min with 100 mL 1% sodium dodecyl sulfate, and the optical density (OD) of the reactions was read at 405 nm with a SpectraMax M2/M2e Microplate Reader (Molecular Devices, Sunnyvale, CA, USA). The endpoint titer was defined as the lowest dilution of serum that gave an OD value three standard deviations above that of the negative control wells. Oral and fecal swab samples were suspended in 500 mL PBS and total DNA was extracted utilizing, the DNeasy Kit and the QIAamp DNA Stool Mini Kit (Qiagen, Valencia, CA, USA), respectively. Extracted DNA served as template in a real-time PCR assay for the detection of the K. pneumoniae khe, rpmA, or magA genes following published protocols (Hartman et al., 2009). 2.6. Serum resistance assay The serum resistance of K. pneumoniae strains was determined using previously published methods (Soto et al., 2010). Briefly, pooled AGM serum from healthy donors (meeting inclusion criteria) was mixed at a 1:1 vol/ vol ratio with an inoculum of 5 104 CFU prepared from mid-log phase cultures. A control group was created by heating pooled serum at 56 8C for 30 min and mixed the same way. The final mixtures were incubated at 37 8C. Colony counts, determined by the serial dilution method, were checked at the baseline (0 h), and at 1 h and 2 h post-
inoculation. All experiments were performed in triplicate on a minimum of three separate occasions to confirm repeatability of results. 2.7. Oxidative-mediated killing assay Hydrogen peroxide (H2O2) (Sigma–Aldrich, St. Louis, MO, USA) and sodium hypochlorite (NaOCl) (Sigma– Aldrich, St. Louis, MO, USA) were used to evaluate each isolates resistance to oxidative compounds following published protocols (McKenna and Davies, 1988; Bonvillain et al., 2011). Fresh preparations of H2O2 were made at concentrations of 4 mM, and 8 mM. Similarly, fresh preparations of NaOCl were made at concentrations of 0.04 mM, and 0.08 mM. An inoculum of 105 CFU/ml, prepared from mid-log phase cultures, was mixed at a 1:1 vol/vol ratio with PBS (controls), H2O2 or NaOCl. The final mixtures were composed of 0, 2 or 4 mM H2O2, or 0, 0.02 or 0.04 mM NaOCl by volume. The mixtures were incubated at 37 8C for 1 h then transferred to ice to halt growth. The average survival percentage was plotted against the control. 2.8. Phagocytosis by peripheral blood mononuclear cell assays Blood samples (4–6 ml) were collected from adult AGM confirmed as healthy on physical examination. Briefly, monkeys were isolated by tunneling into a squeeze cage and anesthetized with ketamine (10 mg/kg intramuscular) and brought to a central husbandry area for examination by a veterinarian. Whole blood (4–6 ml/animal) was collected from the femoral vein using EDTA vacutainer tubes (Becton Dickinson and Company, Sparks, MD, USA), and placed on ice. Animals recovered from anesthesia in their home cages or tunnels of their home cages. They were kept sheltered and under observation until completely recovered. Within 1 h of collection, peripheral blood mononuclear cells (PBMC) were purified using Histopaque 1077 (Sigma–Aldrich, St. Louis, MO, USA) following manufacturer’s instructions. Mononuclear cells were plated at a concentration 1 106 per well in 96-well tissue culture plates (Becton Dickinson and Company, Sparks, MD, USA). After a 4 h incubation at 37 8C, wells were washed with PBS to remove non-adherent cells followed by a 1 h incubation at 37 8C with a suspension of 1 106 bacteria in cell culture medium to obtain an
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
multiplicity of infection (MOI) of 1:20 (PBMC to bacteria). Final estimation of attached cells was done by counting the pooled non-adherent cells in culture media and PBS wash. The population of attached cells was >95% monocytes as determined by microscopic examination of Wright’s stained plates. After incubation, the wells were washed three times with PBS, and PBS supplemented with 0.5% Triton X-100 (Sigma–Aldrich, St. Louis, MO, USA) was added to lyse PBMC. Adhered and phagocytized bacteria were quantified by plating appropriate dilutions on agar plates. Similar methods were used in the internalization assays; however, after incubation of PBMC with the bacterial suspension, the wells were washed with PBS and incubated for 1 h with fresh medium containing gentamicin (100 mg/ml) (Sigma–Aldrich, St. Louis, MO, USA) to kill extracellular bacteria. PBMC were lysed, and intracellular bacteria were quantified as described above. 2.9. Detection of K. pneumoniae-mediated cytopathogenicity Cytopathogenicity was assessed by measuring the release of cytosolic lactate dehydrogenase (LDH) into the cell culture medium, which reflects a loss of plasma membrane integrity in infected PBMC. LDH concentration was measured using the colorimetric Cytotox 96 Kit (Promega, Madison, WI, USA) according to the manufacturer’s instructions. The percentage of cytopathogenicity was calculated as 100 [(experimental release spontaspontaneous release)]/[total release spontaneous release)], where spontaneous release is the amount of LDH in the supernatant of uninfected cells and the total release is the activity in cell lysates. 2.10. Statistical analysis The statistical software Stata version 13 (StataCorp LP, College Station, TX, USA) was used for all analyses described below and a Bonferroni correction was applied in evaluating all pairwise comparisons. 2.10.1. Serum resistance Complement-mediated killing (serum killing factor) present in normal or heat-inactivated serum was combined with the isolate factor to create a combined factor. Similarly, serum killing factor was combined with hypermucoid phenotype (present or absent) to create a new combined factor. An ANOVA model was run with the log transformed bacterial counts as the outcome and the following factors: combined serum killing-hypermucoid, time, and the interaction between the two factors. 2.10.2. Oxidative-mediated killing assay Analyses for the impact of the oxidizing agents (hydrogen peroxide and hypochlorous acid) were run separately but in a similar manner. In both instances, an ANOVA model was run with the log transformed bacterial counts as the outcome and the following factors: combined oxidizing agent-isolate, time, and the interaction between the two factors.
137
2.10.3. Attachment and uptake by peripheral blood mononuclear cells An ANOVA model was used to evaluate whether there was a significant difference in the number of bacteria recovered from PBMC, between each of the bacterial isolates. 2.10.4. K. pneumoniae-mediated cytotoxicity An ANOVA model was run to evaluate whether there was a significant difference in cytosolic LDH released by infected AGM PBMC, between each of the bacterial isolates. 3. Results 3.1. Detection of HMV phenotype and HMV associated genes and serotypes The identity of K. pneumoniae isolates was confirmed after amplification of the K. pneumoniae specific khe gene. The HMV phenotype associated genes magA and rmpA were detected in the KS1 isolate, also identified as presenting a K1 serotype (Table 1). In contrast, only the rmpA gene was detected by real-time PCR in the KS3 isolate; later identified as K5 serotype (Table 1). Both KS1 and KS3 isolates displayed the HMV phenotype (positive for the string test) (Fig. 1). The non-HMV K. pneumoniae, ATTC 700603 strain, and KS5 isolate contained neither of the HMV associated genes, and were negative for the string test (data not shown). 3.2. Inclusion criteria Real-time amplification of oral and fecal swab material detected the khe in 27/61 oral swabs and in 35/61 fecal swabs. The khe gene was detected in both the oral and fecal swabs in 18/61 animals. The rmpA gene was detected in an oral swab of only 1/61 monkeys; the magA gene was not detected in any sample. Antibodies against K. pneumoniae isolate KS1, KS3, KS4 and KS5 were detected in 44/61, 40/ 61, 50/61 and 36/61 serum samples, respectively. Only monkeys with no detectable K. pneumoniae antibodies and real-time PCR negative for the detection of the khe, rmpA or magA genes were considered K. pneumoniae negative; a prerequisite for inclusion as blood donors in this project. 3.3. K. pneumoniae isolates displaying the HMV phenotype are significantly more resistant to serum killing than nonHMV isolates After a 1 h incubation, only non-HMV isolates (KS4 and KS5) had significantly reduced CFU counts (p < 0.0001) (Fig. 2A–D). One hour post-incubation in normal serum, HMV isolates (KS1 and KS3) showed significantly greater CFU counts than non-HMV isolates (KS4 or KS5) (p < 0.0001). Similarly, 2 h post-incubation, significant reductions in CFU counts were observed in non-HMV strains, and no viable bacteria were detected in the KS5 isolate. In contrast, 74% of KS1 and 83% of KS3 HMV isolates were still viable at this time point (Figs. 2A–D and 3A). No significant difference was detected in the survival rate of KS1 and KS3 isolates (p > 0.05). Similarly, no significant
138
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
Fig. 1. Example of the string-test in hypermucoid (HMV) K. pneumoniae isolates used in this study. A. Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+). B. Klebsiella pneumoniae KS3 (khe+, rmpA+, magA).
difference was detected between the non-HMV isolates KS4 and KS5, after 1 h (p = 0.28), but KS5 CFU were significantly reduced after 2 h of incubation (p < 0.001). When incubated with heat-inactivated serum, no killing was observed in any isolate, suggesting that the killing observed in non-HMV bacteria was due to the action of complement (Figs. 3B and 4). 3.4. K. pneumoniae isolates displaying the HMV phenotype are significantly more resistant to oxidative-mediated killing than non-HMV isolates When incubated at a concentration of 0.02 mM NaOCl, no significant difference was observed in survival rates among any of the isolates (p > 0.05). However, when incubated at a concentration of 0.04 mM NaOCl, HMV isolates had significantly higher survival rates than nonHMV isolates (p < 0.05) (Fig. 5A and B). Similarly, when incubated with 2 mM H2O2, no significant difference in
survival rates was observed among any of the isolates (p > 0.05). However, when incubated with 4 mM H2O2, HMV isolates KS1 and KS3 were significantly more resistant to oxidative-mediated killing than non-HMV isolates (p < 0.05) (Fig. 5C and D). 3.5. K. pneumoniae isolates displaying the HMV phenotype are protected from phagocytosis Attachment and uptake of K. pneumoniae isolates by AGM PBMC was investigated in vitro. KS1 attached to AGM PBMC significantly less than did the HMV KS3 and nonHMV isolates (Fig. 6). Although few KS1 bacteria attached to the AGM PBMC, no KS1 was taken up by the phagocytes (Fig. 7). The KS3 HMV isolate was taken up at significantly lower rates than the non-hypermucoid KS5 isolate (p < 0.05) (Fig. 7). Uptake of the K. pneumoniae ATTC 700603 (KS4) isolate was not evaluated since it is resistant to gentamicin.
Fig. 2. Serum-mediated killing of K. pneumoniae with defined phenotype. Survival and growth of Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+) (A), Klebsiella pneumoniae KS3 (khe+, rmpA+, magA) (B), Klebsiella pneumoniae KS4 (khe+, rmpA, magA) (C), and Klebsiella pneumoniae KS5 (khe+, rmpA, magA) (D) in normal serum. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
139
Fig. 3. Serum-mediated bacterial killing of hypermucoid (HMV) and non-HMV K. pneumoniae isolates. Survival and growth of HMV and non-HMV Klebsiella pneumoniae in normal serum (A) and normal serum that had been heat inactivated at 55 8C for 30 m (B). The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
3.6. K. pneumoniae isolates displaying the HMV phenotype are significantly more cytotoxic to AGM PBMC than non-HMV isolates After 1 h of incubation, both HMV isolates were found to be significantly more cytotoxic to AGM PBMC than nonHMV isolates (p < 0.05) (Fig. 8). Among non-HMV isolates, KS5, was found to be significantly more cytotoxic than KS4 (p < 0.05). 4. Discussion The emergence of HMV and hypervirulent strains of K. pneumoniae is a concern in human and veterinary medicine. The potential of these strains to acquire multidrug resistance genes, lack of an objective, routine
diagnostic test in the clinical microbiology laboratory, lack of vaccines for humans and animals against these strains, their capacity to persist in the environment and to infect a wide-range of hosts, and, a lack of understanding of pathogenesis of the diseases caused by these strains, creates the potential for serious consequences in humans and animals. Non-human primates (NHPs) are not only extremely important animals in biomedical research, but also are naturally susceptible to K. pneumoniae infections. Due to their close phylogenetic relationship to humans, they are recognized as an intermediate animal model between humans and rodents, and are an indispensable model for human diseases (Yeager et al., 2012; Jasinska et al., 2013). In comparison with other model organisms such as mice, NHPs show remarkable similarities to humans
Fig. 4. Effect of serum on serum-mediated killing on Klebsiella pneumoniae of defined phenotype. Survival and growth of Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+) (A), Klebsiella pneumoniae KS3 (khe+, rmpA+, magA) (B), Klebsiella pneumoniae KS4 (khe+, rmpA, magA) (C), and Klebsiella pneumoniae KS5 (khe+, rmpA, magA) (D) in normal serum that had been heat inactivated at 55 8C for 30 m. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
140
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
Fig. 5. Effect of NaOCl and H2O2 on the survival of Klebsiella pneumoniae of defined phenotype. Survival of Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+), Klebsiella pneumoniae KS3 (khe+, rmpA+, magA), Klebsiella pneumoniae KS4 (khe+, rmpA, magA), and Klebsiella pneumoniae KS5 (khe+, rmpA, magA) 1 h post-incubation in different concentrations of sodium hypochlorite (NaOCl) or hydrogen peroxide (H2O2). Percent survival was calculated using controls suspended in PBS as 100% viable bacteria. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
in behavior, neurologic, reproductive and immunological development, and physiologic functions; providing a more suitable animal model of human infectious disease (Shen et al., 2013). Moreover, similar to human clinical cases, AGM naturally infected with HMV K. pneumoniae presented with abdominal, cerebral and pulmonary abscesses, that were associated with peritonitis, bronchopneumonia, mesenteric lymphadenitis, typhlocolitis, and hydroureter (Kawai, 2006; Twenhafel et al., 2008).
Thus, AGM are an ideal host to investigate the interaction of the immune system with this bacterium. Because of this, our main objective was to investigate and compare the survival and resistance of HMV and non-HMV isolates when challenged with primate innate immune components, and to further investigate the influence of the HMV phenotype on virulence. Two clinically relevant non-HMV isolates were used for comparative analysis.
Fig. 6. Influence of K. pneumoniae phenotype on attachment to primate PBMCs. Adhesion of Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+), Klebsiella pneumoniae KS3 (khe+, rmpA+, magA), Klebsiella pneumoniae KS4 (khe+, rmpA, magA), and Klebsiella pneumoniae KS5 (khe+, rmpA, magA) by African green monkey peripheral blood mononuclear cells was performed as described. The results are expressed as a percentage of inoculated bacteria. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
Fig. 7. Influence of K. pneumoniae phenotype on PBMC uptake. Uptake of Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+), Klebsiella pneumoniae KS3 (khe+, rmpA+, magA), and Klebsiella pneumoniae KS5 (khe+, rmpA, magA) by African green monkey peripheral blood mononuclear cells was determined as described. The results are expressed percentage of inoculated bacteria. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
B.L. Cox et al. / Veterinary Microbiology 176 (2015) 134–142
Fig. 8. Influence of K. pneumoniae phenotype on PBMC cytotoxicity. Cytotoxicity of Klebsiella pneumoniae KS1 (khe+, rmpA+, magA+), Klebsiella pneumoniae KS3 (khe+, rmpA+, magA), Klebsiella pneumoniae KS4 (khe+, rmpA, magA), and Klebsiella pneumoniae KS5 (khe+, rmpA, magA) in African green monkey peripheral blood mononuclear cells was determined at 1 h post inoculation. Cytotoxicity was assayed by release of LDH from infected cells as described in Section 2. The error bars represent standard errors for triplicate samples, and the results shown are representative of three independent experiments. Treatments with different letters are significantly different from one another at p < 0.05.
The HMV phenotype of K. pneumoniae has been associated with the presence of virulence genes, magA and rmpA, and with serotypes K1, K2 and K5. The importance of the K antigen, magA and rmpA and HMV phenotype in K. pneumoniae virulence has been shown in a mouse model (Simoons-Smit et al., 1986; Lin et al., 2011; Fung et al., 2011), but as in vivo challenges have not been conducted in AGM, a direct comparison between species is not possible. In mice, magA+/rmpA+ and magA/rmpA+ K. pneumoniae isolates were found to be significantly more virulent than magA, rmpA, and an unencapsulated isogenic K1 mutant (DK1, magA+, rmpA+) isolate (Fung et al., 2011). magA+/ rmpA+ and magA/rmpA+ K. pneumoniae were significantly more invasive and virulent in vivo, and presented significantly different cytokine and chemokine responses than magA/rmpA isolates (Fung et al., 2011). Cytokines in the liver and serum from mice infected with low-doses (
