Diabetes mellitus is a documented risk factor for melioidosis, a tropical infection caused by Burkholderia pseudomallei. The increased susceptibility of diabetic individuals to infections with other pathogens has been associated with immune dysregulation. However, the impact of diabetes on the functional responses of dendritic cells (DC) and macrophages during B. pseudomallei infection has not been investigated. This study compared the responses of macrophages and DC towards B. pseudomallei using bone marrow-derived DC (BMDC) and peritoneal elicited macrophages (PEM) isolated from streptozotocin-induced diabetic C57BL/6 mice exhibiting hyperglycaemia for 9 days (acute) or 70 days (chronic) and age-matched nondiabetic C57BL/6 mice. Following coincubation of BMDC and PEM with a highly virulent B. pseudomallei isolate, maturation, bacterial internalization plus intracellular survival and cytokine gene expression profiles were assessed. No significant differences in functional responses of BMDC or PEM isolated from acute diabetic and nondiabetic mice were observed. However, significant differences in BMDC and PEM function were observed when chronic diabetic and nondiabetic mice were compared. This study demonstrates that diabetic mice with extended periods of uncontrolled hyperglycaemia have impaired DC and macrophage function towards B. pseudomallei, which may contribute to the high susceptibility observed in clinical practice.
Melioidosis is an emerging tropical infectious disease caused by the Gram-negative soil saprophyte, Burkholderia pseudomallei. Both the incidence and the mortality rate of melioidosis are high, even in Australia, despite early diagnosis and therapeutic intervention (Cheng & Currie, 2005; Malczewskie et al., 2005). Diabetes mellitus is the most commonly associated risk factor for melioidosis with 40–60% of patients having diabetes mellitus as a comorbidity (Suputtamongkol et al., 1999; Malczewskie et al., 2005).
Diabetic subjects are considered to have dysregulated immune responses, increasing their risk of developing infection (Marhoffer et al., 1992; Currie et al., 2000). Currently, there is limited information on how the diabetic status of an individual impacts on immune responses towards B. pseudomallei infection. It has been demonstrated that in response to B. pseudomallei, neutrophils are rapidly recruited and play a crucial role in early cytokine production within sites of infection (Easton et al., 2007; Chanchamroen et al., 2009). Chanchamroen (2009) recently demonstrated that B. pseudomallei internalization and migration by neutrophils isolated from diabetic patients with very poor glycaemic control were significantly impaired, compared with neutrophils from healthy controls. Other innate immune cells, including dendritic cells (DC) and macrophages, play an important role in the development of immunity to B. pseudomallei (Leakey et al., 1998; Elvin et al., 2006; Barnes et al., 2008; Williams et al., 2008; Chanchamroen et al., 2009; Charoensap et al., 2009). The C57BL/6 mouse is partially resistant to B. pseudomallei infection, and is widely used as a model for chronic melioidosis (Leakey et al., 1998). Comparison of C57BL/6 mice and susceptible BALB/c mice has demonstrated differences in the functional responses of both macrophages and DCs when stimulated with B. pseudomallei (Leakey et al., 1998; Williams et al., 2008). However, it is not known what impact diabetes has on the functional response of these cells towards B. pseudomallei.
Streptozotocin has been extensively used in experimental animal models to chemically induce diabetes by causing pancreatic β cell cytotoxicity (Lu et al., 1998; Szkudelzki, 2001; Srinivasan & Ramarao, 2007). Therefore, using nondiabetic and streptozotocin-induced diabetic C57BL/6 mice, this study investigated how diabetic status and persistence of hyperglycaemia affected the ability of bone marrow-derived DCs (BMDC) and peritoneal elicited macrophages (PEM) to respond to stimulation with B. pseudomallei.
Materials and methods
Eight- to 12-week-old male C57BL/6 mice from James Cook University Small Animal Breeding Facility were used. Institutional ethical approval for all animal experimentation was granted by the James Cook University Animal Ethics Committee (A987 and A1248).
Streptozotocin-induced diabetes in C57BL/6 mice
To induce diabetes, mice received intraperitoneal injections of streptozotocin (Sigma) dissolved in sodium citrate buffer (pH 4.5) at a dose of 55 mg kg−1 for 6 days (days 0–5). Nondiabetic control animals received intraperitoneal injections of sodium citrate buffer only. On day 10, peripheral blood was drawn from the lateral tail vein and blood glucose levels were assessed using an Ascensia Esprit 2 Blood Glucose Monitor. Mice with a blood glucose level of ≥13 mmol L−1 were considered diabetic and sacrificed following uncontrolled hyperglycaemia for 9 days (acute; sacrificed on day 19 postinitial streptozotocin injection; total mice=15, n=5 × 3 experiments) or 70 days (chronic; sacrificed on day 80 postinitial streptozotocin injection; total mice=16, n=8 × 2 experiments). Male mice were used because female C57BL/6 mice did not develop hyperglycaemia in a consistent rate following streptozotocin treatment (data not shown).
Isolation of BMDC
Isolation of BMDC from mouse femurs (n=3 per treatment group) was carried out using a method described previously (Williams et al., 2008). Briefly, bone marrow cells from each animal were cultured separately in BMDC medium containing supplements including 10% supernatant from Ag8653 myeloma cells transfected with the gene encoding murine granulocyte macrophage colony-stimulating factor. Lutz (1999) demonstrated previously that on day 10 of culture ∼90% of nonadherent cells are BMDC. On day 10 of culture, nonadherent BMDC were washed and seeded into 96-well Nunc culture plate (5 × 104 BMDC per well, In Vitro Technologies P/L) for fluorescence-activated cell sorting analysis or 24-well Nunc culture plates (1 × 106 BMDC per well, In Vitro Technologies P/L) for internalization and cytokine assays.
Isolation of PEM
To isolate PEM, mice (n=5 per treatment group) were injected intraperitoneally with 2.5 mL of Brewer's thioglycollate medium (BD Biosciences) using previously described protocols (Barnes & Ketheesan, 2007). PEM from each mouse within an experimental group were pooled, washed and seeded in triplicate into 24-well plates (1 × 106 PEM per well, In Vitro Technologies P/L). Cells were left to adhere overnight at 37 °C in 5% carbon dioxide (CO2). To remove nonadherent cells, monolayers were washed twice with phosphate-buffered saline (PBS) (pH 7.2), then fresh culture media was added to PEM.
Burkholderia pseudomallei isolates
A clinical B. pseudomallei isolate of high virulence (NCTC 13178) was grown to log phase in tryptone soya broth and used to stimulate BMDC and PEM at a multiplicity of infection of 1 : 1 (Ulett et al., 2000; Barnes & Ketheesan, 2007; Williams et al., 2008). Co-cultures were incubated for 4 h in 5% CO2 at 37 °C. Extracellular bacteria were then removed by adding kanamycin (250 µg mL−1; Sigma). Burkholderia pseudomallei isolates have varied sensitivity or resistance to kanamycin. Because this isolate, NCTC 13178, is sensitive to kanamycin, we and other previously published studies have used kanamycin in standard antibiotic protection assays (Utaisincharoen et al., 2001; Feterl et al., 2008; Williams et al., 2008).
Flow cytometry analysis of BMDC maturation
After 10 days of culture, the ex vivo-expanded BMDC were left untreated, infected with B. pseudomallei or stimulated with Pseudomonas aeruginosa lipopolysaccharide (50 ng mL−1– positive maturation control). After 22 h, BMDC were harvested and fluorescently labelled with fluorescein isothiocyanate-conjugated anti-CD11c (DC cell-surface marker, BD Biosciences), Phycoerthyrin (PE)-conjugated anti-CD86 (DC maturation marker, BD Biosciences) and Peridinin-chlorophyll proteins (PerCP)-conjugated anti-CCR7 (DC maturation marker, BD Biosciences). Analysis of labelled BMDC was performed using a FACScan™ with cell-quest software (version 5.2, BD Biosciences). Cells positive for CD11c were analysed for the expression of CD86 and CCR7. Mature DC had the phenotype CD11c+CD86hiCCR7hi. The percentage increase in DC maturation following treatment with B. pseudomallei (or lipopolysaccharide) was calculated as %CD11c+CD86hiCCR7hi treated−%CD11c+CD86hiCCR7hi untreated.
Assessment of Fluoresbrite® YG microsphere internalization
Fluoresbrite® YG microspheres, 2 µm (Polysciences), were added to BMDC and PEM cultures at a ratio of cells: beads of 1 : 1. After 4 h, cells were removed and fixed in ice-cold 4% paraformaldehyde for 20 min. Cells were then pelleted (300 g, 8 min, 4 °C), resuspended in 2 mL PBS (pH 7.2), then layered over 5 mL PEG8000 (20% w/v in water) and centrifuged (2000 g, 12 min, 4 °C) to remove extracellular microspheres. The cell pellet was then resuspended in standard azide buffer and analysed using a FACScan™ with cell-quest software (version 5.2, BD Biosciences).
Assessment of internalization and survival of B. pseudomallei
To determine phagocytic and bactericidal capacity of BMDC and PEM, internalization and survival were assessed using protocols described previously (Williams et al., 2008). Briefly, bacterial uptake by BMDC was assessed after 4-h incubation with B. pseudomallei [%uptake=internalized bacteria (CFU) at 4 h/inoculating dose × 100]. At this time point, kanamycin (250 µg mL−1) was added to parallel co-cultures to remove extracellular bacteria. Bacterial survival within BMDC and PEM was assessed 1 and 18 h after the addition of kanamycin [%bacterial survival=surviving bacteria (CFU) at 1 or 18 h/internalized bacteria (CFU) at 4 h × 100]. To enumerate intracellular B. pseudomallei at each time point, BMDC and PEM were washed twice, then treated with 0.1% Triton-X to disrupt cells and release the internalized B. pseudomallei. CFUs were determined from serial dilutions of lysed cell suspensions, plated in triplicate on Ashdown agar and incubated for 48 h at 37 °C.
Assessment of cytokine gene expression
To determine the changes in cytokine gene expression by BMDC and PEM in response to B. pseudomallei, total RNA was extracted from BMDC and PEM 18 h after the removal of extracellular B. pseudomallei using a commercial RNA extraction kit (RBC) according to the manufacturer's instructions. DNase 1 (Fermentas) was used to digest any contaminating genomic DNA. Total RNA (1–2 µg) was used to synthesize first-strand cDNA with the aid of an anchored oligo(dT) primer (Sigma) and Revert Aid M-MuLV reverse transcriptase enzyme (Fermentas) and stored at −80 °C until required. Standard reverse transcription-PCR reactions were set up to assess the cytokine gene expression of interleukin-10 (IL-10), IL-12 and IL-18 for BMDC and IL-1β, IL-12 and tumor necrosis factor-α (TNF-α) for PEM (Table 1). IL-12 and IL-18 were selected as they are classical cytokines known to drive strong production of interferon-γ (IFN-γ) and differentiation of T cells into Th1 cells. The expression of IL-10 was also assessed in BMDC as it has a suppressive effect, and when expressed while DC present antigen to naïve T cells drives the differentiation of adaptive regulatory T cells (de Jong et al., 2005). Expressions of IL-1β and TNF-α PEM were selected because they are important proinflammatory cytokines produced by macrophages that promote antimicrobial activity. Furthermore, hyperproduction of these cytokines can impact negatively on the host during B. pseudomallei infection (Ulett et al., 2000; Koo & Gan, 2006). Macrophages are also a potential source for IL-12 production during early phases of B. pseudomallei infection (Ulett et al., 2000). Following electrophoresis, PCR products were analysed with genesnap and genetools software, to quantify the mRNA ratio of each cytokine standardized against β-actin.
|Interleukin 1β (166 bp)||5′-TGTGAAATGCCACCTTTTGA-3′|
|Interleukin 10 (254 bp)||5′-ACCAAAGCCACAAAGCAG-3′|
|Interleukin 12p40 (128 bp)||5′-AGACCCTGCCCATTGAACTG-3′|
|Interleukin 18 (504 bp)||5′-TTATTGACAACACGCTTTAC-3′|
|Tumor necrosis factor-α (111 bp)||5′-AAGAGGCACTCCCCCAAAAG-3′|
|Interleukin 1β (166 bp)||5′-TGTGAAATGCCACCTTTTGA-3′|
|Interleukin 10 (254 bp)||5′-ACCAAAGCCACAAAGCAG-3′|
|Interleukin 12p40 (128 bp)||5′-AGACCCTGCCCATTGAACTG-3′|
|Interleukin 18 (504 bp)||5′-TTATTGACAACACGCTTTAC-3′|
|Tumor necrosis factor-α (111 bp)||5′-AAGAGGCACTCCCCCAAAAG-3′|
Significance between groups was analysed by univariate anova using the spss statistic program (version 17). Groups were considered significantly different when a P value ≤0.05 was returned.
Maturation of BMDC in the presence of B. pseudomallei
Mature BMDC were defined as CD11c+CD86hiCCR7hi. An increase in mature BMDC isolated from both diabetic and nondiabetic mice was found in response to stimulation with NCTC 13178. No difference in maturation was observed between BMDC isolated from acute diabetic mice and nondiabetic mice (Fig. 1). Maturation of BMDC isolated from chronic diabetic mice was reduced compared with BMDC isolated from nondiabetic mice; however, this trend was not statistically significant. Similar trends in maturation were observed for chronic-derived BMDC stimulated with either B. pseudomallei or lipopolysaccharide (50 ng mL−1) (data not shown).
Internalization capacity of BMDC and PEM
To ensure the functionality of BMDC and PEM, the ability to internalize Fluoresbrite® YG microspheres was assessed. After 4 h, the percentage of BMDC and PEM containing Fluoresbrite® YG microspheres was found to be similar for cells from both acute diabetic and nondiabetic mice (data not shown). Similarly, chronic diabetes of the host did not impair the ability of BMDC and PEM to internalize Fluoresbrite® YG microspheres, with no difference observed between cells derived from chronic diabetic and nondiabetic mice (Fig. 2a). The ability to internalize B. pseudomallei was similar for BMDC and PEM isolated from acute diabetic and nondiabetic mice (data not shown). However, internalization of B. pseudomallei was significantly decreased in chronic diabetic-derived BMDC (3.09% or 0.6 × 104 CFU) compared with non-diabetic-derived BMDC (5.88% or 1.3 × 104 CFU; P<0.05, Fig. 2b). No significant difference was observed between chronic diabetic (7.95% or 3.0 × 104 CFU) and non-diabetic-derived PEM (8.80% or 3.3 × 104 CFU; Fig. 2b).
Bacterial survival within diabetic and non-diabetic-derived BMDC and PEM
To determine bacterial survival, the percentage of internalized B. pseudomallei surviving within BMDC and PEM was assessed after 1 and 18 h. After both time points, intracellular survival of B. pseudomallei within acute diabetic-derived BMDC was similar to that within non-diabetic-derived BMDC (1 h, data not shown), with <1% of internalized B. pseudomallei remaining after 18 h (0.16% or 9.7 × 101 CFU acute diabetic; 0.08% or 7.7 × 101 CFU nondiabetic; Fig. 2c). In contrast, survival of B. pseudomallei was increased in chronic diabetic-derived BMDC compared with nondiabetic BMDC after 1 h (data not shown) and 18 h (2.22% or 12.8 × 101 CFU and 0.5% or 6.7 × 101 CFU, respectively; P<0.05, Fig. 2c). Similar trends were observed for B. pseudomallei infected PEM isolated from diabetic and nondiabetic animals. Intracellular survival of B. pseudomallei within acute diabetic-derived PEM was equivalent to that within non-diabetic-derived PEM at both 1 h (data not shown) and 18 h time points (0.37% or 2.0 × 102 CFU and 0.26% or 1.8 × 102 CFU, respectively; Fig. 2d). However, for PEM isolated from chronic diabetic mice, the intracellular survival of B. pseudomallei was increased compared with non-diabetic-derived PEM at both 1 h (data not shown) and 18 h time points (2.35% or 6 × 102 CFU and 0.82% or 2.6 × 102 CFU, respectively; Fig. 2d).
Cytokine gene expression by BMDC
In response to B. pseudomallei, the gene expression of IL-12, IL-18 and IL-10 was significantly increased in BMDC isolated from both acute diabetic and nondiabetic mice when compared with uninfected BMDC (Fig. 3a, c and e). While no differences in the gene expression of IL-12, IL-18 and IL-10 were observed between acute diabetic and non-diabetic-derived BMDC (Fig. 3a, c and e), significant differences in gene expression were observed between chronic diabetic and non-diabetic-derived BMDC following exposure to B. pseudomallei (Fig. 3b, d and f).
When stimulated with B. pseudomallei, the gene expression of IL-12 by chronic diabetic-derived BMDC was significantly decreased compared with non-diabetic-derived BMDC, and significantly decreased compared with that expressed by uninfected chronic diabetic-derived BMDC (Fig. 3b). However, when stimulated with lipopolysaccharide, the gene expression of IL-12 was similar for both chronic diabetic and non-diabetic-derived BMDC (Fig. 3b). Like IL-12, significant differences in the gene expression of IL-18 was observed between chronic diabetic and non-diabetic-derived BMDC when stimulated with B. pseudomallei, but not when stimulated with lipopolysaccharide (Fig. 3d). The expression of IL-10 by chronic diabetic-derived BMDC stimulated with B. pseudomallei was similar to B. pseudomallei-stimulated non-diabetic-derived BMDC, and was significantly higher than that expressed by unstimulated chronic diabetic-derived BMDC (Fig. 3f). A similar gene expression of IL-10 was observed when chronic diabetic and non-diabetic-derived BMDC with either B. pseudomallei or lipopolysaccharide (50 ng mL−1) (Fig. 3f).
Cytokine gene expression by PEM
The gene expression of IL-1β, IL-12 and TNF-α was significantly increased in acute diabetic and non-diabetic-derived PEM exposed to B. pseudomallei when compared with uninfected PEM (Fig. 4a, c and e). Similar levels of gene expression were observed for IL-1β and IL-12 in both acute diabetic and non-diabetic-derived PEM (Fig. 4a and c). The gene expression of TNF-α by acute diabetic-derived PEM stimulated with B. pseudomallei was significantly decreased compared with non-diabetic-derived PEM stimulated with B. pseudomallei (P<0.05, Fig. 4e). However, TNF-α gene expression by acute diabetic-derived PEM stimulated with B. pseudomallei was still significantly higher than that expressed by uninfected acute diabetic-derived PEM.
The gene expression of IL-1β, IL-12 and TNF-α was also significantly increased in chronic diabetic and non-diabetic-derived PEM co-cultured with B. pseudomallei when compared with uninfected PEM (Fig. 4b, d and f). However, the level of IL-1β, IL-12 and TNF-α gene expression by chronic diabetic-derived PEM stimulated with B. pseudomallei was significantly lower than that expressed by non-diabetic-derived PEM stimulated with B. pseudomallei. Non-diabetic-derived PEM demonstrated significantly higher gene expression of IL-1β, IL-12 and TNF-α when stimulated with B. pseudomallei compared with when stimulated with lipopolysaccharide (Fig. 4b, d and f). PEM isolated from chronic diabetic and nondiabetic hosts induced similar levels of IL-1β, IL-12 and TNF-α gene expression in response to lipopolysaccharide stimulation.
In countries endemic for melioidosis, up to 60% of patients also have diabetes mellitus (Suputtamongkol et al., 1999; Cheng & Currie, 2005; Malczewskie et al., 2005; Kandasamy & Norton, 2008). Recently, Chanchamroen (2009) demonstrated that in response to B. pseudomallei, internalization and migration function of neutrophils isolated from diabetic individuals was significantly impaired compared with neutrophils isolated from healthy controls. The importance of both DC and macrophage function in response to B. pseudomallei infection has been demonstrated (Leakey et al., 1998; Williams et al., 2008). However, to date no studies have investigated the function of DCs or macrophages from diabetes in response to B. pseudomallei. Therefore, our study compared BMDC and PEM isolated from streptozotocin-induced diabetic C57BL/6 mice exhibiting hyperglycaemia for 9 days (acute diabetic) or 70 days (chronic diabetic) to age-matched, nondiabetic C57BL/6 mice.
Our findings indicate that a short hyperglycaemic period did not significantly alter BMDC and PEM function in response to B. pseudomallei. The importance of the length of time between induction of experimental diabetes and investigation of macrophage function from nondiabetic and streptozotocin-induced diabetic hosts has been recently reported (Ma et al., 2008). Only mice with chronic hyperglycaemia (112 days) had significantly reduced numbers of splenic and peritoneal macrophages with a significantly reduced capacity to phagocytose allogenic BALB/c T cells compared with macrophages isolated from nondiabetic controls (Ma et al., 2008). In the present study, chronic diabetic mice experienced uncontrolled hyperglycaemia for 70 days before the isolation of BMDC and PEM. In response to stimulation with a highly virulent B. pseudomallei isolate, the current study has demonstrated significant differences in the functional responses of BMDC and PEM isolated from chronic diabetic C57BL/6 mice compared with those from nondiabetic mice.
Previous studies have demonstrated that in response to heat-killed B. pseudomallei, BMDC undergo maturation to stimulate T-cell proliferation (Elvin et al., 2006; Healey et al., 2006). Furthermore, DCs cultured from healthy donor peripheral blood mononuclear cells, matured in the presence of paraformaldehyde-fixed B. pseudomallei and stimulated a Th1-biased T-cell differentiation (Charoensap et al., 2008). In the current study, co-culture with B. pseudomallei induced a similar maturation response in both diabetic and non-diabetic-derived BMDC. Phenotypically mature BMDC (CD11chi and CD86hi) also had high expression of the chemokine receptor, CCR7, which promotes migration of DC to secondary lymphoid organs. Our results indicate that in response to stimulation with B. pseudomallei, BMDC isolated from both diabetic and nondiabetic C57BL/6 mice undergo similar upregulation of important maturation markers that are necessary for migration to secondary lymphoid organs, and for activation of T-cell responses.
We have demonstrated previously differences in DC maturation in response to B. pseudomallei between susceptible (BALB/c) and partially resistant (C57BL/6) mouse strains (Williams et al., 2008). In that study, following co-culture with B. pseudomallei, BMDC isolated from BALB/c mice demonstrated significantly increased maturation response in comparison with C57BL/6-derived BMDC. However, this increased phenotypical maturation response in BALB/c-derived BMDC was linked to decreased bactericidal function. In the current study, chronic diabetes did not impair BMDC and PEM internalization of Fluoresbrite® YG microspheres. Interestingly, internalization of B. pseudomallei by chronic diabetic-derived BMDC was significantly inferior to that internalized by non-diabetic-derived BMDC, although this was not seen for chronic diabetic-derived PEM. When the level of microsphere internalization was compared with B. pseudomallei internalization, our observations indicate that there is a general decreased capacity to internalize B. pseudomallei by both chronic diabetic and non-diabetic-derived BMDC and PEM. Similar findings have been reported for human neutrophils where, compared with other intracellular bacteria (Salmonella enterica serovar Typhimurium and Escherichia coli), significantly less B. pseudomallei is internalized (Chanchamroen et al., 2009). The findings of the current study suggest that B. pseudomallei may interfere with internalization by BMDC and PEM and that this reduced capacity to internalize B. pseudomallei is further diminished in BMDC isolated from chronic diabetic hosts.
Macrophages play an important role during the early phases of B. pseudomallei infection and are thought to contribute to determining innate susceptibility or resistance of the host. Unlike macrophages, the main effector function of DCs is not bacterial clearance, rather highly specialized antigen presentation that provides potent activation of naïve T cells (Charoensap et al., 2008). Recently, Charoensap (2008) reported similar growth kinetics of B. pseudomallei within human monocyte-derived DCs and macrophages. In the current study, we observed that chronic diabetic-derived BMDC were efficient at killing internalized B. pseudomallei compared with non-diabetic-derived BMDC. While the percentage of bacteria surviving in both chronic and non-diabetic-derived BMDC was low, it was significantly higher in chronic diabetic-derived BMDC. It is possible that DC facilitate the dissemination of B. pseudomallei, and consequently chronic diabetic-derived BMDC could traffic a higher bacterial burden to secondary lymphoid organs. Given the virulence of this organism, the low number of bacteria surviving within these DC is important. Previous studies have shown that as few as 1 × 103 CFU intranasally or 5 × 103 CFU intravenously of this B. pseudomallei isolate, NCTC 13178, causes fatality in 50% of C57BL/6 mice (Barnes & Ketheesan, 2005). Burkholderia pseudomallei survival was also higher in PEM isolated from chronic diabetic mice compared with those from nondiabetic mice. The reduced ability of chronic diabetic-derived BMDC and PEM to clear B. pseudomallei during the early phases of infection may suggest that infection may be established in chronic diabetic hosts following a lower inoculating dose.
Cytokine profiles generated by different subsets of DCs drive the polarization of either Th1, Th2 or regulatory T cells, thus providing pathogen-specific responses (de Jong et al., 2005). DCs readily secrete IL-12 in response to various pathogens, including bacteria, stimulating IFN-γ production by naïve Th cells promoting Th1 polarization and responses. The expression of IL-18 by DCs on its own stimulates weak IFN-γ expression by CD4+ T cells. Combined IL-12 and IL-18 expression by DCs drives strong IFN-γ expression due to the presence of IL-12 upregulating IL-18 receptor complexes (de Jong et al., 2005). An increased expression of IL-12 and IL-18 by B. pseudomallei-stimulated nondiabetic BMDC suggests a capacity to polarize a Th1-type response to B. pseudomallei (de Jong et al., 2005). In contrast, the expression of IL-12 and IL-18 by chronic diabetic-derived BMDC in response to B. pseudomallei was significantly decreased compared with non-diabetic-derived BMDC. In addition, the expression of IL-10 was increased in both B. pseudomallei-stimulated chronic and non-diabetic-derived BMDC. IL-10, which can negatively regulate Th1 and Th2 polarization, has also been shown to inhibit full maturation of DCs and downregulate the expression of major histocompatibility complex class II molecules (de Jong et al., 2005). The expression of IL-10, coupled with significantly decreased IL-12 and IL-18 expression suggests that chronic diabetic-derived BMDC may not provide adequate stimulation for development of protective Th1 responses in the presence of B. pseudomallei.
Proinflammatory cytokine gene expression profiles for IL-1β, IL-12 and TNF-α were also significantly decreased in B. pseudomallei-stimulated chronic diabetic-derived PEM, compared with non-diabetic-derived PEM. During the early stages of B. pseudomallei infection, the cytokine gene profiles expressed in response to B. pseudomallei also play an important role in the development of susceptibility or resistance (Ulett et al., 2000; Koo & Gan, 2006). In this study, we have observed the hypoproduction of proinflammatory cytokines IL-1β and TNF-α and the decreased expression of the Th1 polarizing cytokine IL-12 by PEM isolated from chronic diabetic C57BL/6 mice. This supports the reduced bactericidal capacity observed for chronic diabetic-derived PEM against B. pseudomallei and may contribute to dysregulated adaptive immune responses against B. pseudomallei.
The suppressor of cytokine signalling (SOCS) proteins have been implicated as potential effectors driving the dysregulated cytokine profiles observed in type 2 diabetes (T2D) (Pradhan et al., 2001; Spranger et al., 2003; Ueki et al., 2004; O'Connor et al., 2007). A study by Ueki (2004) demonstrated that obesity and lipopolysaccharide-induced endotoxaemia correlated with an increased expression of SOCS3. Furthermore, SOCS3 was able to cause insulin resistance by interfering with tyrosine phosphorylation of insulin receptor substrate (IRS)-1 and IRS-2. Similarly, O'Connor (2007) demonstrated a constitutive overexpression of SOCS3 by macrophages isolated from the peritoneum of T2D mice. Such research provides evidence that by acting as negative regulators of hormone/cytokine signalling, SOCS proteins are providing a link between T2D and altered cytokine function (Ueki et al., 2004; O'Connor et al., 2007). The increased expression of SOCS3, is a mechanism that is used by B. pseudomallei to enable intracellular survival (Ekchariyawat et al., 2005). By activating the expression of SOCS3, B. pseudomallei inhibits the expression of inducible nitric oxide synthase and IFN-γ to escape killing by normal macrophages. It is therefore possible that SOCS3 is being overexpressed in our chronic diabetic animals, aiding the ability of B. pseudomallei to further suppress the antimicrobial activity and the cytokine expression of BMDC and PEM isolated from chronic diabetic animals compared with nondiabetic animals.
The current study provided valuable information regarding the function of BMDC and PEM isolated from diabetic mice in response to B. pseudomallei. Acute hyperglycaemia did not affect BMDC and PEM function in response to B. pseudomallei. However, BMDC and PEM isolated from C57BL/6 mice with chronic diabetes, were less efficient at internalizing and killing B. pseudomallei and generated cytokine gene expression profiles that may not favour the development of Th1-type immune response. Additionally, proinflammatory cytokines that contribute to host resistance against B. pseudomallei infection were reduced in chronic diabetic-derived PEM. These results provide further evidence that chronic hyperglycaemia impairs innate immune responses to B. pseudomallei. Further investigations are warranted to understand the effect of diabetes on the innate and adaptive immune responses to B. pseudomallei.
We acknowledge the contribution of Associate Professor Leigh Owens with the statistical analysis of data.