Introduction

Melioidosis, a potentially fatal disease seen predominantly in South-East Asia and northern Australia is caused by the facultative intracellular bacterium, Burkholderia pseudomallei. Transmission is most likely to occur following exposure of cuts or abrasions to contaminated soil or surface water.^{1, 2} There is broad variation in the clinical presentations, severity and duration of illness in patients infected with B. pseudomallei. Disease manifestations range from chronic pulmonary infection to fatal septicaemia.^{3, 4} While it is recognized that the severity of the disease falls along a continuous spectrum, clinical presentations are often categorized into acute, subacute and chronic forms of disease.

Characterization of a BALB/c–C57BL/6 mouse model of the acute and chronic forms of human melioidosis has facilitated recent studies on the pathogenesis of B. pseudomallei infection in vivo.^{5, 6, 7} BALB/c mice are highly susceptible to infection with B. pseudomallei and provide a suitable model with which to study the acute septicaemic form of human melioidosis. Burkholderia pseudomallei infection in BALB/c mice induces abscessation in spleen and liver, and a rapidly progressing bacteraemia resulting in death within 4 days.⁷ In this model of acute melioidosis, a massive production of proinflammatory cytokines at days 2 and 3 postinfection is thought to contribute to the host's death.⁵ Following i.v. inoculation with 37 c.f.u. of virulent B. pseudomallei, bacterial loads in liver and spleen of BALB/c mice increase exponentially, reaching 10⁸ c.f.u./mL and 10⁷ c.f.u./mL, respectively, prior to the host's death by day 4.⁷ In contrast, C57BL/6 mice are relatively resistant to infection, as reflected by the absence of bacteraemia in these mice. Resistance of C57BL/6 mice is not complete, however, because persistent bacterial loads can be demonstrated in liver and spleen at day 3 postinfection.⁷ However, these bacterial loads are typically 100- to 1000-fold fewer than those in BALB/c mice.⁷ Steadily increasing bacterial loads in spleen and liver of C57BL/6 mice from day 3 postinfection lead to a fatal outcome after several weeks.^{5, 7}

The mechanisms leading to the development of either acute or chronic melioidosis are yet to be elucidated. Studies on several intracellular bacteria have demonstrated that innate host resistance involves the mobilization of specific subsets of leucocytes to the inflammatory site.^{8, 9} In several models of intracellular infection, chemokines and CSF have been shown to regulate these inflammatory processes.^{8, 10} Chemokines are a family of secreted proteins with strong chemotactic properties for particular inflammatory cell types, making them integral components of the initial stages of an inflammatory response. Many chemokines are also involved in activation of the leucocyte populations they attract.^{11, 12} Colony stimulating factors induce the proliferation and differentiation of granulocytes, monocytes and other related haematopoietic cells. They are also important regulators of the functional activity of many mature cell types, including macrophages and neutrophils.^{13, 14}

Development of a cell-mediated immune response to infection with B. pseudomallei is considered to be critical for the survival of the host.¹⁵ Recent studies in patients with melioidosis have demonstrated that elevated levels of the chemokines interferon--inducible protein 10 (IP-10) and monocyte interferon--inducible protein (Mig) correlate with disease severity and clinical outcome.¹⁶ However, the kinetics and magnitude of chemokine and CSF responses in acute versus chronic B. pseudomallei infection are unknown. We have therefore used reverse transcriptase–polymerization chain reaction (RT-PCR) and histology to assess the production of chemokine and CSF mRNA during infection and to monitor changes in cellular composition in the livers and spleens of BALB/c and C57BL/6 mice infected with virulent B. pseudomallei.

Materials and Methods

Mice

Experiments were performed with 8- to 16-week-old BALB/c and C57BL/6 mice purchased from the James Cook University Small Animal Breeding Unit (Townsville, Queensland, Australia). Animal experiments were carried out according to the guidelines of the National Health and Medical Research Council (NH&MRC), with ethics approval from the James Cook University Animal Ethics Committee (A536).

Burkholderia pseudomallei

A highly virulent strain of B. pseudomallei (NCTC 13178), which was originally isolated from a patient admitted to the Townsville General Hospital, was used in the study. Identification of the isolate was made using API 20NE (bioMerieux, La Balme, France) in addition to colonial morphology on Ashdown agar. The LD₅₀ of this isolate was shown to be <10 c.f.u. in BALB/c mice in a previous study.¹⁷

Inoculation of bacteria and determination of bacterial load

Mice were infected i.v. with 25 c.f.u. of B. pseudomallei suspended in sterile PBS via the lateral tail vein. Control mice received sterile PBS only. At days 0, 1, 2, 3, 4, 7 and 14 following inoculation, five mice were killed with CO₂ and their livers and spleens excised. For BALB/c mice, samples were obtained at days 0, 1, 2 and 3 because all mice succumbed to infection by day 4 of postinfection. Samples were bisected aseptically and a half immediately wrapped in aluminium foil, immersed in liquid nitrogen and stored at –80°C until required for RNA extraction.⁵ Bacterial load was assessed in the remaining organ half. Tissue was homogenized in sterile PBS and serial dilutions were plated onto Ashdown agar for colony counts, as described elsewhere.⁷ The detection limit of bacteria ranged between 2 c.f.u./mL and 2 10⁹ c.f.u./mL of organ homogenate. Statistical analysis of bacterial load data between BALB/c and C57BL/6 mice at individual time points was performed using two-tailed Student's t-test for independent samples within the SPSS (version 8.0) software package. Data are expressed as the means SEM (n = 5). P-values < 0.01 were considered significant.

Reverse transcription–polymerization chain reaction for chemokine expression

Tissue was homogenized and total RNA was extracted from the livers and spleens as described elsewhere.^{5, 17} One microgram of total RNA was reverse-transcribed in a 20 L reaction mixture containing 0.5 g Oligo(dT)₁₅ (Promega, Annandale, NSW, Australia); 1 First Strand Buffer (50 mmol/L Tris-HCl (pH 8.3), 75 mmol/L KCl, 3 mmol/L MgCl₂; Gibco-BRL Life Technologies, Melbourne, Vic., Australia); 10 mmol/L dithiothreitol (dTT; Gibco-BRL Life Technologies); 500 mol/L deoxynucleoside triphosphate (dNTP) mix (Promega); and 200 U of Superscript II™ reverse transcriptase (Gibco-BRL Life Technologies). DNase was used to remove genomic DNA from total RNA.¹⁸

Two microlitres of cDNA were amplified with Taq DNA polymerase (1 U; Promega) for 30 cycles (initial step of 95°C for 2 min, then denaturation at 94°C for 50 s, primer annealing at 60°C for 50 s, and primer extension at 72°C for 1 min). In preliminary experiments, 30 cycles was shown to lie in the linear portion of the curve for the amount of PCR product produced. Sense and antisense primers for the chemokines tested were taken from published data^{19, 20, 21, 22, 23} (Table 1) and were synthesized by Gibco-BRL Life Technologies. The optimal MgCl₂ concentration for cytokine-specific primer pairs ranged from 1 mmol/L to 4.5 mmol/L. Primers were used at a final concentration of 1 mol/L in a 100 L PCR mixture containing 200 mol/L dNTP mix (Promega); 1 PCR buffer (1 mmol/L Tris-HCl (pH 9.0), 5 mmol/L KCl, 0.01% Triton X-100; Promega); and 1–4.5 mmol/L MgCl₂ (Promega). For each cDNA sample, duplicate PCR reactions were prepared using -actin primers instead of cytokine primers as an internal positive control standard. Negative controls using water instead of cDNA were also included in each run. Non-reverse transcribed total RNA for each sample was subjected to PCR amplification using -actin primers to confirm that amplification was based solely on cDNA. Polymerization chain reaction-assisted mRNA amplification was performed for three separately prepared cDNA samples.

Table 1 - Chemokine- and colony stimulating factor (CSF)-specific primer pair sequences used in this study.

Full table

Polymerization chain reaction products were electrophoresed in 2.0–3.0% agarose gels, stained with 0.5 g/mL ethidium bromide and visualized with a UV transilluminator. A 123 base pair (bp) ladder (Gibco-BRL Life Technologies) was used as the DNA marker.

Histology

In a parallel experiment, another series of five C57BL/6 and BALB/c mice were inoculated i.v. with 25 c.f.u. of B. pseudomallei. Control mice received sterile PBS only. At days 1, 3 and 14 (for C57BL/ 6 mice only) mice were killed with CO₂ and their livers and spleens excised for histology. At day 3 only, organs were bisected aseptically and a half used to determine bacterial load as described earlier. The remaining tissue was processed for histological examination. Livers and spleens were fixed in 10% neutral-buffered formalin at room temperature for at least 48 h. After embedding in paraffin, 4 m sections were cut and stained with haematoxylin–eosin (H&E). Histological examination was carried out using light microscopy.

Results

Bacterial growth kinetics

Bacterial numbers in BALB/c liver and spleen increased rapidly, reaching approximately 1 10⁶ c.f.u. by day 3, prior to host death. In contrast, bacterial loads in the livers and spleens of C57BL/6 mice were approximately 1000-fold fewer than in BALB/c mice at day 3 (Figure 1). However, 14 days after infection, levels of B. pseudomallei recovered from C57BL/6 tissue were similar to those detected in BALB/c mice at day 3 (10⁶ c.f.u.).

Figure 1.

Bacterial load in the (a) spleens and (b) livers of () BALB/c and () C57BL/6 mice at days 1, 3 and 14 following i.v. inoculation with 25 c.f.u. of Burkholderia pseudomallei strain NCTC 13178. At various time intervals following infection, livers and spleens were excised and homogenized in PBS. Burkholderia pseudomallei numbers were determined by serial dilution on Ashdown agar. Results are shown as the mean bacterial load (log₁₀ c.f.u./mL) for five mice SEM (detection limit of 2 c.f.u./mL, therefore < 2 c.f.u./mL taken as 0 c.f.u./mL). A rapid increase in bacterial load in BALB/c livers and spleens was observed by day 3. In contrast, bacterial loads in C57BL/6 mice remained comparatively low until day 14 postinfection. *Significantly higher bacterial loads in BALB/c mice at day 3 postinfection (P < 0.01).

Full figure and legend (18K)

Detection of chemokine and colony stimulating factor mRNA

Patterns of induction of chemokines and CSF in the spleens and livers of C57BL/6 and BALB/c mice are illustrated in Figure 2 and 3, respectively. Constant levels of the housekeeping gene, -actin, were detected in all samples (Figure 2,3). These results confirm the semiquantitative nature of the method used and allow a comparison of chemokine- and CSF-specific mRNA responses during the course of experimental infection.

Figure 2.

(Previous page) Chemokine and CSF mRNA responses in spleens of BALB/c and C57BL/6 mice infected i.v. with 25 c.f.u. of Burkholderia pseudomallei. At various time intervals following infection (days 0–14), spleens were excised and total RNA was extracted. Genomic DNA was removed and cDNA was subjected to polymerization chain reaction using specific primers. Molecular weight markers (M) are shown in the left-hand lane of each gel. Data shown are representative of three mice at each time point. Constant levels of the house keeping gene, -actin, were detected in all samples. Maximal expression of mRNA for all chemokines and CSF investigated was demonstrated at day 3 in BALB/c mice and day 14 in C57BL/6 mice. IP-10, interferon--inducible protein 10; Mig, monocyte interferon--inducible protein; MCP-1, monocyte chemoattractant protein-1; KC, cytokine-induced neutrophil chemoattractant; MIP-2, macrophage inflammatory protein-2.

Full figure and legend (168K)

Figure 3.

Chemokine and CSF mRNA responses in livers of BALB/c and C57BL/6 mice infected i.v. with 25 c.f.u. of Burkholderia pseudomallei. At various time intervals following infection (days 0–14), livers were excised and total RNA was extracted. Genomic DNA was removed and cDNA was subjected to polymerization chain reaction using specific primers. Molecular weight markers (M) are shown in the left-hand lane of each gel. Data shown are representative of three mice at each time point. Constant levels of the house keeping gene, -actin, were detected in all samples. Maximal expression of mRNA for all chemokines and CSF investigated was demonstrated at day 3 in BALB/c mice and day 14 in C57BL/6 mice.

Full figure and legend (189K)

Patterns of chemokine and CSF induction were similar for liver and spleen, although responses were typically greater in spleen. Baseline levels of mRNA for the majority of chemokines and CSF in spleen were either not detected or were very low (Figure 2). However, constitutive expression of regulated upon activation, normal T-cell expressed and secreted (RANTES) chemokines was observed in both C57BL/6 and BALB/c mice. A slight increase in the expression of RANTES was apparent in spleen of both strains of mice during infection (Figure 2). A chemokine- or CSF-specific response to B. pseudomallei infection, as demonstrated by an increase in gene transcription, was clearly demonstrated because baseline levels of chemokine and CSF mRNA other than RANTES were minimal. With the exception of RANTES, increased amounts of mRNA for all chemokines and CSF were detected in the spleens of BALB/c mice by day 1 and 2 of infection (Figure 2). The level of induction of these chemokines and CSF continued to rise over 3 days prior to host death, with greatest expression observed at day 3 (Figure 2). In spleens of C57BL/6 mice, increased amounts of mRNA for all chemokines were also observed following infection (Figure 2). Increased expression was most pronounced for IP-10 and Mig. In contrast to BALB/c mice, levels of mRNA in spleens of C57BL/6 mice were greatest between days 4 and 7 of infection (Figure 2). In spleens of C57BL/6 mice, the high level of mRNA for IP-10 was maintained for the duration of the assay period (14 days).

The kinetics of induction for chemokine and CSF mRNA differed between the spleens of C57BL/6 and BALB/c mice. While both strains of mice showed evidence of increased induction of IP-10 and Mig at day 1, early increases in mRNA levels of macrophage CSF (M-CSF) and macrophage inflammatory protein-2 (MIP-2) mRNA were demonstrated only in BALB/c mice (Figure 2). Similarly at day 2, levels of the chemokines cytokine-induced neutrophil chemoattractant (KC), MIP-2 and monocyte chemoattractant protein-1 (MCP-1), M-CSF and granulocyte-macrophage CSF (GM-CSF) had increased in spleens of BALB/c but not of C57BL/6 mice (Figure 2). At day 3, for spleens of BALB/c mice, all chemokines and CSF investigated in the present study had increased markedly. In contrast, spleens of C57BL/6 mice demonstrated increased mRNA expression only for IP-10, Mig, RANTES, M-CSF, MIP-2 and KC (Figure 2).

Low baseline levels of mRNA for Mig, IP-10, KC and RANTES were detected in the livers of both BALB/c and C57BL/6 mice (Figure 3). Increased levels of mRNA for G-CSF, M-CSF and GM-CSF were not detected in livers of C57BL/6 mice (Figure 3). In contrast, in BALB/c mice, very slight increases in the level of mRNA for G-CSF, M-CSF and GM-CSF were observed at day 3 postinfection (data not shown). Increased levels of mRNA for IP-10, Mig, RANTES, MCP-1, KC and MIP-2 were detected in livers of both strains of mice following infection (Figure 3). However, as for spleen, the kinetics of induction differed between C57BL/6 and BALB/c mice. In livers of BALB/c mice, maximal expression of these chemokines and CSF occurred between days 2 and 3 of infection. In C57BL/6 mice, mRNA levels for the majority of chemokines and CSF were maximal between days 4 and 7 following infection, with the exception of Mig (day 3; Figure 3). A moderate increase in the level of mRNA for Mig persisted in the livers of C57BL/6 mice for the duration of the assay period (14 days; Figure 3).

In livers of C57BL/6 mice, increased mRNA expression of the neutrophil chemoattractant KC was the earliest detected (day 1; Figure 3). Both KC and IP-10 mRNA levels were increased at day 1 in livers of BALB/c mice, while expression of IP-10 mRNA increased by day 2 in livers of C57BL/6. At 2 days postinfection, increased mRNA expression of KC, IP-10, Mig, MIP-2 and MCP-1 was demonstrated in BALB/c livers. By day 3 in BALB/c livers, mRNA expression of all chemokines and CSF investigated had increased (Figure 3). In contrast, Mig was the only chemokine with increased levels of mRNA in livers of C57BL/6 mice at the same time point (day 3; Figure 3).

Patterns of chemokine and CSF expression in livers and spleens reflected bacterial loads (Figure 1). Levels of viable B. pseudomallei were greatest in spleen, as were chemokine and CSF responses. In spleens of BALB/c mice, induction of all chemokines and CSF investigated were maximal at day 3 postinfection, correlating with the highest bacterial numbers (Figure 1a,Figure 2). In spleens of C57BL/6 mice, while levels of viable B. pseudomallei were low initially (days 1 and 3), a dramatic increase was observed by termination of the study at 14 days (Figure 1a). Peak bacterial numbers in spleens of C57BL/6 mice (14 days; Figure 1a) reflected increasing expression of the majority of chemokines and CSF investigated (7–14 days; Figure 2). The exception was with mRNA expression for Mig, which was greatest at day 3 postinfection (Figure 2). Similar observations were made for viable B. pseudomallei and chemokine/CSF induction in livers of BALB/c and C57BL/6 mice (Figure 1b,Figure 3). In livers of BALB/c mice, the greatest bacterial load and expression of mRNA for all chemokines and CSF investigated was observed at day 3 postinfection. Increasing bacterial load in livers of C57BL/6 mice between days 3 and 14 (Figure 1b) correlated with the period of greatest mRNA expression of chemokines and CSF (Figure 3).

Histology

To examine the type of cellular response to B. pseudomallei infection, histological assessments of the livers and spleens of BALB/c and C57BL/6 mice killed at days 1, 3 and 14 were carried out (Figure 4,Figure 5, respectively). Because BALB/c mice succumbed to infection by day 4, tissue from these mice was examined only at days 1 and 3.

Figure 4.

Histological examination of (a,c) BALB/c and (b,d,f) C57BL/6 livers following infection with 25 c.f.u. Burkholderia pseudomallei. (e) Control mice received PBS only. Livers were collected at (a,b) day 1, (c,d) day 3 and (f) day 14 postinfection for histology. At day 1, early neutrophil (n) accumulation was evident in (a) BALB/c mice, whereas lesions in (b) C57BL/6 mice contained both neutrophils and macrophages (m). Macrophage accumulation was apparent at day 3 in (c) BALB/c mice, although this was more pronounced in (d) C57BL/6 mice. (c) Extensive necrosis (N) was evident in BALB/c mice at day 3, preceding death. (f) By day 14, lesions in C57BL/6 mice comprised predominantly macrophages and involved a greater area of tissue.

Full figure and legend (367K)

Figure 5.

(Previous page) Histological examination of spleens of (a,c,e) BALB/c and (b,d,f,h) C57BL/6 mice following infection with 25 c.f.u. Burkholderia pseudomallei. (g) Control mice received PBS only. Spleen were collected at (a,b) day 1, (c–f) day 3 and (h) day 14 postinfection for histology. (a,b) At day 1, early neutrophil (n) accumulation was evident only in BALB/c mice. (c) Multiple necrotic lesions (N) were observed throughout the spleen by day 3. (e) These lesions were diffuse and contained some macrophages (m) and neutrophils. (d) In C57BL/6 mice, lesions at day 3 consisted of well-defined areas of necrosis. (f) Necrotic tissue and cellular debris were surrounded by fibroblasts (f) and macrophages. (h) By day 14, lesions were larger and often coalesced (arrows).

Full figure and legend (308K)

A rapid inflammatory response consisting predominantly of neutrophils was observed as early as day 1 in the livers of BALB/c mice (Figure 4a). Increases in mRNA for CXC at day 1 in livers of BALB/c mice (Figure 2) correlated with an early neutrophil influx. Macrophages were detected at inflammatory sites by day 3 in BALB/c livers (Figure 4c). But, compared to C57BL/6 livers at day 3 (Figure 4d), lesions in BALB/c livers were predominantly composed of necrotic tissue with relatively few viable inflammatory cells (Figure 4c). Necrosis was extensive in livers of BALB/c mice at day 3 postinfection and numerous thrombi were also present in the blood vessels. A similar cellular response was observed in spleens of BALB/c mice (Figure 5). Compared to control mice, a mild neutrophilic influx was evident in spleens of BALB/c mice at day 1 postinfection (Figure 5a). Similar to their livers, early inflammatory foci in spleens of BALB/c mice progressed to widespread necrotic lesions by day 3 (Figure 5c). Multiple inflammatory foci were dispersed throughout the red and white pulp, with evidence of disseminated intravascular coagulation. Necrotic areas contained remnants of the early neutrophilic influx and were not well defined (Figure 5e). The extent of tissue involvement correlated with the high bacterial loads seen in the livers and spleens of BALB/c mice at day 3 (Figure 1).

At day 1 postinfection, small numbers of inflammatory foci were observed in livers of C57BL/6 mice (Figure 4b). These lesions contained a mixture of macrophages and neutrophils, in which the number of macrophages present was slightly greater (Figure 4b). By day 3, macrophages were the predominant cell type in lesions of C57BL/6 livers (Figure 4d) and had surrounded central areas comprising neutrophils, necrotic cells and nuclear debris. Fewer lesions were seen in livers of C57BL/6 compared with BALB/c at day 3 following infection. By day 14 following B. pseudomallei infection, an increase in the number of lesions in C57BL/6 livers was observed. However, unlike the terminal stages of infection in BALB/c mice (Figure 4c), the inflammatory response in C57BL/6 livers resulted in densely packed accumulations of macrophages and lymphocytes (Figure 4f). No lesions were detected in the spleens of C57BL/6 mice at day 1 (Figure 5b) but, by day 3, consolidated, well-defined inflammatory foci were present (Figure 5d). Lesions consisted of a central area of suppurative inflammation surrounded by a narrow zone of necrosis containing cellular debris (Figure 5f). A zone of macrophages and fibroblasts was adjacent to the necrotic area (Figure 5f). By day 14 multiple abscesses were evident in the livers and spleens of C57BL/6 (Figure 4f,Figure 5h).

Discussion

Recruitment and migration of granulocytes, lymphocytes and monocytes to sites of infection is a crucial event during an immune response. Paradoxically, while these processes are central for the effective clearance of a pathogen, they can, in some instances, contribute to the pathogenesis of disease.²⁴ Chemokines play an important role in the process of inflammatory and immune reactions and are subdivided into subfamilies according to the presence or absence of conserved cysteine (C) residues.^{12, 25} The CXC or chemokine subfamily are subdivided into two groups according to the presence or absence of the amino acid group, glutamate-leucine- arginine (ELR), adjacent to the CXC motif. Interleukin 8, LPS-induced CXC chemokine (LIX), KC and MIP-2 are ELR⁺ and preferentially attract neutrophils.^{26, 27} On the other hand, the CXC chemokines Mig and IP-10 lack the ELR motif (ELR^–) and are typically chemotactic for monocytes and T lymphocytes.^{25, 27} Members of the CC or subfamily demonstrate chemotactic activity for monocytes, lymphocytes, eosinophils and basophils, but not neutrophils.^{25, 26} CC chemokines include RANTES, MCP-1 and MIP-1.

Colony stimulating factors are glycoproteins produced by various cell types throughout the body. They are necessary for the survival, proliferation and differentiation of haematopoietic progenitor cells of the myeloid and erythroid lineage. Four CSF have been described in mice (GM-CSF, G-CSF, M-CSF and IL-3) and humans (GM-CSF, G-CSF, pluripoietin and human urine CSF), and these show considerable structural and functional homology.¹³ Macrophage CSF enhances the survival and activation of cells of the monocyte lineage, while G-CSF boosts the infection-fighting ability of neutrophils.²⁸ Granulocyte-macrophage CSF has a dual role and is able to augment the accumulation and activation of both neutrophils and macrophages.²⁸

Using animal models of infection with intracellular bacteria such as Mycobacterium tuberculosis and Listeria monocytogenes, investigators have demonstrated the importance of particular types of chemokines and CSF in mediating a successful host immune response.^{8, 10} For example, in pulmonary tuberculosis, macrophages and lymphocytes rather than neutrophils are important for host survival.^{10, 29} Monocyte chemoattractant protein 1 and MIP-1, which are chemotactic for mononuclear cells, are upregulated during infection with M. tuberculosis.¹⁰ Similarly, the induction of MCP-1 and CSF, which results in a predominantly monocytic infiltrate, is necessary for protection against infection with L. monocytogenes.^{8, 30}

Studies in patients with melioidosis have demonstrated a link between severity of disease and the levels of cytokines and chemokines in serum, including IFN-, TNF-, IL-6 and IL-8.^{31, 32, 33} Interferon , TNF- and IL-6 have also been implicated as important determinants of disease progression and outcome in murine melioidosis.^{5, 17} Furthermore, induction of mRNA for LIX has been demonstrated in BALB/c mice infected with B. pseudomallei (GC Ulett et al., unpubl. data, 2001). Lipopolysaccharide-induced CXC chemokine is considered to be a functional homologue of IL-8 in mice because of its potent chemotactic activities for neutrophils.³⁴ However, the role of chemokines and CSF during infection with B. pseudomallei has not been adequately investigated. The present study examined the level of expression of a broad range of chemokines and CSF during both the early and later stages of infection in an animal model of human melioidosis.

The present study demonstrated that C57BL/6 and BALB/c mice mount a substantial chemokine and CSF response following exposure to a virulent strain of B. pseudomallei. Induction of mRNA for neutrophil chemoattractants (KC, MIP-2, G-CSF and GM-CSF) and monocyte/lymphocyte chemoattractants (MCP-1, IP-10 and Mig, M-CSF and GM-CSF) occurred in both BALB/c and C57BL/6 mice following infection with B. pseudomallei. However, the kinetics and magnitude of mRNA induction differed between the two mouse strains. These differences were associated with contrasting cellular responses to infection and reflected loads in liver and spleen.

While many of the chemokines and CSF responses detected in C57BL/6 mice were similar to those in BALB/c mice, the kinetics of mRNA expression and cellular inflammatory response differed considerably. In C57BL/6 mice, maximal levels of chemokine and CSF expression were observed between days 3 and 7 of the study, coinciding with increasing bacterial loads in the liver and spleen. Low-level increases in mRNA expression for the mononuclear cell chemoattractant Mig in spleen and the neutrophil chemo-attractant KC in liver correlated with small inflammatory infiltrates in these organs, comprising predominantly macrophages. An increase in macrophage number and frequency of lesions in C57BL/6 livers at day 3 was associated with a large increase in Mig mRNA expression at the same point in time. At day 3 in spleen, a mixture of mRNA for both ELR⁺ and ELR^– CXC chemokines was observed, together with M-CSF. Similarly, cellular infiltrates were also mixed and were observed as central regions of necrotic tissue and neutrophils surrounded by mononuclear leucocytes. As abscessation became more widespread in C57BL/6 spleens at day 14, the expression of mRNA for the neutrophil chemo-attractants KC, MIP-2 and GM-CSF and the mononuclear cell chemoattractant MCP-1 increased. In contrast to spleen, no GM-CSF or MCP-1 mRNA was detected in C57BL/6 livers at day 14. Large increases in mRNA expression of CSF in BALB/c mice at day 3 may reflect the failure of leucocytes to contain B. pseudomallei infection at sites of inflammation, and the subsequent high demands this may place on further inflammatory cell recruitment.

In the present study, we observed a neutrophil-dominant infiltration in BALB/c mice compared to C57BL/6 mice during the early stages of B. pseudomallei infection. An early neutrophilic influx and an increase in mRNA expression for KC, MIP-2, Mig and IP-10 in the present study, as well as increases in IFN-, TNF- and IL-12 in previous studies,⁵ suggest that BALB/c mice are able to mount a strong innate immune response toward B. pseudomallei infection. However, the significantly greater bacterial loads in BALB/c mice at day 3, together with their failure to attract mono-nuclear cells to the same extent as C57BL/6 mice suggests BALB/c mice may not be able to mount an appropriate cellular response to B. pseudomallei. A study by Jones et al. demonstrated the ability of B. pseudomallei to resist killing by neutrophil defensins, thus facilitating the escape of bacteria from neutrophils and the subsequent interaction with less bactericidal mononuclear phagocytes.³⁵ The results of the present study, together with those of Jones et al., indicate that an early neutrophilic influx is not sufficient to eradicate B. pseudomallei from tissues of BALB/c mice.³⁵ In C57BL/6 mice, the infiltration of macrophages within the first 3 days of infection may serve to adequately contain B. pseudomallei for a longer period than in BALB/c mice, allowing the generation of an adequate immune response. This is reflected in the significantly lower bacterial loads in the spleens and livers of C57BL/6 mice in the first 3 days postinfection. In the present study and others,^{5, 17} the hyperproduction of cytokines, such as IL-6 and TNF-, as well as CC and CXC chemokines, and CSF in BALB/c tissues at day 3 may be an attempt to bring the B. pseudomallei infection under control. However, increased concentrations of proinflammatory cytokines and chemokines, combined with high levels of bacterial toxins, are likely to contribute to tissue destruction, triggering disseminated intravascular coagulation and multiple organ system failure. These results are similar to the course of infection seen in humans with acute melioidosis. The comparatively low numbers of macrophages at sites of infection in livers and spleens of BALB/c mice by day 3 implies an inability of the chemokine response in these tissues to attract mononuclear leucocytes. Consequently, bacterial growth and replication proceeds, leading to bacteraemia.⁷ There are several possibilities that may explain the apparent inability of BALB/c mice to attract macrophages to sites of inflammation. While a broad range of chemoattractants were investigated in the present study, contributions from additional CXC and CC chemokines must be considered. Critical factors in determining the accumulation of a particular leucocyte subset are considered to be both the ratio of local and systemic chemokine concentrations,³⁶ and the interaction of particular chemotactic groups in a precise sequence.³⁷ Moreover, the role of chemokine receptors in B. pseudomallei infection has not yet been studied. As chemokine action is dictated by the spectrum of chemokine receptor expression,^{38, 39, 40} it is possible that a decrease in CC or ELR^– CXC chemokine receptor expression contributes to the lack of macrophages in BALB/c mice. Finally, studies involving Legionella pneumophila and M. tuberculosis have indicated an immunomodulatory role for neutrophils through the production of cytokines such as IL-12 and TNF-.^{39, 41} Increased expression of both CXC and CC chemokines has been reported during infection with M. tuberculosis, despite observations that the predominant cell types observed at sites of infection were macrophages and lymphocytes.⁴² We suggested that despite the production of neutrophil chemoattractants such as MIP-2, the chemotactic properties of these CXC chemokines might be modulated or neutralized in the presence of other chemokines or cytokines.⁴² A cross- regulatory role in chemokine production has been demonstrated for a number of cytokines including IL-2, IL-4, IFN- and IL-10.^{10, 43, 44} Hence, it is likely that cytokines also play an important role in regulating chemokine and CSF production during infection with B. pseudomallei. Further investigations into the role of neutrophils, chemokines and their receptors in the early stages of B. pseudomallei infection is therefore warranted.

In summary, the present study demonstrated disparate expressions of mRNA for several chemokines and haematopoietic cytokines in a murine model of acute and chronic human melioidosis. These discrepancies are associated with differences in bacterial growth and the composition of cellular infiltrate and may underlie the highly susceptible phenotype of BALB/c mice to B. pseudomallei infection. Investigations are currently being undertaken to evaluate the role of particular leucocyte subsets during acute and chronic murine melioidosis.

References

1. Guard RW, Khafagi FA, Brigden MC, Ashdown LR. Melioidosis in Far North Queensland: A clinical and epidemiological review of twenty cases. Am. J. Trop Med. Hyg. 1984; 33: 467–73. | PubMed | ChemPort |
2. Ashdown LR, Guard RW. The prevalence of human melioidosis in Northern Queensland. Am. J. Trop. Med. Hyg. 1984; 33: 474–8. | PubMed | ChemPort |
3. Kingston CW. Chronic or latent melioidosis. Med. J. Aust. 1971; 2: 618–21. | PubMed | ChemPort |
4. Leelarasamee A, Bovornkitti S. Melioidosis: Review and update. Rev. Infect. Dis. 1989; 11: 413–25. | PubMed | ChemPort |
5. Ulett GC, Ketheesan N, Hirst RG. Cytokine gene expression in innately susceptible BALB/c mice and relatively resistant C57BL/6 mice during infection with virulent Burkholderia pseudomallei. Infect. Immun. 2000; 68: 2034–42. | Article | PubMed | ChemPort |
6. Hoppe I, Brenneke B, Rohde M et al. Characterisation of a murine model of melioidosis: Comparison of different strains of mice. Infect. Immun. 1999; 67: 2891–900. | PubMed | ChemPort |
7. Leakey AK, Ulett GC, Hirst RG. BALB/c and C57Bl/6 mice infected with virulent Burkholderia pseudomallei provide contrasting animal models for the acute and chronic forms of human melioidosis. Microb. Pathog. 1998; 24: 269–75. | PubMed | ChemPort |
8. DiTirro J, Rhoades ER, Roberts AD et al. Disruption of the cellular inflammatory response to Listeria monocytogenes infection in mice with disruptions in targeted genes. Infect. Immun. 1998; 66: 2284–9. | PubMed | ISI | ChemPort |
9. Orme IM, Cooper AM. Cytokine/chemokine cascades in immunity to tuberculosis. Immunol. Today 1999; 20: 307–12. | Article | PubMed | ChemPort |
10. Mohammed KA, Nasreen N, Ward MJ, Antony VB. Helper T cell type 1 and 2 cytokines regulate C-C chemokine expression in mouse pleural mesothelial cells. Am. J. Respir. Crit. Care Med. 1999; 159: 1653–9. | PubMed | ChemPort |
11. Lukacs NW. C-C chemokines differentially alter interleukin-4 production from lymphocytes. Am. J. Pathol. 1997; 150: 1861–8. | PubMed | ISI | ChemPort |
12. DeVries ME, Ran L, Kelvin DJ. On the edge: the physiological and pathophysiological role of chemokines during inflammatory and immunological responses. Semin. Immunol. 1999; 11: 95–104. | Article | PubMed | ISI | ChemPort |
13. Metcalf D. The molecular biology and functions of the granulocyte-macrophage colony-stimulating factors. Blood 1986; 67: 257–67. | PubMed | ISI | ChemPort |
14. Gasson JC, Weisbart RH, Kaufman SE et al. Purified human granulocyte-macrophage colony stimulating factor: Direct action on neutrophils. Science 1984; 226: 1339–42. | Article | PubMed | ISI | ChemPort |
15. Brown AE, Dance DAB, Chaowagul W, Webster HK, White NJ. Activation of cellular immune responses in melioidosis patients as assessed by urinary neopterin. Trans. R. Soc. Trop. Med. Hyg. 1990; 84: 583–4. | Article | PubMed | ChemPort |
16. Lauw FN, Simpson AJH, Prins JM et al. The CXC chemokines gamma interferon (IFN-)-inducible protein 10 and monokine induced by IFN- are released during severe melioidosis. Infect. Immun. 2000; 68: 3888–93. | Article | PubMed | ChemPort |
17. Ulett GC, Ketheesan N, Clair T et al. Cellular and cytokine responses in experimental melioidosis: Virulence influences the induction of TNF-, IL-1 and IL-6 during the innate phase of immunity. In: Proceedings of the 2nd Congress of the Federation of Immunological Societies of Asia–Oceania. Italy: Monduzzi, 2000; 137–44.
18. Dilworth DD, McCarrey JR. Single-step elimination of contaminating DNA prior to reverse transcriptase PCR. PCR Appl. Methods 1992; 1: 279–82. | ChemPort |
19. Murray LJ, Lee R, Martens C. In vivo cytokine gene expression in T cell subsets of the autoimmune MRL/Mp-lpr/lpr mouse. Eur. J. Immunol. 1990; 20: 163–70. | PubMed | ChemPort |
20. Bongers M, Liehl E, Barsig J. One-step RT-PCR to detect cytokine/ chemokine induction in macrophages. Focus 1999; 21: 66–8.
21. Neumann B, Emmanuilidis K, Stadler M, Holzmann B. Distinct functions of interferon- for chemokine expression in models of acute lung inflammation. Immunology 1998; 95: 512–21. | Article | PubMed | ISI | ChemPort |
22. Ehlers S, Mielke MEA, Blankenstein T, Hahn H. Kinetic analysis of cytokine gene expression in the livers of naive and immune mice infected with Listeria monocytogenes. J. Immunol. 1992; 149: 3016–22. | PubMed | ChemPort |
23. Szilvassy SJ, Weller KP, Lin W et al. Leukemia inhibitory factor upregulates cytokine expression by a murine stromal cell line enabling the maintenance of highly enriched competitive repopulating stem cells. Blood 1996; 87: 4618–28. | PubMed | ISI | ChemPort |
24. Strieter RM, Standiford TJ, Huffnagle GB, Colletti LM, Lukacs NW, Kunkel SL. 'The Good, the Bad, and the Ugly': The role of chemokines in models of human disease. J. Immunol. 1996; 156: 3583–6. | PubMed | ISI | ChemPort |
25. Rollins BJ. Chemokines. Blood 1997; 90: 909–28. | PubMed | ISI | ChemPort |
26. Vaddi K, Keller M, Newton RC. The Chemokine Facts Book. London: Academic Press, 1997.
27. Teruya-Feldstein J, Jaffe ES, Burd PR. The role of Mig, the monokine induced by interferon-, and IP-10, the interferon--inducible protein-10, in tissue necrosis and vascular damage associated with Epstein-Barr virus-positive lymphoproliferative disease. Blood 1997; 90: 4099–105. | PubMed | ChemPort |
28. Sowter HM, Corps AN, Evans AL et al. Expression and localisation of the vascular endothelial growth factor family in ovarian epithelial tumours. Lab. Invest. 1997; 77: 607–14. | PubMed | ISI | ChemPort |
29. Saunders BM, Cooper AM. Restraining mycobacteria: Role of granulomas in mycobacterial infections. Immunol. Cell Biol. 2000; 78: 334–41. | Article | PubMed | ChemPort |
30. Wing EJ, Waheed A, Shadduck RK. Changes in serum colony-stimulating factor and monocytic progenitor cells during Listeria monocytogenes infection in mice. Infect. Immun. 1984; 45: 180–4. | PubMed | ChemPort |
31. Brown AE, Dance DAB, Suputtamongkol Y. Immune cell activation in melioidosis: Increased serum levels of interferon- and soluble interleukin-2 receptors without change in soluble CD8 protein. J. Infect. Dis. 1991; 163: 1145–8. | PubMed | ChemPort |
32. Friedland JS, Suputtamongkol Y, Remck DG. Prolonged elevation of interleukin-8 and interleukin-6 concentrations in plasma and of leucocyte interleukin-8 mRNA levels during septicemic and localised Pseudomonas pseudomallei infection. Infect. Immun. 1992; 60: 2402–8. | PubMed | ChemPort |
33. Lauw FN, Simpson AJH, Prins JM et al. Elevated plasma concentrations of interferon (IFN)- and the IFN--inducing cytokines interleukin (IL)-18, IL-12 and IL-15 in severe melioidosis. J. Infect. Dis. 1999; 180: 1878–85. | Article | PubMed | ChemPort |
34. Rovai LE, Herschman HR, Smith JB. The murine neutrophil-chemoattractant chemokines LIX, KC, and MIP-2 have distinct induction kinetics, tissue distributions and tissue-specific sensitivities to glucocorticoid regulation in endotoxemia. J. Leukoc. Biol. 1998; 64: 494–502. | PubMed | ISI | ChemPort |
35. Jones AL, Beveridge TJ, Woods DE. Intracellular survival of Burkholderia pseudomallei. Infect. Immun. 1996; 64: 782–90. | PubMed | ChemPort |
36. Call DR, Nemzek JA, Ebong SJ et al. Ratio of local to systemic chemokine concentrations regulates neutrophil recruitment. Am. J. Pathol. 2001; 158: 715–21. | PubMed | ChemPort |
37. Gutierrez-Ramos JC, Lloyd C, Kapsenberg ML et al. Non-redundant functional groups of chemokines operate in a coordinate manner during the inflammatory response in the lung. Immunol. Rev. 2000; 177: 31–42. | Article | PubMed | ChemPort |
38. Tsai WC, Strieter RM, Mehrad B et al. CXC chemokine receptor CXCR2 is essential for protective innate host response in murine Pseudomonas aeruginosa pneumonia. Infect. Immun. 2000; 68: 4289–96. | Article | PubMed | ISI | ChemPort |
39. Tateda K, Moore TA, Newstead MW et al. Chemokine- dependent neutrophil recruitment in a murine model of Legionella pneumonia: Potential role of neutrophils as immunoregulatory cells. Infect. Immun. 2001; 69: 2017–24. | Article | PubMed | ISI | ChemPort |
40. Mantovani A, Allavena P, Vecchi A, Sozzani S. Chemokines and chemokine receptors during activation and deactivation of monocytes and dendritic cells and in amplification of Th1 versus Th2 responses. Int. J. Clin. Lab. Res. 1998; 28: 77–82. | Article | PubMed | ChemPort |
41. Pedrosa J, Saunders BM, Appelberg R et al. Neutrophils play a protective nonphagocytic role in systemic Mycobacterium tuberculosis infection in mice. Infect. Immun. 2000; 68: 577–83. | Article | PubMed | ChemPort |
42. Rhoades ER, Cooper AM, Orme IM. Chemokine response in mice infected with Mycobacterium tuberculosis. Infect. Immun. 1995; 63: 3871–7. | PubMed | ChemPort |
43. Kasama T, Strieter RM, Lukacs NW et al. Regulation of neutrophil-derived chemokine expression by IL-10. J. Immunol. 1994; 152: 3559–69. | PubMed | ISI | ChemPort |
44. Loetscher P, Seitz M, Baggiolini M, Moser B. Interleukin-2 regulates CC chemokine receptor expression and chemotactic responsiveness in T lymphocytes. J. Exp. Med. 1996; 184: 569–77. | Article | PubMed | ISI | ChemPort |

Acknowledgements

This study was funded by a Program Grant from James Cook University (Townsville, Queensland, Australia).