Different lines of evidence from patients with melioidosis, including the elevation of pro- and anti-inflammatory cytokines in serum, suggest the importance of innate immunity in determining the outcome of infection [
2,
10]. There is also evidence in humans showing that recovery from melioidosis may depend, in part, on cell-mediated arm of adaptive immune system [
11]. The latter is partially dependent on the type of Th cell differentiation which is controlled in part by activated DCs. Studies of experimentally induced melioidosis in different inbred strains of animals have shown the importance of genetic background in determining the course of the infection [
12]. Despite numerous studies in animal melioidosis models using mutant and wild type bacteria [
13,
14] a complete understanding of the mechanisms involved in pathogenesis, particularly in humans, has yet to be realized. We felt that additional approaches looking at the effects of infection on isolated human phagocytic cell populations would contribute new insights into the pathogenic mechanism of the disease. We also reasoned that using virulent
Bp as well as its genetically related avirulent natural counterpart,
Bt, might highlight aspects of the host response that play a role in virulence and pathogenesis. Moreover, the use of primary human cells, instead of animal cells or cell lines, would yield results that are more relevant in regard to the human disease. We had previously demonstrated that this same
Bp strain adhered to and invaded a non-phagocytic human lung alveolar cell line more effectively than
Bt-UE5 [
6]. In the present study, we continued this same approach, but this time using primary human phagocytic cells instead of a non-phagocytic human and animal cell lines. After internalization, we found that
Bt strain UE5 but not
Bp strain 844, had a reduced survival rate and failed to multiply in the Mφs (Figure
1A). We have limited data confirming that another
Bp stain (1026b) could also survive and readily replicate inside primary human macrophages (unpublished). From the current information regarding the different characteristics between these 2
Burkholderia species, we suspect that exopolysaccharide capsular components [
13] or the type-III secretion system cluster 1 [
15], which are absent in
Bt, may be responsible for the differences in their intracellular survival kinetics. Currently, we have data showing that, compared with its wild type counterpart, a capsule mutant of
Bp that lacks the exopolysaccharide component was at least 10
5-fold less virulent in BALB/c mice and it also survived and replicated poorly inside primary human Mφs (manuscript submitted).
Furthermore, our results showed that following IFN-γ activation, both MoDCs and Mφs acquired an enhanced ability to kill
Bp-844. The significance of IFN-γ in conferring resistance against experimental melioidosis has been reported previously by several groups of investigators [
16,
17] including our own group [
18]. Unactivated MoDCs, while possessing an antigen-presenting capacity for the stimulation of the adaptive immune response that is superior to that of Mφs, were unable to inhibit the intracellular growth of either
Burkholderia strains (Figure
1B). However, after IFN-γ pre-activation, an enhanced intracellular killing capacity was observed in the MoDCs infected with
Bp-844 (Figure
1D).
In vivo, IFN-γ can be produced by NK cells or Th1 cells [
16], but their arrival at the initial site of infection might be too late to provide an effective antibacterial response. We previously showed that injecting BALB/c mice with CpG ODN to enhance IFN-γ production and innate immunity in general could effectively protect the animals against a subsequent challenge with virulent
Bp [
18]. Furthermore, we found that if the CpG ODN administration was given closer to the time of bacterial challenge, the degree of protection was also reduced. As demonstrated in the present study, the enhanced
in vitro killing of Bp-844 by human MoDCs and Mφs pre-stimulated with IFN-γ (Figure
1C) is in accord with the previously mentioned
in vivo observations in experimental animals. Taken together, it appears that the host innate immune response is able to distinguish
Bp-844 from
Bt-UE5 (Figure
1), and this may in turn influence the outcome of primary infection by
Bp. Once the infection has progressed beyond this stage, the host adaptive immune system will be activated but may not be able to distinguish between these 2 bacteria, as demonstrated by our results showing the similar patterns of costimulatory molecule expression, cytokine production and strong bias toward a Th1 polarization (Figures
2 and
3). In addition, the results of our
in vitro studies showing that human MoDCs that have been exposed to
Bp-844 have predominant biased effects on Th1 cell differentiation (Figure
3) are consistent with the results of a previous study showing strong Th1 cell-mediated immune responses in melioidosis patients who had recovered from melioidosis [
11]. Moreover, a mixed Th1/Th2 response has also been reported in animal studies [
12]. The detection of some activated T cells exhibiting a mixed Th1/Th2 response producing both intracellular IFN-γ and IL-4 in this study (Figure
3) is in accord with the animal study just mentioned. Further investigations using the human phagocytic cell system described in the present study are needed if one is to more fully understand the pathogenesis of human melioidosis. We are currently investigating the possible role and involvement of polymorphonuclear cells in the disease process in humans.