Polymorphonuclear neutrophil response to Burkholderia pseudomallei in patients with diabetes mellitus

Background and objectives: Burkholderia pseudomallei is the causative agent of melioidosis, a potentially fatal disease endemic in Bangladesh. Diabetes mellitus (DM) is a risk factor for increased susceptibility to B. pseudomallei infection. A few studies have been conducted to identify the underlying immunological mechanism responsible for increased susceptibility of individuals with diabetes mellitus to B. pseudomallei infection. The present study investigated the polymorphonuclear neutrophil (PMN) response to B. pseudomallei in terms of phagocytosis and early respiratory burst in individuals with diabetes mellitus. Materials and Methods: A total of 5 cases of DM and 5 age and sex matched non-diabetic healthy individuals were enrolled in the study to determine the early respiratory burst and phagocytic ability of PMN to B. pseudomallei. The effect of B. pseudomallei on phagocytic ability and early respiratory burst of PMN was determined by phagocytic assay and nitroblue tetrazolium (NBT) test respectively. The response of PMN treated with B. pseudomallei was compared with that of Escherichia coli. Results: There was no significant (p>0.05) difference in phagocytosis of B. pseudomallei by PMN between diabetic and non-diabetic cases (21.8±4.64 percent vs 29.25±5.5 percent). But in both diabetic and non-diabetic cases, significantly (p˂0.05 and p˂0.01) reduced rate of phagocytosis of B. pseudomallei by PMN was observed compared to E. coli (21.8±4.64 vs 65±5.36; 29.25±5.5 vs 71.25±5.59). Similar results were obtained in terms of phagocytic index. Mean percentage of formazan positive PMN from diabetic cases was not significantly different (p>0.05) from nondiabetic healthy cases when cells were treated with B. pseudomallei or E. coli. In both diabetic and healthy individuals, mean percentage of formazan positive PMN treated by B. pseudomallei was not significantly different from that by E. coli. Conclusion: The observations revealed that B. pseudomallei was equally capable of inhibiting the phagocytic ability of PMN from both diabetic and non-diabetic individuals. This antiphagocytic property might play an important role in the pathogenesis of melioidosis. IMC J Med Sci 2019; 13(2): 005. EPub date: 01 September 2019


Introduction
Burkholderia pseudomallei, a motile gram-negative facultative intracellular bacterium, is the causative agent of melioidosis which ranges from asymptomatic infection, to localized or disseminated abscess to fatal septicemia [1]. The global burden of melioidosis is 165,000 human cases per year, of which 89,000 (54%) die. The bacterium is intrinsically resistant to a wide range of antibiotics and more than 70% of cases die due to treatment with ineffective antibiotics [2][3]. Bangladesh is an endemic country for melioidosis [4]. In Bangladesh, so far, 35 culture-confirmed melioidosis cases were identified from 2001 to 2016; however, true extent of the disease is unknown because of unfamiliarity of the organism to physicians and microbiologists of the country [4]. Human gets infection mainly via traumatic inoculation. After entry into the host, the organism enters into macrophage and may cause latent infection in immunocompetent host and reactivate in immunosuppressed condition [3,5]. Understanding the pathogenesis of B. pseudomallei and the role of host immune response are essential to realize the course of the disease.
Diabetes mellitus is the most common risk factor for melioidosis, and is a co-morbid condition in more than 50% of all melioidosis cases [3]. The risk of diabetic people getting melioidosis is exceedingly higher than the rest of the population [6][7]. Diabetic individuals with poor glycemic control have defects in immune responses against infections [8]. However, the effect of DM on immunopathogenesis of B. pseudomallei infection is not clear yet. In BALB/c mice, the virulence of B. pseudomallei isolates from DM patients is significantly lower than that of isolates from patients without any risk factor; suggesting immunopathological changes due to diabetes increases the susceptibility to otherwise innocuous strain of B. pseudomallei [9].
The impact of diabetes in B. pseudomallei infection clearly delineates that the host immune response dictates the outcome of B. pseudomallei infection. In addition to acquired immune response, innate immune responses might play a critical role for the initial control of infection. Polymorphonuclear neutrophil (PMN), a cell of innate immunity, is considered the first line of defense against invading microorganisms. In vivo findings suggest that depletion of PMN from mice resulted in an acute and rapidly lethal B. pseudomallei infection, indicating the importance of PMN in early clearance [10]. Mice lacking NADPH oxidase, an important enzyme in the generation of the neutrophil respiratory burst, were more susceptible to B. pseudomallei disease [11].
Diabetes mellitus, the primary predisposing condition for melioidosis, is associated with impaired chemotaxis, phagocytosis, oxidative burst, and killing activity of neutrophils [12]. Increased incidence of melioidosis is also noted in neutropenic patients [13], as well as patients with chronic granulomatous disease [14]. Interestingly, treatment of melioidosis with the neutrophil-differentiating cytokine granulocyte colony-stimulating factor (G-CSF) showed mixed results [15]. It has reported to reduce the mortality of melioidosis patients in Australia [15], but is only associated with prolonged survival in Thai patients [16]. Thus, all these in vivo reports suggest the important role of neutrophils in controlling B. psedomallei infection, but how B. pseudomallei affects neutrophil is not clear yet.
In this study, we aimed to observe PMN responses to B. pseudomallei in Bangladeshi people with diabetes mellitus. This would help us to understand the role of innate immunity in the pathogenesis of B. pseudomallei.

Materials and methods
Study population and sample collection: Cases of diabetes mellitus of different duration and age and sex matched apparently healthy non-diabetic individuals were enrolled in the study. All diabetic cases were on different oral hypoglycemic agents. Cases with chronic disease, known hematological disorders, acute infection, leucocytosis and fever in last one month were excluded. About 2-3 ml of blood was collected from each individual with aseptic precautions in a sterile heparin tube and kept in ice until used. All assays to measure the PMN response to B. pseudomallei and E. coli were done within 2-3 hrs of blood collection. Informed consent was obtained from all study population prior to collection of blood sample. The study was approved by the Ethic Review Board of Bangladesh Institute of Research and Rehabilitation in Diabetes, Endocrine and Metabolic Disorders (BIRDEM).

Bacterial strains and cell preparation: B.
pseudomallei strain (CS6887) isolated from a Bangladeshi melioidosis patient and Escherichia coli ATCC 25922 strain were used to measure the early respiratory burst and phagocytic function of PMN. The response of PMN to B. pseudomallei (CS6887) was compared to that of E. coli. The bacteria were stored in trypticase soya broth (TSB, HiMedia Laboratories Pvt. Ltd., India) with 15% glycerol at -20 0 C until used. A single colony of B. pseudomallei or E. coli was suspended separately in 5 ml TSB in two tubes and incubated aerobically overnight at 37 0 C to obtain a cell suspension of 3×10 8 colony forming units per ml (CFU/ml).
Phagocytic assay: Phagocytic assay was performed as previously described [17]. Briefly, 20 µl of growth of either B. pseudomallei or E. coli in TSB (6×10 6 cells) was added to 500 µl of whole blood sample in two tubes and mixed thoroughly by gentle shaking. The tubes were incubated aerobically at 37 0 C for 30 minutes. After 30 minutes, the mixture was mixed with gentle shaking. Then, duplicate smears were made on two glass slides from each tube and air dried. Before making smear, the glass slides were soaked in xylene overnight and then washed with absolute ethanol and air dried for minimizing the clumping of neutrophils [18]. One slide was stained with Leishman stain and another one with 0.5% safranin stain. Smears were examined under the oil immersion lens and 200 neutrophils were counted ( Figure-1a, 1b). The percentage of PMN with phagocytosed bacteria was calculated by: {(Number of PMN with phagocytosed bacteria ÷ Total PMN counted) x 100}. The phagocytic index per neutrophil was estimated by the formula: {Total number of intracellular bacteria ÷ Total PMN with phagocytosed bacteria counted} [17].

Nitro blue tetrazolium (NBT) test:
Early respiratory burst of PMN to B. pseudomallei was determined by NBT test as previously described [19][20]. Formation of formazan by reduction of NBT dye following antigenic stimulation of PMN indicates occurrence of early respiratory burst in the cell.
Twenty microliter of either live B. pseudomallei or E. coli in TSB (6×10 6 cells) was added to 500 µl of whole blood sample in two tubes and mixed thoroughly by gentle shaking. E. coli was used to compare the response of PMN to B. pseudomallei while TSB without any bacteria served as a negative control. The tubes were incubated aerobically at 37 0 C for 30 minutes. Then 500 µl of 0.2% NBT solution was added to each of the above mentioned tubes and mixed thoroughly by gentle shaking. Solution of 0.2% NBT was prepared by dissolving 2 mg NBT dye (Abcam, UK) with 40 µl absolute ethanol; the dissolved solution was made up to 1 ml by adding 960 µl phosphate buffered saline (PBS). To dissolve completely, the solution was heated at 60 0 C for 20 minutes. This solution was made freshly for each batch of test. The tubes were again incubated aerobically at 37 0 C for 25 minutes after adding NBT solution. After 25 minutes, the tubes were mixed with gentle shaking. Then, duplicate smears were made on two glass slides from each tube and air dried. The glass slide was prepared as described above. One slide was stained with Leishman stain and another one with 0.5% safranin stain. Smears were examined under oil immersion lens and 200 neutrophils were counted. Neutrophils with a single, large, dense and deep-blue colored cytoplasmic deposit of formazan (reduced NBT, Figure- Table-3) in formazan positive PMNs was observed between B. pseudomallei and E. coli in both diabetic and nondiabetic healthy cases.

Discussion
In recent time, an increase in melioidosis cases has been observed in Bangladesh. DM is the major risk factor for melioidosis [3]. The aim of this study was to determine neutrophil responses to B. pseudomallei in diabetic patients in vitro. PMN is the first line host defense against bacterial infection. After invasion of bacteria, PMNs migrate to the infection site, phagocytose and kill the bacteria by oxygen-dependent or oxygenindependent pathway [21].
In the present study, PMN from both diabetic and non-diabetic healthy participants showed significantly lower rate of phagocytosis of B. pseudomallei compared to that of E. coli. The phagocytic index (number of intracellular bacteria/PMN) was also significantly lower for B. pseudomallei compared to E. coli in both groups. Earlier, Mulye et al. also showed that in comparison to E. coli, B. pseudomallei was less efficiently phagocytosed by PMNs of both diabetic and healthy subjects [22]. Capsule might be a factor that could have prevented PMN to recognize and internalize the offending bacteria. Such antiphagocytic property due to the presence of carbohydrate capsule have been described for a number of gram-positive and negative bacteria namely Streptococcus pneumoniae, Staphylococcus aureus, Neisseria meningitidis, Haemophilus influenzae and Klebsiella pneumoniae [23][24]. However, in the present study, although B. pseudomallei and E. coli both were capsulated, B. pseudomallei was significantly less phagocytosed than E. coli by PMN from both study groups. This result suggests that B. pseudomallei might have additional virulence factor(s) that is more effectively preventing phagocytosis by PMN. So, further studies are required to investigate anti-phagocytic property of B. pseudomallei in more detail.
However, in the present study, PMN showed no significant differences in the rate of phagocytosis and phagocytic index for both B. pseudomallei and E. coli between diabetic and non-diabetic participants. Similar observation was reported earlier which showed no difference in the rate of phagocytosis of B. pseudomallei by PMN of diabetic and non-diabetic cases [25]. However, the study reported reduced rate of phagocytosis in diabetic cases with very poor glycemic (HbA1c>8.5%) control [25]. Another study group found reduced PMN uptake of virulent K1/K2 Klebsiella pneumoniae in diabetic patients compared with healthy subjects [26].
In oxygen dependent killing mechanism, NADPHoxidase enzyme complex present in PMN is activated, transfers electron to oxygen molecule and forms superoxide. This superoxide kills bacteria. In this study, NBT dye was used to detect superoxide formation. NBT dye is reduced to insoluble deep-blue or purple colored deposit called formazan by transfer of electron by NADPHoxidase enzyme complex. Therefore, formazan formation indicates superoxide formation in terms of early respiratory burst within PMN [21].
In the present study, the early respiratory burst in PMN following stimulation with either B. pseudomallei or E. coli was not significantly reduced in diabetic than non-diabetic population. Though there was no significant reduction of early respiratory burst in PMN from both diabetic and non-diabetic cases for B. pseudomallei and E. coli, we did not assess the actual killing of the ingested pseudomallei can quickly escape from the endosome/phagosome of host cells and persists within the cytoplasm of those cells. Therefore, B. pseudomallei can spread from cell to cell avoiding the host extracellular environment [27]. This might play an important role to in the pathogenesis of melioidosis among diabetics as their other immune parameters are compromised [28]. Also, it is needed to see if there is any alteration of oxygen independent killing of B. pseudomallei by PMN of diabetic patients.
In the present study, the rate of phagocytosis, phagocytic index and early respiratory burst of PMN for both B. pseudomallei and E. coli were though less in diabetic than that of non-diabetic cases, these were not significant. This could be due to low number of cases tested or because of short duration (around 5 years) of diabetes in our cases. Therefore, further study with large number of cases is needed to see the response of PMN to B. pseudomallei in diabetics.