Introduction

Burkholderia pseudomallei is a saprophytic bacterium found in the soil and water of its endemic area [1]. The cause of infection is mostly from contamination by the microbes through skin abrasion and less commonly from inhalation or ingestion [1]. Therefore, constituents of the environment create an important reservoir for this bacterium. Soil is the most complex biomaterial on the planet, containing a wide variety of interacting microbes to build their unique communities. The mechanisms for their survival in ecological environments include adherence, production of toxic metabolites, secretion of antimicrobial substances, and competition for nutrients [2]. Understanding the interference of such mechanisms that provide balance to each ecosystem will be beneficial for selecting antagonistic organisms against human or plant pathogens to be used for controlling microbes and lead to benefits for human and animal health as well as economics as a whole. Transmission of food-borne human pathogens from soil and the environmental spreading of infectious diseases are still major public health problems in several countries [3•]. Among the strategies currently used to control and prevent such infectious diseases, the use of living organisms to control other organisms (biocontrol) has been used for a long time especially for food-borne pathogens [4,5,6,7, 8]. In addition, plant pathogens from soil are another target to which biocontrol can be and has been applied [9,10,11]. As mentioned earlier, examples of biocontrol are already reported to control pathogens in food or soil using natural antagonists such as another bacteria or virus of the bacteria (bacteriophages) [12, 13].

A bacteriophage is a bacterial virus that acts as an obligate parasite. It is generally highly specific and strictly against a single bacterial strain; however, some bacteriophages have a broad host range [14]. The lytic bacteriophages are usually the main phages that are used to control bacteria. The obstacles of bacteriophages as biocontrol agents are as follows: firstly, some bacteriophages infect bacteria and incorporate their nucleic acid into the bacterial genomes, leading to a prophage which will not lyse the bacteria; secondly, as bacteriophages use bacteria as their host for propagation, the numbers of bacteria tend to be reduced but not eliminated by the phages; and lastly, some phages have broad host ranges and may affect other, perhaps beneficial, bacteria in the environment.

Microbes have been an important source of antibiotics for a long time because they can produce a wide spectrum of antimicrobial substances. Presently, several of them may still be a good source of antimicrobial peptides that can be used to fight against several kinds of organisms [15, 16]. They have been used successfully in controlling plant pathogens in soil and thus decrease the contaminated food-borne pathogens [17, 18, 19•]. Apart from those applications, biocontrol from environmental sources using bacteria against human diseases such as melioidosis is very limited.

This review article therefore discusses the soil environment of B. pseudomallei and is focused on the possible uses of bacteriophages and bacteria from the environment as a biocontrol of this disease-causing pathogen.

Environmental Factors and B. pseudomallei

The presence of B. pseudomallei in its ecological habitat was found to be correlated with some physicochemical factors in the environment in the endemic areas [20, 21, 22••, 23, 24••]. From soil surveys in two unused lands of the endemic area in Khon Kaen province, northeast Thailand, these surveys revealed the correlations of the bacteria with low soil pH (4.4–6), a low C/N (carbon/nitrogen) ratio of approximately 18, and a high extractable iron level of approximately 45 ppm [20, 21]. The studied areas are mostly sandy soil, and the organism was found mainly at the 5–30-cm depth. The endemic area of melioidosis disease in northeast Thailand was reported to be acidic [23], and this was confirmed to be one of the physicochemical factors that differed positive from negative soil sample areas. The results were correlated with information from the Agricultural Development Research Center (ADRC) in Thailand, which indicated that sandy soil covered 80% of northeast Thailand where acidic and infertile soils are common in that area. When 61 rice fields that covered the east, central, and northeast of Thailand were sampled at a 30-cm depth, however, the pHs and water contents were not significantly different between positive and negative soils for B. pseudomallei [24••]. The pHs of the soil varied ranging from 4.9 to 8.1, but the average pHs between the three regions were not significantly different. When taking a closer look at the average pH of soil in this study, pH > 6 was found in both B. pseudomallei-positive and B. pseudomallei-negative soils. The average water content was > 10% in both positive and negative soils that were reported to correlate with the presence or absence of the bacterium. Moreover, the types of soil in these regions were different so that the average of all samples may have altered the interpretation of factors correlated with the presence of the bacterium. A study of rice fields in Laos showed a majority of sandy soil types and acidity similar to the northeast of Thailand, but the bacteria predominated in a soil depth of more than 30 cm [25]. This seemed to be related to high moisture content and low total nitrogen and carbon [25]. Again, when the C/N ratio was calculated, both positive and negative soils showed a low C/N ratio of approximately 8–9. This finding goes together with other studies that showed that a low C/N ratio is correlated with the presence of the bacterium [19•, 20] and an increased ratio of > 40:1 could suppress the growth of B. pseudomallei [26]. Other factors such as chloride, organic carbon, phosphorus, potassium, manganese, or exchangeable cations have also been studied but did not show any significant effects when studied in farmlands and between the high and low prevalence areas on Castle hill in Townsville, Australia [22••, 27•]. As the culture of the bacteria is difficult to achieve together with the presence of the unculturable form of the bacteria in the soil [21, 28], negative cultivation may be misleading and variable results, therefore, could be obtained. Molecular methods have been suggested to be used for detecting the presence of the bacteria in the physicochemical studies of soil [25]. According to the results of the molecular method of detection, the comparison between undisturbed sites and environmentally manipulated areas in Australia for the habitat of B. pseudomallei indicated a positive association of B. pseudomallei in rich soils in grassy areas at undisturbed sites while the presence of livestock and a lower pH of soil were associated with manipulated areas [29]. The authors also concluded that B. pseudomallei may spread due to the changes in land management. On the whole, the outcome of factors studied in rice fields indicated a correlation of B. pseudomallei with nutrient depletion [24••, 27•] while high-manure areas with a decrease in C/N ratio and easy-to-digest carbon sources for microbes in farmland were also associated with the presence of the bacteria [22••]. B. pseudomallei is a bacterium that can tolerate diverse conditions from dry, sandy, and infertile soils to wet, clay, and high-manure soils. Therefore, comparisons of factors associated with the presence of the bacterium in different types of soil, different land usage of either undisturbed or manipulated areas, and different depths of water tables in each study area or country result in controversial results of physicochemical properties of soil. A soil survey in rice fields in the northeast of Thailand was reported to be 28% positive for B. pseudomallei with an average of 700 cfu while the empty areas were 80% positive with a 378 cfu [30]. Soil from farmlands disturbed with manure and rice fields that were loaded with fertilizers and tillage during cultivation could affect the ecosystem of the soil and consequently affect the physicochemical properties of the soil. Soil microcosm experiments to simulate various combinations of physicochemical factors in different soil types were conducted and resulted in data that were different from the environmental study [31••]. The combinations of pH, NaCl, iron and carbon-to-nitrogen ratios (C/N) on viability of bacteria cells in the developed microcosms showed that the moisture of soil, pH > 8, NaCl > 1%, and C/N ratios > 40:1 significantly reduced the viable bacteria while increasing the concentrations of iron significantly increased the bacterial growth [26]. Iron in soil is another controversial factor found in several studies as mentioned by Manivanh et al. [25]. The extractable iron (Fe2+) is the reduced form that microbes can use and that was increased under acidic environment. Extractable iron may then be a better indicator than total iron to avoid misinterpretation in an environmental study. The details of various physicochemical factors effecting to the presence B. pseudomallei in the soil are shown in Table 1.

Table 1 Physicochemical factors that affect the growth of B. pseudomallei in soil
Full size table

Therefore, to identify the physicochemical factors correlated with the presence of B. pseudomallei in soil in future studies, investigators should avoid collection of a large quantity of soil or composites of soil samples that may lead to misleading conclusions. Moreover, interpretations of each factor should not only reflect increases or decreases or correlations or negative correlations with the bacterium presence, but rather the range of each factor as low or high during the correlation analysis. Comparisons or interpretation of these factors between undisturbed and natural manipulated areas should also be performed with care.

The concept of using physicochemical factors to inhibit or suppress the bacterial growth or viability of the cells for controlling the bacteria is still difficult to apply due to the involvement of large and different usage areas, soil complexity, and differences in each ecosystem of both planted species and the soil community that always needs optimization [27•]. From all of those studies, it can be hypothesized that the use of physiochemical factors to either support or inhibit the B. pseudomallei and result in controlling the bacterium is unlikely, but information concerning these factors should help the biocontrol of the disease in soil.

Biological Factors and B. pseudomallei

Several reports demonstrated that abiotic factors in soil could affect the presence or absence of B. pseudomallei [20, 21, 22••, 24••]; however, only a few factors were significantly involved and those biotic factors were not included in the studies. Bacteriophages (phages) [34,35,36] and other soil saprophytic bacteria [37] were found to be present in the same ecosystems as B. pseudomallei and showed ability to lyse the bacterium. Phages that could infect B. pseudomallei were also found in soil with the absence of the bacterium indicating the possibility that the phages were affecting the density of this bacterium in the environment [38••]. Phages can act by integrating their DNA into the host genome and cause damage by infecting, multiplying, and lysing the host cell at the end of its propagation. The first lytic bacteriophage for B. pseudomallei was isolated from stagnant water in Hanoi in 1956, but it was not characterized [39]. Later, several groups of researchers isolated and characterized B. pseudomallei phages from the soil environment [36, 37, 38••, 39, 40•]. The ST79 lytic phage, isolated from soil in Khon Kaen, Thailand, gave clear lytic plaques on B. pseudomallei lawn [36]. The lysis cassettes, which the enzymes used to lyse B. pseudomallei, are composed of a holin, a peptidase M15A or endolysin, and the lysB and lysC genes [40•]. Both lytic and lysogenic phages of B. pseudomallei are summarized in Table 2. There are several prophage and prophage-like elements inside the B. pseudomallei genome [44] that may contribute to the genetic and phenotypic diversity of B. pseudomallei, while the lytic phages might be used to control the bacteria [42]. The presence of phages specific for B. pseudomallei may affect the distribution of pathogens in the soil [35]. As bacteriophages have been used for controlling various food pathogens [5, 45, 46], the use of lytic phages as a biocontrol for B. pseudomallei has therefore also been proposed [36, 40•]. One obstacle of this hypothesis, however, is that the balance of the bacteriophage life cycle with the host that may lead to not totally clearing the host- and phage-resistant bacteria [47] may be developed. The use of phage cocktails in other bacteria has been reported to overcome this problem [48, 49]. Modified phage derivatives of ST79 that are specific to B. pseudomallei were developed to increase the lysis ability and reduce the biofilm formation of B. pseudomallei [50]; however, regrowth of the bacterial host was observed after the lysis cycle [36], which emphasizes the problem that has to be overcome if this phage or phage cocktails will be used. Up until now, there is no published work showing the successful use of bacteriophages as biocontrols for B. pseudomallei. In addition to the bacteriophage, various genera of bacteria have been reported to be used as biocontrols such as Agrobacterium, Bacillus, and Pseudomonas [51, 52•, 53, 54]. Most of these are already approved to be used as commercial products. One example of a bacterium is Bacillus amyloliquefaciens [52•], which was successfully used as a biocontrol against plant pathogens. Biocontrol is more likely to be successful for some pathogens than others, and food-borne and plant pathogens are the two main groups where biocontrol is beneficial and for which products are available in the market [55, 56].

Table 2 Bacteriophages that infect B. pseudomallei
Full size table

For melioidosis, in the last 10 years, only a few bacteria were reported to interfere or inhibit B. pseudomallei. The same genera of bacteria which are Burkholderia ubonensis [57] and Burkholderia multivorans [58] were shown to have some antagonistic effects on B. pseudomallei. B. ubonensis produces specific antagonistic compounds active against B. pseudomallei [57]. This compound was found to be a pepsin-sensitive peptide similar to a bacteriocin-like compound that may be purified and characterized [57]. Another bacterium, B. multivorans, was discovered from soil [58]. The antagonist effect was demonstrated by the presence of B. multivorans and was inversely related to the presence of B. pseudomallei [58]. The antagonistic activity of secreted compounds from this bacterium was also reported. As B. multivorans was found in the same area of B. pseudomallei and is more resistant to a broad range of temperature, pH, and salt than B. pseudomallei, the possible use of it as a biocontrol has therefore been discussed. These two bacteria, however, are human pathogens; therefore, the use of B. multivorans for biocontrol of B. pseudomallei may not be possible.

Another candidate for biocontrol of B. pseudomallei is B. amyloliquefaciens, which was isolated from soil that was negative for B. pseudomallei. A few isolates of them have been reported to produce both peptides and non-peptide metabolites that can kill B. pseudomallei and inhibit a broad range of other pathogenic bacteria [37]. The inhibition spectrum was possible from the synergistic effect of the non-protein and peptide compounds. The picture highlighting the active compounds that gave large clear zones on B. pseudomallei culture lawns is shown in Fig. 1. When B. amyloliquefaciens isolates were cocultured with B. pseudomallei, the numbers of the B. pseudomallei bacteria decreased by 5 logs in 72 h [37]. B. amyloliquefaciens produces spores that can tolerate environmental changes and resist desiccation. These spores, when germinated, secrete metabolites that can kill B. pseudomallei, and therefore, it could be used as biological controls for B. pseudomallei in the environment.

Fig. 1
figure 1

Antimicrobial activity of B. amyloliquefaciens against B. pseudomallei. The inhibitory activity of culture supernatants from B. amyloliquefaciens KKU1, KKU3, and KKU14 isolates when tested against B. pseudomallei using the agar well diffusion method and is seen by clear zones. MM minimal medium (control)

Full size image

The use of any microbes for controlling B. pseudomallei in soil should be applied when the soil is able to absorb large amounts of water such as during the rainy season. This will allow interaction of microbes and increase the possibility of biocontrol. The treatment should be continuous, and long-term observation is needed to see changes in the decreases of bacterium in soil due to biocontrol treatment.

Conclusion

Both abiotic and biotic factors in soil were proven to contribute to the presence and absence of B. pseudomallei and may also play a crucial role in providing an uneven distribution of the bacterium in soil. The discovery of bacteriophages and some antagonistic bacteria in soil which can secrete metabolites to inhibit or kill B. pseudomallei is a new research trend and gives potential for biocontrols to reduce the risk of infection when humans and animals are exposed to B. pseudomallei due to contact with contaminated soil and water in the endemic areas.