Discussion
Burkholderia is a genus of proteobacteria which harbors pathogenic and non-pathogenic members.B. pseudomallei is a gram-negative, aerobic, highly pathogenic rod-shaped bacterium. It is the causative agent for melioidosis, an infectious disease with high morbidity and mortality rates in endemic regions [53]. It is a CDC federal select agent for which there are no current vaccines available for human treatment. Molecular research on Burkholderia has focused on vaccine production mainly because of its likely use in bioterrorism but also because it exhibits intricate resistance to multiple groups of antibiotics, especially while growing in biofilms [54]. Because of the pathogenicity of B. pseudomallei , a BSL3 lab is required to work with the bacteria hence manipulations have been made to generate B. pseudomallei strains 1026b Δasd , Bp82, JW270 which are all non-pathogenic BSL2 strains. On the other hand,B. thailandensis is a non-pathogenic strain which rarely causes disease in humans [35, 55]. Most Burkholderia biofilm studies report on biofilm using BSL2 strains. The purpose of this study was to determine if the genetic manipulations made to the pathogenic 1026b strains affected biofilm structure, formation or constituents.
Gram-negative bacteria can form surface adherent biofilms or a biofilm formed at the air-liquid interface which is called a pellicle [1]. The biofilm formation process is activated in response to various signals such as quorum sensing [56, 57], nutrient availability [58], stress [59], and surface material as reviewed in Maganaet al , 2018 [60]. Research has shown that B. pseudomallei and B. thailandensis are capable of producing biofilms [10, 11, 13, 17, 18, 21, 61, 62] and are known to export large amounts of eDNA [25, 63]. Another biofilm characteristic is the ability to persist after treatment which causes disease relapse [54, 64] - a feature widely attributed to a cohort of biofilm cells called persisters. Biofilm is involved in the antibiotic resistance of bacterial disease, relapse, and persistent cells [65, 66]. eDNA exportation has been observed in both Gram-positive and Gram-negative bacteria and has now been associated with biofilm facilitation and composition [23, 67].
Much like Salmonella [68], Burkholderia is capable of producing either a surface adherent biofilm or a pellicle (Fig 1a-c). This phenomenon was also demonstrated in a previous Burkholderiastudy which showed both pellicle and adherent biofilm [23]. The conditions and reasons for the change from surface adherent to pellicle biofilm remains unknown in Burkholderia . Mangalea et al . [8] reviewed that Burkholderia pellicles are typically present in the environment and surface adherent biofilms more consistent with rice paddies. This is quite similar to conditions observed in this study, 37°C representing the higher environmental temperature and 37°C a representative temperature for the cooler rice paddies. This suggests that temperature amongst other environmental conditions indicative of the two different environments may play a role in determining the type of biofilm formed. In Salmonella , curli is responsible for the switch [68] whereas, in C. crescentus, biofilm is initially formed as a monolayer of cells that subsequently develop into a three-dimensional structure and as the biofilm matures, it becomes more cohesive and less adherent [69]. Previous studies have confirmed that some pellicles of some organisms are composed of eDNA with close cell contact [23], sucrose, and proteins [70].
This study characterized the major biofilm components of B. pseudomallei 1026b, four isogenic 1026b mutants and B. thailandensis E264. Since B. pseudomallei is typically found in the soils with relatively lower temperatures than a mammalian host, we grew all strains at temperatures that reflect each environment. Fig. 1d shows that there were no temperature-related differences in biofilm formation between the strains grown at RT and 37°C, so all remaining experiments were performed at 37°C. Interestingly, the amount of biofilm formed by all the B. pseudomallei and B. thailandensisstrains at 72 hours showed no significant difference (Fig. 1e).
Fluorescent microscopy and molecular assays (Fig. 2a and 3) show that the biofilm composition of B. thailandensis is significantly composed of proteins and eDNA and lower levels of polysaccharide.B. pseudomallei 1026b Δasd and Bp82 consists mainly of polysaccharides with lower protein and eDNA levels, respectively. However, JW270 appears to have consistently low protein and polysaccharide concentrations but relatively large amounts of eDNA. JW270 is a CPS I mutant [40] and is unlikely to produce the same amount of polysaccharides as 1026b Δasd or Bp82. We speculate that in order to compensate for the loss of polysaccharide, there is an increased export of eDNA to serve as a skeletal biofilm structure. To confirm our speculations, we compared the concentration of the biofilm constituents of DD503, the parental strain of JW270 which lacksamrAB-oprA , a multidrug efflux system (Fig. S1). The deletion of the AmrAB-oprA efflux pump alone did not affect concentrations of biofilm protein or eDNA, but significantly increased the glucose concentration (Fig 6). JW270 was generated by deleting the wcbcluster responsible for CPS I polysaccharide in DD503 [40]. Deletion of both the AmrAB-oprA efflux pump and the wcb operon led to the significant decrease in both biofilm protein and polysaccharide, but an inverse increase in the amount of eDNA exported and found in a biofilm (Fig. 6).
The differences observed in the molecular composition were not attributed to the extraction method or cell lysis during extraction. A plate count was performed before and after incubation with TE buffer to evaluate cell lysis and the number of cell count obtained before and after extraction remained constant. Florescence live/dead ratio was also carried out and Fig 2a shows that the biofilm harbor viable cells. It is also noted that as with most biofilms, these biofilms are subject to enzymatic degradation. Proteinase K/glucosidase had the most eradicative effect on E264 while DNase seemed to have a more profound effect on theB. pseudomallei derivatives.
In Fig. 1b-c, we observed the pellicles of Burkholderia . Morphologically, JW270 and Bp82 produced thicker pellicles, but 1026bΔasd and E264 formed thinner pellicles. Microscopically in Fig 1c, the pellicles look like mesh work anchored to the walls of the wells and also shows adherent cells attached to it. In the fluorescent image in Fig. 5a, florescent dyes were used to further illustrate the composition of each strain. The data from Fig. 5a supported by the assays used in Figs. 5b and 5c, also showed that the composition of adherent biofilms (Figs 2a and 3) and pellicles (Fig. 5a-c) are quite similar.
Morphological and chemical differences exist between the biofilms ofB. thailandensis and the B. pseudomallei derivatives 1026bΔasd and Bp82, but the JW270 biofilm exhibited similarities to the E264 biofilm. We posit that the differences between B. pseudomallei JW270 and 1026b Δasd or Bp82 is related to the capsule deletion necessitating the need for JW270 to export higher amounts of eDNA required for biofilm formation as observed in B. thailandensis E264 biofilm. This hypothesis is backed by fig 6a-c which shows that asides from glucose, the concentration of protein and eDNA remain unchanged between B. pseudomallei 1026b and DD503 but once the wcb operon is deleted in DD503 to generate JW270, there is a decrease in protein content but a higher export rate for eDNA. Glucose level In DD503 increased in comparison to 1026b but drops to baseline in JW270 keeping in mind that the wcb operon responsible for CPS I polysaccharide. An ELISA test confirms that the difference betweenB. pseudomallei 1026b Δasd /Bp82 and B. pseudomallei JW270/ B. thailandensis E264 is the presence of a capsule. An anti-capsule antibody reacted with B. pseudomallei1026b Δasd /Bp82 but not B. pseudomallei JW270/B. thailandensis E264. Biofilm assay confirmed that there was no difference in biofilm formation between all strains used.
In summary, we propose that the lack of the wcb operon is the driving force behind the large amounts of eDNA exported by JW270. These differences in biofilm composition may play a role in the immune response of host organisms and also illustrate the survival of persistent cells that cause disease relapse in a host organism.