Effect of Homocysteine on Biofilm Formation by Mycobacteria

Abstract

Mycobacteria show peculiar aggregated outgrowth like biofilm on the surface of solid or liquid media. Biofilms harbor antibiotic resistant bacteria in a self-produced extracellular matrix that signifies the bacterial fate to sedentary existence. Despite years of research, very little is known about the mechanisms that contribute to biofilm formation. LuxS has been previously known to play a role in biofilm formation in Autoinducer-2 dependent manner. We here show the effect of LuxS product-homocysteine, on the biofilm forming ability of non-tuberculous mycobacteria, Mycobacterium smegmatis and Mycobacterium bovis BCG showing AI-2 independent phenotypic effect of LuxS. Exogenous supplementation of homocysteine in the culture media leads to aberrant cording, pellicle outgrowth, and biofilm formation. Thus, our study contributes to the better understanding of the mechanism of mycobacterial biofilm formation and sheds light on the role of LuxS product homocysteine. In addition, we highlight the contribution of activated methyl cycle in bacterial quorum sensing.

Introduction

Biofilms are congregation of heterogeneous population of microorganisms that are surface associated and encapsulated in a self-produced matrix [1]. This architecturally complex structure helps the bacteria to evolve, disseminate and acquire extraordinary level of resistance to most of the therapeutic agents by providing a stable permeability barrier [2, 3]. Biofilm related infections are emerging due to low susceptibility of microbial consortia to antimicrobial drugs and host immune responses [4, 5]. Microorganisms like Candida albicans, Pseudomonas aeruginosa, Staphylococcus epidermidis and Mycobacterium tuberculosis pose serious health concerns due to biofilm mediated resistance to environmental challenges [6–9]. Mycobacterium genera includes many environmental organisms which are non-pathogenic but can cause human infections in special circumstances [10]. These non-tuberculous mycobacteria (NTM) are opportunistic bacteria that can cause chronic infections or outbreaks associated with environmental sources [11]. Biofilms formed by NTMs are tolerant to anti-bacterial agents, thus posing a serious health challenge to control nosocomial infections. Therefore, a proper understanding of the molecular events involved in this transition of planktonic cells to pellicle growth can provide useful insights into drug resistance mechanism [12].

S-adenosyl methionine (SAM) is a major methyl group donor in all the methylation reactions (Fig. 1) [13]. Previous reports have identified Autoinducer-2 (AI-2), a product of LuxS, as an important molecule for interspecies communication in bacteria [14, 15]. However, this interpretation is complicated by dual role of LuxS in signaling and activated methyl cycle (AMC), a crucial metabolic pathway. AI-2 is derived from SAM through a series of enzymatic reactions [16, 17]. SAM forms the toxic intermediate S-adenosyl homocysteine (SAH) on donating methyl group. This S-adenosyl homocysteine (SAH) is further hydrolyzed to S-ribosyl homocysteine (SRH) [18]. Where LuxS cleaves SRH to form homocysteine and AI-2 precursor, S-adenosyl homocysteine hydrolase (SahH) synthesizes homocysteine directly from SAH [19]. LuxS mutant in Streptococcus sanguinis and Escherichia coli has been shown to be defective in biofilm formation. Complementation with SahH and not AI-2 restores this biofilm defect [20], indicating an unexplored role of AMC in bacterial quorum sensing.

Fig. 1.

Activated methyl cycle with homocysteine as a key intermediate. Schematic representation of activated methyl cycle. Homocysteine is synthesized either by a two-step mechanism involving Pfs and LuxS enzyme to produce AI-2 precursor and homocysteine or directly from S-adenosylhomocysteine in a one-step reaction using SahH enzyme

Homocysteine is a key intermediate for de novo synthesis of methionine in the cell which is further utilized for the production of SAM [19]. Role of SahH in maintaining the metabolic flux of SAM and SAH and its regulation through phosphorylation has been previously documented [19]. SahH mediated biofilm restoration of LuxS deficient cells prompted us to understand this concept in detail. Since SahH and LuxS both form homocysteine, we study the effect of exogenous homocysteine supplementation on the biofilm formation of M. smegmatis and M. bovis BCG. Our study explored the involvement of AMC in cording, pellicle formation and quorum sensing. Such phenotypic analysis revealed a novel link between biofilm formation and AMC.

Materials and Methods

Bacterial Growth Conditions

The mycobacterial cultures were grown as described earlier. M. smegmatis mc² 155 and M. bovis BCG cells were cultured and maintained as described before [21–23]. The cells were grown in Middlebrook 7H9 (Difco) broth supplemented with 0.5% glycerol, 0.5% Tween-80 (Sigma) and 1% ADC (bovine serum albumin [fraction V], dextrose, and catalase; Difco) for M. smegmatis and 5% for M. bovis BCG at 37 °C with shaking at 220 rpm for 2–4 days. Middlebrook 7H11 (Difco) was used as solid media containing 0.5% glycerol and 10% OADC (oleic acid, bovine serum albumin [fraction V], dextrose, and catalase) supplement. Effect of homocysteine was studied in Sauton’s minimal media (0.5 g KH₂PO₄, 0.5 g MgSO₄∙7H₂O, 2 g citric acid, 0.05 g ferric ammonium citrate, 60 mL glycerol, 4 g asparagine, 0.1 mL 1% ZnSO₄ and make it to 1 L with distilled water. The pH was adjusted to 7.4 using 1 M NaOH and the media was sterilized before using).

Cording Formation in M. smegmatis

Cording formation was assessed for Mycobacterial cultures as previously described method [24]. M. smegmatis cells were grown till exponential phase in 7H9 media with 0.5% Tween-80. 5 μL of this starter culture was dropped onto a solid agar medium (Difco) with or without homocysteine (Sigma). The plate was then incubated at 37 °C for 5 days. The colonies thus appeared was imaged for the cording phenotype.

Pellicle Formation

The M. smegmatis cells were grown till exponential phase in 7H9 media supplemented with 0.5% Tween-80 and the secondary cultures were inoculated in 5 mL Middlebrook 7H9 media with no Tween at an OD₆₀₀ = 0.01 as described. The cultures were kept at 37 °C for 48 h without shaking for pellicle formation as described [25].

Biofilm Formation in M. smegmatis and M. bovis

The M. smegmatis and M. bovis cells were grown until log phase in Sauton’s minimal media with 0.5% Tween-80 and secondary cultures were inoculated in Sauton’s minimal media in the absence of Tween-80 to an OD = 0.01 and aliquoted in a 96-well plate in triplicate. Homocysteine was added at different concentrations (0.1, 0.2, 0.3, 0.4 and 0.5 mM) for M. smegmatis and (0.1, 0.5 and 0.8 mM) for M. bovis. The plates were sealed and incubated at 37 °C for 5 days without shaking and biofilms were observed in control wells. Images were captured after biofilm development was seen in the control wells as described previously [26].

Results

Homocysteine Affects Cording of Mycobacteria

Macrophages form extracellular traps in response to extracellular aggregates of mycobacteria [27]. Such aggregates are tight bundles or serpentine cords formed by the bacteria. Previous studies reflect the involvement of gene methylation in cellular aggregation. We here study this mycobacterial surface property in the presence of homocysteine. M. smegmatis cells grown on a solid 7H11 agar with varying concentrations of homocysteine (0, 0.3 and 0.5) for 5 days showed distinct colony morphologies (Fig. 2). The bacterial growth showed cords in the center of the colony. The cord formation is altered in the center with increasing concentration of homocysteine. Thus, exogenous homocysteine is seen to modulate the molecular factors which are involved in cord formation, giving rise to colonies with altered cording phenotype.

Fig. 2.

Physiological effects of homocysteine. M. smegmatis mc² 155 cells were grown on solid agar medium containing 0.3 and 0.5 mM homocysteine. The images show modified cording phenotype at the centre of the colonies in the presence of homocysteine

Homocysteine Inhibits Biofilm Forming Ability of M. smegmatis

The effect of SahH complementation on regaining the biofilm forming ability of S. sanguins [28] led us to explore the role of its product-homocysteine in forming biofilm. The lipid rich mycobacterial cell wall forms a surface pellicle in liquid media when grown in the absence of detergent. M. smegmatis mc² 155 cells were grown in a static condition with increasing concentrations of homocysteine (0–0.5 mM) in the absence of detergent to analyze the pellicle formation at air–liquid interface (Fig. 3a). Significant pellicle formation occurs at the surface of broth without homocysteine. With increase in homocysteine concentration, cells become defective in pellicle formation with maximum defect observed at 0.5 mM homocysteine.

Fig. 3.

Effect of homocysteine on M. smegmatis pellicle and biofilm formation. a M. smegmatis cells were grown in liquid media with increasing homocysteine concentration in the absence of Tween-80 at static condition. The images showed decreased pellicle formation on addition of 0.4 mM homocysteine. b M. smegmatis cells were grown in minimal media under non-shaking conditions and in the absence of Tween-80 to develop biofilm. The images show decrease in biofilm as concentration of homocysteine (Hcy) is increased. Biofilm formation was completely absent after 0.4 mM homocysteine

Pellicles have been recognized as biofilm at air–liquid interface. Since, homocysteine addition exhibits aberrant pellicle phenotype, we investigated the consequences of elevated homocysteine on biofilm formation. M. smegmatis cells were grown with increasing concentrations of homocysteine (0, 0.1, 0.2, 0.3, 0.4 and 0.5 mM) in static conditions. M. smegmatis cells in a detergent-free media form biofilm at air–liquid interface. Cells grown in absence of homocysteine showed biofilm formation after 5 days (Fig. 3b). In contrast, with subsequent increase in homocysteine concentrations, bacteria were unable to form biofilm. At 0.4 mM homocysteine, biofilm forming ability of the cells was completely lost. Thus, addition of homocysteine to M. smegmatis cells significantly affect its capacity to form cords and biofilms.

Homocysteine Affects Biofilm Formation in M. bovis BCG

M. bovis BCG is an attenuated form of M. bovis, a member of M. tuberculosis complex and is known to cause bovine tuberculosis [29]. Previous reports suggested M. bovis BCG biofilm formation and its role in disease pathogenesis [30]. Our study show the metabolite mediated abrogation of biofilm formation in M. bovis BCG. The results corroborate with our hypothesis as M. bovis BCG showed similar phenotype as that of M. smegmatis. We used a higher concentration of homocysteine (0.8 mM) with this strain. As we increased the concentration of homocysteine, the cells kept on losing the biofilm forming ability with complete loss at 0.8 mM homocysteine (Fig. 4). Thus, our results show that elevated homocysteine concentrations completely abrogated the biofilm forming tendency of mycobacteria, thus affecting its pathogenesis.

Fig. 4.

Homocysteine affects M. bovis BCG biofilm formation. M. bovis BCG was grown in the same condition as that of M. smegmatis cells. Loss in biofilm formation was observed at 0.8 mM homocysteine

Discussion

The air–liquid interface is a favorable niche for mycobacteria due to increased accessibility to oxygen [31]. Hypoxia induced surface drifting of bacteria forms a biofilm like structure known as pellicle [32]. The molecular basis of this hypha like growth remains ill defined, however, such pellicles provide a drug tolerant phenotype to the bacteria. Mycobacteria forms multicellular community by forming cord like structures [27]. We tested these surface characteristics of mycobacteria through modifying metabolite levels of activated methyl cycle in the cellular system. Thus, addition of homocysteine to the cell media confers a distinct phenotype to the colonies and surface biofilms.

Mycobacterium smegmatis colonies were allowed to form cords in the presence and absence of homocysteine. The normal cording feature in the absence of homocysteine was altered in the center of the colony at 0.3 and 0.5 mM homocysteine. This led us to further examine the pellicle formation in mycobacteria. On titrating homocysteine from 0.1 to 0.5 mM, there is a significant abrogation in the pellicle growth. Since, pellicles have recently been identified as biofilms, we tested the biofilm forming ability of M. smegmatis and M. bovis BCG in presence of homocysteine. Biofilm formation of mycobacteria was reduced when grown in minimal media with exogenous homocysteine supplementation, with a complete loss at 0.5 mM in M. smegmatis and 0.8 mM in M. bovis BCG. For M. bovis BCG, the biofilm defect was observed at a higher concentration of 0.8 mM which could be due to its slow metabolism or higher pathogenicity. Virulent bacteria showed decreased cell wall permeability leading to less effective transport of extracellular substances [33].

Biofilm formation in mycobacteria has been extensively studied with many breakthroughs to develop major genetic tools [34]. There have been major implications of biofilm formation in bacterial virulence and pathogenesis. Biofilm defective mutants of Mycobacterium avium failed to colonize and translocate through bronchial epithelial cells [35]. Also, Mycobacterium ulcerans colonizes the host and disseminate using biofilm which causes the pathological feature of Buruli ulcers [36, 37]. Thus, a proper knowledge of such underlying mechanisms of virulence is a major concern. The lipid rich mycobacterial cell wall consists of mycolic acids in the inner; and glycolipids and phospholipids on the outer surface giving a distinct surface characteristic to mycobacteria [38]. During the course of biofilm formation, mycobacteria alters the mycolic acid composition from long chain fatty acids (C₇₀–C₉₀) in free living planktonic state to shorter chain version (C₅₆–C₆₈), possibly to aggregate and develop into a mature biofilm [2]. A recent study has established the role of GroEL in the mycolic acid biosynthesis thus affecting biofilm formation [26, 39]. Mycolic acid methylation through Mma4 has a major role in M. tuberculosis virulence as mma4 mutants were susceptible to IL-12 dependent killing [40]. The methylated sugars and lipids on the mycobacterial cell wall confer hydrophobicity to the cell [41]. Such methylated sugar containing glycopeptidolipids are important for biofilm formation and colony morphology [42, 43].

Such methylation events occur in a SAM-dependent manner, which are required for chemotaxis, cell growth and development. Affecting SAM or methionine biosynthesis have been known to affect infection related processes [44, 45]. Homocysteine rules an important metabolic status where it is required for metabolite synthesis such as SAM, SAH, adenosine, methionine and cysteine and its upregulation either through SahH or its artificial supplementation in the medium directly affects cellular level of methylation. Eukaryotic SahH has been well known to regulate RNA or DNA methylation [46, 47]. SahH mediated biofilm formation in LuxS mutant and disruption of biofilm in a homocysteine supplementation medium hints the involvement of methylation in bacterial surface characteristics and thus quorum sensing.

Since biofilms are antibiotic tolerant [48] and are less susceptible to host immune responses, developing treatments against them requires a deep understanding of their physiological traits. We found that dysregulation of methionine synthesis through homocysteine is what impairs the biofilm formation in cultures with increased homocysteine. Thus, this defect in biofilm formation due to additional homocysteine shows the quorum sensing contribution of the methyl cycle.

Abbreviations

SAM

S-adenosyl methionine

AI-2

Autoinducer-2

AMC

Activated methyl cycle

SAH

S-adenosyl homocysteine

SahH

S-adenosyl homocysteine hydrolase

NTM

Non-tuberculous mycobacteria

SRH

S-ribosyl homocysteine

Author’s Contributions

RV conceived and designed the experiments. YH and YS contributed materials, reagents and analysis tools. All authors have read and approved the manuscript.

Funding

This work was supported by the Council of Scientific and Industrial Research (CSIR), India, J.C. Bose fellowship (SERB) (to Y.S.); and CSIR senior research fellowship (to R.V.).

Compliance with Ethical Standards

Competing interest

The authors declare that they have no competing interest.