New Insight into Vitamins E and K₁ as Anti-Quorum-Sensing Agents against Pseudomonas aeruginosa

ABSTRACT

Today, antivirulence compounds that attenuate bacterial pathogenicity and have no interference with bacterial viability or growth are introduced as the next generation of antibacterial agents. However, the development of such compounds that can be used by humans is restricted by various factors, including the need for extensive economic investments, the inability of many molecules to penetrate the membrane of Gram-negative bacteria, and unfavorable pharmacological properties and cytotoxicity. Here, we take a new and different look into two frequent supplements, vitamin E and K₁, as anti-quorum-sensing agents against Pseudomonas aeruginosa, a pathogen that is hazardous to human life and responsible for several diseases. Both vitamins showed significant anti-biofilm activity (62% and 40.3% reduction by vitamin E and K₁, respectively), and the expression of virulence factors, including pyocyanin, pyoverdine, and protease, was significantly inhibited, especially in the presence of vitamin E. Cotreatment of constructed biofilms with these vitamins plus tobramycin significantly reduced the number of bacterial cells sheltered inside the impermeable matrix (71.6% and 69% by a combination of tobramycin and vitamin E or K₁, respectively). The in silico studies, besides the similarities of chemical structures, reinforce the possibility that both vitamins act through inhibition of the PqsR protein. This is the first report of the antivirulence and antipathogenic activity of vitamin E and K₁ against P. aeruginosa and confirms their potential for further research against other multidrug-resistant bacteria.

INTRODUCTION

During the Golden Age of antibiotics, starting with the discovery of penicillin, many antimicrobial compounds were gradually introduced, but, due to the improper and long-term use of antibiotics, microbial resistance soon emerged and developed worldwide, leading to the worrying growth of difficult-to-treat infections (1, 2). It is estimated that every year, almost 700,000 people die in the world as a result of antibiotic resistance, which may increase to 10 million by 2050 if nothing is done to rectify this health issue (3). In addition to the development of antibiotic resistance, the ability of some bacteria to form biofilms facilitates refractory infection and complicates medical treatments (4, 5). Biofilms are bacterial consortia that attach to and grow on various surfaces inside a complex self-produced polymeric matrix. This complex structure protects the residing cells from immune system responses and offers a site for attachment of other cells. Bacteria within mature biofilms can be 1,000 times more resistant to antibiotics than planktonic forms or free-floating cells (6). In 2017, the World Health Organization (WHO) introduced a list of pathogens hazardous to human life to guide research and development about new antibiotics. One of these pathogens of critical priority is Pseudomonas aeruginosa (7).

P. aeruginosa is a nonfermentative, aerobic, Gram-negative bacterium belonging to a group of pathogen species called ESKAPE (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, P. aeruginosa, and Enterobacter) (2, 3, 8). It is one of the main pathogens associated with opportunistic and nosocomial infections. The infection can become an especially serious threat in patients with cystic fibrosis (CF), critical burns, AIDS, and immunocompromised hosts with endocarditis, sepsis, and so on (3, 9, 10). Clinically, the treatment of P. aeruginosa infections has been associated with difficulties because of its metabolic adaptability supported by various virulence factors and enzymes, significant resistance to antibiotics, and biofilm mode of colonization (1, 3).

In 2012, it was reported that P. aeruginosa infection occurs in about 30% and 60% of CF patients aged 6 to 10 and 18 to 24 years, respectively. Although recent studies show that the rates are declining, P. aeruginosa has remained a more important CF pathogen than the others have and has proved difficult to eliminate. In these patients, current treatments to control chronic P. aeruginosa infection include prescribing tobramycin inhalation solution or aztreonam solution as well as nonantibiotic CF therapies. However, prescribing inhaled antibiotics has limited clinical value due to the low penetration into infection sites, the impenetrability of biofilms, inactivation by sputum, bacterial phenotypic alterations, and more (11). Therefore, the discovery of new antibacterial compounds is not enough, and there is an imperative need to find alternative approaches to counter P. aeruginosa.

Compounds with no intervention in bacterial growth or viability mechanisms may be ideal and induce less resistance than conventional antibiotics, since they attenuate pathogens rather than kill them (1, 12, 13). A promising and attractive strategy relies on targeting an intercellular communication mechanism known as quorum sensing (QS) (4, 14). Broadly speaking, the QS mechanism coordinates a series of functions between bacteria as a community through small chemical molecules known as autoinducers. QS interferes in many vital biological behaviors, including adaption, expression of virulence factors, biofilm development, antibacterial resistance, and bacterial motility (1, 15). Selective inhibition of one or several particular QS pathways in P. aeruginosa has long been suggested to disrupt pathogenicity without affecting viability and developing bacterial resistance (15). The other benefit of the QS inhibition strategy is the avoidance of the possible destruction of the valuable gut microbiota (4).

Generally, the P. aeruginosa QS network employs three major regulatory cascades, known as lasI/R, rhlI/R, and pqs, which determine the production and release of many virulence factors, including pyocyanin, protease, lipase, elastase, rhamnolipids, etc., and biofilm development (8, 16). Overall, more than 350 genes are regulated by QS in P. aeruginosa, of which almost 30% encode toxic virulence factors (4). Each cascade performs its regulatory functions through a major related protein: LasR, RhlR, or PqsR (MvfR). Moreover, in each regulatory pathway, a specific signaling molecule acts as an autoinducer to activate the corresponding protein: 3-oxo-C12-homoserine lactone (3-oxo-C12-HSL), C4-homoserine lactone (C4-HSL), and 3,4-dihydroxy-2-heptylquinoline (PQS), respectively. Additionally, the metabolic precursor of PQS, 2-heptyl-4-hydroxyquinoline (HHQ), can bind to the PqsR protein (Fig. 1). While the interaction between PQS/HHQ and PqsR takes place, the expression of several genes was triggered, including the transcription of pqsABCDE operon, which, in turn, leads to the release of various virulence factors, such as pyocyanin, elastase, rhamnolipids, lectins, and HCN, in addition to affecting biofilm formation (1, 8, 17, 18). This interaction also has positive feedback on PQS/HHQ biosynthesis, which increases the concentration of these signaling molecules exponentially (19). As a consequence of the central role of PqsR in the P. aeruginosa QS network, inhibition of this protein by small molecules structurally similar to its autoinducers could reduce both pathogenicity and antibacterial resistance (1). The present study aimed to investigate the probable anti-QS activity of vitamin E (alpha-tocopherol) and vitamin K₁ (phytonadione) on the P. aeruginosa PAO1 strain. Based on the chemical structural similarity of these vitamins to PQS/HHQ (Fig. 1), we hypothesized that they reduce the pathogenicity of P. aeruginosa by interfering with the interaction of PqsR and its autoinducers.

FIG 1.

Chemical structure of autoinducers involved in the major regulatory cascades of the Pseudomonas aeruginosa quorum-sensing (QS) network (a, 3-oxo-C12-HSL; b, C4-HSL; c, PQS; d, HHQ) and the tested compounds (e, vitamin E; f, vitamin K₁).

RESULTS AND DISCUSSION

Finding and developing novel pharmaceuticals that can be used by humans is a challenging mission due to the decade-long laboratory investigations and subsequent numerous clinical trials. This process is laborious and requires extensive economic investments without a guarantee of success, generally due to the poor pharmacological and pharmaceutical characteristics of newly discovered bioactive compounds (13). During recent decades, many compounds targeting the P. aeruginosa QS network have been identified and evaluated both in vitro and in vivo, and their ability as antivirulence agents has been confirmed. However, most of these anti-QS molecules are not suitable to use as medications because of unfavorable pharmacological properties or cytotoxicity (2, 13). Therefore, based on the chemical structural similarity of vitamin E/K₁ to the natural ligands of PqsR, we chose these vitamins as almost-safe anti-QS agents, and the selected concentrations were considered within pharmaceutical ranges or less.

First, the probable MIC of each vitamin was estimated. MIC was considered the minimum concentration of the tested compounds with inhibitory effects against growth promotion of P. aeruginosa. Therefore, the minimum concentration of each pharmaceutical without a change in red color was selected as the MIC (20). The MIC values were determined as 5 mg·ml⁻¹ for vitamin E (dl-alpha-tocopherol), 2.5 mg·ml⁻¹ for vitamin K₁, and 2 μg·ml⁻¹ for tobramycin.

The effect of each vitamin on the biofilm formation of P. aeruginosa was then evaluated by colorimetric assay using crystal violet. As shown in Fig. 2, compared to the positive control, biofilm formation was partially inhibited by both vitamins (62% by vitamin E and 40.3% by vitamin K₁ at the MIC as well as 36.7% by vitamin E and 63.1% by vitamin K₁ at 1/2 MIC).

FIG 2.

Bar diagram indicating reduced biofilm formation of Pseudomonas aeruginosa PAO1 in the presence of vitamin E and K₁. *, P  < 0.05.

This favorable effect was also observed against the production of virulence factors. During the invasion and spread of P. aeruginosa, the activities of virulence factors play an important role in extensive tissue damage. The greater activities of these factors make the infection symptoms more severe and the treatment process more difficult (5). As displayed in Fig. 3, the production of pyocyanin (74.6% and 60.31%), pyoverdin (60.6% and 59.5%), and protease (86.5% and 42.5%) was significantly inhibited in the presence of vitamin E and K₁, respectively.

FIG 3.

Bar diagram indicating significantly reduced virulence factor production. (A) Pyocyanin. (B) Pyoverdine. (C) Protease. *, P  < 0.05.

The most interesting results were obtained in the simultaneous use of the vitamins and tobramycin. Biofilms are almost impenetrable to antibiotics after they have formed. However, the combination treatment of bacteria within the biofilms by tobramycin and vitamin E/K₁ was effective, and the population of live bacteria significantly decreased (71.6% by a combination of vitamin E and tobramycin and 69% by a mixture of vitamin K₁ and tobramycin), which resulted in the lessening of the indicator red color (Fig. 4). The results also indicated that tobramycin could not affect biofilm bacteria per se, as the absorbance was not significantly different from that of the control (only 4.2% reduction).

FIG 4.

Bar diagram indicating significantly reduced population of live bacteria. *, P  < 0.05.

The lower vulnerability of the cells within the biofilm to the antibiotic can be due to several factors, including altered metabolic states caused by nutrients and oxygen distribution gradients, expression of specific genes, and extracellular matrix biogenesis and composition, which depend on environmental conditions, such as access to nutrients, the incidence of stressors, and interbacterial communication. This protective scaffold establishes particular environments for the exchange of genetic material between them, especially antibiotic resistance genes. On other hand, the extracellular matrix components can interact with antibiotics and limit the flow of accessibility to the biofilm members. For instance, it is reported that the presence of alginate, a polyanionic exopolysaccharide of Pseudomonas biofilm construction, protects them from aminoglycosides due to the ionic bonds with these positively charged antibiotics. Moreover, the cyclic glucans existing in Pseudomonas biofilms prevent the action of the aminoglycoside kanamycin through a similar mechanism. In strains that do not produce alginate, the two other polysaccharides, Pel and Psl, participate in biofilm architecture. Although these polysaccharides are cationic molecules, they interact with eDNA, which is negatively charged and can bind with aminoglycosides indirectly (21).

The adherence and colonization of bacteria to medical and implant devices in the presence of vitamin E was investigated by Gómez-Barrena et al. They tried to improve the orthopedic material capacity in bacterial adhesion and the resulting infection. They used alpha-tocopherol doped or blended ultrahigh-molecular-weight polyethylene (VE-UHMWPE) to examine the adsorption of collection or clinical strains of S. aureus and Staphylococcus epidermidis as the most common bacterial species isolated from orthopedic infections. After several experiments, the authors found that although collection strains showed fewer tendencies to adhere to VE-UHMWPE, a number of clinical strains did not support this effect, and the results represented intrinsic intra- and interspecies differences (22). An assessment by Banche et al. about the adhesive properties of S. aureus and Escherichia coli, as the most frequent bacteria in periprosthetic joint infection, in several kinds of UHMWPE also showed a significant reduction in bacterial adhesion on vitamin E-blended UHMWPE (23). Banche and her colleagues came to the same conclusion about the anti-adherent performance of vitamin E-blended UHMWPE against Candida albicans in another study (24). Campoccia et al. found similar results when they challenged poly-(d,l) lactic acid (PLA) blended with vitamin E/vitamin E acetate with the biofilm-producers S. aureus and S. epidermidis and introduced both enriched polymers as gentle anti-infective biomaterial (25). Nevertheless, these studies generally focused on the anti-adhesive and anti-biofilm role of UHMWPE rather than on vitamin E.

In another study, led by Lee et al., the effectiveness of two biodegradable hydrogels from vitamin E-functionalized polycarbonates was evaluated. The results revealed that both hydrogels were effective against both Gram-positive and Gram-negative bacteria. Similarly, these hydrogels were efficient in the eradication of biomass and viability reduction of S. aureus and E. coli residing in the biofilms (26).

On other hand, Harper et al. produced soft nanomaterials from alpha-tocopherol (α-T) and alpha-tocopherol phosphate (α-TP) to understand their antimicrobial activity against biofilm development of Streptococcus oralis and Streptococcus mutans as saliva bacterial species. In fact, their aim was to fabricate soft nanomaterials with the ability to bind to the teeth, which facilitate and control the penetration of antimicrobial used in the oral biofilms. Their observations exhibited that (+) α-TP could bind strongly to the teeth and had reasonable bacteriostatic and biofilm penetration characteristics with superior activity against S. oralis compared to S. mutans (27). Moreover, Ebersole et al. believed that patients with lower levels of alpha-tocopherol than the normal range confront an increased severity of periodontitis, which is a result of unregulated responses of the host immune system against the chronic bacterial biofilm trouble (28). However, in an overview, most of these investigations showed indications of the natural biological antioxidant properties of vitamin E than of its antimicrobial effect.

Recently, Vergalito et al. studied the role of vitamin E in the reduction of biofilm formation in a wide range of human pathogens implicated in the onset of health care-associated infections, including P. aeruginosa, Pseudomonas putida, E. coli, K. pneumonia, A. baumannii, Proteus mirabilis, S. epidermidis, and S. aureus. Moreover, the effect of vitamin E against the genus Staphylococcus colonization and infection of urinary catheters was investigated. They finally concluded that vitamin E could interfere with and reduce bacterial biofilm formation of all the strains tested. Furthermore, the colonization of S. aureus and S. epidermidis on the lumen of a silicone catheter decreased after vitamin E application. However, the authors emphasized that further studies are necessary to better clarify the antibacterial mechanisms of this vitamin (29).

As far as we know, the role of vitamin K₁ in biofilm formation has not been studied before. However, several studies have dealt with the other types of vitamin K (including vitamins K₂ and K₃). Singh et al. described menadione (vitamin K₃) as the production stimulant for polysaccharide intercellular adhesin (PIA) in menadione-auxotrophic S. aureus small-colony variants (SCVs). This process results in the upregulation of EPS matrix production in S. aureus SCVs, which makes them more resistant to antibiotics and cell wall inhibitors as well as having more biomass content with larger microcolonies than wild-type biofilms (30). On the other hand, Mashruwala et al. found that the absence of menaquinone (vitamin K₂) influence on a protein of S. aureus, called SrrAB, increased biofilm formation and cell lysis. In other words, the absence of menaquinone makes SrrAB nonresponsive (31). This effect of menaquinone was confirmed by Choi et al. They targeted the menaquinone biosynthesis pathway through a series of synthetic compounds to reduce methicillin-resistant S. aureus (MRSA) growth. In contrast, menaquinone supplementation neutralized this effect and rescued MRSA growth. Some of these compounds also exhibited inhibitory activity against the growth of MRSA biofilm at high concentrations (32).

On the contrary, after evaluation of vitamin K treatment on the expression of two genes (icaA and icaR) that are involved in biofilm formation of methicillin-resistant S. aureus, Pasandideh et al. introduced vitamin K as the inhibitor of icaA and icaR genes and subsequent biofilm formation (33).

Finally, the in silico study supported our in vitro experiments (Table 1). The molecular modeling of the protein PqsR-vitamin E/K₁ interaction revealed that both vitamins interact more effectively than natural ligands (PQS/HHQ). Moreover, as expected, vitamin E had lower binding energy and inhibition constant, which could explain its more antibiofilm and antivirulence factor effects. It should be noted, however, that these results were obtained under static conditions, and the protein was considered a rigid form (Fig. 5).

TABLE 1.

Computational factors calculated through in silico study of the interaction between PqsR and vitamin E/K₁

Computational factor	PQS	HHQ	Vitamin E	Vitamin K1
Binding energy (kcal/mol)	−7.38	−7.54	−8.13	−7.71
Inhibition constant (μM)	3.91	2.98	1.1	2.22
Hydrogen bond partner	Ile 236	Leu 197	Ile 236

FIG 5.

Polar (left) and nonpolar (right) interactions of PQS (A), HHQ (B), vitamin E (C), and vitamin K₁ (D) with the active site of PqsR.

Conclusions.

In conclusion, altogether, our results presented that vitamin E and K₁ significantly suppress biofilm formation and virulence factor production of P. aeruginosa. This antipathogenic effect is important, especially due to the crucial role of P. aeruginosa in many opportunistic infections. On the other hand, the outcomes were achieved in the usual and pharmaceutical concentrations of two supplements that are frequently consumed. The molecular modeling tools also supported the hypothesis that both vitamins act by means of the inhibition of PqsR as well as provide a prospect of this interaction. However, further studies are needed to clarify the exact mechanism(s) of action related to these vitamins.

MATERIALS AND METHODS

Chemicals and materials.

P. aeruginosa PAO1 (Nottingham wild type) was used in all experiments. All tests were conducted in Mueller-Hinton agar (MHA) and Mueller-Hinton broth (MHB) from HiMedia (India). Tobramycin was purchased from Sigma-Aldrich (USA). Vitamins E and K₁ were acquired from Osve Pharmaceutical Company, Iran. Other chemicals, including 2,3,5-triphenyltetrazolium chloride (TTC), chloroform, HCl, and skimmed milk, were obtained from Merck (Germany).

Determination of the MIC.

The MIC values of vitamins E and K₁ and tobramycin against P. aeruginosa PAO1 were determined using the serial dilution method. Serial concentrations of vitamin E (10, 5, 2.5, 1.25, 0.625, and 0.312 mg·ml⁻¹), vitamin K₁ (5, 2.5, 1.25, 0.625, and 0.312 mg·ml⁻¹), and tobramycin (16, 8, 4, 2, 1, 0.5, and 0.25 μg·ml⁻¹) were prepared in MHB, and 180 μl of each dilution was inoculated separately in a 96-well culture plate. To each well containing the pharmaceutical, 20 μl of the P. aeruginosa PAO1 isolate (10⁶ CFU/ml) was added. The plates then were incubated at 37°C. After 24 h, the water-soluble salt, TTC, was appended (5 mg·ml⁻¹) to each well as a biological indicator, and the plate was incubated for a further 3 h.

Biofilm assay.

The commonly used crystal violet (CV) staining procedure was applied for the biofilm mass assay (34). Briefly, 180 μl MHB, enriched by 2.5% glucose and containing MIC or 1/2 MIC of vitamin E/K₁, was inoculated in a flat-bottomed 96-well plate, and then 20 μl of the provided bacterial suspension (10⁶ CFU/ml) was added to each well. At least three repetitions were considered for each compound. Both negative (MHB enriched by glucose) and positive (MHB supplemented with glucose and contaminated with P. aeruginosa) controls were also included. The culture plate was incubated for 48 h; however, the medium was replaced with fresh enriched MHB after 24 and 36 h, consecutively.

The bulk liquid of each well was then removed completely and washed twice with normal saline (NS). It was exposed for air drying, and then 50 μl 0.03% CV was added to each well. The contents of each well were removed completely after 10 min, and the wells were again washed twice with NS. Finally, 100 μl 96% ethanol was decanted and incubated for 10 min at room temperature (RT). The constructed biofilm was evaluated by absorbance at 590 nm and compared with the controls.

Virulence factor assays.

The production of pyocyanin, pyoverdin, and protease in the presence of vitamins E and K₁ was assessed as described in the literature using standard methods, with slight modifications (35–37).

(i) Pyocyanin assay.

A single colony was removed from the overnight culture of P. aeruginosa PAO1 at 37°C and transferred into 15 ml MHB enriched by 2.5% glucose. Following 24 h of aerobic growth at 37°C with shaking at 120 rpm, cultures were centrifuged, the supernatant was removed, and the cells were again suspended in 10 ml fresh MHB containing glucose and the MIC of vitamin E/K₁. The new cultures were incubated for an additional 24 h under aerobic conditions as mentioned above. For each compound, at least three repetitions were considered.

Pyocyanin was extracted by adding 5 ml chloroform (divided into 2, 2, and 1 ml) and subsequently reextracted with 2 ml HCl 0.1 M (divided into 1, 0.5, and 0.5 ml) from the organic phase. The optical density of the resulting pink solution was measured at 385 nm.

(ii) Pyoverdine assay.

As described above, a colony of P. aeruginosa PAO1 bacteria was added to 10 ml MHB enriched by glucose and incubated for 24 h at 37°C. The bacterial cultures then were centrifuged, and the cells were resuspended in 5 ml glucose-enriched MHB containing the MIC of vitamin E/K₁. The cultures were incubated for an additional 12 h using the shaker incubator. At least three repetitions were taken for each compound. The culture flasks were also covered to avoid light. Next, 200 μl supernatant related to each compound was inoculated to the dark 96-well plate. The fluorescence intensity was recorded immediately (excitation wavelength, 400 nm; emission wavelength, 460 nm) and reported as relative fluorescence units (RFU).

(iii) Protease assay.

The assessment of the protease activity of P. aeruginosa supernatant was performed using skimmed milk. P. aeruginosa was cultured as indicated previously. The supernatant of the last culture that was grown in the presence of vitamin E/K₁ (200 μl) was added to the tubes containing 1.25% skimmed milk (1 ml). The tubes were incubated at 37°C for 2 h. For each compound, at least three repetitions were considered. The proteolytic degradation of skimmed milk was quantified at 500 nm by measuring the alteration in turbidity.

Combination treatment of P. aeruginosa by vitamin E/K₁ and tobramycin.

The most effective concentrations of vitamins E and K₁ were used in combination with tobramycin. The test steps were similar to the biofilm assay. Briefly, the bacterium was cultured in a flat-bottomed 96-well plate for 48 h at 37°C. The enriched MHB was replaced with fresh medium after 24 h and 36 h. In the last 12 h, the medium was replaced with enriched MHB containing the MIC of tobramycin and one of the vitamins simultaneously. The culture medium containing compounds was then removed completely, and 200 μl of glucose-enriched culture medium containing TTC (5 mg·ml⁻¹) was added to each well and incubated for another 4 h at 37°C. At least three repetitions were considered for each compound. The intensity of the resulting red color was considered an indicator of the living cell number and measured at 450 nm. The results were compared with the controls.

In silico studies.

To form a better understanding of the disrupting role of vitamins E and K₁ in the P. aeruginosa QS network, we examined the interaction of these molecules with PqsR protein using AutoDockTools software (version 1.5.4) (38).

Each ligand molecule was primarily drawn and converted to a three-dimensional (3D) structure with ChemOffice Professional 16 (Perkin Elmer, USA). The molecular energy then was minimized with HyperChem 7.0 software using the AM1 semiempirical method. The optimized energy molecules were considered flexible molecules with free rotation and docked in the PqsR active site.

The 3D structure of the protein was obtained from Protein Data Bank (PDB) code 4JVD (39). The cocrystallized native ligand was removed from the PqsR active site. The protein molecule was improved by adding all polar and nonpolar hydrogen atoms as well as Kollman charges. The grid box was adjusted to the active-site amino acids, and the docking parameter files were generated by means of the Lamarckian genetic algorithm. The number of runs was adjusted to 50 replications without changing the other settings.

The results of the docking calculation were ordered based on binding energies. Finally, the conformations and interactions of the protein molecule were evaluated by ViewerLite 4.2.

Statistical analysis.

The results of each experiment were analyzed and compared by the one-way analysis of variance method (Tukey’s or Tamhane's T2 test) using SPSS software (version 22). The statistical significance level was set at a P value of <0.05.