It is therefore more than urgent to renew the therapeutic arsenal and strategies against bacterial infections, particularly by developing:

(i) New antibiotics with new mechanisms of action, with the high probability of selecting in the short or medium term some resistant strains to these new antibiotics (León-Buitimea et al., 2020)

(ii) Alternatives to antibiotics (bacteriocins, bacteriophages, immunotherapies, antimicrobial peptides, predatory bacteria, inhibitors of bacterial virulence or resistance factors) with less impact on the selection of resistant strains (Hauser et al., 2016; León-Buitimea et al., 2020; Shao et al., 2020).

Inhibiting bacterial resistance factors/genes expression or mechanism seems to be a promising approach to restore bacterial susceptibility to already available antibiotics (Laws et al., 2019; Melander and Melander, 2017). On the other hand, reducing bacterial pathogenicity by inhibiting the production of virulence factors would benefit the infected organism by giving its immune system more time to fight the infection (Ruimy and Andremont, 2004). It is established that genes expression involved in the production of bacterial resistance and virulence factors is under the control of a cell-to-cell communication termed quorum sensing (QS), which is a mechanism used by bacteria to detect their critical cell numbers through producing and perceiving diffusible signal molecules in order to coordinate a common behaviour (Allegretta et al., 2017; Shao et al., 2020). As a result, QS inhibitors have been identified as a promising solution for controlling multi-drug resistant pathogens (Lamin et al., 2022).

Pseudomonas aeruginosa, is one of the multi-drug resistant pathogens listed by WHO and classified in the most critical pathogens group. In spite of the advances in antimicrobial therapy, P. aeruginosa infection remains associated with high mortality, from 18 to 61% (Kim et al., 2014; Zhang et al., 2020). P. aeruginosa was selected for investigation, as the mechanism that controls the transcription  of  genes  involved  in  its  virulence  is  well known and is controlled by quorum sensing. In P. aeruginosa, QS coordinates AHL-mediated expression of virulence genes. It is a key opportunistic pathogen that leads to severe acute and chronic nosocomial infections in immune-compromised patients (Bielecki et al., 2008; Malgaonkar and Nair, 2019).

Numerous P. aeruginosa infections are marked by typical skin manifestations such as burn infections and over-infection of old wounds (Spernovasilis et al., 2021). Multi-resistant P. aeruginosa has posed a major challenge to conventional antibiotics and therapeutic approaches, which show low efficacy and cause serious side effects (Shao et al., 2020). Only Cefiderocol was listed in 2020 as an antibiotic with consistent clinical and microbiological efficacy against specific strains of P. aeruginosa (Bassetti et al., 2021). New medicines are thus required. Interfering with quorum sensing has been shown to be very effective in reducing P. aeruginosa pathogenicity (Mancuso et al., 2021). One of the recommended options involves natural antimicrobial drugs, such as natural plant compounds (Moradi et al., 2020). Medicinal plants are excellent chemical antimicrobial active agents (Mulat et al., 2020; Rasamiravaka et al., 2017). Plants produce various antimicrobial compounds, such as phenolics, terpenoids, flavanones, and quinones, due to their similarity in chemical structure to QS signals (acyl-homoserine lactone, AHL) and their ability to damage signal receptors (Chaudhry et al., 2021; Teplitski et al., 2021). Previous research has identified some plants as anti-QS (Moradi et al., 2020). But many others need to be further investigated for their ability to interfere with bacterial QS.

Burkina Faso's flora abounds with various plants from which anti-QS compounds can be sought (Tibiri et al., 2020). Ageratum conyzoides was selected according to previous studies which showed anti-microbial activity on specific Gram-positive and Gram-negative bacterial species such as P. aeruginosa, Escherichia coli, Shigella dysentery, and Streptococcus aureus, without any effect on the growth of these pathogens (Akinyemi et al., 2005; Pintong et al., 2020). Earlier studies reported a minimum bactericidal concentration of 160 mg/mL. Chah et al. (2006) reported a 90% wound healing rate with methanolic extract (6%), with no inhibition on the growth of the involved bacteria. A. conyzoides found in Burkina Faso, has long been used in folk medicine for infectious and skin diseases treatment (Nacoulma, 1996). The leaves were mainly used as poultices on wounds, burns, gastrointestinal pains, and anthrax (Nacoulma, 1996; Nébié et al., 2004). It is possible that the plant acts through another process that reduces the production of bacterial virulence factors.

The aim of the study is to provide a scientific justification for the application of A. conyzoides in traditional medicine for the treatment of skin infections, old wounds and burns with over-infection. Specifically, the  objective  was  to  highlight  the anti-quorum potential and the impact of A. conyzoides on the control of virulence factors associated with P. aeruginosa pathogenicity.

Chemicals

p-Iodonitrotetrazolium, elastin congo red, acetic acid, hydrochloric acid, 3-(N-morpholino) propane sulfonic acid (MOPS), Folin-Cioralteu reagent, perchloric acid, glacial acetic acid, vanillin-glacial acetic, aluminium trichloride, carbenicillin (antibiotic) hexanoyl homoserine lactone molecule (HHL), gallic acid, vanillic acid, ellagic acid, caffeic acid, sinapic acid, rutin, quercetin and kaemferol O-nitrophenyl- β?D-galactopyranoside, nutrient agar, and Lauria-Bertani (LB) broth medium were obtained from Sigma-Aldrich (Germany). Solvents (n-hexan, chloroform, ethyl acetate, n-butanol) of analytical grade were provided from sigma Aldrich (Belgium). 

Bacterial strains, plasmids and growth conditions

P. aeruginosa PAO1 and Chromobacterium violaceum CV026 were provided from the Plant Biotechnology Laboratory (Université Libres de Bruxelles). Bacteria (10⁶ UFC/mL) were grown in LB broth (pH 7, 175 rpm) at 37°C for P. aeruginosa PAO1 and 30°C for C. violaceum CV026. P. aeruginosa derivatives harboring plasmid (pPCS1001, pPCS1002, pβ03, pLPR1, pβ02, pβ01 and pTB4124) were streaked onto LB-MOPS broth (50 mM, pH 7.2, 175 rpm) supplemented with carbenicillin (300 µg/mL).

Plant material extraction

A. conyzoides L. (Asteraceae) samples were collected in Gampela in August 2014 (Ouagadougou, Burkina Faso) and formerly identified. Voucher specimen has  been deposited at the national herbarium of Burkina Faso (HNBU) under code 8755. The whole plant material was washed and dried at room temperature. The dried material was reduced into powder and extracted by maceration for 24 h with methanol containing 20% water. The extract was concentrated in a vacuum evaporator and used for a liquid-liquid fractionation with n-hexane, chloroform, ethyl acetate and n-butanol successively. Collection and experimental research on this plant were in accordance with national guidelines in Burkina Faso.

MIC and MBC assay

The minimal inhibitory concentrations (MIC) of the extract were determined by broth microdilution method (Eloff, 1988). An overnight bacterial culture was diluted with LB broth to obtain a starting inoculum (106 CFU/mL). Each inoculum (180 µL) was incubated with a serial concentration of extracts ranging from 5 to 0.049 mg/mL. Bacteria growth was studied using p-iodonitrotetrazolium staining. After 18 h of incubation, 50 µL of INT (0.2 mg/mL) was added to each well and incubated at 37°C for 30 min. A red colour indicated bacterial growth. To assess minimum bactericidal concentration (MBC), 20 µL aliquots of all dilutions showing no bacterial growth were spread on LB agar plates (37°C, 24 h). The MIC were determined as the lowest concentration that inhibits bacterial growth, and the highest concentration that produces no bacterial colonies on a solid medium was chosen as the MBC (Ouedraogo and Kiendrebeogo, 2016).

Bacteria kinetic growth assay

Kinetics growth was assessed for 48 h for CVO26 or 18 h for PAO1 as previously described (Rasamiravaka et al., 2018). A 5 mL volume of bacterial suspension was aliquoted into six or eight sterile tubes and grown (37°C for PAO1, 30°C for CV026) under continuous stirring (175 rpm). At regular time intervals (6 h for CVO26 and 3 h for PAO1), the turbidity or optical density of the bacterial culture at 600 nm is measured from the bacterial suspension. Bacterial cultures were then centrifuged at 3000G at 24°C for 5 min. The supernatant was removed and the bacterial pellet re-suspended with 5 ml of a sterile NaCl solution (9‰). Briefly, a range of bacterial dilutions (10^-5 to 10^-8) was generated with sterile NaCl solution (9‰). 100 μl of each dilution was spread on LB agar plate and incubated for 18 h (Rasamiravaka et al., 2018). Bacterial growth was assessed by determination of colony forming units (CFU/mL) according to the formula:

CFU = Number of colonies / (Dilution × volume).

Violacein production in C. violaceum CV026 assay

Violacein production induced by hexanoyl-L-homoserine lactone (HHL) was evaluated during 48 h. Violacein was extracted from the supernatant of the culture using the previously described method of Saqr et al. (2021), with some modifications. CV026 inoculum (100 µL) was incubated for 24 and 48 h with 1.880 mL of LB broth supplemented with HHL and samples (20 µL). Salicylic acid was used as a positive control. 1 mL of culture from each tube was centrifuged at 7000 rpm for 10 min in order to precipitate the insoluble violacein. The supernatant was removed and 1 ml of DMSO was added to the pellet. The solution was vortexed for 30 s to solubilise the violacein and then centrifuged again at 7000 rpm for 10 min. 200 μL of the supernatant containing the violacein was introduced into 96-well microplates and the absorbance was measured at 585 nm. Violacein amount was assessed by using the absorbances ratio 585 nm/600 nm.

Pyocyanin production in P. aeruginosa PAO1

The pyocyanin kinetic production was evaluated during 18 h (King et al., 2016). Briefly, 150 μL of samples were inoculated with 750 μL of PAO1 inoculum in LB medium for 8 and 18 h of growth. At each time point, the bacterial culture was centrifuged at 3000 G, 24°C for 5 min. 500 μL of chloroform was added to 1 mL of bacterial supernatant and then vortexed to extract pyocyanin. To extract pyocyanin further, 400 µL of the lower chloroform phase containing pyocyanin was added to 300 µL of aqueous 0.2 N hydrochloric acid in Eppendorf tubes. The mixture was vortexed and then centrifuged to separate the two phases with pyocyanin in the upper aqueous phase. The absorbance at 380 nm of 200 μL of aqueous phase was measured to assess the pyocyanin produced by the bacteria. Salicylic acid was used as a positive control. Pyocyanin amount was assessed by using the absorbance ratio 380 nm/600 nm.

Elastase assay

Elastase production was detected using elastin Congo red (ECR) (Ahmed et al., 2019). The method consists of the degradation of elastin bound to Congo Red by a bacterial suspension containing elastase. P. aeruginosa PAO1 was grown with samples and control (DMSO 1% and salicylic acid) for 8 and 18 h at 37°C at 175 rpm in LB. The tubes were then centrifuged at 3000 g for 5 min at 24°C. 250 μL of congo red elastin (5 mg/ml) dissolved in Tris-HCl buffer (0.1 M   Tris-HCl   pH 8; 1 mM CaCl₂)   was    added   to   750 μL  of supernatant in each Eppendorf tube. The mixture was incubated at 37°C for 16 h at 200 rpm. The degradation of elastin causes solubilisation of Congo Red during this incubation phase, producing a red solution, whereas the undegraded Elastin-Red-Congo complex is insoluble in water. The reaction mixture was centrifuged (3000 g, 10 min) to separate the non-soluble part, and the absorbance of 200 μl of supernatant was read at 495 nm to estimate the elastin activity.

Rhamnolipids assay

Rhamnolipid production was assessed by methylene Blue method (Rasamiravaka et al., 2016). This method relies on measuring the absorbance of the rhamnolipid-methylene blue complex moving through the chloroform phase. P. aeruginosa PAO1 were grown with samples and control (DMSO 1% and salicylic acid) for 8 and 18 h in 5 mL of LB medium (37°C, 175 rpm). All tubes were centrifuged (3000 g for 10 min) and 4 ml of supernatant recovered and filtered (0.1 μm Millipore filters). The pH of the supernatant was then adjusted to 2.3 ± 0.2 with 1N hydrochloric acid. Rhamnolipids were extracted by stirring the supernatant with 4 ml ethyl acetate (centrifugation at 100 g for 1 min). The upper phase containing the rhamnolipids was collected in a 30 ml tube. The process was repeated three times and the extracts were collected and evaporated to dryness. The extracts of rhamnolipids were then dissolved in 4 mL of chloroform and mixed with 400 μl of freshly prepared methylene blue solution. The methylene blue solution was prepared by adding 200 μl of methylene blue reagent (1.4%, w/v in ethanol) and 4.8 mL of distilled water adjusted to pH 8.6 ± 0.2 with 50 mM borate buffer. Tubes were vortexed for 5 min and incubated at room temperature for 15 min to allow complexation of the methylene blue to the rhamnolipids. 1 ml of the chloroform blue phase was added to a 2 ml Eppendorf tube and vortexed for 20 s with 500 μl of 0.2 N hydrochloric acid, allowing the complexed methylene blue to be transferred to the acid phase. Tubes are centrifuged at 100 g for 1 min and left at room temperature for 10 min. 200 μL of the acidic upper phase is transferred to a 96-well microplate well and the absorbance is measured at 638 nm.

β?Galactosidase assay

All reporter strains of PAO1 were incubated in LB-MOPS-Carbenicillin for 8 and 18 h (50 µL, 37°C, 175 rpm) supplemented with samples (Extract and DMSO). In 12-well plates, 940 μL of LB-MOPS-Carbenicillin, 10 μL of samples (dissolved in DMSO), and 50 μL of PAO1 bacterial inoculum were incubated at 37°C. After incubation, the absorbances were read at 600 nm. 300 μL of permeabilization buffer was added to the initial mixture, then 75 μL+50 µL of buffer was incubated for 20 min at 37°C. The samples were then used to perform the β?galactosidase assay with O-nitrophenyl- β?D-galactopyranoside. 600 μL of ortho-nitrophenyl-β-D-galactoside (oNPG) substrate buffer was added and the tubes were allowed to incubate for 5 to 60 min at 37°C. The reaction was stopped with 700 μL of stop buffer (1 M Na₂CO₃). The absorbance of 100 μL of the supernatant was read at 420 nm. β?Galactosidase activity has been expressed in Miller units (Bonneau et al., 2020).

Phytochemical analysis

Determination of total polyphenolic, total flavonoid, and total triterpenoid content

Total polyphenolic, flavonoids, and terpenoids contents in extract were determined  using  previous  spectrometric  method  (Olech  et al., 2020). The polyphenol content was assessed with 25 μL of extract solution (0.1 mg/ml) and 125 μL of FCR solution (0.2N). After 5 min of incubation, 100 μL of sodium carbonate solution (75 g/L) was added. The absorbance was read after 1 h of incubation, at 760 nm against a standard curve of gallic acid (200 mg/L, R²=0.9952). Total polyphenolics were expressed as mg/g gallic acid equivalents (GAE). Flavonoid content analysis was performed with 100 μL of AlCl₃ (2% in methanol) and 100 µL of extract (1 mg/ml in methanol). Absorbance was read at 415 nm after 10 min from a quercetin standard curve (R²=0.9995). Total flavonoid was expressed as mg/g quercetin equivalent (QE). Triterpenoid content was assessed by the following process: 100 µL of sample (10 mg/mL in methanol), 150 µL of glacial acetic acid vanillin solution (5%) and 500 µL of perchloric acid were used as the reaction solution. The mixture was placed into a water bath (60°C) for 45 min, then cooled in an ice bath and 2.5 mL of glacial acetic acid was added. Absorbance was read at 548 nm against standard curve of ursolic acid (R²=0.9966). Total triterpenoids were expressed as mg/g ursolic acid equivalent (UAE).

HPLC-DAD analyses

The chromatography analysis was conducted according to the protocol described by Meda et al. (2011). HPLC-DAD-UV-Visible chromatographic analysis of the phenolic compounds of the active fraction was carried out under isocratic conditions by using a C18 reverse phase column (4.6 mm × 250 mm) packed with 5 µm diameter particles. The mobile phase consisted of methanol, acetonitrile, water (40:15:45, v/v/v) containing 1.0% acetic acid. The mobile phase was filtered through a 0.45 µm membrane filter and degassed by ultrasound before use. Eight standards including five (05) phenolic acids (gallic, vanillic, ellagic, caffeic and sinapic acids) and three flavonoids (rutin, quercetin and kaempferol) were used. Each standard was dissolved in methanol to an initial concentration of 400 µg/mL and then cascade diluted to 6.25 µg/mL in 1 mL vials. The fraction was also dissolved in methanol. The injection rate was 0.5 mL/min and the injection volume was 10 µL. Chromatographic data were recorded over 210 to 400 nm, and integrated at 271 nm, 327 nm for phenolic acid and 365 nm for flavonoids. The used solvent was HPLC grades (Sigma-Aldrich, Belgium). Quantification of the compounds was done by peak area integration against the standard curves.

Statistical analysis

Experiment was performed in triplicate and data were expressed as mean ± SD. GraphPad prism Software was used for statistical analysis (GraphPad software Inc., San Diego, CA, USA); one-way or two-way ANOVA followed by the Tukey or Bonferonni test on GraphPad at the value ≤ 0.001 was considered significant.

Anti-QS effect

Effect on bacterial growth

The MICs for hydro-methanolic extract were 5.0 mg/mL for P. aeruginosa PAO1 and 2.5 mg/mL for C. violaceum CV026. For two bacteria the MBC values were 5 mg/mL (> 5 mg/mL for PAO1). MBC/MIC ratio (<4) for both strains  indicated  a  bactericidal   effect   of   extract.  For concentrations below the MIC, A conyzoides extract should not induce the bacteriostatic or bactericidal activity. So, at a minimum concentration of 100 µg/mL, corresponding to CMB/CMI >32 ratio, the strains should be tolerant to the plant extract which will allow evaluation of its intrinsic effect on QS-dependent bacterial factors.

Effect of extract on bacterial kinetics growth

Data of Figure 1a showed the C. violaceum kinetic in the presence of hydro methanolic extract in relation to DMSO 1%  for   48 h.   An   exponential   growth  phase  (0-24 h) followed by a stationary phase (24-48 h) was observed. The exponential phase started immediately at the beginning of the incubation. In both phases, the same growth pattern of C. violaceum CV026 was observed. The results showed that the samples (methanolic extract (MeOH) and DMSO 1% exhibited substantially similar growth kinetics. Figures 1b shows the P. aeruginosa PAO1 kinetic data in the presence of methanolic extract in relation to DMSO 1% during 18 h of growth. As shown in CV026 growth, as soon as the samples were added, the bacterial strains started immediately growing in two phases. After bacteria growth reached the exponential phase at 12 h, a stationary phase was followed until 18 h.

In both phases, the same growth pattern of PAO1 was observed. The results showed that the samples (methanol 80% extract and DMSO1%) exhibited substantially similar growth kinetics effects. Methanolic extract (100 µg/mL) did not significantly affect the cell viability according to CFU quantification.

Effect on violacein production in C. violaceum CVO26

The methanol extract and its fractions were  screened  for their effect on HHL-induced violacein production by CV026. As shown in Figure 2a, after treatment of sub-MIC concentration, the extract hexane (HF) and chloroform (CF) fractions induced a significant inhibition of production of violacein production (OD585 nm/600 nm) during 24 and 48 h. Compared to negative control, the violacein production inhibition >50% was found with the hexane and chloroform fractions. For the ethyl acetate (EAF) and butanol (BF) fractions, the inhibition was less noticeable (30%). CF was a best inhibiter of violacein production than methanolic extract and salicylic acid.

Importantly, the presence of the samples in the medium had no adverse effect on bacterial growth (Figure 1a). It is therefore conceivable that the negative effect on violacein production is not driven by growth inhibition but rather by disruption of the resistant quorum-sensing systems. On this basis, the impact of the extract and its fractions were investigated on the production of QS-dependent virulence factors, such as pyocyanin, elastase, rhamnolipid and selected genes involved in P. aeruginosa QS system.

Effect on pyocyanin production

Figure 2b shows the pyocyanin amount produced par P. aeruginosa PAO1 in the presence of methanolic extract (MeOH) and fractions in relation to controls. As shown in Figure 2b, at 8 and 18 h, MeOH extract, hexane fraction (HF) and chloroform fraction (CF) have  a significant impact (p < 0.001) on pyocyanin production (OD380 nm/600 nm). This impact was more evident for the HF and CF fractions compared to salicylic acid used as positive control. These two fractions may contain similar or different compounds that express the same effects on pyocyanin production. The effect of butanol fraction (BF), was weaker than methanolic extract (MeOH extract), while no significant effect was found with the ethyl acetate fraction (EAF).

Effect on elastase and rhamnolipids production

Rhamnolipids and lasB elastase are also important virulence factors in the virulence of P. aeruginosa. The results of this elastase and rhamnolipids assay were summarized in Figure 3. At 8 and 18 h, CF induced the lowest amount of elastase (Figure 3a) and rhamnolipids production (Figure 3b). The effect of CF fraction was more significant from 18 h onward, compared to salicylic acid. For the EAF and BF fractions, their effects were either similar to or less than that of MeOH extract. The effect of HF was similar to that of the stock extract and less than the CF fraction. It was suggested that a molecular difference in the composition of HF and CF. CF appears to contain compounds that are more effective than HF, and which effectively repress the production of these virulence factors. The CF fraction has therefore been studied for its ability to modulate the expression of virulence genes.

Effect on QS genes in P. aeruginosa PAO1

If CF interfered with QS mechanisms, it should be reflected on the transcription of QS-regulated and QS-regulatory genes. In order to highlight any interference with the QS genes expression in P. aeruginosa PAO1, it was followed that the transcription rate over 8 and 18 h of growth (Table 1). After 8 h of growth, the CF caused a significant decrease in the expression of the rhlR gene lasI, lasR, lasB, rhlR compared to the negative control. This effect was more pronounced after 18 h (except for LasI). Thereafter, the expression of the rhamnolipid synthesis genes, rhlA as well as the elastase synthesis gene lasB was significantly reduced at 18 h, whereas at 8 h no effect was evident. In contrast, salicylic acid reduced the expression of all QS-dependent genes after 8 h, but had a limited impact after 18 h. Indeed, only the lasR and rhlR genes were affected while the other had an equivalent expression to the DMSO condition. Compared to the extract and salicylic acid, the impact of CF was more noticeable on the rhlR gene both at 8 and 18 h (P<0.001) whereas on the lasR regulatory gene, no significant difference was observed after 18 h. In order to prove whether the decrease in β-galactosidase activity is really associated with a reduction in the expression of QS-related genes and not with a general effect on transcription/translation mechanisms, the promoter activity of the aceA gene was examined in P. aeruginosa PAO1. The addition of samples has no negative impact on aceA gene transcription, reflecting that they affect the expression of QS-related genes without affecting the transcriptional machinery of P. aeruginosa PAO1 (Table 1).

Phytochemical analyses

Polyphenolic, flavonoid and terpenoid contents

Total polyphenol, flavonoid and terpenoid contents were analysed in the active fraction (CF) and the stock extract (MeOH extract). The amount of polyphenol, flavonoid and terpenoid compounds in the MeOH extract was 123.33 ± 5.6 mg/g GAE, 54.67 ± 1.54 mg/g QE and 15.60 ± 0.6 mg/g UAE, respectively. For CF fraction, the values were 683.14 ± 60.39 mg/g GAE, 115.21 ± 2.52 mg/g QE and 37.74 ± 0.29 mg/g UAE, in polyphenol, flavonoids and terpenoids, respectively. Mostly, high content of total polyphenol, flavonoids and terpenoids was found mainly in the active fraction (CF).

HPLC-DAD analysis

The identification of compounds from the CF fraction was performed by injection on HPLC-DAD. The retention times of the different peaks were compared to the retention times of the different standards and identified (Figure 4). The summary information is listed in Table 2. Four (04) of the peaks were recognized, gallic acid (R_t=5.507 min), vanillic acid (R_t=7.364 min) and ellagic acid (R_t=8.279 min), and sinapic acid (R_t=8.209 min). Other compounds such as caffeic acid, quercetin, rutin and kaempferol, were also identified but their presence was  revealed  in  trace form (except for quercetin) (Table 2), while the rest of the peaks were not identified. The concentration of the identified compounds indicates a higher proportion of ellagic and sinapic acids (Table 2).

A. conyzoides is an herbal medicine commonly used in traditional medicine to  treat old  wounds,  infected  burns and dermatoses (Chahal et al., 2021; Igbinosa and Eribo, 2016). A great deal of A. conyzoides-oriented research has been carried out to discover compounds to control multidrug resistant pathogens (Kotta et al., 2020). Antibacterial activities of some extracts have been reported on several pathogenic strains such as S. aureus, E. coli, and P. aeruginosa (Igbinosa and Eribo, 2016; Odeleye et al., 2014). Beyond the bactericidal and bacteriostatic  effect,  bioactive  compounds  of  the  plant disruption of bacterial virulence. The study indicates that A. conyzoides exerts anti-virulence activities rather than a bactericidal effect against the Gram-negative opportunistic pathogen P. aeruginosa. The report is the first in vitro investigation of the anti-QS properties of A. conyzoides from Burkina Faso.

C. violaceum CV026 has a low human health impact, but is widely used as a reporter strain in QS screening (Zhu et al., 2011). As a result, chloroform fraction (CF) obtained from methanol (MeOH) extract reduces QS-mediated violacein production in C. violaceum, as well as pyocyanin, elastase, rhamnolipid production, and biofilm formation in P. aeruginosa PAO1 with no effect on bacterial growth. The effect of CF fraction was more significant from 18 h onwards, compared to salicylic acid, an inhibitor of quorum sensing (Ahmed et al., 2019; Prithiviraj et al., 2005). At a concentration (100 µg/mL) below the MIC (5 mg/mL), no bacteriostatic or bactericidal effects were detected. This observation supports the findings of Chah et al. (2006) who reported the lack of inhibition on PAO1 growth by the methanolic extract, and those of Odeleye et al. (2014) whose results indicted a  sensitivity  of 160 mg/mL well above the MIC value recorded in this study. Virulence factors and the biofilm formation examined in this study are under QS control (Paluch et al., 2020); hence reduction/inhibition of these factors indicates possible QS antagonism, which may result either from a direct effect on gene expression and/or protein synthesis downstream of AHLs, or from an inhibition of AHL synthesis (Dekimpe and Déziel, 2009). In QS, the rhl system has been shown to directly control pyocyanin biosynthesis through the transcription factor RhlR and its autoinducer C4-HSL (Dekimpe and Déziel, 2009; Haque et al., 2018). C4-HSL forms a complex with RhlR to activate the transcription of lasB elastase genes and those involved in initiating pyocyanin (by phz operon) and rhamnolipids (rhlAB operon) biosynthesis (Cruz et al., 2020; Dekimpe and Déziel, 2009; Hendiani et al., 2019). According to the results, the CF significantly down-regulated the RhlR gene and caused the reduction of pyocyanin, rhamnolipids and lasB-elastase biosynthesis. This is evidence to support that CF may possess RhlR inhibitory can act to control disease-causing bacteria through activity against P. aeruginosa virulence.  

Although the mechanism of action of CF is unclear, it is evident here that it acts upstream of the virulence genes (Hendiani et al., 2019).

Herein, the liquid-liquid fractionation of the A. conyzoides MeoH extract led to obtain a high content of polyphenol, flavonoids and terpenoids within the active fraction (CF). Recent works show evidence that several phenolic compounds, terpenoids and flavonoids have anti virulence effects (Asfour, 2018; Bouyahya et al., 2017; Santos et al., 2021). The HPLC results revealed high concentration of ellagic, sinapic, and gallic acids and low concentration of vanillic acid and quercetin. These compounds have been reported previously in A. conyzoides (Dang Xuam et al., 2004; Okunade, 2002). Most of them have been previously pointed out as QSinhibitors (Francesca et al., 2020; Manner and Fallarero, 2018; Onem et al., 2021; Quecan et al., 2019). Previous investigation has shown ellagic acid derivatives from Terminalia chebula Retz down-regulate the expression of QS genes, thereby attenuating the virulence of P. aeruginosa PAO1 (Sarabhai et al., 2013). A combination of ellagic acid and tetracycline was determined to effectively inhibit biofilm    formation   by    Propionibacterim   acnes without affecting its growth (Sankar et al., 2016). Vanillic acid was identified as an active leader responsible for anti-quorum sensing activity in Actinidia deliciosa extract (Sethupathy et al., 2017). It also inhibited the violacein production in C. violaceum ATCC 12472 (Sivasankar et al., 2020). Ouyang et al. (2016) had shown that quercetin as an effective inhibitor of QS, and virulence factors including pyocyanin, protease and elastase in P. aeruginosa (Ouyang et al., 2016). Furthermore, the expression levels of lasI, lasR, rhlI and rhlR were significantly reduced, in response to 16 µg/mL quercetin (Ouyang et al., 2016). Widely distributed in nature, gallic acid is a phenolic compound extensively studied for its numerous biological activities including anti-QS (Borges et al., 2014). Gallic acid had also been reported to  be  an autoinctor-1 type QS inhibitor (Zhang et al., 2020). The reduced expression of virulence genes in the CF fraction could be attributed to the presence of these compounds (Nain et al., 2020), or/and to other unidentified phytochemicals.

Altogether, these non-bactericidal anti-virulence properties and the reported antimicrobial activities, provide additional evidence and support the long history use of this plant in traditional medicine for the treatment of skin infectious disease (Nacoulma, 1996). Indeed, the reduction of QS genes and the end-effect on virulence factors’ productions allows to explain the historical use of A. conyzoides in Burkina Faso. These observations also provide an opportunity to extrapolate on how this plant could  be  used  to  control   antibiotic   resistant  bacteria.

Research is currently in progress to identify and isolate the bioactive compound.

The authors have not declared any conflict of interests.

This study was carried out with funding from the "Centre National de l'Information, de l'Orientation Scolaire et Professionnelle, et des Bourses" (CIOPB) of Burkina Faso. This research was supported by the supply of bacterial strains provided by the Plant Biotechnology Laboratory (Université Libre de Bruxelles ULB).