Biofi lm inhibition activity of traditional medicinal plants from Northwestern Argentina against native pathogen and environmental microorganisms

Introduction: Plants have been commonly used in popular medicine of most cultures for the treatment of disease. The in vitro antimicrobial activity of certain Argentine plants used in traditional medicine has been reported. The aim of this study was to investigate the antimicrobial, anti-biofi lm, and anti-cell adherence activities of native plants (Larrea divaricata, Tagetes minuta, Tessaria absinthioides, Lycium chilense, and Schinus fasciculatus) collected in northwestern Argentina. Methods: The activities of the fi ve plant species were evaluated in Bacillus strains and clinical strains of coagulase-negative Staphylococcus isolated from northwestern Argentina and identifi ed by 16S rDNA. Result: Lycium chilense and Schinus fasciculatus were the most effective antimicrobial plant extracts (15.62μg/ml and 62.50μg/ml for Staphylococcus sp. Mcr1 and Bacillus sp. Mcn4, respectively). The highest (66%) anti-biofi lm activity against Bacillus sp. Mcn4 was observed with T. absinthioides and L. divaricate extracts. The highest (68%) anti-biofi lm activity against Staphylococcus sp. Mcr1 was observed with L. chilense extract. T. minuta, T. absinthioides, and L. divaricata showed percentages of anti-biofi lm activity of between 55% and 62%. The anti-adherence effects of T. minuta and L. chilense observed in Bacillus sp. Mcn4 refl ected a difference of only 22% and 10%, respectively, between anti-adherence and biofi lm inhibition. Thus, the inhibition of biofi lm could be related to cell adherence. In Staphylococcus sp. Mcr1, all plant extracts produced low anti-adherence percentages. Conclusion: These fi ve species may represent a source of alternative drugs derived from plant extracts, based on ethnobotanical knowledge from northwest Argentina.


INTRODUCTION
Resistance to antibiotics has become a major public health problem worldwide as it reduces the effectiveness of treatments and increases morbidity, mortality, and health-care costs (1) .This challenge is a natural consequence of the adaption of infectious pathogens to the types of antimicrobials used in multiple areas, including medicines, plants and animals raised as food, and disinfectants used in farms, hospitals, and households (2) (3) .
Research on plants as potential sources of new and effective antimicrobials with novel modes of action is well established.Although active constituents may occur in plants in low concentrations, plant extracts may, in many cases, represent a superior source of antimicrobial compounds than synthetic drugs (4) .
The use of plants for the treatment of multiple ailments has been a common practice in the popular medicine of most cultures, even where the precise causation of a disease and/or the mechanisms of its cure have not always been understood (5) (6) .Exploratory research on plants as sources of new antimicrobial compounds has led to a joint interest in fundamental research in the fi elds of agriculture, chemistry, and medicine.However, studies on the antimicrobial activity of plant extracts have been largely restricted to the analysis of their bacteriostatic and bactericidal properties (7) .
There are some existing reports on the in vitro antimicrobial activity of ethanolic extracts of some Argentine plants used in traditional medicine (8) (9) , although few studies evaluating antimicrobial activity as well as anti-biofilm activity in relation to pathogenic microorganisms have been reported (10) (11) .Biofi lm is the predominant mode of growth for bacteria in most natural, industrial, and clinical environments.Biofi lms typically consist of densely packed, multi-species populations of cells, encased in a self-synthesized polymeric matrix and attached to a tissue or surface (12) .This mode of growth protects bacteria from environmental stresses, that is, the treatment of biofi lms with antibiotics or other biocides is often ineffective for eradication (13) .Biofilm formation is therefore a major challenge in many contexts, ranging from industrial corrosion and biofouling (14) to chronic and nosocomial infections (15) .
The purpose of the present work was therefore to investigate and demonstrate the presence of antimicrobial, anti-biofi lm, and anti-adherence activities in native plants collected in northwestern Argentina (specifi cally, Larrea divaricata, Tagetes minuta, Tessaria absinthioides, Lycium chilense, and Schinus fasciculatus), which have been traditionally used for a variety of folk medicinal purposes.

Microorganisms and fermentation conditions
Bacillus strains were isolated from a variety of sources in northwestern Argentina, such as soils contaminated with hydrocarbons and natural soils from a riverbank.Soil suspensions were created and heated at 80°C for 15 min, plated onto Luria-Bertani (LB) agar, and incubated at 37°C.Clinical isolates of coagulase-negative Staphylococcus (CoNS) were also obtained from a dialysis center in the City of San Miguel de Tucumán.All isolated bacteria were characterized using conventional biochemical methods, namely catalase, bacitracin susceptibility, and coagulase testing.Liquid cultures of Bacillus and CoNS were prepared in 125-ml fl asks containing 10ml of LB medium on an orbital shaker at 37°C.

Molecular characterization
Total deoxyribonucleic acid (DNA) was extracted from cells harvested in the mid-exponential growth phase as described previously (16) .Polymerase chain reaction (PCR) amplifi cation was performed in a 25-µl reaction mix containing 2.5µl 10X STR reaction buffer (Promega, Wisconsin, United States), 20ng total DNA, 0.5µM of each primer, and 1U Taq DNA polymerase (Promega).Primers of the sequence 27F (5´-AGAGTTTGATCMTGGCTCAG-3´) and 1492R (5´-GGTTACCTTGTTACGACTT-3´) were used to generate partial sequences of 16S ribosomal DNA (rDNA).Amplifi cation conditions were as follows: 5 min at 94°C; 35 cycles of 1 min at 94°C, 2 min at 50°C, and 2 min at 72°C; and 7 min at 72°C for the fi nal extension.PCR products were analyzed by electrophoresis on 2% (wt/vol) agarose gels.
DNA sequencing was performed by Macrogen Services(Seoul, South Korea).Sequences were compared and aligned with the GenBank database using Basic Local Alignment Search Tool (BLAST) software.Partial nucleotide sequences of the 16S rDNA gene of Bacillus sp. and CoNS Staphylococcus sp. were deposited in the GenBank database under accession numbers KP872905 and KP872906, respectively.

Plant material and extraction
In the present study, samples of fi ve native plants from four families were collected in order to evaluate their antibacterial and/or anti-biofi lm activities (Table 1) (17)(18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31) .The leaves of these plants were collected in the locality of Tafí del Valle in the province of Tucumán in northwestern Argentina (Figure 1A).We focused our study on leaves because they represent a sustainable resource and, although bark has been traditionally used, leaf extracts tend to show higher activity.The identity of the plant material was confi rmed and voucher specimens were entered into the Miguel Lillo Herbarium at the National University of Tucuman.Botanical and vernacular names, traditional uses, and reported chemical compositions are shown in Table 1.The collected leaves were dried at room temperature in a well-ventilated room and ground to fi ne powder using a mortar and pestle.This powder was then macerated in 80% ethanol for 48h and then fi ltered through Whatman n° 1 fi lter paper.
The solid residue was suspended in 50% ethanol until dissolution.The extracts were then concentrated to dryness using a rotary evaporator (Figmay, Córdoba Argentina) under reduced pressure and at a temperature not exceeding 50°C.For antimicrobial and anti-biofi lm studies, several concentrations of each prepared powder were evaluated (8-500μg/ml).

Determination of sub-inhibitory concentration and minimum inhibitory concentration
The microorganisms used in this study were cultured in Mueller-Hinton (MH) broth and incubated at 37°C for 24h prior to assays.All cultures were standardized by standard dilution in MH broth (4°C) and characterized by experimental transmittance [T=30, optical density (OD) 610 ].Growth at 4°C was stopped by immersion in ice-cold water.From this standard dilution, a 50 ml working suspension was prepared with corrected transmittance T=95 (25µl of T=30 in 5ml of refrigerated MH broth), corresponding to 1.5 × 10 8 colonyforming units (CFU)/ml for the Bacillus sp.strain and 1.2 × 10 8 CFU/ml for the Staphylococcus sp.strain.This dilution was maintained at 4°C to ensure cellular quiescence.Inoculum (100µl) was added to each well containing sample plant extracts.
The effect of plant extracts on the growth of Bacillus and CoNS Staphylococcus isolates was evaluated using a standard broth microdilution methodology.Serial dilution was employed to determine the minimum inhibitory concentration (MIC) [the lowest plant metabolite concentration (µg/ml) that inhibited bacterial growth] and sub-inhibitory concentration (SIC) [the highest extract concentration (µg/ml) below the MIC that did not inhibit bacterial growth] concentrations in the range 8-500µg/ml after 24h of growth.
The plant extracts at each concentration were prepared in dimethyl sulfoxide (DMSO) solvent, aliquoted into multi-well polystyrene plates, and dried to allow DMSO evaporation (to avoid the toxic effects of DMSO on bacteria).Negative controls (cells + MH broth), positive controls [cells + MH broth + antibiotics (chloramphenicol and gentamicin)], and media controls (MH) were included.Positive controls for antibiotics were prepared at 8-500µg/ml using serial dilution.All tests were performed in quadruplicate.

Biofi lm formation and anti-biofi lm effects of plant extracts
Biofilm production was carried out with the method in polystyrene multi-well cell culture plates.The biofi lmproducing CoNS Staphylococcus strains 263 and 1174 (isolated from bovine mastitis and provided by the Faculty of Veterinary Medicine, Buenos Aires University) and a non-producing CoNS Staphylococcus strain (ATCC 12228) were used as controls (1) .
The inoculum of each microorganism was standardized through standard dilution with LB (4°C) and characterized by experimental transmittance (T=30, OD 610 ).From this standard dilution, a 50ml working suspension was prepared with corrected transmittance T=95 (25µl of T=30 in 5ml of LB broth at 4°C) corresponding to 9 × 10 8 CFU/ml for Bacillus sp.strain and 1.6 × 10 8 CFU/ml for Staphylococcus sp.strain.To detect biofi lm-producing bacteria, 100µl of inoculum from each bacterial suspension and medium control (LB broth) were deposited in triplicate on multi-well polystyrene plates, and the plates were incubated at 37°C for 24h.After incubation, medium was extracted by inversion and wells were washed twice with phosphate-buffered saline (PBS) (8.1mM Na 2 HPO 4 , 1.5mM KH 2 PO 4 , 140mM NaCl, and 3mM KCl; pH 7.2).The contents of wells were emptied by inversion and plates were dried upside-down on absorbent paper.Once dry, biofi lms appeared as whitish, opaque membranes and were fi xed with methanol for 7 min (32) .Plates were emptied again by inversion and air-dried.Once dry, an aqueous solution of 0.1% safranin was used as a stain.After 4 min, plates were again inverted, washed with water, dried upside-down (1) .
Biofi lm formation was evaluated using a microplate reader at 490nm (Thermo Scientifi c TM Multiskan TM GO Ultraviolet-Visible microplate spectrophotometer,United States) .The strains considered as the best producers of biofi lm were those with OD 490 values greater than those of the control strains (OD 490 ˃ 0.08) (1) .
Next, the selected bacterial strains were analyzed in the absence and presence of plant extracts.To determine the effects of plant extracts on biofi lm formation, SIC of plant extracts (10-100µg/ml as appropriate for each microorganism) were added to the test wells and allowed to dry prior to inoculation.Inoculum (corresponding to 9 × 10 8 CFU/ml for Bacillus sp.strain and 1.6 × 10 8 CFU/ml for Staphylococcus sp.strain) was added in the form of a bacterial suspension at the zero hour time point of biofi lm formation, and biofi lms were allowed to develop for 24h as described above.Each bacterial strain inoculated in the absence of a plant extract was considered as a positive control for biofi lm formation.

Bacterial adhesion assay
Inoculum was prepared as a 1:10 dilution of microorganism cultures (24h) using fresh LB broth followed by incubation at 37°C for 2h with shaking at 100rpm (inoculum had OD 610 =0.100, corresponding to 1.5 × 10 8 CFU/ml).Inoculum of 200µl was then added to multi-well polystyrene plates at 37°C for 3h in the presence and absence of plant extracts at SIC. Next, the medium wasextracted by inversion, plates were washed twice with PBS, and bacterial adhesions were fi xed with 25% formaldehyde for 30 min.Wells were then dried and stained with 0.1% safranin.Cell adherence measurement was carried out in a microplate reader at 490nm.The percentage of inhibition of bacterial adhesion was determined as described above.

Statistical analysis
Statistical analysis was performed using SigmaPlot 12.0 and Minitab (Minitab, Inc.) software version 14 for Windows.Normality distribution was evaluated by Shapiro-Wilk test and analysis of variance (ANOVA) was used to evaluate the mean differences among treatments.Two-way ANOVA and subsequent comparisons were performed using the Holm-Sidak test and Bonferroni post-test as appropriate.Results were presented as the mean Standard Deviation (SD), and differences were accepted as signifi cant for p < 0.05 (5% of the signifi cance level).

Isolation and selection of indigenous spore-forming bacteria and coagulase-negative Staphylococcus strains
A total of 53 spore-forming bacteria were isolated from natural environments, and 22 CoNS strains were isolated from a dialysis center (1) .From this total of 75 native isolates, 24 were selected as biofi lm producers and the Mcn4 and Mcr1 strains were identifi ed as the best biofi lm producers compared with control strains (Figure 1B).Partial sequencing of the 16S rDNA gene confi rmed the biochemical and morphological characterization of both strains.A BLAST search and alignment analysis showed a similarity of 99% for strain Mcn4 with Bacillus licheniformis, hereinafter referred to as Bacillus sp.showed SIC values between 15.62-62.5µg/mlagainst this microorganism.

Determination of sub-inhibitory concentration and minimum inhibitory concentration
The L. divaricata extract showed remarkable activity against the genus Staphylococcus, with an SIC value equal to that of gentamicin (15.62µg/ml), which was used as the reference antibiotic.Furthermore, T. absinthioides extract showed similar antimicrobial activity against Staphylococcus sp.Mcr1 to that of chloramphenicol (Table 2).

Effect of plant extracts on biofi lm formation
The anti-biofi lm activity of the plant extracts on Bacillus sp.Mcn4 and Staphylococcus sp.Mcr1 was also tested at selected SIC values.
The best anti-biofi lm activity (66%) against Bacillus sp.Mcn4 was observed with the extract from T. absinthioides at 100µg/ml (Figure 2A) and with L. divaricata extract across the entire range evaluated (10-100µg/ml), with no signifi cant differences observed for the latter extract across the extract concentrations evaluated (p = 0.67) (Figure 2A).
Anti-biofi lm activity against Staphylococcus sp.Mcr1 was also evaluated at selected SIC values (Figure 2B).The highest anti-biofi lm activity was observed with L. chilense extract (68%).In this case, anti-biofi lm activity was dose dependent, with the highest inhibition observed at 50µg/ml (p = 0.02).Tagetes minuta, T. absinthioides, and L. divaricata showed percentages of anti-biofi lm activity with values between 55% and 62% (Figure 2B).For these extracts, no signifi cant difference was observed between the two concentrations evaluated (10µg/ml and 50µg/ml: p = 0.17, p = 0.05, and p = 0.59 for T. minuta, T. absinthioides, and L. divaricata, respectively).As also observed against the genus Bacillus, S. fasciculatus extract showed only a low level of biofi lm inhibition activity (Figure 2B).

Effect of plant extracts on cell adherence
The anti-adherence effects of the plant extracts were also evaluated.Samples selected for testing included the plant extract concentrations that had previously shown the most effective inhibition of biofi lm (Figure 2).
The anti-adherence effects of T. absinthioides and L. divaricata extracts against Bacillus sp.Mcn4 were found to be lower than their anti-biofi lm activity, indicating that their ability to inhibit biofi lm formation took place during a step subsequent to cell adherence (Figure 3A).The percentage of anti-adherence observed with T. minuta and L. chilense extracts refl ected a difference only of 22% and 10%, respectively, between antiadherence and biofi lm inhibition (Figure 3A).
Schinus fasciculatus showed the same percentage of antiadherence effect as its percentage of anti-biofi lm activity, with no signifi cant difference observed between these (p = 0.55).Thus, for this plant extract, the inhibition of biofi lm formation could be related to cell adherence.
In contrast, all plant extracts tested showed low antiadherence percentages against Staphylococcus sp.Mcr1, suggesting that their biofi lm inhibition abilities may be related to a later step in biofi lm formation.In all of these cases, the anti-biofi lm effect was more than 50% higher than the antiadherence effect (Figure 3B).

DISCUSSION
The more potent antimicrobial activities observed for plant extracts were against the Staphylococcus strain.The antimicrobial activities produced by extracts of the genera Schinus and Lycium against Staphylococcus sp.Mcr1 were comparable to those observed for gentamicin.This antimicrobial effect of the genus Schinus against Staphylococcus was also observed by Yao et al. (19) and Gehrke et al. (33) , with both of these sources reporting an MIC value of 125µg/ml.However, in the present work, a more effective MIC of 31.25µg/ml(Table 2) was observed.The similarities in the activities observed for S. fasciculatus and L. chilense could be due to the presence of fl avonoids in both extracts.The antibacterial properties of these compounds are increasingly being reported, and raw extracts from plants with a history of use in traditional medicine have now been screened for antimicrobial activities by numerous research groups (34) (35) .However, it is also possible that the high antimicrobial activity shown by S. fasciculatus was due to disruption of the bacterial membrane caused by the presence of terpenoids, which inhibit Adenosine triphosphate (ATP) synthesis (36) (37) .
However, ethanolic extract of L. divaricata showed potent antimicrobial activity against Staphylococcus sp.Mcr1.Davicino et al. (18) have reported that the fraction obtained from an ethanolic extract of L. divaricata contained several phenolic compounds, which may be responsible for the cytotoxicity observed against the Staphylococcus genus.
The antimicrobial activity observed with T. minuta and T. absinthioides may be due to the presence, not only of fl avonoids, but also of essential oils of the family Asteraceae (to which these species belong; Table 1), which have been reported as compounds with antimicrobial activity (38) (39) .
The SIC values determined for each plant extract were used to evaluate biofi lm inhibition.
Biofi lm eradication is diffi cult to achieve because of host defenses and inherent resistance to antibiotics and biocides.Furthermore, there are several mechanisms used to explain the resistance of biofi lms to antimicrobials (40) , which makes it diffi cult to predict the behavior of biofi lm cells.In the present study, the anti-biofi lm activity of plant extracts at SIC was evaluated after 24h and the percentage of biofi lm inhibition were observed (Figure 2).By comparing the values obtained, it can be seen that plant extract concentrations signifi cantly affected the removal and/or inactivation of biofi lms.
Extracts from these medicinal plants may infl uence biofi lm formation by damaging microbial membrane structures (4) , inhibiting peptidoglycan synthesis, and/or modulating quorum sensing (41) .With regard to the last of these mechanisms, several approaches involving quorum interference have been studied and they represent the most recent strategies to counteract staphylococcal infections, among others.Quorum interference can be achieved by several alternative strategies, including quorum sensing disruption through inhibition of signal molecule biosynthesis, signal molecule inactivation, and blockage of signal transduction (41) (42) .In fact, fl avonoids are among the compounds that exert anti-biofi lm effects via quorum sensing inhibition (43) .In the present study, all plant extracts that showed biofi lm inhibition contain fl avonoids, and as such, these compounds may be responsible for the biofi lm inhibition observed, although further studies are required to investigate this.
Biofi lm formation occurs stepwise, and one step of relevant importance is cell adhesion that occurs prior to biofilm formation (44) .In view of the importance of this step in biofi lm production, the anti-adherence effects of the plant extracts under study were also evaluated.
For S. fasciculatus, the inhibition of biofi lm formation observed may be related to cell adherence.Considering that adherence is a prerequisite for colony formation, preventing bacterial adhesion would appear to be an ideal strategy for reducing biofi lm formation (44) .
For the plant extracts evaluated in this study, anti-biofi lm and anti-adherence properties against the genus Bacillus and anti-biofi lm activity against CoNS isolates have not been previously reported.These fi ndings may represent novel, alternative approaches to biofilm control, especially in relation to infectious diseases caused by CoNS (1) .Furthermore, the biodeterioration of metals in industrial processes has detrimental effects on the environment with economic implications.The production of biofi lm by Bacillus spp.affects the level of metal corrosion in industrial processes (45) .Thus, the development of a preventive strategy against bacterialmediated corrosion is necessary.
The plant extracts studied here showed antimicrobial activity as well as biofi lm inhibition against the genera Bacillus and Staphylococcus.These fi ndings may represent the fi rst of many steps towards the development of new antimicrobial and antibiofi lm drugs using extracts from plants used in traditional folk medicine in northwestern Argentina.Given the importance of developing new strategies for controlling the growth of biofi lm because of its pathogenic association with several diseases or infl uence on the level of metal corrosion in industrial processes, further investigation of the natural biofi lm inhibition exhibited by Argentinean plant extracts is clearly warranted.

FIGURE 1 .
FIGURE 1. (A): Anti-microbial and anti-biofi lm activity of medicinal plants from Northwestern Argentina against native isolates from the genera Bacillus and Staphylococcus.(B): A total of 53 spore-forming bacteria were isolated from natural environments and 22 coagulase-negative Staphylococcus (CoNS) were isolated from a dialysis center.Of these, 24 were selected as biofi lm producers.The best biofi lm producing strains were identifi ed as Bacillus sp.+ Mcn4 (strain 3) and Staphylococcus sp.Mcr1 (strain 12) (p < 0.01).MIC: miminimum inhibitory concentration; SIC: sub-inhibitory concentration.

Table 2
shows the SIC and MIC values obtained for the fi ve plant extracts.These results refl ect a variable degree of antibacterial activity against the Bacillus and Staphylococcus genera, and indicate that dose-dependent antimicrobial activity was observed for all plant extracts.Extracts from L. chilense and S. fasciculatus showed the best antimicrobial properties, with SIC values of 62.50µg/ml for Bacillus sp.Mcn4 and 15.62µg/ml for Staphylococcus sp.Mcr1 for both extracts (Table 2).Furthermore, both L. chilense and S. fasciculatus showed antimicrobial activity against Bacillus strains, with an SIC value of 62.50µg/ml, the same SIC observed for chloramphenicol (Table 2).The antimicrobial activities shown by the genera Schinus and Lycium against Staphylococcus sp.Mcr1 included SIC values equal to those observed for gentamicin (15.62µg/ml).

Table 2 )
. However, these plant extracts showed high antimicrobial activity against Staphylococcus sp.Mcr1.It is notable that all plant extracts

TABLE 2
Effects of Larrea divaricata, Tagetes minuta, Tessaria absinthioides, Lycium chilense, and Schinus fasciculatus on the viability of bacteria from the genera Bacillus and Staphylococcus.
MIC (µg/ml) and SIC (µg/ml) were determined, with values expressed as means from triplicate assays. 1 MIC = the lowest plant metabolite concentration (µg/ml) that inhibited bacterial growth. 2 SIC = the highest extract concentration (µg/ml) below the MIC that did not inhibit bacterial growth.MIC: miminimum inhibitory concentration; SIC: sub-inhibitory concentration.