Introduction

Biofilm is a surface-associated community of microorganisms that is embedded in exopolymeric matrix (Donlan 2001; Douglas 2002). Extracellular polymeric substances (EPS) form a barrier that impedes the penetration of antimicrobial agents into the biofilm structure (Tote et al. 2010). The phenotype of biofilm cells changes along with the up- and downregulation of many genes (Donlan and Costerton 2002; Haussler and Fuqua 2013). The ability to form tenacious biofilms results in protection against low and high temperatures and low water activity, i.e., conditions employed in food industry technologies for storage of products: (Chmielewski and Frank 2003). Sterilization eliminates the planktonic forms of microorganisms but biofilms are 1000× more resistant and may be responsible for secondary infections (Bridier et al. 2015; Singh et al. 2014). The recent realization of the emergence of antimicrobial resistance has emphasized the need for better understanding of biofilm-related behavior of food microflora and potentially pathogenic contaminants in food industry and lead to an increased interest in elucidating the role of surface-microorganism behavior (see Table 1).

Table 1 Selected studies on biofilm formation by Candida spp., Staphylococcus spp., P. aeruginosa and E. coli
Full size table

Candida spp. are food pathogens and contaminants, as well as commensal microorganism without negative influence on healthy organisms. Candida species adhere readily to surfaces and form biofilms (Kuhn 2002); they also easily acquire resistance to host defense mechanisms and antifungal drugs (Kuhn et al. 2004; Ramage et al. 2005). Their pathogenicity manifests mainly in immunocompromised patients (Wang et al. 2014). The high resistance of Candida sp. biofilms is caused by the extracellular matrix (Donlan 2001; Rajasekharan et al. 2014).

Pseudomonas aeruginosa often forms part of natural microflora of vegetables, meat, milk, and other food products (Franzetti and Scarpellini 2007). It is considered to be the causative agent of off-flavors and unpleasant textural changes in finished food products (Martin et al. 2011; Van Tassell et al. 2012). P. aeruginosa is not only a food contaminant but also a pathogen responsible for nosocomial infections, bacteremia, pneumonia, and urinary tract infections. It belongs to pathogenic bacteria that may possess a broad resistance toward antibiotics and biocides (Leid et al. 2009; Page and Heim 2009). The natural resistance of P. aeruginosa to many antimicrobial drugs is usually connected with the ability to produce extracellular polymers and biofilm formation.

Staphylococcus spp. are one of the major etiological agents of food poisoning (Sievert et al. 2013) and a major milk pathogen. Besides contamination of foods, Staphylococcus sp. is also responsible for the increasing incidence of biomaterial infection, extended hospital stays, and patient mortality (Laverty et al. 2015).

Escherichia coli strains are among the common opportunistic pathogens found in humans; their pathogenicity is often connected with the biofilm formation (Paulo et al. 2010).

Synthetic additives are often used to lessen the probability of food microbial contamination and their side effects are one of the studied issues (Tepe et al. 2005). The importance of using nontoxic natural agents as replacements of synthetic ones is therefore evident. Some plants produce substances with pronounced antimicrobial properties that give them selective advantage in their natural environment (Zhao et al. 2005). Such compounds might be potentially used for food preservation (Delamare et al. 2007). Antimicrobial activity was demonstrated in essential oils, extracts from plants used in traditional Chinese medicine (Coenye et al. 2012; Marinas et al. 2015), but also in Ficus sp. (Awolola 2014) that contains substances such as resveratrol, pterostilbene, baicalein, and others (Huang 1999; Wu and Wen 2009).

Many procedures and biocidal agents such as benzalkonium chloride, and sodium hypochlorite are employed in food industry to eliminate microbial biofilms and suppress biofilm formation (Pagedar and Singh 2015). The efficacy of commonly applied biocides is dependent on many factors such as the cleaning regime, concentration, contact time, and properties of cleaned surface.

Plants are a rich source of biologically active compounds, many of which possess antimicrobial activity (Palombo 2011). We have focused on three such compounds: baicalein, resveratrol, and its structural analogue pterostilbene. Baicalein is found in high concentrations in medicinal plants, Scutellaria baicalensis and Oroxylum indicum cells (Dinda et al. 2017). Both resveratrol and pterostilbene are also found in plants, especially in fruits and vegetables. Resveratrol has a wide potential in medicinal applications due to its anti-aging, anti-carcinogenic, anti-inflammatory, and anti-oxidant properties (Cottart et al. 2010; Kolouchova et al. 2005). All three substances have long been studied due to their positive influence on human health, stemming from their presence as active compound in traditional medicine applications (baicalein) or their antioxidative properties (resveratrol, pterostilbene). Their antimicrobial activity was discovered much later and their antibiofilm properties are only now being discovered. The antibiofilm activity of natural substances is often strain or genera-specific, therefore it is necessary to study a wide range of microorganisms and experimental approaches to obtain comprehensive information leading to potential applications. The antimicrobial and antibiofilm activity of baicalein, resveratrol, and pterostilbene has been observed against selected microorganisms (Zeng et al. 2008; Chen et al. 2016; Hu et al. 2017; Cho et al. 2015; Lee et al. 2014b). For potential application of these substances, it is necessary to study their influence on both planktonic and biofilm populations, as it is known that their properties vary significantly. This is intensively studied in medicine, as pathogenic microorganisms freely switch between free and adhered form of life which diminishes the effectivity of standard antimicrobials. In this regard, the baicalein effectivity was studied (Serpa et al. 2012), as well as the stilbenes (Lee et al. 2014a; Li et al. 2014; Nimmy et al. 2014). The novelty of our study is the combination of the approaches of observing their effect on both initial adhesion and biofilm eradication, as well as comparison with substance efficiency toward the planktonic cells.

One of the advantages of all three compounds is their low cytotoxicity. Baicalein was reported to show almost no toxicity to human normal epithelial, peripheral and myeloid cells (Dinda et al. 2017). Toxicity study of an herbal formulation containing Scutellaria baicalensis extract with 2 g/L of baicalein on nude mice showed no toxicities. The nude mice ingested the extract in a diet and were monitored for 13 weeks, showed no abnormalities in their behavior (Donald et al. 2012). The stilbenes such as resveratrol and pterostilbene also were reported to show low toxicity in normal hemopoietic stem cells (Tsai et al. 2017) and pterostilbene oral administration to nude mice (100 μg/kg per day) for 8 weeks did not produce signs of acute toxicity (Kosuru et al. 2016).

The present study was undertaken to evaluate the antibacterial/anti-biofilm properties of the natural compounds baicalein, resveratrol, and pterostilbene against selected microorganisms—Candida albicans, Staphylococcus epidermidis, Pseudomonas aeruginosa, and Escherichia coli—that cause food and nosocomial infections.

Material and methods

Microbial strains and growth media

Candida albicans DBM 2164, Staphylococcus epidermidis DBM 3179, and Escherichia coli DBM 3125 were kindly provided by the Collection of Microorganisms at UCT Prague. Pseudomonas aeruginosa NRRL B-59189 was obtained from the ARS Culture Collection, Bacterial Foodborne Pathogens and Mycology Research Unit, National Center for Agricultural Utilization Research, USA.

C. albicans was pre-cultured (30 °C) before each experiment in yeast-peptone-dextrose medium (YPD), S. epidermidis (37 °C) in Tryptone Soya Broth medium (TSB), and P. aeruginosa and E. coli (30 °C) in Luria Broth medium (LB). All microbial strains were stored at − 70 °C. The microbial strains were pre-cultured for 24 h to achieve exponential phase of growth (100 ml in Erlenmeyer flasks, 100 rpm).

Biologically active agents

The natural substances (baicalein, resveratrol and pterostilbene) were purchased from Sigma-Aldrich and were dissolved in dimethylsulfoxid (final concentration of DMSO in the medium was 1% in all assays). This DMSO concentration was proven not to interfere with the growth and evaluation methods by independent control cultivations.

Determination of minimum inhibitory concentration of planktonic cells

The response of planktonic cells to various concentration of natural substances was characterized as the minimal inhibitory concentration. Minimum inhibitory concentrations (MICs) for planktonic cells by baicalein, resveratrol and pterostilbene were determined by microdilution method according to (Sharma et al. 2010). The cultivation of planktonic cells was carried out in flat-bottomed microtiter plates using Bioscreen C analyzer (Oy Growth Curves Ab Ltd., Finland). Aliquots of 30 μl of standard cell suspensions of the microorganisms (A600nm = 0.1) in growth medium were transferred into the microtiter plates with serially diluted antimicrobial agent and growth medium to a final volume of 320 μl, followed by incubation for 24 h at 30 or 37 °C (depending on microbial strain). The MICs were assigned as the lowest concentration that did not allow visible growth of more than 20% (MIC80) or 50% (MIC50) after overnight incubation according to the definition by Andrews (2001). Experiments were performed in triplicate. Control cultivations without the natural substances were included.

Determination of minimum adhesion inhibition concentration and the agent efficiency against biofilm formation

For the initial adhesion (biofilm formation) testing and the minimum adhesion inhibition concentration (MAIC) determination, the substances were added at the beginning of the cultivation.

The biofilm cultivation was performed in 96-well microtiter plates, into which aliquots of 200 μl of standard cell suspensions (A600nm = 0.6) in growth medium and the appropriate agent were transferred into each well. The microtiter plate was covered with a lid held by semipermeable parafilm and was placed into an orbital shaker (150 rpm) at 30 or 37 °C (according to microorganism) to allow for biofilm development and the cultivation was carried out for 24 h. After 24 h, each well was washed three times with saline and biofilm formation on the bottom of wells was evaluated by XTT reduction assay and the crystal violet staining method. Minimum adhesion inhibition concentration (MAIC) was determined by the XTT reduction assay—MAICs were assigned as the lowest concentration that did not exhibit more than 20% (MAIC80) or 50% (MAIC50) of metabolic activity of biofilm cells. The total biofilm biomass was determined by the crystal violet staining method. As control, cultivations without the natural substances were used. All experiments were performed in 8 parallels.

Determination of minimum biofilm eradication concentrations and the agent efficiency against pre-formed biofilms

For the biofilm susceptibility testing and minimum biofilm eradication concentration (MBEC) determination, the substances were added to a preformed, 24-h-old biofilm. The biofilm cultivation was performed in 96-well microtiter plates. Aliquots of 200 μl of standard cell suspensions (A600nm = 0.6) in growth medium were transferred into a 96-well microtiter plate, followed by cultivation for 24 h at 30 or 37 °C (according to microorganism). The microtiter plate was covered with a lid held by semipermeable parafilm and was placed into an orbital shaker (150 rpm) at 30 or 37 °C (according to microorganism) to allow for biofilm development and the cultivation was carried out for 24 h. After 24 h (biofilm development without agent), each well was washed three times with saline, followed by the addition of growth medium and serially diluted agents. The cultivation continued for the next 24 h (30 or 37 °C at 150 rpm in an orbital shaker), after which each well was washed three times with saline and biofilm on the bottom of wells was evaluated by XTT reduction assay and the crystal violet staining method. Minimum biofilm eradication concentrations (MBEC) were determined by the XTT reduction assay—MBECs were assigned as the lowest concentration that did not exhibit more than 20% (MBEC80) or 50% (MBEC50) of metabolic activity of biofilm cells. The total biofilm biomass was determined by the crystal violet staining method. Biofilm populations were quantified by XTT reduction assay and the crystal violet staining method. As control, cultivations without the natural substances were used. All experiments were performed in 8 parallels.

XTT reduction assay

Minimum adhesion inhibitory concentrations (MAIC) and minimum biofilm eradication concentrations (MBEC) of natural substances were determined using XTT (2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-5-[(phenylamino)carbonyl]-2H–tetrazoliumhydroxide) reduction assay (Zhou et al. 2012). For each assay, the wells were washed three times with saline to remove planktonic cells and into each well was added 155 μL of glucose solution (36.1 mg/mL), 40 μL of XTT solution (1 mg/mL), and 5 μL of menadione solution (0.11 mg/mL). The plate was incubated in dark at 30 or 37 °C for 3 h and then a 100 μl volume from each well was transferred into 96-well microtiter plate and the colorimetric change was determined using spectrophotometric reader (Tecan, Switzerland) at 492 nm. Experiments were performed in 8 parallels.

Crystal violet staining

The total biofilm biomass was determined by staining the biofilm with crystal violet (CV). In the quantification process, each well was washed three times with saline and into each well was added 100 μL of 0.1% filtered CV (Carl Roth, Germany) solution. The plate was incubated for 20 min at room temperature. Afterwards, each well was washed three times with saline. CV bound to the biofilm biomass was released by adding 200 μL of 96% ethanol (Penta, Czech Republic) and, after incubating for 10 min at room temperature, measured using spectrophotometric reader (Tecan, Switzerland) at 580 nm. Experiments were performed in 8 parallels.

Light microscopy—Cellavista device

In one representative sample from eight parallels, the area populated by biofilm was visualized by Cellavista device, as previously described (Kvasničková et al. 2015). Cellavista device (Synentech, Germany) is a fully automated cell imager that incorporates bright-field inverted microscope with high-resolution camera and software for image analysis.

Statistical analysis

The significance of difference between control and substance efficiency on adhesion inhibition and preformed biofilm was determined by one-way analysis of variance (ANOVA) and Tukey’s HSD test.

Results and discussion

Determination of minimum inhibitory concentrations of plant-derived agents

The MIC50 values (inhibition of planktonic cells) of baicalein, pterostilbene and resveratrol are shown in Table 2. The inhibitory properties of specific compounds were dependent on the microorganism type, but pterostilbene was found as the most effective substance with lowest MIC50 against all studied microorganisms. Baicalein had inhibitory properties only against planktonic cells of C. albicans. Inhibition of planktonic growth by resveratrol was observed in C. albicans, S. epidermidis, and E. coli.

Table 2 Minimum inhibitory concentrations (MIC50 and MIC80) of baicalein, resveratrol and pterostilbene for C. albicans DBM 2164, S. epidermidis DBM 3179, P. aeruginosa NRRL B-59189 and E. coli DBM 3125
Full size table

The MIC50 of pterostilbene for C. albicans (5 mg/L) that we found was very low and was near that of amphotericin B (2 mg/L) reported by Melo et al. (2011). The MIC80 of the agent for C. albicans (10 mg/L) was lower than the value determined by Li et al. (2014) (32 mg/L). Resveratrol, a close analogue of pterostilbene, had a MIC50 value against C. albicans 160 mg/L, a value similar to the results reported by other articles (Jung et al. 2005; Jung et al. 2007; Okamoto-Shibayama 2010; Weber et al. 2011). Though it was described to decrease the mitochondrial activity of cells in some strains, it is also known to show no inhibitory effect against other C. albicans strains (Collado-Gonzalez et al. 2012; Weber et al. 2011). Baicalein MIC50 for C. albicans (13–26 mg/L) was similar to that reported by Serpa et al. (2012), but significantly lower than the 500 mg/L reported (as baicalein MIC) by Wang et al. (2014). The mechanisms of action of baicalein were reported as induction of programmed cell death in C. albicans cells (Dai et al. 2009), promotion of the formation of blastoconidia, and formation of filaments (Serpa et al. 2012).

We found the MIC50 (60 mg/L) against P. aeruginosa only for pterostilbene and no MIC80 was found for any of the agents in the studied concentration ranges. The MIC80 against another gram-negative bacterium, E. coli, were determined for both pterostilbene and resveratrol. MIC80 for resveratrol (250 mg/L) against E. coli was lower than the value reported by Paulo et al. (2010) who observed resveratrol activity toward E. coli (> 400 mg/L) but did not observe any inhibition for P. aeruginosa (Paulo et al. 2010).

P. aeruginosa exhibits resistance to many biocides such as isopropanol, peracetic acid, sodium hypochlorite, iodophore or hydrogen peroxide, which are able to inhibit only planktonic cells, albeit in high concentrations (Tote et al. 2010). The most usual cleaning cycle in dairy industry uses a 15-min contact time of the biocide with the surface. Like the usually employed concentrations, this may be insufficient for P. aeruginosa elimination. P. aeruginosa MIC (for planktonic cells) were determined at 300 mg/L for benzalkonium chloride and sodium hypochlorite and about 50-fold lower for iodophore (Pagedar and Singh 2015).

In S. epidermidis, we found MIC80 for both pterostilbene and resveratrol, but baicalein did not inhibit the growth below the concentration 100 mg/L and its MICs were not found. Available data confirm the high concentration of baicalein necessary to inhibit the growth of many bacterial strains, both gram positive and negative. Luo et al. (2016) reported baicalein MIC50 for P. aeruginosa at 1000 mg/L while Enterococcus isolates were found to be inhibited by concentrations in excess of 256 mg/L (Chang et al. 2007). Gram-positive bacteria are inhibited by resveratrol in the growth phase of the bacterial cell cycle (Paulo et al. 2010). According to Yun et al. (2012), the antibacterial activity of baicalein against S. aureus involved membrane permeabilization, protein synthesis inhibition and modification of the activity of succinate dehydrogenase, malate dehydrogenase and DNA topoisomerase I and II (Yun et al. 2012).

Inhibition of biofilm formation by plant-derived agents

In Table 3 is shown the effect of substances on the biofilm formation determined by the XTT method as the minimal adhesion inhibiting concentration (MAIC). In Fig. 1, the influence of a series of concentrations of agents on total biofilm biomass determined by the crystal violet is shown. Crystal violet staining (CV) and determination of metabolic activity of cells by the XTT assay are two of the most often used methods used for the evaluation of biofilm development, but there is an obvious distinction due to the difference in their principle. Crystal violet staining determines total biofilm biomass—both metabolically active and inactive cells and EPS (Hawser et al. 1998; Ramage et al. 2001) while the XTT assay is based on the reduction of XTT sodium salt by dehydrogenases of metabolically active cells and reports on actual metabolic activity of the cells (Bizerra et al. 2008; O'Toole et al. 2000).

Table 3 Minimum adhesion inhibitory concentrations (MAIC) and minimum biofilm eradication concentrations (MBEC) of baicalein, resveratrol and pterostilbene for C. albicans DBM 2164, S. epidermidis DBM 3179, P. aeruginosa NRRL B-59189 and E. coli DBM 3125 (determined by the XTT reduction assay)
Full size table
Fig. 1
figure 1

Inhibition of total biofilm biomass (determined by the crystal violet staining) in initial adhesion of C. albicans DBM 2164, S. epidermidis DBM 3179 and E. coli DBM 3125 by baicalein, resveratrol, and pterostilbene. Error bars represent standard deviation. Control (no agent, 100%), *p < 0.05; **p < 0.01; ***p < 0.001

Full size image

A significant inhibition of biofilm formation was observed in E. coli. Pterostilbene had the lowest MAIC50 (40 mg/L) followed by baicalein (80 mg/L). Similar values for baicalein and resveratrol were reported by Bakkiyaraj et al. (2013). Resveratrol was reported to reduce swimming and swarming motilities and repress several key motility and flagellar genes (flhD, fimA, FimH, motB), without inhibiting the growth of planktonic cells (Lee et al. 2013). The degree of resistance may be strain dependent; thus, the E. coli DBM 3125 we studied showed higher resistance to resveratrol than the E. coli strain O157:H7 studied by Lee et al. (2013) whose biofilm was significantly reduced by a resveratrol concentration of 10 mg/L.

In S. epidermidis, baicalein at 80 mg/L concentration inhibited the biofilm formation of by 50%, pterostilbene exhibited a similar effect at 50 mg/L, and resveratrol at 170 mg/L. Among gram-positive bacteria, high concentrations of resveratrol are necessary for their inhibition, e.g., Propionibacterium acnes biofilm formation was found to be inhibited by 80% by resveratrol in concentration 3.2 g/L (Coenye et al. 2012).

In our work, pterostilbene at the highest concentration of 20 mg/L decreased Candida biofilm formation only by 30% (data not shown), resveratrol inhibited it with MAIC50 88 mg/L and baicalein, which induces apoptosis of C. albicans cells by suppressing extrusion of the drug due to inhibition of efflux pumps (Dai et al. 2009; Huang et al. 2008), at 200 mg/L. The resistance of Candida biofilms is an issue in both food industry and medicine (Mah and O’Toole 2001).

Beside the MAIC and MBEC determined by the XTT reduction assay, the total biofilm biomass during biofilm formation was determined by the crystal violet staining (Fig. 1).

P. aeruginosa was found to be the most resistant to the agents. The anti-biofilm activity of the agents that influenced the total biofilm biomass of P. aeruginosa (determined by crystal violet staining), was omitted, because none of the substances tested had any significant effect on the inhibition of P. aeruginosa. The highest adhesion inhibition which was obtained under the experimental conditions corresponded to a 10% decrease in metabolic activity of cells during biofilm formation in comparison with control (data not shown). Cho et al. (2013) showed 50% inhibition of P. aeruginosa biofilm formation by resveratrol (100 mg/L) while the highest concentration of resveratrol (170 mg/L) used in our study did not significantly inhibit the P. aeruginosa biofilm formation. P. aeruginosa exhibits a high resistance toward a variety of antimicrobials such as aminoglycosides and macrolide antibiotics (Pompilio et al. 2015). Pathogenesis and biofilm formation by P. aeruginosa are associated with the production of extracellular virulence factors and cell motility (Wahman et al. 2015). The weak penetration of antimicrobials into biofilm is the result of increased production of EPS in the biofilm and consequent decrease of diffusion of antibiotics and biocides (Dynes et al. 2009; Pompilio et al. 2015).

Baicalein inhibited the total biofilm biomass formation (Fig. 1a) in S. epidermidis, but was not active against E. coli. In C. albicans, baicalein even led to 50% increase higher total biofilm biomass, even though the metabolic activity of cells was suppressed (MAIC50 400 mg/L was found—see Table 3). Similar results were reported by Cao et al. (2008) who observed the inhibitory effect of baicalein on the metabolic activity of C. albicans cells and determined a MAIC50 of 200 mg/L. A twofold increase in total biofilm biomass was observed at the baicalein concentration of 200 mg/L. Baicalein therefore decreases the metabolic activity of C. albicans cells but promotes the total biofilm biomass production.

The results obtained by methods observing the viability of sessile cells (e.g., XTT reduction assay) and the total biofilm biomass (crystal violet staining) must therefore be carefully compared when used for the determination of the inhibitory effect of substances on the biofilm, especially for certain genera of microorganisms.

Resveratrol (Fig. 1b) had a concentration-dependent inhibitory effect on both bacteria, S. epidermidis and E. coli, as well as on the yeast C. albicans. Similar results were observed by Selma et al. (2012), and resveratrol at high concentration (3.2 g/L) also inhibited the adhesion of E. coli O15:H17 and the biofilm formation of gram-positive Propionibacterium acnes (Coenye et al. 2012).

Increasing concentrations of pterostilbene (Fig. 1c) were found to inhibit biofilm formation of C. albicans and E. coli. The effect of pterostilbene S. epidermidis biofilm biomass was found to be opposite; with increasing concentration of the agent the total biofilm biomass increased, but did not match the detected metabolic activity of the cells. At 20 mg/L, the adhesion monitored as the total biofilm biomass increased by 133%, while the metabolic activity of cells in biofilm decreased by 20%. The anti-biofilm effect of pterostilbene is based on its antiadhesive and anti-morphological transition activities (Li et al. 2014). Li et al. (2014) also observed inhibition of C. albicans biofilm formation by 60% (4 mg/L pterostilbene) and 90% (32 mg/L pterostilbene) and a decrease of the cellular surface hydrophobicity of C. albicans (Li et al. 2014).

Eradication of pre-formed biofilm by plant-derived agents

Minimum biofilm eradication concentrations (determined by the XTT reduction assay) are shown in Table 3. The influence of a series of concentrations of agents on total biofilm biomass determined by the crystal violet is shown in Fig. 2.

Fig. 2
figure 2

Inhibition of total biofilm biomass (determined by the crystal violet staining) of pre-formed biofilm of C. albicans DBM 2164, S. epidermidis DBM 3179 and E. coli DBM 3125 by baicalein, resveratrol, and pterostilbene. Error bars represent standard deviation. Control (no agent, 100%), *p < 0.05; **p < 0.01; ***p < 0.001

Full size image

Resveratrol and pterostilbene displayed similar effect on E. coli biofilm eradication as on inhibition of biofilm formation, MBEC50 was found for both agents.

S. epidermidis pre-formed biofilm metabolic activity was effectively inhibited by both resveratrol and pterostilbene. Both substances were more efficient in biofilm eradication than in the inhibition of biofilm formation. The MBEC50 values (100 and 25 mg/L, respectively) are almost half the MAIC50 values (170 and 50 mg/L, respectively).

Baicalein exhibited less significant influence on S. epidermidis biofilm eradication, MBEC was not found and the total biofilm biomass was not significantly affected. The concentration of 400 mg/L decreased the amount of mature biofilm by a mere 20%. Cui et al. (2016), who studied the anti-biofilm effect of sage oil on S. aureus, found significant biofilm eradication only above the concentration of 2000 mg/L.

The resveratrol MBEC50 value we found for the yeast C. albicans was 88 mg/L. Pterostilbene at a concentration of 20 mg/L reduced the total biofilm biomass by 15%. These agents thus may contribute to the eradication of C. albicans biofilm, which is often found very resistant against the commonly used antibiotics. It should be noted that Melo et al. (2011) did not find fluconazole and amphotericin B MBEC50 and MBEC80 values for C. albicans, even though the monitored concentrations were greater than 2000 mg/L.

As in biofilm formation, the pre-formed biofilms of the gram-negative bacterium P. aeruginosa were not significantly inhibited by the agents and therefore the results of total biofilm biomass are not shown. Only the highest concentration of baicalein (400 mg/L) reduced the amount of P. aeruginosa biofilm by 10% (data not shown).

Resveratrol eradicated the total biofilm biomass of preformed biofilms of C. albicans, S. epidermidis, and E. coli with a similar intensity and was the most effective—with increasing concentration of resveratrol the total biofilm biomass was reduced, the reduction at 170 mg/L being almost 35%. This effect was visualized by light microscopy in Fig. 3.

Fig. 3
figure 3

Effect of resveratrol on pre-formed C. albicans biofilm. a Control (no agent). b Resveratrol 170 mg/L (scale bar 200 μm)

Full size image

The studied plant-derived compounds may be an advantageous alternative to the commonly used biocides such as EDTA, benzalkonium chloride, sodium hypochlorite, or iodophore that have often high MICs (hundreds of mg/L) and MBECs (up to thousands of mg/L) (El-Sharif and Hussain 2011; Pagedar and Singh 2015). Although antibiotics have lower MICs, their MBECs are high (up to hundreds and thousands of mg/L) (Barchiesi et al. 1998; Huang et al. 2008; Pompilio et al. 2015; Serpa et al. 2012; Zhou et al. 2012) and nevertheless their application in food industry is impossible.

Conclusion

Overuse or misuse of antibiotics for the control of infections and the difficulties in eradication of biofilm formation and resistance of microorganisms increases the pressure on finding new techniques for CIP processes. Our study demonstrates the ability of natural substances derived from plants to inhibit microbial biofilms that occur in both the food industry and medicine. Baicalein, resveratrol and pterostilbene, all originally isolated from plants, have been shown to inhibit the formation of biofilms and disrupt preformed biofilms. In particular, resveratrol and pterostilbene displayed high inhibition of biofilm cell viability for the opportunistic pathogens under study. The highest resistance to these substances was found in Pseudomonas aeruginosa. In Escherichia coli, all three compounds caused a significant drop of cell viability in biofilm formation and eradication of pre-formed biofilm. Our results show that these new antimicrobials can be used with good effect in food industry and medicine and can contribute to an improvement of strategies designed to ensure better food safety and quality.