Bioﬁlm Degradation of Nontuberculous Mycobacteria Formed on Stainless Steel Following Treatment with Immortelle ( Helichrysum italicum ) and Common Juniper ( Juniperus communis ) Essential Oils

: Nontuberculous mycobacteria, like other opportunistic premise plumbing pathogens, produce resistant bioﬁlms on various surfaces in the plumbing system including pipes, tanks, and ﬁttings. Since standard methods of water disinfection are ineffective in eradicating bioﬁlms, research into new agents is necessary. Essential oils (EOs) have great potential as anti-bioﬁlm agents. Therefore, the pur-pose of this research was to investigate the potential anti-bioﬁlm effect of common juniper ( Juniperus communis ) and immortelle ( Helichrysum italicum ) EOs. Minimum inhibitory concentrations (MIC), minimum bactericidal concentrations (MBC), and minimum effective concentrations of EOs on Mycobacterium avium , M. intracellulare , and M. gordonae were tested. Additionally, bioﬁlms on the surface of a stainless steel disc were treated with single or mixed concentration of EOs, in order to investigate their degeneration via the bacterial count and confocal laser scanning microscopy (CLSM). H. italicum EO showed the strongest bioﬁlm degradation ability against all Mycobacteria strains that were tested. The strongest effect in the bioﬁlm degradation after the single or mixed applications of EOs was observed against M. gordonae , followed by M. avium . The most resistant was the M. intracellulare bioﬁlm. Synergistic combinations of J. communis and H. italicum EOs therefore seem to be an effective substance in bioﬁlm degradation for use in small water systems such as baths or hot tubs.


Introduction
Mycobacteria originated 150 million years ago [1]. The genus Mycobacterium is the only member of the family Mycobacteriaceae from the order Actinomycetales and the class Actinomycetes. Today, more than 200 species belong to the genus Mycobacterium, which include obligate and opportunistic pathogens and saprophytes [2]. Nontuberculous mycobacteria (NTM) are a heterogeneous group of environmental bacteria mainly isolated from water, soil, dust, various animals, milk, and dairy products [3]. Although mostly apathogenic, nowadays, they increasingly represent important environmental opportunistic pathogens [4]. Mycobacterium avium and M. intracellulare are members of the Mycobacterium avium complex (MAC). These are slow-growing unpigmented mycobacteria that form smooth, flat, transparent colonies. MACs are the most frequently isolated pathogenic NTM species from respiratory samples [5]. M. gordonae is a mycobacterium that forms smooth The natural EOs of common juniper (Juniperus communis) and immortelle (Helichrysum italicum) used in this research were purchased from IREX AROMA d.o.o., Zagreb, Croatia. The EOs were produced in 2018. Gas chromatography and mass spectrometry (GC/MS) analyses of EOs were done [31]. EOs have been shown to have chemical composition characteristic for the said essential oils. Each EO was dissolved in dimethylsulfoxide (DMSO; Kemika, Zagreb, Croatia) to obtain a stock suspension, which was stored in sterile glass vials in the dark at 4 • C prior to use.

Sterile Tap Water Sample
In all experiments, tap water from the public water supply system of the city of Rijeka was used. Physicochemical parameters of tap water in Rijeka are regularly monitored by authorized Croatian testing laboratories certified to provide chemical analysis of drinking water and show values that rarely deviate. The water is colorless and odorless, with a normal temperature parameter depending on seasonal variations. It has low turbidity, neutral to slightly alkaline pH (from 7.5 to 8.0), low conductivity (0.211-0.250 mS cm −1 at 20 • C), and moderate total hardness (135 mg L −1 ). According to these parameters, it is considered as medium hard. The tap water sample in a glass bottle was left at room temperature for two days to allow for dechlorination. The water sample was then autoclaved at 121 • C for 15 min and cooled to room temperature and stored at 4 • C until use.

Checkerboard Synergy Method
To determine the potential interaction effect of J. communis and H. italicum EOs on NTM, the checkerboard synergy method was used, as described previously [33][34][35]. Briefly, stock solutions and serial two-fold dilutions of each EO were prepared in 7H9S. These dilutions were arrayed in a grid pattern, with the J. communis EO dilution series running perpendicular to that of the H. italicum EO. The combinations of concentrations of each EO tested are shown in the results section ( Figure 1). An inoculum of each Mycobacterium isolate (10 6 CFU mL −1 ) was prepared in 7H9S and added along with 0.015% resazurin solution (Sigma-Aldrich, Darmstadt, Germany) to wells containing diluted EOs. Positive (bacterial inoculum in 7H9S) and negative (7H9S) growth controls were prepared. Additionally, the antibiotic amikacin (Sigma-Aldrich, Darmstadt, Germany) was also tested against all three mycobacteria. The final concentration of DMSO as a solvent was approximately 10% and its effect was tested against the selected mycobacteria. The plates were incubated for four days under aerobic conditions at 30 • C (M. gordonae) or 37 • C (M. avium and M. intracellulare), and then dilutions from each well were inoculated on 7H10S in duplicate and incubated for a further four weeks. Fractional inhibitory concentration or fractional bactericidal concentration [36] and fractional inhibitory concentration index (FIC i ) or fractional bactericidal concentration index (FBCi) were determined as previously described by Bassole et al. and White et al. [20,37]. Baes on the FIC i or FBC i values, a combination of EOs was considered synergistic if FIC i /FBC i was ≤0.5, additive if FIC i /FBC i was >0.5 and ≤1.0, indifferent when FIC i /FBC i was >1.0 and ≤4, and antagonistic if FBC i was >4 [38].

Effect of Juniperus communis and Helichrysum italicum Essential Oils on Mycobacterial Biofilm on Stainless Steel Discs in Sterilized Tap Water
The effect of different concentrations of J. communis and H. italicum EOs as well as synergistic or additive combinations of these EOs on the degradation of the biofilm of M. avium, M. intracellulare, and M. gordonae was tested on stainless steel discs (diameter, 5 mm; American Iron and Steel Institute (AISA) type 316) in sterilized tap water (STW). The discs were left overnight in 70% ethanol (Kemika, Zagreb, Croatia), rinsed with distilled water, air dried, dry heat sterilized at 160 • C, and then aseptically transferred to the wells of microtiter plates (24-well microtiter plates, Falcon, Becton Dickinson, Franklin Lakes, New Jersey, USA). Then, a suspension of 10 6 CFU mL −1 of mycobacterial cells was prepared in STW and added to wells containing discs to form a biofilm. The plates were incubated for 72 h at Cell viability assays were performed (Live/Dead BacLight Bacterial Viability Kit; Invitrogen, Carlsbad, California, USA) according to the manufacturer's protocol. Briefly, a biofilm of M. avium and M. intracellulare was grown for three days on round stainless steel discs. These were exposed to the individual effect of J. communis or H. italicum EOs, in synergistic or additive combinations, for 22 h at 37 • C. The stainless steel discs were carefully washed with STW to remove planktonic cells. Fluorescent-stain working solution was prepared by adding 3 µL of the SYTO ® 9 stain and 3 µL of the propidium iodide (PI) stain to 1 mL of filter-sterilized water. This staining solution was then applied to the surface of the disc and incubated in the dark for 15 min. The samples were then washed with sterile saline to remove excess dye. Fluorescence from the stained cells was observed using an Olympus confocal microscope FV300 (Olympus Optical Company, Tokyo, Japan) with a 40x LCPlanF objective. The excitation/emission maxima for these dyes are around 480/500 nm for the SYTO ® 9 stain and 490/635 nm for PI. Simultaneous dual-channel imaging was used to display green and red fluorescence. The obtained images were saved

Statistical Analysis
All assays were repeated three times. Experimental data were expressed as means with standard deviations and analyzed using STATISTICA commercial software, 12.0 (StatSoft, Tulsa, OK, USA). Differences between groups of samples were analyzed using the Kruskal-Wallis ANOVA on ranks test, while the effects of EOs on mycobacterium were tested using the Mann-Whitney U test. Differences with p < 0.05 were considered to be statistically significant.

Checkerboard Synergy Method
The MIC and MBC values obtained for J. communis and H. italicum EOs against M. avium, M. intracellulare, and M. gordonae were 1.6 mg mL −1 and 3.2 mg mL −1 , respectively (Figures 1 and 2). For the control antibiotic, amikacin, MIC was 0.002 mg mL −1 for M. avium, 0.001 mg mL −1 for M. intracellulare, and 0.0005 mg mL −1 for M. gordonae ( Figure 1). The DMSO growth control showed that the concentration applied did not affect the growth of the mycobacteria being tested.
The best effective combination of low synergistic combinations of the EOs to achieve a high efficacy against M. avium in the checkerboard synergy method was 0.8 mg mL

Effect of Juniperus communis and Helichrysum italicum Essential Oils on Mycobacterial Biofilm on Stainless Steel Discs in Sterilized Tap Water
As can be seen in Figure 3, H. italicum EO was more effective than J. communis EO at degrading biofilm formed in STW on stainless steel AISI 316 discs for all treatments of mycobacteria. Almost all of the treatments (excluding the concentration of J. communis MIC for M. avium and M. intracellulare) caused statistically significant biofilm degradation (p < 0.05) using both EOs, when compared to the control group. In the control group, the 2 × MIC concentration of H. italicum EO led to the most substantial degradation of the biofilm. No statistically significant differences were found in M. avium and M. intracellulare biofilm degradation with H. italicum EO at concentrations of either the MIC and 2 × MIC (Figure 3a,b). In contrast, the M. gordonae (Figure 3c) biofilm showed statistically significant biofilm degradation using either the MIC or 2 × MIC of H. italicum EO (p < 0.05). J. communis EO demonstrated a lower effectiveness on biofilm degradation in all treatments and no statistically significant differences were found for any of the mycobacteria.
Subinhibitory synergistic concentrations of J. communis and H. italicum EO did not degrade biofilms of M. gordonae formed on stainless steel discs in STW in a statistically significant manner (Figure 4c). Furthermore, significant degradation of M. avium biofilm by J. communis and H. italicum EOs was observed using concentrations of 0.8 mg mL −1 and/or 0.012 mg mL −1 (Figure 4a). Meanwhile, a subinhibitory concentration of H. italicum EO (1.6 mg mL −1 ) degraded biofilms formed by M. intracellulare (p < 0.05; Figure 4b).
The combination of subinhibitory concentrations of J. communis and H. italicum EOs had a significant effect (p < 0.05) on the degradation of all mycobacteria biofilms formed on stainless steel discs.

Effect of Juniperus communis and Helichrysum italicum Essential Oils on Mycobacterial Biofilm on Stainless Steel Discs in Sterilized Tap Water
As can be seen in Figure 3, H. italicum EO was more effective than J. communis EO at degrading biofilm formed in STW on stainless steel AISI 316 discs for all treatments of mycobacteria. Almost all of the treatments (excluding the concentration of J. communis MIC for M. avium and M. intracellulare) caused statistically significant biofilm degradation (p < 0.05) using both EOs, when compared to the control group. In the control group, the 2 x MIC concentration of H. italicum EO led to the most substantial degradation of the biofilm. No statistically significant differences were found in M. avium and M. intracellulare degrade biofilms of M. gordonae formed on stainless steel discs in STW in a statistically significant manner (Figure 4c). Furthermore, significant degradation of M. avium biofilm by J. communis and H. italicum EOs was observed using concentrations of 0.8 mg mL −1 and/or 0.012 mg mL −1 (Figure 4a). Meanwhile, a subinhibitory concentration of H. italicum EO (1.6 mg mL −1 ) degraded biofilms formed by M. intracellulare (p < 0.05; Figure 4b). The combination of subinhibitory concentrations of J. communis and H. italicum EOs had a significant effect (p < 0.05) on the degradation of all mycobacteria biofilms formed on stainless steel discs.

Cell Viability of Biofilm on Stainless Steel Discs Treated with Juniperus communis and Helichrysum italicum Essential Oils
In order to further investigate the anti-biofilm properties of J. communis and H. italicum EOs, confocal laser scanning microscopy (CLSM) analyses were performed ( Figure 5). Some regions of the biofilm appeared yellow because of overlapping green and red cells.
The CLSM results indicate a strong synergistic effect of J. communis and H. italicum EOs on biofilm eradication of both bacterial strains. M. intracellulare was more sensitive, with more total red fluorescence (154.7 AU) than M. avium (137.9 AU). Individual treatment with H. italicum EO showed a better anti-biofilm effect than that with J. communis EO on both Mycobacterium species, although M. intracellulare was again the more sensitive species.         Lower total fluorescence of the M. intracellulare biofilm and a lower number of cells in the biofilm (Figure 6) may indicate both biofilm destruction and cell detachment due to EO action.

Discussion
Mycobacterium avium and other NTMs belong to the specific group of waterborne microorganisms named opportunistic premise plumbing pathogens, which are normal inhabitants of premise plumbing systems and can cause infections in immunocompromised patients [12,40].
Cell wall hydrophobicity and aggregation ability in liquid media are key factors in their pathogenicity and biofilm formation as well as play a crucial role in NTM resistance

Discussion
Mycobacterium avium and other NTMs belong to the specific group of waterborne microorganisms named opportunistic premise plumbing pathogens, which are normal inhabitants of premise plumbing systems and can cause infections in immunocompromised patients [12,40].
Cell wall hydrophobicity and aggregation ability in liquid media are key factors in their pathogenicity and biofilm formation as well as play a crucial role in NTM resistance to disinfectants, acidic environments, and high ambient temperatures. The growth of NTM in the form of biofilms on glass, copper, galvanized steel or plastic results in their resistance to antimicrobials and disinfectants [41,42].
Due to the extreme resistance of OPPP biofilms including those of NTM, new approaches or new substances are needed to fight biofilm formation and destruction. Our previous studies have demonstrated the great potential of EOs as inhibitors of mycobacterial adhesion or biofilm formation on polystyrene as well as inhibitors of adhesion to living cells (amoebae and HeLa cells) [27,32,33,43]. The EOs of J. communis and H. italicum from coastal regions of Croatia have been shown to be particularly effective. In this study, we used commercial EOs produced from the same manufacturer two years apart, and although deviations in the amount of α-pinene and some other compounds can be seen, all repeated experiments gave the same MIC concentrations and synergistic effects of combinations of these two oils. Obviously, the antibacterial effect against NTM is not a result of a single dominant component, but the effect of the interaction of different components in these oils [27,32].
Haziri et al. [44] found moderate to high antimicrobial activity of J. communis EO against S. aureus, E. coli, and Hafnia alvei, while P. aeruginosa was shown to be resistant to this oil. Klančnik et al. [24] analyzed the effect of J. communis EO on the adhesion of C. jejuni to AISI 304 stainless steel. They reported that the adhesion of C. jejuni to AISI 304 under the influence of J. communis EO was reduced by more than 90%.
Monoterpenes, α and β-pinene, sabinene, and β-myrcene, together make up at least a quarter, and sometimes more than two-thirds of the chemical composition of J. communis and H. italicum EO, with the remainder consisting of sesquiterpenes, primarily γ-curcumene and neryl acetate. However, when we tested α-pinene as an individual compound against M. avium and M. intracellulare, its MIC/MBC/MIC values were three times higher than those of the J. communis EO and twice as high as those of the H. italicum EO [27]. In an experiment with M. gordonae, α-pinene had the same MIC value as J. communis EO, however, the MBC and MIC values were two and three times higher, respectively. We could assume that the antimycobacterial activity of J. communis EO and H. italicum EO can be attributed to α-pinene, but it is more likely that it could be due to the synergistic activity of several major compounds within these EOs.
M. avium, M. intracellulare, and M. gordonae demonstrated an abundant biofilm forming ability on stainless steel in STW. In our previous study, M. avium and M. intracellulare formed biofilms on polystyrene, but the number of bacteria was lower by two logarithmic units than in this study [32]. M. avium produced larger volume biofilms than M. intracellulare, which coincides with data from research studies [4]. The highest degree of adhesion of M. avium was observed on galvanized stainless steel, followed by stainless steel, polyvinyl chloride, glass, and copper. Factors enhancing the adhesion of M. avium to the surface are the roughness and hydrophobicity of the substrate as well as the presence of zinc, calcium, and magnesium [41]. Fast-growing and saprophytic species of M. chelonae, M. fortuitum, M. gordonae, and M. tarrae were identified in 90% of the polymicrobial biofilms found in the water supply systems of households and water treatment plants. In polymicrobial biofilms, M. avium, M. intracellulare, and M. xenopi, are predominantly present on faucets and shower heads [14]. It has been observed that Methylobacterium spp., like M. avium, rapidly forms a biofilm of a characteristic pink color in water supply systems [45].
Esteban et al. [46] studied biofilm formation by unpigmented fast-growing mycobacteria on a plastic surface in three different media. Biofilm formation was monitored at room temperature in Middlebrook 7H9S, STW, and phosphate buffered saline with 5% glucose (PBS 5% GLU). All the examined/analyzed species showed a sigmoid growth curve in 7H9S and STW. In 7H9, they initially had a characteristic lacy growth pattern with a delicate reticulate structure, which was then firmly formed and covered the entire surface by the 28th day of incubation. In STW, they showed the same growth pattern, but the fully developed biofilm was formed only on day 63, while in PBS 5% GLU, they did not manage to cover the entire surface within 69 days. It has long been known that low-nutrient media reduce the amount of biofilm produced by slow-growing NTMs [45]. However, in a study by Esteban et al. [46], STW was observed to be a better biofilm development medium than PBS 5% GLU. The authors concluded that such behavior may be due to a multitude of chemicals present at low levels in STW, serving as nutrients for mycobacteria. Ambient temperature was recognized as another important factor affecting biofilm development. Incubation of cultures was performed at room temperature, naturally present in the environment [46]. Our study into the effects of temperature on the mycobacterial biofilm formation revealed a temperature of 25 • C to be the most favorable for biofilm formation. At this temperature, J. communis EO showed the weakest anti-adhesion and antibiofilm activity against NTM on polystyrene [33].
M. gordonae is a saprophytic, environmental NTM. According to our study, M. gordonae on stainless steel AISI 316 in STW, after 72 hours, produced a significantly larger volume biofilm than M. avium and M. intracellulare, which confirmed previous observations found in scientific papers of a significant presence of M. gordonae in the biofilm on metal surfaces of water supply systems [13]. NTMs in the aqueous medium showed greater sensitivity to the action of J. communis EO and H. italicum EO, than was observed in the nutritive liquid medium 7H9S [27]. The reason for this may be that nutrient-rich broth stimulates the multiplication of mycobacteria, which makes them more sensitive to the effect of EOs, or it could be due to the greater solubility of EOs in this nutrient-rich medium. In contrast, an aqueous medium slows down the multiplication of mycobacteria and promotes their resistance.
The degradation of three-day biofilms of M. avium, M. intracellulare, and M. gordonae on AISI 316 was most strongly affected by H. italicum EO at a concentration of 2 × MIC. The greatest degree of degradation of biofilm, at this concentration of H. italicum EO, was observed in M. gordonae, followed by M. avium, whereas the most resistant was the biofilm of M. intracellulare. Thus, in studies conducted with M. gordonae and M. avium, we revealed its exceptional ability to form a biofilm. However, the biofilm of M. gordonae, in contrast to the biofilm of M. avium and M. intracellulare, is more sensitive to degradation caused by H. italicum and J. communis EO activity.
Increased resistance of M. avium biofilm was observed by Carter et al. [47] who recently reported that clarithromycin could inhibit M. avium if administered before the formation of a biofilm in the respiratory system and becomes ineffective after the formation of a biofilm by this mycobacterium. Due to the altered M. avium phenotype in the biofilm, its response to antimicrobial therapy is limited, which is a key problem in the treatment of pulmonary mycobacteriosis caused by this mycobacterium [48]. The most likely explanation for the synergistic action of the EOs is that compounds from each EO have a different target site, combined with improved diffusion and distribution of each EO and components in the bacterial cell, inhibition of common biochemical pathway, inhibition of protective enzymes, and action on the specific resistance mechanism [20,49].

Conclusions
Our study showed that the tested EOs, when used at subinhibitory synergistic concentrations, had a greater effect on the degradation of mycobacterial biofilms grown on stainless steel than when they were applied individually at inhibitory concentrations. This allowed for the application of low non-toxic concentrations in biofilm eradication. Synergistic combinations of J. communis and H. italicum EOs could therefore potentially be applied in new ways to prevent the adhesion and biofilm formation of NTM, not only in the water supply system as a reservoir of NTM and a source of human infections, but also on artificial materials used in medicine or in the case of infections associated with biofilm formation.