Chemical Composition and Biological Activities of the Essential Oils of Leptospermum petersonii and Eucalyptus gunnii

The aim of this study was to characterize the chemical composition and to evaluate the antimicrobial and phytotoxic properties of the essential oils (EOs) obtained from leaves of Leptospermum petersonii chemotype “Variety B” and Eucalyptus gunnii, native to Australia. Geranyl acetate, γ-terpinene, geraniol, terpinolene, α-pinene, p-cimene, and linalool were the main components in L. petersonii EO, confirming also the existence of several chemotypes in such taxa; on the other hand, 1,8-cineole, trans-sabinene hydrate acetate, globulol, longicyclene, terpinolene, and camphene were present in major amounts in the E. gunnii EO. Chemical analysis of L. petersonii revealed that it belongs to the variety “B.” E. gunnii EO showed good antibacterial activity, with an MIC of 0.5 and 2 μg/mL against Staphylococcus aureus, and Pectobacterium carotovorum, respectively. The activity of E. gunnii EO was stronger than L. petersonii EO, whose maximum MIC reached 5 μg/mL. E. gunnii and L. petersonii EOs were particularly effective in inhibiting the biofilm formation by S. aureus, already at a concentration of 0.01 μg/mL. The other strains were resistant to both EOs up to a dose of 0.05 μg/mL. The maximum inhibition on biofilm formed by P. carotovorum was recorded for E. gunnii EO, reaching a value of 93.12% at 1.0 μg/mL. This is the first manuscript which studies the biofilm inhibition by EOs and evaluates their effects on biofilm metabolism. Both EOs were more effective against P. carotovorum. In addition, even though L. petersonii EO 0.1 μg/mL was unable to inhibit biofilm formation by Escherichia coli, it decreased the metabolic activity of the biofilm to 78.55% compared to control; furthermore, despite it inducing a relatively low inhibition (66.67%) on biofilm formation, it markedly affected metabolic activity, which decreased to 16.09% with respect to the control. On the contrary, L. petersonii EO 0.5 μg/mL induced a 79.88% inhibition of S. aureus biofilm, maintaining a high metabolic activity (90.89%) compared to the control. Moreover, this EO showed inhibitory activity against radical elongation of Solanum lycopersicum and the germination of radish. On the contrary, E. gunnii EO showed no phytotoxic activity.

INTRODUCTION chemical composition or properties of E. gunnii EO (Lucia et al., 2008;Elaissi et al., 2010;Bugarin et al., 2014). It has a weak antiinflammatory effect (Dhakad et al., 2018) and major antibacterial properties against Pseudomonas aeruginosa when compared to EOs from other species of Eucalyptus genus (Elaissi et al., 2010;Salehi et al., 2019). To date, the phytotoxic activity, as well as the anti-biofilm properties, of these two EOs against several bacterial strains such as Staphylococcus aureus, P. aeruginosa, Listeria monocytogenes, and Pectobacterium carotovorum have not been investigated.
The persistent use of synthetic herbicides may impede sustainable agricultural production, causing ecological and environmental concerns such as increased resistance of weeds, environmental pollution, and health hazards (Rolim de Almeida et al., 2010). EOs are valuable sources of lead molecules in agriculture for weed and pest management. It is thus pertinent to explore and characterize the phytotoxic properties of aromatic plants.
In addition, biofilm formation and bacterial antibiotic resistance constitute problems of particular bearing for human health. For this reason, research aimed at identifying new antimicrobial compounds from natural sources such as medicinal plants are very topical.
Biofilm formation by S. aureus and P. aeruginosa is important when studying infections of the upper airways, because it perpetuates antigen presentation, leading to chronic inflammation and complicating clinical treatments (Camporese, 2013). Similar problems are caused by the uropathogenic Escherichia coli, whose biofilms are difficult to eradicate from the surface of hospital catheters (Bernal-Mercado et al., 2019), and L. monocytogenes, which form biofilms on food products (Djordjevic et al., 2002). Finally, P. carotovorum biofilm causes soft rot in food crops due the release of exoenzymes and increases bacterial resistance during plant disinfection (Gutierrez-Pacheco et al., 2018). P. carotovorum subsp. brasiliense was reported to form biofilm-like aggregates inside xylem vessels of tomato and potato plants (Kubheka et al., 2013). From this point of view, several EOs have already been tested (Hosseini Nezhad et al., 2012). Biofilm cells profoundly differ from planktonic cells of the same species because of phenotypic and metabolic changes that regulate some cell events, such as sporulation, starvation survival, adhesion, rough-smooth phase variations, etc. This also led to differences in the susceptibility of planktonic and biofilm cells to antimicrobial agents because adhesion-dependent phenotypic changes may be important as diffusion barriers or as growth rate-dependent changes in this important phenomenon. The metabolic changes can also make these cells inherently resistant to a particular antibiotic agent, virtually developed against specific targets in planktonic cells.
The present study was carried out to characterize the micromorphological features of L. petersonii and E. gunnii leaves by scanning electron microscopy (SEM) and to investigate the chemical composition of their EOs, as well as their antimicrobial and phytotoxic activities. The antibacterial activity was evaluated against some Gram-positive (S. aureus and L. monocytogenes) and Gram-negative (E. coli and P. aeruginosa) pathogenic strains and against the phytopathogen, P. carotovorum. The potential in vitro phytotoxicity was evaluated against germination and radical elongation of Raphanus sativus L.

Plant Material
Branches with leaves of L. petersonii, belonging to the chemical variety "B CT Australian Rose, " were obtained from plantations situated in the Byron Bay hinterland, Northern New South Wales, Australia. Branches bearing young rounded leaves of E. gunnii were collected from pruning material of trees growing on a private land in Sardinia (Italy). Taxonomic identification of plant samples was carried out by MV for E. gunnii and GT for L. petersonii.

Chemicals and Reagents
Ethanol-FineFix working solution was obtained from Milestone s.r.l., Bergamo, Italy. The bacterial culture medium, PBS, DMSO, tetracycline, ciprofloxacin, and MTT were supplied by Sigma, Milano, Italy. All other reagents were of analytical grade.

Scanning Electron Microscopy
Samples of leaves (1-1.5 cm 2 ) were fixed in a 70% ethanol-FineFix working solution (Milestone s.r.l., Bergamo, Italy) for 24 h at 4 • C, and then gradually dehydrated by ethanol series (Chieco et al., 2012). Subsequently, samples were processed in a critical point dryer (CPD processor, K850 2M Strumenti s.r.l., Rome, Italy), mounted on aluminum stubs using glued carbon tabs, and sputter-coated with gold. Specimens were analyzed and photographed using an FE-SEM (Supra VP-40; Zeiss, Oberkochen, Germany) at an accelerating voltage of 20 kV.

Isolation of Essential Oils
Fresh plant materials were subjected to steam distillation until no significant increase in the volume of the collected EO was observed (3 h).
The EO yield (w/v, %) was calculated according to the following equation: where W 0 is the plant material weight distillated and V EO is the EO volume obtained.
Essential oils were dried on Na 2 SO 4 and stored in a darksealed vial with nitrogen headspace until analysis.

GC-FID Analysis
Analytical GC was performed on a Perkin-Elmer Sigma-115 gas chromatograph equipped with an FID and a data handling processor. The separation was achieved using a HP-5 MS fused-silica capillary column (30 m × 0.25 mm i.d., 0.25 µm film thickness). Column temperature: 40 • C, with 5 min initial hold, and then to 270 • C at 2 • C/min, 270 • C (20 min); injection mode splitless (1 µL of a 1:1000 n-hexane solution). Injector and detector temperatures were 250 and 290 • C, respectively. Analysis was also run by using a fused silica HP Innowax polyethylene glycol capillary column (50 m × 0.20 mm i.d., 0.25 µm film thickness). In both cases, helium was used as a carrier gas (1.0 mL/min).

GC/MS Analysis and Identification of Single Constituents
Analyses were performed on an Agilent 6850 Ser. II apparatus, fitted with a fused silica DB-5 capillary column (30 m × 0.25 mm i.d., 0.33 µm film thickness), coupled to an Agilent Mass Selective Detector MSD 5973, ionization energy voltage 70 eV, electron multiplier voltage energy 2000 V. Mass spectra were acquired in the range 40-500 amu, scan time 5 scans/s. Gas chromatographic conditions were as reported above, with a transfer line temperature of 295 • C. Most constituents were identified by comparison of their Kovats retention indices (Ri) [calculated in relation to retention time of n-alkanes (C 10 -C 35 )], with either those of the literature (Jennings and Shibamoto, 1980;Davies, 1990;Adams, 2007;Goodner, 2008), by accurate analysis of mass spectra on both columns and by their comparison with those of authentic compounds available in our laboratories by means of NIST 02 and Wiley 275 libraries (Wiley, 1998). The components' relative concentrations were obtained by peak area normalization.

Determination of the Antibacterial Susceptibility by Agar Diffusion Assay
Different concentrations of Eos, ranging from 0.01 to 2.0 µg/mL, diluted in sterile DMSO (Sigma-Aldrich Italy, Milan, Italy) were spotted onto the plates previously inoculated with 0.5 McFarland (1.5 × 10 7 cells/mL) bacterial suspension (Densitometer cell density turbidity 0.3-15.0 McFarland, CAMLAB, Cambridge, United Kingdom) (Fratianni et al., 2016). After 10 min in sterile conditions, the plates were incubated for 24 h at 37 or 28 • C, depending on the strain. The diameter of the clear zone shown on plates (inhibition zone) was accurately measured by using an Extra steel Caliper mod 0289, mm/inch reading scale, precision 0.05 mm (Mario De Maio, Milan, Italy). Sterile 1% DMSO and tetracycline (7 µg/mL, Sigma-Aldrich Italy, Milan, Italy) were used as negative and positive controls, respectively. The experiments were performed in triplicate and results expressed as mean ± SD.

Minimal Inhibitory Concentration (MIC)
The MIC values were calculated through the application of the resazurin microtiter-plate assay (Sarker et al., 2007). The EOs were dissolved in sterile DMSO. Twofold serial dilutions were prepared to obtain 50 µL of the EOs in serially descending concentrations in each well. Thirty-five µL of 3.3 × strength iso-sensitized broth and 5 µL of resazurin, used as indicator solution, were added to reach a final volume/well of 240 µL with several volumes of sterile Muller-Hinton broth (Sigma-Aldrich, Milan, Italy) previously set. Finally, 10 µL of bacterial suspension was added to each well to reach a concentration of about 5 × 10 5 cfu/mL. Sterile DMSO and ciprofloxacin (Sigma-Aldrich Italy, Milan, Italy, 1 mg/mL in DMSO) were used as negative and positive controls, respectively. Multiwell plates were prepared in triplicate and incubated at 37 • C for 24 h. The lowest concentration at which a color change occurred (from dark purple to colorless) revealed the MIC value.

Biofilm Inhibitory Activity
The effect of the EOs on bacterial adhesion ability was assessed in flat-bottomed 96-well microtiter plates according to the method of O'Toole and Kolter (1998), using EO concentrations ranging from 0.01 to 2.0 µg/mL (corresponding to volumes ranging from 1 to 20 µL, respectively).
In each well, the overnight bacterial cultures were adjusted to 0.5 McFarland (1.5 × 10 7 cells/mL, Densitometer cell density Kingdom) with fresh culture broth. Then, 10 µL of the diluted cultures was distributed in each well, and different volumes of the extracts and Muller-Hinton broth were added, to reach a final volume of 250 µL/well. Microplates were completely covered with parafilm, to avoid the evaporation of samples with relative loss of volume and incubated for 48 h at different temperatures (depending on the strain). Planktonic cells were removed and the attached cells were gently twice washed with sterile physiological saline. After that, 200 µL of methanol was added to each well, retaining it for 15 min to fix the sessile cells. Methanol was then discarded, and each plate was left until complete dryness of samples. The staining of the adhered cells was obtained by adding 200 µL of 2% w/v crystal violet solution to each well that was left for 20 min. Wells were gently washed with sterile physiological solution and left to dry. Two hundred microliters of glacial acetic acid 20% w/v were added to allow the release of the bound dye. The absorbance was measured at OD = 540 nm (Varian Cary Spectrophotometer model 50 MPR, Cernusco sul Naviglio, Italy). The percent value of biofilm inhibition was calculated with respect to control (cells grown without the presence of the EOs). Triplicate tests were done, and the average results were taken for reproducibility.

Metabolic Activity of Biofilm Cells
The effect of different concentrations of Eos, ranging from 0.01 to 2.0 µg/mL on the metabolic activity of biofilm cells, was evaluated through the MTT colorimetric method according to Kairo et al. (1999) and Fratianni et al. (2019), using 96-well microtiter plates.
The overnight bacterial cultures were adjusted to 0.5 McFarland and treated as described in Section "Biofilm Inhibitory Activity." After 48 h incubation, bacterial suspension was removed and 150 µL of PBS and 30 µL of 0.3% MTT (Sigma, Milan, Italy) were added, keeping microplates at 37 • C. After 2 h, the MTT solution was removed, two washing steps were performed with

Phytotoxic Activity
The These seeds are usually used in phytotoxicity assays because they easily germinate and are well known from a histological point of view. Radish, lettuce, garden cress, and tomato seeds were purchased from the Blumen Group s.r.l. (Emilia Romagna, Italy), Italian ryegrass seeds were purchased from Fratelli Ingegnoli Spa (Milano, Italy), and purslane seeds from W. Legutko s.r.l. (Jutrosin, Poland). The seeds were surface sterilized in 95% ethanol for 15 s and sown in Petri dishes (Ø = 90 mm), containing three layers of Whatman filter paper, and impregnated with 7 mL of distilled water (control) or 7 mL of the tested solution of EO. The germination conditions were 20 ± 1 • C, with a natural photoperiod. The EOs, dissolved into water-acetone mixture (99.5:0.5), were assayed at the doses of 100, 10, 1, and 0.1 µg/mL. Controls performed with water-acetone mixture alone showed no differences in comparison to controls in water alone. Seed germination was observed directly in Petri dishes every 24 h. A seed was considered germinated when the protrusion of the root became evident (Bewley, 1997). After 120 h (on the fifth day), the effects on radicle elongation were measured in cm. Each determination was repeated three times, using Petri dishes containing 10 seeds each. Data were expressed as mean ± SD for both germination and radical elongation.

Statistical Analysis
All experiments were carried out in triplicate. Data from each experiment were statistically analyzed using GraphPad Prism 6.0 software (GraphPad Software Inc., San Diego, CA, United States) followed by comparison of means (two-way ANOVA) using Dunnett's multiple comparisons test, at the significance level of p < 0.05.

Micromorphological Characterization
The leaf anatomical features of L. petersonii and E. gunnii have been scarcely investigated, although leaves are the main section of the plant from which EOs are extracted. Concerning E. gunnii, most recent studies have been focused on the epicuticular waxes of the leaf surface, considering their morphology, composition, function, and biosynthesis, and also regarding the modulatory effects of different stress factors (Shepherd and Griffiths, 2006) and regeneration processes after removal (Huth et al., 2018). A brief anatomical description of L. petersonii leaf anatomy has been reported by Johnson (1980) in his comprehensive revision of the genus Leptospermum, taking into account 40 different species.
Our FE-SEM investigation has highlighted many typical xeromorphic adaptations against water loss, such as sunkenstomata, thick cuticles, and a waxy epidermis in the leaves of L. petersonii (Figures 1A,B), and even more distinctly in those of E. gunnii (Figures 2A,B). The cuticle on the leaf epidermal surfaces of E. gunnii is smooth or slightly striated and shows cuticular papillae and epicuticular waxes shaped as small granules ( Figure 2B). These features appear similar to those reported by Migacz et al. (2018) for Eucalyptus dunni Maiden. The mesophyll was dorsiventral in L. petersonii (Figures 1C,D), whereas it appeared isobilateral in E. gunnii (Figures 2C,D). The oil glands of both species have previously been found to develop by schizogony of cell walls (Carr and Carr, 1970;Johnson, 1980). In our observations, these schizogenous cavities appeared scattered in the mesophyll, especially in the sub-epidermal region, on both leaf sides (Figures 1, 2C,D). In transversal sections of E. gunnii leaves, many prismatic crystals and druses of calcium oxalate were also found, mainly around the ribs and near the oil glands (Figures 2C,D, arrows). On the contrary, prismatic crystals were absent on the leaf epidermal surface, being one of the diagnostic features useful to differentiate Eucalyptus species (Migacz et al., 2018).
The leaf morphological features detected in the examined species can be useful for their identification or in the quality control of herbal material used for EO extraction, to avoid accidental or intentional adulteration.

Essential Oil Yields
Steam distillation of the chemical variety "B CT Australian Rose" of L. petersonii and E. gunnii branchlets and leaves furnished EOs in 0.5 and 2% yields, respectively. The yield of our sample of L. petersonii was higher than other species of genus Leptospermum. In fact, Windsor and Brooks (2012) in their study observed a yield of 0.01-0.06% for Leptospermum laevigatum, 0.06-0.2% Leptospermum speciosum, and 0.2-0.4% for Leptospermum trinervium. Van Vuuren et al. (2014) demonstrated that for L. petersonii annual EO yields ranged from 0.4% (in February) and 1.8% (in March). Moreover, Demuner et al. (2011) showed that flower L. petersonii (0.5%) EO was lower than that found for the leaves (3.7%) in the same period.
Also, for genus Eucalyptus, the yields varied according to the species. In fact, Elaissi et al. (2010) analyzed 13 species from Tunisia and demonstrated yields ranging from 0.5%, for E. gunnii to 3.9% for Eucalyptus cinerea. Moreover, comparing different Eucalyptus species, Lucia et al. (2008) showed a significant relationship between total EO yields and 1.8cineole concentration.
This shows that there are different factors that can lead to quantitative and qualitative differences in the EOs such as soil, season, temperature, geographic origin, and ecological role of the plant organs that produce EOs (Demuner et al., 2011).
The composition of the EO of L. petersonii chemical variety "B CT Australian Rose" of this study was partially in agreement with previous data. In fact, structural analogs of geraniol, such as geranial (34.7-29.9%) and neral (19.7-23.5%), have been reported as the main constituents of L. petersonii EO. On the other hand, constituents reported in high amounts in other studies, such as citronellal (33.9-11.4%) or citronellol (17.5%), were absent in our sample (Demuner et al., 2011;Kim and Park, 2012;Van Vuuren et al., 2014).
The different composition of the EOs confirmed the existence of several chemotypes in L. petersonii, as reported by Brophy et al. (2000). The most common chemical variety with a pleasant odor is the "A" type, which consists mainly of aldehydes such as neral, geranial, citronellal, and monoterpene hydrocarbons (γ-terpinene, α-pinene, p-cymene). On the contrary, the chemical variety "B, " with a rose-like odor, is quite rare and contains geraniol and geranyl acetate as main constituents, followed by γ-terpinene and terpinolene (Brophy et al., 2000).
There have been no large studies on the rare chemical variety "B": the sample analyzed in this study was evidently derived from L. petersonii chemical variety "B." In fact, our results agree with Brophy et al. (2000), who reported the chemical composition of three EOs belonging to this variety, with geranyl acetate (ranging from 21 to 38%) and geraniol (ranging from 21 to 29%) as the main components.
The chemical composition of the analyzed E. gunnii EO is quite similar to data reported in literature. 1,8 Cineole was confirmed as the main component, as described in previous studies, with a percentage ranging from 17.9 to 67.8% (Lucia et al., 2008;Bugarin et al., 2014). Conversely, spathulenol (16.5%) was reported as the major component of E. gunnii EO from Tunisia (Elaissi et al., 2011) and it was identified also in E. gunnii EO from Argentina with a percentage of 12.3% (Lucia et al., 2008). This compound was present in a very low amount (0.6%) in our EO. Moreover, viridiflorol and globulol were present in a high percentage in Tunisian EO (11.5 and 12.5%, respectively) (Elaissi et al., 2011). These compounds were also present in our sample, in particular globulolin in a similar quantity (10.3%), while viridiflorol was present in a lower amount (0.6%).

Antibacterial Activity
The antibacterial activity of the EOs was evaluated against Grampositive and Gram-negative pathogenic strains, through the    inhibition zone test and the determination of the MIC. Results are shown in Figure 3 and in Table 3, respectively. The behavior of the two EOs against Gram-positive (S. aureus and L. monocytogenes) and Gram-negative (E. coli, P. aeruginosa, and P. carotovorum) strains was completely different. E. gunnii EO showed the best antibacterial activity against Gram-positive strains. Indeed, at highest concentration (10 µg/mL), the inhibition zone vs S. aureus and L. monocytegenes was 2 and 2.5 times greater than those obtained using the same concentration of L. petersonii EO (Figure 3). Moreover, the highest concentration of E. gunnii EO was much more effective than the reference antibiotic against S. aureus and even more so against L. monocytogenes.
The different behavior showed by the two EOs at the highest concentration could be ascribable to their different phytochemical composition. The presence of 1.8-cineole (33%) and trans-sabinene hydrate-acetate (15%) in E. gunnii EO probably contributes to the strong antibacterial activity of this EO, according to Nazzaro et al. (2013). Indeed, 1.8-cineole is a well-known antibacterial agent with proven activity against all the microorganisms used in our experiments (Caputo et al., 2017). Furthermore, γ-terpinene and trans-sabinene hydrateacetate could enhance the E. gunnii EO ability to block the bacterial growth (Nazzaro et al., 2013), probably due to a synergistic mechanism (Dorman and Deans, 2000). Finally, globulol (10.3%), which was identified in other Eucalyptus EOs, also possesses antibacterial activity (Sebei et al., 2015).
On the contrary, there is no positive correlation between the high amount of geranyl-acetate in L. petersonii EO and antibacterial efficacy vs L. monocytogenes (Lis-Balchin and Deans, 1997).
Considering the Gram-negative microorganisms tested, the results obtained would allow differentiating two subgroups for resistance/sensitivity to the L. petersonii and E. gunnii EOs, a first sub-group represented by E. coli and P. aeruginosa, and a second one by P. carotovorum. E. gunnii EO showed good activity against E. coli, although it was weaker than that exerted on other E. coli strains (Bugarin et al., 2014). It is well known that the resistance/sensitivity to an antimicrobial agent may vary among different strains of the same species (Del Monte et al., 2015). However, we should stress that the inhibition zone induced by E. gunnii EO was almost double than tetracycline ( Figure 4A). E. gunnii EO also showed strong activity against P. aeruginosa, with an inhibition diameter twice with respect to tetracycline ( Figure 4B). Thus, this EO can represent a product with a wide potential against pathogenic species such as E. coli, P. aeruginosa, L. monocytogenes, and S. aureus, resulting in particular interest due to the increased number of microbial species showing resistance to antibiotic drugs (Heras et al., 2015;Hartmann et al., 2019).
Leptospermum petersonii EO was usually less effective in comparison with E. gunnii EO, as indicated by the MIC values (Table 3). Moreover, despite exercising an inhibitory activity against E coli at all the concentrations used, its efficacy (width of the inhibition zone) was always less than tetracycline ( Figure 4A). On the contrary, it was more effective vs Pseudomonas, with inhibition zones of about 20 mm, practically double with respect to tetracycline. Indeed, in this case, the inhibiting force was similar to that exerted by the E. gunnii EO (Figure 4B).The effectiveness shown by L. petersonii EO against P. aeruginosa and S. aureus (Figures 3, 4B) is in accordance with Demuner et al. (2011), which demonstrated a strong efficacy of Leptospermum spp. EOs against these bacteria. The composition of L. petersonii EO from Brazil and South Africa, with considerable amounts of citronellal, geranial, and neral, may confer forceful antimicrobial activity (Van Vuuren et al., 2014).
An MIC test confirmed the more powerful activity of E. gunnii EO (Table 3), with MIC values which ranged between 0.5 and 2.0 µg/mL (against S. aureus and P. carotovorum, respectively). The activity of E. gunnii EO was generally stronger than L. petersonii EO, whose maximum MIC value reached even 5 µg/mL. MIC values of L. petersonii EO were usually higher against L. monocytogenes ( Table 3).
Pectobacterium (formerly Erwinia) carotovorum is a Gramnegative bacterium belonging to the Enterobacteriaceae family, which affects several crops such as potato, pineapple, maize, and Results are reported as percent of inhibition respect to the control (inhibition = 0). Results are the mean of three experiments ± SD. EC: E. coli; LM: L. monocytogenes; PC: P. carotovorum; PSA: P. aeruginosa; SA: S. aureus; EUC: E. gunnii EO; LEPT: L. petersonii EO. African violet (Mehrsorosh et al., 2014). It causes soft rot and blackleg of potatoes and vegetables, as well as slime flux on many different tree species. The soft rot erwinias can be present on plant surfaces and in soil where they may penetrate within the plant via wound sites or through natural openings on the plant surface, e.g., lenticels.
In the antibacterial tests, we observed completely opposed behavior of the two EOs investigated. E. gunnii EO resulted active already at 2 µg/mL (Table 3), causing an inhibition zone of 7 mm (Figure 5). The highest concentration (10 µg/mL) showed an inhibition zone of 17 mm, almost double with respect to the diameter obtained with tetracycline ( Figure 5).
To our knowledge, this is the first time that the antibacterial activity of E. gunnii EO against P. carotovorum has been evaluated. Such activity was stronger than those of EOs obtained from other species belonging to the Eucalyptus genus, such as Eucalyptus caesia Benth., Eucalyptus camaldulensis subsp. obtusa (Blakely Brooker and M. W. McDonald), and Eucalyptus gomphocephala A. Cunn. ex DC., and much more effective with respect to some Lamiaceae EOs (Mehrsorosh et al., 2014;Salem et al., 2015).
This high antibacterial activity against the Gram negative phytopathogen P. carotovorum makes this EO an ideal candidate to treat and prevent bacterial infections in crops; in addition, through the use of new species-specific technologies, such as the encapsulation of EO in mesoporous silica nanoparticles, this EO has high efficacy when applied to this and other phytopathogens (Cadena et al., 2018).

Biofilm and Metabolic Activity of Biofilm Cells
Eucalyptus gunnii and L. petersonii EOs were tested on the above bacterial strains for inhibition of biofilm formation and metabolic activity. In our study, for the first time, biofilm inhibition by these two EOs (Table 4) was integrated with EO effects on biofilm metabolism (Table 5), thus increasing available information on their antibacterial power and potential applications.
The activity of E. gunnii EO against the uropathogen E. coli was of particular interest. Indeed, at a concentration of only 0.5 µg/mL, this EO causes a biofilm formation inhibition of 80.28%. These results were in accordance with other previous studies (Karpanen et al., 2008;Hendry et al., 2009), which have ascertained the antimicrobial activity of Eucalyptus EO vs E. coli and Staphylococcus epidermidis biofilm formation.
Moreover, E. gunnii EO decreased the metabolic activity of the E. coli biofilm cells to 55.6% with respect to the control, as observed by MTT assay ( Table 5). This means that E. gunnii EO was capable not only of strongly inhibiting the E. coli biofilm formation but also to limiting all metabolic changes, which make cells more resistant to the antibacterial agents. L. petersonii EO, although less strong than the previous one, still shows inhibitory properties against E. coli biofilm formation (96.22%) and decreases the metabolic activity (48.24%) at 1.0 µg/mL (Tables 4, 5, respectively). The activity of E. gunnii and L. petersoni EOs against E. coli could be particularly important in clinical settings and patient management.
When EOs were tested to evaluate their ability to block L. monocytogenes biofilm formation, their behavior was different and L. petersonii EO showed, in general, the best antibacterial activity. Indeed, it inhibits the L. monocytogenes biofilm formation at 0.1 µg/mL, with an inhibition rate reaching 78.41% at 2 µg/mL. Moreover, it was able to influence the biofilm cells metabolism by decreasing their activity to 44.30% compared to the control ( Table 5). On the contrary, E. gunnii EO was completely inactive until 0.1 µg/mL, showing an inhibition of biofilm formation of 4.8 and 50.21% at 0.2 and 0.5 µg/mL, respectively ( Table 4). The latter concentration (0.5 µg/mL) showed a decrease of biofilm cells metabolism to 60.35% with respect to control ( Table 5).
Essential oils investigated showed a weak inhibitory activity regarding both biofilm growth (35.43 and 20.15% by E. gunnii and L. petersoni EOs, respectively, Table 4) and metabolic activity of P. aeruginosa (77.74 and 83.55% by E. gunnii and L. petersoni EOs, respectively, Table 5).
This confirmed the weak antibacterial activity of Eucalyptus species against biofilm formation, as previously reported for Eucalyptus globulus Labill. EO (Sambyal et al., 2017), although its action could be enhanced through innovative biotechnological processes. For instance, nanoemulsion FIGURE 6 | Phytotoxic activity of L. petersonii EO against germination of R. sativus (A) and radical elongation of S. lycopersicum (B), 120 h after sowing. Results are the mean of three experiments ± SD. **p < 0.01; ***p < 0.001 compared to control (ANOVA followed by Dunnett's multiple comparison test).
containing E. globulus EO showed antimicrobial and anti-biofilm activities against some Gram-negative bacteria, commonly found in immunosuppressed patients, such as P. aeruginosa and C. albicans (Quatrin et al., 2017).
The two EOs were particularly effective against S. aureus (Table 4), inhibiting biofilm formation by 10.37 and 14.78%, respectively, at a concentration of 0.01 µg/mL. The activity of E. gunnii EO on S. aureus was stronger than L. petersonii EO at a concentration above 0.01 µg/mL ( Table 4). The inhibitory effect of E. gunnii EO against S. aureus confirmed the capability of the EOs recovered by species belonging to the Eucalyptus genus to block the biofilm formation by this methicillin resistant strain (Junka et al., 2019); its activity was more powerful than E. globulus EO (Merghni et al., 2018). The effectiveness shown by both EOs could be taken into consideration, once again, in patient management, mainly to avoid S. aureus infections that concern the central venous, dialysis (Gahlot et al., 2014), and urinary catheters (Murugan et al., 2016), as well as arteriovenous shunts (MacRae et al., 2016) and sutures (Henry-Stanley et al., 2010). Interestingly, L. petersonii EO at 0.5 µg/mL induced a strong inhibition of S. aureus biofilm formation (79.88%), while maintaining a high metabolic activity (90.89%) compared to the control.
Eucalyptus gunnii EO was particularly active in inhibiting the biofilm formation of the Gram-negative phytopathogen P. carotovorum. Indeed, at 1 µg/mL, it was capable of avoiding almost completely the biofilm formation (93.12%, Table 4) and to strongly affect the metabolic activity of the few biofilm cells (22.57% respect to control, Table 5). On the contrary, L. petersoni EO at a concentration of 2.0 µg/mL induced a relatively lower inhibition (66.67%, Table 4) on biofilm formation, while also markedly affecting metabolic activity (16.09% with respect to control) ( Table 5)

Phytotoxic Activity
A delay in germination or any other adverse effect on plants caused by specific substances is defined as phytotoxicity (Baumgarten and Spiegel, 2004). Moreover, secondary metabolites produced by plants, micro-organisms, viruses, and fungi, can alter the growth of the target species, with both positive and negative effects, through a phenomenon called "allelopathy" (Zeng et al., 2010).
Parameters to analyze the effects of natural or chemical substances on the growing of selected vegetal species are: relative or absolute germination and relative root elongation (Barral and Paradelo, 2011).
In this study, the two EOs were evaluated for their activity against germination and radical elongation of radish, lettuce, garden cress, tomato, rye grass, and purslane. L. petersonii EO showed inhibitory activity against the germination of R. sativus ( Figure 6A) and radical elongation of S. lycopersicum ( Figure  6B). The treatment of seeds with a concentration of 100 µg/mL inhibited germination of R. sativus; all doses tested seemed to be active against radical elongation of S. lycopersicum. E. gunnii EO showed no phytotoxic activity on the tested seed (data not shown). Its principal component, 1,8 cineole, was inactive against several of the tested seeds, as shown in our previous studies (Caputo et al., 2018).
Few previous studies have reported data about the phytotoxicity and/or allelopathy of extracts and or EOs of plants belonging to the Myrtaceae family. Imatomi et al. (2013) studied the phytotoxic potential of leaf extracts of Myrcia tomentosa Glaz. and showed that ethyl acetate extract was the most active on the roots of lettuce, tomato, and onion, as well as on the shoots of lettuce and tomato. Instead, ethyl acetate extract of leaves of Blepharocalyx salicifolius (Kunth) had a high phytotoxic activity against Triticum aestivum L., Sesamum indicum L., Echinochloa crusgalli L. Beauv., and Euphorbia heterophylla L. (Habermann et al., 2015(Habermann et al., , 2016. Methanol extract of Eugenia flavescens DC leaves showed significant seed germination inhibition of Mimosa pudica and Senna obtusifolia (Cantanhede Filho et al., 2017). Moreover, Callistemon viminalis EO showed dose-dependent allelopathic activity on L. sativa seeds (de Oliveira et al., 2014). Myrtus communis L. EO inhibited seed germinations and seedling growths of Amaranthus retroflexus L., Chenopodium album L., Cirsium arvense (L.) SCOP., Lactuca serriola L., and Rumex crispus L. (Kordali et al., 2016).
In this study, we evaluated for the first time the phytotoxicity of L. petersonii EO; in fact, no previous studies have reported similar data about this or other plants from the genus Leptospermum.
On the contrary, the phytotoxic and allelopathic activities of several Eucalyptus species are well known in both natural and modified ecosystems (Chu et al., 2014) and some EOs from Eucalyptus spp. have been proposed for use in agricultural and forestry management (Ramezani et al., 2008). Despite the fact that the Eucalyptus genus is probably among the most investigated for its allelopathic properties, no studies are currently available on the activity of E. gunnii EO on seeds growing.

CONCLUSION
This is the first study that provides a detailed SEM analysis of the leaf micromorphology of L. petersonii and E. gunnii. The qualiquantitative analyses showed a superimposable phytochemical profile for E. gunnii EO with respect to other previously published data. In addition, it highlighted the phytochemical composition of the EO of a quite rare L. petersonii chemical variety "B CT Australian Rose, " with a rose-like odor and containing geraniol and geranyl acetate as main constituents. This is also the first time that the effects of these EOs on the biofilm formation biofilm cells metabolic activity were investigated.
This could open new perspectives for the application of L. petersonii and E. gunnii EOs.
Despite both EOs showing a good antimicrobial activity, the E. gunnii EO was the strongest one. This is probably due to the high content of 1,8-cineole, although a synergism with γ-terpinene and trans-sabinene hydrate-acetate could be postulated in light of previous literature data about the antimicrobial activity of the pure compound 1,8-cineole against the same pathogens.
In addition, both EOs were particularly effective against S. aureus biofilm formation, with E. gunnii EO showing the strongest activity not only with respect to L. petersonii EO, but also compared to other EOs isolated from other Eucalyptus species.
On the contrary, L. petersonii chemical variety "B CT Australian Rose" EO showed inhibitory activity against germination and radical elongation of R. sativus and S. lycopersicum, respectively, while no phytotoxic activity was detected for E. gunnii EO.
Considering these results, further investigations on these EOs could provide useful applications for both the treatment of human microbial infections and for agronomic management.

DATA AVAILABILITY STATEMENT
All datasets generated for this study are included in the article.

AUTHOR CONTRIBUTIONS
VD, FN, LaC, and DT designed the project. MV and GT distilled the EOs. FN and FF performed the antibacterial assays. LuC studied the phytotoxic activity and performed the data analysis. LaC carried out the micromorphological characterization. AS and LuC performed the chemical characterization. All authors contributed to writing and revising the manuscript.