Antimicrobial Activity of Five Essential Oils against Bacteria and Fungi Responsible for Urinary Tract Infections

Urinary tract infections are frequently encountered in small animal practice. Escherichia coli and Enterococcus spp. are the most common agents associated to these infections, even though other bacteria and yeasts, such as Candida albicans and Candida famata, may be involved. In view of the increasing problem of the multi-drug resistance, the aim of this study was to investigate the antimicrobial activity of essential oils obtained from star anise (Illicium verum Hook.f.), basil (Ocimum basilicum L.), origanum (Origanum vulgare L.), clary sage (Salvia sclarea L.) and thymus (Thymus vulgaris L.) against multidrug-resistant strains of Escherichia coli, Enterococcus spp., Candida albicans and Candida famata previously isolated from dogs and cats with urinary tract infections. Enterococci were resistant to Illicium verum and Salvia sclarea, such as Candida to Salvia sclarea. Thymus vulgaris and Origanum vulgare essential oils showed the best activity against all the tested pathogens, so they could be proposed for the formulation of external and/or intravesical washes in small animals.


Introduction
Infections of the urinary tract are frequent and sometimes can induce severe threat both in human and veterinary medicine, mostly affecting dogs and cats. Urinary tract infections (UTI) may be localized to the upper tract (kidney and adjacent ureter) or the lower tract (bladder and adjacent urethra) and more than one organ is often involved [1]. These infections are usually caused by bacteria, mainly those of the intestinal microflora. Escherichia coli and Enterococcus spp. are the most frequent agents encountered in UTI cases [2][3][4]. Even though infections by haematogenous route are possible, bacteria usually colonize the genito-urinary tract by ascendant route. In view of the anatomic structure, females are more prone to UTI than males [1].
Yeasts belonging to Candida genus, in particular Candida albicans, are reported as responsible for fungal infections of the lower urinary tract, both in dogs and cats [5,6]. The organism is a commensal of digestive and genito-urinary tract of both healthy people and animals [7]. It can also act as opportunistic pathogen of skin and mucosae both in receptive animals and in immunocompromised patients [8]. Conversely, other Candida species are ubiquitous yeasts and can provoke UTI, when carried by the

Essential oils Analysis
The chemical composition of the tested EOs is reported in Table 1. All the five oils were rich in monoterpenes. In detail, the main terpenes identified in O. vulgare and T. vulgaris EOs were carvacrol (65.9%) and thymol (52.6%), respectively, followed by p-cymene (15.3%) only in T. vulgaris. The main compounds of S. sclarea and O. basilicum were linalyl acetate (54.7%) and linalool (46.0%), respectively. I. verum EO was mainly composed by the phenylpropanoid anethol (89.8%).

Agar Disc Diffusion Method
The results of the agar disc diffusion method testing E. coli and Enterococcus isolates against 21 antibiotics are summarized in Table 2.
All the strains resulted multi-resistant, even though with different resistance patterns. High percentages of no-sensitive (resistant or intermediate) strains against several antibiotics were detected, both among the tested E. coli and Enterococcus spp. isolates (Table 3).

Antimycotic Activity
Selected yeasts showed different patterns of resistance to conventional antimycotic drugs. In detail, all C. albicans isolates were resistant to voriconazole and to itraconazole, 9/12 to fluconazole, while MICs yielded from C. famata were high in 3 cases out of 4 for both fluconazole and itraconazole, and in 1/4 case for voriconazole. Caspofungin resulted active for all yeasts isolates.
Selected EOs showed different efficacy against tested yeasts. T. vulgaris EO yielded the lowest overall MICs and was effective versus all C. famata and versus 11/12 C. albicans isolates with MICs lower than 1 mg/mL. O. vulgare and O. basilicum appeared less active, even if the lowest MIC among all tested EOs (0.01%) was obtained by O. vulgare versus a strain of C. albicans. Conversely, S. sclarea was ineffective versus all examined fungi, at highest concentration tested. I. verum showed the widest range of activities, resulting completely not effective (>19.52 mg/mL) against four C. albicans isolates and showing a MIC of 0.19 mg/mL versus two C. famata and one C. albicans. Results in detail of MICs expressed both as percentage and as mg/mL are reported in Table 5. No apparent relationship among the profiles of resistance to conventional drugs and MIC values of selected EOs was observed.

Discussion
The present investigation reports the activity of selected EOs, against multidrug-resistant both bacterial and fungal organisms causing UTI in pet carnivores. At the best of our knowledge it is the first study that refers about drug-resistant veterinary isolates of E. coli, Enterococcus spp., C. albicans and C. famata.
The selected E. coli and Enterococcus spp. strains, previously isolated from dogs and cats with severe cases of UTI, resulted not sensitive (resistant and intermediate) to several antibiotics. In detail, one E. coli isolate was not sensitive to any tested antibiotic and some strains were sensitive only to one or two out of the twenty-one tested antibiotics.
The evaluation of the antibacterial activity of the selected EOs showed promising results, especially with O. vulgare and T. vulgaris. These EOs, in fact, resulted effective against all the tested isolates showing MIC values ranging from 0.15% (v/v) (0.293 mg/mL) to 0.6% (v/v) (1.183 mg/mL) for O. vulgare, and from 0.07% (v/v) (0.146 mg/mL) to 1.25% (v/v) (2.342 mg/mL) for T. vulgaris. These results are corroborated by previous investigations that found relevant antimicrobial properties of oregano and thyme EOs related to their main components carvacrol and thymol, but also other minor constituents such as the monoterpene hydrocarbons γ-terpinene and p-cymene [18]. I. verum and S. sclarea resulted moderately effective against E. coli strains, whereas no activity was observed when they were assayed versus Enterococcus spp. isolates. This different activity could be related to the dissimilar structure of Gram-positive and Gram-negative bacteria cell wall, as also supposed by Benmalek et al. [19] who found similar results when tested I. verum EO against E. coli and Staphylococcus aureus. Activity of I. verum against enterococci is scantly investigated; however, Hawrelak et al. [20] found a very weak effectiveness of this EO against Enterococcus faecalis. About S. sclarea, our results are comparable to those obtained by Frydrysiak et al. [21] who found weak antibacterial activity of Salvia officinalis and Salvia lavandulaefolia, whereas they are totally in disagreement with other studies in which strong activity of S. officinalis EO was observed against both Gram-negative bacteria, such as E. coli, and Gram-positive bacteria including enterococci [22]. This difference could be related to the original plant species and/or to variability among the bacterial strains.
Basil EO showed a good antimicrobial activity, with MIC values ranging from 0.15% (v/v) (0.285 mg/mL) to 1.25% (v/v) (2.287 mg/mL), against all the selected E. coli strains, except for one isolate (n.986) that showed no-sensitivity to 19 antibiotics among the 21 tested. Very weak activity was observed against Enterococcus spp. isolates. Antibacterial activity of O. basilicum EO was demonstrated in other studies that assayed it against multi-drug resistant clinical isolates including E. coli [23] and Enterococcus [24]. Interestingly, O. basilicum EO chemical composition reported by Sienkiewicz et al. [23] differed considerably from the EO used in the present study being estragole the main represented compound (86.4%). Antimicrobial activity of O. basilicum EO is considered mainly related to eugenol that in our study is present in a moderate amount (11.5%); this could explain the lower MIC values found with respect to the other EOs.
Even though some among the tested oils did not show a very strong antimicrobial activity, the comparison to the standard drugs could be significant in some particular cases. For instance, the overall results obtained with E. coli strain n.876 are interesting, because this isolate was not-sensitive to all the tested antibiotics, but sensible to all the EOs, with low MIC values. These results suggest that there is no correlation between the sensitivity to conventional antibacterial drugs and to EOs. Such natural products could be an alternative when treating some clinical cases.
Although mycotic cystitis is occasionally signaled in carnivores, causative agents investigated in the present study showed a wide range of resistance to commonly used antimycotic drugs. Fluconazole represents the preferred drug in the treatment of Candida UTI in human patients [25], while echinocandins and newer azoles are not routinely recommended, due to their low urinary concentrations [26]. Furthermore, itraconazole is the sole antimycotic drug allowed for systemic administration in veterinary medicine, and it is registered for treating feline microsporiasis, only. In this view, the local application of EOs formulations would be welcome.
T. vulgaris appeared as the most effective EO, probably due to its high content of thymol (52.6%). This monoterpene compound, together with carvacrol, appeared to be able to completely block ergosterol synthesis at the MIC values [27], making porous the membrane and provoking the yeast cell killing. Carvacrol is mostly contained in O. vulgare EO (about 66%) that, in the present study, showed a good efficacy, although with slightly higher MICs when compared to T. vulgaris. Moreover, these compounds are reported as able to restore antifungal susceptibility to fluconazole in resistant Candida strains [28]. Eugenol is contained in moderate amounts (11.5%) in O. basilicum EO and shares with thymol the ability to damage cell membrane in C. albicans [29]. This EO resulted active against part of selected yeasts (2/4 C. famata and 5/12 C. albicans). Anethol is the main component (about 90%) of I. verum EO, and it is reported to have a weak antimicrobial action [30].
Antimicrobial activities of I. verum and its more represented component, anethol, have been extensively studied, mainly against phytopathogenic fungi [31][32][33][34], indicating a strong antifungal activity. Antifungal properties of I. verum EO and its main features have been extensively revised by Wang et al. [35], indicating a versatile use of this compound in ethnobotany. At the best of our knowledge, the only assays of this EO against zoopathogenic fungi are referred to dermatophytes [36] and to some agents of mycotic otitis in pets [37]. In both cases I. verum EO showed a poor activity. Moreover, in the present study it did not yield univocal results, showing wide differences in MIC values. The marked differences among the results obtained by us, would indicate a yeast individual sensitivity, considered that, for C. albicans, MIC were ranging from more than 10 to 0.1 mg/mL. For these reasons individual checking for the in vitro efficacy of EOs should be carried out prior to establish an alternative treatment.
An interesting finding would be the apparent absence of relation between the sensitivity to EOs with regards to the multidrug resistance.
Topical application of EOs based formulations has been reported in murine models of Candida vaginitis [38,39], indicating the feasibility of this route of administration. However, considered the wide variability of drug/EOs sensitivity patterns among the different isolates of Candida, a preliminary evaluation of the in vitro activity of selected compounds should be performed, to identify the most suitable EO to treat Candida UTI in pets. This assessment would be advisable, considering that other non-albicans species such as C. famata, C. kefyr, C. inconspicua, C. rugosa, C. dubliniensis, and C. norvegensis, although rarely isolated, are now considered emerging species, as their isolation rate has increased between 2-and 10-fold in human patients over the last 15 years [40].
All EOs were purchased from the producer (FLORA ® , Pisa, Italy) and maintained at 4 • C in dark glass vials until used. They were microbiologically analyzed for quality control before antibacterial and antimycotic tests.

Gas Chromatography-Mass Spectrometry Analysis
The GC analysis was performed as previously described [44].

Bacterial Strains
Seven E. coli and eight Enterococcus spp. strains were employed in the study. The strains were previously isolated from female dogs and cats with urinary tract infections. E. coli and Enterococcus spp. strains were typed using API 20E and API 20STREP System (BioMérieux, Marcy l'Etoile, France), respectively and stored in glycerol broth at −80 • C.
Antibacterial activity of each EO was tested by Kirby-Bauer agar disk diffusion method following the procedures reported by Clinical and Laboratory Standards Institute (CLSI) [46] with some modifications. In details, each EO and mixture was diluted 1:10 in dimethyl sulfoxide (DMSO, Oxoid Ltd.) and one absorbent paper disk was impregnated with 10 µL of each dilution, respectively. A paper disk impregnated with 10 µL of DMSO was included as negative control. All tests were performed in triplicate.
Microdilution assay was performed in 96-well microtitre plates following the protocol previously described [43]. Briefly, the test was carried out in a total volume of 200 µL including 160 µL of brain hearth infusion broth (BHIB, Oxoid Ltd.), 20 µL of each bacterial suspension and 20 µL of each oil with final EOs concentrations ranging from 10% to 0.03%. Plates were incubated at 37 • C for 24 h. The same assay was performed simultaneously for bacterial growth control (tested bacteria and BHIB) and sterility control (tested oil or mixture and BHIB). All tests were executed in triplicate. The MIC value was defined as the lowest concentration, expressed as mg/mL, of EO at which microorganisms show no visible growth.

Fungal Strains
Four strains of C. famata and twelve of C. albicans were used for the assays. All the yeasts were clinical isolates from female dogs. Identification was achieved using physiological tests such as cultivation onto Corn Meal Agar (Sigma Aldrich, Milano, Italy) and germ-tube. Microscopy and biochemical profile evaluated by ID 32 (BioMérieux), were performed, also. When a doubtful profile was obtained, a final identification was carried out by molecular methods.

Conclusions
UTIs are frequent reason for antimicrobial treatment in dogs and cats. Choice of antibiotics and/or antifungal products should follow the in vitro evaluation of causative agent sensitivity. However, sometimes these microorganisms persist in the urogenital tract and/or a new infection can occur. The use of EOs has been proposed for the treatment of human UTIs, so these natural products could be evaluated in veterinary medicine, also, mainly when clinical healing cannot be achieved by using conventional drugs. Moreover, EOs could be used for relapses prevention. An in vitro evaluation of the isolated pathogens sensitivity to different EOs should be performed to select an oil for UTI treatment. As reported in literature for other microorganisms, in the present study T. vulgaris and O. vulgare EOs showed the strongest antimicrobial activity against E. coli, Enterococcus spp., C. albicans and C. famata, so they could be proposed for the formulation of external and/or intravesical washes, after a careful evaluation of both cytotoxicity and therapeutic index.

Conflicts of Interest:
The authors declare no conflict of interest.