Potent In Vitro Activity of Citrus aurantium Essential Oil and Vitis vinifera Hydrolate Against Gut Yeast Isolates from Irritable Bowel Syndrome Patients—The Right Mix for Potential Therapeutic Use

Background: Irritable bowel syndrome (IBS) is a functional disorder without any pathological alteration, in which the alterations of the Candida/Saccharomyces ratio of the gut microbiota, the balance of pro and anti-inflammatory cytokines and the brain-gut-microbiome axis are important for the development and progression of IBS. The aim of the study was to identify natural products, including essential oils or hydrolates, which were contextually harmless for the gut beneficial strains (e.g., Saccharomyces spp.) but inhibitory for the pathogenic ones (Candida spp.). Methods: The effectiveness of 6 essential oils and 2 hydrolates was evaluated using microbiological tests, carried out on 50 clinical isolates (Candida, Saccharomyces and Galattomyces species) and 9 probiotic strains (Saccharomyces cerevisiae, Lactobacillus species, Akkermansia muciniphila and Faecalibacterium prausnitzii) and immunological and antioxidant assays. Results: The study led to a mixture based on a 1/100 ratio of Citrus aurantium var. amara essential oil / Vitis vinifera cv Italia hydrolate able to contextually reduce, in a concentration-dependent manner, the ability of Candida species to form hyphal filaments and have an interesting immunomodulatory and anti-oxidant action. This mixture can potentially be useful in the IBS treatment promoting the restoration of the intestinal microbial and immunological balance.


Introduction
Irritable bowel syndrome (IBS) is a gastrointestinal disorder that affects around 11% of the population globally [1]. Despite being a chronic condition, contrary to the intestinal bowel disease (IBD) that is a chronic inflammation of the intestine, IBS is a functional disorder without any pathological alteration identified to date. However, IBS usually involves the brain-gut-microbiome axis but it can develop after an enteric infection, with the stress exacerbating the disorder-associated symptoms [2]. Symptoms reported in either IBS or IBD patients are almost the same but differ with respect to their extent. Generally, patients with IBD have more severe gastrointestinal symptoms compared with IBS. [3]. There is a growing evidence that the gut microbiota alteration (e.g., decrease in the relative abundance of a beneficial microorganism like Akkermansia muciniphila) and the psychology of the patient are both important for the development and progression of IBS [2]. Since 1985, authors have correlated the presence of Candida species in the intestinal tract with IBS [4] but only twenty years later other authors showed that yeasts are able to cause IBS symptoms in patients sensitized by Candida products, antigens and cross-antigens [5]. Other studies showed that the integration of Saccharomyces cerevisiae in the diet of patients with IBS was able to significantly reduce abdominal pain/discomfort and bloating and to improve the minimal clinically relevant threshold up to 10% [6][7][8].
In 2018, another study reported on the association between the relative abundance of A. muciniphila and the pain modulation in IBS patients [9]. Finally, studies showed that the concomitant decrease in the abundance of Saccharomyces species (i.e., S. cerevisiae and S. cerevisiae var. boulardii) and the increase of Candida species in the intestinal or vaginal microbial communities correlated with a decreased expression of the anti-inflammatory cytokine IL-10 and an increased expression of the pro-inflammatory cytokines IL-6, IL-8 and TNF-α [10][11][12][13].
Integrated treatments for IBS include the combined administration of conventional and complementary medications.
The latter encompass Chinese medicine, homeopathy and phytotherapy [14,15]. Several studies have recently evaluated the efficacy of natural products, such as essential oils (EOs), in the treatment of IBS symptoms [16][17][18][19][20][21][22]. Some of the EOs studied derive from aromatic plants of the Mediterranean area, such as Foeniculum vulgare, Pimpinella anisum, Mentha × piperita and Citrus aurantium. The most active components of EOs are terpenes, whereas the hydrophilic hydrolates (Hys) are by-products of EO distillation (also known as hydrolates or aromatic waters) that contain up to 1% terpenic components of the EOs [23].
The aim of the present study was to evaluate the effectiveness of six EOs and two Hys, alone or in combination, on both yeast (Candida, Saccharomyces and Galattomyces) isolates from IBS patients and probiotic strains (Saccharomyces and Lactobacillus).

EO and Hy Natural Substances
The six EOs studied were as follows-Citrus lemon var. femminello (femminello lemon), supplied by Exentiae Srl (Catania, Italy); C. aurantium var. amara (bitter orange), Citrus lemon (lemon) and Citrus reticulata (mandarin), supplied by Misitano and Stracuzzi Company (Messina, Italy); Monarda didyma (bee balm), supplied by the DISTAL of the Università di Bologna (Bologna, Italy) [24]; and Cinnamomum zeylanicum (cinnamon) supplied by APA-Ct (Forlì, Italy). All the Citrus species EOs were obtained from their respective fruit epicarps by pressing, whereas the M. didyma EO was obtained from the plant aerial parts and the C. zeylanicum EO was obtained from the cortex, both through steam distillation. The two Hys studied were from M. didyma and Vitis vinifera cv Italia, supplied by DISTAL of the Università di Bologna (Bologna, Italy). In particular, the M. didyma Hy was obtained from the same distillation process as for the M. didyma EO [25], whereas V. vinifera, for the first time in this study, was obtained according to the following protocol. Briefly, vegetal material (ripe berries and grape-stalks) of the V. vinifera cv Italia was collected from the plants grown in the Sicily region, monitored with a visual inspection to eliminate the parts apparently infested by bacteria and/or fungi and cut in small pieces that were subjected to distillation in a Clevenger-type apparatus. After one hour of distillation, only a sample of Hy (no EO was visualized) was collected.

Fungal and Bacterial Organisms
Fifty clinical isolates of fungal species (29 Candida albicans, 3 Candida glabrata, 1 Candida krusei, 4 Candida parapsilosis, 1 Candida zeylanoides, 8 Galactomyces species and 4 S. cerevisiae) were used, which had been recovered from fecal samples of IBS patients at the Dipartimento di Gastroenterologia e Epatologia, Ospedale G. da Saliceto of Piacenza, Italy. Patients were selected according to ROME III diagnostic criteria for IBS; all patients presented abdominal pain/discomfort for almost 3 days/weeks. The 62% of IBS patients had diarrhea (> 3 discharges/day) and the 38% suffered of constipation (<3 discharges/weeks). The Institutional Review Board of the Dipartimento di Gastroenterologia e Epatologia, Ospedale "G. da Saliceto" of Piacenza, Italy, approved the protocol for the obtainment of the patients' fecal samples. Isolation and identification of fungal cultivable single-cell pure colonies from stool sample of the IBS patients were performed as previously described [24]. Two fluconazole-resistant Candida tropicalis isolates were also used, obtained from clinical stool samples of patients hospitalized at the Fondazione Policlinico Universitario A. Gemelli IRCCS of Rome, Italy, only to evaluate the effectiveness of natural products on the fungal morphology. For comparison purposes, nine probiotic strains were included, five non-commercial strains (A. muciniphila DSMZ 22959, Faecalibacterium prausnitzii DSMZ17677, Lactobacillus acidophilus LA3SACCO, Lactobacillus plantarum ATCC 14917, Lactobacillus rhamnosus DSMZ 20021) and four commercial strains (S. boulardii CBS5926, L. acidophilus LA-14, Lactobacillus casei RO215, Lactobacillus helveticus ROO52).

Broth Microdilution Susceptibility Testing
Broth microdilution (BMD) susceptibility tests was performed on all the 59 microbial organisms according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) international guidelines. RPMI broth (Irvine Scientific, St. Ana, CA, USA) was used for fungal organisms, Man, Rogosa and Sharpe (MRS) broth (Sigma Aldrich, St. Louis, MO, USA) for Lactobacillus species organisms and Brucella broth (Thermo Scientific, West Sussex, UK) for A. muciniphila and F. prausnitzii organisms. All the bacterial probiotic strains were allowed to grow in an anaerobic condition. Briefly, for each organism a cell suspension equal to 5 × 10 5 cfu/mL was obtained and 50 µL of the suspension was dispensed in each well of a 96-well microtiter plate, which contained 50 µL of serial concentrations ranging from 4% v/v (40 mL/L) to 0.06% v/v (0.6 mL/L) for each EO and from 50% v/v (500 mL/L) to 1.67% v/v (16.7 mL/L) for each Hy. Tween 80 was utilized, at 0.05% v/v, as emulsifier for EO. Suitable positive controls, only with culture medium, were made for each microbiological strain. After 24/48 h of incubation at 37 • C, minimum inhibitory concentration (MIC) values were determined by spectrophotometric reading at 450 nm (EL808, Biotek, Winooski, VT, USA). To determine the minimum cytocidal concentration (MCC) or the minimum fungicidal concentration (MFC), 5 µL of the content of each well were seeded on standard medium agar plates, which were incubated at 37 • C for 24/48 h. For each natural substance (EO or Hy)/organism combination tested, the MIC was defined as the lowest concentration that allowed a complete inhibition of the organism's growth compared with the growth in the substance-free control and the MCC as the lowest concentration that allowed the death of 99.9% or more of the initial inoculum organism. All tests were performed in triplicate. For C. aurantium var. amara, C. zeylanicum or M. dydima EOs, the BMD susceptibility tests were also performed with co-cultures of two randomly selected organisms (i.e., C. albicans clinical isolate 3.1 and S. cerevisiae clinical isolate 14.3), starting from a 5 × 10 5 cfu/mL cell suspension obtained with a mixture (ratio 1:1) of the two isolates. Then, 10 µL of the content of each well showing the organism(s)' visible growth was seeded on the Brilliance Candida agar chromogenic medium (Oxoid, Thermo Scientific, Basingstoke, UK) plates, which was incubated at 37 • C for 48 h. After incubation, the blue (C. albicans) and purple (S. cerevisiae) colonies were counted to obtain the cfu/mL values for each organism, respectively. Finally, to assess the effect of the C. aurantium var. amara EO and V. vinifera cv Italia combination, a checkerboard broth microdilution susceptibility testing was performed with the two cultured fungal organisms and sub-MIC concentration of C. aurantium var. amara EO (from 0.50% v/v to 0.06%) and V. vinifera cv Italia Hy (from 0.25% v/v to 0.06% v/v). Then, the growth percentage at each combination's concentration was determined as the ratio between the number of cfu/mL obtained for that concentration and the number of cfu/mL obtained for the substance-free control. Finally, the content of each well was visualized directly with an inverted microscope and, after scraping of the bottom (50 µL), with an optical microscope to study the fungal morphology.

Peripheral Blood Mononuclear Cell (PBMC) Collection and ELISA Testing
Peripheral blood mononuclear cells (PBMCs) from human blood derived from refrigerated, banked healthy donor blood were collected with HISTOPAQUE-1077 (Sigma Aldrich, St. Louis, MO, USA) solution according to the manufacturer's protocol. After collection, 250 µL of a suspension of 1.2 × 10 6 cells/mL were cultured in sextuplicate in a 48-well cell culture plate at 37 • C overnight. PBMCs were incubated in the presence of serial dilutions of the V. vinifera Hy (from 50% v/v to 6.25% v/v) alone or in combination with C. aurantium var. amara EO (from 0.50% v/v to 0.06% v/v) or with lipopolysaccharide (LPS, 1 µg/mL). Untreated PBMC cells were used as control. After 24 h of incubation at 37 • C, the content of 4 replicates was mixed and centrifuged (400 × g for 10 min at room temperature) to separate the cells from the culture medium, which was stored at −20 • C until testing. The remaining two replicates were treated with 5 µL of Alamar Blue reagent (Invitrogen, Carlsbad, CA, USA) in order to study the cellular vitality. To evaluate the expression of IL-6, IL-8, TNF-α and IL-10 cytokines, ELISA tests (AVIVA Systems, San Diego, CA, USA) were performed using 100 µL of the aforementioned culture medium, according to the manufacturer's instructions. In each well, the content of cytokines was evaluated with optical reading using a microplate reading set to 450 nm. A calibration curve with six serial dilutions of the standard was included in each experiment, which was repeated three times.

DPPH (2,2-Diphenyl-1-Picryl-Hydrazyl-Hydrate) Assay
The DPPH antioxidant assay was performed to evaluate the scavenging activity of the V. vinifera Hy. Serial dilutions of the Hy (100 µL) were mixed with the 400-µM DPPH solution (100 µL) in a 96-well microtiter plate. Methanol was used as a negative control. The mixtures were incubated for 30 min at 37 • C in the dark and the change in spectrophotometric absorbance at 517 nm was measured. Mean values were obtained from triplicate experiments. The percentage of radical scavenging activity per sample was calculated using the following equation: where A DPPH is the absorbance of the DPPH solution only and A Extract is the absorbance of the tested sample. The IC50 value for the V. vinifera Hy was defined as the concentration of organic compound at which 50% of DPPH radicals was quenched and was expressed as mean ± standard deviation.

GC-MS Analysis
A V. vinifera cv Italia Hy sample (50 mL) was extracted with diethyl ether (100 mL). The collected organic layers dried over dry Na2SO4 and deprived of the solvent at reduced pressure furnished about 21.5 mg of an oily mixture (0.043%). The resulting Hy extract (1 mg) as well as a C. aurantium var. amara EO sample (1 mg) were separately diluted in 1mL methanol and used for Gas Chromatography-Mass Spectrometry (GC/MS) analysis. A GC-MS Perkin Elmer Clarus 500 instrument equipped with flame ionization detector (FID) and a Restek Stabilwax fused-silica capillary column (length 60 m × 0.25 mm ID) was employed. Helium was used as the carrier gas with a flow rate of 1 mL/min and the oven temperature program was as follows-5 min at 60 • C then a gradient of 5 • C/minute to 220 • C. One µL of sample was manually injected at 280 • C into the Gas Chromatography injector in the splitless mode. Relative percentages for quantification of the components were calculated by electronic integration of the GC-FID peak areas. Identification of the constituents was performed on MS library search (Wiley and Nist). Linear retention indices (LRIs) of each compound were calculated using a mixture of aliphatic hydrocarbons (C8-C30, Ultrasci) injected directly into GC injector.

Cytotoxicity Assays
Human gingival fibroblasts were obtained as described from by Radicioni G. et al. [26]. Cells were grown in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal calf serum (FCS) and were not used beyond the fifth step. The cytotoxicity of serial dilutions of the mixture of C. aurantium var. amara EO (from 0.50% v/v to 0.03% v/v) and V. vinifera cv Italia Hy (from 50% v/v to 3.25% v/v) for 24 and 48 h was evaluated. Untreated culture cells were used as control. Tetrazolium compound [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt; MTS] (Promega, Madison, WI, USA) was used, which was diluted in basic medium. Then, 1 × 10 4 cells were suspended in 200 mL of basic medium to be coated in 96-well cell culture plates and grown at confluence for 24 h. Cell vitality was evaluated after the addition of the aforementioned serial dilutions at 37 • C for 24 and 48 h. The MTS analysis was performed according to the manufacturer's protocols. Each experiment was done in triplicate.

Statistical Analysis
All data obtained in triplicate from each experiment were expressed as means ± standard deviation (SD). The response of S. cerevisiae and C. albicans to C. aurantium var amara EO concentration was fitted by means of mathematical models, using the Sigma Plot 10 (Systat Software Inc., Chicago, IL, USA) software. The best fitting model was chosen between linear and exponential decay function. The two respective equations are: where: y0, intercept; a, slope (Equation (1)), initial level (Equation (2)); b, rate constant. All the results obtained from the broth microdilution susceptibility testing were presented as relative values, assuming the average of S. cerevisiae and C. albicans as positive control (Ctrl+) = 100%. The data were treated with the GraphPad V 4.0 software. A two-way ANOVA for the factors clinical isolates (2 levels) and EO-Hy combined treatments (13 levels including Ctrl+) was run on these data. The CoStat 6.3 package (CoHort Software, Berkeley, CA, USA) was used for the ANOVA.
All the results, obtained by cytotoxicity study, were expressed as mean percentage of vitality and data treated with GraphPad V 4.0 software. The treated group means were compared by analysis of variance (ANOVA) followed by a multiple comparison of means using the Dunnett's test. Values with p < 0.05 were considered significant. Table 1 shows the MIC values of the 38 Candida species isolated from IBS patients against the EOs of C. lemon, C. lemon var. femminello, C. reticulata, C. aurantium var. amara, C. zeylanicum and M. didyma tested at the 0.50% v/v to 0.06% v/v concentrations. The MIC50 values were > 0.50% v/v for C. lemon, C. lemon var. femminello and C. reticulata, 0.50% v/v for C. aurantium var. amara, ≤ 0.06% v/v for C. zeylanicum and 0.25% v/v for M. didyma. Table 2 shows the MIC values of the 7 probiotic strains and the 12 Galactomyces species and S. cerevisiae isolates against the EOs of C. aurantium var. amara, C. zeylanicum and M. didyma (the only three EOs showing better activity) tested at the 4.0% v/v to 0.06% v/v concentrations. It was chosen to test lower EO concentrations for the isolates from pathogenic species (i.e., Candida species) and higher EO concentrations for the isolates from beneficial species (i.e., probiotic or Galactomyces/S. cerevisiae species). This choice would ensure the activity toward the pathogenic species and non-toxicity toward the beneficial species eventually present in the gut microbiota of IBS patients. The MIC50 values were Based on these findings, the C. aurantium var. amara, C. zeylanicum and M. didyma EOs were tested against two co-cultured clinical isolates of C. albicans (clinical isolate 3.1) and S. cerevisiae (clinical isolate 14.3), which were representative of both pathogenic and beneficial isolates included in the study. Using 4% v/v to 0.015% v/v concentrations, it was found that MIC and MFC values were equal for all the three EOs. Expectedly, the C. zeylanicum EO was the most effective (MFCs for C. albicans and S. cerevisiae, 0.015% v/v and 0.06% v/v, respectively), compared to the M. didyma EO (MFCs for C. albicans and S. cerevisiae, 1% v/v) and the C. aurantium var. amara EO (MFCs for C. albicans and S. cerevisiae, 2% v/v). Therefore, the inhibitory capability of the C. aurantium var. amara EO was further investigated by counting the cfu/mL of both C. albicans and S. cerevisiae recovered after their incubation with sub-MIC concentrations (≤ 2% v/v) of the EO. Figure 1 shows that the 2% concentration of C. aurantium var. amara EO was able to completely suppress both organisms. Interestingly, S. cerevisiae exhibited an exponential trend of increased tolerance to the EO (R 2 = 0.71) that was statistically significant and different (p < 0.05) from the linear trend of C. albicans (R 2 = 0.65).

Immunomodulatory and Antioxidant Activities of the V. vinifera cv Italia Hy
The immunomodulatory activities of both V. vinifera cv Italia and M. didyma Hys were studied, using human PBMCs. Data showed that PBMCs treated with M. didyma Hy died whereas those treated with V. vinifera cv Italia Hy did not lose viability. Therefore, the effects of the V. vinifera cv Italia Hy and, in parallel, LPS alone or in combination with the Hy has been investigated, on the levels of IL-10 and TNF- produced by PBMCs. Table 3 summarizes both the mean and standard deviation values of IL-10 and TNF- under the tested conditions, while figure 2 shows the relationship of the two cytokines in each condition.  As expected, neither M. dydima nor V. vinifera cv Italia Hys showed antimicrobial activities against all the 59 microbial strains above mentioned.

Immunomodulatory and Antioxidant Activities of the V. vinifera cv Italia Hy
The immunomodulatory activities of both V. vinifera cv Italia and M. didyma Hys were studied, using human PBMCs. Data showed that PBMCs treated with M. didyma Hy died whereas those treated with V. vinifera cv Italia Hy did not lose viability. Therefore, the effects of the V. vinifera cv Italia Hy and, in parallel, LPS alone or in combination with the Hy has been investigated, on the levels of IL-10 and TNF-α produced by PBMCs. Table 3 summarizes both the mean and standard deviation values of IL-10 and TNF-α under the tested conditions, while Figure 2 shows the relationship of the two cytokines in each condition.  As shown in table 3, the levels of IL-10 and TNF- in untreated PBMCs were 2.6 ± 0.07 pg/mL and 47.98 ± 2.8 pg/mL, respectively, whereas those in PBMCs treated with LPS were 518 pg/mL ± 3.0 pg/mL and 2002.9 pg/mL ±14.4 pg/mL. The IL-10 and TNF- values of PBMCs treated with 50% of the V. vinifera cv Italia Hy and those of PBMCs treated with both 50% of the V. vinifera cv Italia Hy and LPS had values close to those of the untreated PBMCs ( Figure 2, letters a and e in CTR circle). The values of PBMCs treated with concentrations other than 50% of V. vinifera cv Italia Hy alone varied based on the Hy concentration tested. In particular, only TNF- values were higher than those of PBMCs, while both IL-10 and TNF- values were lower than those of PBMCs stimulated with LPS alone (Figure 2, letters b, c, d in Hy circle). PBMCs treated with concentrations other than 50% of V. vinifera cv Italia Hy in combination with LPS had a strong reduction of IL-10 values, compared to PBMCs stimulated with LPS only.
The DPPH assay was used to study the anti-oxidant activity of the V. vinifera cv Italia Hy. Data show that the Hy exhibited low activity because of an IC50 of 4 ×10-3 ± 0.001 g/L of organic component, which corresponds to the one present in a 95.6 ± 2.1% v/v of Hy solution (i.e. to the one of the almost pure Hy). Figure 3 shows the results of checkerboard BMD testing of C. aurantium var. amara EO (from 0.50% v/v to 0.06% v/v) and V. vinifera cv Italia Hy (from 25% v/v to 6.25% v/v) against C. albicans (clinical isolate 3.1) and S. cerevisiae (clinical isolate 14.3). As showed by ANOVA test, concentrations equal to or lower than 0.12% v/v for EO, regardless of Hy concentration, did not produce any statistically significant inhibition for the two isolates, whereas for concentrations of the EO higher than 0.12% v/v the isolate of S. cerevisiae shows a statistically significant growth inhibition. . Letters e, f, g and h are referred to samples treated with scalar dilution of the Hy (50% v/v, 25% v/v, 12.50% v/v and 0.06% v/v respectively) and LPS (1 µgr/mL). Letters i, l, m and n are referred to PBMCs treated with a 1:100 v/v mixture of the Hy, (50% v/v, 25% v/v, 12.50% v/v and 0.06% v/v respectively) and the EO (0.50% v/v, 0.25% v/v, 0.12% v/v and 0.06% v/v respectively) plus LPS. Table 3, the levels of IL-10 and TNF-α in untreated PBMCs were 2.6 ± 0.07 pg/mL and 47.98 ± 2.8 pg/mL, respectively, whereas those in PBMCs treated with LPS were 518 pg/mL ± 3.0 pg/mL and 2002.9 pg/mL ±14.4 pg/mL. The IL-10 and TNF-α values of PBMCs treated with 50% of the V. vinifera cv Italia Hy and those of PBMCs treated with both 50% of the V. vinifera cv Italia Hy and LPS had values close to those of the untreated PBMCs (Figure 2, letters a and e in CTR circle). The values of PBMCs treated with concentrations other than 50% of V. vinifera cv Italia Hy alone varied based on the Hy concentration tested. In particular, only TNF-α values were higher than those of PBMCs, while both IL-10 and TNF-α values were lower than those of PBMCs stimulated with LPS alone (Figure 2, letters b, c, d in Hy circle). PBMCs treated with concentrations other than 50% of V. vinifera cv Italia Hy in combination with LPS had a strong reduction of IL-10 values, compared to PBMCs stimulated with LPS only.

As shown in
The DPPH assay was used to study the anti-oxidant activity of the V. vinifera cv Italia Hy. Data show that the Hy exhibited low activity because of an IC50 of 4 ×10-3 ± 0.001 µg/µL of organic component, which corresponds to the one present in a 95.6 ± 2.1% v/v of Hy solution (i.e., to the one of the almost pure Hy). Figure 3 shows the results of checkerboard BMD testing of C. aurantium var. amara EO (from 0.50% v/v to 0.06% v/v) and V. vinifera cv Italia Hy (from 25% v/v to 6.25% v/v) against C. albicans (clinical isolate 3.1) and S. cerevisiae (clinical isolate 14.3). As showed by ANOVA test, concentrations equal to or lower than 0.12% v/v for EO, regardless of Hy concentration, did not produce any statistically significant inhibition for the two isolates, whereas for concentrations of the EO higher than 0.12% v/v the isolate of S. cerevisiae shows a statistically significant growth inhibition. Nutrients 2020, 12, x FOR PEER REVIEW 10 of 16 Then, 1:100 v/v mixtures of EO (0.50% v/v to 0.06% v/v) and Hy (50% v/v to 6.25%) were evaluated for morphological microscopy examination and five of seven Candida species (exceptions were one C. glabrata and one C. tropicalis [0.4 R]) isolates displayed a concentration-dependent reduced capability of forming hyphal filaments ( Table 4). As expected, any morphological alterations for the S. cerevisiae isolate tested were observed. Additionally, none of the EO/Hy mixture concentrations was able to affect the nine probiotic isolates as well the A. muciniphila and F. prausnitzii isolates (potentially usable as probiotics).

Effectiveness of the C. aurantium var. Amara EO and V. vinifera cv Italia Hy Combination
It was assessed whether the EO/Hy mixture concentrations had immunomodulatory activities on human PBMCs when used in combination with LPS.
As shown in Figure 2   Then, 1:100 v/v mixtures of EO (0.50% v/v to 0.06% v/v) and Hy (50% v/v to 6.25%) were evaluated for morphological microscopy examination and five of seven Candida species (exceptions were one C. glabrata and one C. tropicalis [0.4 R]) isolates displayed a concentration-dependent reduced capability of forming hyphal filaments ( Table 4). As expected, any morphological alterations for the S. cerevisiae isolate tested were observed. Additionally, none of the EO/Hy mixture concentrations was able to affect the nine probiotic isolates as well the A. muciniphila and F. prausnitzii isolates (potentially usable as probiotics).
It was assessed whether the EO/Hy mixture concentrations had immunomodulatory activities on human PBMCs when used in combination with LPS.
As shown in Figure 2, a linear decrease of IL-10 and TNF-α values were observed in PBMCs treated with both LPS and different concentrations of the EO/Hy mixture. In particular, at the lowest concentration tested (6.25% v/v Hy and 0.625% v/v EO) the values of cytokines are similar to those expressed by the PBMC stimulated with LPS only. Whereas, by increasing the tested concentrations of the EO/Hy mixture, the IL-10 and TNF-α values decrease linearly (Figure 2, letters l, m and n in Mix + LPS circle) until they become similar to those of the PBMCs (Figure 2, letter i in CTR circle).
Furthermore, it was investigated whether the EO/Hy mixture concentrations had cytotoxic activities on human gingival fibroblasts. As shown in Figure 4, none of the concentrations tested was capable of producing statistically significant effects on the fibroblasts after 24 h and 48 h of incubation.
Furthermore, it was investigated whether the EO/Hy mixture concentrations had cytotoxic activities on human gingival fibroblasts. As shown in Figure 4, none of the concentrations tested was capable of producing statistically significant effects on the fibroblasts after 24 h and 48 h of incubation. Finally, GC-MS analyses performed on both the EO and the organic fraction from Hy showed that the major component of EO was the limonene (93.93%) and the major components of the Hy were limonene (17.68%), cis-geraniol (25.40%) and -linalool (38.81%) ( Table 5).

Discussion
Few studies have been done on the effectiveness of essential oils in treating patients with IBS and most of them have been made in the last 10 years. In addition to the now known properties of the essential oil of Mentha × piperita that make it an interesting treatment not only for IBD but also for IBS [22], other essential oils, especially those obtained from aromatic plants with known carminative activity, have been recently studied in the treatment of IBS. Mahboubi M. [18] reviewed the action of Finally, GC-MS analyses performed on both the EO and the organic fraction from Hy showed that the major component of EO was the limonene (93.93%) and the major components of the Hy were limonene (17.68%), cis-geraniol (25.40%) and β-linalool (38.81%) ( Table 5).

Discussion
Few studies have been done on the effectiveness of essential oils in treating patients with IBS and most of them have been made in the last 10 years. In addition to the now known properties of the essential oil of Mentha × piperita that make it an interesting treatment not only for IBD but also for IBS [22], other essential oils, especially those obtained from aromatic plants with known carminative activity, have been recently studied in the treatment of IBS. Mahboubi M. [18] reviewed the action of Zataria multiflora EOs in the treatment of gastrointestinal problems, especially in IBS. As reported, 60 drops oral daily dose of Z. multiflora essential oil (2%) can relieve the symptoms of IBS without any adverse effects. Portincasa P. et al. [20] and Di Ciaula A. et al. [16] investigated the activity of the EOs of F. vulgaris in association with Curcuma longa extract in IBS patients. The two studies concluded that the administration of these natural compounds significantly improved symptoms and quality of life in IBS patients within 2 months from the treatment. Mosaffa-Jahromi M. et al. [19] have demonstrated that a treatment of 4 weeks with encapsulated P.anisum EO (200 mg for 3 times a day) is able to improve IBS symptoms better than placebo or M. piperita capsules (Colpermin ® ) (p < 0.0001). Finally, Agah S. et al. [27] studied the effect of 20 drops of Cuminum cyminum EO on 57 patients with IBS selected according to the ROME II diagnostic criteria. Data showed that IBS symptoms had a statistically significant reduction during and after treatment.
According to the most recent literature, at the base of the IBS there is an alteration of the intestinal microbiota in favor of microbial strains generally present in lower percentages (especially fungal strains). For this reason, we chose to investigate the antimicrobial action of some EOs and Hys on both fungal and probiotic strains. All the natural products studied were obtained from plants potentially cultivable in the Mediterranean area with the exception of C. zeylanicum, which was taken as a reference given its known high antimicrobial activity [28]. Furthermore, considering that the IBS symptoms are deeply conditioned by an altered brain-gut-microbiome axis, some of the EOs in study have been chosen to potentially act not only on microbial strains but also on the central nervous system (CNS). Therefore, 4 out of the 6 EOs studied belong to the Citrus genus which, besides being a genus widely present in Italy, is mainly used in aromatherapy in the treatment of nervous disorders and anxiety. Particularly, the C. aurantium var. amara EO is used as a natural mild sedative, hypnotic and anticonvulsant agent. It is used in treating insomnia, sleep disorders, anxiety, epilepsy and seizures. It has been also reported for its anti-spasmodic effects and, thanks to the presence of limonene, for its gastro-protective and ulcer healing actions by increasing the gastric production of mucus, which is useful as a secondary intervention in the treatment of chronic inflammatory diseases. Similarly, C. limon and C. reticulata EOs are used in phytotherapy to relieve symptoms of anxiety [29]. Instead, the M. didyma is an aromatic plant included in the BELFRIT (Belgium, France and Italy) list, native of Northern Europe but cultivable in Italy; its EO has two phenols as major components (thymol and carvacrol) and therefore shows a marked antimicrobial activity [24].
The study of MICs (and MCCs), against clinical isolates of Candida spp., indicates that no EOs belonging to the Citrus genus have an action comparable to that of M. didyma (MIC50 = 0.25% v/v) and C. zeylanicum (MIC50 ≤ 0.06 % v/v) EOs. All the EOs obtained from Citrus have a MIC50 > 0.5 % v/v with the exception of the EO of C. aurantium var. amara that showed a MIC50 = 0.50% v/v, slightly different from that of M. didyma. Based on this result, the action of the 3 EOs with major antifungal activity has been tested towards both probiotic strains and beneficial clinical isolates belonging to the intestinal microbiota ( Table 2). The data indicate that both the above are generally more resistant to the action of EOs than the pathogenic fungal isolates. In particular, the EO of C. aurantium var. amara has a MIC50 value (MIC50 ≥ 4% v/v) at least 3 dilutions higher against the beneficial clinical isolates than against the clinical isolates of Candida species (MIC50 = 0.50% v/v).
These data are in agreement with those available in the literature highlighting that the probiotic strains have higher MIC values than the pathogenic strains [3,30,31] and that thanks to the different MIC values between probiotic and pathogenic strains it is possible to administer probiotics and EOs together to cure pathogenic infections in human gut [32].
To evaluate the effectiveness of the three best EOs on the rebalance of beneficial and pathological clinical strains, we performed the BMD susceptibility tests using the EOs of C. aurantium var. amara, M. dydima and C. zeylanicum against two co-cultured randomly selected fungal clinical isolates (C. albicans clinical isolate 3.1 and S. cerevisiae clinical isolate 14.3).
The choice to co-culture one clinical isolate of C. albicans with one of S. cerevisiae is motivated by the evidence that in IBS patients a greater colonization of fungal strains belonging to the genus Candida, at the expense of the ones of S. cerevisiae, has been demonstrated [4,5,33]. If on the one hand it is known that S. cerevisiae is a yeast generally well represented in the intestinal microbiota of healthy subjects that, besides having a known anti-inflammatory action, is able to compete for the territory with Candida spp., on the other we know that IBS patients are characterized by a higher abundance of C. albicans (61.8%) compared to healthy subjects (48.75%) [33]. Moreover, IBS C. albicans isolates were different from those of healthy subjects not only genotypically but also phenotypically as these clinical isolates of C. albicans were able to grow at high temperatures and low pH compared to those isolated from healthy subjects [33]. These data suggest that IBS could be characterized by the presence of potentially virulent C. albicans strains and motivate our choice.
Our data show that, when co-cultured, the two fungi have the same MFC but, by analyzing the fungal growth in the presence of sub-MIC of C. aurantium var. amara EO, it is possible to identify a different modulating action of this EO towards the two clinical strains. In particular, for concentrations equal to or lower than 2% v/v of C. aurantium var. amara EO, the strain of C. albicans is significantly more inhibited than that of S. cerevisiae (Figure 1). Therefore, this activity could be useful in an attempt to rebalance the relative abundance of each clinical species in IBS patients.
The variation of the intestinal microbiota also involves the variation of the intestinal inflammatory microenvironment. Indeed, it is known that the decrease in S. cerevisiae and the increase in Candida species is generally related to an increase in the pro-inflammatory pattern in IBS patients [5]. As described in Results, an interesting immunomodulatory and anti-oxidant property of the V. vinifera cv Italia Hy was identified; on the contrary, the Hy of M. dydima was excluded because toxic for human PBMCs culture.
The V. vinifera is a plant genus recognized by both traditional and consolidated use for the treatment of microcirculation problems thanks to its important anti-oxidant, anti-inflammatory, spasmolytic and immunomodulated actions. The first sources on the use of V. vinifera as medicament are the Egyptian papyri, the Sumerian tablets, the writings of Hippocrates of Cos, Celsus, Galen and Paracelsus. Currently, natural products obtained from V. vinifera are found on the market in the form of dry extracts or soft aqueous extracts [34][35][36] but it is hard to find the hydrolate of V. vinifera because it is an unusual natural product obtained from the distillation of a non-aromatic plant. This means that our data are innovative showing that it is obtained from a still unexplored natural product that, thanks to its immunomodulatory and anti-oxidant activity, could be exploited in the chronic low-grade inflammation that establishes in patients with IBS [37].
Tests developed to identify the best combination of C. aurantium var. amara EO and V. vinifera cv Italia Hy have shown that a serial dilution of a mixture made with a 1:100 ratio of the two natural products (starting from a concentration of 0.50% v/v of C. aurantium var. amara EO and 50% v/v of V. vinifera cv Italia Hy) caused an interesting inhibition of the ability of C. albicans species to form hyphae ( Table 4). As reported in a recent review, the morphological plasticity of Candida albicans cells makes this fungus able to adapt to the different niches present in the gut of the host organisms. Generally, the production of C. albicans hyphae and pseudo-hyphae is typical of cellular models of virulence specialized in the commensalism [38]. The above are in agreement with what has been observed on C. albicans clinical isolates of IBS patients [33] where the gut microbiota is characterized by a greater abundance of C. albicans cells, more virulent than in healthy subjects. In this context, the use of one of the effective serial dilutions based on C. aurantium var. amara EO and V. vinifera cv Italia Hy, able to inhibit the production of hyphae and pseudo-hyphae of C. albicans strains, could prove to be successful in IBS treatment; especially considering that the same serial dilution of the two natural compounds also has the ability to linearly inhibit in a concentration-dependent manner the production of the IL-10 and TNF-α cytokines from PBMCs grown in presence of the pro-inflammatory stimulus (LPS) (Figure 2). Rather than the use of intestinal cells such as CaCo2, the ex vivo experiments carried out on PBMC were performed in order to evaluate the action of natural compounds in modulating the human systemic immunological response. This latter activity would enhance the effectiveness of the mixture inducing the reduction of the pro-inflammatory framework generally present in patients and able to favoring the growth of Candida species and promoting the establishment of the appropriate microenvironment for the development of beneficial probiotic strains including A. muciniphila and F. prausnitzii recently proposed as markers of the syndrome [39].
Finally, the mixture identified in this study appears to be suitable for an oral use in patients with IBS because, in addition to showing no in vitro cytotoxic activity (Figure 4), it appeared to be composed mainly of the monoterpene limonene because all the other chemical constituents were present in traces (Table 5). In a study, the limonene has been designated as a compound with low oral toxicity based on its lethal dose (LD50) and repeated-dose toxicity studies. Therefore, except for topical applications, for which attention must be paid to avoid its irritating potential, the use of limonene is widely granted in the pharmaceutical, cosmetic and food industries for its many properties [40].

Conclusions
In conclusion, even though further in vivo studies to evaluate the effectiveness as well as the possible presence of adverse and/or toxic effects are necessary, our study concludes that a mixture based on C. aurantium var. amara EO and V. vinifera cv Italia Hy (ratio 1:100) can be useful in the treatment of IBS. The use of this mixture in an integrated treatment with conventional treatments could favor the growth of both probiotic and beneficial gut strains, disfavoring pathogenic fungal strains and rebalancing the intestinal inflammatory microenvironment through the reduction of the expression of IL-10 and TNF-α cytokines and, in parallel, of the pro-inflammatory microenvironment.