Carvacrol Codrugs: A New Approach in the Antimicrobial Plan

Objective The increasing prevalence of antibiotic-resistant bacterial infections led to identify alternative strategies for a novel therapeutic approach. In this study, we synthesized ten carvacrol codrugs – obtained linking the carvacrol hydroxyl group to the carboxyl moiety of sulphur-containing amino acids via an ester bond – to develop novel compounds with improved antimicrobial and antibiofilm activities and reduced toxicity respect to carvacrol alone. Method All carvacrol codrugs were screened against a representative panel of Gram positive (S. aureus and S. epidermidis), Gram negative (E. coli and P. aeruginosa) bacterial strains and C. albicans, using broth microdilution assays. Findings Results showed that carvacrol codrug 4 possesses the most notable enhancement in the anti-bacterial activity displaying MIC and MBC values equal to 2.5 mg/mL for all bacterial strains, except for P. aeruginosa ATCC 9027 (MIC and MBC values equal to 5 mg/mL and 10 mg/mL, respectively). All carvacrol codrugs 1-10 revealed good antifungal activity against C. albicans ATCC 10231. The cytotoxicity assay showed that the novel carvacrol codrugs did not produce human blood hemolysis at their MIC values except for codrugs 8 and 9. In particular, deepened experiments performed on carvacrol codrug 4 showed an interesting antimicrobial effect on the mature biofilm produced by E. coli ATCC 8739, respect to the carvacrol alone. The antimicrobial effects of carvacrol codrug 4 were also analyzed by TEM evidencing morphological modifications in S. aureus, E. coli, and C. albicans. Conclusion The current study presents an insight into the use of codrug strategy for developing carvacrol derivatives with antibacterial and antibiofilm potentials, and reduced cytotoxicity.

ion trap mass spectrometer (San Jose, CA) equipped with an electrospray ionization (ESI) source. The capillary temperature was set at 300°C and the spray voltage at 4.25 kV. The fluid was nebulized by use of nitrogen (N 2 ) as both the sheath gas and the auxiliary gas. The identity of all new compounds was confirmed by NMR data and LC-MS/MS system; homogeneity was confirmed by thin-layer chromatography (TLC) on Merck 60 F 254 silica gel. Solutions were routinely dried over anhydrous sodium sulfate prior to evaporation. Chromatographic purifications were performed on a Merck 60 70-230 mesh ASTM silica gel column.
General method for acetylation. Acetic anhydride (6.6 mmol) was added to a stirred solution of alkylated cysteine derivatives (5.5 mmol) in acetic acid (11.1 mL) and the reaction mixture was left under stirring at room temperature for 1 h. The solvent was removed under vacuum giving the corresponding N-acetylated aminoacids which were used without further purification.
N-Ac-Cys(Trt)-OH (16). Yield: 59%; R f = 0.11, CH 2 Cl 2 :MeOH (9:1); 1  General method for alkylation. A solution of commercially available NAC (15) (6.13 mmol) and KOH (13.48 mmol) in MeOH (1 mL) was added with propargylic bromide (7.36 mmol) and left for 1 h under reflux. After evaporation of the solvent, the residue was taken up with water, acidified with HCl 1N (pH = 2), then extracted with AcOEt. The organic layer was dried over anhydrous Na 2 SO 4 and evaporated under vacuum to give product as oil, which were used without further purification.

Antimicrobial activity
Two Gram-positive (S. aureus ATCC 29213 and S. epidermidis ATCC 35984) and two Gramnegative (E. coli ATCC 8739 and P. aeruginosa ATCC 9027) strains, and one fungal strain (C. albicans ATCC 10231) were used for the detection of antibacterial activity of carvacrol, NAC, Ac-Cys(Allyl)-OH, and carvacrol codrugs 1-10. All tested compounds were dissolved in 20% DMSO to prepare the stock solution (100 mg/mL). The final concentration of DMSO in each well appeared to be not toxic for all studied microorganisms, as confirmed by the solvent control test included in the study.
The effects of carvacrol codrugs 1-10 and reference compounds on planktonic cells were evaluated by Minimum Inhibitory Concentration (MIC) and Minimum Bactericidal Concentration (MBC) determination using the broth microdilution method according to EUCAST guidelines [33]. Bacterial suspensions, grown in Mueller-Hinton Broth (MHB) at logarithmic phase, were incubated on microtiter plates at a concentration of 5 x 10 5 CFU/mL, with several dilutions (0.07-40 mg/mL) for 24 h at 37°C. The MIC was defined as the lowest concentration of substances giving a complete inhibition of visible growth in comparison with a control well, and the MBC was determined as the lowest concentration at which no bacterial growth occurred on Mueller-Hinton Agar plates.
The MIC detection of C. albicans ATCC 10231 was performed using the broth microdilution method according to EUCAST guidelines [34] in RPMI 1640 plus 2% glucose with a final inoculum of 1-5 x 10 5 CFU/mL for 24-48 h at 37°C. The Minimum Fungicidal Concentration (MFC) was determined as the lowest concentration of substances at which no fungal growth occurred on Sabouraud agar plates.
Data were obtained from at least three independent experiments performed in duplicate. The susceptibility of the bacterial strains against ciprofloxacin was used as an internal standard during MIC determinations.
Antibiofilm assay. The efficacy on established biofilm was evaluated by determining Biofilm Inhibitory Concentration (BIC) and Biofilm Eradication Concentration (BEC) according to the method described by Johnson et al. [35] with some modifications. Bacterial suspensions, grown in Tryptic soy broth supplemented with 0.5% (v/v) glucose at logarithmic phase, as well as C. albicans grown in RPMI 1640 plus 2% glucose, were incubated on flat-bottomed microtiter plates at a concentration of 5 x 10 5 CFU/mL. After 24 h of incubation at 37°C, the planktonic cells were gently removed and wells were washed with sterile phosphate-buffered saline solution (PBS) pH 7.3 and filled with carvacrol, carvacrol codrug 4, Ac-Cys(Allyl)-OH with dilutions ranging from the MIC values to a maximum concentration of 50 mg/mL. The OD 600 was measured at time 0 and after incubation for 24 h at 37°C. The BIC values were determined as the lowest concentrations where no growth occurred in the supernatant fluid, confirmed by no increase in optical density compared with the initial reading. The BEC values were determined as the lowest concentrations at which no bacterial growth occurred on Tryptic soy agar for bacteria and on Sabouraud agar for C. albicans. Data were obtained from at least three independent experiments performed in duplicate.
Hemolytic Activity. Experimental protocols to study hemolysis on human red cells were approved by Ethics Committee of the University "G. d'Annunzio" Chieti-Pescara. Human red cells were from two volunteers, who signed the informed consent for this study.
Hemolytic activity of carvacrol codrugs 1-10 was tested against human red blood cells (h-RBC). Fresh human blood, collected with EDTA, was centrifuged at 3000 rpm for 5 min then washed three times with PBS pH 7.3. Red blood cells were diluted to 4% in PBS and incubated with different codrugs concentrations ranging from 25 to 0.05 mg/mL. After 1 h of incubation at 37°C, the suspensions were sedimented by centrifugation and the release of hemoglobin was determined by absorbance measurement at 405 nm and compared with a 0% hemolysis control (PBS) and a 100% hemolysis control (PBS with 1% v/v Triton X-100). The percentage of hemolysis was calculated using the following equation: Hemolysis ð%Þ ¼ ½ðOD 405 sample À OD 405 0% lysis controlÞ=ðOD 405 100% lysis control À OD 405 0% lysis controÞ X 100 Transmission electron microscopy (TEM). For transmission electron microscopy, microrganisms were fixed with 2.5% glutaraldehyde in 0.1 M cacodylate buffer, postfixed with 1% OsO 4 , dehydrated in graded ethanol and embedded in Epon 812. Ultrathin sections (60 to 80 nm) were then mounted on 200-mesh copper grids and stained with uranyl acetate and lead citrate for ultrastructural observation. Sections were photographed using a Philips 268 D electron microscope (FEI).

Pharmacokinetic studies
HPLC-UV assays. The HPLC system consisted of a Waters 600 HPLC pump (Waters Corporation, Milford, MA, USA), equipped with a Waters 2996 photodiode array detector. The column was a Kinetex RP-C8 column (3.0 x 150 mm, 5 μm). The mobile phase consisted of 0.1% TFA in acetonitrile/0.1% TFA in 95% H 2 O and 5% acetonitrile 50:50, under isocratic conditions at a flow rate of 1 mL/min. The UV detector was set at a length of 220 nm.
Solubility. The water solubility was determined by stirring an excess of the compound in 1 mL of water for 15 min at room temperature. Then the mixture was filtered using a Millipore filter (0.45 μm) and the concentration of the compound in its saturated solution was determined by HPLC analysis [36].
Lipophilicity. The calculated clogP values were determined using ACD LogP software package, version 4.55 (Advanced Chemistry Development Inc., Toronto, Canada).
Kinetics of chemical hydrolysis. The chemical hydrolysis rate of compound 4 was studied in solution of 0.02 M hydrochloric acid buffer of pH 1.3, as non-enzymatic simulated gastric fluid (SGF), and a 0.02 M phosphate buffer of pH 7.4 at 37°C, as simulated intestinal fluid (SIF). Reactions were initiated by adding 1 mL of 10 -4 M stock solution (in acetonitrile) of the compound to 10 mL of the appropriate aqueous buffer solution, containing 20% acetonitrile. At appropriate time intervals, samples of 20 μL were withdrawn and analyzed by HPLC. Pseudo-first-order rate constants (k obs ) for the hydrolysis of the compounds were then calculated from the slopes of the linear plots of log (% residual compound) against time. The experiments were run in triplicate and the mean values of the rate constants were calculated [37].
Kinetics of enzymatic hydrolysis. The enzymatic hydrolysis was evaluated in human and rat plasma. Stock solution of drug was prepared in methanol (1 mg/mL) which did not have effects on the enzymatic degradation (data not shown). A volume of 200 μL of this solution was added with pre-heated (37°C) plasma fractions (4 mL) previously diluted with 0.02 M phosphate buffer (pH 7.4) to give a final volume of 5 mL (80% plasma). Aliquots (100 μL) were taken at various times and deproteinized by mixing with 200 μL of 0.01 M HCl in methanol. After centrifugation for 5 min at 5000 × g, 10 μL of the supernatant layer were analysed by chromatography as described above. The amounts of remaining intact compound were plotted as a function of incubation time [38].
Drug stability testing in intestinal fluids. To assess enzymatic stability, hydrochloric buffer with pepsin (10 or 40 mg/mL) and phosphate buffer with pancreatin (10 or 40 mg/mL) were used. Buffer solution (250 μL) was preincubated at 37°C and 50 μL of drug stock solution (50 μM in ethanol and FASSIF 1:9) were added and shaken at 37°C and 650 rpm. Samples of 100 μL were withdrawn at various times and 100 µL of ice-cold acetonitrile containing 0.5 v/v % formic acid were added to stop enzymatic activity [39]. Samples were vortexed and centrifuged at 2°C and 10000 rpm for 10 min. The drug content in the supernatant was analyzed by HPLC.
PAMPA Method. Parallel Artificial Membrane Permeability Assay (PAMPA) was used to measure the permeability coefficient (P e ) of codrug 4 through the artificial membrane to predict oral absorption. The carvacrol codrug 4 permeability was determined using PAMPA-GI protocols already developed [37]. Briefly, donor and acceptor plates were assembled to form a "sandwich" and incubated at room temperature for 18 h. P e can be calculated from the equation below: where P e is the effective permeability coefficient (cm × s -1 ), V D is the volume of the donor compartment (0.15 cm 3 ) and V A is the volume of the acceptor compartment (0.30 cm 3 ), A is the effective filter area (0.28 cm 2 ), t is the incubation time for the assay (s), [drug] acceptor is the concentration of the compound in the acceptor compartment at assay completion, and [drug] equilibrium is the concentration of the compound at theoretical equilibrium.

Results and Discussion
Synthesis of carvacrol codrugs 1-10 Carvacrol codrugs 1-10 were synthesized as outlined in Figs 2-4 employing solution phase procedures by elongation of the suitably protected Cys aminoacid chain in the C direction. Acetylation of aminic groups of cysteine derivatives 11-14, and 22 was performed as previously described [40] to obtain compounds 16-19, and 23 in good yields (Figs 2-3). Alkylation of commercial NAC (15) was performed using propargyl bromide in dry MeOH for 1 h at 60°C in basic conditions to afford the derivative 20 in good yield (Fig 1). The syntheses of the carvacrol codrugs 1-10 were successfully carried out by using N,N'dicyclohexylcarbodiimide (DCC), which proves to be an effective catalyst for the conversion of carboxylic acids to esters and amides (Figs 2-4). Particularly, compounds 16-20, 23-26 and 28 were treated with DCC in DMF/DCM for 1 h at room temperature; after the addition of carvacrol and DMAP, the mixture was left under stirring for 15 h at room temperature. The carvacrol codrugs 2-7 and 9-10 were obtained and purified on silica gel using DCM/AcOEt as eluant (yields 30-47%).
A further step was necessary to obtain the final compounds 1 and 8; in case of carvacrol codrug 1 the deprotection of the cysteine-SH group was reached when the corresponding protected precursor 21 was treated with TIPS, TFA in DCM for 48 h under nitrogen atmosphere at room temperature (yield 88%) (Fig 2); the removal of the Boc protecting group on 27 by treatment with TFA for 2 h at room temperature afforded the corresponding trifluoracetate 8 in quantitative yield (Fig 4). Before performing biological studies, all carvacrol codrugs 1-10 were fully characterized by 1 H-NMR, 13 C-NMR spectra and the purity was checked by HPLC analysis. A grade of purity higher than 98% after purification was obtained for all the final compounds.

Antimicrobial activity of carvacrol codrugs 1-10
In this study, the antimicrobial activity of carvacrol codrugs 1-10, carvacrol, NAC, and Ac-Cys (Allyl)-OH (the last three as reference compounds) was evaluated against pathogenic microorganisms and the results are outlined in Table 1.
When considering antimicrobial potency against Gram positive and Gram negative, carvacrol codrug 4 resulted the most active with bacteriostatic and bactericidal values equal to 2.5 mg/mL for all bacterial strains except for P. aeruginosa ATCC 9027, for which MIC and MBC values were 5 mg/mL and 10 mg/mL, respectively. We observed a lower antibacterial potency for all the other carvacrol codrugs (MIC values ranging from 2.5 mg/mL to 10 mg/mL) ( Table 1). We also tested the reference compounds: a) the antimicrobial effect exerted by NAC -MIC and MBC values ranged from 5 to 20 mg/mL for Gram positive and Gram negative tested bacteria-was similar to those reported by Aslam and Darouiche [41], suggesting that this amino acid may competitively inhibit bacterial utilization of cysteine or react via its sulfhydryl group with bacterial membranes; b) Ac-Cys(Allyl)-OH showed a better antimicrobial activity (MIC values from 2.5 to 10 mg/mL) respect to NAC probably due to the presence of the allyl moiety that might be a more offensive tool against bacteria, as previously reported [30]. In fact, the better antibacterial activity of Ac-Cys(Allyl)-OH compared to NAC is reflected in the corresponding carvacrol codrugs 4 and 1, respectively, suggesting that the presence of the allylcysteine moiety in 4 improves the antibacterial properties. However, compound 4 showed a weaker potency against P. aeruginosa confirming the much lower outer membrane     permeability than E. coli [42] probably due to the presence of efflux pumps that can transport anti-infective agents out of the periplasmatic space and to the reduced number and alteration of porins. In this study, the same set of ten carvacrol codrugs was also tested against C. albicans ATCC 10231. Most of them, including carvacrol, showed antifungal activity with MIC values ranging from 0.15 mg/mL to 5 mg/mL (Table 1). In particular, carvacrol codrugs 6, 1, 8 displayed fungicidal activity with MIC values equal to 0.15, 0.3, and 0.6 mg/mL, respectively. On the other hand, NAC was inactive against this fungal strain (MIC 20 mg/mL) while Ac-Cy(Allyl)-OH displayed antifungal activity with MIC and MFC equal to 2.5 mg/mL. However, the carvacrol codrug 1, containing the NAC moiety, showed a good antifungal activity while carvacrol codrug 4, containing the allyl-cysteine, was less active. This antifungal behavior is opposite to the antibacterial one for these carvacrol codrug 1 and 4. The antifungal activity is related to the presence and hydrophobicity of the carvacrol structure that is able to distribute into lipids of fungal membrane and interfere with its integrity causing fungal death. However, the overall lipophilicity of carvacrol codrugs 1 and 4 does not correlate with the results since codrug 1 is less hydrophobic (calculated LogP 2.66) than codrug 4 (calculated LogP 3.77) suggesting a potential involvement of the free thiol group in the antifungal activity of the carvacrol codrug. This consideration is in accordance with Bertling et al. [31] that support the hypothesis the thiol group is able to inactivate the gliotoxin produced by C. albicans by oxidation reactions. Also in the case of the best antifungal carvacrol codrug 6 (MIC 0.15 mg/mL) the hydrophobicity (calculated LogP 2.61) does not seem to be linked to the activity indicating that the presence of selenium might promote further interactions with the fungal membrane destabilizing it.
Antibiofilm effect of carvacrol codrug 4. BIC and BEC for carvacrol, Ac-Cys(Allyl)-OH, and the most active carvacrol codrug 4 were determined for all studied microorganisms ( Table 2). BIC and BEC values of the three compounds for S. aureus ATCC 29213 were twofold (2MIC) or four-fold (4MIC) greater than the concentration required to inhibit growth in suspension (MIC value); for S. epidermidis ATCC 35984 BIC and BEC values for carvacrol codrug 4 were 2MIC and 8MIC, equal to MIC and 2MIC for carvacrol, and 4MIC for Ac-Cys (Allyl)-OH; for E. coli ATCC 8739, Ac-Cys(allyl)-OH showed BIC and BEC values of 2MIC and 8MIC, while codrug 4 and carvacrol showed higher values: 8MIC, 16MIC, and 64MIC, respectively. BIC and BEC values of codrug 4, carvacrol, and Ac-Cys(Allyl)-OH coincided with MIC and 2MIC values for P. aeruginosa ATCC 9027. As shown in Table 2, for C. albicans ATCC 10231 BIC and BEC values of codrug 4 were 8MIC, for carvacrol these coincided with MIC and MFC, and for Ac-Cys(Allyl)-OH BIC and BEC were 2MIC.
Our results outlined that carvacrol possesses antimicrobial and antibiofilm properties but, except for S. epidermidis ATCC 35984 and C. albicans ATCC 10231, did not show an interesting and considerable effect on bacterial sessile phase respect to the planktonic one. Escherichia coli ATCC 8739 mature biofilm was more affected by Ac-Cys(Allyl)-OH and codrug 4 than carvacrol, suggesting that carvacrol is able to reduce bacterial biofilm formation interfering with the quorum sensing signaling mechanism but not with the mature biofilm, as reported by Burt et al. [43]. Probably, carvacrol alone is unable to penetrate the microbial biofilm matrix of E. coli while the conjugation to Ac-Cys(Allyl)-OH renders the codrug 4 able to promote the permeabilization and destabilization of the bacterial membrane. Hemolytic Activity. In vitro hemolytic assay represents a suitable screening tool to verify the in vivo toxicity to host cells. In the present study, the hemolytic activities of codrugs 1-10, carvacrol and NAC, as reference compounds, were evaluated (Fig 5). All analyzed codrugs produced human blood hemolysis below 50% at their MIC values, except for codrugs 8 and 9 which showed hemolytic activity values over 50%. In particular, carvacrol codrugs 7 and 10 hemolytic activities were never more than 12% at the tested concentrations; codrug 1 never expressed hemolytic activity higher than 40% and for carvacrol codrugs 2, 3, 5, and 6 the hemolytic activities were more than 50% only over 20 mg/mL of concentration. The carvacrol codrugs 4, 8, and 9 displayed hemolytic activity over 50% at 5 mg/mL, 3 mg/mL, and 10 mg/ mL, respectively. These data correlate with the hydrophobicity parameters: the carvacrol codrugs 8 and 9, showing the highest LogP values, resulted the most toxic compounds. Furthermore, carvacrol codrug 8 was obtained as trifluoracetate salt, which is substantially more toxic than hydrochloride salt even though they are both able to cross the intestinal membranes [44].
In particular, the hemolytic activity of carvacrol codrug 4 was evaluated in a wider range of values between 6.02-2.96 mg/mL to verify if the hemolysis occurs around the MIC value (2.5 mg/mL) for S. aureus ATCC 29213, S. epidermidis ATCC 35984, and E. coli ATCC 8739. Results showed that codrug 4 resulted not toxic at the MIC values for S. aureus ATCC 29213, S. epidermidis ATCC 35984, E. coli ATCC 8739, and C. albicans ATCC 1023.
Transmission Electron Microscopy of carvacrol codrug 4. To elucidate the physiological effects of carvacrol codrug 4 against S. aureus, E. coli, and C. albicans transmission electron microscopy was utilized. As illustrated in Fig 6, the micrographs clearly demonstrated that the growth of S. aureus, E. coli, and C. albicans in media containing carvacrol codrug 4 at 12.5 mg/ mL for 3 h generated profound changes in cell morphology [45]. We found that carvacrol codrug 4 had no effect on the envelope in S. aureus, but induced abnormal septum formation with irregular features (Fig 6B, arrows) probably with the same mechanism of action of some antibiotics such as penicillin [46], erythromycin [47], and vancomycin [48]. On the contrary, E. coli treated cells were evidently affected on the cell wall: alteration of outer membrane ( Fig  3D), formation of membrane blebbing with numerous electron-dense bubbles protruding from the cell surface (Fig 6D-6c and 6b) and electron-dense granules internalized into the cytoplasm were observed (Fig 6D-d). In C. albicans we observed the disintegration of the membranes ( Fig  6F) confirming that the carvacrol codrug 4 hydrophobicity allows it to interact with the fungal cell membrane and to interfere with its integrity. In fact, compound 4 could distribute into lipids of fungal membranes, rendering them more permeable and damaging their integrity, consequently causing mycelia death [49].

Pharmacokinetic studies of carvacrol codrug 4
The pharmacokinetic parameters were studied only for carvacrol codrug 4 since it resulted the compound with the best antimicrobial activity.
Solubility is one of the most important parameters to reach the desired concentration of drug in systemic circulation for achieving required pharmacological response. Table 3 shows the water solubility of codrug 4 and its lipophilicity (LogP and cLogP). Our data revealed that codrug 4 has a low water solubility (0.15 mg/mL) and a cLogP value (3.77) similar to carvacrol (3.35), suggesting that this hydrophobic character could allow codrug 4 to interact with the bacterial membrane just like carvacrol.
The precondition of codrug strategy is that the codrug can be converted to the parent drugs to exert the pharmacological effect. For this purpose, the chemical and enzymatic stabilities of carvacrol codrug 4 were evaluated in three buffers (pH 1.3, 5.0, and 7.4), in simulated fluids (SGF and SIF) with different concentrations (10 mg/mL and 40 mg/mL) of enzymes (pepsin and pancreatin), and in human and rat plasma. As reported in Table 4, codrug 4 resulted more stable at pH 1.3 (t ½ = 75.38 h) and pH 5.0 (t ½ = 62.08 h), respect to pH 7.4 (t ½ = 22.9 h). These values suggest that codrug 4 is able to pass unhydrolyzed the g.i. tract. In the presence of increasing concentrations of enzymes, the rate of hydrolysis of codrug 4 was significantly faster at pH 7.4 than 1.3; in fact, in the presence of pepsin (10 mg/mL) at pH 1.3, codrug 4 resulted  stable with a t ½ higher than 74 h (Table 4) (Table 4).
These data evidenced that codrug 4 resulted 13-fold more stable at pH 1.3 respect to pH 7.4 in the presence of pepsin and pancreatin (10 mg/mL), suggesting that the hydrolysis of ester linkage is preferred at physiological pH (Table 4). Codrug 4 showed the same rate of hydrolysis in 80% human and rat plasma (t ½ = 0.50 h) suggesting that both plasma may have more selective enzymes for the hydrolysis of ester bond (Table 4). Taken together, these results point out that codrug 4 is quite stable to potentially cross unmodified the acidic environment of the stomach, to be absorbed intact from the intestine after oral administration, and able to release carvacrol and Ac-Cys(Allyl)-OH after enzymatic hydrolysis.
Depending on the phospholipid type, PAMPA can mimic different adsorption/permeation environments. For predicting the ability of carvacrol codrug 4 to diffuse through the gastrointestinal tract, lecithin in dodecane has been used for g.i. permeation studies (PAMPA-GI) ( Table 5). To carefully reproduce the g.i. tract, the permeability of codrug 4 was determined at pH 5.0, 6.5, and 7.4. Since our carvacrol codrug 4 has a low solubility in the conventional solution used for the PAMPA, permeability tests were performed in the presence of cosolvents (2% of Tween 80 and 10% of MeOH). The employed cosolvents did not change the permeability of the phospholipid layer at the investigated concentrations, as previously reported [37]. After 18 h of incubation, codrug 4 showed good permeability through PAMPA-GI membrane (P e > 1 x 10 -6 cm/s) at each value of pH, thus suggesting a good absorption along g.i. tract.

Conclusion
In summary, we synthesized ten carvacrol codrugs with the aim of developing innovative antimicrobial compounds. However, Ac-Cys(Allyl)-CAR (4) resulted the most active compound against S. aureus, S. epidermidis, and E. coli with MIC values of 2.5 mg/mL and endowed with low toxicity. Even though antibacterial activity of Ac-Cys(Allyl)-CAR is not particularly good, pharmacokinetic data evidenced a good stability at pH 1.3 respect to pH 7.4 in the presence of pepsin and pancreatin, suggesting that, after absorption it might be able to release carvacrol and Ac-Cys(Allyl)-OH due to the enzymatic activity. Further experiments will be necessary to evaluate the in vivo antimicrobial activity of Ac-Cys(Allyl)-CAR and assess the real hydrolysis of the ester bond that would guarantee the release of the free hydroxyl group of carvacrol able to form hydrogen bonds and act as a transmembrane carrier of monovalent cations through the bacterial cytoplasmatic membrane.
In conclusion, new approaches like codrug strategy applied to natural products (essential oils), could be useful in the antimicrobial plan to overcome the issue of the antibiotic resistance.