In Vitro and In Vivo Comparison of Changes in Antibiotics Susceptibility of E. coli and Chicken's Intestinal Flora after Exposure to Amoxicillin or Thymol

This study aims at verifying, in vitro, the extent to which the use of amoxicillin or thymol induces the selection of resistant bacteria and at evaluating in vivo their effects on the development of antimicrobial resistance in the intestinal flora of poultry. E. coli strain was subcultured on agar plates containing increasing concentrations of either amoxicillin or thymol. Thereafter, minimal inhibitory concentrations (MICs) of thymol, amoxicillin, and two other antibiotics, tylosin and colistin, were determined using the microdilution method. Groups of chicks were subjected to a 2-week regime of either amoxicillin or thymol added to their drinking water. During the treatment with either thymol or amoxicillin, the total aerobic mesophilic flora (TAMF) was counted on thymol-gradient plates or amoxicillin-gradient plates and the MICs of antibiotics and thymol for E. coli isolates were determined. The in vitro test showed that for E. coli, which had been serially subcultured on increasing concentrations of amoxicillin, a 32-fold increase in MIC values for amoxicillin and a 4-fold increase for colistin and tylosin were noted. However, the MIC of thymol for this strain remained constant. For the E. coli, which had been serially subcultured on increasing concentrations of thymol, no change in the MIC values for antibiotics and thymol was observed. The in vivo test confirmed the in vitro one. It demonstrated that exposure to amoxicillin induced a selection of antimicrobial resistance in TAMF and intestinal E. coli, whereas exposure to thymol did not. The results showed that the group receiving thymol had a lower consumption index compared to the other groups. This study demonstrates the feasibility of this natural product as an alternative solution to the current use of antibiotics in poultry farming.


Introduction
For decades, antibiotics have been used in poultry as growth promoters [1]. Van Boeckel et al. [2] estimate that by 2030, a total of 105,596 (±3605) tons of antibiotics will be consumed in feed animal production globally. Industry researchers assert that antimicrobial growth promoters are essential to sustaining increases in productivity and contribute to the lowering of the cost of chicken products for consumers [3]. However, the use of antibiotics as growth promoters increases the risk of the development of antimicrobial resistance [4,5]. Such use induces the selection of multidrug-resistant bacteria, which in turn reduces the efficacy of antibiotic therapy in both animals and humans that are colonized with resistant bacteria [1,6]. In fact, this widespread use of antimicrobials in livestock contributes to the emergence of antimicrobial-resistant bacteria and has significant public health implications: antimicrobial-resistant bacteria of animal origin can be transmitted to humans through the environment and food products (and to agricultural workers by direct contact [2]. It is for this reason that the European Union banned the use of antibiotics as growth promoters in 2006 [7,8]. Our laboratory has demonstrated the antimicrobial activity of essential oils and their major compounds [9][10][11][12]. Among the various constituents of EOs, thymol, the major component of the essential oil of thyme and oregano, has been shown to have an antibacterial effect on several bacterial species [10,[13][14][15]. ese results suggest that this substance could be used as an alternative to antibiotics for poultry. is study aims to verify, in vitro and in vivo, the extent to which the use of antibiotics induces the selection of resistant bacteria and to compare its effect to thymol.
is phenolic major compound was dispersed in a 0.2% sterile agar suspension [16]. ree antibiotics were used: amoxicillin, colistin, and tylosin. ey were purchased from Sigma-Aldrich (France). ese antibiotics were dispersed in distilled sterile water.

Determination of MIC.
e MICs were determined by microdilution assays in 96-well plates conforming to the standards of the CLSI [17]. Ten concentrations of each agent were prepared in sterile tubes. ey were carried out by successive dilutions 1/2 in Mueller Hinton broth for antibiotics and in MH broth containing agar at 0.2% for thymol. 20 μl of each concentration was then added to each well containing 160 μl of MHB. Bacterial suspensions were prepared by taking colonies from 24 h cultures on TSA plates. e colonies were suspended in a sterile 0.9% aqueous solution of NaCl. e density was adjusted to the turbidity of a 0.5 McFarland Standard (10 8 colony-forming unit (CFU/mL)) [9]. ese suspensions were diluted in MH broth and plated in 96-well plates at a density of 5 × 10 5 CFU/well. After the plates were incubated at 37°C for 18 h, 40 μl of 0.5% triphenyltetrazolium chloride (TTC) was added to each well. After two hours of incubation, the MIC corresponded to the lowest concentration that does not produce a red color [9].

Animals and Breeding Conditions.
e chicks used in this study were a day old (approximately 40 g). ey were divided into groups of ten and housed in separate cages. e photoperiod was adjusted on a daily basis to 12 hours of light and 12 hours of darkness. At the beginning of the experiment, the ambient temperature was 32°C. It was reduced by 2 to 3°C each week to finally reach 23°C at the end of the experiment. Chicks were given ad libitum access to feed and water. ey received a maize-based feed diet that was free of antibiotics and antiparasitics.

Treatment in Drinking Water
(i) ymol is the active principle of NP ® (15% of thymol), produced by the Industrial Laboratory of Veterinary Alternatives (LIAV) in Morocco. In addition to thymol, the NP contains other excipients that provide stability and solubility. In industrial poultry farms, NP ® is administered orally in drinking water at a rate of 1 g/L/day from the first day to 40 th day of chicken's age.
(ii) Amoxicillin: Amoxy 70 ® , in powder form at a concentration of 700 mg/g, was purchased from Novovet, Casablanca, Morocco. In industrial poultry farms, it is administered orally in drinking water at a rate of 60 mg/L of body weight/day from the first day to the fifth day. e animals were randomly divided into three experimental groups of 10 chicks each: Group 1 (n � 10): animals receiving 48 mg/L of amoxicillin in drinking water Group 2 (n � 10): animals receiving 1 g/L of NP in drinking water (equivalent to 0.15 g/L of thymol) Group 3 (n � 10) control group: animals receiving drinking water Treatment with amoxicillin and with NP lasted 15 days (from day 7 to day 21 of the chick's age). e antimicrobial resistance of TAMF was evaluated on days 7, 14, and 21 by counting on plates containing a linear gradient from peak to trough antimicrobial concentrations of either amoxicillin or thymol. e evaluation of the antimicrobial resistance in E. coli isolates was performed by the evaluation of the MIC values of three antibiotics (amoxicillin, colistin, and tylosin) and thymol. Once a week (on days 7, 14, and 21), 1 g of fresh feces sample from each group was collected and solubilized in 9 ml of physiological serum, and then dilutions were made.

Antibiotic Gradient Plates.
e gradient plates were prepared as described by De Vecchi et al. [18]. Gradients were prepared in Petri dishes, on which two layers of Plate Count Agar (PCA, Biokar ® ) were poured. e bottom layer consisted of Plate Count Agar containing either amoxicillin or thymol at a maximum concentration (C max ) allowed to harden with the plate slanted sufficiently to cover the entire bottom.
e top layer, added to the dish in the normal position, did not contain any amoxicillin or thymol. For amoxicillin, three gradients were prepared from three maximum concentrations: 3, 6, and 12 μg/mL. For thymol, a maximum concentration of 500 μg/mL was used. 100 μl of the diluted feces samples were homogeneously spread onto 2 Veterinary Medicine International each plate and incubated at 37°C for 24 h. e gradient plates were analyzed by counting the colonies growing on 4 parts of each plate: from 0 to 25% C max , from 25% C max to 50% C max , from 50% C max to 75% C max , and from 75% C max to C max as shown in Figure 1.  [19]. ree E. coli strains were picked randomly each week.

Evaluation of the Antimicrobial
e MICs of amoxicillin, colistin, tylosin, and thymol were then determined by the microdilution method as previously described.

Performance Parameters.
e impact of different treatments on the following zootechnical parameters was evaluated: body weight, weight gain, feed intake, and the consumption index (CI) [20].

Statistical Analysis.
e results were expressed as mean values ± SEM (standard error of the mean). In order to compare the three groups where the independent variables were the number of TAMF, body weight, or body weight gain and the dependent variable was time, a one-way analysis of variance followed by Tukeyʼs multiple comparison test (ANOVA followed by Tukeyʼs test) was performed using Graph Pad Prism software, version 5.03. Differences of p < 0.05 were considered statistically significant. Table 1 reports the MIC values of thymol and the three antibiotics: amoxicillin, tylosin, and colistin before and after seven passages of E. coli ATCC25922 in the amoxicillin-containing plates. e results show that before subculture, the E. coli ATCC 25922 was found to be relatively sensitive to low concentrations of antibiotics. However, after seven passages in amoxicillin, a general increase in the MICs was observed for the three antibiotics.

Effect of Amoxicillin on In Vitro Susceptibility of E. coli ATCC 25922 to Antibiotics.
e MIC values for amoxicillin increased 32-fold. e MIC values for tylosin and colistin increased 4-fold. In the case of the thymol, no increase in the MIC was observed; it remained constant at 250 μg/mL.

Effect of ymol on In Vitro
Susceptibility of E. coli ATCC 25922 to Antibiotics. When E. coli ATCC 25922 was subcultured on increasing concentrations of thymol, no growth developed on concentrations more than 300 µg/ mL. After the subculturing onto thymol's concentrations of 100, 200, and 300 µg/mL, the evaluation of the MIC values of the three antibiotics and the thymol did not show any increase in the MIC values for neither thymol nor the three antibiotics tested. e respective MICs of amoxicillin, tylosin, and colistin were of 0.8 μg/mL, 1.6 μg/mL, and 0.16 μg/mL. e MIC of thymol also remained constant at 250 μg/mL.

Evaluation of the Antimicrobial Resistance to Amoxicillin in TAMF.
To evaluate the effects of exposure to amoxicillin and thymol on the development of antimicrobial resistance in total aerobic mesophilic flora, groups of 10 chicks each were exposed to a 2-week course of amoxicillin or NP (thymol) added to the drinking water at a dose of 48 mg/L and 1 g/L, respectively. In the control group, chicks did not receive any form of drug. Samples of feces from different groups were taken each week (day 7, day 14, and day 21) and diluted. en, 100 μl of the diluted feces samples were spread on amoxicillin-gradient plates prepared as previously described. e total number of strains grown on amoxicillingradient plates is reported in Tables 2-4.
On day 7 (before treatment), and for the three groups of animals, there is a TAMF growth of 3.10 5 CFU/g on an amoxicillin concentration less than 0.75 μg/mL ( Table 2). For Group 1 that received amoxicillin, after a one-week treatment (day 14), the growth of TAMF of 10 5 CFU/g was noted on an amoxicillin concentration of 3 μg/mL. On day 21, the growth was noted over the entire surface of amoxicillin-gradient plates of the three maximum concentrations tested 3 μg/mL (Table 2), 6 μg/mL (Table 3), and 12 μg/mL (Table 4). For Group 2 treated with NP (day 7 to day 21), a TAMF growth around 10 3 CFU g −1 was noted on an

Evaluation of the Antimicrobial Resistance to ymol in TAMF.
e total number of strains grown on thymolgradient plates is reported in Table 5. It shows that on day 7 (before treatment), no growth was detected on the thymolgradient plates. On days 14 and 21, TAMF growth is noted on a thymol concentration lower than 250 µg/mL. It was noted in the three groups.  Figure 2 shows the evolution in time of the body weight and the body weight gain of the different groups. e results show that from the second week, the chicks' body weight of Group 1 and Group 2 began to differ significantly (p < 0.05) from that of the control group (Figure 2(a)). e body weight gain of the poultry in the two treated groups was found to be significantly (p < 0.05) higher than that of Group 3 (Figure 2(b)).

Feed Intake and Consumption
Index. e evolution in the period of time for the feed intake and the consumption index of the different groups is shown in Table 7.
roughout the experiment, the control group showed the highest consumption index.

Discussion
In the present research study, an in vitro assessment of the effect of amoxicillin and thymol on antimicrobial resistance in a strain of E. coli ATCC 25922 was performed. e effects on the antimicrobial resistance of the intestinal flora of animals in vivo particularly the total mesophilic aerobic bacteria and intestinal E. coli were evaluated. e in vitro test: in order to demonstrate the effect of amoxicillin on the selection of resistance mechanisms in E. coli ATCC 25922, successive subcultures of this strain on increasing concentrations of amoxicillin were made. After  Amoxicillin-gradient (µg/mL) 0-3 3-6 6-9 9-12 0-3 3-6 6-9 9-12 0-3 3-6 6-9 9-12 is subcultured strain is considered to be resistant to amoxicillin (MIC > 8 μg/mL) according to EUCAST's recommendations [22]. ese results corroborate with those obtained by Cebrian et al. [23] who showed that the in vitro exposure of Salmonella strains to amoxicillin     Our results also show that E. coli strain became less sensitive to tylosin and colistin, while these two antibiotics were not added to the subculture medium. e MICs of tylosin and colistin increased by 4-fold compared to the starting MICs. is increase in MIC of colistin and tylosin can be explained by the existence of cross-resistance obtained between amoxicillin and the other two antibiotics.
ese results corroborate those obtained by Toprak et al. [26] who analyzed the evolution of resistance in E. coli under selection with chloramphenicol, doxycycline, and trimethoprim, and showed that after 20 days of culture with increasing concentrations of each antibiotic, the resistance levels increase dramatically. Moreover, the authors reported that the wholegenome sequencing of the evolved strains showed mutations specific to resistance to the antibiotic in use and resistance to multiple antibiotics (cross-resistance). As for the MIC of thymol, it did not change; it remained constant at 250 μg/mL after subculture on increasing concentrations of amoxicillin, which leads us to consider that the selection of amoxicillin resistance does not affect the sensitivity to thymol. In order to test the in vitro effect of thymol on the antimicrobial resistance in the same strain of E. coli, successive subcultures of E. coli ATCC 25922 on increasing concentrations of thymol were made. During this subculture, no growth was detected on the plates containing a concentration of thymol greater than 300 µg/mL. e determination of MIC values was carried out after the subculture on the three concentrations 100, 200, and 300 µg/mL. e results showed that thymol does not induce the resistance selection phenomenon neither to thymol itself nor to the tested antibiotics. Several studies have reported that the use of essential oils or their major compounds does not induce the selection of resistant strains [27]. Ohno et al. [28] tested the effect of 13 essential oils on Helicobacter pylori strain and showed that these EOs are bactericidal against H. pylori without inducing the selection of resistant bacteria. Gomes Neto et al. [29] have also shown that exposure of a strain of S. aureus to infrainhibitory concentrations of the essential oil of Rosmarinus officinalis and 1,8-cineole does not induce the selection of resistant strains to these two agents.
e in vivo test: to confirm the in vitro obtained results and to verify whether the use of either amoxicillin or thymol in the drinking water of the chicks will cause a selection of resistant bacteria in vivo, a test was carried out by treating groups of chicks with amoxicillin or NP which contains thymol as an active principle. en, samples of chick feces were collected to evaluate the sensitivity of total aerobic mesophilic flora and intestinal E. coli. e results of TAMF's counting on amoxicillin-gradient plates showed that following amoxicillin administration to the drinking water (day 7 to day 21), total aerobic mesophilic flora became less sensitive to amoxicillin; this is illustrated by the growth over the entire gradient surface of the three maximum concentrations of amoxicillin (3, 6, and 12 µg/mL) by the day 21, whereas before the addition of amoxicillin to the drinking water, growth was only observed at concentrations less than 0.75 µg/mL. However, for the group receiving NP, the growth of TAMF is noted just on concentrations of amoxicillin less than 1.5 µg/mL. ese results show that the exposure to amoxicillin exerts a selective pressure for the emergence of resistance in TAMF, whereas the addition of the NP does not induce this phenomenon. For the control group, on days 14 and 21, growth was observed on amoxicillin concentrations less than 2.25 µg/mL whereas, on day 7, growth was observed on concentrations less than 0.75 µg/mL. ese observations suggested that environmental sources including feed, water, and air may be the main factors in the colonization of the chicks' intestinal tracts by less sensitive bacteria [30,31]. As for the count of TAMF on thymol-gradient plates, the same results were noted for the three groups; no growth was noted on day 7, whereas growth was noted on a thymol concentration less than 250 µg/mL on days 14 and 21. e absence of growth on day 7 can be explained by the low bacterial load in the feces suspension prepared on day 7 or the absence of certain bacterial strains which subsequently colonize the intestine from the feed, the water, or the air [30].
After observing the effect of NP and amoxicillin on the intestinal TAMF, it was necessary to confirm this effect on intestinal E. coli isolates. e results obtained show a significant increase by 16-fold in MIC values of amoxicillin for E. coli isolated from the group that received amoxicillin. ese results are similar to those obtained by Van der Horst et al. who showed that the addition of amoxicillin, enrofloxacin, or oxytetracycline in the chicken's drinking water induced the selection for resistant E. coli [32]. Similarly, Miranda et al. and Simoneit et al. reported that the administration of amoxicillin or other antibiotics in drinking water induced the selection and development of antimicrobial resistance in E. coli strains from chickens [33,34]. Moreover, Pouwels et al. [35] reported that amoxicillin, which is mainly prescribed for human infections, is associated with increased resistance against various antibiotics is result confirms the presence of the phenomenon of cross-resistance. Concerning the control group, the MIC values for amoxicillin increased 2-fold, for tylosin 4-fold, and for colistin 8-fold by day 21, although this group did not receive any kind of drug. is can be explained by the spread of less sensitive strains from a contaminated environment. Strains of antimicrobial-resistant E. coli were isolated from the feed, water, and litter of chickens in India [36]. Da Costa et al. and Rossato et al. also reported that feed is a source of antimicrobial-resistant E. coli [37,38]. For the group receiving NP, the MIC values noted are lower compared to those noted in the control group, and an increase in the MIC by 2-fold for amoxicillin and tylosin and 4-fold for colistin was observed. Regarding the MIC of thymol, it was 250 µg/mL for the three groups. is confirms the results obtained in vitro which show the absence of development of thymol resistance.
To evaluate the effect of NP on the zootechnical performances of animals, we evaluated the body weight, the weight gain, the feed consumption, and the consumption index. e results obtained showed that the weight of the treated groups was significantly higher than that of the control group and that group receiving NP had the lowest consumption index. is can be explained by the reduction of the bacterial load by thymol, which affects intestinal integrity [39]. With a balanced intestinal flora, food may be more easily absorbed, which explains the difference in growth between animals. A study conducted by Lee et al. showed that thymol increases the activity of chicken's intestinal amylase which improves chicken growth by increasing the digestibility of nutrients and the regulation of the intestinal microflora [40]. Ezzak Abd El-Hack et al. [41] also demonstrated the valuable potential of thymol to enhance the growth performances, digestive enzyme activity, nutrient bioavailability, immunity, and general health of poultry. Suresh et al. [1] have also recommended the use of essential oils and their compounds as alternatives to antibiotics as it could minimize the risk of antibiotic resistance in livestock. erefore, thymol could represent a natural alternative to antibiotics currently used in poultry farming as reported by Yang et al. [42].

Conclusion
e results of these experiments lead us to conclude that thymol does not induce the selection of antimicrobial-resistant bacteria. However, it has a significantly positive effect on the zootechnical performance of the animals, making it a good, safe alternative that meets the breeders. is natural product could be an efficient alternative that would have positive effects on the fight for antimicrobial resistance that is observed in human medicine and that originates from the administering of antibiotics to farm animals.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that there are no conflicts of interest related to this research paper.