Chemical composition, antioxidant effects and antimicrobial activities of some spices’ essential oils on food pathogenic bacteria

Thymus vulgaris, Cinnamomum zeylanicum and Ocimum gratissimum are spices widely used as aroma enhancers and food preservatives. This work assessed the chemical composition, antioxidant and antimicrobial effect of their essential oils on some food pathogenic bacteria, namely, Staphylococcus aureus, Citrobacter freundii, Enterobacter cloacae, Morganella morganii, Escherichia coli, Salmonella typhimurium, Proteus vulgaris and Shigella flexineri. After chemical analyses of the essential oils by gas chromatography and gas chromatography coupled to mass spectroscopy, the antimicrobial effects were subsequently assessed by disk and microdilution methods, while the antioxidant evaluations were performed by free radical scavenging activity. T. vulgaris essential oil composed of p-cymene (45.90%) and thymol (23.72%) which exhibited the highest inhibitory diameters of 20.33±0.58 and 18.00±1 mm, respectively, on the growth of S. aureus and C. freundii. O. gratissimum essential oil with thymol as major compound (47.11%) was more active to inhibit the growth of C. freundii and S. flexineri with respective inhibitory diameters of 18±1.73 and 16±2 mm. Essential oil from dry leaves of C. zeylanicum containing cinnamaldehyde (82.23%) and linalool (12.12%) was found to have the lowest values for minimal inhibitory concentration (MIC) and minimal bactericidal concentration (MBC) (≤3.53 mg/mL) considering the growth of S. flexineri, C. freundii and E. cloacae. Regarding the antioxidant effect, C. zeylanicum and T. vulgaris essential oil showed the most scavenging effect with half-maximal DPPH scavenging concentration (SC50) of 2.5 × 10 -5 and 6.5 × 10 -5 mg/ml, respectively. Their antioxidant effects were higher than conventionally used antioxidants in food products, butylhydroxyltoluene (BHT) and vitamin C.


INTRODUCTION
Spices are used as flavour and aroma in many processed and cooked food. Different amounts of spices are used by different populations all over the world. Chemical components mainly belonging to the essential oil (EO) fraction are responsible for the flavour and aroma of spices. Some of these compounds have medicinal properties and can exert bactericidal and bacteriostatic effects (Nwinyi et al., 2009). These antimicrobial effects are provided by the presence of peptides, alkaloids and EO compounds, which are the major components in these plants (Okigbo and Igwe, 2007). Associated to these antimicrobial effects, spices have been attributed antioxidant properties. In fact, these activities due to compounds that, when present at low concentrations compared to that of an oxidizing substrate, markedly delay or prevent its oxidation (Cavero et al., 2005). Hence, food or supplements containing spices or their compounds may contribute to alleviate the pathologies associated with the presence of reactive oxygen species (ROS) originating from oxidative processes in human cells (Nwinyi et al., 2009). The whole plant, part of it, flowers or fruits are generally used as spices. Added to this, the different agro-ecological environment where the spices are produced can influence the composition and therefore their biological activity. Applications of spices for their antimicrobial and antioxidant properties have been proposed by some authors (Hernández-Ochoa et al., 2014;Burt, 2004;Essia Ngang et al., 2014) and these properties have been related to the composition of the spices (Hyldgaard et al., 2012). In this perspective, it is important to evaluate the effectiveness of the antimicrobial and antioxidant properties of spices typical for certain locality, because of important differences in activity that may occur as a result of the different chemical composition. This is the rational that brought us to assess the chemical composition, the antimicrobial and antioxidant properties of EOs from Thymus vulgaris, Cinnamomum zeylanicum and Ocimum gratissimum.
T. vulgaris is the most widely cultivated species of the Laminaceae family (Alonso, 2004;Al-Bayati et al., 2008). It is used as spice and traditional drug for antispasmolytic, antibacterial and antifungal illnesses (Özgüven and Tansi, 1998). Regarding C. zeylanicum, many species are reported throughout the world, but the most used are those from Sri-Lanka (Richard and Multon, 1992). Its EO is rich in cinnamaldehyde and eugenol and has shown antimicrobial activities on food borne pathogens (Paranagana et al., 2003;Chang et al., 2001). On the other hand, C. zeylanicum that belongs to the family of Leguminocaeae, has its origin in Asia (Ndoye, 2001). In many parts of Africa, the plant is used to cure upper respiratory tract infections, diarrhoea, pneumonia, teeth, and gum disorder (Nwinyi et al., 2009).

Plant
Samples of T. vulgaris, O. gratissimum and C. zeylanicum were collected from spice sellers in Bafoussam (West region), Yaounde (Center region) and Lolodorf (South region) of Cameroon, respectively. After collection, the spices were identified by the Cameroon National Herbarium and a voucher specimen deposited with the following codes N° 25746/SRF/Cam, 5817/SRF/Cam, and 22309/SRF/Cam for T. vulgaris, O. gratissimum, and C. zeylanicum, respectively. For C. zeylanicum, based on the work of Richard and Multon (1992), who attributed these activities to different parts of the plant, leaves and stem barks were separated for the study.

Preparation of EOs
The EOs of the spices were extracted by hydro distillation using a Clevenger apparatus. Oils recovered were dried over anhydrous sodium sulphate and stored at +4°C until usage. The extraction yield was calculated as the ratio of oil recovered mass over the spices mass used multiplied by 100 (Ndoye, 2001).

Gas chromatography (GC)
GC analyses were performed on a Varian gas chromatograph, model CP-3380, with flame ionization detector containing two silica capillary columns: HP5 J&W Agilent (5%-Phenylmethylpolysiloxane) capillary column (30 m × 0.25 mm i.d. × 0.25 μm film) and Supelcowax 10 (polyethylene glycol) fused capillary column (30 m × 0.25 mm i.d. × 0.25 μm film); N 2 was the carrier gas at 0.8 ml/min; injection type 0.1 μl of pure sample, split ratio 1:100; injector temperature 220°C, detector temperature 250°C; temperature program 50 to 200°C at 5°C/min, then kept at 200°C for 10 min. The linear retention indices of the components were determined relative to the retention times of a series of n-alkanes.

Qualitative analysis
The identification of the constituents was based on comparison of their relative retention times and mass spectra with either that of authentic samples or with published data in the literature (Adams, 2007).
Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution License 4.0 International License tiphymurium, Proteus vulgaris, Enterobacter cloacae, Shigella flexneri, Morganella morganii, Escherichia coli, and Citrobacter freundii were obtained from the culture collection of the Microbiology Laboratory of the Department of Microbiology, University of Yaoundé I, Cameroon. Strains, originally stored at -80°C were sub-cultured twice in Brain Heart Infusion (BHI) at 37°C for 24 h and conserved on Muller Hinton slants at 4°C throughout the work. These strains were chosen for their implication in food borne diseases outbreaks. Prior to each test, a loop of the strain was inoculated in Mueller Hinton broth and incubated at 37°C for 24 h. Based on preliminary assessments, the overnight cultures were diluted in order to use about 10 6 cells/ml for the antimicrobial susceptibility test.

Antimicrobial activities of the EOs
The disk method was performed according to NCCL (2009) method with some adaptations. Briefly, 100 μl of an inoculum of 10 6 cells/ml were spread on sterile Mueller Hinton agar plates and let to dry at ambient temperature. 10 μl of the EO solubilised in 10% Tween 20 solution in order to obtain a concentration of 500 mg/ml were deposited on a 6 mm diameter disk and placed at the centre of the plate containing the microorganism. A disk containing 10 μl of 10% Tween 20 solution was used as control, while disk containing 10 μg of gentamicin was adopted as reference. Inhibition diameters (intended as the zone surrounding the disk with no bacterial growth) were measured after incubation at 37°C for 24 h on each of the three repetitions performed and expressed as mean ± standard deviation. The inhibition diameter was calculated as the difference between the test and the control.
For minimal inhibitory concentration (MIC) and minimal bactericidal concentration (MBC), microdilution method was used according to Bajpai et al. (2008) and Yèhouenou et al. (2010) in a 96 wells plate with single well covers. EOs diluted in 10% dimethyl sulfoxide (DMSO) were added to one of the series of wells containing the strain culture at 10 6 cells/ml and 0.02 mg/L of phenol red in order to have a concentration of 452.5 mg/L. Subsequently, series of 2 fold dilutions were prepared until a final concentration of the EO of 0.022 mg/L. A well containing the culture and phenol red without EO was used as positive control, while another well without the strain was used as negative control. After 24 h of incubation at 37°C, wells where the phenol red had changed to yellow were considered as having microbial growth, hence, indicating that the EO concentration used was not active. The MIC was considered as the lowest concentration where the phenol red did not change the initial red colour. Gentamicin used as reference antimicrobial was tested with the same methodology up to a concentration of 1600 μg/L.
For the MBC, 100 μl of the wells without visible bacterial growth were inoculated in a fresh Mueller Hinton broth and incubated at 37°C for 24 h. Following this, the lowest concentration of the EO which did not permit any growth was considered as MBC. Results for the MIC and MBC were reported as mean of the three repetitions and their standard deviation.

Free radical scavenging activity determination
Stable 2,2-diphenyl-1-picrylhydrazyl radical (DPPH) was used for the determination of the free radical scavenging activity of the EOs according to Brand-William et al. (1995) adapted to EO as described by Agnaniet et al. (2004) SC 50 (concentration of the tested substance that provides 50% scavenging activity) was determined graphically on the best curve fitting the experimental points representing the percentage scavenging activities as a function of the antioxidant concentration.
The data was represented as mean of the three replicates and their standard deviation.

Antibacterial activities of EOs
Disk diffusion method was used to assess the inhibitory leaves EOs against the microorganisms tested. In general, the EOs that were tested at 500 mg/L (this means 50 μg of the EO deposited on the disks) gave lower or comparable inhibition zone compared to those obtained with gentamicin whose disk contained 10 μg of the antibiotics. The MIC and MBC of EOs and that of gentamicin used as reference were assessed by the microdilution method. The results presented in Table 3 are expressed in mg/L for the EOs and μg/L for gentamicin. Due to the results obtained with the disk method, C. zeylanicum fresh leaves were not tested since it was demonstrated to have the same composition and antimicrobial potential as C. zeylanicum dry leaves. T. vulgaris and O. gratissimum EOs showed the most potent antimicrobial activity. In fact, it was possible to obtain a MIC and MBC with the EOs of these spices on all the microorganisms tested except on E. Cloacae, which were not sensitive to T. vulgaris up to 452.5 mg/ml. Gentamicin chosen as reference antibacterial compound was more active than the EOs. The MIC and MBC reported in Table 3 were the lowest when S. aureus was exposed to it (to what? gentamicin or EO; according to Table 3 it should be gentamicin). Among the Gram negative bacteria, S. typhimurium and M. morgani were the most sensitive to this antibiotic. Regarding the EOs, C. zeylanicum dry leaves EO was the most active to inhibit the growth of S. flexneri, C. freundii and E. cloacae than the other EOs tested. C. zeylanicum dry stem bark EO was the less active EO in general. Moreover, C. zeylanicum EOs, no matter the part chosen were not active on S. typhimurium and E. coli up to 452.5 mg/L.

Antioxidant properties of the EOs
The antioxidant properties of the EOs were assessed using the DPPH scavenging method. Table 4 represents the DPPH scavenging percentages obtained with EOs and the reference antioxidants BHT and vitamin C. From these data, a comparison index, namely, the SC 50 was obtained as described in material and methods and also reported in the same table. The higher the SC 50 , the lower the scavenging activity of the substance. It can be observed that C. zeylanicum EOs obtained from dry and fresh leaves had the best scavenging properties (2.5 × 10 -5 mg/ml) even compared to the reference compounds.
T. vulgaris EO also demonstrated a good antioxidant property as its SC 50 was 6.5 × 10 -5 mg/ml. These values are lower than that of BHT and vitamin C which exhibited SC 50 of 245 × 10 -5 and 11 × 10 -5 mg/ml, respectively. In general, among the substances tested, BHT was the less efficient in scavenging the DPPH radicals.

DISCUSSION
Testing and proving of biological activity of EOs is one of the major aims of the study of natural substances. It is now accepted by all the scientific community that these studies should pay attention to the variability of the Where, Conc. = concentration of essential oils, BHT or Vitamin C in mg/ml; DS% = %DPPH scavenging activity, and SC 50 = the concentration necessary for the scavenging of 50% DPPH radical from solution.
biological material concerned. In fact, it is known that the composition of EOs can be different due to the species, the agro-ecological factors and the part of the plant that is being analysed (Bruneton, 1999;Brada et al., 2007). In this specific study, the EOs tested showed different composition considering the species. Within the same species, dry and fresh leaves EOs of C. zeylanicum had roughly the same composition, but different from that of dry stems. The compositions of the EOs of these different parts of C. zeylanicum were also different from that of the same plant studied in Asia and Mid Orient. Richard and Multon (1992) and Cuvelier (1997) observed that EOs of leaves of C. zeylanicum had eugenol and campheme as their major components whereas the barks were rich in cinnamaldehydes. Later on, Yèhonenou et al. (2010) when analysing EOs of leaves of the same plant reported similar composition to that of this work, confirming that the agro-ecological factors can be determinant to the composition of EOs of the same species. To support these findings, in some countries C. zeylanicum is a big tree while those harvested in Cameroon are very small plants with a stem that cannot be easily dissociated from the bark. Regarding T. vulgaris EO, the composition obtained in this work indicated that the major components were pcymene and thymol. These results are similar to those of Moghtader et al. (2012), but differ in the relative percentages of the compounds. In Moghtader et al. (2012), thymol was the major component followed by pcymene, which is opposite to the findings in this work. Many chemo-types of T. vulgaris have been identified in literature. Thymol, carvacrol, thymol and even borneol were identified as chemo-types for T. vulgaris EOs collected from different parts of Morocco (Chebli et al., 2003). Few years ago, Imelouane et al. (2009) discovered a camphene chemo-type in Eastern Morocco. The T. vulgaris from Bafoussam, Cameroon, can be considered as a p-cymene chemo-type.
O. gratissimum EO analysed in this work had thymol as the major compound and also contained p-cymene and ɣ-terpinene. This is in accordance with the conclusions of Orwa et al. (2009) who observed that eugenol, thymol and carvacrol are the major compounds generally found in the EO of this plant.
The antimicrobial properties of the EOs tested in this work were assessed by disk diffusion and microdilution methods. The first method can be considered as an explorative assessment that helps in a preliminary screening. In fact, this permitted to assess that dry leaves and fresh leaves EOs of C. zeylanicum had the same activity and hence only the dry leaves EOs were used to assess the MIC and MBC. EO from the leaves of C. zeylanicum was the most active and that of the dry stem was the less active, in particular on S. aureus, S. typhimurium and E. coli. In fact, this EO contains 68% of α-terpinene, 10% of carvacrol and 10.4% of thymol. Among these compounds, carvacrol and thymol are mostly endowed with antimicrobial properties (Burt, 2004;Chiasson et al., 2005). The high content in cinnamaldehyde (>80%) of the leaves EO of this plant is probably the main reason of its antimicrobial activity. As a matter of fact, cinnamaldehyde antimicrobial properties according to Walsh et al. (2003) acts on bacteria by inhibiting the biosynthetic enzymes and can also, according to Oussalah et al. (2006) contribute in the reduction of bacteria intracellular pH. Hadri et al. (2014) observed that EO from C. zeylanicum barks with 91% cinnamaldehyde inhibited the growth of S. aureus and E. coli, while this was not the case in our study with an EO containing more than 80% of the same compound. These different results could be explained by the additional 8.5% of cinnamaldehyde acetate of the EO or by the strain variability.
T. vulgaris did not show antimicrobial activity on the growth of E. cloacae. The antimicrobial activity of this EO has been generally attributed to the high content in thymol and carvacrol (Sokovic et al., 2007;Khodaei Motlagh et al., 2014). In our study, thymol was present at 23% and p-cymene at 46%, while carvacrol was only 2%. The synergy between thymol and carvacrol may be the base of the antimicrobial properties of this EO. Knowing that thymol is a more potent antimicrobial than p-cymene, the inversion of the percentage contains those of these compounds in O. gratissimum EO (thymol 47% and pcymene 14%) gave way to a more important antimicrobial activity. It was observed that MIC and MBC of O. gratissimum EO were generally lower than those obtained with T. vulgaris for the same microorganism. Beside the fact that antimicrobial activities of thymol are well known, the interaction of all the molecules can have a synergic effect that may potentiate the activity of the whole EO. In this regard, a synergic activity has been reported for many compounds (Jayaprakasha et al., 2003;Matasyoh et al., 2007). The EOs activities are attributed to their disorganising effect of the molecules on the cell membrane, leading to a non control of the exchange between cytoplasm and the external medium (Bouhdid et al., 2009;Gardini et al., 2009). After entering into the microorganism's cytoplasm, other mode of action could be the acidification (Wright, 2002) and the destruction of the genetic material (Brackman et al., 2008;De Martino et al., 2009).
The antioxidant activity of the EOs was also assessed and compared to BHT, a primary antioxidant with the ability to inactivate free radicals by transferring a proton (Cuvelier, 1997) and vitamin C as a secondary antioxidant that blocks oxygen and metals that can faster the oxidation (Cuvelier, 1997). In the case of this study, C. zeylanicum leaves EO was the most active in scavenging the DPPH followed by the EO of T. vulgaris. This may be due to the presence of linalool in larger extent than that of cinnamaldehyde. Although an in vivo antioxidant property have been reported for cinnamaldehyde in rat kidney (Gowder and Devaroy, 2006), this aromatic aldehyde has a low proton transferring property than linalool. On the other hand, thymol's percentage is higher in O. gratissimum EO compared to T. vulgaricus. The latter gave the best scavenging properties. The presence of 4% linalool and 45% p-cymene may be the reason for this activity. De Oliveira et al. (2015) observed that p-cymene has an antioxidant potential in vivo as well as a neuroprotective agent in the brain. The reason behind the good scavenging performances of the EOs with respect to the reference compounds may be the interactivity of all the compounds composing the EOs. It can be observed from these results that these EOs can well replace BHT in the limitation of oxidation in food and tissues.

Conclusion
EOs of T. vulgaris, O. gratissimum and C. zeylanicum showed a high composition variability depending on the species, and within the same species of C. zeylanicum, between leaves and stem. This variability was also highly influenced by the agro-ecological factors as demonstrated by comparison of the results with those in the literature. Moreover, the studied antimicrobial properties were of large spectra for T. vulgaris and O. gratissimum EOs, while that of C. zeylanicum leaves was more active to inhibit the growth of S. flexneri and C. freudii than the other EOs. These EOs demonstrated to be good natural antioxidants that may be used in food industries and in the prevention of oxidative degenerative diseases.