Using Rosemary Essential Oil as a Potential Natural Preservative during Stirred-like Yogurt Making

The popularity of rosemary has grown as a natural alternative over the synthetic supplements due to its potential health benefits. The rosemary plant has been utilized to preserve food due to its ability to prevent oxidation and microbial contamination. The reason for this study was to determine the phytochemical components and antimicrobial activity of rosemary essential oil (REO) and the effect of REO addition (0.5 and 0.7%) on the chemical, microbiological, and sensory properties of stirred-like yogurt (SLY) during 16 days of storage at 4 °C. The obtained data observed that REO exhibited antimicrobial action against Escherichia coli, Staphylococcus aureus, and Salmonella marcescens, as well as fungi (Aspergillus flavus) and yeasts (Candida albicans). Increased REO to 0.7% accelerated (p < 0.05) the development of lactic acid bacteria (LAB) in SLY (8.3 log cfu/g) and delayed yeast growth up to 12 days. Molds and coliforms were also not found in the SLY samples with REO. In comparison to control samples, sensory results showed that the addition of REO improves the overall acceptance of SLY (p < 0.05). In conclusion, the current study found that REO could be used as a natural preservative during the production of SLY to extend shelf-life and promote LAB development.


Introduction
Consumers are generally aware of healthful, nutritious food products with a longer shelf-life. In food products, microbiological, enzymatic, physical, and chemical changes occur, resulting in a loss of quality, nutritional value, and safety [1]. Using synthetic preservatives, such as benzoic acid, to prevent food deterioration results in a kerosene-like odor [2]. As a result, the necessity for natural plant-derived antimicrobial activity (Bio-preservatives) to replace these artificial preservatives has grown [3,4]. Natural antioxidant ingredients were added to dairy products to boost antioxidant activity and anti-inflammatory characteristics, as well as providing a variety of health benefits [5]. In fact, essential oils have antimicrobial properties against yeasts, molds, and bacteria [3,6].
Natural phenolic compounds that have been found in plants and vegetables may decrease the risk of some diseases because of their antioxidant and free radical inhibition potentials imparted by the benzene ring and the hydroxyl group in their structures [7,8]. Rosemary (Rosmarinus officialism L.) oil is used as a food seasoning for food [9]. Rosemary essential oil (REO) has traditionally and largely been used as a medicinal herb, with a number of properties, such as anti-inflammatory, analgesic, astringent, antimicrobial, antirheumatic, carminative, antifungal, and antioxidant [10]. Yogurt is a coagulated milk

Chemical Composition
Titratable acidity was determined (as lactic acid) according to Sadler and Murphy [18]. Total solids (TS) and total protein contents (TP) were determined according to the AOAC guidelines [19]. Fat content was determined by Gerber method according to Kleyn et al. [20].

Determination of Total Phenolic Compounds
Total phenolics content was determined using the method of Singleton et al., and the results are reported in mg of gallic acid equivalents (mg GAE/100 g sample) [21].

Determination of Total Flavonoids
The flavonoid content was measured as aluminum chloride colorimetric according to Marinova et al., and the results are represented as mg catechin equivalents (mg catechin/100 g sample) [22].

Gas Chromatography-Mass Spectrometry (GC-MS) Analysis
The GC-MS analysis of the essential oil samples was carried out at the Department of Analytical Chemistry, Faculty of Science, Assiut University, to determine the volatile and semi-volatile chemicals. The extract of rosemary was made with isopropanol and ethyl acetate. The GC-MS analysis was performed with a Thermo Scientific TM TRACE 1300 coupled to an ISQ-7000 and a Thermo Scientific TM TG-6MB 5 ms (30 m×0.250

Proximate Composition Analysis 2.4.1. Chemical Composition
Titratable acidity was determined (as lactic acid) according to Sadler and Murphy [18]. Total solids (TS) and total protein contents (TP) were determined according to the AOAC guidelines [19]. Fat content was determined by Gerber method according to Kleyn et al. [20].

Determination of Total Phenolic Compounds
Total phenolics content was determined using the method of Singleton et al., and the results are reported in mg of gallic acid equivalents (mg GAE/100 g sample) [21].

Determination of Total Flavonoids
The flavonoid content was measured as aluminum chloride colorimetric according to Marinova et al., and the results are represented as mg catechin equivalents (mg catechin/100 g sample) [22].

Gas Chromatography-Mass Spectrometry (GC-MS) Analysis
The GC-MS analysis of the essential oil samples was carried out at the Department of Analytical Chemistry, Faculty of Science, Assiut University, to determine the volatile and semi-volatile chemicals. The extract of rosemary was made with isopropanol and ethyl acetate. The GC-MS analysis was performed with a Thermo Scientific TM TRACE 1300 coupled to an ISQ-7000 and a Thermo Scientific TM TG-6MB 5 ms (30 m×0.250 mm×1.00 m) column from Thermo Scientific. The temperature in the GC oven was kept at 100 • C for 15 min, then increased to 150 • C at a rate of 10 • C/min, and then to 200 • C at a rate of 5 • C/min. Furthermore, the temperature was raised to 250 • C at a rate of 5 • C per minute, and then to 280 • C at a rate of 5 • C per minute. Helium gas was used as the carrier, with a flow rate of 0.5 mL/min. The mass spectrometer was set to electron ionization mode, with a temperature of 320 • C for the ion source and 280 • C for the MS transfer line. The NIST 17 mass spectrum library (mainlib) was used to identify volatiles, and the results are represented as a percentage of the total GC area. To prepare inoculum for bioassay, bacterial strains were individually cultured for 48 h in 100 mL conical flasks containing 30 mL nutrient broth medium. Fungi were grown for 7 days in 100 mL conical containing 30 mL Sabouraud's dextrose broth. Nutrient agar and Sabouraud's dextrose agar were, respectively, used for bacteria and fungi. After solidification of the media, 5 mm diameter cavities were cut in the solidified agar (4 cavities/plate) using sterile cork borer. After that REO was added (50 µL/cavity). Cultures were then incubated at 28 • C for 48 h in case of bacteria and up to 7 days in case of fungi. Results were read as the diameter (in mm) of inhibition zone around cavities.

•
Minimum inhibitory concentrations (MICs): To determine the minimum inhibitory concentrations (MICs), REO was diluted with dimethyl sulfoxide (DMSO) to prepare a series of concentrations (12.5, 25, 50 and 100%), which were pipetted into the cavities (50 µL/cavity) and similarly assayed as mentioned before and the least concentration (below which no activity) was recorded as the MIC.

Standard Plate Count Technique
Fresh and after 4, 8, 12, and 16 days, the microbiological analysis was performed by using the usual plate count procedure, a total bacterial count (TBC) was plated on nutrient agar medium, and enumeration was completed [23]; before microbiological enumeration, the plates were incubated at 32 • C for 48-72 h. The MRS agar medium was used to obtain lactic acid bacterial (LAB) counts, and plates were incubated at 37 • C for 48 h [24]. In MacConkey broth media, coliform bacteria were discovered, and tubes were cultured for 24 h at 32 ± 1 • C [25]. Yeasts were counted on malt extract agar medium [26], whereas mold counts were counted on potato dextrose agar media, and the plates were cultured for 5 days at 25 ± 1 • C [23].

Sensory Evaluation
Sensory analysis of SLY samples was delineated by Kamel et al.'s guidelines with certain adjustments [27]. Flavor (45 points), body and texture (40 points), color and appearance (15 points), and overall acceptability (100 points) were all tested fresh and after 4, 8, 12, and 16 days of storage.

Statistical Analysis
The influence of REO addition, time, or their interaction on the features of SLY samples was explored statistically using R software (R x 643.3.3, Vienna, Austria) by ANOVA, testing a GLM for each variable. When significant differences were discovered at p < 0.05, the least significant difference (LSD) comparison test was used to separate the means.

Phytochemical Components in REO
Total phenolics and total flavonoids: The total phenolic content in REO was recorded as 203.6 mg/kg, whereas total flavonoids concentration in our study was 488.98 mg catechin/100 g sample.

Antimicrobial Activity of REO Minimum Inhibitory Concentration (MICs):
The diameter of inhibition zones was measured and taken as an indicator of the antimicrobial effect, as shown in Table 2. The range of inhibition zone diameter was varied from 0.0 to 14 mm in the studied bacteria. The MICs of REO were tested against Escherichia coli, Salmonella marcescens, and Staphylococcus aureus, and we found that adding 25% of REO gave the lowest inhibitory effect, with a 6 mm diameter inhibition zone. For 50% of REO extracts, the zones ranged from 7 to 8 mm. The data in the same table also indicate that the inhibition zone for REO (100% concentration) recorded 7 and 9 mm against Aspergillus flavus and Candida albicans, respectively. The amount added in each pore is 50 µL; Nd: Not determined; * DMSO: Negative control by dimethyl sulfoxide.

•
Total solids: Data presented in Figure 3 illustrate the TS percentages of SLY samples without and with REO addition. The obtained data observed that TS percentages in SLY samples varied between 13 and 14.66%, and non-significant differences were found between control samples and other SLY samples after REO addition. Moreover, there were non-significant decreases in total solid content in all SLY samples during the storage period.

•
Total solids: Data presented in Figure 3 illustrate the TS percentages of SLY samples without and with REO addition. The obtained data observed that TS percentages in SLY samples varied between 13 and 14.66%, and non-significant differences were found between control samples and other SLY samples after REO addition. Moreover, there were non-significant decreases in total solid content in all SLY samples during the storage period. • Total protein: Data presented in Figure 4 illustrate the total protein (TP) percentages of SLY samples without and with REO addition. The obtained data showed that the addition of REO caused a significant decrease in TP percentages, while the control samples recorded the highest amount of TP percent up to 8 days of storage. However, the influence of REO addition on protein content appeared on day 12 of storage, as T1 and T2 samples recorded a higher value of protein content than control. • Total protein: Data presented in Figure 4 illustrate the total protein (TP) percentages of SLY samples without and with REO addition. The obtained data showed that the addition of REO caused a significant decrease in TP percentages, while the control samples recorded the highest amount of TP percent up to 8 days of storage. However, the influence of REO addition on protein content appeared on day 12 of storage, as T1 and T2 samples recorded a higher value of protein content than control. • Total protein: Data presented in Figure 4 illustrate the total protein (TP) percentages of SLY samples without and with REO addition. The obtained data showed that the addition of REO caused a significant decrease in TP percentages, while the control samples recorded the highest amount of TP percent up to 8 days of storage. However, the influence of REO addition on protein content appeared on day 12 of storage, as T1 and T2 samples recorded a higher value of protein content than control.   Data presented in Figure 5 illustrate the fat percentages of SLY samples without and with REO addition. The obtained data showed that the addition of REO caused a significant boost in fat content percentages, while the control samples recorded the lowest amount of fat percent, as the T2 samples recorded a much higher value. In contrast, data showed non-significant differences in the fat content during storage in all treatments.

PEER REVIEW 13 of 22
Data presented in Figure 5 illustrate the fat percentages of SLY samples without and with REO addition. The obtained data showed that the addition of REO caused a significant boost in fat content percentages, while the control samples recorded the lowest amount of fat percent, as the T2 samples recorded a much higher value. In contrast, data showed non-significant differences in the fat content during storage in all treatments.

Microbiological Changes
Data presented in Table 4 illustrate the TPC as well as LAB (Log cfu/mL) of SLY samples without and with REO addition. The obtained data showed that the TBC and LAB increased significantly with increasing storage periods up to 16 days in all treatments. Moreover, control samples recorded the highest and lowest TBC and LAB, respectively. Furthermore, the addition of REO caused a substantial decrease and increase in TBC and LAB, respectively. The LAB count increased in all SLY samples during storage, indicating that REO addition boosted LAB count. Regarding yeasts and molds counts, the data showed that yeasts were detected in the control samples after 8 days of storage; they appeared after 12 days in SLY samples supplemented with REO (T1 and T2), while molds appeared only in the control sample after 8 days of storage, demonstrating the role of REO addition in slowing down the spoilage rate. On the other hand, the data observed that coliform group counts were not detected in all treatments. This might be due to the highhygienic condition during the making of SLY and the development in the acidity.

Microbiological Changes
Data presented in Table 4 illustrate the TPC as well as LAB (Log cfu/mL) of SLY samples without and with REO addition. The obtained data showed that the TBC and LAB increased significantly with increasing storage periods up to 16 days in all treatments. Moreover, control samples recorded the highest and lowest TBC and LAB, respectively. Furthermore, the addition of REO caused a substantial decrease and increase in TBC and LAB, respectively. The LAB count increased in all SLY samples during storage, indicating that REO addition boosted LAB count. Regarding yeasts and molds counts, the data showed that yeasts were detected in the control samples after 8 days of storage; they appeared after 12 days in SLY samples supplemented with REO (T1 and T2), while molds appeared only in the control sample after 8 days of storage, demonstrating the role of REO addition in slowing down the spoilage rate. On the other hand, the data observed that coliform group counts were not detected in all treatments. This might be due to the high-hygienic condition during the making of SLY and the development in the acidity.

Sensory Evaluations
Data presented in Table 5 illustrate the organoleptic properties, such as flavor, body and texture, and appearance and overall acceptability of SLY samples, without and with REO addition. The obtained data showed that REO-supplemented samples had the greatest mean flavor, body and texture, and overall acceptance scores, while the control samples had the lowest scores. The data in the same table revealed that the impact of storage time on sensory characteristics, as the addition of REO (T1 and T2), improved flavor and general acceptability during storage up to 16 days in most treatments.

. Total Phenolics and Total Flavonoids
Our data were lower than those of Adris et al., who found that total phenolic content was 29.23 mg gallic acid equivalent/g in the methanolic extract of rosemary [28]. Furthermore, Olmedo et al. found that REO contains 14.01 mg gallic acid/g of total phenol content [29]. The solvent used in oil extraction could explain the discrepancy. Adris et al. reported higher flavonoid content (6.59 mg catechin equivalent/g) in rosemary methanolic extract [28], when compared with our findings.

The Chemical Analysis and Identification of REO by GC/MS
Olmedo et al. found that camphor (35.70 g/100 g) was the most important component in REO, followed by verbenone (26.20 g/100 g) and b-caryophyllene (15.80 g/100 g) (24). In addition, another study showed that rosemary oil contains 45 volatile compounds, and eucalyptol (1,8-cineole) was the most abundant (33.15%), followed by camphor (10.31%), α-pinene (8.11%), isocaryophyllene (7.02%), bornyl acetate (5.66%), α-terpineol (4.92%), and camphene (4.22%), while differences in geographic location [30] may explain the differences between our study and previous studies. Generally, the natural antioxidant properties are mainly attributed to their phenolic contents; thus, their antioxidants action is like synthetic phenolic antioxidants [31]. Moreover, phenolic compounds are well known as radical scavengers, metal chelators, reducing agents, hydrogen donors, and singlet oxygen quenchers [32]. Therefore, natural antioxidants can protect the human body from free radicals and could retard the progress of many chronic diseases, as well as lipid oxidative rancidity in foods [33]. In this respect, it might be that REO has an antioxidant effect, mainly attributed to its phenolic contents.

Physico-Chemical Characterization
• Acidity: Ali et al. found that the addition of rosemary extracts during yogurt manufacturing affected the Gram-positive bacterial cell surface and inhibited its growth [77]. Likewise, the addition of oregano and rosemary essential oils to cream cheese declines the acidity value when compared with control samples [29]. Moreover, yogurt fortified with fish oil/γoryzanol showed a lower acidity than the plain yogurt [78]. The acidity value increased significantly during storage, and similar findings were noticed by El-Kholy et al. [34]. Indeed, the acidity percentages increased during storage due to the increase in lactic acid bacteria counts, which converted the lactose into lactic acid [79], whereas previous study found that the addition of REO to the yogurt samples did not show differences in titratable acidity and pH values during the storage period, and the titratable acidity of the control samples increased substantially [15].

•
Total solid: Our results are in agreement with Al-Soudy et al. [80], who found non-significant differences between the control sample and drinking yoghurt samples (infused with herbal extract) in total solids, due to the little quota of the herbal extract, whereas El-Sayed et al. found that fortified Labneh with Moringa oleifera oil increased the TS percentages [81]. Our results showed a non-significant decrease in total solid content during the storage period; similar results were obtained by Ghalem and Zouaoui, who found that dry matter percentage decreased in control yogurt samples with storage period [15].

•
Total protein: Results obtained by El-Kholy et al., who found that the protein content decreased in cheese samples after the addition of REO (from 17.18% in control samples to 16.91% in cheese made with REO) [34], were in agreement with our study. The decline in protein content in our study after 8 days of storage could be attributed to proteolytic activity by added microorganisms (76). Our findings are consistent with the results of El-Kholy et al. and Thabet et al. [34,82].

•
Fat: Our results were consistent with a previous study, conducted by Ghalem and Zouaoui, as they observed a slight increase in yogurt fat with REO addition [15]. Moreover, the addition of REO extracts to the cheese showed a significant (p ≤ 0.05) higher fat content than the control cheese samples [34]. Our data illustrated that fat content during storage matched the results reported by Ghalem and Zouaoui [15].

Microbiological Changes
Generally, the data revealed that there were decreases in TBC and yeasts and molds counts in SLY with increasing REO levels in all treatments. This may be due to the effect of REO on TBC and yeasts and molds. Indeed, REO contains monoterpenes, such as αpinene, 1,8-cineole, and borneol (Table 1), which have strong antibacterial and antimicrobial activities on food product deterioration pathogens [83].
Control samples had higher counts of TBC and yeasts and molds than those of SLY with REO in all treatments. Similar findings were reported by El-Kholy et al. [34]. On the other hand, Ali et al. [77] discovered that rosemary extract had a detrimental effect on the starter cultures Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus. Our data illustrated the effect of REO addition in slowing down the spoilage rate and similar observations were reported by Ghalem and Zouaoui and Ali et al. [15,77].

Sensory Evaluations
The enhancement effect of REO addition on sensory properties in our study matched the findings reported by El-Kholy et al., as they showed that adding REO to ultra-filtrated soft cheese samples improved sensory characteristics, while the control samples had lower scores [34]. According to other findings, adding 0.14 g/L of REO to yogurt samples resulted in the best flavor, taste, and texture, whereas increasing the previous concentration resulted in the worst flavor, taste, and texture [15]. The control samples degraded in all sensory quality properties due to aromatic chemicals (such as acetone and acetaldehyde) and lactic acid generation during storage [84]. These results are in agreement with those reported by Ali et al., who proposed that the addition of aqueous extract of rosemary affected the sensory properties of yogurt (flavor, body and texture, appearance, and overall grade), wherein an increasing concentration of rosemary extract increased the score of flavor, body and texture, appearance, and overall grade [77].

Conclusions
The results of antibacterial characteristics of REO indicated that it could be used as a natural preservative in the manufacture of SLY. In addition, SLY samples with REO addition showed acceptable flavor, body and texture, appearance, and overall acceptance. Based on the analysis, SLY with REO at 0.5% and 0.7% performed equally in this study, which indicates that adding REO at 0.7% would not provide additional benefits when compared with REO at 0.5% in SLY. Therefore, REO at 0.5% could be utilized in the manufacture of SLY.