Phytochemical profiles of lemon verbena (Lippia citriodora H.B.K.) and its potential application to cookie enrichment

Abstract In this research, phytochemical properties of lemon verbena and oxidative stability of the fat component in cookies (contain lemon verbena powder and EO) were investigated. The essential oil (EO) profile and polyphenol compounds were identified by GC/MS and HPLC, respectively. Different concentrations of lemon verbena powder and EO were added to the cookies in comparison with TBHQ. The oxidative stability of fat component in cookies (peroxide value, p‐Anisidine, TOTOX value), along with the physicochemical (pH, acidity, weight loss, and moisture content) and sensory properties of the cookies were evaluated over a period of 6 months during storage at room temperature. The main constituents of EO are geranial (27.21%), neral (20.01%), spathulenol (7.28%), and limonene, while trans‐Ferulic acid (6.71 mg/g), Hesperidin (1.87 mg/g), and ρ‐Coumaric acid (0.04 mg/g) were measured as main phenolic compounds. The peroxide value increased in all samples for the first 2 months of storage and then decreased as hydroperoxide was converted to secondary oxidation products. The p‐Anisidine value increased in all samples during storage. This parameter was lower in cookies containing lemon verbena EO and TBHQ treatments. Sensory evaluations of cookies showed that lemon verbena EO had positive effects on the aroma and taste of cookies during storage, whereas lemon verbena powder had adverse effects on mouthfeel and consumer acceptance. The results showed that lemon verbena can increase the eating quality, prolong the shelf life, and maintain the integrity of bakery products with high‐fat content.

Lemon verbena (Lippia citriodora H.B.K.) is a medicinal and aromatic plant with wide usage belongs to Verbenaceae family. Lemon verbena is an annual plant that originates in South America and is nowadays grown commercially in North Africa (Morocco), Southern Europe, and various parts of Iran (Casamassima et al., 2013). The plant contains various phenolic and flavonoid compounds that validate its use as a natural antioxidant (Naser Aldeen et al., 2015). It has traditionally been used as tea seasoning and herbal tea due to its pleasant aroma and taste (Pascual et al., 2001). Therefore, it seems that lemon verbena can be used for enriching cookies and can be acceptable by consumers.
Accordingly, the objectives of this study are a) to study the phytochemical properties of lemon verbena and identify the chemical composition of essential oil (the plant growth in Iran), as well as its alcoholic extract, via GC/MS and HPLC, respectively; b) to determine the antioxidant activity of lemon verbena essential oil via the DPPH method and by drawing a comparison with the synthetic antioxidant TBHQ; and c) to evaluate and compare the use of essential oil and lemon verbena powder in cookies, their physicochemical and sensory properties and oxidative stability of cookies fat for a 6-month period of shelf life.

| Materials
The raw materials for making cookie dough were wheat flour, margarine, whole egg, sugar, vanilla, skim milk, salt, and baking powder,

| Plant preparation
One batch of Lemon verbena (L. citriodora) leaves was obtained from the HPRC in the harvest season (November 2019). Immediately after harvest, fresh lemon verbena leaves were cleaned. Healthy leaves were selected to be included in sample preparation. The plant was dried at ambient temperature (28 ± 2°C) for 72 hr.

| Essential oil (EO) extraction
Shade-dried leaves of lemon verbena were subject to water distillation for 3 hr in an all-glass apparatus, which eventually generated yellow essential oil. The Clevenger method was used for extracting the essential oil from the lemon verbena leaves. The isolated oil was dried over anhydrous sodium sulfate and stored in a laboratory freezer (−18°C). The extraction efficiency was calculated based on the dry weight of leaves (v/w).

| Plant extraction and biochemical measurements
Plant materials were extracted according to the maceration method, as described by Wojdyło et al., (2007) with some minor modifications. The antioxidant activity of the extract was determined by a spectrophotometric method based on the reduction of a methanol solution from DPPH (Oke et al., 2009). Assays on flavones and flavonols were carried out according to Popova et al., (2004), with some minor modifications, and the results were reported as mg quercetin/g dry weight of plant sample (Popova et al., 2004). Total flavonoid content was calculated according to the formation of a flavonoid-aluminum complex, while quercetin was selected as a standard (Menichini et al., 2009). Total phenolic contents were analyzed using the Folin-Ciocâlteu colorimetric reagent, and gallic acid was used as a standard (Wojdyło et al., 2007).

| Gas Chromatography/Mass Spectrometry (GC/MS) Analysis
Identifying the components of essential oils involved using the gas chromatography (GC) and gas chromatography-mass spectrometric (GC-MS) analyses. To identify the constituents of the lemon verbena essential oil, an Agilent Technologies-7890A gas chromatograph device was used. The type, length, diameter, and thickness of the column were HP-5-MS, 30 m, 0.32 mm, and 0.25 μm, respectively.
The temperature program of the column was set to change within the range of 60-210°C at a rate of 4°C/min. Nitrogen carrier gas was used at a flow rate of 0.5 ml/min. The gas chromatograph was connected to a mass spectrometer (GC-MS) (Agilent Technologies-5975C). The column type was HP-5MS, being 30 m in length, 0.25 mm in diameter, and 0.25 μm in thickness. The temperature program was 280°C, and helium carrier gas was used at a flow rate of 1 ml/min. The relative percentage of each component of EO was determined based on chromatogram peak area and compared with the total area by using the normalization method of the GC/ FID peak areas. Retention indices were determined using retention times of n-alkanes (C 8 -C 25 ) that were injected after the volatile oil under the same chromatographic conditions. The retention indices for all components were determined by using n-alkanes as standard.
Identifying the spectra involved using a data bank of mass, retention time, Kovats index, and a study of mass spectra per essential component and pattern of spectral refraction, compared with standard spectra and with the use of reputable sources (Adam, 2001).

| Identification of polyphenol by HPLC
Extraction, separation, and quantification of phenolic compounds were performed according to method of Mišan et al., (2011) with some modifications. Plant extracts were prepared by macerating 200 mg dried samples with a solution of methanol/acetic acid mixture (85:15) for 24 hr at 4°C and subsequently extracted in an ultrasonic bath at room temperature for 15 min. The resulting suspension was then centrifuged at 10,000 rpm for 20 min at 0°C. To remove compounds such as chlorophylls and lipids, the supernatant was extracted with 1 ml n-hexane and centrifuged at 10,000 rpm for 10 min.
After removing the supernatant, the resulting solution was used for the analysis of both total phenolic contents and their components.
The HPLC system employed consisted of a high-performance liquid chromatography (Agilent 1,200 series) equipped with a UV-Vis multi-wavelength detector at 280 and 330 nm. Data were evaluated using a ChemiStation Software (Agilent Technologies) data processing system. The separation of components was achieved by an Agilent, XDB-C18, 5 μm, 4.6 × 150 mm column, at a flow rate of 1 ml/min. Solvent gradient was performed by varying the proportion of solvent A (methanol) to solvent B (2% acetic acid in water) to separation chlorogenic acid in 330 nm and other compounds in 280 nm. The total running time and postrunning time were 30 and 10 min, respectively. The column temperature was 30°C. The volumes of samples and standards injected were 20 μl which was done automatically using autosampler (Mišan et al., 2011). 2.2.6 | Dried leaves powder preparation Lemon verbena leaves were shade-dried at room temperature (28 ± 2°C). The dry lemon verbena leaves were ground and powdered by a kitchen blender before being mixed in the cookie dough.
Then, beaten whole eggs (75 g) were mixed with water (100 g) and then added to make dough. The dough was divided accurately to make each piece 20 g in weight. It was then rounded and sheeted into cookie shapes (5 cm diameter with 3 mm thickness) before baking at 180°C for 20 min in a convection oven. The cookies were packed in polyethylene bags and stored at room temperature (25°C) for 6 months. The cookie samples were examined after 0, 2, 4, and 6 months of storage.

| Extraction of fats
The cookies were ground and their fat content was obtained, as described by Hallabo (1977) via cold extraction with n-hexane. About 100 g of each sample was shaken with 200 ml n-hexane in a metabolic shaker for 24 hr before being filtered. In a fat-solvent mixture, the n-hexane evaporated after filtration at room temperature by a vacuum pump under a laboratory hood. The fat residues were obtained after being dissolved and were relocated to a glass tube for storage at freezing temperature (−18°C) for further analysis.

| Oxidative stability
The stability of cookie's fats was examined periodically at intervals of 60 days during the 6 months of storage at ambient temperature.
The assessments involved pH, acid value, peroxide value, p-Anisidine value, and TOTOX value, according to previous methods described by AACC (2000) and A.O.C.S. (1993).

| Moisture content
Moisture content was determined by gravimetric heating (130 ± 2°C for 3 hr) using a 5 g ground cookie sample on a preweighed petri plate, according to the 44-15 AACC method (AACC, 2000). After heating, petri plates cooled in a desiccator at room temperature and the loss of weights were reported as moisture content (%).

| Baking loss and weight loss
Baking loss (%) was calculated as the difference between the weight of the cookie before and after the baking process. Weight loss was determined by weighing cookies after 2, 4, and 6 months of storage.
All measurements were carried out in triplicate.

| Sensory evaluation
Cookie samples were evaluated for their sensory characteristics on a five-point hedonic scale. This involved assessing color, appearance, flavor, taste, texture, mouth feel, and overall acceptability. Ten trained panelists volunteered to test the samples.

| Statistical analysis
Experimental and organoleptic data were analyzed for variance (ANOVA) and significant differences were identified according to Tukey test (HSD: High significant differences) (p <.05), using the JMP 8.0 software. The analyses were carried out in triplicate, and the results were presented as average values.

| Biochemical analysis
According to the results, the inhibitory power against DPPH radicals in lemon verbena extract was 77.35 ± 0.21%. In this regard, Farahmandfar et al., (2018) reported that increasing the concentration of lemon verbena essential oil from 200 to 3,200 ppm in sunflower oil caused the free radical scavenging activity of DPPH to increase from 10% to 56%. Furthermore, Choupani et al., (2014) reported that the level of DPPH radical scavenging activity by the methanol extract of lemon verbena was 85% and that this amount was higher than the ethanol, acetone, and aqueous extracts.
A variety of phytochemical compounds in medicinal plants exhibit an array of protective and therapeutic properties that function effectively and are necessary for preventing the occurrence of disease. One group of these phytochemical compounds is phenolic compounds. The results showed that the amount of total phenolic compounds in the methanol extract of lemon verbena leaf was 49.2 ± 0.11 mg/g (Table 1). This was higher than the amount reported in a study by Choupani et al., (2014). The researchers reported that the total phenolic compounds in the methanol, ethanol,
In previous cases of research on the components of lemon verbena essential oil, Argyropoulou et al., (2007) reported that the number of

| Qualitative-quantitative high-performance liquid chromatographic analysis
The different components of lemon verbena extract were fractionated and identified via the HPLC technique. In total, 17 different standards were injected, and phenolic compounds were detected at 280 nm. Trans-ferulic acid, hesperidin, and ρ-coumaric acid were identified by making a comparison between the inhibition time, which was detected in the chromatogram, and the illustration of standard curves. According to HPLC analysis, the most important components of lemon verbena extract were trans-ferulic acid (6.71 mg/g), followed by hesperidin (1.87 mg/g), whereas the least significant ingredient in lemon verbena extract appeared to be ρ-coumaric acid (0.04 mg/g). More details of the components in lemon verbena extract are presented in Table 3. Bilia et al., (2008) reported that the most important compounds identified in the Components less than 0.05% was not reported.

| Thermal stability of cookie's fat
Fat content was extracted from cookies and evaluated for pH, acid value, peroxide value, p-Anisidine value, and TOTOX value to determine the functionality of various concentrations of essential oil and lemon verbena powder in comparison with TBHQ. The results according to variance analysis showed significant effect of treatment on measured factors (Table 4). Changes in pH were directly related to the amount and the release rate of free fatty acids, as well as the buffering properties of components in the cookie formulation.
According to Figure 1, there was no significant difference between the pH of different formulations, apart from the pH of the control sample after baking (p <.01). The pH of the control sample was the lowest of all after baking, which indicates fatty acid degradation and the breakdown of triglycerides which generated free fatty acids  in cookies during storage was greater than the ability of essential oil and powder obtained from cardamom and cinnamon in cookies (Badei et al., 2002), drumstick (Moringa oleifera) (Reddy et al., 2005), amla leaves (Emblica officianalis), raisins (Vitis vinifera) in biscuits, and rice bran extract in cookies (Bhanger et al., 2008). can be broken down into more complex products including ketones, aliphatic aldehydes, alcohols, and hydrocarbons, collectively known as secondary products of oxidation (Kolakowska, 2003). Therefore, the safety of food products that contain fat is calculated based on the progress of oxidation processes. Tracking and measuring primary and secondary products of fat oxidation can help control the quality of bakery products, but because the destruction and change of primary and secondary products of fat oxidation are done continuously over time, measuring one of these parameters alone cannot yield accurate results of the fats oxidation intensity. Thus, primary oxidation products (the peroxide value being an indicator) and secondary oxidation products (the p-Anisidine being an indicator) are measured simultaneously. This can serve as an optimum way to control the oxidation of fat products (Bialek et al., 2016;Pegg, 2001).
When the peroxide value is between 10 and 20 meq./kg of fat, the food has technically become rancid but is still acceptable Abbreviation: ns: Not significant correlation *Significant correlation in levels of 5%;; **Significant correlation in levels of 1%.

TA B L E 5
The correlation coefficient between most important properties of cookie's fat in terms of flavor and taste. However, if this number exceeds 20 meq./kg, the food becomes no longer acceptable by the consumer because the strong rancidity affects the food flavor as well (Pearson, 1970). On the other hand, the maximum peroxide value being allowed in a cookie is 2 meq./kg, but because in this study,  (Bialek et al., 2016). In the same report, it was claimed that the highest peroxide value was observed in the ninth week of storage, after which time the peroxide value decreased significantly in the cookies.
The peroxide value is generally used for determining the amount of oxidative degradation in oils, fats, and fatty foods. However, the rapid conversion of hydroperoxides to secondary oxidation products makes the peroxide value an incomplete criterion for determining the oxidative degradation of oils. To yield more accurate results, it is recommended that other quality tests be carried out on oils and fats (Bhanger et al., 2008). One of the most valid tests that can determine the oxidative stability of oils and fats is the p-Anisidine test.
In all cookie samples, the p-Anisidine value increased significantly over time (p <.01). While considering that the p-Anisidine value represents the amount of secondary oxidation compounds, the decrease in peroxide value from the second month of storage onwards occurred parallel to an increase in the p-Anisidine value (Figure 4).
During the 6 months of storage, p-Anisidine levels increased from 3.02 to 8.60 in cookies that contained lemon verbena powder. In cookies containing lemon verbena essential oil, it increased from 2.37 to 8.06, whereas it increased from 2.64 to 7.28 in cookies containing TBHQ. In the control sample, however, the p-Anisidine value increased from 4.56 to 9.35. The highest p-Anisidine value was observed in the control sample after 6 months of storage, whereas the lowest was observed after baking cookies that contained 5,000 ppm  (Pourfarzad et al., 2011). Due to their strong ability to absorb water, plant fibers can act as moisture absorbers and maintain moisture in the product during storage. Therefore, cookies containing higher proportions of lemon verbena powder had more moisture and edibility after 6 months. The analysis of variance showed that the essential oil did not have a significant effect on the moisture content of cookies during storage and, in this regard, there was no significant difference among the different concentrations of essential oil, synthetic antioxidants, and the control group. Sudha et al., (2007) reported that adding 25% apple pomace to cake formulations ultimately improved the moisture content and soluble fiber in cakes, increasing them by 1 and 5.64%, respectively (Sudha et al., 2007). On the other hand, Nanditha et al., (2009) reported that the use of turmeric powder did not have significant effects on the final moisture content of biscuits (Nanditha et al., 2009).

| Baking loss and weight loss
The effects of variables were different on baking loss and weight loss during storage. Baking loss denotes the amount of mass released in the form of gas under the influence of heat in the baking process.
Higher amounts of baking loss indicate that more moisture is removed from the product through baking, and this leads to a decrease in the shelf life of the baked products (Choi et al., 2007). According to this, the lowest amount of baking loss was observed in samples storage occurs in response to various factors such as packaging permeability to moisture, vapor pressure of the product, the equilibrium relative humidity of the surrounding atmosphere of product, package volume, storage temperature, etc. (Cauvain & Young, 2009).
A lower rate of moisture loss from the product is associated later on with a lower weight loss through storage. An enhanced level of weight loss during storage reduces the shelf life and eating quality of bakery products (Cauvain & Young, 2009). According to Figure 7, the lowest amounts of weight loss through storage were observed in cookies that had the formulations F3, F2, F1, F4, and F7. Plant fibers and their ability to retain high amounts of moisture limited the rate of weight loss in cookies containing lemon verbena powder. The results showed a strong, significant correlation between the rates of weight loss during storage and the rates of change in the moisture content of different cookies.

| Sensory evaluation
The cookies scored differently and significantly in terms of sensory evaluation ( taste. The highest taste score was given to 2000 ppm lemon verbena essential oil. As the concentration of lemon verbena essential oil increased from 2000 to 5,000 ppm, the taste score decreased but not significantly. As expected, the lowest taste scores were given to the control group and to the samples with TBHQ. Adding lemon verbena powder to the cookies caused a slight increase in scores on mouthfeel, as compared to the control group, but this increase was not significant. The highest score on mouthfeel was given to cookies containing lemon verbena essential oil. By increasing the concentration of lemon verbena essential oil, the score on mouthfeel increased. By increasing the concentration of lemon verbena powder, however, this score decreased. The presence of rough lemon verbena particles which had not been ground well caused a coarse mouthfeel. Secondly, the absorption of free water by plant fibers in the powder made consumers experience a drier mouthfeel while chewing, and so the mouthfeel score decreased in response to higher concentrations of lemon verbena powder in the cookies. In terms of overall acceptability after the first day of baking, the most acceptable formulation was F5 and the worst was F8 (i.e., the control sample).
The results showed that longer durations of storage led to significant effects on all sensory parameters. The sensory score for cookies that were stored for a total of 6 months at room temperature decreased significantly. The most substantial decrease in this score was attributed to the textural properties of the cookies.
Parallel to the increase in weight loss and the decrease in moisture content, the cookies became stale and, thus, the texture grew significantly stiff. In evaluating all sensory properties, the storage time had the least effect on the appearance and color of the cookies. As the storage time increased, the flavor and aroma scores of cookies containing essential oil and lemon verbena powder were similar and did not differ significantly. This means that, through time, the flavor and aroma scores decreased faster in cookies containing essential oil than in those containing lemon verbena powder. The volatile nature of essential oils is associated with a high rate of deterioration in their compounds through time, especially when the essential oils are used freely in food formulations. On the other hand, adding aromatic herbal powders, such as verbena powder, is likely to have more stable, enduring effects and turns out to be more successful in maintaining flavor over time. According to the sensory scores, the score on overall acceptance for F3 and F7 cookies was below "3" after 6 months of storage at ambient temperature (Table 6). This indicates that the consumer cannot accept these formulations after the said storage time. The decrease in the overall acceptance score in the F3 formulation was because of a decrease in mouthfeel score.
Nonetheless, it can be corrected by thoroughly grinding the verbena powder so as not to leave large, coarse particles in the cookies.
Another remedy can be to increase the initial moisture content of the formulation. Similar results were reported by Delvarianzadeh

| CON CLUS ION
Adding the essential oil, extract, or powder of lemon verbena, Through storage at room temperature, it was observed that the acid value, peroxide value, p-Anisidine value, and TOTOX value increased at much slower rates in cookies containing lemon verbena essential oil, as compared to the control sample. At a concentration of 5,000 ppm, lemon verbena essential oil had stronger inhibitory effects on oxidation than what TBHQ did. According to sensory evaluation, cookies containing lemon verbena powder received relatively low scores on acceptance, while cookies containing lemon verbena essential oil received higher scores due to their good flavor and aroma. Compared to synthetic antioxidants, medicinal plants such as lemon verbena can be added to a variety of foods. Their role as healthy additives, along with their functional properties, can deliver strong remedial effects, nutritional properties, and desirable sensory properties. The clinical study of the effect of consuming products containing lemon verbena essential oil on the health of consumers would be worthwhile.

ACK N OWLED G M ENTS
This project was supported by the Department of Food Science and Technology, Jahrom University, Iran. The authors are very grateful for their technical support.

CO N FLI C T O F I NTE R E S T
The authors confirm that there is no known conflict of interest associated with this publication. This article does not contain any studies with human participants or animals performed by any of the authors.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.