Oxidative and Microbial Stability of a Traditional Appetizer: Aubergine Salad

Abstract

An eggplant-based salad, called aubergine salad (AS), is a traditional appetizer and as such, is quite popular in the Mediterranean area. It is widely produced either on a home scale or on an industrial scale and widely consumed. However, there are cases where preservatives (such as sodium benzoate and potassium sorbate) are added in order to extend the shelf life of the product. In the present study, the stability of this delicatessen against oxidation and microbial spoilage was evaluated, with or without preservatives. The physicochemical properties of the salad were evaluated, along with the tocopherol content, resistance to oxidation, and microbial count. According to the results, it is evident that the induction period of AS is 16% (in the case that preservatives were used) and 26% (in the case without preservatives) increased, compared to a control sample (plain soybean oil). This can be attributed to the increased content in tocopherols, and more specifically to α-tocopherol. Furthermore, the addition of preservatives resulted in increased storage days and a reduction of microorganisms. However, in both cases, the AS-prepared salad exhibited a self-stabilization ability after 13 days, negating the need for preservatives.

1. Introduction

In recent years, salads called traditional, or delicacies, have become very popular in both Europe and the United States. As a consequence, their sales, as industrial products, have increased significantly [1]. Generally, traditional foods are a major part of human diets, in countries around the world, since they are closely related to local culture [2]. Therefore, they have an important impact on human nutrition. Their production is based on a variety of recipes and ingredients, such as mayonnaise, starch, yogurt, cheese, vegetables, dates, fish, potatoes, and various types of sauces, in addition to local ingredients [1]. One such traditional appetizer is the eggplant or aubergine (Solanum melongena L.) salad, made for the first time in the Holy Monasteries of Mountain Athos, Greece, or Melitzanosalata (originated from the Greek name of the plant: melitzana), which is very popular and widely consumed in many countries and especially in the Mediterranean region. This salad is so popular that aside from the industrial products, it is also widely prepared on a home scale. This delicatessen is traditionally made of grilled eggplants, grilled red peppers, vegetable oil, vinegar, garlic, and salt.

It is established that, except for lipid oxidation in food products, there are two other important indicators of food quality: microbial load and the impact of bacterial pathogens and yeasts. Food spoilage involves any changes that make food unpleasant and dangerous for human consumption [3]. Food spoilage is both a health and an economic issue for consumers, leading to the recognition of microbial food safety as a major global issue for consumers, researchers, and the food industry. Food contamination is one of the most significant causes of food spoilage [4] since it can occur at any stage of food production. For vegetables and vegetable-based products, according to the EU commission legislation, Escherichia coli (E. coli), Listeria monocytogenes and salmonella are the most commonly examined type of microorganisms in order to assess the safety of the food product [5].

It is accepted that the inadequate handling conditions of salads in catering, delicatessen, or homemade preparation can sometimes allow the survival of microorganisms and promote their growth. Today, although chemical preservatives (used to extend the shelf life of the product) are supposed to prevent and delay food spoilage, foods without chemical preservatives seem to be more appealing to consumers, as these kinds of products are reminiscent of “homemade” food and appear to be a more healthy alternative [1]. While consumers today increasingly prefer less additions of chemical preservatives, foods (including AS), must, undoubtedly, be safe for human health and, at the same time, have a prolonged shelf-life [6].

Conventional approaches to determining the microbial load of a product for consumer safety are based on cell-counting methods [7]. The plate-counting method is the most widely employed technique for detecting the presence of yeasts in different kinds of edible products, but has a main disadvantage compared to other analytical methods; it requires a long time to obtain the results. As a consequence, there has been a great interest in quantifying microorganisms, based on other methods. Therefore, various alternatives have been proposed, such as methods that are based on the immediate enhancement of yeasts [8] and bacteria [9]. Adenosine triphosphate (ATP) bioluminescence is an alternative, relatively rapid method, suitable for detecting microorganisms in food through online monitoring of bacterial contamination in edible products (foods and beverages) [7]. It is sensitive and, moreover, does not require a culture step or expensive equipment [10].

Aside from food spoilage, chemical deterioration is also a major issue. Deterioration occurs mostly due to oxidation of the food components. Two factors determine the stability to oxidation of food lipids: (I) their fatty acid composition and (II) the presence of antioxidant compounds [11,12]. It is well established that tocopherols have a major antioxidant activity and, therefore, prevent the chemical oxidation of lipids of food products [13]. In addition, many researchers have monitored the relationship between tocopherol content and the progression of oil oxidation during storage [14,15]. This is due to the fact that the stability of the oil contained in traditional salads also characterizes the stability of the final product.

Even though foods (such as products reminiscent of “homemade” salads) without chemical preservatives seem to be more appealing to consumers, it is mandatory to ensure the safety of these traditional salads, free of preservatives. The aim of this study is to determine the physicochemical characteristics (such as pH, acidity, and water activity), tocopherol content, the stability to oxidation, and the microbial stability of AS with and without preservatives. To the best of our knowledge, this is the first time that this traditional product is examined in terms of resistance to oxidation and its microbial flora development.

2. Materials and Methods

2.1. Chemical and Reagents

Sodium benzoate, potassium sorbate, n-hexane, and substrates used to determine microbial contamination (Ringer’s solution, MacConkey broth, and Sabouraud broth) were purchased from Merck (Darmstadt, Germany). The reagents (Lumit-PM, Lumit-Buffer, NRS, NRB, and Somase) used in the ATP bioluminescence method to determine the microbial contamination were obtained from Lumac B.V. (Landgraaf, the Netherlands). Petroleum ether was purchased from Sigma-Aldrich (Taufkirchen, Germany). The dl-α-tocopherol, γ- and δ-tocopherol standards were purchased from Alfa Aesar (Karlsruhe, Germany).

2.2. Aubergine Salad Samples

Traditional AS containing preservatives or not were examined in this study. The AS was prepared using ingredients from a local market according to a simple homemade recipe. More specifically, the grilled eggplant fruits were well mixed with ground Florina (Capsicum annuum L.) peppers (a kind of long red peppers) cultivated in the northern region of Western Macedonia (Greece), soybean oil, vinegar, cloves of garlic, and salt, using a food processor. The ingredient proportions are shown in

Table 1. % Percentage of ingredients of aubergine salad.

Ingredient	% Percentage of Total Weight
Eggplant	80
Florina peppers	8
Soybean oil	4.5
Vinegar	4
Garlic	2
Salt	1.5

. The salad was divided into two equal batches. The preservatives sodium benzoate (0.05% w/w) and potassium sorbate (0.05% w/w) were added to the first batch, while none were added to the second. After that, each batch was separated into 100 g sub-batches, which were packaged in sterile, sealed plastic containers. All containers were stored at 5 °C. For conducting the experiments, three batches of AS were prepared, on a monthly basis, during summer 2021, with freshly bought ingredients, so as to have a wider variety of samples.

2.3. Determination of Physicochemical Parameters

The pH of the samples was measured using a digital pH meter (Hanna Inst., Smithfield, Smithfield, RI, USA). The sample was diluted with distilled water (1:1 w/v) and agitated for 30 min. After centrifugation for 5 min at 4000 rpm, the pH of the supernatant was recorded. The titratable acidity was determined according to the method ISO 750:1998 [16] and the results were expressed as a percentage of acetic acid (w/w). Water activity (a_w) was measured by a a_w meter (FA-st1, GBX France Scientific Instrument, Romans-sur-Isère, France) at 23 °C ± 1 °C.

2.4. Extraction of Oil and Determination of Oxidation Stability

The lipid phase extraction from all AS preparations was performed according to the method described by AOAC 935.60 [17]. In brief, 20 g of sample were placed in a glass bottle and 80 mL of petroleum ether were added. After thorough mixing for 10 min, the sample was centrifuged (5 min, 4000 rpm), and the supernatant was transferred to a glass round flask. The extraction step was repeated three more times to the residue, and the supernatants were pooled. The solvent was removed at a rotary evaporator, and the oil was collected. The stability to oxidation of extracted oil was determined as described by Lalas et al. [18] using a Rancimat 743 (Metrhom Ltd., Herisau, Switzerland). The temperature was set to 90 °C and the airflow to 15 L/h. The results were expressed as induction period (in hours).

2.5. Determination of Tocopherol Composition by HPLC-FLD

The tocopherol (vitamin E) content of the lipid phase of AS samples was determined as described by Lalas et al. [18]. The analysis was carried out on a Shimadzu Prominence CBM-20A liquid chromatograph (Shimadzu Europa GmbH, Duisburg, Germany) equipped with a SIL-20AC autosampler and a CTO-20AC column oven. The detection was performed using a Shimadzu RF-10AXL fluorescence detector (FLD) set at 278 nm (excitation) and 339 nm (emission). The flow rate of the mobile phase (n-hexane: 2-propanol: absolute ethanol, 97.5: 2.0: 0.5% v/v/v) was 1.0 mL/min. A 20 μL sample (diluted 1:5 w/v with n-hexane) was injected into the HPLC. The column used was a Waters μ-Porasil (125 Å, 10 μm, 3.9 mm × 300 mm) (Waters Corp., Waltham, MA, USA).

2.6. Determination of Microbial Population by ATP Bioluminescence

The method used to determine microbial population via ATP measurement was described in detail by Lalas et al. [6] and is in agreement with the results of colony-forming units measurement with ISO 4833-1:2013 [19], ISO 21527-1:2008 [20], and ISO 5541-1:1986 [21] standards for total plate counts, enumeration of yeasts, and enumeration of coliforms, respectively [7]. For the preparation of the samples, 3 portions of AS (10 g each) were mixed with 90 g of (A) Ringer solution, (B) Sabouraud medium and (C) MacConkey broth. After 2 min of homogenization with a Stomacher, sample A was analyzed directly, whereas samples B and C were analyzed after incubation for 24 h at 35 °C. Afterwards, for the determination of relative light unit (RLU) values, 200 μL of the sample was aseptically transferred to a cuvette, 100 μL of Somase/NRS reagent was added, and the sample was incubated for 45 min. Then, half of the above mixture was placed into cuvettes, which were then placed into a Lumac Biocounter 1800 (Landgraaf, the Netherland). After the automatic addition of 100 μL of the NRB solution, the samples were left for 10 s followed by the addition of 100 μL of the Lumit-PM reagent.

2.7. Statistical Analysis

Data are expressed as mean value ± standard deviation of three replicate analyses. One-way analysis of variance (ANOVA) was used to determine the statistical significance of the differences between mean values; p < 0.05 was used as the criterion for statistically significant differences.

3. Results and Discussion

3.1. Physicochemical Parameters Determination

Food preservation methods usually include pH control, chemical preservatives, dehydration, reduction of water activity (a_w), etc. Thus, in the present work, these parameters were studied. Results are presented in

Table 2. Results of physicochemical parameters examined for AS with and without preservatives.

Parameter	AS without Preservatives	AS with Preservatives
pH	4.16 ± 0.02	4.14 ± 0.02
Acidity (% w/w acetic acid)	0.39 ± 0.03	0.41 ± 0.03
aw	0.99 ± 0.01	0.99 ± 0.01

. Potassium sorbate is generally recognized as a safe preservative used in various food products [22]. Sodium benzoate and potassium sorbate are extremely useful in the preservation of food and beverages and are usually added together to obtain a combined effect [23]. It is noteworthy that in foods sold as ready-to-eat (meaning not intended to be cooked or reheated), often, the quantity of chemical preservatives added is increased, compared to others.

In our work, the pH value of AS samples (with and without preservatives) was 4.15, with no statistically significant (p > 0.05) differences between samples. It has previously been reported that grilled eggplant fruits (which consist of the biggest part of AS) have pH values ranging from 4.5 to 5.3 [24]. The low pH of the final product can be attributed to the other ingredients of the recipe and, especially, to the high acidity of the vinegar.

Regarding the acidity (% w/w acetic acid), the value calculated for both samples (with or without preservatives) was about 0.40% (with no statistically significant differences between samples). The pH and acidity of the AS affect, as expected, the organoleptic characteristics of the final product and, are expected to contribute positively to its preservation.

In addition to the above, water activity is used to predict food stability and safety, since it can affect microbial growth, the rates of deteriorative reactions, and chemical/physical properties [25]. In particular, a_w turns out to be a useful indicator of the microbiological stability of food products and has an important impact on the sensory properties of foods, such as taste, aroma, texture, and their chemical and biological reactivity (e.g., lipid oxidation and (non-)enzymatic activity) [26]. The higher the value of a_w, the more unbound water exists in the food and the higher the susceptibility to microbial spoilage. Therefore, a_w is used as a parameter for product development and quality control in the food industry [27,28]. During the present study, the value of water activity (a_w) of the samples of AS was 0.99, presenting no significant differences between samples. This value is relatively high and hints toward food deterioration and increased susceptibility to microbial growth.

As a means of comparison, a commercially available AS was also analyzed in terms of the abovementioned physicochemical parameters. The pH of the sample was 3.9 (probably due to the different percentages of ingredients added), the acidity was 0.51% and the water activity was found to be 0.99. Therefore, the as-prepared AS samples closely resemble the commercially available ones.

3.2. Determination of Oxidation Stability and Tocopherol Composition

The induction period of the oils extracted from AS and control sample is shown in

Table 3. Induction period and tocopherol content of the lipid phase extracted from aubergine salad (AS) samples and control sample (soybean oil); superscript letters denote statistically significant differences (p < 0.05) within a column.

Sample	Induction Period (h)	Tocopherol Content (mg Per kg of Oil)
		α-tocopherol	γ-tocopherol	δ-tocopherol
Soybean oil (control sample)	27.0 ± 0.8 a	254 ± 11.2 a	1114 ± 40.2 a	399 ± 12.7 a
AS with preservatives	31.4 ± 0.5 b	293 ± 13.9 b	1092 ± 38.8 a	474 ± 21.6 b
AS without preservatives	34.0 ± 0.8 c	312 ± 16.1 b	1084 ± 39.9 a	438 ± 18.9 c

. In all cases, the induction period of the extracted oil from AS either with or without preservatives seemed to be significantly higher than that of the control sample. This can be attributed to the presence of compounds with antioxidant activity, leading to higher stability of AS, compared to soybean oil. These antioxidants can derive from all plant-based ingredients, which are good sources of antioxidants. However, a higher contribution is expected from the eggplant fruits (the content of polyphenols in eggplant was not found to be affected by grilling [29]) and red Florina pepper, since they consist of the larger part of the AS [29,30,31]. Moreover, the results showed that the extracted oil from the salad without preservatives showed the highest (significant at p < 0.05) induction period (higher oxidation resistance), compared to the extracted one from the salad with preservatives. This difference cannot be attributed to the addition of the preservatives, since no prooxidant activity has been reported so far for these compounds. Therefore, the observed difference should be attributed to another parameter (vide infra).

Results regarding the composition of tocopherols in all samples are presented in Table 1. All three tocopherol species were detected in the samples. The content in the major tocopherol (i.e., γ-tocopherol) was not found to be statistically significant among the different samples. However, increased contents of α- and δ-tocopherols were observed for both samples, compared to the soybean oil. Therefore, the oxidative stability differences cannot be correlated with the tocopherol content of the samples. However, the high content of AS in tocopherol may have a positive impact on the shelf-life of the product, since it not only serves as a nutritional supplement, but also as an antioxidant compound.

3.3. Determination of the Microbial Activity by (ATP) Bioluminescence

The ATP bioluminescence assay was chosen for this study because it is a rapid method for determining the microbial count of a product on an industrial scale, requires minimal to no sample pretreatment compared to traditional methods, and can detect the total bacterial viability quickly without sacrificing reliability, and is an excellent way to detect bacterial contamination non-specifically.

Figure 1 shows the mean RLU values of AS, with and without preservatives, in the Ringer solution. As can be seen, a minor increase in the RLU was recorded during the first day of the storage followed by an abrupt decrease. This increase can be attributed to the fast development of microorganisms until the AS was refrigerated. Thereafter, a slow increase in the RLU is recorded, which corresponds to the lag phase of a microorganism growth curve. The recorded increase is low, compared to the fast development rate of microorganisms. This is due to the non-optimal conditions for microbial growth (storage temperature: 4 °C, limited oxygen in the package, and presence of antimicrobial compounds). Next, an abrupt increase in the RLU is observed on day 8, followed by a steady decrease and stabilization after day 15 at a low level. Moreover, it can be observed that the RLU values of the samples without preservatives, almost in all cases, are higher than the RLU values of those with preservatives since potassium sorbate and sodium benzoate, used in the recipe, assisted in the inhibition of the growth of microorganisms.

In addition to the above, the same procedure was followed after incubation for 24 h, at 35 °C on two selective substrates (MacConkey and Sabouraud) in order to examine coliforms and fungi. Incubation was performed since the initial microbial count of yeasts and coliforms was lower than 10⁴ microorganisms per mL of sample. Figure 2 shows the RLU mean values of AS (in MacConkey broth), with and without preservatives. As expected, the RLU values of the samples without preservatives were much higher, at the beginning of the incubation compared to the RLU values of the samples with preservatives. A decreasing trend was recorded for the RLU values. This not only signifies that the coliforms did not increase during the storage time, but also their amount decreased, increasing the shelf time of the product. The difference in the profiles between the two samples can be ascribed to the addition of preservatives, which efficiently inhibited microbial growth. Figure 3 shows the RLU mean values of AS (incubated in Sabouraud liquid medium) with and without preservatives. As also expected, the RLU values of samples without preservatives were much higher compared to the RLU mean values of samples with preservatives. Additionally, a gradual decrease in RLU was observed during the first days, followed by a short increase between the 4th and 6th day and then again, a decrease in both samples, on the following days. The above differences can be attributed to the presence of preservatives that hinder the growth of yeasts (as exhibited by the respective sample). Moreover, this increase in the RLU during days 4 and 6 can be attributed to better adaptation of the yeasts on AS and an enhanced growth rate, which is later on followed by the death phase of the yeasts.

In both cases (MacConkey and Sabouraud substrate/with and without preservatives), the mean RLU values showed the same trend. The RLU values in the Sabouraud incubated samples were significantly (p < 0.05) higher than those in MacConkey, which means that there was a very high number of yeasts in the salad samples. It is obvious that the physicochemical characteristics of the AS, such as low pH values/high acidity, do not allow the growth of most bacteria, in contrast to the increase in the population of yeasts. It is already showcased that the pH can act as a bacterial inhibitor, by many scientists [32,33]. In particular, it seems that most microorganisms grow best near pH 7.0 (6.6–7.5), while few grow below pH 4.0 [34]. However, the results of a previous study showed that strict conditions are required, i.e., low pH and the addition of potassium sorbate in order to control the growth of Listeria monocytogenes in delicatessen salads [22]. Our results are in accordance with previous research, showcasing that ready-to-eat vegetable salads mixed with vinegar-based salad dressings are more susceptible to yeast deterioration during refrigerated storage because of their lower pH, whereas this is not the case with bacteria [35]. Indeed, many yeast species grow at pH values as low as 2.0–2.5 [35]. Sabouraud (selected media for yeasts) had a higher microbial count than MacConkey (selective medium for coliforms) because yeast cells are bigger than bacterial cells and, therefore, contain more ATP per cell, resulting in higher RLU values [7]. In addition, the amount of ATP in a eukaryotic cell can be up to 10⁷ times more than that of a prokaryotic cell, making it far easier to detect ATP from eukaryotic cells than from microbial cells [36].

AS showed its self-stabilization ability after 13 days, even without the use of preservatives. This phenomenon was also observed in our previous work [7] during the study of the self-stabilization ability of another traditional Greek ready-to-eat deli salad, the tzatziki salad. Therefore, it can be concluded that the use of preservatives is not obligatory in order to stabilize AS.

Finally, there is a potential correlation between the microorganism amount and the resistance of the samples to oxidation. The lipid phase of the salad without preservatives showed better oxidative condition (longer Induction Period). Taking into account that the sole difference between the AS with and without preservatives was the microbial load, such a correlation may be possible. Microorganisms somehow protected the oil phase of the AS by acting as free radicals and oxygen scavengers. It is common knowledge that microorganisms possess defense mechanisms against oxidative damage. In order to do so, they produce metabolites to inhibit oxidation [37]. One type of microorganisms, acknowledged for this property is lactic acid bacteria (LAB). They possess multiple oxidation defense mechanisms and as a result, they can inhibit oxidative damage to the host organisms [37]. Additionally, their antioxidant potential was verified in the study of Lin et al. [38]. In our case, no LAB were added in AS, as probiotics. Aside from LAB, there are also multiple other microorganisms known to produce metabolites with antioxidant activity, such as gallic acid, ferulic acid and ellagic acid, deriving mainly from fungi [39]. As stated in the study of Chandra et al. [39], there are over 23,000 metabolites of microbial origin with various bioactivities, including antioxidant activity. Therefore, it is possible that the microorganisms contained in the AS produce some metabolites that bestow oxidative stability to the product. This is further strengthened by the fact that the sample without the preservatives (that exhibit antibacterial activity), that contains more microorganisms, exhibits better oxidative stability. However, future work should be focused on the investigation of this possible relationship.

4. Conclusions

The appetizer AS is a unique product that can be found either as a homemade salad or as an industrial product. It is widely consumed in the Mediterranean region and therefore, its storage conditions are important to ensure consumer safety. In this context, preservatives are usually added in the industrial products to extent shelf life, which is not always welcomed by the consumers. As indicated by the results of the present work, AS can be characterized as an oxidatively stable and microbially self-stabilized product (negating the need for the addition of preservatives), which makes it attractive for consumers who prefer preservative-free products in resemblance of homemade ones.