Effects of Drying on the Biochemical Composition of Atherina boyeri from the Tunisian Coast

The effects of solar and experimental drying on the protein concentration and total fatty acid (TFA) content of the muscles of Atherina boyeri (sand smelt) were studied. The percentage of protein was 18.3% for fresh fish and 14.2% for sun-dried fish, while TFA content was 4.9 g/100g and 0.5 g/100g. After the drying experiment, the percentage of protein was 23% and the fatty acid (FA) content was 2.8 g/100g. The results show that natural or experimental drying favors saturated fatty acids. The n-3 polyunsaturated fatty acids (PUFAs) are less sensitive to heat than the n-6 PUFAs. From a nutritional point of view, it seems that the drying conditions, where parameters are T = 50 ̊C, moisture = 30% and air speed = 2 m/s, would be the most beneficial for the preservation of sand smelt.


Introduction
Fish is important for humans for its nutritional qualities and also for the choice it offers in taste [1], texture, and the form in which it is marketed: whole, fillets, fresh, frozen, salted, smoked, dried, or processed [2]. Fish meat is the part that is most used by consumers. The fish meat is the most used in part for human consumption. The lipids of the flesh have a beneficial effect on human health due to the presence of n-3 polyunsaturated fatty acids (PUFAs) [3].

Sampling and Biological Parameters
Fresh samples of A. boyeri came from the region of Monastir. These samples were collected in winter 2010. The average total length (TL) and the average total weights (TW) of the individuals are shown in Table 1.

Drying Conditions
The fish were divided into several groups, each consisting of six samples (n = 6). One group was subjected to solar drying (SN). Another group was analyzed fresh (F). The other fish was subjected to different experimental conditions. It was dehydrated by convective drying at variable air speeds (1 m/s and 2 m/s) at different temperatures (45˚C, 50˚C, 60˚C, and 70˚C) and relative humidity (20%, 30%, and 40%). The experimental conditions are summarized in Table 2.
The drying tests were conducted in a closed-loop drying system using forced hot air convection. The average diameter of the fish was 14 mm, and the initial water content varied between 75% and 80%. The conditions of the drying air were kept constant during each test. A high precision balance is equipped with a data output enables via a software acquisition weight during the drying process. The temperature inside and on the surface of A. boyeri was measured using thermocouples connected to a data acquisition channel. The air flow in the test section was perpendicular to the surface of the product to be dried in order to obtain optimum air-fish contact.

Determination of Moisture Content
This measurement was made on fresh fish and dried fish. The fish was weighed prior to being placed in an oven to determine the wet weight (P h ). It was then placed in an oven at a temperature of 105˚C for 24 hours until a constant weight was reached (P s ). Moisture was calculated using the following formula:

Extraction of Total Fatty Acids
The extraction of total fatty acids (TFAs) was performed in the presence of chloroform-methanol (2:1 v/v) [12]. The total lipids obtained were stored in chloroform-methanol-butylatedhydroxytoluene (BHT) at −28˚C. For further analysis, the fatty acids were transformed into methyl esters, according to Cecchi et al. [13]. The quantification of the fatty acids is based on an internal standard not present in our samples, methyl nonadecanoate or C 19:0 (Sigma Aldrich, Corporate Headquarters, St. Louis, MO).

Identification and Quantification of Fatty Acids
Methyl esters of TFAs were separated, identified, and quantitated by gas chromatography using a HP 6890 gas chromatograph with a split/splitless injector with electronic pressure control and a flame ionisation detector was used for the analysis. Separation was performed with a 30 m HP Innowax capillary column with an internal diameter of 250 μm and a 0.25 μm film thickness, the stationary polar phase of the column being polyethylene glycol.

Protein Determination
The determination of protein content was based on a solution of serum albumin [14]. The sample was incubated for 30 minutes away from light and then the optical density at a 540 nm was measured.

Statistical Analysis
Mean comparison was performed by analysis of variance (ANOVA) followed by Duncan test at the significance level of p < 0.05. All statistical analyses were carried out using the software program SPSS version 13 (SPSS Inc., Chicago, Illinois).

Results
The percentages of moisture protein and total fatty acids (TFA) of fresh and sun-dried A. boyeri are shown in Table 3.

Effects of Experimental Drying on Protein Content
After subjecting A. boyeri to different experimental drying conditions, the protein content of the flesh varied significantly. The results are shown in Table 4. A maximum of 23% protein was obtained with E2. The effects of different temperatures were observed with E2 (23%), E6 (18%), and E8 (18%); the difference between the three conditions was significant. Protein content varied significantly with moisture, with a significant difference between E2 (23%) and E3 (19.6%). The drying rate appeared to have a destructive effect on the protein content; the destructive effect of protein content was greater with low drying speed of 1 m/s. the percentages were 23% and 17.6% for E2 and E9, respectively.

Effects of Drying on TFA Content
TFA content after extraction was expressed in g per 100 g of fish. The TFA contents in fresh and sun-dried A. boyeri were 4.9 ± 1.9 g/100g and 0.5 ± 0.0 g/100g respectively ( Table 3). Drying may be a necessary step in food preservation. We subjected A. boyeri to different experimental drying conditions. The TFAs obtained are shown in Table 4. Drying appeared to affect the TFA content of the flesh of A. boyeri. The values varied from 4.9 g/100g Table 3. Water (%), protein (%), and total fatty acid content (g/100g) of fresh and sun-dried A. boyeri (mean ± SE, n = 6). Variations in temperature and moisture content led to significant variations in TFA content between the experimental conditions. At 60˚C, the moisture content did not cause significant variation between E2 (2.8 g/100g) and E3 (2.6 g/100g). By contrast, at 70˚C, a significant difference was found between E4 and E1 where the TFA content was 2.6 g/100g and 1.8 g/100g, respectively. The drying rate did not seem to affect the TFA content. No significant difference was found between E3 and E9.

In Fresh Fish and Sun-Dried Fish
Natural drying (SN) promotes the formation of saturated fatty acids that reach a percentage of 67.2%, while in the fresh fish, the percentage of saturated fatty acids is 42.6% of the TFAs ( Table 5).

Under Experimental Drying
For fresh fish, natural drying seemed to affect MUFA content; while PUFA content remained stable (Figure 1). There seemed to be an inverse relationship between the percentage of SFAs and MUFAs. The most favorable conditions for the preservation of A. boyeri were those that showed a minimum percentage of SFA and a maximum percentage of unsaturated fatty acids (MUFA + PUFA). The results of E5, E6, and E7 met these conditions.
According to Figure 2, solar drying (SN) affected the percentage of unsaturated fatty acids. The PUFAs that were most sensitive to experimental drying were those of the n-6 family. Fatty acids of the n-3 family did not seem to be affected by experimental drying; the content was higher than in fresh fish. Two points stood out, E1 and E5, with respective percentages of 30.1 and 30.4% of the TFAs. Among the MUFAs, n-9 fatty acids were more resistant to experimental drying than n-7 fatty acids.

Discussion
The drying of fish is performed to reduce some of the so-called "free" water that is a vector of various infections and is involved in the degradation reactions. Drying helps promote preservation. During natural drying, the fish is exposed to air and natural light. Drying in the sun can be performed for 3 to 10 days, but drying times of 1 to 3 days are more common [15].
The protein content of fresh A. boyeri was 18.3%. With natural drying, the surface of the fish dries faster and hardens, locking moisture inside, which slows down the drying process and promotes the degradation of proteins [16]. In this study, we obtained a protein content of 14.2 g/100g. Protein destruction is accelerated when the products are subjected to high temperatures for an extended period of time [16]. The total fatty acid (TFA) content of fish affects the biochemical characteristics postmortem. In this study, the TFA content of fresh fish was 4.9 g/100g; it was 0.5 g/100g with sun-dried fish. When drying conditions change (E1 to E9), the contents of total fatty acids vary from 1.1 g/100g to 2.8 g/100g.
The exposure of A. boyeri for extended periods to solar drying led to the degradation of lipids. The formation of lipid oxidation by-products reduces the nutritional quality and increases health risks [17] because they are associated with aging, altered membranes, heart disease, and cancer [18]. Photo-oxidation explains the low trans fat content of fish exposed to the air [19].
Taking into account productivity, one might think that drying with warm air would take less time than natural drying. In E2, E3, E4, and E9, the TFA content increased in parallel with the temperature. However, a high temperature of 70˚C (E1) can alter the final TFA content, which was 1.8 g/100g. It seems that the E2 conditions would be ideal for the preservation of AGT in A. boyeri.
In fresh A. boyeri, palmitic acid was the major component of the SFAs, with a relative percentage of 57.5% (of SFAs), followed by stearic acid with a relative percentage of 29.3% (of SFAs). The profile of fatty acids in A. boyeri was distinguished by the presence of myristic acid (C 14:0 ), with a relative percentage of 9% (of SFA). During natural drying, the respective percentages of these fatty acids were 51.4%, 42.2%, and 2.9% (of SFA). These three fatty acids cannot be considered as a whole because they differ in their structure, metabolism, cellular functions, and even their deleterious effects in case of excess [20]. Myristic acid represents a small proportion of the TFAs in the animal body (between 0.5% and 2% of TFA) [21].
A. boyeri, in addition to being a source of protein, contains large amounts of PUFAs. In fresh fish, the percentage was 34.7% (of TFA), and between 33.2% and 41.4% (TFA) with experimental drying. The n-6 PUFA content was 11% (of TFA) in fresh fish. This category is represented by linoleic acid (LA, C 18:2n-6 ), acid which has a percentage of 5.1% and by arachidonic acid (ARA, C 20:4n-6 ) with a respective percentage of 5% (TFA). As shown in Table 5, under different experimental conditions, ARA was the main n-6 PUFA except with E5. The percentages of n-3 PUFAs were higher than those of n-6 PUFAs, as shown by the n-3/n-6 ratios in Table 5. The highest ratios of 4.8 and 4.2 were observed with E5 and E8, respectively. This ratio is useful for comparing the nutritional value of fish oils [24]. Fish and fishery products rich in n-3 fatty acids and low in n-6 fatty acids are considered good for human health [25]. An increase in the n-3/n-6 ratio is essential to help the body use n-3 fatty acids. A low ratio indicates that the enzymes that convert fatty acids to their active forms are likely to be used by n-6 PUFAs [26]. The percentage of EPA + DHA is responsible for variations in the n-3/n-6 ratios [26].
Research has shown that there are significant health benefits of a diet rich in EPA and DHA. DHA and ARA are important in neonatal health [27] [28], in particular in eye and brain development. A number of countries including Canada and the United Kingdom, and organizations such as the World Health Organization (WHO) and North Atlantic Treaty Organization have advocated dietary recommendations for n-3 PUFAs. These recommendations are 0.300 to 0.500 mg/day EPA + DHA [29]. Based on the results in Table 5, these recommendations can easily be met by consuming 100 g fresh A. boyeri (i.e., 680 mg EPA + DHA), or 200 g of dried fish with E6 (540 mg EPA + DHA), E7 (632 mg EPA + DHA) or E8 (492 mg EPA + DHA).
The PI index [30] shows the relationship between the unsaturated fatty acid composition of tissue and its susceptibility to oxidation. It provides information on the quality of the product that when the value of PI is high, polyunsaturated fatty acids are susceptible to be oxidized. The higher the PI value, the higher fat fish is likely to be oxidized. Factors that influence lipid oxidation are either intrinsic, such as the fatty acid composition of lipids (number and position of unsaturation), the presence of pro-oxidants (heme, metal ions, and enzymes) or natural antioxidants (tocopherols and carotenoids), and external factors such as temperature, light, oxygen, water activity, and the conditions of storage and processing [31]. The PI of fresh A. boyeri tissue was 176; the PI of sundried A. boyeri was 83. Although the conditions E1, E5, E6, and E7 had the highest values on the PI index, they also showed a greater sensitivity to oxidation than other conditions. The values that were closest to that of fresh fish were 187 (E2) and 195 (E4). The values that are closest to fresh fish are 187 and 195 and they correspond to drying conditions E4 and E2.