The Sander lucioperca (S. lucioperca) is an important freshwater fish species in Tunisian farming. It was introduced for the first time in Tunisia by the ONP (National Office of Fishing) in 1968 (CTA, 2009) in the dam of Nebhana using 30, 000 eggs which were incubated in floating boxes. Following its introduction, the recruitment of S. lucioperca in dams was made naturally and the stock recovered on its own. Currently, this species is present in most Tunisian dams (CTA, 2009). The production of S. lucioperca in Tunisian dams had a first phase which was characterized by an exponential increase of up to 192.4 tons in 2006. By 2010, the total capture of this species was estimated at 188.7 tons, representing 16.9% of the national production of freshwater fish in Tunisia. Most S. lucioperca production comes from the reservoir of Sidi Salem with an average of 84.28 tons/year, followed by the dams of Sidi Barrak with an average of 36.83 tons per year (DGPA, 2010).

The Tunisian fishery market is dominated by the marine species and the demand for freshwater fish is limited, despite their high nutritional value. The price of S. lucioperca is low and not adapted to the zander fishery costs. In order to develop the prospects for these markets, the image of fresh fish must be revitalized and its high nutritional qualities must be better shown. Like marine fish, S. lucioperca is a valuable source of nutrients and moisture (Guler et al., 2007). In Tunisia, this specie is sold as a fresh product. The application of a smoking process is important to encourage the consumption of freshwater fish and to conserve the highest value of tissues with good acceptance. Smoking, such as hot, cold, liquid and electrostatic treatments, is considered the most traditional conservation method in marine products with lower value to ameliorate it taste and adopt its sensory proprieties such as taste, color... etc. (Ünlüsayin et al., 2001). In addition, this traditional procedure has become a means of producing diversified products with high added value, since it offers an additional marketing preference for some aquatic species, such as freshwater fish, which are not so appreciated by consumers. In this context, the consumption of smoked freshwater fish has increased considerably in recent years because it preserves almost all the freshness of tissues (Rorvik, 2000).

Regardless of the diverse reports on the characteristics of S. lucioperca flesh, no information on lipid alteration or biochemical analysis of these freshwater fish is available following the smoking process. The aim of the current study was to assess the nutritional value of S. lucioperca fillets and the alterations in their lipid quality following hot and cold smoking treatments.

The total lipid and protein contents of fresh fillets were similar to those found in other species from freshwater, such as Clarias gariepinus (Kumolu-Johnson and Ndimele, 2001). Generally, the lipid content of the fillets varied among species (1.46 to 5.77 g/100g) and was affected by the fishing season (1.20 to 18.4 g/100g) (Osman et al., 2001). The ash levels of fresh fillets were similar to those found in other freshwater species such as Ctenopharyngodon idella and Cyprinus carpio (Afkhami et al., 2011). The apparent increases in lipid as well as some proximate component contents after both cold and hot smoking were mainly due to water loss by evaporation during the process. These increases were more pronounced after the hot smoking process than the cold one. In the current experiment, the difference between cold and hot smoking treatments could be attributed to the fillet composition, thickness, and weight, as well as to the smoking temperature and time. These findings are in agreement with those described by De Souza Franco et al., (2010) for Matrinxa fillets, in that the final moisture content was about 25% after hot and cold smoking processes.

The fatty acid profile reflects the highest nutritional quality of zander, which was similar to a previous study (Guler et al., 2007). Several reports have evidently shown the importance of fish lipids especially long-chain n-3 PUFAs for human health and nutrition (Sidhu, 2003). Significant differences (p < 0.05) were found in SFAs and PUFAs of the edible fillets after the smoking process as well as between the cold and hot processes. The apparent increases in the SFA amount, especially after hot smoking was mainly due to water loss by evaporation during this process.

The major fatty acids for the group of neutral lipids were palmitic acid (C16:0), oleic acid (C18:1n-9), and arachidonic acid (C20:4n-6). The high levels of C18:1 and C16:0 were in harmony with other reports (Osman et al.,2001; Passi et al., 2002), where the major SFAs and MUFAs identified in the fish lipids typically enclosed 16 and 18 carbon atoms, respectively. The noticeable reduction in PUFAs may indicate the degradation of these unsaturated fatty acids throughout smoking treatments. Such results were confirmed by the higher increases in PV and TBARS levels from the hot process than the cold one.

Polar lipids were similar to the neutral ones after both cold and hot smoking processes. In fact, PUFAs showed a significant variation after the hot smoking process. Such declines were in line with Azad shah et al., (2009), who reported significant decreases in PUFAs in herring filets during the smoking process. Because of its great amount of unsaturation, DHA declined by 26.5% after the hot smoking process. PUFAs are particularly susceptible to oxidation and degradation processes through enzymatic and chemical oxidation, which produces a large amount of volatile compounds (Kolakowska, 2002). The oxidative phenomena of membrane lipids might be one of the factors involved in the development of the distinguishing taste and flavor of the analyzed fillets during the smoking process. Triqui and Reineccius (1995) showed that lipid oxidation was associated with the decline in n-3 PUFAs, which was probably related to the change in the flavor in anchovy flesh during ripening.

The levels of peroxide value, malonedialdehyde (thiobarbituric acid reactive substances) and free fatty acids of S. lucioperca fillets following hot and cold smoking processes were lower than the values reported for Euthynnus lineatus (Chaijan et al., 2006). Higher PV levels were possibly due to the hot temperature of smoking and the high content of polyunsaturated fatty acids known to be very sensitive to oxidative reactions. Previous reports by Hawrysh (1990) showed that lipid oxidation occurs when oxygen molecules react with unsaturated fatty acids to produce lipoperoxides, which are influenced by various elements such as temperature, water activity, pH, and chemical environment (Ashie et al., 1996). According to Connell (1995), the acceptable limit for human consumption of PV of oils is 20 meq O₂/kg oil. In our present study, none the analyzed fillet oil samples reached such a value after the hot smoking process. Furthermore, hydroperoxides are decomposed into a wide range of carbonyl compounds, such as aldehydes, ketones, and other compounds which contribute to off flavors in foods as well as color and texture deteriorations (Kolakowska, 2002). Consequently, the increase in secondary oxidation products was determined through the thiobarbituric acid reactive substances.

The initial value for TBARS was 0.49 mg MDA/kg oil, signifying that lipid oxidation did not take place during the post-mortem treatment. Our experimental analysis showed that TBARS levels increased in both smoked fillets to reach a value of 1.31 and 0.78 mg MA/kg oil after hot and cold smoking, respectively, which was below the permissible limit reported by (Ozden and Erkan, 2006). Goulas and Kontominas (2005) revealed that the initial TBA value of chub mackerel (Scomber japonicus) amounted to 0.23 mg MDA/kg oil, with this value reaching 0.54 and 0.47 mg MDA/kg after the application of hot and cold smoking processes. The higher TBA value noted with the smoking procedure could be attributed to the dehydration of fish fillets and to the augmented oxidation of unsaturated fatty acids as a consequence of smoking at moderately high temperatures (up to 70 °C). Our results were in line with others given by Goktepe and Moody (1998), who detected a two-fold elevation in the TBA value of catfish fillets after smoking at a temperature up to 82 °C.

Free fatty acids (FFAs) were determined as a marker of lipolysis in the smoked fillets of S. lucioperca. The hydrolysis of acylglycerol esters (mainly triacylglycerols) is one of the imperative changes occurring in fish muscle lipids during post-mortem with the discharge of FFAs. In general, the development of FFAs in fish oil during storage is known to be related to some features such as the initial lipid content, the lipolytic activity and the temperature. The percentage of FFAs in fresh S. lucioperca represented 1.25% of total lipids. Such a level was lower than those found in herring oil by Aidos et al., (2001). Usually, the FFA content in crude fish ranges from 1 to 7% (Bimbo, 1998). The FFA level increased significantly (p < 0.05) after the smoking process compared to fresh fish to reach 2.27% and 4.29% of total fatty acids in cold and hot smoked fillets, respectively. Increases were significantly different between both treatments as well. Our obtained results indicated that both treatments were under the acceptable limit (IFOMA, 1981). Such a rise suggested that S. lucioperca lipid hydrolysis was produced by lipases or phospholipases, particularly in the hot smoking process. Azad Shah et al., (2009) reported that free fatty acid contents increased from 4.6 to 5.1% within 4 days of natural drying and then progressively increased up to 6.86% after 10 days.

The results showed that hot and cold smoking processes have significant effects on lipid quality. PV, TBARS and FFA levels increased more after the hot smoking process than the cold one, indicating a partial degradation of lipids. Our results suggest that S. lucioperca is a good source of lipids for the human diet. It is recommended to use the cold smoking process since it better conserves the chemical composition of tissues. Additional investigation on smoking conditions should be carried out to minimize such oxidation and hydrolysis in fillets, which could be associated to off-flavors and off-odors.