MATERIALS AND METHODS

Sample Preparation
Seventy six golden mullet (Liza aurata Risso, 1810) samples were obtained from Southern Caspian Sea in October-December, 2006. The average lengthof golden mullet was 41.75 cm and the average weights of the fish and its roe were about 0.875 and 0.208 kg, respectively. Prior to analysis, the fish were gutted and head, tail, fins, viscera and skin were removed. Fish roes were used as raw caviar. All fish and roe samples were immediately transferred to the laboratory.

Lipid Extraction and Fatty Acid (FA) Analysis
The fish were filleted and homogenized. Lipids were extracted from the homogenized edible portion of flesh by the standard AOAC official method 948.15 (Helrich, 1990) using petroleum ether for 6 h in soxhelet extraction apparatus. Lipids of the roe samples were obtained using the same method after the membrane were removed from roes. The fatty acid compositions of golden mullet fillet and roe oil were determined by Capillary Gas Chromatography (CGC) technique. For the determination of fatty acid composition, the oil samples were converted to their corresponding methyl esters by BF3-methanol esterification by the AOCS official method Ce 2-66 (AOCS, 1972).

The fatty acid methyl esters were quantified by gas-liquid chromatography method using a packed column, D.E.G.S -15% and Flame-Ionization Detector (FID) in Simadzo-14 A gas chromatograph (Japan). Helium (99.999% pure) was used as the carrier gas at a flow rate of 45 mL min^–1. The detector and injector temperatures were chosen as 210 and 200°C, respectively. The oven temperature was set to 140°C for 5 min and heated to 190°C with a heating rate of 5°C min^–1 and the temperature of the oven was also isothermal. Peaks were identified by comparing the retention times with those of a mixture of standard methyl esters (Sigma Chemical Co. Ltd., Poole, UK). All of the other chemicals used in the experiments were analytical grade (Merck, Darmstadt, Germany).

The lipid and fatty acid contents between fillet and roe oil were compared statistically, using t-test and Excel for windows, XP.

RESULTS AND DISCUSSION

The fatty acid compositions of golden mullet and its roe oil are shown in Table 1. In addition, the averages of its fatty acids series are shown in Table 2.

The results showed that palmitic acid (C16:0) was the dominant saturated fatty acid in all samples. The predominance of C16:0, C16:1 and C18:1, fatty acids in golden mullet and its roe may be attributed to the fish diets (Sengor et al., 2003). Fatty acid composition of fish lipid was highly dependent on a number of factors, especially fish diets (Fowler et al., 1994; Karakoltsidis et al., 1995; Okkes et al., 1996; Sathivel et al., 2002).

Table 1:	Fatty acid composition of Caspian Sea golden mullet fillet and roe oil (Liza aurata) (g/100 g lipid)
The data are expressed as the average of three samples. Different letter(s) show significant difference at p<0.05 in fillet and roe oil

Table 2:	Average of fatty acids series in Caspian Sea golden mullet fillet and roe oil (Liza aurata) (g/100 g lipid)

Levels of palmitoleic (C16:1) and oleic (C18:1) acids of golden mullet fillet oil were 17.32 and 17.09%, respectively. Other unsaturated fatty acid contents of golden mullet fillet oil, such as linoleic (C18:2), α-linolenic (C18:3) and docasahexaenoic (C22.6) were significantly higher than roe oil (p<0.05) with 5.96, 8.72 and 3.55%, respectively. Table 2 shows total lipid contents in roe and the fillet are 10.32±1.33% and 9.25±1.05%, respectively.

Major fatty acids of roe oil were C16:0, C16:1, C18:1, C18:2 C18:3, C20:5 and C22:6. These results agree with the results reported by Lu et al. (1979) and Sengor et al. (2003). They reported that the major saturated and unsaturated fatty acids of gray mullet roe were C16:0, C16:1 and C18:1, respectively. Palmitic acid contents of fillet and roe oil were 14.39 and 6.45%, respectively.

Table 2 shows fatty acids profiles contents of roe oil comparing with those of mullet fillet oil. The total monounsaturated fatty acid contents of mullet fillet and roe oil were 34.41 and 40.48%, respectively.

On the other hand, total polyunsaturated fatty acid contents of mullet fillet and roe oil were 21.96 and 27.75%, respectively.

Regarding to the lipid percentage and the amounts of C14:0, C16:0, C18:2, C18:3, C20:5 and C22:6 fatty acids, there is different significance between the fillet and roe, statistically (p<0.05). The difference between ω-3 and ω-6 unsaturated fatty acids in both fillet and roe oil was also significant, statistically (p<0.05).

Table 2 show total of saturated, unsaturated, mono and polyenoic fatty acids were 22.49, 56.37, 34.41 and 21.96% in fresh fillet, respectively. Also, the mentioned (FA) series were 10.98, 68.59, 40.48 and 27.75% in fish roe oil, respectively.

Total lipid contents in roe were higher than the fillet (Table 2). In addition, rate of ω-3 + ω-6 was 21.96 and 27.75% in fillet and roe oil, respectively. These results defines that golden mullet is one of the best sources for unsaturated oil.

In conclusion, the present study revealed the fatty acid compositions of golden mullet fillet and roe oil in spawning period of golden mullet, which have not previously been studied in Iran. C16:0, C16:1, C18:1, C18:3 and C22:6 were the most abundant fatty acids in fish and roe oil samples. It has been known there is significant seasonality in genders in amount of fatty acid in some fish like Capoeta capoeta umbla (Okkes et al., 1996) and Esox luxius (Hedayatifard and Shabani, 2005). To test this hypothesis, further experiments should be carried out on golden mullet fillet oil of different seasonal periods.

ACKNOWLEDGMENTS

Author thank the Prof. Dr. Amin Keyvan (Islamic Azad University), Prof. Robert G. Ackman (Canadian Institute of Fisheries Technology) and Dr. M. Yousefian (Caspian Sea Ecology Research Center) for their scientific helps. Author thank also the Persia Fisheries Industries Company for their support.