Fresh fish samples of M. hubbsi were provided by the fishery Mar del Chubut, Comodoro Rivadavia, in October 2011, March 2012, June 2012 and September 2012. Fifteen fish were randomly selected for each sampling, measured, weighed, eviscerated and then hand filleted. These procedures were carried out in a cold room (+4 °C). The different body parts were separated and stored on ice for 1–3 h before mincing. The raw material mixtures were separated into two groups: livers (L) and viscera without livers (V-L). The skin, head and backbone were not included in the experiments. Following Stansby (1986), the raw material from each group was pooled in a single sample and minced with a manual mincer with the addition of butylated hydroxytoluene as an antioxidant (Guil Guerrero et al., 2011). Both L and V-L samples were stored under nitrogen at −20 °C until analysis.

Moisture, ash and protein were analyzed according to the AOAC methodology (Official Methods of Analysis, 1990), total lipids were determined by Soxhlet (Hamilton, 1992), and carbohydrates were calculated by difference.

For the analysis of fatty acids, total lipids were transesterified according to Lepage and Roy (1984) with acetyl chloride in methanol 1:20 v /v. After the reaction, the tubes were cooled in water and the reaction mixture was diluted with 1 mL water and extracted three times with 1 mL hexane. The hexanic phase was dried under a nitrogen stream at atmospheric pressure and at room temperature. Fatty acids as fatty acid methyl esters (FAME) were re-suspended with 100 µL hexane for injection into a Hewlett-Packard 589 gas chromatograph with a flame ionization detector. A capillary column Varian Factor Four VF-23 ms (30 m×0.25 mm×0.25 µm; Varian INC Lake Forest, CA, USA) was used. Hydrogen was used as carrier gas. The injection split was set at 25/1. The temperature in the column was held at 120 °C for 1 min, then increased at 10 °C·min⁻¹ until 160 °C (ramp 1), and finally increased to 3 °C·min⁻¹ until 200 °C (ramp 2). The injector and detector temperatures were 240 °C and 260 °C, respectively. Determinations were carried out in triplicate. A standard of FAME (Supelco Inc., Supelco Park, Bellfonte, USA) was run under identical conditions to identify compounds on the basis of their retention times. Fatty acids were quantified using heptadecanoic acid (C17:0) as internal standard. Thus, an aliquot of 125 µg heptadecanoic acid dissolved in 5 µL toluene was added to the biological samples before transesterification. Quantitative data for fatty acid content represent the mean of three independent experiments.

The experiments were performed in triplicate and conventional statistical methods were used to calculate the mean and standard deviations. The results are presented as mean ± standard deviation of determinations for triplicate samples.

Biomass distribution in the samples, i.e. percent of weight for L and V-L, are shown in Table 1. Liver comprised 3.5% of the weight of the fish. Taking into account that the annual volume of capture in Chubut is on average 65,600 tons (Ministerio de Agricultura, Ganadería y Pesca de la Nación, 2014), nearly 2,296 tons of liver would be currently available for the extraction of oils.

A highest lipid recovery (27.0 to 41.8% of total lipids) was obtained from the liver fraction in all the months analyzed, although the highest was in June, which is autumn/winter in South America (Table 2). Guil Guerrero et al. (2011) reported a lipid recovery of 36% from M. hubbsi liver from European coasts in June, which is spring/summer in Europe. This value is similar to that obtained in this work in October (36%). Taking into account the high availability of lipids measured and the volume of capture reported for Chubut, it would be possible to obtain between 620–960 annual tons of liver oil. Protein levels in the viscera without livers (V-L) (11.9–14.1%) were higher than those in the liver (L) (8.8–10.9%) in all the months analyzed. The protein content in the rest of the viscera on dry basis was between 42 and 76%, whereas minerals were between 4.6 and 10%. This makes the rest of the viscera a very interesting material, because it can provide high-quality offal meal (Guillaume et al., 2001). Offal meal is very simple to obtain and has the advantage that it does not generate new residues (Yano et al., 2008).

The chemical composition of the FAMEs from the liver is shown in Table 3. All the months analyzed showed similar profiles, with palmitic acid (C16:0), oleic acid (18:1 n9), DHA (22:6 n3), EPA (20:5 n3) and palmitoleic acid (16:1) as the main constituents. Throughout the year, saturated fatty acids (SFAs) accounted for 29–33%, polyunsaturated fatty acids (PUFAs) accounted for 29–33%, and monounsaturated fatty acids (MUFAs) accounted for 36–39% of the total fatty acid composition (Table 4). Although the value of SFA content was higher than that reported for Uruguay and Spain (Mendez, 1997; Guil Guerrero et al., 2011), the PUFAs/SFAs ratio (0.9 to 1.1%) indicates that hake liver oil is a good supplement of PUFAs. Table 4 also shows that the total amount of n3 PUFAs (26–30%) was higher than that of total n6 PUFAs, with the latter being only 2–3% of the total lipid profile of hake liver. It is known that the n3/n6 ratio can be used as an index to compare the relative nutritional values of fish oils from different species (Huynh and Kitts, 2009). The n3/n6 ratio for hakes from San Jorge Gulf is very high (10.1–16.0), unlike that found for Mediterranean (4.3) and European hakes (9.4) (Guil Guerrero et al., 2011). The values reported for European, Mediterranean and Uruguayan hake species (Guil Guerrero et al., 2011; Mendez, 1997) indicate that the DHA content is higher than the EPA content. In this work, both EPA and DHA were found in similar proportions, at all catch times. The fatty acid content in the tissues comes from two different sources: de novo biosynthesis and diet. The latter varies with the habitat. EPA + DHA values reported for hakes from other parts of the world range between 21.3 and 24.9% (Mendez, 1997; Guil Guerrero et al., 2011). Values found in hakes from Chubut are higher (23.2–27.6%). According to that explained by Guil Guerrero et al. (2011), this difference could be due to the differences in water temperature in their respective habitats.

In aquaculture, the n3/n6 ratio is a very important feature. However, most of the commercial fish feed is based on plant oils, because of reduced costs (Tacon and Metían, 2008; Guillou et al., 1995). As a consequence, the n3/n6 ratio in the fish muscle is reversed, and a re-feeding period with fish oil-based feed before slaughter is necessary. The resulting fillets in general have lower nutritional quality in EPA than fish fed animal oil-based feed (Montero et al., 2005; Izquierdo et al., 2005). This is true for marine fish because they do not have the necessary enzymes (elongases and desaturases) for the conversion of EFAs like EPA, DHA and arachidonic acid from n6 linoleic acid (Izquierdo et al., 2005; Guillou et al., 1995). The absence of EFAs can affect the normal development of fish. As fish is the main source of PUFAs for humans (Montero et al., 2005), a deficiency in EFAs in fish can affect the health of consumers. Humans are also unable to synthesize EFAs (Guillou et al., 1995). On the other hand, extracting the lipid content from fish viscera increases the quality of fish meal, because the meal is priced on the basis on the protein content (Yano et al., 2008). Palmitic and oleic fatty acids are the main constituents of Patagonian hake oil. They are useful for formulations based on fish oil because fish use both acids as an energy source during the production of eggs and the spawning time, respectively (Medina et al., 2014). Fish oil is used mainly in the aquaculture industry, but increasingly for human consumption as well, with a better-paying market, particularly for nutraceutical purposes (Italy; Food and Agriculture Organization of the United Nations 2014). Regarding the high n3 PUFA content found in M. hubbsi liver oil from Patagonia, it is important to emphasize how beneficial the consumption of foods rich in n3 PUFAs is for human health. The beneficial health effects of the omega-3 fatty acids EPA and DHA include low rates of coronary heart disease, asthma, type 1 diabetes mellitus, multiple sclerosis, benefits related to cancer, inflammatory bowel disease, rheumatoid arthritis, psoriasis, and mental health. Moreover, the balance of omega-6/omega-3 fatty acids is an important determinant in maintaining homeostasis, normal development, and mental health throughout the human life cycle (Simopoulus 2011a).