CHANGES IN CHEMICAL AND PHYSICOCHEMICAL CHARACTERISTICS DURING THE PRODUCTION OF TRADITIONAL SREMSKA SAUSAGE

: The aim of this trial was to investigate changes in chemical and physico-chemical parameters during the production of traditional Sremska sausage (dry fermented sausage) from pork of three pig breeds: Mangalitsa (MA), Moravka (MO) and Swedish Landrace (SL). Analyses of all variants of sausages were carried out after stuffing (day 0) and on production days 3, 7, 14 and 21. The reduction in moisture during production caused the increase in protein, fat and ash contents (p<0.001) in all three variants of sausages, were found to be within the range for this type of sausages. Higher fat content in MA and MO sausages compared to SL variant was most likely a result of the different chemical composition of the meat from pigs of autochthonous breeds. All three sausage variants had a similar final pH value, but the mildest drop of pH was determined in MA sausages. Pig breed significantly affected (p<0.05) all three indicators of oxidative changes (thiobarbituric acid value, peroxide value and free fatty acid content). It was found that they were higher in SL compared with MA and MO sausages and to significantly increase during the production process.


Introduction
Today, dry fermented sausages are produced from meat of modern pig breeds and manufacturing technology is based on the use of controlled ripening rooms and rapid curing techniques, resulting in lower production time and improved product safety (Flores et al., 1997;Marco et al., 2008). The sausages obtained are excellent in appearance, but their typical sensory characteristics are poor and most often they have a vigorous acidic taste that is not accepted by the consumer (Sanz et al., 1998). Opposed to them, traditionally produced dry fermented sausages are made from meat of late maturing breeds of pigs and by spontaneous meat fermentation at low temperatures, without additives (nitrate, nitrite, glucono-γ-lactone, etc) and starter cultures (Marcos et al., 2007). Sausages produced in this way have usually very high sensory quality.
Sremska sausage is popular dry fermented sausage in Serbia and all Balkan region. It is characterized by specific hot taste, aromatic and spicy flavour, dark red colour and hard consistency (Stanišić et al., 2012). Traditionally, it was produced from the meat of autochthonous pig breeds such as Mangalitsa and Moravka and in traditional smoking house. Today, however, it is produced from modern pig breeds in controlled ripening rooms and with the use of rapid curing techniques, which gives them a slightly modified characteristics in comparison to traditional products.
The scientific knowledge of traditional produced Sremska sausage is limited and its quality is very variable, because there is very little uniformity in the production by different homemade producers and meat industries. In order to preserve the quality of traditionally Sremska sausage and to describe production process, this trial was set to investigate changes in chemical and physico-chemical parameters of Sremska sausage manufactured in traditional manner from pork of three pig breeds: Mangalitsa, Moravka and Swedish Landrace. Mangalitsa and Moravka breeds were selected as autochthonous Serbian pig breeds (Petrović et al., 2010), while Swedish Landrace, was chosen as the most represented commercial pig breed in Serbia.

Materials and Methods
All pigs used in the study were bred at the farm of the Institute for Animal Husbandry (Belgrade, Serbia). The diet of animals consisted of concentrated commercial feed administered "ad libitum". Water was provided using automatic feeding troughs. Pigs were slaughtered when they attained their target slaughter weight of 105 ± 5.0 kg. Sremska sausages were prepared in a meat processing plant of the Institute for Animal Husbandry (Belgrade, Serbia). Three different types of Sremska sausages were produced, each from the meat of different breeds of pig: Swedish Landrace (SL), Mangalitsa (MA) and Moravka (MO).
For the production of Sremska sausage, meat from shoulder was used and back fat in the ratio of 75:25. All three variants were produced on the same day and in an identical manner: meat and fat were chopped and minced to 8 mm particle size and mixed with in a cutter (Seydelman K60, Germany), whereupon they were transferred to a mixer and the same amounts of ingredients were added: 2.2% NaCl, 0.3% glucose, 0.17% garlic (powder), 0.55% hot red paprika (powder) and 0.5% sweet red paprika (powder). The sausage mixture was stuffed into natural Changes in chemical and physicochemical … 707 casings (pig small intestines) of around 32 mm diameter. After stuffing (day 0) the sausages were drained in a cold store (4 ± 1°C) for 12 h, for the surface to dry, after which they were hung to dry in a traditional smoking house. The ripening was as follows: the first stage lasted 14 days in a traditional smoking house at 10-15°C with 75-90% relative humidity (RH), where they were smoked for 6 h each day; last 7 days (from day 15 to day 21) sausages were processed in a drying room at 14-16°C with about 75% RH, to reach about 35.0% moisture content. The total processing time was 21 day. Sampling of all variants of sausages was carried out after stuffing (day 0) and on production days 3, 7, 14 and 21.
The proximate composition of sausages was determined in the following manner: moisture content by drying samples at 105°C (ISO 1442(ISO , 1997; protein content by Kjeldahl method and multiplying by factor 6.25 (ISO 937, 1978); total fat content by Soxhlet method (ISO 1443(ISO , 1973, and ash content by mineralization of samples at 550 ± 25°C (ISO 936, 1998).
pH value was measured by pH-meter Hanna, HI 83141 (Hanna Instruments, USA), equipped with an puncture electrode. The pH meter was calibrated using standard phosphate buffers (ISO 2917(ISO , 1999. The 2-thiobarbituric acid (TBA) method was performed according to Tarladgis et al. (1960). The absorbance was measured on a spectrophotometer (Spekol 1300, Analytic Jena, Germany) at 532 nm. TBA values were calculated against standard curve of malonaldehyde (MDA) and expressed as mg MDA/kg sample.
Peroxide value was determined by method described in the ISO 3960 (1977), and peroxide values were expressed as milliequivalents of active oxygen per kg of fat (mEq O 2 /kg).
Free fatty acid content was expressed as g oleic acid/100 g fat, after the titration with 0.1N NaOH and determining the total acidity, as described in ISO 660 (2011).
An analysis of variance (ANOVA) using the General Linear Model (GLM) procedure of the SPSS 20.0 software (IBM SPSS Statistics, Version 20, IBM Corp, USA) was performed for all variables considered. If the effect of main factor (breed or time) was found significant, LSD test was used to evaluate the significance of difference at p<0.05.

Results and Discussion
The changes in proximate composition during ripening of three variants of Sremska sausages are shown in Table 1. The reduction in moisture during ripening caused the increase in protein, fat and ash contents (p<0.001) in all three variants of sausages. The fat contents of the sausages on the day of preparation were close to, but did not exactly match between variants. Sausages made from the meat of Mangalitsa (MA) and Moravka (MO) at the beginning (day 0) had the lower moisture content (p<0.05), but also the higher fat content (p<0.01), compared with SL sausages. This discrepancy was most likely a result of the different chemical composition of the meat from pigs of autochthonous breeds, which have been found to have a higher intramuscular fat content, compared with modern pig breeds (Petrović et al., 2010).
At the end of ripening, moisture content decreased to the level from 27.16% (MA) to 30.21% (SL). Such low moisture content is typical for similar products in Greece, Hungary and Croatia (Kozacinski et al., 2008). Although a SL sausages had significantly the highest water content at the beginning (p<0.05), at the end of the production process (day 21) no significant difference in water content between SA and MA variants were establish. This could be explained by higher share of intramuscular fat in MA sausages, which was reported to extent the dehydration process by decreasing moisture diffusivity coefficient (Arnau et al., 1997).
Some studies have indicated the occurrence of lower protein content in meat from autochthonous pig breeds compared to meat from commercial pig breeds (Kim et al., 2008;Parunović et al., 2012), thus partly explaining the slightly lower protein content at the end of the ripening (day 21) in MA and MO sausages compared with SL variants (p<0.05), in the current study.
The ash values significantly decreased with time in all three variants of sausages (p<0.001), which is in disagreement with findings of (Salgado et al., 2005), who have reported that ash content remained constant during the ripening process of Chorizo, a traditional Spanish fermented sausage. 2.91 ± 0.18 a 3.12 ± 0.09 b 4.20 ± 0.16 cA 5.18 ± 0.14 d 5.49 ± 0.11 eAB *** p ns ns ** ns * a-e Different letters within the same row denote significant differences between means A-C Different letters within the same column denote significant differences between means ns (p≥0.05); * (p<0.05); ** (p<0.01); *** (p<0.001) During the fermentation lactic acid is formed, and the pH is reduced (Lücke, 1994). In this trial pH dropped moderately and reached the minimum on day 14 of the process, and remained approximately the same until the end of the production process (day 21), for sausage variant MO and SL ( Figure 1). As contrast to them, the variant MA had a significant decline in the pH value from the fourteenth to the twenty-first day when the minimum pH is determined. As the fat content was highest in the MA sausages (Table 1), this results is in agreement with the trend reported in similar studies (Olivares et al., 2010;Lorenzo and Franco, 2012), where a greater decrease of pH values were obtained for low fat sausages. However, some authors found no significant effect of fat level on pH (Liaros et al., 2009). At the end of the production process, all variants of sausages had similar pH value, which ranged from 5.0 to 5.1, which is lower than values reported by other authors (5.2 to 6.4), for naturally fermented dry sausages (Comi et al., 2005). These established lower pH values are probably the result of added sugar in the sausage stuffing and the possible presence of sugar in spices, such as paprika (Oberdick, 1988). The oxidative process during ripening of Sremska sausages was evaluated by the TBA values, peroxide index and FFA content in order to determine how it was affected by the different meat type (Figure 2).
Initial TBA values were approximately the same in all three variants of sausages, however, by the end of the production process SL sausages had the highest TBA value compared with MA and MO variants (Figure 2a). TBA values increased in all three variants of sausages during production through the manufacture process and this was also reported by other authors (Fanco et al., 2002;Lorenzo and Franco, 2012). In this trial, the values for this parameter, both in the initial phase and in the final product, are lower than those found in the literature for other dry-fermented sausage varieties (Lorenzo et al., 2000). The relatively low TBA values at the end of the process (0.12-0.19 mg MDA/kg) were similar to those described by Fanco et al. (2002).
Peroxide values increased significantly (p<0.05) up to 14 days and then slightly decreased at day 21, in all three sausage variants (Figure 2b). The peroxide values in this trial were lower than one alleged by Fanco et al. (2002) of 16 and 28 meq O 2 /kg in Androlla (a Spanish dry-cured pork sausage) at 0 and 14 days, respectively, and Salgado et al. (2006) who found that the average peroxide value in homemade Chorizo was 12.85 meq O 2 /kg of fat. Pig breed had a significant effect on the value of peroxide, which was higher in the SL sausages compared to MO and MA sausages (p<0.05).
The FFA (expressed as % of oleic acid) during the production of Sremska sausages (Figure 2c), increase slightly in the first seven days of ripening and no statistically significant difference were established between the groups of sausages. However from day 7, there was the increase in the FFA content, which was very similar in MA and MO sausages, as opposed to SL variant where the increase of FFA content was more rapidly until the end of ripening (p<0.05). These results are in agreement with the results of Bañón et al. (2010), that sausages made from meat of indigenous breeds of pigs had significantly lower acidity values (p<0.001) compared to those made from the meat of modern pig breed.
As shown in Figure 2 pig breed significantly affected (p<0.05) all three indicators of oxidative changes. It was found that they were higher in SL compared with MA and MO sausages. This results, are in disagree with findings of Liaros et al. (2009) andFranco (2012), that the higher degree of oxidative changes are established in sausages with higher fat content. However, a different fatty acid composition of meat from autochthonous pig breeds may be one of the reasons for this inconsistency.

Conclusion
Traditionally produced Sremska sausage made from meat of autochthonous pig breeds (Mangalitsa nad Moravka), compared to the one produced from meat of Swedish Landrace breed, is characterised by higher fat content and consequently lower water content throughout the production process. The reduction in moisture during ripening caused the increase in protein, fat and ash contents in all three variants of sausages.
The mildest drop of pH was determined in MA sausages, although the final pH values were approximately the same in all variants. Oxidative changes, evaluated by the TBA values, peroxide index and FFA content, were significantly increasing during the production process, but were higher in SL compared with MA and MO sausages.

Acknowledgment
This paper is a result of the research within the projects TR31081, financed by the Ministry of Education, Science and Technological Development, Republic of Serbia.