PaPer 96 Ital. J. Food Sci., vol. 28 - 2016 - Keywords: CIE L*, a*, b*, horse Salami, microbiological aspect, physicochemical properties, Sensory attributes, texture profile analysis - PHYSICOCHEMICAL, MICROBIOLOGICAL AND COLOUR ATTRIBUTES OF HORSE SALAMI ESTABLISHED DURING THE RIPENING PERIOD D. KOvAčEvIća, K. MASTANjEvIća *, j. PLEADINb and j. FRECEc aDepartment of Food Technology, Faculty of Food Technology University of Osijek, F. Kuhača 20, HR-31000 Osijek, Croatia bLaboratory for Analytical Chemistry, Croatian Veterinary Institute, Savska 143, HR-10000 Zagreb, Croatia cDepartment of Biochemical Engineering, Faculty of Food and Biotechnology University of Zagreb, Pierrotijeva 6, HR-10000 Zagreb, Croatia *Corresponding author: Tel. +385 31 224 298 Fax +385 31 207 115, email: kresimir.Mastanjevic@ptfos.hr AbstrAct changes in physicochemical, colour, textural, microbiological and sensory attributes occurring during the processing of Horse salami and established on manufacturing days 0, 7, 14, 21, 28, 42, 60, 90 were studied. significant changes (P<0.05) in physicochemical parameters attributa- ble to moisture loss, as well as changes in colour and textural properties were observed during the fermentation and ripening stage. Proteolysis and lipolysis, coming as a result of endogenous enzymatic activity and high lactic acid bacteria and staphylococci counts, contributed to specific organoleptic properties of the final product. sensorial profiling showed a significant (P<0.05) acid taste, lactic acid odour and flavour intensity, and low fat/lean ratio and smokiness and saltiness values. Final Horse salami products were microbiologically safe, the dominant microbial popu- lation thereby being Lactobacillus plantarum, Lactococcus lactis ssp. lactis, Enterococcus faecium and Staphylococcus xylosus. mailto:kresimir.Mastanjevic%40ptfos.hr?subject= Ital. J. Food Sci., vol. 28 - 2016 97 INtrODUctION Horse salami, an indigenous croatian meat product, is a dry fermented sausage made of horse meat supplemented with pork fatback, salt and spices. In croatia, the tradition of Horse sa- lami production is kept by the Italian minority populating the eastern part of the country (in specific, the western slavonian region). In the past, this product had been known as “the dish of the poor”; nowadays, it represents a highly ap- preciated autochthonous croatian meat product having a great potential to become a PGI (Pro- tected Geographical Indications) & PDO (Protect- ed Designation of Origin). Although horse meat has a high nutrition, as well as a high miner- al value (due to its vitamin b and iron content, respectively) (bADIANI et al., 1997; FrANcO et al.,, 2011), human consumption is negligible in comparison with other conventional types of meat like pork, beef or chicken (LOMbArDI et al., 2005). Horse meat used for the production of Horse salami is obtained from horses slaugh- tered at the end of their (5-year or longer) life- cycle. the meat has no appreciable organolep- tic qualities. Its original colour is deep red lard- ed with yellow fat, while the meat is tough to chew due to the connective tissue maturation (LItWINcZUK et al., 2008; tAtEO et al., 2008). Horse salami has specific sensorial proper- ties (smell and taste) attributable mainly to dry- ing and smoking, but also to ripening, as well as to enzymatic, lactic acid bacteria and mould activity. the recipe is 130 years old and the sole difference in final products coming from various producers boils down to the difference in mass fraction of fatback used in the salami preparation (ranging from 12 to 15%). the pro- duction of the traditional Horse salami main- ly takes place on small farms; we are therefore talking a small-scale production seasonal in its nature, fluctuating on a year-by-year basis de- pendent on weather conditions. In light of the foregoing, standardization of the Horse salami production becomes imperative. Dry sausages produced in various European countries, main- ly spain and Italy, have been extensively stud- ied for their physicochemical composition, col- our and textural properties (cAsIrAGHI, et al., 1996; GIMENO et al., 2000; brUNA et al., 2001; sPAZIANI et al., 2009). However, scientific infor- mation on this croatian indigenous dry sausage, which would efficiently contribute to its charac- terization and production standardisation, is vir- tually non-existent. therefore, the aim of this study was to inves- tigate, for the first time ever, physicochemical composition, microbiological and sensorial at- tributes of the dry-fermented sausage known as Horse salami and the changes occurring during 90 days of its manufacturing. Investigations also included instrumental measurements of colour and texture of the studied salami on certain pro- cessing days, as well as the isolation and identi- fication of autochthonous microbial population and gathering of other data needed for micro- biological safety evaluation of the final product. MAtErIAL AND MEtHODs The manufacturing process samples of traditional Horse salami (24 units) were manufactured in a small-scale facility in the western slavonian region (the Eastern croatia). All samples were prepared using traditional pro- cedures that made no use of additives such as starter cultures supplemented with nitrites, ni- trates or ascorbic acid (namely, the production of traditional croatian meat products does not involve the use of additives). such a tradition- al production takes about 3 months (90 days). Horse salami is made of meat of older (5+ years), worn-out horses, mainly of the Hrvatski Posavac breed. After slaughtering, fat and connective tis- sue are carefully removed from the horse meat. this is especially important when it comes to fat, because horse fat has a particularly unpleasant smell and taste. the meat is then grinded using a grinding plate having holes measuring 6 mm in their diameter and left to rest overnight (12 hours at the minimum) in a special container equipped with a decantation hole. Grinded horse meat is then mixed with pig fatback represent- ed in the amount of 12%. before its mixing with the horse meat, the fatback is grinded using a grinding plate having holes measuring 10 mm in their diameter. the mixture of meat and fat is then mixed with salt added in the amount of 2.2%, red paprika powder added in the amount of 0.2%, hot red paprika powder added in the amount of 0.3%, garlic added in the amount of 0.2%, and black pepper added in the amount of 0.3%. In the subsequent course, the mixture gets to be stuffed into a horse small intestine (roughly 50 cm long and 50 mm wide in diame- ter) or into collagen casings (of the same dimen- sions). thereafter, the Horse salami is smoked on a dry hard wood (hornbeam, beech and its sawdust) every few days (for 2-3 hours) for the total of four weeks. At this stage, the tempera- ture and relative humidity should be kept at 18 to 20ºc and 70 to 90%, respectively. After smok- ing, the Horse salami is left to ripen. this stage is the longest and should take about two months, throughout which period the salami should be kept in a dark room at the temperature rang- ing from 14° to 17ºc, with the relative humidity ranging from 70 to 80%. After that, Horse sala- mi is ready for consumption. Within this study frame, samples of Horse salami were taken on the processing days 0, 7, 14, 21, 28, 42, 60 and 90. In total, 24 samples were produced; at each processing stage, three samples were taken for the analyses. 98 Ital. J. Food Sci., vol. 28 - 2016 ANALytIcAL MEtHODs Physicochemical parameters before the analysis, the sausage samples were homogenised using a knife mill Gridomix GM 200 (retsh, Germany) and prepared according to IsO 3100-1:1975. Water content was determined gravimetri- cally (IsO 1442:1997) at 103°c (Epsa 2000 bari, croatia), while the ash content was estab- lished according to IsO 936:1998, by virtue of burning the samples at 550 °c (LV9/11/P320 Nobertherm, Germany). total protein content was determined using the Kjeldahl method (IsO 937:1978) that made use of an Unit 8 basic di- gestion block (Foss, sweden) and a Kjeltec 8400 automated distillation & titration device (Foss, sweden). the total fat content was determined using the soxhlet method (IsO 1443:1973), which involves digestion of a sample in acidic environment followed by fat extraction with pe- troleum ether using a soxtherm 2000 Automat- ic device (Gerhardt, Germany). the determina- tion of collagen content was performed through the analysis of hydroxyproline according to IsO 3496:1994 that made use of a spectrophotome- ter (Hach Dr/4000U, Germany). sodium chlo- ride content was determined using the internal titration method (trAJKOVIĆ et al., 1983). In this analysis, 2 g of each sample were homog- enized with sand and 3 mL of water. the con- tent was transferred into a 100 mL-volumetric flask, stirred and placed for 15 min into a water bath at 100 °c. After cooling, the flask was filled with water up to the mark and filtered. An ali- quot (25 mL) of the filtrate was transferred into an Erlenmeyer flask containing a few drops of K 2 crO 4 indicator (62 g/100 mL of water) and ti- trated with 0.1 M-AgNO 3 until a persistent red- dish colour was obtained. sodium chloride con- tent was calculated based on the expenditure of titration reagent and its concentration. pH values were determined in a homogenate diluted with distilled water (1:10, p/v) using pH/ Ion 510 – bench pH/Ion/mV Meter (Eutech In- struments Pte Ltd/ Oakton Instruments, UsA) according to the pH/Ion 510 Instruction Man- ual. Water activity (a w ) was determined at the room temperature (20°±2ºc) using a rotronic Hygrolab 3 (rotronic AG, bassersdorf, switzer- land). All chemicals used for analyses of physico- chemical parameters were of an analytical grade. For each sample, three independent measure- ments were made. Instrumental determination of colour Instrumental colour measurements (those of L*, a*, and b* values) were performed using a Hunter-Lab Mini scanXE (A60-1010-615 Mod- el colorimeter, Hunter-Lab, reston, VA, UsA). the instrument was standardized on each oc- casion using a white ceramic plate (L 0 = 93.01, a 0 = -1.11, and b 0 = 1.30). the cIELAb space values (L*, a* and b*) (cIE, 1976) correspond to lightness, greenness (-a*), redness (a*), blue- ness (-b*) or yellowness (b*). the colour meas- urements performed on the Horse salami took place at the room temperature (20°±2ºc). Each sample was cut in slices and colour-measured at ten different spots. Texture Profile Analysis texture Profile Analysis (tPA) was performed using a tA.Xt2i sMs stable Micro systems tex- ture Analyzer (stable Microsystems Ltd, surrey, England) equipped with a P/75 aluminium cylin- drical probe. this involved cutting the samples into 1.5 cm-thick slices and their double com- pression so as to downsize them to 40% of their original thickness. Force-time curves were re- corded at the across-head speed of 5 mms-1 and at the same recording speed. the following pa- rameters were quantified (bOUrNE, 1978): hard- ness (kg), i.e. the maximum force required to compress the sample; springiness (ratio), i.e. the ability of the sample to recover its original form after the cessation of the deforming force; cohe- siveness (ratio), i.e. the extent to which the sam- ple could be deformed prior to rupture; chewi- ness (kg), i.e. labour required to masticate the sample before swallowing, which represents the product of hardness multiplied by cohesiveness and springiness; and finally resilience (ratio), so as to determine how well the product “fights to regain its original position”. these parameters were obtained using the texture Expert for Win- dows (Version 1.0) stable Micro systems. With each sample, eight determinations of texture pa- rameters were made. Microbiological analysis After aseptically removing and discarding the casing, 10 g of the product were recovered in an aseptic manner, homogenized in 90 ml of the sterile 0.5%-saline solution and serially dilut- ed before their planting on a non-selective (pep- tone yeast extract glucose agar, biolife, Milano, Italy), PcA-agar (standard plate count agar) (bi- olife, Milano, Italy) and the following selective media: Mrs-agar (biolife, Milano, Italy) intended for lactic acid bacteria growth and baird-Parker agar (Merck, Darmstadt, Germany) intended for staphylococci growth. the plates were incubated under conditions specified in table 1. Isolation and identification of microbial population in the final product classical microbiological and biochemical (API) methods (table 1) were used for the isola- tion and identification of the natural microbial population in the traditionally produced Horse Ital. J. Food Sci., vol. 28 - 2016 99 salami (i.e. in the final product obtained after 90 production days). ten grams of the sample were homogenized in 90 mL of sterile 0.5% sa- line solution and serially diluted before planting on a non-selective medium (peptone yeast ex- tract glucose agar, biolife, Milano, Italy) and se- lective media under conditions specified in table 1. colonies randomly taken from selected plates were identified on the basis of their morphology, Gram-staining, cell morphology and catalase re- action. the identity of bacteria species was fur- ther confirmed using the API identification kits (bioMérieux, France). Sensorial analysis the final Horse salami product (obtained af- ter 90 days) was subjected to a quantitative de- scriptive analysis performed by a panel of seven (3 male and 4 female) trained experts according to IsO 6658:2005 standard. the panellists had completed a preliminary three session-training in order to familiarize themselves with the sam- ples under investigation. Fourteen attributes were examined and rated on a 5-point scale, “1” thereby standing for “poorly perceived or absent” and “5” standing for “intensely perceived”. Dur- ing these three training sessions, the descrip- tors to be targeted by the analysis were agreed table 1 - classical microbiological and biochemical (API) methods of isolation and identification of microbial population applied in the Horse salami analyses. Microorganism Nutrient media Incubation conditions API test Salmonella sp. RP-broth, XLD 37°C API 20 E (Biolife, Italy) 24-48 h V4.1 Enterobacteriaceae VRBG 37°C API 20 E (Biolife, Italy) 24 h V4.1 Staphylococcus aureus BP 37°C API Staph (Biolife, Italy) 48 h V4.1 Coagulase negative staphylococci (CNS) BP 37°C API Staph (Biolife, Italy) 48 h V4.1 Sulphite reducing clostridia Sulphite agar 37°C - (Biolife, Italy) 72 h Listeria monocytogenes Fraser broth 37°C API Listeria Palcam agar 24 h V1.2 (Biolife, Italy) Lactic acid bacteria MRS agar 30°C API 50 CHL (Biolife, Italy) 48-72 h V5.1 API 20 STREP V7.0 Yeasts Sabouraud agar 25°C API 20 C (Biolife, Italy) 48-72 h AUX V4.0 Yeasts upon. the latter included as follows: 2 external attributes (appearance, hardness), 4 attributes descriptive of a slice (fat/lean ratio, easy peeling capability, colour intensity, sliceability), 5 attrib- utes descriptive of perceptions during mastica- tion (flavour intensity, juiciness, smokiness, acid taste, saltiness) and 3 attributes descriptive of the product smell (spice odour, lactic acid odour, mould odour). the sausage samples were coded using a three-digit code and presented in form of oblique slices approximately 0.4 cm thick. Wa- ter was provided to clean the panellists’ palate between analyses. Data analysis Differences between the average values of the same physicochemical, colour, texture, microbi- ological and sensory parameters were analyzed using the analysis of variance (ANOVA) and the Fisher’s least significant difference test (LsD), with statistical significance being set at P<0.05. Moisture, fat, protein, collagen and Nacl con- tent, pH, a w, colour and textural parameters were subjected to correlation analysis (Pearson´s cor- relation test) so as to determine their possible statistically meaningful relationships. statisti- cal analysis was carried out using statistica Ver. 8.0 statsoft Inc. tulsa, OK, UsA. 100 Ital. J. Food Sci., vol. 28 - 2016 rEsULts AND DIscUssION Physicochemical parameters basic chemical composition, salt (Nacl) con- tent, pH values and water activity (a w ) of the Horse salami, established at various processing stages, are given in table 2. the average initial moisture content of the Horse salami found to be 61.91% had significantly decreased (P<0.05) as the processing went on due to smoking and dry-ripening typical of dry fermented sausag- es (LIZAsO et al., 1999; PErEZ-ALVArEZ et al., 1999; sALGADO et al., 2005; sALGADO et al., 2006; LOrENZO et al., 2012). Higher moisture losses were observed in the first 21 processing days and on day 28, which is characteristic for this type of product (< 40%) and dry sausages in general (PLEADIN et al., 2014). Further rip- ening leads to additional moisture content re- duction, so that the lowest value (28.51%) was determined on manufacturing day 90. In 2012, LOrENZO and co-workers reported higher ini- tial and final moisture values for the foal sal- chichon. this can be explained by the fact that horse meat has a lower water content as com- pared to foal meat (LItWINcZUK et al., 2008; LAN- ZA et al., 2009; tAtEO et al., 2008), as well as by the longer ripening period of the Horse salami. the final moisture content was also lower than in similar dry sausages coming from spain (GI- MENO et al., 2000; rUbIO et al., 2007; LOrEN- ZO et al., 2012), which can also be attributed to a longer ripening period of the Horse salami. the highest amount of proteins (30.53%) was determined on day 90. the results are consist- ent with the published literature data, which show that due to prolonged drying and ripen- ing (weight loss of up to 50%) and a high share of lean meat used in stuffing preparation, mois- ture and protein content in ripened dry-ferment- ed sausages tend to be similar (30-40%), indi- cating a high nutritional value of the final prod- uct (PLEADIN et al., 2014). the average fat content of the Horse salami had increased significantly (P<0.05) from day 1 to day 90 (from 13.84 to 28.54%), in proportion to the duration of the Horse salami ripening pro- cess and dehydration, i.e. the continuous reduc- tion of water content in the product; the same goes for the protein and collagen content (table 2). Fat as a substantial component of fermented sausages has multiple functions; it represents a concentrated energy source (9 kcal/g) and the source of essential fatty acids and fat-soluble vi- tamins (MELA, 1990). Furthermore, it is contrib- uting to the fullness of flavour, texture and soft- ness of the product, all of the aforementioned be- ing relevant for the quality and acceptability of the product in question (OLIVArEs et al., 2010). Hydrolysis and oxidation of fatty acids that oc- cur during the ripening process largely contrib- ute to the taste of fermented sausages (OrDON- EZ et al., 1999). the final fat content was low- er, while the final protein content turned out to be higher than in spanish and Italian dry fer- mented sausages (DELLAGLIO et al., 1996; rU- bIO et al., 2008). the average initial ash content was 3.13% and had increased significantly (P<0.05), reach- ing the ultimate value of 5.72%, whereas water activity (a w ) (table 2) had decreased significant- ly (P<0.05) during the smoking and dry-ripen- ing period (from 0.96 to 0.78). changes in mass fraction of individual basic constituents and wa- ter activity decrease seen after 90 days of Horse salami production (table 2) are mostly caused by the drying process, i.e. the loss of water oc- curring during ripening. changes in pH values seen during the pro- cessing of the Horse salami are presented in ta- ble 2. pH value had decreased during the first 21 days of processing (from 5.58 to 4.71), possibly as a result of the presence of organic acid pro- duced by bacteria (LUcKE, 1994). this pH drop is typical of most dry fermented sausage (PErEZ- ALVArEZ et al., 1999; GIMENO et al., 2000; LI- ZAsO et al., 1999; MUGUErZA et al., 2002; bOZ- KUrt and bAyrAM, 2006; VAN scHALKWyK et al., 2011). At the final processing stage, pH val- ues increased to 4.94, possibly due to the liber- ation of peptides, amino acid and ammonia re- table 2 - basic chemical composition, salt content, a w and pH of the Horse salami established during the manufacturing process. Processing time (days) 0 7 14 21 28 42 60 90 Moisture (%) 61.91a±0.06 55.67b±0.01 48.92c±0.09 43.22d±0.06 37.57e±0.01 35.16f±0.04 31.61g±0.06 28.51h±0.02 Fat (%) 13.84h±0.03 15.83f±0.03 17.71e±0.01 18.55d±0.02 18.59d±0.04 20.59c±0.02 25.45b±0.06 28.54a±0.12 Protein (%) 17.05h±0.04 22.48g±0.01 23.73f±0.01 24.36e±0.08 27.62d±0.04 27.95c±0.05 29.34b±0.23 30.53a±0.05 Collagen (%) 0.63e±0.11 1.19d ±0.11 1.56cd±0.07 2.05c±0.21 2.06c±0.09 2.82b±0.15 2.84b±0.12 3.93a±0.10 Ash (%) 3.13g±0.02 3.73f±0.06 4.56e±0.04 4.87d±0.01 4.94d±0.01 5.36c±0.05 5.45b±0.06 5.72a±0.01 Salt (NaCl) (%) 2.30g±0.04 2.71f±0.05 3.29e±0.03 3.63d±0.05 3.75c±0.05 3.81c±0.05 4.24b±0.02 4.51a±0.03 aw 0.96a±0.01 0.93ab±0.02 0.91b±0.01 0.88c±0.01 0.87c±0.03 0.86c±0.04 0.86c±0.01 0.78d±0.01 pH 5.58a±0.03 4.99b±0.05 4.74f±0.10 4.71g±0.16 4.72fg±0.08 4.76f±0.06 4.81d±0.10 4.93c±0.07 Values are means ±SD obtained with three measurements. Values displayed in the same row and tagged with different letters (a-h) are significantly different (P<0.05). Ital. J. Food Sci., vol. 28 - 2016 101 sulting from a proteolityc reaction (sPAZIANI et al, 2009). the final pH was lower than in most dry fermented sausages (5.2 to 5.8) (bOVEr-cID et al., 2001; rUbIO et al., 2007; rOsErIO et al., 2010), which can be explained by horse meat properties in terms of higher glycogen content as compared to pork, beef and foal meat (LAW- rIE and LEDWArD, 2006). the salt content of the Horse salami had sig- nificantly increased during processing (P<0.05) (table 2). Literature sources have reported the average mass fraction of salt in dry sausage stuffing to range from 2.0% to 2.6%, and that in final products to range from 3.3% to 4.3% (OcK- ErMAN and bAsU, 2007; stAHNKE and tJENEr, 2007). In this study, mass fraction of salt (Nacl) established during the Horse salami manufac- turing process ranged from 2.31% to 4.51%. Instrumental colour properties the cIELAb space (L*, a* and b*) values of the Horse salami were significantly affected (P<0.05) by the length of smoking and ripen- ing period (table 3). Lower lightness L* values seen with an increased length of processing are probably related to the dark colour of the Horse salami coming as a consequence of browning. A similar decrease in L* values during ripening was reported by bOZKUrt and bAyrAM (2006) for turkish sucuk, and by LOrENZO et al. (2012) for foal salchichon. redness (a*) had significantly (P<0.05) de- creased at all processing stages. similar lower a* values were seen during the ripening of span- table 3 - colour parameters of the Horse salami established during the manufacturing process. Processing time (days) 0 7 14 21 28 42 60 90 L* 46.67a±1.08 43.01b±1.05 41.81c±0.37 40.54d±0.30 38.75e±0.35 33.77f±1.27 33.28f±0.92 31.28g±0.89 a* 17.71ab±0.40 17.29ab±2.03 18.54a±3.44 16.16ab±2.65 15.49b±0.39 12.07c±0.62 10.86cd±01.06 8.15d±0.69 b* 20.32a±1.49 18.01bc±2.52 17.98bc±2.63 16.51c±2.22 13.39d±0.43 13.14d±0.80 12.14d±2.04 9.11e±0.58 Values are means ±SD obtained with ten measurements. Values displayed in the same row and tagged with different letters (a-g) are significantly different (P<0.05). ish pork dry sausages and foal salchichon, as reported by PErEZ-ALVArEZ et al. (1999) and LOrENZO et al. (2012). Lowering of a* values can possibly be explained by total or partial dena- turation of nitrosomyoglobin coming as a result of lactic acid production. L* and a* values low- er than those reported by LOrENZO et al. (2012) can probably be related to the nature of horse meat, which is darker and redder than foal (tA- tEO et al., 2008) yellowness (b*) had decreased from 20.32 to 9.11 and had varied significantly (P<0.05) during the production process. the decrease in b* val- ues seen with the prolongation of the processing time was also reported by other authors (PErEZ- ALVArEZ et al., 1999; LOrENZO et al., 2012) and explained by the decrease in concentration of ox- ymyoglobin coming as a result of oxygen con- sumption executed by microorganisms. Texture Profile Analysis texture Profile Analysis (tPA) parameters of the Horse salami established during the smok- ing and dry ripening period are presented in ta- ble 4. Average hardness values had significantly increased (P<0.05) from 0.32 to 20.54 kg as the processing went by. this can be related to the coagulation of muscle protein coming as a re- sult of low pH values and sausage drying (bOZ- KUrt and bAyrAM, 2006). springiness and cohesiveness had signifi- cantly decreased (P<0.05) during the processing (from 0.76 to 0.64 and from 0.67 to 0.43, respec- tively). springiness is related to elastic proper- table 4 - Parameters obtained by virtue of textural Profile Analysis (tPA) of the Horse salami during the manufacturing process. Processing time (days) 0 7 14 21 28 42 60 90 Hardness (kg) 0.32h±0.01 3.67g±0.19 4.61f±0.09 6.19e±0.21 9.91d±0.11 14.94c±0.25 17.58b±0.81 20.54a±0.92 Springiness 0.76a±0.03 0.62def±0.03 0.68bc±0.01 0.73ab±0.02 0.58f±0.01 0.61ef±0.02 0.66cd±0.01 0.64cde±0.03 Cohesiveness 0.67a±0.04 0.51c±0.04 0.48cd±0.01 0.65ab±0.12 0.46cd±0.01 0.46cd±0.01 0.43d±0.03 0.43d±0.03 Gumminess (kg) 0.26f±0.01 1.87e ±0.15 2.26e±0.07 4.02d±0.50 4.56d±0.06 6.87c±0.13 7.73b±0.50 8.83a±0.54 Chewiness (kg) 0.20g±0.02 1.16f±0.09 1.54e±0.04 2.94d±0.40 2.64d±0.04 4.19c±0.10 5.10b±0.49 5.65a±0.55 Resilience 0.19a±0.04 0.15bcd±0.02 0.16bc±0.03 0.18ab±0.05 0.15bcd±0.01 0.12e±0.01 0.13de±0.02 0.14cde±0.01 Values are means ±SD obtained with eight measurements. Values displayed in the same row and tagged with different letters (a-h) are significantly different (P<0.05). 102 Ital. J. Food Sci., vol. 28 - 2016 ties, so that the decrease in this textural prop- erty of the Horse salami is most likely to be re- lated to the removal of water (bOZKUrt and bA- yrAM, 2006). Increases in gumminess and chewiness val- ues (from 0.26 to 8.83 and from 0.20 to 5.65, respectively) seen during the Horse salami pro- cessing were statistically significant (P<0.05). Increase in chewiness values indicates that the Horse salami becomes tougher during the rip- ening period (sZcZEsNIAK, 2002), possibly due to moisture loss. resilience values established at the beginning and at the end of the processing were 0.19 and 0.14, respectively. significant changes in resil- ience during smoking and ripening failed to be observed (P>0.05) (table 4). Microbial counts Microbial flora changes seen during manu- facturing are shown in Fig. 1. the initial bacte- rial counts were 6.09 log cFU g-1 for total via- ble count (tVc), 5.29 log cFU g-1 for lactic bac- teria (LAb), and 3.88 log cFU g-1 for Staphylo- coccus spp, respectively. relatively low bacteri- al counts in the salami stuffing indicate a good hygienic quality of the raw materials. tVc, LAb and Staphylococcus spp counts had significant- ly increased during the ripening period (P<0.05). this increase in bacterial count is typical of most naturally dry fermented European sau- sages (KOZAČINsKI et al., 2008). At the end of the Horse salami production process, the mean values were 9.10, 7.79 and 5.10 log cFU g-1, re- spectively. As reported by many studies, microorganisms most represented during the ripening of cured sausages and meat products are LAb (LIZAsO et al., 1999; sAMELIs and GEOrGIADOU, 2000), whose counts tend to remain stable through- out the ripening period. Within the frame of this study, high LAb counts had been found during the first 28 ripening days, which can be related to a substantial pH drop witnessed during that period (table 2). LAb inhibit the growth of path- ogenic and spoilage bacteria by virtue of forma- tion of lactic acid, acetic acid and possibly bac- teriocins (LUcKE, 2000). Isolation and identification of microbial population Native sausage products are of a higher quality than those obtained by virtue of controlled fer- mentation with the addition of industrial start- ers (LEbErt et al., 2007). Many authors sup- port the view that indigenous microflora or mi- croorganisms present in traditional sausages originate from raw materials or the manufactur- ing environment (MAUrIELLO et al., 2004; rANt- sIOU et al., 2005). this microbiota is commonly referred to as “the house flora” (GArcIA-VArO- NA et al., 2000). therefore, in this study, the isolation and identification of autochthonous microbial pop- ulation inhabiting the Horse salami was per- formed. the results of a microbiological analy- sis (table 5) showed the dominant microflora to be the lactic acid bacteria strain termed Lacto- bacillus plantarum, Lactococcus lactis ssp. lactis, and Enterococcus faecium while the most repre- sented coagulase-negative staphylococci strain was S. xylosus. the yeast Candida famata/De- baryomyces hansenii was found as well, which is in agreement with the results of NIELsEN et al. (2008), who stated that halophilic yeasts most frequently isolated from fermented meat prod- Fig. 1 - changes in microbial counts seen during the processing of the Horse salami (mean±standard devia- tion obtained with three samples). Ital. J. Food Sci., vol. 28 - 2016 103 ucts are Debaromyces hansenii, Candida famata, Candida zeylanoides, Trichosporon sp., Crypto- coccus sp. and Rhodotorula sp. yeasts also play an important role in the maturation of sausag- es, since their lipolytic and proteolytic activity contributes to the development of sensory char- acteristics of fermented sausages (KOVAČEVIĆ, 2001; ALAGIĆ et al., 2008). In the Horse salami samples, bacteria of the Salmonella genus, Enterobacteriaceae, sulphite- reducing clostridia, L. monocytogenes or S. au- reus were not found; however, API biochemical tests uncovered the presence the Listeria grayi bacterium which is non-pathogenic (table 5). Is- sues sometimes emerging with this type of fer- mented meat product are short shelf-life and poor hygienic surroundings, but the sausages produced in this investigation were proven to be microbiologically safe. It should be pointed out that biochemical (API) tests gave very good results (identification of one species with ID > 98,2-99.9 %). the isolated lactic acid bacteria L. lactis ssp. lactis. L. plantarum and E. faecium could be used as starter cultures for meat products. L. plan- tarum as an autochthonous meat microflora is widely spread in nature (sALAMA et al., 1995; AyAD et al., 2001), L. lactis ssp. lactis in ferment- ed sausage has rarely been reported so far and therefore further studies must to include detail molecular identification of isolated strains be- cause API identification is not 100% precisely. the interest in exploring the potential of new strains isolated from different natural ecosys- tems to the effect of aroma compounds produc- tion has recently increased (AyAD et al., 2001; FrEcE et al., 2009; bAbIĆ et al., 2011, FrEcE et al., 2014). Metabolic properties of the L. plantar- um, E. faecium and L. lactis species have both di- rect and indirect influence on organoleptic, nu- tritional and hygienic quality of fermented prod- ucts. More and more research is focused on the isolation and identification of autochthonous functional starter cultures with the aim of de- veloping new functional meat products that will be recognised and labelled as autochthonous to the region in which they are produced (bAbIĆ et al., 2011, FrEcE et al., 2014, FrEcE et al., 2014 a,b). therefore, L. plantarum, E. faecium, L. lactis and S. xylosus as potential functional autoch- thonous starter cultures will be thoroughly in- vestigated in the future. Further studies will be carried out to detail phenotypic, genotypic and physiological characterization of isolated strains of staphylococci and LAb. Sensory characteristics complex interaction between physicochem- ical, biochemical and microbiological process- es, playing a role in formation of chemical com- pounds, and the modification of molecules re- sponsible for the texture and appearance of the final product also determine its sensory char- acteristics. Average scores given by the panellists at the end of the Horse salami manufacturing process are shown in Fig. 2. As for the external attrib- utes, the Horse salami scored highly when it comes to hardness (4.10±0.71) and low when it comes to appearance (3.60±0.43). It was high- ly rated for its sliceability, but low-rated when it comes to its colour intensity, fat/ lean ratio and easy peeling capacity. After slicing, the high- est scores were obtained for the fat distribution (4.78±0.67), while the fat/lean ratio scored low (2.22±0.44). table 5 - biochemical (API) results of the final product obtained after 90 days of manufacturing. Microorganism Values log CFU g-1±SD API test Salmonella sp. - - Enterobacteriaceae - - Staphylococcus aureus - - CNS (Coagulase negative staphylococci) 5.10±1.5 S. xylosus Sulphite reducing clostridia - - Listeria sp. - Listeria grayi Lactic acid bacteria 7.79±1.3 L. lactis ssp. lactis, Lactobacillus plantarum, Enterococcus faecium Yeasts 3.25±1.2 Candida famata/Debaryomyces hansenii Fig. 2 - Mean values of sensory properties of the final Horse salami. 104 Ital. J. Food Sci., vol. 28 - 2016 table 6 - Pearson´s correlation coefficients established between basic chemical composition, salt content, aw, texture and instrumental colour parameters. Hardness Springiness Cohesiveness Gumminess (kg) Chewiness (kg) Resilience L* a* b* (kg) Moisture (%) -0.95** 0.19 0.68** -0.96** -0.94** 0.48 0.95** 0.89** 0.95** Fat (%) 0.95** 0.058 -0.64 0.94** 0.95** -0.35 -0.91** -0.95** -0.93** Protein (%) 0.96** -0.28 -0.77** 0.95** 0.93** -0.51 -0.95** -0.90** -0.97** Collagen (%) 0.95** -0.09 -0.59 0.95** 0.94** -0.37 -0.94** -0.96** -0.96** Ash (%) 0.92** -0.10 -0.61 0.94** 0.93** -0.44 -0.95** -0.87** -0.91** Salt (NaCl) (%) 0.94** -0.03 -0.62 0.94** 0.95** -0.33 -0.92** -0.90** -0.95** a w -0.86** 0.12 0.51 -0.86** -0.84** 0.20 0.84** 0.89** 0.94** Values marked with ** are statistically significant (P<0.05). regarding the attributes that describe per- ceptions during mastication, Horse salami was highly rated for its flavour intensity (4.10±0.44), juiciness (4.24±0.21) and acid taste (3.79±0.17), and low-rated for its saltiness and smokiness (2.41±0.31 and 2.20±0.19). During the fermentation of dry sausages, LAb produce lactic acid (MAtEO et al., 1996) respon- sible for the sour taste (LOtONG et al., 2000) and odour of the product, while mould odour is to be associated with 1-octen-3-ol, which spreads a typical mushroom odour (MEyNIEr et al., 1998). In the present study, all three attributes scored highly (lactic acid taste 4.24±0.18; lactic acid odour 4.31±0.22; mould odour 3.82±0.15). As for the smell descriptors, lactic acid (4.3±0.22) and mould odour (3.8±0.15) were dominant, while the spice odour scored low (3.00±0.28). Correlation between the parameters Instrumental colour parameters of the Horse salami, established during its processing, were significantly inversely correlated (P<0.05) to the protein, fat, ash, collagen and salt content. Mois- ture content and a w values exhibited a signifi- cant direct correlation (P<0.05) to the instru- mental colour parameters (table 6). relation- ships between the moisture, protein, fat, ash, collagen and salt content and a w on one hand, and hardness, gumminess and chewiness on the other, were also significant (P<0.05) (that between moisture and a w being an inverse one). Pearson’s correlation coefficients indicated that springiness and resilience are not significantly (P>0.05) correlated to the basic chemical com- position, salt content and a w (table 6). cONcLUsIONs this study investigated into the changes in physicochemical, colour, textural, microbiologi- cal and sensorial properties of the Horse salami as an indigenous croatian dry fermented sau- sage. During 90 days of manufacturing, major changes in physicochemical, colour and textur- al properties took place during the fermentation and ripening stage, pointing to proteolysis and lipolysis phenomena coming as a result of en- dogenous enzymatic activity, as well as to high lactic acid bacteria and staphylococci counts contributing to the specific organoleptic attrib- utes of the final product. sensorial profiling of the final Horse salami showed a significant acid taste, lactic acid odour and flavour intensity, and low fat/lean ratio, smokiness and saltiness val- ues. the final product was proven to be micro- biologically safe, the dominant microbial popu- lation being L. lactis ssp. lactis, L. plantarum, E. faecium and S. xylosus. rEFErENcEs Alagić D., Kozačinski L., Filipović I., Zdolec N., Hadžiosma- nović M., Njari b., Kozačinski Z. and Uhitil s. 2008. Mi- crobiological changes during ripening of fermented sau- sages of horsemeat. Meso. 10: 200. Ayad E.H.E., Verheul A., Engels W.J.M., Wouters J.t.M. and smit G. 2001. Enhanced flavour formation by combina- tion of selected lactococci from industrial and artisanal origin with focus on completion of a metabolic pathway. J. Appl. Microb. 90: 59. babić I., Markov K., Kovačević D., trontel A., slavica A., Đugum J., Čvek D., svetec I. K., Posavec s. and Frece J. 2011. Identification and characterization of potential autochthonous starter cultures from a croatian “brand” product “slavonski kulen”. Meat sci. 88: 517. badiani A., Nanni N., Gatta P., tolomelli b. and Manfredi- ni M. 1997. Nutrient profile of horsemeat. J. Food com- pos. Anal. 10: 254. bourne M.c. 1978. texture Profile Analysis. Food technol.- chicago. 32: 62. bover-cid s., Izquierdo-Pulido M. and Vidal-carou M.c. 2001. Effectiveness of a Lactobacillus sakei starter cul- ture in the reduction of biogenic amine accumulation as a function of the raw material quality. J. Food Pro- tect. 64: 367. bozkurt H. and bayram M. 2006. colour and textural attrib- utes of sucuk during ripening. Meat sci. 73: 344. bruna J.M., Ordonez J.A., Fernandez M., Herranz b. and De la Hoz L. 2001. Microbial and physico-chemical chang- es during the ripening of dry fermented sausages super- ficially inoculated with or having added an intracellu- lar cell-free extract of Penicillium aurantiogriseum. Meat sci. 59: 87. Ital. J. Food Sci., vol. 28 - 2016 105 casiraghi E., Pompei c., Dellaglio s., Parolari G. and Virgili r. 1996. Quality attributes of Milano salami, an Italian dry-cured sausage. J. Agr. Food chem. 44: 1248. cIE. 1976. “colorimetry: official recommendations of the in- ternational commission on illumination”. comisión Inter- nationale de l’Èclairage [International commission on Il- lumination], cIE No. 15 (E-1.3.1). Paris, Fr. Dellaglio s., casiraghi E. and Pompei c. 1996. chemical, physical and sensory attributes for the characterization of an Italian dry-cured sausage. Meat sci. 42: 25. Franco D., rodríguez E., Purriños L., bermúdez r. and Lor- enzo J.M. 2011. Meat quality of “Galician Mountain” foals breed. Effect of sex, slaughter age and livestock produc- tion system. Meat sci. 88:292. Frece J., Kovačević D. and Markov K. 2014.a National Patent P20130089A: “Formulation of bacterial starter cultures for the production of dry sausage and its use„ (15.8. 2014.) Hrvatski glasnik intelektualnog vlasništva 17/2014. Frece J., Kovačević D. and Markov K. 2014.b National Pat- ent P20130569A: “Use of probiotic bacterial cultures Lac- tobacillus plantarum 1K for the production of functional foods„ (9.10.2014.) Hrvatski glasnik intelektualnog vla- sništva 1/2015. (2.1. 2015.) Frece J., Kos b., svetec I. K., Zgaga Z., beganović J., Leboš A. and Šušković J. 2009. synbiotic effect of Lac- tobacillus helveticus M92 and prebiotics on the intes- tinal microflora and immune system of mice. J. Dairy res. 76: 98. Frece J., Kovačević D., Kazazić s., Mrvčić J., Vahčić N., Delaš F., Ježek D., Hruškar M., babić I. and Markov K. 2014. comparison of sensory properties, shelf live and micro- biological safety of industrial sausages produced with autochthonous and commercial starter cultures (start- er cultures for sausages production). Food technol. bi- otech. 52: 307. Garcia-Varona M., santos E. M., Jaime I. and rovira J. 2000. characterisation of Micrococcaceae isolated from differ- ent varieties of chorizo. Int. J. Food Microbiol. 54:189. Gimeno O., Ansorena D., Astiasaran I. and bello J. 2000. characterization of chorizo de Pamplona: Instrumental measurements of colour and texture. Food chem. 69:195. IsO 3496:1994. Meat and meat products. Determination of hydroxyproline content. IsO 1442:1997. Meat and meat products. Determination of moisture content. IsO 1443:1973. Meat and meat products. Determination of total fat content. IsO 3100-1:1975. Meat and meat products. sampling and preparation of test samples. Part 1: sampling. IsO 936:1998. Meat and meat products. Determination of total ash. IsO 937:1978. Meat and meat products. Determination of nitrogen content. IsO 6658:2005. sensory Analysis - Methodology - Gener- al guidance. Kovačević D. 2001. “chemistry and technology of meat and fish”. Faculty of food technology, Josip Juraj strossmayer University of Osijek, Grafika. Osijek. croatia. Kozačinski L., Drosinos E., Čaklovica F., cocolin L., Gas- parik-reichardt J. and Vesković s. 2008. Investigation of microbial association of traditionally fermented sau- sages. Food technol. biotechnol. 46: 93. Lanza M., Landi c., scerra M., Galofaro V. and Pennisi P. 2009. Meat quality and intramuscular fatty acid com- position of sanfratellano and Haflinger foals. Meat sci. 81: 142. Lawrie r.A. and Ledward D.A. 2006. “Lawrie’s meat science” (7th Ed.). crc Press. boca raton, FL. Lebert I., Leroy s., Giammarinaro P., Lebert A., chacornac J. P., bover-cid s., Vidal-carouc M.c. and talona r. 2007. Diversity of microorganisms in the environment and dry fermented sausages of small traditional French proce- ssing units. Meat sci. 76: 112. Litwinczuk A., Florek M., skalecki P. and Litwinczuk Z. 2008. chemical composition and physicochemical properties of horse meat from the longissimus lumborum and semiten- dinosus muscle. J. Musc. Food. 19: 223. Lizaso G., chasco J. and beriain M.J. 1999. Microbiological and biochemical changes during ripening of salchichon, a spanish dry cured sausage. Food Microbiol. 16: 219. Lombardi-boccia G., Lanzi s. and Aguzzi A. 2005. Aspects of meat quality: trace elements and b vitamins in raw and cooked meats. J. Food compos. Anal.18: 39. Lorenzo J.M., temperan s., bermudez r., cobas N. and Purrnios L. 2012. changes in physico-chemical, microbi- ological, textural and sensory attributes during ripening of dry-cured foal salchichon. Meat sci. 90: 194. Lotong V., chambers E. and chambers D.H. 2000. Deter- mination of the sensory attributes of wheat sourdough bread. J. sens. stud. 15: 309. Lucke, F.K. 1994. Fermented Meat-Products. Food res. Int. 27: 299. Lucke, F.K. 2000. Utilization of microbes to process and pre- serve meat. Meat sci. 56: 105. Mateo J., Dominguez M.c., Aguirrezabal M.M. and Zumalac- arregui J.M. 1996. taste compounds in chorizo and their changes during ripening. Meat sci. 44: 245. Mauriello G., casaburi A., blaiotta G. and Villani F. 2004. Isolation and technological properties of coagulase nega- tive staphylococci from fermented sausages of southern Italy. Meat sci. 67: 149. Mela D.J. 1990. the basis of dietary preference. trends Food sci. tech. 1: 55. Meynier A., Genot c. and Gandemer G. 1998. Volatile com- pounds of oxidized pork phospholipids. J. Am. Oil chem. soc. 75: 1. Muguerza E., Fista G., Ansorena D., Astiasaran I. and blouk- as J.G. 2002. Effect of fat level and partial replacement of pork backfat with olive oil on processing and quality characteristics of fermented sausages. Meat sci. 61: 397. Nielsen D.s., Jacobsen t., Jespersen L., Koch A.G. and Ar- neborg N. 2008. Occurence and growth of yeasts in pro- cessed meat products-Implications for potential spoila- ge. Meat sci. 80: 919. Ockerman H.W. and basu L. 2007. reduction and consump- tion of fermented meat products. ch 2. “Handbook of fer- mented meat and poultry”. F. toldrá (Ed.), p. 9. black- well Publishing, Ames, IA. Olivares A., Navarro J.L., salvador A. and Flores M. 2010. sensory acceptability of slow fermented sausages based on fat content and ripening time. Meat sci. 86:251. Ordonez J.A., Hierro E.M., bruna J. and Hoz L. 1999. chang- es in the components of dry-fermented sausages during ripening. crit. rev. Food sci. 39: 329. Perez-Alvarez J.A., sayas-barbera M.E., Fernandez-Lopez J. and Aranda-catala V. 1999. Physicochemical charac- teristics of spanish-type dry-cured sausage. Food res. Int. 32: 599. Pleadin J., Krešić G., barbir t., Petrović M., Milinović I. and Kovačević D. 2014. changes in basic nutrition and fat- ty acid composition during production of “slavonski ku- len”. Meso. 16: 522. rantsiou K., Urso r., Iacumin L., cantoni c., cattaneo P. and comi G. 2005. culture-dependent and -independ- ent methods to investigate the microbial ecology of Ital- ian fermented sausages. Appl. Environ. Microb. 71: 1977. roseiro L.c., Gomes A., Gonçalves H., sol M., cercas r. and santos c. 2010. Effect of processing on proteolysis and bi- ogenic amines formation in a Portuguese traditional dry- fermented ripened sausage “chouriço Grosso de Etrem- oz e borba PGI”. Meat sci. 84: 172. rubio b., Martinez b., sanchez M.J., Garcia-cachan M.D., rovira J. and Jaime I. 2007. study of the shelf life of a dry fermented sausage “salchichon” made from raw ma- terial enriched in monounslaturated and polyunsaturat- ed fatty acids and stored under modified atmospheres. Meat sci. 76: 128. rubio b., Martinez b., Garcia-cachan M.D., rovira, J. and Jaime I. 2008. Effect of the packaging method and the storage time on lipid oxidation and colour stability on dry fermented sausage salchichon manufactured with raw 106 Ital. J. Food Sci., vol. 28 - 2016 material with a high level of mono and polyunsaturated fatty acids. Meat sci. 80: 1182. salama M.s., Musafija-Jekni, t., sandine W.E. and Giovan- noni s.J. 1995. An ecological study of lactic acid bacte- ria: isolation of new strains of Lactococcus including Lac- tococcus lactis subsp. cremoris. J. Dairy sci. 78: 1004. salgado A., Fontan M.c.G., Franco I., Lopez M. and carbal- lo J. 2005. biochemical changes during the ripening of chorizo de cebolla, a spanish traditional sausage. Effect of the system of manufacture (homemade or industrial). Food chem. 92: 413. salgado A., Fontan M.c.G., Franco I., Lopez M. and car- ballo J. 2006. Effect of the type of manufacture (home- made or industrial) on the biochemical characteristics of chorizo de cebolla (a spanish traditional sausage). Food cont. 17: 213. samelis J. and Georgiadou K.G. 2000. the microbial asso- ciation of Greek taverna sausage stored at 4 and 10 de- grees c in air, vacuum or 100% carbon dioxide, and its spoilage potential. J. Appl. Microbiol. 88: 58. spaziani M., del torre M. and stecchini M.L. 2009. changes of physicochemical, microbiological, and textural prop- Paper Received November 4, 2015 Accepted April 22, 2015 erties during ripening of Italian low-acid sausages. Pro- teolysis, sensory and volatile profiles. Meat sci. 81: 77. stahnke L. H. and tjener K. 2007. Influence of processing parameters on cultures performance. ch 18. In “Hand- book of fermented meat and poultry”. F. toldrá (Ed.), p. 187. blackwell Publishing, Ames, IA. szczesniak A.s. 2002. texture is a sensory property. Food Qual. Prefer. 13: 215. tateo A., De Palo P., ceci E. and centoducati P. 2008. Phys- icochemical properties of meat of Italian Heavy Draft horses slaughtered at the age of eleven months. J. Anim. sci. 86: 1205. trajković J., Mirić M., baras J. and Šiler s. 1983. Analize životnih namirnica. tehnološko metalurški fakultet. be- ograd, yU. Van schalkwyk D.L., McMillin K.W., booyse M., Witthuhn r.c. and Hoffman L.c. 2011. Physico-chemical, microbi- ological, textural and sensory attributes of natured game salami produced from springbok (Antidorcas marsupia- lis), gemsbok (Oryx gazella), kudu (Tragelaphus strep- siceros) and zebra (Equs burchelli) harvested in Namib- ia. Meat sci. 88: 36.