Effect of Low-Salt Processing on Lipolytic Activity, Volatile Compound Profile, Color, Lipid Oxidation, and Microbiological Properties of Four Different Types of Pastırma

Abstract

Pastırma is a traditional dry-cured meat product made from whole pieces of cattle or water buffalo carcasses. Sixteen or more types of pastırma can be produced from different parts of the carcass. This study investigated the effect of low salt processing (3% NaCl) on the lipolytic enzyme activity, volatile profile, color, lipid oxidation, and microbiological properties of commonly produced types of pastırma (kuşgömü, sırt, bohça, and şekerpare). In the study, 5% NaCl level was used as the control group. For all pastırma types, the pH changed between 5.5 and 6.0. The a_w value was less than 0.90 for the pastırma types. The L* value increased when the salt level decreased from 5% to 3% (p < 0.05); however, the salt level did not affect the a* and b* values (p > 0.05). Reducing the salt level increased the neutral lipase activity and decreased the TBARS. As the salt level increased, the acid lipase activity increased in the bohça pastırma, and the phospholipase activity increased in the kuşgömü and sırt pastırma (p < 0.05). Furthermore, while Micrococcus/Staphylococcus constituted the dominant microbiota in pastırma types, a 5% salt level led to a decrease in the number of lactic acid bacteria. The volatile compounds were more affected by salt level than by pastırma type. The correlation analysis showed that there are some differences between 3% and 5% salt levels and the use of a 3% salt level increases the abundance of the compounds. The correlation analysis also revealed that there are differences between the pastırma types in terms of the volatile compounds and that kuşgömü pastırma differs from other pastırma types.

1. Introduction

One of the most popular food additives is sodium chloride (NaCl). It enhances flavor and ensures microbiological safety in meat products [1]. However, excessive salt consumption can increase blood pressure and lead to cardiovascular disease [2,3]. Moreover, sodium is accepted as the etiological factor for some diseases such as obesity, various types of cancer, kidney stones, and osteoporosis [4]. Therefore, the World Health Organization (WHO) suggests restricting adult sodium intake to <2 g/day (5 g/day salt) to lower blood pressure and reduce the risk of coronary heart disease, cardiovascular disease, and stroke [5].

Salt plays an important role in biochemical and chemical reactions such lipolysis, lipid oxidation, and proteolysis in processed meat products [6,7] and contributes to the characteristic flavor of dry-cured meat products [8,9]. Moreover, salt, which is effective in ensuring the solubility of myofibrillar proteins, plays a key role in texture development [10,11]. Meat products produced from whole pieces such as dry-cured ham are usually exposed to 3–5% NaCl in the dry-curing step [12], and the total processing time of some products may exceed one year [13]. In pastırma, a traditional dry-cured meat product, salt is used in quantities of up to 10% in traditional production, and the production process of pastırma can take up to one month, depending on the size of the muscle or muscles used and the processing conditions [14]. Pastırma, like other dry-cured meat products made from whole pieces, is a significant source of sodium [15]. In a study that aimed to limit the sodium content in the pastırma, it was reported that using 50% KCl instead of sodium chloride during production did not cause any changes in the microbiological, physical, or chemical characteristics of the product [15]. Another study on reducing sodium content determined that KCl can replace 15% NaCl without changing the sensory properties of pastırma [16]. On the other hand, it has been stated that pastırma can be produced using 5% salt under controlled conditions without altering its quality [17].

The raw material used for pastırma production shows significant differences in terms of both shape and muscle type and structure. Sixteen or more pastırma types can be produced from a cattle or water buffalo carcass [9]. Pastırma types are named according to the part from which they are obtained. Sırt pastırma is obtained from the loin and posterior rib, şekerpare and bohça are obtained from the round, and kuşgömü is obtained from the tenderloin [18]. TS 1071 categorizes pastırma into three classes: first, second, and third. In this classification, kuşgömü and sırt are included in the first class, and şekerpare and bohça are included in the second class [19]. The salt content in pastırma varies depending on raw material and processing conditions [9]. In fact, some studies on pastırma have reported significant differences between pastırma types in terms of salt content [20,21,22]. In a study investigating different salt levels (3, 6, and 9%), some of the quality parameters of the sırt pastırma were examined [7]. However, there is no information available regarding how reducing the amount of sodium chloride (NaCl) in other types of pastırma affects the lipolytic enzyme activity, volatile profile and other quality parameters. This study was carried out to establish the effect of low-salt processing (3% NaCl) on the lipolytic enzyme activity, volatile profile, color, lipid oxidation, and microbiological properties of four different pastırma types (kuşgömü, sırt, bohça, and şekerpare). In this study, a 5% NaCl level was also used as the control group.

2. Materials and Methods

2.1. Material

In this study, two beef carcasses (postmortem 24 h) were used as the raw material. A total of 16 pieces of pastırma (kuşgömü, sırt, bohça and şekerpare) were obtained from carcasses (Figure 1). The excess fat and connective tissue on the surface were removed.

2.2. The Production of Pastırma

A climate chamber (Reich, Germany) was used under controlled conditions during the pastırma production and the production stages of pastırma are shown in

Table 1. The production stages of pastırma.

Production Stages	Temperature (°C)	Relative Humidity (%)	Time
Curing	4 ± 1	-	2 days
First drying	15 ± 1	80 ± 2	6 days
First pressing (15 kg weight per 1 kg meat)	7 ± 1	-	20 h
Second drying	20 ± 1	70 ± 2	5 days
Second pressing (15 kg weight per 1 kg meat)	25 ± 1	-	7 h
Third drying	20 ± 1	70 ± 2	5 days
Çemen coating	7 ± 1	-	24 h
Drying with çemen	20 ± 1	70 ± 2	8 days

. Two different salt levels (3% and 5%) were used in various pastırma types (kuşgömü, sırt, bohça and şekerpare). Before the curing stage, the meats were cut from the muscle surface at a 45° angle to avoid exceeding 2/3 of the thickness, and during the curing phase, 3% or 5% NaCl, 150 mg/kg sodium nitrite, and 0.3% (w/w) sucrose were added to the meat pieces. After the third drying period, the surface of pastırma was coated with çemen made from 1200 mL of water, 500 g of Trigonella foenum graecum flour, 350 g of garlic, and 150 g of red pepper. Finally, drying with çemen was performed.

2.3. Determination of pH, a_w and Thiobarbituric Acid-Reactive Substance Content (TBARS) Values

Buffer solutions were used to calibrate the pH meter (Mettler Toledo Ion S220, Greifensee, Switzerland). After that, 10 g of sample, to which 100 mL of water was added, was homogenized in an Ultra-Turrax (IKA T25, Staufen, Germany) for 60 s and the pH was measured.

A water activity device (Novasina AG CH-8853, Lachen, Switzerland) was used to determine the a_w. Six different salt solutions were used for calibration at a temperature of 25 °C. Afterwards, the samples in plastic containers were placed in the measurement chamber, and the measurements were taken at 25 °C.

Next, 2 g of sample and 12 mL of TCA solution containing trichloroacetic acid, ethylene diamine tetra acetic acid, and propyl gallate were homogenized in Ultra-Turrax. After filtration, 3 mL of thiobarbituric acid (0.02 M) was added to the mixture and maintained in a boiling water bath for 40 min. The cooled samples were centrifuged for 5 min, and 530 nm was used for the absorbance measurement. The results are presented in µmol malondialdehyde (MDA)/kg [23].

2.4. Determination of Color Values

Color values (L*, a* and b*) were determined according to the criteria given by the International Commission on Illumination based on three-dimensional color measurements (Commision Internationele de I’E Clairage). The color of the pastırma slices of 2 mm thickness was determined using a Chroma Meter (CR-400 Konica Minolta, Osaka, Japan).

2.5. Lipolytic Enzyme Activities

Next, 5 g samples and 25 mL of 50 mM disodium phosphate buffer were homogenized in an ice bath with Ultra-Turrax. The homogenized sample was centrifuged for 20 min at 10,000× g at 4 °C, the supernatant was filtered using glass wool, and the filtrate was used for enzyme assays. Motilva et al. [24] were used to measure acid and phospholipase activities, while Yalınkılıç et al. [15] modified their procedures for neutral lipase activity. The reaction buffer contained 0.8 mg/mL bovine serum albumin (BSA), 0.5 mg/mL Triton X-100, 0.2 mol/L disodium phosphate, 0.1 mol/L citric acid, for acid lipase, 0.2 mol/L disodium phosphate, 0.1 mol/L citric acid, 150 mmol/L sodium fluoride, 0.8 mg/mL BSA, 0.5 mg/mL Triton X-100 for phospholipase, and 0.22 mol/L Tris and 5 mg/mL BSA for neutral lipase assays. Phospholipase and acid lipase activities were determined at pH 5.0, and neutral lipase activity was determined at pH 7.5 using 4-methylumbelliferyl oleate as a substrate. The reaction mixture was incubated for 20 min at 37 °C, and excitation and emission measurements were performed using a fluorescence spectrophotometer (Cary Eclipse, Agilent, Santa Clara, CA, USA) at wavelengths of 335 and 460 nm, respectively. A unit of enzyme activity (U) was defined as the quantity of enzyme that hydrolyzes 1 μmol of substrate in one hour at 37 °C. Enzyme activities are given as units in 1 g of dry matter.

2.6. Determination of Volatile Compounds

Next, 5 g of sample was placed in 40 mL vials (Supelco, Bellefonte, PA, USA). After the volatile compounds were collected in the headspace of the vials at 30 °C for 1 h, the fiber (CAR/PDMS, 75 μm, Supelco, Bellefonte, PA, USA) was inserted into the headspace for 2 h, and the adsorption of volatile compounds was achieved. The volatile compounds were detected by mass spectrometry (MS, Agilent Technologies 5973) after fiber was injected into a gas chromatograph (GC, Agilent Technologies 6890 N). DB-624 (J&W Scientific, Santa Clara, CA, USA, 60 m, 0.25 mm i.d., 1.4 μm film) was used as a column. The oven temperature was 40 °C for 5 min and subsequently increased to 110 °C at a rate of 3 °C/min than 150 °C at 4 °C/min and 210 °C at 10 °C/min. Helium was used as the carrier gas (1 mL/min). It was maintained at this temperature for 12 min, so the total process took 56.33 min. Volatile compounds were identified using the library of mass spectrometry (NIST, WILEY, FLAVOR) and standard substances. The results are given as Au × 10⁶ [14].

2.7. Microbiological Analysis

A pastırma sample weighing 25 g was homogenized in 225 mL of sterile physiological saline (0.85%) using a stomacher for 2 min. Decimal dilutions prepared from this homogeneous mixture were used for the analysis. De Man, Rogosa, and Sharpe Agar (MRS, Merck, Darmstadt, Germany) was used to count the lactic acid bacteria and incubated for 2 days at 30 °C in anaerobic conditions. Micrococcus/Staphylococcus counts were determined on Mannitol Salt Phenol Red Agar (MSA, Merck), for 2 days at 30 °C. The Enterobacteriaceae were enumerated on Violet Red Bile Dextrose Agar (VRBD, Merck) and incubated for 2 days at 30 °C in anaerobic conditions [25]. For the yeast–mold, Rose-Bengal Chloramphenicol Agar (RBC, Merck) was used, and plates were incubated for 3–5 days at 25 °C [26].

2.8. Statistical Analysis

The effects of pastırma type and salt level on a_w, pH, TBARS, instrumental color, lipolytic enzyme activity, volatile profile, and microbiological properties were examined using two-way ANOVA with the mixed model, where dependent variables (pastırma type and salt level) were deemed to be fixed effects and replicates (two block) were deemed to be random effects. Duncan’s test was performed to compare the mean values and their interactions at the significance level of p < 0.05. All statistical analyses were carried out using IBM SPSS 20.0 software. In addition, to determine the relationship between salt levels and volatile compounds, and between pastırma type and volatile compounds, the correlation heat map was carried out using ChiPlot (https://www.chiplot.online/, accessed on 30 May 2025) [27].

3. Results and Discussion

The effects of salt reduction on the pH, a_w, and color of the pastırma types are listed in

Table 2. Overall effects of pastırma type and salt level on pH, a_w and color values (mean value ± standard deviation).

	pH	aw	L*	a*	b*
Pastırma Type (PT)
Kuşgömü	5.90 ± 0.07 a	0.890 ± 0.013 a	35.69 ± 0.64 b	33.84 ± 1.29 b	17.23 ± 0.42 b
Sırt	5.89 ± 0.02 a	0.874 ± 0.013 b	36.64 ± 2.20 b	36.09 ± 0.73 a	20.20 ± 1.20 a
Bohça	5.88 ± 0.07 a	0.887 ± 0.017 a	32.73 ± 1.01 c	32.00 ± 0.73 c	16.41 ± 0.44 b
Şekerpare	5.93 ± 0.09 a	0.887 ± 0.014 a	41.72 ± 0.64 a	30.12 ± 2.02 d	19.46 ± 2.02 a
Significance	NS	**	**	**	*
Salt Level (SL)
3%	5.90 ± 0.04 a	0.890 ± 0.008 a	37.92 ± 3.04 a	32.67 ± 2.53 a	18.58 ± 2.18 a
5%	5.90 ± 0.07 a	0.872 ± 0.009 b	35.99 ± 3.40 b	33.09 ± 2.95 a	18.07 ± 1.77 a
Significance	NS	**	**	NS	NS
PT × SL	**	NS	NS	NS	NS

a–d: Any two means in the same column that have the same letters in the same section are not significantly different at p > 0.05, ** p < 0.01, * p < 0.05. NS: not significant.

. Pastırma type and salt level did not have significant effects on the average pH. In addition, the interaction of pastırma type and salt level had a significant effect on the pH value of pastırma (p < 0.05). The use of 5% salt in production decreased the pH of the bohça, while it increased the pH in the kuşgömü (p < 0.05). (Figure 2). In addition, the pH of pastırma usually ranges between 5.5 and 6.0. In this study, the pH value was determined to be within this range in all groups. During pastırma production, the pH decreases initially but increases later due to proteolysis [9]. Proteolysis significantly impacts the quality of raw-cured meat products made from whole pieces, including dry-cured ham and pastırma [14,28,29]. The amount of NaCl also has an important effect on the proteolysis of dry-cured meat products [30,31]. In addition, the intensity of proteolysis can be influenced by the anatomic location of the muscle [32,33]. In the present study, it was estimated that differences in pH between pastırma types were dependent on the degree of proteolysis. In addition, the fiber type profile of the meat pieces used can lead to differences in the pH value [34]. Type I fibers have lower glycogen content than type II fibers, and the pH drop is limited in muscles rich in type I fibers [35]. Furthermore, the postmortem pH drop is slower in muscles rich in oxidative fibers, while the pH drop is more rapid in muscles rich in glycolytic fibers [36].

There was no significant difference in the average a_w value among the pastırma types and the average a_w value was found to be below 0.90 in all samples. The use of 5% salt decreased the a_w value further, but the average a_w value was below 0.90 for both salt levels and even the salt level x pastırma type interaction was not found to be significant (Table 2). During curing, significant mass transfer occurs with diffusion as a result of the intake of salt into the cell and the removal of water by changing the salt concentration with the osmotic pressure [37]. With the progress of the process, as a result of the removal of water, the product becomes stable for microbiologically as the a_w decreases [10].

Color is one of the parameters that influences consumer acceptance. In this study, the pastırma type had a very significant effect (p < 0.01) on the L* and a* values, and a significant effect (p < 0.05) on the b* value. The highest and lowest L* values were detected for the şekerpare and the bohça types, respectively. The a* value, which represents the intensity of the red color, was highest in the sırt pastırma (Table 2). The color differences through pastırma types are due to differences in fiber type and, thus, in the amount of myoglobin. In addition, it is estimated that the anatomical location has an effect on this result. Akköse et al. [18] and Çakıcı et al. [20] also found significant differences in pastırma types in terms of L*, a* and b* values. In this study, reducing the quantity of salt from 5% to 3% caused an increase in the L* value (p < 0.01). This result is due to the lower a_w value observed in the presence of 5% salt. Similarly, Bermudez et al. [38] reported that the L* value decreases because of the increase in salt concentration, which is related to the moisture content.

Dry-cured meat processing can affect the structure and stability of lipids through lipolysis and lipid oxidation [6,39,40]. Lipolysis is usually catalyzed by lipases, which originate from endogenous sources such as neutral lipases, acid lipases, and phospholipases [41]. Lipolysis produces free fatty acids, change in proportion to acid, phospho-, and neutral lipase activities [42]. The lipolytic activity on triglycerides was attributed to neutral lipase (optimum pH 7.0–7.5) and acid lipase (optimum pH 5.5), because the pH value ranged from 5.5 to 6.2 during the processing of most dry-cured hams. Thus, both neutral and acid lipases positively affected the hydrolysis of glycerol esters and cholesterol [43]. In this study, the mean pH value of pastırma ranged from 5.88 to 5.93. The highest activity was detected for neutral lipase in pastırma types, followed by acid lipase and phospholipase. These results show that neutral lipase is more effective for lipolysis in the pastırma. Similarly, Wang et al. [44] reported that neutral lipase activity was higher than acid lipase and phospholipase activity in the semimembranosus and biceps femoris muscles in Xuanwei ham. Liu et al. [41] also reported that the highest enzymatic activity was determined in neutral lipase during ripening of dry-cured ham in both low-sodium group and control group. In another study, the triceps brachii muscle showed the highest enzyme activity among the biceps femoris, longissimus dorsi, and triceps brachii muscles, but no significant difference was found between the other two muscles. Moreover, it has been reported that the energy required is provided from fatty acids in oxidative muscles such as the triceps brachii and from carbohydrates in glycolytic muscles such as the longissimus dorsi; therefore, oxidative muscles have higher lipolytic activity [45]. In this study, sırt pastırma showed higher acid lipase activity than şekerpare, kuşgömü, and bohça (

Table 3. Overall effects of pastırma type and salt level on enzyme activity and TBARS (mean value ± standard deviation).

Factors	Acid Lipase (U)	Phospholipase (U)	Neutral Lipase (U)	TBARS (µmol MDA/kg)
Pastırma Type (PT)
Kuşgömü	0.86 ± 0.09 b	0.77 ± 0.07 a	1.29 ± 0.06 a	30.16 ± 4.11 a
Sırt	1.11 ± 0.13 a	0.55 ± 0.11 b	1.14 ± 0.10 b	30.22 ± 4.89 a
Bohça	0.70 ± 0.14 c	0.36 ± 0.07 c	0.91 ± 0.10 c	23.47 ± 2.49 b
Şekerpare	0.66 ± 0.04 c	0.29 ± 0.05 d	0.98 ± 0.10 c	22.40 ± 1.88 b
Significance	**	**	**	**
Salt Level (SL)
3%	0.86 ± 0.17 a	0.54 ± 0.21 a	1.12 ± 0.14 a	25.45 ± 4.68 b
5%	0.81 ± 0.25 b	0.45 ± 0.19 b	1.04 ± 0.20 b	27.68 ± 5.24 a
Significance	*	**	*	*
PT × SL	**	**	NS	NS

a–d: Any two means in the same column that have the same letters in the same section are not significantly different at p > 0.05, ** p < 0.01, * p < 0.05, NS: not significant.

). In terms of phospholipase activity, the highest and lowest values were detected in kuşgömü and şekerpare, respectively. Neutral lipase activity was found to be similar to acid lipase activity in bohça and şekerpare. The highest activity in terms of this enzyme was found in kuşgömü, while the lowest activity was observed in bohça. These results are thought to be due to the metabolic properties (glycolytic or oxidative) of the muscles. Hernandez et al. [45] also reported that the oxidative triceps brachii muscle exhibited higher acid and neutral lipase activity than the glycolytic muscles. This finding biochemically supports the difference in neutral lipase activity between the kuşgömü (oxidative) and the bohça (glycolytic). Flores et al. [46] found that acid lipase activity was higher in the trapezius muscles and lower in the masseter muscle between the longissimus dorsi, semimembranosus, biceps femoris, masseter, and trapezius muscles. In a study conducted on in dry-cured ham, acid lipase activity was also determined to be significantly higher in M. semimembranosus compared to M. biceps femoris during storage [43].

In this study, a decrease in the investigated enzyme activities was detected as the salt level increased (Table 3). A similar result was noted by Cao et al. [42]. Increased salt concentration and low water activity can reduce lipolytic enzyme activity [47]. Another study on dry-cured ham found that low-salt conditions promote neutral and acid lipase activities. The same study reported that low salt content speeds up the decomposition of phospholipases. Additionally, it was emphasized that phospholipase is the main endogenous enzyme responsible for forming free fatty acids [41].

The effects of pastırma type and salt level interaction on the acid lipase and phospholipase activities of the samples are shown in Figure 3a and Figure 3b, respectively. There was a significant decrease in acid lipase activity as the salt content of the bohça increased. In addition, phospholipase activity decreased in all pastırma types with increasing salt level, and this decrease was found to be significant (p < 0.05) in the sırt and kuşgömü pastırma. Depending on the salt level and muscle structure, the diffusion of salt into the tissue changes, and thus, it is estimated that the abovementioned enzyme activities decrease due to the decrease in water activity. Toldra et al. [48] reported the inhibitive effect of salt on neutral lipase and acid esterase activities.

Pastırma type significantly affected TBARS, a biomarker of lipid oxidation (p < 0.01). In addition, the kuşgömü and sırt types had higher TBARS values than the bohça and şekerpare (Table 3). Hazar et al. [49] also reported that the TBARS values of kuşgömü and sırt were higher than those of other types. It is thought that this is caused by the change in the fat ratio, especially in different muscles. Otherwise, the salt level significantly affected the TBARS, and the TBARS value increased as the salt level increased from 3% to 5% (Table 3). Other studies have also shown that increasing the salt level in dry-cured meat products increases the TBARS [7,38,50]. This result is thought to be due to the salt’s prooxidant activity. Salt’s prooxidant activity is achieved by enabling the access of oxidizing agents to the lipid substrate by disrupting cell membrane integrity, releasing iron ions from molecules, or inhibiting the function of antioxidant enzymes [51].

Catalase-positive cocci (mainly coagulase-negative staphylococci) are the predominant microorganisms in the microbiota of pastırma [14]. As shown in

Table 4. Overall effects of pastırma type and salt level on microbiological properties (mean value ± standard deviation) (log cfu/g).

Factors	Micrococcus/ Staphylococcus	Lactic Acid Bacteria	Yeast–Mold	Enterobacteriaceae
Pastırma Type (PT)
Kuşgömü	7.47 ± 0.36 a	4.53 ± 1.49 b	4.93 ± 0.85 c	<2
Sırt	6.55 ± 0.29 b	3.82 ± 0.69 c	4.91 ± 0.73 c	<2
Bohça	7.85 ± 0.25 a	5.45 ± 0.89 a	6.64 ± 0.31 a	<2
Şekerpare	7.63 ± 0.44 a	5.25 ± 1.00 a	5.82 ± 0.55 b	<2
Significance	**	**	**
Salt Level (SL)
3%	7.40 ± 0.68 a	5.40 ± 1.13 a	5.59 ± 0.95 a	<2
5%	7.38 ± 0.60 a	4.76 ± 1.20 b	5.56 ± 0.99 a	<2
Significance	NS	**	NS
PT × SL	NS	**	**

a–c: Any two means in the same column that have the same letters in the same section are not significantly different at p > 0.05, ** p < 0.01. NS: not significant.

, Micrococcus/Staphylococcus were the predominant microbiota in all pastırma types, and the lowest average was detected in the sırt pastırma. These microorganisms are not affected by the increase in salt level because of their salt resistance (Table 4). Significant differences were observed among the pastırma types in terms of lactic acid bacteria (LAB). However, while the LAB were not affected by the salt level in sırt or bohça, the increase in the salt level in the other pastırma types caused a decrease in the LAB (Figure 4a). Salt can cause osmotic pressure inside and outside microorganism cells and disrupt the normal physiological activities of microorganisms [52]. The highest average yeast–mold count of pastırma types was detected in the bohça samples (Table 4). The change in salt level only affected the yeast–mold count of the kuşgömü and decreased as the salt level increased (Figure 4b). The Enterobacteriaceae count was below the detectable limit (<2 log cfu/g) in all groups. Other studies have shown that the number of Enterobacteriaceae decreases as production progresses [14,53].

As shown in

Table 5. Overall effects of pastırma type and salt level on volatile compounds (mean value ± standard deviation) (AU × 10⁶).

Compound	KI	Salt Level		Pastırma Type				PT × SL
		3%	5%	Kuşgömü	Sırt	Bohça	Şekerpare
Aldehydes
Acetaldehyde	623	36.93 ± 5.86 a	31.73 ± 4.49 b	32.81 ± 5.93 a	37.57 ± 8.22 a	34.21 ± 2.94 a	33.05 ± 4.14 a	NS
3-methylbutyraldehyde	686	0.93 ± 1.47 b	3.02 ± 4.89 a	1.97 ± 3.70 ab	0.27 ± 0.44 b	4.80 ± 5.68 a	1.06 ± 1.84 b	NS
2-methyl-2-butenal	788	6.27 ± 4.40 a	5.86 ± 4.17 a	5.91 ± 4.18 a	5.35 ± 3.89 a	8.39 ± 4.24 a	4.91 ± 4.58 a	NS
Hexanal	849	7.83 ± 7.21 a	5.83 ± 6.07 a	11.88 ± 6.41 a	4.48 ± 3.64 b	5.10 ± 3.22 b	5.78 ± 9.32 b	NS
Heptanal	955	0.73 ± 1.09 a	0.70 ± 0.99 a	1.08 ± 1.30 a	0.82 ± 0.96 a	0.00 ± 0.00 a	0.86 ± 1.08 a	NS
2-heptenal	1019	0.15 ± 0.23 a	0.03 ± 0.09 a	0.13 ± 0.26 a	0.12 ± 0.20 a	0.08 ± 0.12 a	0.04 ± 0.10 a	NS
Benzaldehyde	1026	3.54 ± 2.36 a	1.95 ± 1.43 b	2.20 ± 1.07 a	2.56 ± 1.79 a	4.61 ± 2.74 a	1.95 ± 1.86 a	NS
Octanal	1051	0.75 ± 0.48 a	0.53 ± 0.44 a	0.90 ± 0.57 a	0.69 ± 0.54 a	0.56 ± 0.31 a	0.42 ± 0.31 a	*
Decanal	1267	0.22 ± 0.41 a	0.15 ± 0.21 a	0.43 ± 0.53 a	0.13 ± 0.13 ab	0.00 ± 0.00 a	0.17 ± 0.24 ab	NS
Ketones
2.3-butanedione	657	9.74 ± 8.37 a	13.73 ± 16.31 a	8.06 ± 9.50 b	7.04 ± 4.91 b	25.38 ± 18.18 a	7.92 ± 7.64 b	NS
3-pentanone	733	1.94 ± 2.88 a	1.77 ± 2.31 a	3.6 ± 4.41 a	0.791 ± 0.47 a	2.111 ± 1.32 a	0.961 ± 1.25 a	NS
3-hydroxy -2-butanone	779	5.18 ± 5.85 a	8.25 ± 12.87 a	3.63 ± 4.04 b	4.34 ± 3.99 b	16.31 ± 16.87 a	3.57 ± 4.30 b	NS
2-heptanone	946	1.80 ± 1.34 a	1.46 ± 1.55 a	1.73 ± 1.86 a	1.50 ± 1.55 a	2.02 ± 1.10 a	1.33 ± 1.26 a	NS
2,3-octandione	1025	1.35 ± 1.77 a	0.67 ± 0.98 a	1.07 ± 1.01 a	0.25 ± 0.48 a	1.74 ± 2.51 a	1.13 ± 1.12 a	**
6-methyl-5-hepten-2-on	1048	0.70 ± 0.46 a	0.34 ± 0.25 b	0.39 ± 0.35 b	0.44 ± 0.41 b	0.95 ± 0.46 a	0.39 ± 0.14 b	NS
4-methyl-, 2-hexanone	1052	0.40 ± 0.33 a	0.24 ± 0.24 a	0.32 ± 0.39 a	0.33 ± 0.34 a	0.43 ± 0.21 a	0.22 ± 0.20 a	NS
Alcohols
Ethanol	512	30.53 ± 11.22 a	26.55 ± 11.60 a	36.21 ± 11.21 a	27.93 ± 8.94 ab	18.80 ± 8.22 b	30.24 ± 11.26 ab	NS
Isobutyl alcohol	643	2.81 ± 3.18 a	1.08 ± 0.71 a	2.51 ± 4.62 a	1.74 ± 1.72 a	2.06 ± 0.63 a	1.59 ± 0.83 a	NS
1-penten-3-ol	743	1.28 ± 1.89 a	2.59 ± 3.06 a	2.29 ± 3.32 a	0.92 ± 0.63 a	2.05 ± 1.87 a	2.42 ± 3.50 a	NS
3-methyl-1-butanol	783	1.86 ± 2.80 a	2.66 ± 5.36 a	0.52 ± 0.95 b	0.62 ± 0.06 b	7.16 ± 6.36 a	1.296 ± 2.59 b	NS
2-penten-1-ol	829	1.01 ± 0.81 a	0.89 ± 0.80 a	0.89 ± 0.60 a	0.69 ± 0.52 a	1.60 ± 0.88 a	0.71 ± 0.93 a	NS
1-hexanol	930	10.73 ± 10.36 a	6.07 ± 6.75 a	4.20 ± 1.51 b	8.06 ± 10.03 b	17.97 ± 10.29 a	4.85 ± 5.01 b	NS
2-ethyl-1-hexanol	1084	0.73 ± 0.86 b	1.38 ± 0.87 a	2.05 ± 0.45 a	1.12 ± 0.72 b	0.35 ± 0.81 c	0.58 ± 0.60 bc	NS
1-(2-methoxypropoxy)-2-propanol	1233	0.87 ± 1.53 a	0.84 ± 1.57 a	0.75 ± 0.51 b	2.36 ± 2.40 a	0.07 ± 0.13 b	0.14 ± 0.17 b	NS
Sulfur compounds
2-propen-1-thiol	570	25.39 ± 11.92 a	23.33 ± 10.84 a	18.07 ± 9.40 b	22.21 ± 12.45 b	32.94 ± 10.42 a	25.42 ± 9.07 ab	NS
Allyl methyl sulfide	730	15.15 ± 13.60 a	13.47 ± 14.21 a	13.59 ± 16.57 a	16.03 ± 15.43 a	20.48 ± 12.21 a	8.02 ± 8.68 a	NS
3,3′-thiobis-1-propen	888	100.29 ± 104.40 a	45.56 ± 40.97 b	22.55 ± 12.22 b	85.21 ± 110.47 ab	145.09 ± 93.16 a	51.30 ± 37.73 b	NS
Propyl allyl sulfide	907	0.80 ± 0.81 a	0.43 ± 0.63 a	0.47 ± 0.54 a	0.50 ± 0.75 a	1.20 ± 0.69 a	0.39 ± 0.81 a	NS
2,4-dimethl-thiophene	909	1.53 ± 1.29 a	0.85 ± 0.92 a	0.83 ± 0.47 b	1.28 ± 1.50 ab	2.15 ± 1.26 a	0.67 ± 0.76 b	NS
3,4-dimethl-thiophene	940	1.67 ± 1.40 a	0.89 ± 0.83 a	0.83 ± 1.11 a	1.35 ± 1.58 a	2.29 ± 1.02 a	0.81 ± 0.29 a	NS
Methyl 2-propenyl disulphide	954	12.22 ± 13.22 a	7.11 ± 4.29 a	10.29 ± 4.72 a	15.92 ± 17.61 a	8.96 ± 4.498 a	3.73 ± 2.12 a	NS
Trans propenyl methyl disulphide	969	1.26 ± 1.41 a	0.70 ± 0.50 a	1.34 ± 0.62 ab	1.89 ± 1.57 a	0.49 ± 0.35 bc	0.17 ± 0.13 c	NS
di-2-propenyl disulphide	1135	77.23 ± 53.07 a	42.61 ± 18.58 b	36.97 ± 21.44 a	83.52 ± 73.82 a	70.01 ± 26.24 a	52.61 ± 12.33 a	NS
Esters
Hexanoic acid ethyl ester	1014	1.43 ± 1.93 a	0.81 ± 0.95 a	0.53 ± 0.82 a	2.36 ± 2.16 b	1.08 ± 1.15 a	0.53 ± 1.06 a	**
Acetic acid hexyl ester	1050	0.57 ± 0.36 a	0.43 ± 0.53 a	0.46 ± 0.74 a	0.48 ± 0.39 a	0.66 ± 0.28 a	0.43 ± 0.25 a	NS
2,4-Hexadienoic acid methl ester	1075	0.64 ± 0.40 a	0.81 ± 0.73 a	0.93 ± 0.84 a	0.62 ± 0.51 a	0.79 ± 0.45 a	0.55 ± 0.46 a	NS
Hexanoic acid hexyl ester	1133	2.37 ± 2.24 a	2.06 ± 2.90 a	1.68 ± 1.11 a	1.36 ± 1.04 a	4.42 ± 3.54 a	1.69 ± 2.87 a	NS
Butanoic acid hexyl ester	1221	0.69 ± 1.61 a	0.94 ± 1.81 a	0.41 ± 0.83 b	0.00 ± 0.00 b	2.30 ± 2.67 a	0.71 ± 1.49 ab	NS
Hexanoic acid penthyl ester	1303	0.67 ± 1.13 a	0.23 ± 0.68 a	0.33 ± 0.93 a	0.89 ± 1.54 a	0.45 ± 0.43 a	0.16 ± 0.39 a	NS
Furans
2-pentyl-furane	1021	0.86 ± 0.50 a	0.57 ± 0.24 b	0.51 ± 0.37 a	0.72 ± 0.44 a	1.00 ± 0.55 a	0.70 ± 0.14 a	NS
Terpenes
α-pinene	950	0.64 ± 0.79 a	0.66 ± 1.01 a	0.86 ± 1.01 ab	0.32 ± 0.60 b	1.38 ± 0.99 a	0.14 ± 0.39 b	NS
β-myrcene	998	0.39 ± 0.41 a	0.09 ± 0.10 b	0.38 ± 0.52 a	0.33 ± 0.32 a	0.09 ± 0.16 a	0.16 ± 0.13 a	NS
D-Limonene	1054	2.22 ± 1.31 a	1.13 ± 0.50 b	1.48 ± 0.63 a	1.46 ± 1.31 a	2.46 ± 1.63 a	1.48 ± 0.58 a	NS
Aromatic hydrocarbons
Styrene	935	1.04 ± 0.50 a	0.68 ± 0.48 a	0.83 ± 0.29 a	0.76 ± 0.52 a	1.03 ± 0.53 a	0.87 ± 0.70 a	NS
1-methyl-2-(1-methylethyl)-benzene	1072	2.06 ± 1.69 a	1.15 ± 1.11 a	1.32 ± 0.90 a	1.31 ± 1.42 a	3.00 ± 1.97 a	1.02 ± 0.89 a	NS
Aliphatic hydrocarbons
Propene	<500	7.58 ± 6.49 a	5.28 ± 4.04 a	4.58 ± 2.65 b	4.96 ± 6.60 b	11.83 ± 5.83 a	5.16 ± 3.43 b	NS
1,4-Pentadiene	555	62.77 ± 48.75 a	18.79 ± 16.10 b	46.79 ± 48.45 ab	59.89 ± 57.41 a	21.39 ± 11.57 b	35.36 ± 33.56 ab	NS
Hexane	600	0.93 ± 1.75 a	7.23 ± 18.37 a	12.07 ± 24.75 a	0.34 ± 0.52 a	1.76 ± 1.85 a	1.46 ± 2.66 a	NS
Octane	800	1.38 ± 1.72 a	1.45 ± 2.05 a	1.45 ± 2.39 a	0.59 ± 0.80 a	2.46 ± 1.62 a	1.29 ± 2.06 a	NS
Decane	1000	0.62 ± 1.57 b	0.98 ± 1.83 a	1.06 ± 2.04 a	0.13 ± 0.25 a	1.19 ± 1.64 a	0.82 ± 2.21 a	NS
Tridecane	1300	1.78 ± 2.36 a	1.31 ± 3.26 a	2.65 ± 4.76 a	1.50 ± 2.50 a	0.87 ± 0.50 a	1.10 ± 1.49 a	NS
Tetradecane	1400	0.45 ± 0.42 a	0.29 ± 0.49 a	0.45 ± 0.66 a	0.52 ± 0.55 a	0.37 ± 0.21 a	0.15 ± 0.10 a	NS

Results are expressed in Arbitrary Units (×10⁶) as means of 3 replicates of each sample. KI: Kovats index calculated for DB-624 capillary column (J&W Scientific. 60 m. 0.25 mm i.d. 1.4 μm film) installed on a gas chromatograph equipped with a mass selective detector. a–c: Any two means in the same column that have the same letters in the same section are not significantly different at p > 0.05, ** p < 0.01, * p < 0.05. NS: not significant.

, nine aldehydes, seven ketones, eight alcohols, eight sulfur compounds, six esters, one furan, three terpenes, two aromatic hydrocarbons, and eight aliphatic hydrocarbons were detected in all the pastırma samples. In other studies on pastırma, many volatile compounds—including aliphatic hydrocarbons, esters, alcohols, sulfur compounds, aldehydes, terpenes, ketones, aromatic hydrocarbons, and furans—were identified during processing [14,49]. These compounds formed as a result of amino acid catabolism, carbohydrate fermentation, lipid oxidation, and microbial activities [54].

The pastırma type had a significant (p < 0.05) or very significant (p < 0.01) effect on hexanal, 3-methyl butyraldehyde, decanal, 2,3-butanedione, 3-hydroxy-2-butanone, 6-methyl-5-hepten-2-one, 1-hexanol, ethanol, 2-etil-1-hexanol, 3-methyl-1-butanol, 1-(2-methoxypropoxy)-2-propanol, 2-propen-1-thiol, 3,3′-thiobis-1-propen, 2,4-dimethl-thiophene, butanoic acid hexyl ester, hexanoic acid ethyl ester, 1R-alpha-pinene, 1,4- pentadiene and propene. The pastırma types showed a significant difference in terms of TBARS values. The change in the TBARS, which is an indicator of lipid oxidation, parallels the change in the volatile compound profile in pastırma types. Additionally, lipolytic activity, which occurs more frequently during oxidative and carbohydrate degradation, is greater in glycolytic muscles [45], which is thought to be the reason for the difference in volatile compounds in pastırma samples. As a matter of fact, while the glycogen formed aliphatic compounds of less than six carbons, such as acetone, ethanol, and heterocyclic volatile compounds with proteins, amino acids, peptides and polyunsaturated fatty acids formed aldehydes, alcohols, and ketones of more than five carbons through oxidation [55]. The salt level showed a significant (p < 0.05) or very significant (p < 0.01) effect on acetaldehyde, 3-methyl butyraldehyde, benzaldehyde, 6-methyl-5-hepten-2-one, 2-ethyl-1-hexanol, di-2-propenyl disulfide, 3,3′-thiobis-1-propene, 2-pentyl-furane, D-limonene, beta-myrcene, and 1,4-pentadiene. High salt content regulates the activities of different enzymes, affecting lipid oxidation in foods as well as microbial reactions [56]. On the other hand, the pastırma type and salt content interaction, which is one of the main variations, had a very significant effect (p < 0.01) on hexanoic acid ethyl ester and 2,3-octadione, and a significant (p < 0.05) effect on octanal (Table 5).

The results of the correlation heat map show the relationship between salt level and volatile compounds (Figure 5a) and the relationship between pastırma type and volatile compounds (Figure 5b). The use of different salt levels in production caused differences in the abundance of volatile compounds and the use of 3% salt increased the abundance of volatile compounds, especially the abundance of sulfur compounds (Figure 5a). This difference shows the effect of salt on protein degradation and lipid oxidation and asserts its important effect on volatile compounds [57]. On the other hand, the differences were observed between pastırma types in terms of volatile compounds. Two main clusters were formed and the kuşgömü pastırma type was separated from other pastırma types (Figure 5b). It is thought that this result is due to the use of different muscles in the production of pastırma types. In fact, the kuşgömü type is obtained from the tenderloin [18].

4. Conclusions

During the production of pastırma, reducing the salt content from 5% to 3% caused an increase in the L* value, which is an indicator of brightness. The desired level of drying was achieved at both 3% and 5% salt levels, and the a_w value was below 0.90 for all types of pastırma. A decrease in the salt level decreased lipid oxidation and neutral lipase activity. Micrococcus/Staphylococcus and lactic acid bacteria were an important part of the microbiota in all pastırma types at both salt levels. Depending on the oxidative or glycolytic properties of the muscles used as the raw material, differences in volatile compounds were observed among the pastırma types. Additionally, the salt level also affected volatile compounds. As a consequence, it is possible to produce pastırma with typical characteristics using 3% salt in industrial pastırma production, provided that the curing process is applied at 4 °C.