Physicochemical and Sensory Properties of Mihalic Cheese

Abstract

Mihalic is a hard, brined and quite salty cheese. In this study, selected physical, chemical, and sensory properties of 15 Mihalic cheese samples (matured more than six months) in vacuum packaging were investigated using representative samples collected from its major production area in northwestern Turkey. Aroma-active compounds were analyzed by solid phase microextraction and determined by gas chromatography-olfactometry. Average dry matter and fat contents of the cheese were 60.4 and 27.4 g/100 g cheese, respectively. Significant differences were observed among the cheese samples for hardness, adhesiveness, chewiness, and gumminess. Sulfurous, free fatty acids, and barny/animal were major sensory descriptors for the cheese. Methional (boiled potato), (E,E)-2,4-nonadienal (fatty, stale), diacetyl (buttery), 1-octen-3-one (mushroom), ethyl butyrate (sweet, bubble gum), and butyric acid (rancid) were common aroma-active compounds found in Mihalic cheese.

Keywords:

• Mihalic cheese
• SPME
• Aroma
• Physical and chemical properties

INTRODUCTION

Mihalic cheese is one of the traditional cheeses of the Marmara region in Turkey. It is made from raw sheep milk and mostly produced in Bursa-Karacabey and Balıkesir in northwestern Turkey.^[1] Mihalic cheese has been also known as Maglıç, Mahlıç, Kelle, or Manyas.^[2] Mihalic is a hard, brined, and slightly acidic cheese and its color changes from cream to light yellow.^[1] It has a sharp taste and odor and a 3–4 mm diameter rind.^[1,^2] A photograph of Mihalic cheese and a flow diagram of its production are given in Figs. 1 and 2.

Figure 1  Photograph of Mihalic cheese.

Figure 2  Flow diagram of Mihalic cheese production.^[1,2]

Acceptability of cheese depends on several factors including texture and flavor properties.^[3,^4] Texture is an important attribute used for differentiation of cheese types and shaped mainly by the ripening period, composition, proteolysis during ripening, fat distribution, and production methods.^[5,^6] Texture has an effect not only on the sense of feeling in the mouth, but also on flavor perception.^[7] Typical flavor characteristics of cheeses are the results of proteolysis, lipolysis, and degradation of amino acids. Formation of flavor compounds is related to moisture, salt content, pH, and ripening conditions of cheeses^[8,^9] and gas chromatography-olfactometry (GC-O) can be used to describe aroma-active compounds of food samples.^[10,11]

Although several studies have been conducted to characterize Mihalic cheese, results are inconsistent, necessitating more comprehensive studies to generate knowledge. For example, Oner and Aloglu^[12] investigated changes in physicochemical and microbiological properties of Mihalic cheese made from raw milk. These researchers particularly studied fatty acids, amino acids, biogenic amines, and lactic acid bacteria, yeast and mold, coliform group microorganisms in the cheese during a 90-day ripening period. They also found that the amount of myristic, palmitic, stearic, and oleic acids was higher than that of other fatty acids. Chemical and microbial changes in Mihalic cheese during the ripening period were also investigated by Bulut.^[13] Hayaloğlu et al.^[14] studied the chemical and biochemical properties and microstructure of Mihalic cheeses which were packaged under brine or in polyethylene bags during a 360-day ripening period. They found that proteolysis was low due to its high salt content, and the protein matrix was more compact in the cheese packaged with polyethylene bags. Kamber^[2] reported that Mihalic cheese had moisture content ranging from 32.64 to 40.66%, fat free dry matter content from 35.59 to 39.66%, and salt content from 7.49 to 9.34%. In another study, Özcan and Kurdal^[15] showed that total solid, fat, protein, and salt contents of Mihalic cheeses on day two were 52.20, 22.58, 19.90, and 3.83%, respectively. However, no studies are available on determination of textural properties, aroma-active compounds, and sensory properties of Mihalic cheese. The objectives of this study were to determine and compare physical, chemical, sensory properties, and aroma-active compounds of traditional Mihalic cheeses produced in northwestern Turkey.

MATERALS AND METHODS

Samples

In characterization of Mihalic cheeses, a total of 30 cheese samples were collected from producers and local markets. Fifteen out of 30 cheese samples were selected based on representativeness of the region. Mihalic cheese samples were collected from towns of Balıkesir (Havran, Ivrindi, Gonen), Bursa (Karacabey), and Canakkale (Yenice). Ages of the cheese samples were at least six months. Cheese samples were placed in vacuum packaging and stored at 4°C prior to analysis.

Composition of Cheese

Cheese samples were analyzed for dry mater content (g/100 g) by oven-drying method.^[16] pH, titratable acidity (g lactic acid/100 g), salt (g salt/100 g), and ash content (g/100 g) of the cheeses were determined by Bradley et al.^[17] Fat content (g/100 g) was measured by the Gerber method.^[18] Protein content, water soluble nitrogen (WSN), 12 (g/100 g), trichloroacetic acid-soluble nitrogen (TCA-SN), and 5 (g/100 g) phosphotungstic acid-soluble nitrogen (PTA-SN) were determined by the method described by Kjeldahl.^[19]

Color Analysis

Minolta Cr-400 (Minolta, Tokyo, Japan) was used to measure color parameters of cheese samples. Colorimeter was calibrated by a white standard plate. The three coordinates of CIELAB (L*, a*, and b* values) representing lightness, redness, and yellowness,^[21] respectively, were measured from the internal and external surfaces of the cheese samples by SpectraMagic NX software. Color measurements were taken from five different locations of each sample.^[21]

Instrumental Texture Analysis

Before analysis, cheese samples taken from the refrigerator were equilibrated about 30 min at 20°C. Cheese samples were sliced into cubes (15×15×15 mm) for texture profile analysis. TA-XTPlus (Stable Micro Systems, Surrey, England) with a 25 kg load cell was used to determine texture parameters. Cheese cubes were placed under P-36 probe with a pre-test speed of 1 mm/s, test speed of 0.4 mm/s, post-test speed of 0.4 mm/s. Cubes were compressed to 40% of the original portion height, and 5 s was used between two compressions.^[22] As textural parameters, hardness (maximum force required to compress the cheese), adhesiveness (force required to remove the cheese from the probe), springiness (elasticity of cheese after force is removed), cohesiveness (strength of internal bonds of cheese), gumminess (required force to swallow cheese), chewiness (required energy to masticate cheese), and resilience (re-deformation capacity)^[23] were calculated by texture exponent software.

Mineral Composition

K, Ca, Cu, Mg, and Fe elements were analyzed by inductively coupled plasma atomic emission spectrometry (ICP-AES, Varian Pty, Australia). One gram of the cheese sample was placed in a vessel and 10 ml HNO₃ was added to determine the dissolved metals and non-metals. Heating was applied at 100°C in the first 30 min and then temperature was raised to 150°C and kept for 1.5 h. After cooling, 20 ml distilled water was added and filtered through a Whatman 42 filter paper. Samples were diluted (1:30) with distilled water. One milliliter of the solution was taken and 0.9 mL distilled water was added into vial.^[24]

Sensory Analysis

A roundtable discussion with a ten-member panel was conducted to identify the descriptive aroma and taste for the cheeses. Panelists were selected based on willingness to participate and time available. The panelists (25–41 years of age) were staff and graduate students in the Department of Food Engineering at Canakkale Onsekiz Mart University. The panelists were asked to identify and define the desriptive terms from representative cheeses. The panelists quantified the attributes using 15-point product specific scales anchored on the left with “not” and on the right with “very.” Eighty hours of training sessions were performed to familiarize the panelists with the descriptive terms used in sensory analysis. The terms used to define taste and flavor were shown in Table 1. During the training and calibration sessions, intensities of salty taste were perceived as higher than the salty 15 reference. Panelists accordingly extended the scale for this attribute. Panelists used water to rinse their mouth.^[25]

Table 1  Flavor terms with references used for sensory evaluation for Mihalic cheese

Descriptors	Definition	Reference*
Cooked	Aromatics associated with cooked milk	Milk heated to 85°C for 30 min
Whey	Aromatics associated with whey powder	Solubulize 5 g whey powder in 100 mL water
Creamy	Aromatics associated with milkfat	Cream or butter
Sulfurous	Aromatics associated with sulfurous compounds	Boiled mashed egg or Methional
Free fatty acids	Aromatics associated with butyric acid	10 mL butyric acid in methanol
Barny/Animal-like	Aromatics associated with barns and stock	5% Na–caseinate solution in water
Moisty	Aromatics associated with dirty and wet towel	Dirty/wet towel
Fruity	Aromatics associated with fruits	Fresh pineapple, ethyl hexanoate, 20 mg/kg
Nutty	Aromatics associated with nuts	Unsalted raw nuts
Storage	Aromatics associated with warehouse	Assignment by panelist
Yeast/mold	Aromatics associated with yeast/mold	Raw yeast dough, potting soil
Salty	Taste sensation elicited by salts	0.5% sodium chloride solution in water
Sour	Taste sensation elicited by acids	0.08% citric acid solution in water
Sweet	Taste sensation elicited by sugars	2% sucrose solution in water
Umami	Chemical feeling factor elicited by certain peptides and nucleotides	1% monosodium glutamate solution in water
Bite	Chemical feeling factor elicited by carbonation on the tongue	Carbonated water
Bitter	Taste sensation elicited by caffeine	0.08% caffeine solution in Water

Extraction of Aroma Compounds

Aroma compounds were extracted by solid phase microextraction (SPME) and determined by GC-O. Five grams of grated cheese samples were placed in 40 mL amber colored and screw top vial with hole cap PTFE/silicon septa (Supelco, Bellafonte, US) and then 1 g of NaCl was added. Samples were kept at 40°C in a water bath for 30 min to improve mass transfer between the cheese matrix and headspace. Two cm-50/30 μm DVB/Carboxen/PDMS stable flex SPME fiber (Supelco, Bellafonte) was used at a depth of 2 cm for extraction of volatile compounds of cheese.^[24]

GC-O

HP 6890 GC (Agilent Technologies, Wilmington, De, US) equipped with a flame ionization detector (FID), a sniffing port and split/splitless injector were used to analyze the headspace compounds of the cheese samples. Sniffing was achieved on a polar capillary column (HP-INNOWAX (30m length × 0.25 mm i.d. × 0.25 μm film thickness (d_f); J&W Scientific, Folsom, CA, US) and a non-polar column (HP-5 30 m length × 0.32 mm i.d. × 0.25 μm d_f; J&W Scientific). Helium was used as a carrier gas. Inlet pressure was 11.55 psi and total flow was 1 ml min⁻¹. Column effluent was split 1:1 between FID and olfactory port using deactivated fused silica capillaries (90 cm length × 0.25 mm i.d.). The GC oven temperature was programmed from 40 to 200°C at a rate of 10°C min⁻¹, with initial hold of 3 min and final hold time of 20 min. The FID and sniffing port were maintained at the temperatures of 250 and 200°C, respectively. Post peak intensity method was used for determination of aroma intensity.^[26] One sniffer who is experienced with the GC-O technique quantified the odor intensities using a 10-point scale anchored on the left with “not” and on the right with “very.” The sniffer had more than 80 h of experience with the GC-O technique, scale using, and odor description. Aroma-active compounds were identified by comparing retention indices (RI) and odor quality of unknowns with those of authentic standards analyzed at the same experimental conditions by the sniffer during the GC-O procedure. RI were calculated using n-alkane series.^[27]

Statistical Analysis

The multidimensional scaling (MDS) method was used for visualization of the data.^[28] MDS plots the cheese samples by specific characteristics on a map. Similar cheese samples are grouped on the map. MDS plots were drawn using SPSS statistical software.

Table 2  Physicochemical characteristics of Mihalic cheeses

Sample	pH	Titratable acidity (% lactic acid)	Dry matter (%)	Salt (%)	Fat (%)	Ash (%)	Protein (%)	WSN (%)	TCA (%)	PTA (%)
CHS1	5.79	0.37	64.1	6.66	30.5	8.42	24.6	0.35	0.10	0.03
CHS 2	5.92	0.55	60.4	3.38	30.2	4.93	22.2	0.84	0.26	0.11
CHS 3	5.60	0.59	57.8	6.77	25.0	8.14	22.8	0.30	0.06	0.03
CHS 4	5.85	0.48	58.7	6.31	25.2	5.87	24.6	0.34	0.08	0.06
CHS 5	5.83	0.50	63.9	7.25	26.2	9.76	23.7	0.26	0.05	0.03
CHS 6	5.63	0.64	62.2	6.07	28.2	7.75	22.0	0.60	0.11	0.06
CHS 7	5.54	0.64	56.7	4.90	26.2	6.98	21.8	0.48	0.11	0.06
CHS 8	5.67	0.52	56.7	4.44	25.0	5.96	21.2	0.44	0.15	0.03
CHS 9	5.72	0.44	62.4	5.84	33.0	7.38	18.6	0.42	0.13	0.06
CHS 10	5.47	0.68	58.3	4.67	27.5	6.11	21.2	0.52	0.11	0.03
CHS 11	5.22	0.41	61.3	3.27	28.0	5.66	22.7	0.65	0.14	0.08
CHS 12	5.13	0.55	59.5	6.30	25.2	7.78	22.1	0.44	0.10	0.08
CHS 13	5.09	0.50	60.2	6.77	26.5	8.62	21.6	0.51	0.13	0.08
CHS 14	5.10	0.55	63.6	7.95	27.0	10.09	21.6	0.50	0.03	0.03
CHS 15	5.16	1.04	59.8	8.18	27.0	9.58	19.5	0.42	0.09	0.03
Minimum	5.09	0.37	56.7	3.27	25.2	4.93	18.6	0.26	0.03	0.02
Maximum	5.92	1.04	64.1	8.18	33.0	10.09	24.6	0.84	0.26	0.11
Average	5.51	0.56	60.4	5.92	27.4	7.54	22.0	0.47	0.11	0.05

RESULTS AND DISCUSSION

Composition of Mihalic Cheese

Physicochemical parameters of Mihalic cheese are given in Table 2. pH and titratable acidity of the cheese samples ranged from 5.09 to 5.92 and from 0.37 to1.04%, respectively. Differences in pH values could be explained by various levels of milk titratable acidity and presence of different starter cultures.^[29] The highest value (1.04%) of titratable acidity for CHS15 could be due to lactose fermentation and formation of amino and free fatty acids with proteolysis and lipolysis.^[29] Dry matter and fat contents of the cheese ranged from 56.75 to 64.13% and from 25.25 to 33.00%, respectively. CHS1 had the highest amount of dry matter, whereas CHS7 had the lowest. CHS7 might have lost the dry matter components (e.g., water soluble proteins and peptides) to brine.^[30] Differences among the cheese samples with respect to fat contents could be attributed to lipolytic activity of microflora and transmission of fatty acids to brine.^[31] The highest ash (10.09%) and salt contents (8.18%) were observed in CHS14 and CHS15, respectively. Different salt concentrations and brining time are the most important factors affecting the salt content of cheese.^[32] Maximum protein content was found to be 24.63% (CHS1) while the minimum was 18.64% (CHS9). The highest protein content of CHS1 could be attributed to higher dry matter content. In addition, hydrolysis of proteins might have resulted in differences in protein content.^[30] Similar results were also found by other studies.^[12,¹³,¹⁵,^33]

The nitrogen fractions could be used to understand the development of proteolysis. WSN, TCA-SN, and PTA-SN ranged from 0.26 to 0.84, from 0.03 to 0.26, and from 0.028 to0.112, respectively. CHS2 had the highest amount of WSN, TCA-N, and PTA-N whereas CHS5 and CHS14 had the lowest. This difference could be attributed to high casein hydrolysis by rennet and milk proteases in CHS2.^[29] In addition, the lower WSN and TCA-N contents are due to the presence of serum proteins in CH5 and CHS14, which are more resistant to enzymatic proteolysis than other cheeses.^[30]

Table 3  Descriptive statistics of color parameters

Sample	L (External)	−a (External)	b (External)	L (Internal)	−a (Internal)	b (Internal)
CHS 1	81.07	3.07	16.25	73.40	2.66	14.20
CHS 2	78.48	2.23	12.92	78.14	2.03	12.46
CHS 3	83.91	1.06	12.25	73.73	1.48	11.35
CHS 4	86.17	0.97	12.00	75.44	1.82	11.58
CHS 5	88.40	1.43	9.37	78.90	2.85	11.54
CHS 6	77.05	2.71	13.40	69.36	2.63	12.52
CHS 7	75.92	2.03	13.53	72.19	1.81	13.22
CHS 8	79.53	0.88	13.50	73.55	1.45	13.05
CHS 9	83.54	2.21	15.63	75.31	2.17	14.27
CHS 10	78.62	1.90	14.26	69.86	1.87	12.93
CHS 11	84.40	2.53	17.86	71.16	2.94	16.55
CHS 12	82.04	1.72	14.25	71.58	2.46	13.43
CHS 13	83.51	2.45	16.64	72.91	2.58	15.45
CHS 14	80.74	3.41	16.58	66.84	4.07	13.35
CHS 15	84.95	2.55	18.26	77.13	3.42	17.85
Minimum	75.92	0.88	9.37	66.84	1.45	11.35
Maximum	88.40	3.41	18.26	78.90	4.07	17.85
Average	81.89	2.08	14.45	73.30	2.42	13.58

Color

Determination of L* and b* values are important color parameters for cheese. Because L* value shows the lightness and b* represents the yellowness of a sample. L* and b* parameters define both interior and surface of the cheese samples. Table 3 presents the results obtained from the external and internal color analysis of Mihalic cheese. The results of this study showed that surface L* values of CHS5 (88.40) was the highest, whereas that of CHS7 (75.92) was the lowest. CHS5 (78.90) and CHS2 (78.14) have the highest interior L* values. The difference in L* values of the samples may be due to salt distribution in cheese structure and swelling phenomena caused by concentration difference of salt in brine.^[34] Because high salt concentration leads to an increase in lipid oxidation and change in the color of cheese.^[34] Another explanation is that various milk sources and heat treatments in cheese making may have resulted in differences in cheese color.^[35] Because of heat treatment, some of the proteins in cheese form gel particles, which scatter light and increase the L* value of samples.^[36] Findings of the current study are consistent with those of Rohm and Jaros^[37] who found that L* value of Emmental cheese ranged from 75 and 80. Furthermore, these findings are in agreement with previous results by El-Nimr et al.^[38] showing that L* value of Egyptian Gouda cheese was around 80 throughout a storage period of 60 days. The present work agrees with the previous research in that an increase in salt concentration decreased the L* value of Gaziantep cheese.^[34]

Surface b* values showed that CHS11 (17.86) and CHS15 (18.26) had the highest values, whereas CHS5 (9.37) had the lowest (Table 3). Similar results were obtained for interior b* values. CHS15 had the highest value (17.85), whereas CHS3, CHS4, and CHS5 had the lowest values. It is possible that these results are due to different milk sources and bacterial flora on the cheese surface. Cows can pass dietary carotenoids to milk, whereas sheep and goats cannot. Therefore, cheese made from cow milk appears to be yellower.^[39] Also, Brachybacterium, a high salt tolerant microorganism lives on the surface of cheese, giving it a yellow color.^[40] The results corroborate the findings of Saldo et al.^[41] who showed that b* values of hard Garrotxa cheese were around 11. In a previous research on b* values of cheese as a function of ripening time by Buffa et al.,^[42] it was reported that an increase in b* value from 9.98 to 11.05. These results are in agreement with the presented results showing changes between 9.37 and 18.26. Similarly, Trobetas et al.^[43] reported that yellowness (b* value) of grated Graviera hard cheese changes between 14 and 19 with respect to storage time and different treatments. In this study, cheese samples were provided from different sources, and had had different compositions and proteolysis levels. Age of the cheese can also be different. These factors may result in color differences in cheese. Proteolysis, lipolysis, moisture, different ripening conditions, and time could also affect color properties of cheese. Covering the cheese by brine prevents drying of cheese surface, resulting in no detectable color change.^[40]

Texture

Texture is one of the important components of cheese quality. Texture and overall acceptance are more important than the flavor component of cheese.^[44] These factors are the primary consideration for consumers.^[45]. As shown in Table 4, large variations were observed among the cheese samples for hardness, adhesiveness, gumminess, and chewiness measurements. CHS5 had the highest hardness value (129.93), whereas CHS8 had the lowest (24.48). Moreover, the highest values of springiness, gumminess, chewiness, and resilience were obtained for CHS1 (0.88), CHS5 (84.93), CHS5 (73.14), and CHS11 (0.42), respectively.

Table 4  Texture profile parameters of Mihalic cheeses

Sample	Hardness (N)	Adhesiveness (–) (g. sec)	Springiness	Cohesiveness	Gumminess (N)	Chewiness (N)	Resilience
CHS1	80.72	31.21	0.88	0.54	44.83	39.49	0.29
CHS 2	36.70	64.25	0.78	0.51	19.18	15.12	0.26
CHS 3	61.30	41.99	0.85	0.64	40.52	34.59	0.35
CHS 4	51.88	34.37	0.86	0.61	32.61	28.16	0.33
CHS 5	129.93	3.78	0.86	0.64	84.93	73.14	0.34
CHS 6	77.29	112.96	0.82	0.54	42.40	35.18	0.29
CHS 7	36.45	78.27	0.82	0.50	18.66	15.46	0.27
CHS 8	24.44	90.29	0.82	0.61	15.15	12.63	0.32
CHS 9	50.04	36.33	0.87	0.53	27.28	23.84	0.26
CHS 10	36.15	94.32	0.81	0.47	17.33	14.08	0.23
CHS 11	61.94	41.48	0.85	0.71	44.70	38.32	0.42
CHS 12	58.02	44.26	0.87	0.54	32.24	28.12	0.29
CHS 13	67.20	43.45	0.84	0.61	41.85	35.40	0.34
CHS 14	76.85	9.19	0.85	0.44	34.43	29.35	0.20
CHS 15	73.02	10.0	0.84	0.57	42.70	36.43	0.26
Minimum	24.44	3.78	0.78	0.44	15.15	12.63	0.20
Maximum	129.93	112.96	0.88	0.71	84.93	73.14	0.42
Average	61.46	49.08	0.84	0.56	35.92	30.62	0.30

Figure 3 provides MDS results obtained from the texture profile analysis parameters. The cheese samples that are grouped together by texture properties are similar, whereas the samples that fall apart are dissimilar for the same properties. CHS4, CHS13, CHS11, and CHS3 samples are grouped together, especially CHS4, CHS13, and CHS3. Separation of CHS11 from others could be due to high values of cohesiveness and resilience. As can be seen in Fig. 3, CHS5 is clearly separated because this sample had the highest hardness, gumminess, and chewiness values.

Figure 3  Geometrical representation of cheeses in terms of textural characteristics by multidimentional scaling (MDS). Each CHS represents different cheese among 15 samples.

In this work, cohesiveness and chewiness results confirm the findings of Sameh^[46] that “Ras” hard cheese had values between 0.39 and 0.53 and between 12.67 and 25.64 N for cohesiveness and chewiness, respectively, during 180 days of storage. The same research suggested that low chewiness values could be due to high proteolysis and reduction in protein interaction in the cheese.^[46,47] Springiness values for Mihalic cheese samples were in the similar range with the findings of Zhang and Zhao^[47] who reported that springiness values did not change appreciably over 21 days of storage in a semi hard cheese. The values of hardness obtained in the present work were found to be similar to those reported by Cerqueira et al.,^[48] ranging around 25 N for “regional” semi hard cheese that was stored at 20°C for 7 days. It can be concluded that high hardness values are related to decreasing moisture.^[46]

In general, it is possible that differences among samples might be related to the variety of milks, ratio of milk types used in cheese making, ripening conditions, proteolysis level, salt content of brine, and process techniques. For example, CHS5 had the lowest WSN and PTA levels (Table 2). However, the same cheese sample had the highest hardness, gumminess, and chewiness measurements (Table 4). These results were supported by the findings of Delgado and coworkers.^[49] They showed the relationship between textural and proteolytic parameters for a Spanish soft cheese (Torta del Casar). In the same study, they found an increase in casein degradation during maturation while decreasing the hardness and consistency and increasing the adhesiveness in the cheese.

Table 5  Mineral compositions of Mihalic cheeses

Sample	K (mg/100 g)	Ca (mg/100 g)	Cu (mg/100 g)	Mg (mg/100 g)	Fe (mg/100 g)
CHS 1	7.2	304.2	0.9	18.0	0.9
CHS 2	26.1	650.7	0.9	33.3	3.6
CHS 3	9.0	130.5	0.9	13.5	1.8
CHS 4	10.8	324.0	0.9	19.8	0.9
CHS 5	6.3	439.2	1.8	25.2	8.1
CHS 6	12.6	153.9	0.9	14.4	0.9
CHS 7	10.8	127.8	0.9	16.2	0.9
CHS 8	11.7	265.5	0.9	27.9	1.8
CHS 9	16.2	603.9	21.6	29.7	3.6
CHS 10	10.8	189.0	0.9	15.3	0.9
CHS 11	13.5	260.1	0.9	21.6	0.9
CHS 12	9.0	113.4	0.9	10.8	0.9
CHS 13	9.9	131.4	0.9	16.2	0.9
CHS 14	7.2	122.4	0.9	10.8	0.9
CHS 15	9.0	254.7	0.9	21.6	0.9
Minimum	6.3	113.4	0.9	10.8	0.9
Maximum	26.1	650.7	21.6	33.3	8.1
Average	11.7	271.8	2.7	19.8	1.8

K: Potassium, Ca: Calcium, Cu: Copper, Mg: Magnesium, and Fe: Iron.

Mineral Composition

Mineral composition of Mihalic cheese samples is shown in Table 5. Their mineral composition ranges from 6.3 to 26.1 mg/100 g, from 113.4 to 650.7, from 0.9 to 21.6, from 10.8 to 33.3, and from 0.9 to 8.1 mg/100 g for K, Ca, Cu, Mg, and Fe elements, respectively. Figure 4 shows the geometrical representation of the samples in terms of mineral composition. Large distances between CHS2, CHS9 and other samples were observed. In other words, CHS2 and CHS9 samples did not group with other samples with respect to mineral contents. These differences can be explained by the highest concentration of K, Ca, and Mg elements in CHS2 and the highest concentration of Cu element in CHS9 sample. High concentration of Mg in CHS2 may be due to the ratio of milk types used in cheese making. In addition, K, Mg, and Ca concentrations were affected by the season, production method, and brining time of cheese.^[50−52]

Figure 4  Geometrical representation of cheeses in terms of mineral composition by MDS. Each CHS represents different cheese among 15 samples.

Iron (Fe) concentrations of Mihalic cheese samples were in agreement with findings of Baquero et al.^[53] who showed that Fe concentration was between 1.28 and 1.31 mg/100 g for hard goat cheese, depending on the cheese technology, diet, and rennet composition. Concentration ranges of Mg in Mihalic cheese were consistent with those of Korenovská and Suhaj^[54] who found that Emmental and Edam cheese had Mg content between 17.6 and 47.3 mg/100 g originating from United Kingdom. In addition, high level of Ca concentration in Mihalic cheese appears to be similar to the results quoted by Korenovská and Suhaj^[54] who described that the high Ca contents are related to no curd washing in cheese making. The present results for K concentration were lower than those reported earlier (30.5–36.2 mg/100 g).^[55] In this work, only CHS5 and CHS9 had high concentrations of Cu, higher than 0.9 mg/100 g. This could be due to high contact time of milk and cheese with metallic vessels.^[56]

Sensory Analysis

Figure 5 shows the similarities among the cheese samples for sensory attributes. CHS4 and CHS7 grouped together, suggesting that they are similar for sensory attributes. In addition, CHS10 and CHS11 were closely clustered. Sensory analysis showed that sulfurous, free fatty acids, creamy, cooked, whey, and salty were the major descriptive terms of Mihalic cheese samples (Table 6). Salt is an important factor determining Mihalic cheese quality. Salty taste also depends on the brining process and salt distribution in cheese.^[39] Mihalic cheese is brined in two or three steps, depending on cheese plant. In the present work, CHS14 had the highest salty taste score (23.25), and CHS11 had the lowest value (11.80). These results were supported by the compositional analysis (Table 2). CHS14 had higher salt content than most other cheese samples, while CHS11 had the lowest salt content (Table 2). Volatile sulfur compounds consisting of methanethiol, hydrogen sulfide, and dimethyl sulfide play a key role in determining cheese flavor. Increased storage time results in breakdown of sulfur-containing amino acids, such as cysteine and methionine. During storage, methanethiol is oxidized to dimethyl sulfide, dimethyl disulfide, dimethyl trisulfide and other sulfur compounds.^[57] CHS15 had the highest sulfurous value (3.48), whereas CHS11 had the lowest value (1.86). Animal-like odor intensity was the highest for CHS7 (2.42), and it was related to p-cresol, described as a barny odor.^[58] With respect to free fatty acids flavor, CHS14 had the highest value (2.93), whereas CHS11 had the lowest value (1.96). Primary and secondary reactions are responsible for formation and degradation of free fatty acids. Hydrolysis of triglycerides is the main reaction for production of free fatty acids, and acids between C4 and C12 have an important role for contributing specific flavors (rancid, sharp, goaty, and soapy). Free Fatty Acids are associated with pH of the medium, distribution between aqueous and fat phase, cations, and protein degradation products.^[59] Storage flavor, which is not a desirable flavor in cheese, was found at the highest intensity in CHS8 (1.53). Sourness scores were between 1.27 and 2.13 for samples, and the highest value was found in CHS14. Bitterness, a major cheese defect, depends on the accumulation of hydrophobic peptides during proteolysis of casein. In addition, amino acids, amines, amides, long chain ketones, and some monoglycerides play roles in formation of bitter taste.^[60] With respect to bite intensity, CHS14 had the highest value (2.45), whereas CHS11 had the lowest value (0.93). Umami is one of the five basic tastes

Table 6  Sensory attributes of Mihalic cheeses

Sample	Cooked	Why	Crm	Sulf	FFA	Barny/Anim	Moisty	Fruity	Nutty	Str	Yea/Mol	Sour	Bitt	Salt	Swt	Umami	Bite
CHS 4	1.92	1.85	1.95	2.30	2.17	1.25	1.56	0.90	0.57	1.06	1.40	1.35	0.57	18.42	0.77	0.90	1.53
CHS 5	2.00	2.20	1.91	2.58	2.82	2.31	1.42	0.70	0.32	0.80	1.53	1.65	0.41	23.12	0.60	1.38	2.42
CHS 7	2.13	1.92	1.86	2.26	2.38	2.42	1.33	0.55	0.43	0.77	1.47	1.47	0.40	16.12	0.62	0.97	1.07
CHS 10	2.26	2.12	2.27	2.00	2.13	1.23	1.01	0.72	0.51	0.55	1.12	1.86	0.50	13.60	1.06	1.46	1.00
CHS 11	2.15	2.06	2.23	1.86	1.96	1.15	0.93	0.72	0.47	0.66	1.16	1.62	0.47	11.80	1.07	1.20	0.93
CHS 12	1.87	1.65	1.80	2.18	2.23	1.53	1.23	0.57	0.42	0.72	1.27	1.68	0.45	20.07	0.65	1.21	2.20
CHS 14	2.22	1.92	2.02	2.91	2.93	1.96	1.10	0.55	0.40	0.61	1.27	2.13	0.55	23.25	0.67	1.45	2.45
CHS 15	2.07	2.13	1.97	3.48	2.71	1.71	1.13	0.42	0.31	0.52	1.23	2.10	0.43	22.62	0.60	1.55	2.25
Minimum	1.87	1.65	1.80	1.86	1.96	1.15	0.93	0.42	0.31	0.52	1.12	1.35	0.40	11.80	0.60	0.90	0.93
Maximum	2.26	2.20	2.27	3.48	2.93	2.42	1.56	0.90	0.57	1.06	1.53	2.13	0.57	23.25	1.07	1.55	2.45
Average	2.08	1.98	2.00	2.45	2.42	1.71	1.21	0.64	0.43	0.71	1.31	1.73	0.47	18.63	0.76	1.27	1.73

Why: Whey, Crm: Creamy, Sulf: Sulfurous, FFA: Free fatty acid, Barny/anim: Barny/animal like, Moisty: Moisty cloth, Str: Storage, Yea/mol.: Yeast/mold, Bitt: Bitter, Salt: Salty, and Swt: Sweet.

Figure 5  Geometrical representation of cheeses in terms of sensory attributes. Each CHS represents a different cheese among eight selected samples.

and generally defined as monosodium glutamate taste. Glutamic acid, liberated by proteolysis, is responsible for the accumulation of umami taste. This taste is generally used as a flavor enhancer in the food industry.^[61] In this study, CHS15 had the highest umami taste (1.55) and CHS4 had the lowest value (0.90) for this attribute. In addition to glutamic acid, some organic acids, amino acids (Ile, Lys, and Tyr due to the non-polar or hydrophobic side chains), nucleotides, and some γ-glutamyl peptides (γ-Glu-Phe) are responsible for umami taste.^[39,61]

In general, similar results were found by other studies in “Ras” and “Sepet” cheese. Ayad et al.^[62] identified the bitter, rancid, unclean, salty, moldy, and fatty attributes for “Ras” cheese by descriptive sensory analyses. Ercan et al.^[63] observed the sour, bite, free fatty acids, animal like, and sulfurous characteristic terms for “Sepet” cheese. In addition, cooked, whey, and sweet terms were determined as descriptive for Circassian cheese by Guneser and Karagul Yuceer.^[64]

Aroma-Active Compounds

Formations of volatile compounds are due to degradation reactions as glycolysis (lactose and citrate), lipolysis (milk lipids), and proteolysis (caseins). Primary breakdown of constituents triggers the formation of precursors of flavor compounds, but only some of them directly affect the cheese flavor.^[59] A total of 30 volatile compounds were identified in the headspace of Mihalic cheese samples (Table 7). Major aroma-active compounds were aldehydes, ketones, esters, acids, and sulfur compounds and five unknowns. Some aroma-active compounds in Mihalic cheese were detected at low intensities, including acetoin with buttery note and 3-methylthiophene with plastic note. Similar results were also observed in fresh Chevre-style goat cheese and Ezine cheese.^[65,66] Acetoin can be formed by two different reactions as enzymatic condensation of acetaldehyde by acetolactate decarboxylase or from diacetly reduction by diacetyl reductase and carbohydrate metabolism.^[67,68] 3-Methylthiophene is produced via Maillard reaction and known as a major heterocyclic sulfur compound.^[69] Previous studies revealed that this compound was found in Chevre style goat, Ezine, and Circassian cheeses.^[64−66] However, intensities of diacetyl (buttery), ethyl butyrate (sweet, bubble gum), butyric acid (rancid, cheesy), and hexanoic acid (cheesy, sour) were very high in the cheese.

Intensities of butyric acid varied from 6 to –7.7 in the cheese samples. High intensities of butyric acid could be due to its proportion in milk fat and selective hydrolization and synthesization by the cheese microflora.^[59] Previous research findings showed that this compound is found in Cheddar, Ezine, Sepet, Circassian, and Gruyere cheeses.^{[8,64−66,70−72]} In all Mihalic cheese samples, diacetyl intensity was between 3.5 and 7.0. Diacetly results from the oxidative decarboxylation of α-acetolactate and produced by many lactic acid bacteria.^[68] This compound is an important flavor compound of butter, buttermilk, and some young cheeses.^[66,72] Several studies previously reported that this compound is found in Cheddar, Ezine, Circassian, and Sepet Cheese.^[8,64−66] Except in CHS11, hexanoic acid was determined in all samples. CHS5 had the highest hexanoic acid intensity (8.5). This fatty acid originated from the degradation of lactose and aminoacids, and it can be formed from ketones, esters, and aldehydes by oxidation.^[73] This compound was also found in Sepet, Cheddar, Niva, Circassian, and Gruyere cheeses.^{[8,54,59,64,71,74]} Ethyl butyrate intensities of the samples ranged from 1.5 to 5.8. It can be formed by esterification, alcoholysis, acidolysis, or transesterification. Previous studies reported that this compound is found in Ezine, Sepet, Cheddar, and Circassian cheeses.^[8,64−66]

Table 7  Aroma-active compounds of Mihalic cheeses

		Retention index^a			Aroma intensity^c
No	Chemical	HP-5	INNOWAX	Aroma quality^b	CHS4	CHS5	CHS7	CHS10	CHS11	CHS12	CHS14	CHS15
1	Dimethyl sulfide	<600	–	Sulfurous	0.8	5.3	0.7	0.7	0.7	0.5	1.5	0.3
2	Diacetyl	<600	1006	Buttery	5.7	3.5	7.0	7.0	5.3	4.0	6.3	6.5
3	Acetic acid	628	1384	Vinegar, sour	–	4.0	5.7	5.7	–	3.3	4.0	1.5
4	Unknown 1	659	1368	Sulfurous, plastic	–	6.7	1.7	1.7	–	0.5	2.8	–
5	Acetoin	715	–	Buttery	1.0	0.5	0.5	0.5	–	1.0	0.5	–
6	Unknown 2	772	1329	Ester, fruity	0.7	–	3.3	3.3	4.7	2.5	2.3	0.9
7	3-Methylthiophene	784	–	Plastic	1.3	0.4	–	–	–	1.4	1.5	–
8	Ethyl butyrate	813	1063	Sweet, bubble gum	1.5	5.5	5.7	5.7	1.7	4.3	5.8	5.8
9	Butyric acid	848	1579	Rancid, cheesy	7.3	7.0	7.5	7.5	7.7	6.5	6.0	6.5
10	2-Methyl-3-furanthiol	884	–	Vitamin	–	4.0	–	–	–	–	–	–
11	o-Xylene	889	1188	Geranium	5.5	–	2.0	2.0	4.7	–	1.5	2.8
12	Unknown 3	894	–	Dirty, moisty cloth	–	–	–	–	5.0	–	–	–
13	Heptanal	899	1083	Green grassy	7.3	–	4.7	4.7	5.3	–	3.0	2.5
14	(Z)-4-Heptanal	910	1215	Dirty fatty	–	5.3	–	–	–	–	5.8	2.0
15	Unknown 4	916	–	Ester, fruity	7.5	–	–	–	–	–	–	–
16	Methional	921	1392	Boiled potato	1.3	7.3	5.7	5.7	5.7	5.7	7.5	5.5
17	Propionic acid	927	1435	Acid, sour	4.0	–	–	–	–	–	–	–
18	2-Acetylthiazole	943	1409	Popcorn	–	1.0	3.0	3.0	4.0	3.5	0.8	1.3
19	1-Octen-3-one	993	–	Mushroom	2.5	8.3	2.7	2.7	3.0	4.5	2.5	3.3
20	Ethyl hexanoate	1011	1279	Fruity, ester	–	7.0	5.5	5.5	–	0.5	8.0	3.3
21	Hexanoic acid	1061	1743	Cheesy, vinegar, sour	0.7	8.5	7.7	7.7	–	4.3	7.8	7.8
22	2-Nonanone	1074	–	Oxide, soil	1.0	–	2.0	2.0	3.5	5.3	4.5	3.5
23	2-Isopropyl-3-methoxypyrazine	1093	–	Dirty, nutty	–	–	–	–	–	1.5	–	–
24	2-Acetyl-2-thiazoline	1139	1763	Popcorn	–	–	1.8	1.8	–	–	1.8	2.5
25	Nonanal	1146	1426	Fatty, stale, soapy	1.9	4.5	–	–	–	–	2.5	–
26	Ethyl octanoate	1159	1435	Fruity, ester	1.5	3.5	1.5	1.5	0.5	3.5	3.8	2.3
27	(E)-2-Nonenal	1160	1569	Fatty, floral	–	–	–	–	–	–	2.0	–
28	(E,E)-2,4-Nonadienal	1207	1589	Fatty, stale, soapy	5.0	6.3	3.5	3.5	6.5	5.3	5.3	2.8
29	(E,E)-2,4-Decadienal	1302	–	Fried fatty	–	–	–	–	–	–	3.5	–
30	Unknown 5	1388	1674	Oxide fatty	–	–	–	–	–	–	2.8	–

^aRetention indies on INNOWAX and HP5 columns.

^bAroma quality determined on olfactory port.

^cMean aroma intensities given by one sniffer on HP5 column. Odor intensities represent the mean of each sample evaluated in duplicate by one sniffer.

Propionic acid was only identified in CHS4 at an intensity of 4.0. Propionic acid bacteria in this cheese could contribute to lipolysis with its high lipolytic activity.^[75] This compound was also found in Ezine, Niva, and Gruyere cheeses.^[66,71,76] 2-isopropyl-3-methoxypyrazine was only found in CHS12. It contributed a “dirty, nutty note” to cheese with intensity at 1.5. It was also found in British farmhouse Cheddar cheese with “soil like note.”^[58] Zehentbauer and Reineccius^[72] reported that this compound was found in Cheddar cheese. Aldehydes, such as (E)-2-nonenal and (E,E)-2,4-decadienal, were also found only in CHS14. These compounds were also determined as aroma-active compounds in Chevre, Ezine, Circissian, and Gruyere cheeses.^[61−63,71]

CONCLUSION

In conclusion, results of this study showed differences among Mihalic cheese samples in terms of physical, chemical, sensory properties, and aroma-active compounds. As a traditional and regional cheese, Mihalic cheese is a very hard cheese. Texture is related to sensory perception of food products. Cheese samples were different from each other in terms of textural properties including hardness, adhesiveness, chewiness, and gumminess. Major descriptive terms for the samples were sulfurous, free fatty acid like, barny/animal flavors, and salty taste. Characteristic aroma-active compounds were diacetyl (buttery), ethyl butyrate (sweet, bubble gum), butyric acid (cheesy), methional (boiled potato), 1-octen-3-one (mushroom), and (E,E)-2,4 non-adienal (fatty).

ACKNOWLEDGMENTS

The authors would like to thank the panel members for their participation and input during panel training and product evaluation. They also thank the producers who provided the cheese samples.

FUNDING

This research was funded by Scientific Research Fund of Canakkale Onsekiz Mart University (No: 2008/46).