Available online at www.sciencedirect.

com

Food Control 19 (2008) 278285

www.elsevier.com/locate/foodcont

Eects of heat treatment and starter culture on the properties

of traditional Urfa cheeses (a white-brined Turkish cheese)
produced from bovine milk
A. Ferit Atasoy
a

a,*

zer
, Atilla Yetismeyen b, Huseyin Turkoglu c, Barbaros O

Sanlurfa Vocational High School, Department of Food Technology, Harran University, 63010 Sanlurfa, Turkey
b
Faculty of Agriculture, Department of Dairy Technology, Ankara University, Ankara, Turkey
c
Faculty of Agriculture, Department of Food Engineering, Harran University, Sanlurfa, Turkey
Received 5 December 2005; received in revised form 4 April 2007; accepted 10 April 2007

Abstract
The objective of the present research was to determine the eects of heat treatment and starter culture on some properties of Urfa
cheese produced from bovine milk. Proteolysis developed more rapidly in the cheese made with mesophilic starter culture. Whereas
the cheese inoculated with thermophilic starter culture received higher sensory scores. High heat treatment (at 72 C for 5 min) aected
the sensory and chemical properties of Urfa cheese adversely. Therefore, incorporation of low heat treatment (at 65 C for 20 min) is
recommended for the production of Urfa cheese, with addition of a culture of Lactococcus lactis subsp. lactis + Lactococcus lactis subsp.
cremoris.
 2007 Elsevier Ltd. All rights reserved.
Keywords: Urfa cheese; Heat treatment; Starter culture

1. Introduction
Urfa cheese is a traditional semi-hard brined Turkish
cheese type produced mainly in the southeast Anatolia
region of Turkey from raw ovine milk or appropriate mixtures of ovine and caprine milk, without any starter culture. Since lactation period of ovine and caprine milk in
Turkey is very short (approximately 67 months), it is
not always possible to extend Urfa cheese production over
the all year without the use of bovine milk. Recently, the
industrial production of Urfa cheese has been made from
bovine milk, and therefore, today this cheese variety has
gained nationwide popularity.
The nal sensory quality of cheese is aected by the
treatment of milk including heat treatment. Heat treatment
*

Corresponding author. Tel.: +90 414 247 03 91/279; fax: +90 414 247
03 92.
E-mail address: [email protected] (A.F. Atasoy).
0956-7135/$ - see front matter  2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodcont.2007.04.004

to cheese milk is important in order to maintain uniform

quality, promote safety and achieve better production control. The safety aspect is the driving force for making
cheese from pasteurized milk (Pandey, Ramaswamy, &
St-Gelais, 2003). Despite the safety concern, there is still
a large demand for cheeses made from raw milk because
they possess strong and unique avours, and therefore,
raw milk cheeses are popularly sold in many parts of the
world including Turkey. Raw milk cheese has often been
associated with outbreaks of diseases. Especially, cheeseborne salmonellosis, listeriosis and brucellosis are very
common in southeast of Turkey where Urfa cheese produc zer, Uraz, Ylmaz, & Atasoy, 2004).
tion is prevailing (O
Therefore, raw milk cheeses are suggested to be stored
under appropriate conditions at least 90 days before consumption (Kurt, 1996). In order to minimize the risk of
microbial contamination, raw milk was replaced with pasteurized milk in the production of Urfa cheese, without
impairing the overall quality of the nal product.

A.F. Atasoy et al. / Food Control 19 (2008) 278285

Heat treatment of milk prior to cheese-making is not

only an eective way of preventing the detrimental eect
of microorganisms, but also causes changes in physicochemical properties of milk compounds leading, eventually,
to variations in the quality parameters of cheese. In order to
employ appropriate heat treatment norm for cheese milk, a
control method for the upper limit of heat load is needed.
Hardly any routine method exist to measure the upper limit
for heat load to cheese milk that could assure preservation
of the important properties of milk for production of dierent cheese varieties (Ardo, Lindblad, & Qvist, 1999).
Cheese avour is the result of the breakdown of milk
components by indogeneous or exogeneous enzymes
derived from milk, rennet or microorganisms, which produce a series of volatile and non volatile compounds. The
enzymes from cheese-related microorganisms, particularly
lactic acid bacteria, are the chief factors responsible for
the formation of many such compounds that are essential
for cheese avour (Martinez-Cuesta, Fernandez de Palencia, Requena, & Pelaez, 2001). Therefore, starter bacteria
are of critical importance for production of avour compounds in cheeses made from heat treated milk.
The amount of information available about the charac zer,
teristics of industrial Urfa cheese is limited (Atasoy, O
zer, Atasoy, & Akn, 2000, 2002; O
zer,
& Turkoglu, 2003; O
zer, Robinson, &
Atasoy, Yetismeyen, & Deveci, 2004; O
Grandison, 2003). Since Urfa cheese is technologically produced from raw milk, to the best of our knowledge, no study
has, so far, been carried out on the eect of dierent heat
treatments and starter cultures on this cheese variety. Therefore, the objectives of this study were to investigate the
eects of heat treatment (at 65 C for 20 min or at 72 C
for 5 min) and starter culture combinations (Lactococcus
lactis subsp. lactis + Lactococcus lactis subsp. cremoris or
Lactobacillus delbrueckii subsp. bulgaricus + Streptococcus
thermophilus) on the basic compositions, proteolysis, lipolysis and sensory characteristics of the Urfa Cheese.
2. Materials and methods
2.1. Materials
Raw bovine milk supplied from Koc-Ata Dairy
(Sanlurfa, Turkey) was used in the manufacture of cheese
samples. Rennet of animal origin was used to coagulate
milk. The starter cultures were obtained in a freeze dried
from Peyma-Chr. Hansen (Istanbul, Turkey). Mesophilic
starter was a blend of Lc. lactis subsp. lactis and Lc. lactis
subsp. cremoris and thermophilic starter was a blend of Lb.
delbrueckii subsp. bulgaricus and Str. thermophilus in equal
proportions.
2.2. Methods
2.2.1. Cheese-making
Five Urfa cheese-making trials denoted B0, B1, B2, B3
and B4. Urfa cheeses were made from either raw or heat

279

Table 1
Urfa cheese production in the present work
Cheese
trials
B0
B1
B2
B3
B4
a
b
c
d
e

Mesophilic
culturea

Thermophilic
cultureb

Heat treated
milkc

+
+

Heat treated
milkd

+
+
+

+
+

Lc. lactis subsp. lactis and Lc. lactis subsp cremoris.

Lb. delbrueckii subsp. bulgaricus and Str. thermophilus.
At 65 C for 20 min.
At 72 C for 5 min.
+, addition , no addition.

treated milk using starter culture with three experiments

each repeated three times at dierent days. Use of cultures
and heat treatment norms are detailed in Table 1. In the
rst experiment, milk (80 kg) (control cheese) was warmed
up to 32 C. Then raw milk was coagulated with animal
rennet for 90 min and after curdling, the curd was cut into
small cubes, approximately 1 cm3, and left to rest (15 min).
The curd was drained in parzins, special moulds of triangular shape, for about 18 h at room temperature. After
whey separation was complete, the cheese blocks were
scalded in their own whey at 90 C for 3 min. The cheese
blocks were cooled down to room temperature and then
placed into plastic containers aseptically. Cheeses were
brine-salted at 4 C for 90 days. The brine concentration
was 14% (w/v) and amount of brine added was 0.75-fold
of the cheese weight.
In the second and third experiments, milk (each part
160 kg) was heated 65 C for 20 min and 72 C for 5 min
in the vat, respectively, and cooled to 32 C. After cooling,
CaCl2 was added at rate of 0.02%. Each part was divided
into two separate batches. The rst batch of each experiment was inoculated with mesophilic culture at a rate of
1.0% (w/w) and second batch of each experiment was inoculated with thermophilic culture at a rate of 0.5% (w/w).
Milk was rested at 32 C for 30 min to allow growth of
starter bacteria before renneting. The rest of the manufacturing procedure was identical to the control cheese.
2.2.2. Milk analyses
The pH was determined by means of a combined electrode pH-meter (Orion 420). Total solids, fat, titratable
acidity were determined according to TS (1981). Total
nitrogen (TN), was measured by the Kjeldahl method
according to IDF (1993). All analyses were performed in
duplicate.
2.2.3. Cheese analyses
2.2.3.1. Cheese yield. Cheese yield was calculated by using
the formula: Amount of total cheese 100/Amount of
total milk (Lucey & Kelly, 1994).
2.2.3.2. Chemical analyses. In the cheese, total solids, titratable acidity and salt were determined according to TS

280

A.F. Atasoy et al. / Food Control 19 (2008) 278285

(1989). The pH was determined by using a pH-meter

(Orion 420). The fat content was determined by the Gerber
method (TS, 1978). The total nitrogen (TN), water soluble
nitrogen (WSN) and non-protein nitrogen (NPN) were
measured by Kjeldahl method according to Gripon, Desmazeaud, Bars, and Bergere (1975). Proteose-peptone
nitrogen (PPN) was calculated by dierence of WSN from
NPN. The ripening index was calculated using the formula:
WSN 100/TN.
2.2.3.3. Electrophoretic analyses. Mini alkaline urea gel
electrophoresis technique was employed in the determination of proteolysis (Creamer, 1991). The gel concentration
was 30% (w/w) and the ratio of acrylamide/bisacrylamide
was 37.5/1. All the chemicals used were electrophoretic
grade (SIGMA ALDRICH Co. D 82039 Deisenhofen,
Germany).
2.2.3.4. Lipolysis analyses. Acid degree value was determined according to Renner (1986). Total volatile fatty
acids were determined according to Kosikowski (1978).
2.2.3.5. Sensory analyses. The sensory evaluations of the
90 day-old cheeses were carried out with three replications
by 12 trained panellist who were members of Dairy Technology Department of Food Engineering sta. The attributes were organised into color, taste (saltiness, bitterness
and characteristic after taste), aroma (characteristic
aroma), texture (texture and rmness) categories. Color
attribute was scored between 0 (white) and 7 (yellowish).
Characteristic after taste, characteristic aroma and texture
attributes were assessed on a 07 scale (0 = lowest quality,
7 = best quality). Panellists were asked to evaluate the saltiness (0 = no salty, 7 = too salty) and bitterness of cheeses
(0 = no bitter, 7 = too bitter). Firmness was assessed on a
07 scale (0 = soft, 7 = hard). Panellists were wanted to
express the cheese similar sensory characteristic with the
control cheese.
2.2.3.6. Statistical analyses. The experiment was conducted
according to completely randomized blocks arranged in a
factorial design with 2 (heat treatment) 2 (starter culture) 2 (storage period) 3 (replications). Data were subjected to analysis of variance. The signicant means were

separated using Duncans multiple range test using SPSS

software program.
3. Results and discussion
3.1. Milk compositions
Chemical compositions of raw and pasteurized milk are
illustrated in Table 2. In general, the heat treated milk had
lower titratable acidity, fat than raw milk. On contrary,
raw milk had lower pH, total solids, total nitrogen (TN).
However, eects of heat treatments on the properties of
bovine milk were found to be insignicant (P > 0.05).
3.2. Cheese yield
The yield of the cheeses B0, B1, B2, B3 and B4 were 14.1,
14.1, 14.6, 14.8 and 15.5 g cheese 100 g 1 milk, respectively.
In general, heat treatment at dierent temperatures led to a
slight increase in yield (P < 0.05). This minor increase was
thought to be due to the denaturation of whey protein and
their subsequent association with caseins. Similar results
were reported by Lucey and Kelly (1994) and Kosikowski
and Mistry (1997). The type of starter culture used in the
production of Urfa cheese had a slight inuence on the
cheese yield. Cheeses made with thermophilic starter bacteria had slightly higher yield than that of made with mesophilic starter bacteria. This dierence could be due to faster
and more extensive development of acidity by mesophilic
starter (see Table 3) during cheese-making which results
in an increase mineral losses (Lucey & Fox, 1993). On contrary to the heat treatment, eect of starter culture on the
yield was insignicant (P > 0.05).
3.3. Cheese compositions
The basic compositions of the cheese samples are shown
in Table 3. In general, the total solids content of the cheeses
declined during 90 days of storage with more remarkable in
the samples B1 and B3. This decrease may be due mainly to
breaking of peptide bonds and releasing new ionic groups
zer
(Creamer & Olson, 1982). Atasoy et al. (2003), O
zer, Atasoy, et al. (2004) reported
et al. (2000, 2002) and O
that the total solids content of Urfa cheese decreased
throughout storage as a result of extended proteolysis. At

Table 2
Chemical compositions of raw and pasteurized milka

pH
Titratable acidity (g 100 g 1 l)
Total solids (g 100 g 1)
Fat (g 100 g 1)
Total nitrogen (TN) (g 100 g 1)

BR

BP1

BP2

6.68 0.081
0.20 0.007
12.8 0.060
4.14 0.023
0.55 0.005

6.77 0.049
0.19 0.005
13.0 0.020
4.03 0.037
0.56 0.017

6.71 0.040
0.19 0.005
13.0 0.094
4.01 0.021
0.57 0.004

BR = Raw bovine milk; BP1 = Pasteurized bovine milk at 65 C for 20 min; BP2 = Pasteurized bovine milk at 72 C for 5 min.
a
Arithmetic mean of three replicates.

A.F. Atasoy et al. / Food Control 19 (2008) 278285

281

Table 3
The basic compositions of the cheese samplesxa
Cheese samples
b

pH
T.Ac (g 100 g

la)

Total solids (g 100 g 1)

Fat-in-dry matter (g 100 g 1)
Salt-in-dry matter (g 100 g 1)
Total nitrogen (g 100 g 1)

Level of signicance

SD

B0

B1

B2

B3

B4

HT

SC

1
90
1
90
1
90
1
90
1
90
1
90

6.11 0.10b
5.75 0.05c
0.35 0.08a
0.66 0.09a
44.3 0.46ab
43.1 1.32
46.9 0.08
48.1 0.10
8.7 0.20
11.4 0.34a
2.67 0.06
2.49 0.09

4.89 0.09a
4.91 0.01a
0.74 0.15b
0.99 0.06bc
45.9 0.86ab
44.0 0.97
48.9 1.03
47.0 1.84
8.1 0.48
13.7 0.97b
2.79 0.17
2.48 0.08

5.93 0.17b
5.57 0.09bc
0.28 0.03a
0.71 0.15ab
44.9 0.61ab
43.7 0.35
47.1 0.82
46.4 0.15
9.5 0.15
12.7 0.23ab
2.69 0.12
2.57 0.13

5.06 0.11a
4.99 0.08a
0.76 0.05b
0.99 0.14c
46.3 0.97 b
42.0 1.04
49.2 1.56
46.7 0.88
8.8 0.71
13.9 0.63b
2.72 0.11
2.33 0.15

6.10 0.11b
5.48 0.10b
0.30 0.05a
0.75 0.01abc
43.5 0.71a
43.0 0.79
46.7 0.19
46.8 0.76
9.3 0.73
12.2 0.36a
2.56 0.11
2.41 0.08

***

***

***

***

NS
NS
NS
NS
NS
NS
NS

**
*
*

NS
NS
NS
NS

NS
NS

NS
NS

HT = Heat treatment; SC = Starter culture; NS = Non-signicant.

a
Arithmetic mean of three replicates.
b
Storage days.
c
Titratable acidity.
x
Dierent letters indicate statistical dierences within the samples at the same storage day.
*
Signicant at P < 0.05.
**
Signicant at P < 0.01.
***
Signicant at P < 0.001.

the early stages of storage, the eect of starter culture on

the total solids was signicant (P < 0.05). The pHs of
cheeses decreased during storage except the sample B1.
The eect of heat treatment and starter culture on pH were
signicant (P < 0.001). In parallel to the variation in pH
values, the titratable acidity values of the samples increased
until 30-day of ripening, then uctuated during the rest of
storage (data not shown). The increase in titratable acidity
during the rst month of cheese in brine is due mainly to
almost completion of the lactose fermentation and the liberation of amino and free fatty acids following proteolysis
and lipolysis. The decreases in titratable acidities after the
60th day of storage could be due to catabolism of lactic
acid and passing into brine and the production of ammonia
by deamination of free amino acids (FAA) (Azarnia,
Ehsani, & Mirhadi, 1997; Grappin & Beuvier, 1997; Polychroniadou, 1994; Prieto, Urdiales, Franco, Fresno, &
Carballo, 2000). While the eect of heat treatment on
titratable acidity was insignicant (P > 0.05), the type of
starter culture had signicantly aected the development
of acidity (P < 0.05). At the beginning of the storage, the
fat-in-dry matter contents of the samples were between
46.7 g 100 g 1 and 49.2 g 100 g 1 and these gures did
not change signicantly at the end of storage (P > 0.05).
Salt-in-dry matter contents of the cheeses increased up
to 30-day of ripening, then remained almost unchanged.
Regardless of the treatments, the salt (NaCl) penetration
into the cheese was much faster during early stage of storage. Salt is driven into cheese by concentration gradient
between cheese blocks and brine, and this gradient is much
larger at the beginning of ripening (Azarnia et al., 1997).
The salt contents of the cheeses were lower than the values
zer
reported in previous studies (Atasoy et al., 2003; O
zer, Atasoy, et al., 2004). The control
et al., 2000, 2002; O

cheese (raw milk cheese) had lower salt-in-dry matter level

and at the same heat treatment, cheese made with mesophilic culture had higher salt content then its thermophilic
counterpart. The share eect of heating norm and starter
culture on salt-in-dry matter were signicant (P < 0.05).
3.4. Proteolysis of cheeses
As seen in Table 3, the total nitrogen (TN) levels of Urfa
cheeses decreased during storage irrespective of treatments.
Due to the proteolysis, the amount of nitrogen decreases
during ripening, releasing amino acids that are transformed
into volatile compounds. The continuous decrease in TN
content of Urfa cheese throughout cold storage was
zer et al., 2000,
reported earlier (Atasoy et al., 2003; O
zer, Atasoy, et al., 2004). The eect of heat treat2002; O
ment and starter culture on total nitrogen content were
insignicant (P > 0.05).
In order to better understand development of proteolysis in cheese, it is necessary to investigate the nitrogen fractions formed during storage (Law, 1987). Variations in the
nitrogen fractions and ripening indices of Urfa cheese samples are illustrated in Table 4. Cheeses manufactured with
mesophilic culture (B1 and B3) contained more water
soluble nitrogen (WSN) than cheeses manufactured with
thermophilic culture (B2 and B4). The eect of starter culture on WSN level of the cheeses was signicant, with more
remarkable at the later stages of ripening (P < 0.001). The
cheeses inoculated with mesophilic culture and heated at
65 C had higher WSN values than the sample produced
with the same culture but heated at higher temperature
(72 C). Similar trend was noted for the cheeses manufactured with thermophilic culture as well. Similar ndings
were reported by Benfeldt and Sorensen (2001). Eect of

282

A.F. Atasoy et al. / Food Control 19 (2008) 278285

Table 4
Variations of nitrogen fractions and ripening indices of cheese samplesxa
Cheese samples
b

WSN (g 100 g 1)
NPN (g 100 g 1)
PPN (g 100 g 1)
RI (%)

Level of signicance

SD

B0

B1

B2

B3

B4

HT

SC

1
90
1
90
1
90
1
90

0.29 0.05ab
0.46 0.05a
0.05 0.01a
0.20 0.02a
0.25 0.05
0.26 0.03
10.8 1.47abc
18.5 1.37b

0.38 0.01c
0.63 0.03b
0.19 0.05b
0.39 0.02b
0.18 0.04
0.24 0.04
13.5 0.43c
25.2 0.72c

0.26 0.02ab
0.44 0.01a
0.05 0.01a
0.21 0.03a
0.20 0.02
0.22 0.04
9.6 1.09ab
17.1 0.37ab

0.34 0.03bc
0.57 0.01b
0.14 0.00b
0.37 0.01b
0.19 0.02
0.20 0.01
12.3 0.72bc
24.6 1.06c

0.22 0.01a
0.40 0.02a
0.04 0.01a
0.19 0.03a
0.17 0.01
0.20 0.03
8.4 0.46a
15.1 0.37a

NS
NS

**

**

***

**

***

NS
NS
NS
NS

NS
NS
*
***

WSN = Water soluble nitrogen; NPN = Non-protein nitrogen, PPN = Proteose-peptone nitrogen; RI = Ripening index; HT = Heat treatment;
SC = Starter culture; NS = Non-signicant.
a
Arithmetic mean of three replicates.
b
Storage days.
x
Dierent letters indicate statistical dierences within the samples at the same storage day.
*
Signicant at P < 0.05.
**
Signicant at P < 0.01.
***
Signicant at P < 0.001.

heat treatment on water soluble nitrogen content of Urfa

cheeses was insignicant (P > 0.05). The non-protein nitrogen (NPN) levels of the Urfa cheeses increased with ripening. Increasing NPN level in bovine milk Urfa cheese
zer
during ripening were reported (Atasoy et al., 2003; O
zer, Atasoy, et al., 2004). As with
et al., 2000, 2002; O
WSN, NPN levels of the cheeses were signicantly aected
by the starter cultures (P < 0.001). The eect of heat treatment on NPN content of Urfa cheeses was found to be signicant (P < 0.01). The joint eect of starter culture and
heat treatment on NPN content was more pronounced during ripening.
The proteose-peptone nitrogen (PPN) fraction is made
up of large peptides. The initial and nal PPN fractions
are higher than NPN for B0, B2 and B4 cheese. This allows
us to conclude that rennet is the main agent of the proteolysis in the cheeses in question and that the microbial
enzymes have a secondary role in the extended degradation
of peptides into small molecular weight peptides and amino
acids these cheeses. This nding is line with Prieto et al.
(2000) and Franco et al. (2001). The eect of heat treatment
and starter culture on PPN content of Urfa cheeses were
insignicant (P > 0.05).
The process of proteolysis is aected by various factors,
including natural ora of raw milk, starter culture, milk
coagulating enzyme and production parameters. In this
study, the raw milk of the same origin was used for the
cheese manufacturing and the production parameters were
constant for each individual cheese batches, except type of
starter culture and heat treatment norms. The eect of
heating norm on ripening index (RI) of cheeses was insignicant (P > 0.05), on contrary, the starter culture aected
RI values of cheeses signicantly (P < 0.001). At the beginning of storage RI of the samples B0, B1, B2, B3, B4 were
10.8%, 13.5%, 9.6%, 12.3%, 8.4%, respectively. After 90
days of ripening period, these values increased to 18.5%,
25.2%, 17.1%, 24.6% and 15.1%, in the same order.

According to a classication of white cheeses on the basis

of RI values, as proposed by Kurt (1972), B0, B2 and B4
cheeses are classied as unripened (RI < 20%), B1 and B3
cheese are semi-ripened (RI between 20% and 30%).
3.5. Electrophoresis of cheese
The alkaline urea gel electrophoretograms of 1 day and
90 day-old cheeses are illustrated in Fig. 1. The density of
bands representing as1-casein (as1-CN) and b-casein (bCN) decreased throughout ripening and as1-CN degradation products became more evident, except the samples
B0. A more extensive hydrolysis of b-CN was noted in
B0. No band representing as1-casein degradation products
was observed. In general, degradation of casein fractions
in Urfa cheeses was limited throughout storage, indicating
lower degree of proteolysis in the cheeses. This could be
due mainly to high salt/moisture ratio (between 8.6 g
100 g 1 and 10.7 g 100 g 1 in 90 day-old cheeses). Thomas
and Pearce (1981) found that at 4% salt/moisture ratio,
as1-CN was degraded almost completely in Cheddar
cheese, however, increasing the ratio of salt/moisture to
8% in the same cheeses led to a sharp decrease in degradation of this casein fraction. Mulvihill and Fox (1978) and
Thomas and Pearce (1981), showed that the rennet action
on b-CN was also inuenced by the NaCl concentration.
Eects of mesophilic and thermophilic lactic acid bacteria
on the caseins were very limited. These ndings are in good
agreement with previous studies (Farkye, Fox, Fitzgerald,
& Daly, 1990; Law, Fitzgerald, Uniacke-Lowe, Daly, &
Fox, 1993; Oberg, Davis, Richardson, & Ernstrom, 1986).
3.6. Lipolysis of cheeses
Acid degree value and total volatile fatty acids contents
of the experimental cheeses are shown in Table 5. At the
beginning of ripening, the acid degree values of the cheeses

A.F. Atasoy et al. / Food Control 19 (2008) 278285

283

Cheese samples
B0
1

B1
90

90

B2
1

B3
90

B4
90

90

-casein
-casein

s1-casein
s1-casein
degradation
product

Fig. 1. Alkaline urea gel electrophoretograms for the caseins after 1 and 90-days of ripening in the Urfa cheese samples.

B0, B1, B2, B3 and B4 were 2.32, 1.23, 1.07, 1.25, 1.08 mg
KOH 100 g 1 fat, respectively, and these gures increased
to 6.39, 1.81, 1.58, 1.64, 1.60 mg KOH 100 g 1 fat after
90 days of storage, in the same order. The total volatile
fatty acids values of the cheeses were 8.25 (B0), 5.53 (B1),
3.84 (B2), 4.59 (B3), 3.36 (B4) mL NaOH 100 g 1 cheese
in 1 day-old cheeses, and these gures rose to 14.7, 6.84,
5.92, 6.37, 6.12 mL NaOH 100 g 1 cheese in the 90 dayold cheeses, respectively.
B0 (raw milk cheese) had higher acid degree value and
total volatile fatty acids degree levels than B1, B2, B3 and
B4. Similar results were found by McSweeney, Fox, Lucey,
Jordan, and Cogan (1993) and Rehman et al. (2000). This
indicated that native milk lipase was principally responsible
for the hydrolysis of the lipids in bovine Urfa cheese. The
hydrolyzation of lipids in cheese during ripening is catalyzed by indigeneous lipase of the milk and by microbial
lipases (Fox, Guinee, Cogan, & McSweeney, 2000; Franco,

Prieto, Urdiales, Fresno, & Carballo, 2001). The heat treatment of milk completely inactivate the native milk lipase.
Moreover, native milk lipase is optimally active at pH
value of 8.09.0, and is inhibited by NaCl to great extent
(Franco et al., 2001; Vlaemynck, 1992). The gures of salt
content and pH in the experimental cheeses were far from
the the values indicated for optimum activity of milk lipases (pH 5.754.91, salt content 4.916.01%). Eect of
heating norm was signicant on acid degree values and
total volatile fatty acid values (P < 0.05 and P < 0.001,
respectively).
Acid degree value and total volatile fatty acid contents
of the cheeses made with mesophilic (B1 and B3) and thermophilic (B2 and B4) lactic cultures were close to each
other. The lipolytic capacities of Lactobacillus spp. and
Lactococcus spp. are very limited (Fox et al., 2000). Eect
of starter culture on acid degree value and total volatile
fatty acids was insignicant (P > 0.05).

Table 5
Variations of acid degree value and total volatile fatty acids of cheese samplesxa
Cheese samples
b

Acid degree valuec

TVFAd

Level of signicance

SD

B0

B1

B2

B3

B4

HT

SC

1
90
1
90

2.32 0.64b
6.39 2.78b
8.25 0.64d
14.7 2.45b

1.23 0.02a
1.81 0.11a
5.53 0.35c
6.84 0.65a

1.07 0.05a
1.58 0.13a
3.84 0.22ab
5.92 0.22a

1.25 0.05a
1.64 0.05a
4.59 0.08bc
6.37 0.36a

1.08 0.07a
1.60 0.19a
3.36 0.31a
6.12 0.18a

*
*

NS
NS

***

***

NS

HT = Heat treatment; SC = Starter culture; NS = Non-signicant.

a
Arithmetic mean of three replicates.
b
Storage days.
c
mg KOH 100 g 1 fat.
d
Total volatile fatty acids (mL NaOH 100 g 1 cheese).
x
Dierent letters indicate statistical dierences within the samples at the same storage day.
*
Signicant at P < 0.05.
***
Signicant at P < 0.001.

284

A.F. Atasoy et al. / Food Control 19 (2008) 278285

B0, B1, B2, B3, B4

CHARACTERISTIC AROMA
7

BITTERNESS 7

7 CHARACTERISTIC AFTER-TASTE

0
SALTINESS 7

7 COLOR

FIRMNESS 7

7 TEXTURE

Fig. 2. Sensory prole of 90 day-old Urfa cheese samples.

As the control cheese had higher level of lipolysis than

the cheeses made from milk inoculated with mesophilic
or thermophilic lactic starters, it can be concluded that
native lipases and/or non starter lactic acid bacteria
(NSLAB) were primarily responsible for the devolopment
of lipolysis in Urfa cheese.

lists pointed out that B1 and B3 cheeses were soft and adhesive but the acidity in these samples was well-balanced.
Majority of the panel group expressed that B2 cheese had
more or less similar sensory characteristic with the control
cheese (B0).
4. Conclusions

3.7. Sensory evaluation of cheese

Sensory characteristics of 90 day-old Urfa cheeses are
given in Fig. 2. All experimental cheeses were whitish in
color. Cheeses manufactured from milk heated at lower
temperature (65 C) had higher textural characteristics
than the cheeses made from high heat treated (72 C) milk.
Moreover, cheeses manufactured with mesophilic culture
had lower sensory scores for textural characteristic than
that made with thermophilic culture. This result was in harmony with higher protein degradation in the cheese made
with mesophilic culture. According to Tunick et al.
(1993), Tunick, Malin, Smith, and Holsinger (1995),
Tunick, Cooke, Malin, Smith, and Holsinger (1997), there
is a negative correlation between proteolysis and development of texture in cheese. Raw milk cheese (B0) received
higher overall aroma and pungency and avour scores
from the panel group.
Bitterness is one of the most common defects associated
with an excessive casein breakdown, leading, eventually, to
an accumulation of high-molecular weight bitter peptides.
This defect was not detected in the 90 day-old Urfa cheese
and experimental cheeses had almost similar bitterness
scores. High saltiness was the main point that was criticized
by the panellists; although, the fact that traditional Urfa
cheese is a too salty cheese variety (Atasoy, 1999;
Yetismeyen & Yldz, 2001). While 40% of the panellists
found the samples B2 and B4 to be hard and well textured
but poor in acidity. On the other hand, 55% of the panel-

This work clearly demonstrated the role of heat treatment to cheese milk and starter culture on some characteristics of Urfa cheese. Traditional raw milk cheese had
higher sensory scores, but since traditional applications
of scalding and brining were insucient to elimination all
zer, Uraz, et al.,
pathogens from industrial Urfa cheese (O
2004), incorporation of heat treatment into Urfa cheesemaking is inevitable. In the present work, heat treatment
at higher temperature (72 C) adversely aected the chemical and sensory properties of Urfa cheeses. The adverse
eects of heat treatment at lower temperature (65 C) on
properties of the cheeses were less pronounced. Therefore,
it can be concluded that milk used in the production of
Urfa cheese should be heat treated at lower temperature.
While cheeses made by the addition mixed culture consisting of Lc. lactis subsp. lactis + Lc. lactis subsp. cremoris
had higher levels of proteolysis, these samples received
lower preference by the panellists. On contrary, the cheeses
produced with addition of starter culture containing Lactobacillus delbrueckii subsp. bulgaricus + Streptococcus thermophilus had lower levels of proteolysis but higher
sensory preferences.
Future studies should be intensied on the selection of
suitable starter combination(s) to manufacture Urfa cheese
having close physical, chemical and sensory characteristics
to the traditional ones. Also, further studies should be dedicated to lower the salt level of Urfa cheese without imparing the nature of the end products.

A.F. Atasoy et al. / Food Control 19 (2008) 278285

References
Ardo, Y., Lindblad, O., & Qvist, K. B. (1999). Study of methods to
routinely monitor heat load to cheese milk. International Dairy
Journal, 9, 547552.
Atasoy, A. F. (1999). A study on the determination of level of proteolysis in
Urfa cheese sold in local markets. MSc Thesis. Harran University,
Institute of Natural and Applied Sciences, Sanliurfa.
zer, B., & Turkoglu, H. (2003). Ripening and textural
Atasoy, F., O
properties of UF and traditional Urfa cheese produced from raw and
pasteurised bovine milk. 3th GAP Agriculture Congress, Sanlurfa,
Turkey, October 23, 2003. Harran University.
Azarnia, S., Ehsani, M. R., & Mirhadi, S. A. (1997). Evaluation of the
physico-chemical characteristics of the curd during the ripening of
Iranian brine cheese. International Dairy Journal, 7, 473478.
Benfeldt, C., & Sorensen, J. (2001). Heat treatment of cheese milk: Eect
on proteolysis during cheese ripening. International Dairy Journal, 11,
567574.
Creamer, L. K., & Olson, N. F. (1982). Rheological evaluation of
maturing Cheddar cheese. Journal Food Science, 47, 631638.
Creamer, L. K. (1991). Electrophoresis of cheese. Bulletin of IDF 261.
Brussels, Belgium: International Dairy Fedaration (pp.1428).
Farkye, N. Y., Fox, P. F., Fitzgerald, G. F., & Daly, C. (1990). Proteolysis
and avor development in Cheddar cheese made exclusively with single
strain proteinase-positive and proteinase-negative starters. Journal of
Dairy Science, 73, 874880.
Fox, P. F., Guinee, T. P., Cogan, T. M., & McSweeney, P. L. H. (2000).
Fundamentals of cheese science. USA: Walters Kluwer Company.
Franco, I., Prieto, B., Urdiales, R., Fresno, J. M., & Carballo, J. (2001).
Study of the biochemical changes during ripening of Ahumado de
Aliva cheese: A Spanish traditional variety. Food Chemistry, 74,
463469.
Grappin, R., & Beuvier, E. (1997). Possible implications of milk
pasteurization on the manufacture and sensory quality of ripened
cheese. International Dairy Journal, 7, 751761.
Gripon, J. C., Desmazeaud, M. J., Bars, D., & Bergere, J. L. (1975). Etude
du roles des micro-organismes et des enzymes au cours de la
maturation des fromages. Le Lait, 55, 502516.
IDF (1993). Determination of nitrogen content. International IDF
Standard, 20B:1993, Brussels, Belgium: International Dairy
Fedaration.
Kosikowski, F. V. (1978). Cheese and fermented milk foods (2nd ed.). New
York USA.
Kosikowski, F. V., & Mistry, V. V. (1997). The yield of cheese. In
Kosikowski, F. V., and Mistry V. V., (Eds.), Cheese and fermented milk
foods (pp. 622632). Vol. 1. Origins and principles third ed. Westport
CT, USA.
Kurt, A. (1972). Standard methods for milk and dairy products. Publication
No: 252. Erzurum, Turkey: Ataturk University Press.
Kurt, A. (1996). Milk technology. Publication No: 257. Erzurum, Turkey:
Ataturk University Press.
Law, B. A. (1987). Proteolysis in relation to normal and accelerated cheese
ripening. In P. F. Fox (Ed.), Cheese: Chemistry physics and microbiology. Vol. 1: General aspects (pp. 365392). London: Elsevier Applied
Science Publisher Ltd.
Law, J., Fitzgerald, G. F., Uniacke-Lowe, T., Daly, C., & Fox, P. F.
(1993). The contribution of Lactococcal starter proteinases to proteolysis in Cheddar cheese. Journal of Dairy Science, 76, 24552467.
Lucey, J., & Kelly, J. (1994). Cheese yield. Journal of the Society of Dairy
Technology, 47(1), 114.
Lucey, J. A., & Fox, P. F. (1993). Importance of calcium and phosphate
cheese manufacture: a review. Journal of Dairy Science, 76, 1714
1724.
Martinez-Cuesta, M. C., Fernandez de Palencia, P., Requena, T., &
Pelaez, C. (2001). Enzymatic ability of Lactobacillus casei subsp. casei
IFPL731 for avour development in cheese. International Dairy
Journal, 11, 577585.

285

McSweeney, P. L. H., Fox, P. F., Lucey, J. A., Jordan, K. N., & Cogan,
T. M. (1993). Contribution of the indigeneous microora to the
maturation of Cheddar cheese. International Dairy Journal, 3, 613614.
Mulvihill, D. M., & Fox, P. F. (1978). Proteolysis of b-casein by chymosin:
inuence of pH, urea and NaCl. Irish Journal of Food Science and
Technology, 2, 135143.
Oberg, C. J., Davis, L. H., Richardson, G. H., & Ernstrom, C. A. (1986).
Manufacturing of Cheddar cheese using proteinase-negative mutants
of Streptococcus cremoris. Journal of Dairy Science, 69, 29752981.
zer, B., Atasoy, F., & Akn, S. (2002). Some properties of Urfa cheese (a
O
traditional white-brined Turkish cheese) produced from bovine and
ovine milks. International Journal of Dairy Technology, 55(2), 9499.
zer, B. H., Atasoy, A. F., & Akn, M. S. (2000). Eect of pasteurisation
O
and post-manufacture heat treatment on the microbiological and
chemical properties of traditional Urfa cheese. 6th Milk and Milk
products Symposium. Tekirdag, Turkey, May 2226, 2000. Trakya
University.
zer, B. H., Atasoy, A. F., Yetismeyen, A., & Deveci, O. (2004).
O
Development of proteolysis in ultraltered Turkish white-brined
cheese (Urfa type)-eect of brine concentration. Milchwissenschaft,
59(3/4), 146149.
zer, B. H., Robinson, R. K., & Grandison, A. S. (2003). Textural and
O
microstructural properties of urfa cheese (a white-brined Turkish
cheese). International Journal of Dairy Technology, 56, 171176.
zer, H. B., Uraz, G., Ylmaz, E. B., & Atasoy, A. F. (2004). The eects of
O
brine concentration and scalding on survival of some pathogens in
Urfa cheese:a traditional white-brined Turkish cheese. International
Journal of Food Science and Technology, 39, 727735.
Pandey, P. K., Ramaswamy, H. S., & St-Gelais, D. (2003). Evaluation of
pH change kinetics during various stages of Cheddar cheese-making
from raw, pasteurized, micro-ltered and high-pressure-treated milk.
Lebensmittel-Wissenchaft und-Technologie, 36, 497506.
Polychroniadou, A. (1994). Objectives indices of maturity of Feta and
Teleme cheese. Milchwissenchaft, 49(7), 376379.
Prieto, B., Urdiales, R., Franco, I., Fresno, J. M., & Carballo, J. (2000).
Quesucos de Liebana cheese from cows milk:biochemical changes
during ripening. Food Chemistry, 70, 227233.
Rehman, Shakeel-Ur., McSweeney, P. L. H., Banks, J. M., Brechany, E.
Y., Muir, D. D., & Fox, P. F. (2000). Ripening of cheddar cheese made
from blends of raw and pasteurised milk. International Dairy Journal,
10, 3344.
Renner, E. (1986). Milchpraktikum. Fachgebiet Milchwissenchaft, JusticLiebig-Universitat Giessen.
Thomas, T. D., & Pearce, K. N. (1981). Inuence of salt on lactose
fermentation and proteolysis in Cheddar cheese. New Zealand Journal
of Dairy Science and Technology, 16, 253259.
TS (1978). Determination of fat in cheese. TS 3046. Ankara, Turkey:
Turkish Standards Institute.
TS (1981). Standards of raw milk. TS 1018. Ankara, Turkey: Turkish
Standards Institute.
TS (1989). White cheese standard. TS 591. Ankara, Turkey: Turkish
Standards Institute.
Tunick, M. H., Cooke, P. H., Malin, E. L., Smith, P. W., & Holsinger, V.
H. (1997). Reorganization of casein submiceles in Mozzarella cheese
during storage. International Dairy Journal, 7, 149155.
Tunick, M. H., Mackey, K. L., Shieh, J. J., Smith, P. W., Cooke, P. H., &
Malin, E. L. (1993). Rheology and microstructure of low-fat Mozzarella cheese. International Dairy Journal, 3, 649662.
Tunick, M. H., Malin, E. L., Smith, P. W., & Holsinger, V. H. (1995).
Eects of skim milk homogenization on proteolysis and rheology of
Mozzarella cheese. International Dairy Journal, 4, 191483.
Vlaemynck, G. (1992). Study of lipolytic activity of the lipoprotein lipase
in lunch cheese of the Gouda type. Milchwissenchaft, 47, 164166.
Yetismeyen, A., & Yldz, F. (2001). Determination of chemical, microbiological and sensory properties of Urfa cheese sold in Ankara
market. 2nd GAP Agriculture Congress, Sanlyurfa, Turkey, October
2426, 2001. Harran University.