The samples of raw sheep milk analyzed were collected from two dairy breeds, Ouled-Djellal and Rembi, located in the region of Djelfa (300 km south of the capital Algiers). For each breed, three individual milk samples were taken for three months (January, February and March) and three times a month.

The milk was heated and stirred gently for 10 minutes in a water bath at 30-35 °C to allow the rise of the fat surface, then it was skimmed by centrifugation at 3500 x g for 20 min at 4 °C then filtered through glass wool. The operation was repeated 2 to 3 times.

From the skimmed milk, the caseins were separated from the serum proteins by precipitation at their isoelectric point (pH 4.6) by adding dropwise 4N hydrochloric acid, followed by centrifugation at 4000 x g for 20 min at 25 °C.

The pellet containing the caseins was recovered in a minimal volume of distilled water. This operation was repeated 3 times, in the same way as the supernatant containing the serum proteins, in order to eliminate any trace of contamination after adjusting the pH to 7 by addition of 1N sodium hydroxide and acidification and centrifugation in the same forms. The two major groups of proteins obtained (caseins and whey proteins) were dialyzed (cut-off membrane equal to 10,000 Dalton) for 48 hours at 4 °C. against distilled water, renewed twice a day.

The electrophoretic migration of the proteins obtained (total caseins and whey proteins) was carried out under native conditions (native PAGE), or in the presence of dissociating and/or denaturing agents (urea-PAGE or SDS-PAGE).

By applying protocols developed on bovine milk (Laemmli and Favre, 1973; Darling and Butcher, 1975), the electrophoretic migration of ewe’s milk proteins has been monitored and the methods were each time optimized by modifying certain electrophoretic parameters (gel porosity, migration time, amperage, voltage, coloring conditions/discoloration) to have resolving and discriminating profiles.

The electrophoresis was conducted on a system of vertical mini-tanks 10x10 and 10x8cm (Hoefer SE 200) in constant voltage and amperage. After migration, the proteins were fixed with 12% tricholoracetic acid, stained with Coomassie blue and decolorized using water/ methanol/acetic acid mixture.

This electrophoresis was performed according to the method of Hillier (1976) with a polyacrylamide gel (T =12%) in 0.75M Tris-HCl buffer, pH 8.9. Samples (2 mg mL^-1) were solubilized in 75mM Tris-HCl buffer, pH8.9, containing 10% (v/v) glycerol, and 0.01% (w/v) bromophenol blue.

In this conditions, the method described by Laemmli and Favre (1973) was used with a stacking gel (T = 4%, C = 2.7%) in Tris-HCl buffer, pH 6.8 and a separating gel (T = 17%, C = 2.7%) in Tris-HCl buffer, pH8.8.

In order to calibrate the gel, a pre-colored kit of known molecular weight (MW) proteins was used. It was composed by the 7 following entities: a2-Macroglobulin (180,000 Da); P-Galactosidase (116,000 Da); Lactoferrin (90,000 Da); Pyruvate kinase (58,000 Da); Fumarase (48,500 Da); Lactic dehydrogenase (36,500 Da); Triose-Phosphate Isomerase (26,600 Da). After proteins marker migration a standard curve Log (MW) versus migration distance was traced than the MW of the unknown proteins were calculated by resolving generated equation y =6.6015x + 33.438 ( y: Log (MW) ; x: migration distance).

In native conditions, caseins, because of their micellar structure, were difficult to separate. In this case, dissociating agents such as urea and an S-S bridge reducing agent (P-mercaptoethanol) were used.

The method used was that described by Shalabi and Fox (1987) with a stacking gel (T = 4.8%, C = 2.7%) containing 5.7 mol/l urea and a separating gel T = 13%; C = 4.15%) containing the same concentration of urea. The gel buffers were identical to those of the SDS-PAGE and the migrating buffer was similar to the native-PAGE buffer.

In native PAGE, the mobility of protein fractions depends both on their charge and their PM. According to Lin et al. (2010), this method was well-adapted for the separation of serum proteins.

An examination of the electrophoretic diagrams (Figure 1) allowed to drawing two remarks. The first was that the separation profiles obtained show a great deal of similarity and homogeneity between the different samples of the raw sheep milk analyzed. The second showed that some bands were characterized by migration levels similar to those of cow’s milk proteins. The latter migrate into five distinct bands that can be characterized by their electrophoretic mobility as being: Ig, PP, BSA, a-La, and P-Lg, based on the bibliographic data of Egito et al. (2001).

Figure 1

The electrophoretic profile obtained for the raw sheep milk analyzed shows the existence of six bands (1 to 6) of which three well focused (4, 5 and 6). Among these, bands 4 and 6 appeared with an identical migration levels to those of BSA and a-La, respectively. While band 5 was between these two proteins. Furthemore, as reported by Pesic et al. (2011b), whey proteins from ovine milk followed the increasing order of electrophoretic mobility: SA, a-La and P-Lg and the migration level of the last two proteins was lower than that found in bovine milk. This leaded us to hypothesize that the bands 4, 5 and 6 might correspond respectively to SA, a-La and P-Lg. However, confirmation requires the isolation of each fraction and the sequencing of their N-terminal part.

Although ovine P-Lg showed two major variants P-LgA and P-LgB (Amigo et al., 1992; Mayer, 2005; Pesic et al., 2011a and b), these appear only in the form of a single intense band whose electrophoretic mobility was similar to that of the bovine a-La. This observation is in agreement with the results obtained by the latter authors. The highlighting of these two bands may be possible using other techniques such as isoelectrofocusing (Amigo et al., 1992; Moatsou et al., 2005), capillary electrophoresis (Recio et al., 1997), chromatofocusing (Fernandez-Espla, Lopez-Galvez, and Ramos, 1993). With the highest electrophoretic mobility compared to all other proteins, bovine P-Lg was targeted on PAGE-native to detect a possible adulteration of ovine with bovine milk (Pesic et al., 2011a).

Due to their micellar structures and interactions of their groups, electrophoretic separations of caseins require the utilization of agents dissociating hydrogen bonds (such as urea) and S-S bridge reducers (such as 2-mercaptoethanol). In bovine milk, four well separated casein bands were distinguished in the profiles of urea-PAGE. Starting from the deposition and based on their electrophoretic mobility (Pesic et al., 2011a), these bands corresponded to y-CN, P-CN, aS₂- CN and aS₁-CN.

Regardless of breed and sampling period, sheep caseins migrate in five bands of high intensity divided into two distinct zones: zone 1 with two bands and zone 2 with 3 bands (Figure 2). The first zone was located at the boundaries of y-CN and P-CN bovines while the second zone was characterized by a lower electrophoretic mobility than aS bovine caseins. This same behavior was observed by Dall’Olio, Davoli and Russo (1990) and Moatsou et al. (2004). Pesic et al. (2011a) report that k-CN and P-CN were characterized by the same electrophoretic mobility in the milk of the three species (sheep, goat and cattle) and the ovine aS-CN showed a lower migration with greater bands intensity than caprine and bovine milks. The migration of sheep caseins investigated by Dall’Olio et al. (1990) in the decreasing order of mobility was: k-CN, P-CN+k-CN, aS₁-CN and aS₂-CN. Whereas for Trujillo, Casals and Guamis (2000), the order was of type k-CN, as₂-CN, as₁-CN and P-CN. While, Recio et al. (1997) and Clement, Agboola and Bencini (2006), were referred to the following migration order of ovine caseins: aS₂-CN, aS₁-CN, k-CN and P-CN.

Figure 2.

According to the electrophoretic mobility in urea-PAGE, Dall’Olio et al. (1990) and Moatsou et al. (2004) notes that sheep caseins aS and P were migrated in two distinct groups: P-CNs (with two bands called P₁ and P₂ caseins) and aS-CNs (with three bands named aS₁, aS₂ and aS₃-CN according to their electrophoretic mobility). Mayer (2005) notes that bovine aS1-CN can be used as a marker to identify the presence of cow milk in sheep or goat milk.

Under this denaturing and dissociating conditions with sodium dodecyl sulfate as detergent, the ovine whey proteins of our samples were homogeneous for the six visualized bands (indicated from 1 to 6) (Figure 3). Three bands (2, 6 and 7) were intense with a migration levels similar to those observed for bovine SA, P-Lg and a-La. Bands 1 and 3 were characterized by the lower electrophoretic mobility which can corresponded to sheep lactoferrin and/or immunoglobulins belonging to two different classes based on calculated and theoretical PM and referring to bibliographic data. Bands 4 and 5 corresponded to caseins (low casein contamination).

Figure 3

The profiles obtained for serum proteins in SDS-PAGE were consistent with those obtained by Pelmus et al. (2012) by examining the protein polymorphism of sheep milk from the local Romanian breed. These authors also note that sheep proteins showed a low electrophoretic mobility, both for serum proteins and for caseins, compared with bovine proteins.

For caseins, four bands were observed in the electropherogram obtained for the different samples of raw sheep milk collected in the region of Djelfa. These bands, noted 1 to 4 according to their increasing electrophoretic mobility, were well focused but with varying intensities (Figure 4). Vairo Cavali et al. (2008) report that these bands corresponded respectively to aS2-CN, aS1-CN, P-CN and k-CN. Nevertheless, the presence of k-CN was difficult to identify (Calavia & Burgos, 1998).

Figure 4

The measured molecular weights (Table 1 and 2) of ovine proteins of milk samples collected from breeds Ouled-Djellal and Rembi corresponded globally to those reported by different authors. but not corresponded to those reported by Ameur Ameur et al. (2016). Similarly, any major differences in the MW of proteins between the samples of these two breeds were detected. However, the slight variations detected might be due to genetic differences in the dairy females considered.

Table 1

*Ig: immunoglobulin , SA: serum -albumin, PP: proteose-peptone, Ig:

Table 2

CN : casein.