Lyophilized bacterial strains were activated in 10 mL of de Man, Rogosa, Sharpe (MRS) broth at 37˚C for 24 hr. Then, 1 mL of each culture was added to 100 mL of MRS broth and incubated at 37˚C for 24 hr. One mL of each culture was inoculated into 100 mL of autoclaved reconstituted skimmed milk powder (11%) and incubated at 37˚C overnight and used for preparation of fermented milks.

Camel milk (total solids 12.37%, protein 3.21%, fat 4.1%, pH 6.54, and acidity 0.170%) was divided into six parts. Modified starch (0.5%; w/v) was added to each part of milk before heat treatment. Fermented milk was prepared according to Tamime and Robinson [24]. All milks were heated separately at 90˚C for 10 min and rapidly cooled to 42˚C. The first part was inoculation with 0.04% (w/v) of commercial starter culture alone (as control). Each of the other five parts of milk was inoculated with 3% (v/v) of single strain of Lactobacillus (Lb. helveticus B-734, Lb. casei subsp. casei B-1922, Lb. paracasei subsp. paracasei B-4560, Lb. rhamnosus B-1445 and Lb. rhamnosus B-442). Fermentation was carried out at 40˚C until the pH reached 4.69 with samples containing the commercial starter culture (control), Lb. casei, and Lb. rhamnosus B-442. While, the pH values for samples containing Lb. helveticus, Lb. paracasei, and Lb. rhamnosus B-1445 were 5.05, 4.82 and 4.46, respectively. Then, the fermented milks were stored at 4˚C ± 1˚C for 14 days.

Camel milk was analyzed for total solids (%), protein (%) and fat (%) using the AOAC procedures [25]. The pH of milk was measured using a digital pH meter (Martini, Italy). Titratable acidity (lactic acid %) of raw and fermented camel milks was evaluated by titration with 0.1 N NaOH in the presence of phenolphthalein as an indicator. All analyses were performed in triplicate.

Water-soluble extracts were prepared in accordance with the method of Shori and Baba [26]. Fermented milk samples (10 g) were homogenized with 2.5 ml of distilled water then acidified to pH 4.0 with 0.1 M HCl, followed by holding at 45˚C for 10 min and centrifugation at 10,000 g for 10 min at 4˚C. The pH of supernatants was adjusted to pH 7.0 using 0.1 M NaOH and re-centrifuged (10,000 g, 10 min/4˚C) for further precipitation of proteins and salts. The supernatants were collected and kept at −20˚C for further analyses.

1) Cadmium–ninhydrin assay

Total free amino groups in water-soluble extract were assessed by the Cadmium–ninhydrin method described by Folkertsma and Fox [27]. The concentrations of total free amino groups were quantified against a standard curve of leucine (0.0 - 0.3 mM). Results were expressed as leucine equivalent in mM/mL extract. All determinations were performed in triplicate.

2) Sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE)

SDS-PAGE, 12.5% T, was conducted under reducing conditions using the discontinuous buffer system described by Laemmli [28]. SDS-PAGE was performed on fermented camel milk samples using a Mini-PROTEAN electrophoresis cell (Bio-Rad Laboratories, Hercules, CA, USA). Five grams of each sample were stirred with cold acetone (30 mL) for 10 min to get rid of the fat and left to dry at room temperature. The dried powdered fermented milk sample (20 mg) was mixed with 500 µL of sample buffer. Samples were denatured by boiling for 5 min, and then 5 µL of each sample was injected. The data were analyzed by Total Lab software (V1.11).

1) Determination of total phenolic content

Total phenolic contents (TPC) of water-soluble extracts were determined in triplicate using the method developed by Abirami et al. [29]. One and half milliliters of Folin-Ciocalteu’s reagent (diluted 10 times) and 1.2 mL of Na₂CO₃ (7.5% w/v) were added to 300 µL of water-soluble extract. Mixtures were shaken and kept at room temperature for 30 min before measuring absorbance at 765 nm using a spectrophotometer (Pg T80+, England). TPC was expressed as mg gallic acid equivalents (GAE)/mL extract.

2) DPPH radical scavenging activity assay

Scavenging activity of the stable 1, 1-diphenyl-2-picrylhydrazyl (DPPH) radical was determined in triplicate according to the procedure described by Lim and Quah [30]. Two milliliters of 0.15 mM DPPH dissolved in methanol were added to 1 mLof water-soluble extracts, mixed well and left in the dark for 30 min at room temperature. Absorbance (Abs) was measured at 517 nm against distilled water as a blank using a spectrophotometer (Pg T80+, England). Control was prepared by adding 2 mL of DPPH to 1 mL of methanol. The radical scavenging activity of extracts was calculated as follows

Radical scavenging activity % = ( 1 − Abs sample / Abs control ) × 100

3) Ferrous ion chelating assay

Ferrous ion chelating assay (FCA) was carried out as described by Chan et al. [31]. A freshly prepared ferrous sulphate hydrate (FeSO₄∙xH₂O) solution (2 mM) and ferrozine solution (5 mM) were both diluted 20 times before using. Diluted FeSO₄∙xH₂O (1 mL) was mixed with samples of water-soluble extracts (1 mL). Diluted ferrozine (1 mL) was then added to the mixture followed by incubation for 10 min at room temperature. Absorbance (Abs) was measured at 562 nm against distilled water as blank using a spectrophotometer (Pg T80+, England). The absorbance of control was determined similarly by replacing the supernatant with distilled water. The ferrous ion chelating ability (FCA) of extracts was determined in triplicate and calculated using the equation:

FCA % = ( 1 − Abs sample / Abs control ) × 100

4) Ferric reducing antioxidant power

Ferric ion reducing antioxidant power (FRAP) was determined according to Oyaizu [32]. One milliliter of extracts was added to 2.5 mL of phosphate buffer (0.1 M, pH 6.6) and 2.5 mL of potassium ferricyanide (1% w/v). The mixture was then incubated in a water bath at 50˚C for 20 min followed by cooling to room temperature and adding 2.5 mL of trichloroacetic acid (10% w/v). The contents of the tubes were centrifuged at 10,000 × g at 4˚C for 10 min. Next, 2.5 mL of supernatant was removed from each tube, and mixed with 2.5 mL of distilled water and 0.5 mL of ferric chloride solution (0.1% w/v). The mixtures were allowed to stand for 30 min then absorbance was measured at 700 nm using UV/Visible spectrophotometer, Pharmacia-LKB-Ultrospec III (Pharmacia, USA). The assay was done in triplicate. The FRAP values, expressed in mg GAE/mL extract, were derived from a standard curve.

The viscosity of fermented camel milk was measured in triplicate at 10˚C ± 1˚C using oscillatory viscometer (VR 3000M YR Viscometers, Spain), using spindle 2 at speed of 60 rpm.

Fermented milks were evaluated for color, taste, consistency and overall acceptability. The evaluation was conducted by 9 panelists, with experience in sensory evaluation of dairy products, from the Department of Dairy Science and Technology, Faculty of Agriculture, Alexandria University. Samples were presented cooled in plastic cups for sensory evaluation after 1, 7, and 14 days of refrigerated storage. Water was provided between evaluations of samples for mouth rinsing. The evaluation was determined using a 9-point hedonic scale (1 dislike extremely, 9 like extremely).

The data were analyzed by a general linear model procedure (GLM) using SAS statistical analysis software package (SAS Procedure Guide “Version 6.12 Ed.” SAS Institute Inc., Cary, 2004). The statistical analysis was performed using one-way analysis of variance (ANOVA). Means were compared by Duncan’s test at the significance level of P ≤ 0.05. Pearson’s correlation coefficient was used to calculate the correlation.

Cadmium-ninhydrin method was used to estimate proteolysis in fermented milk during a 14-day storage period. Table 1 shows the proteolysis degree (mM leucine equivalent/mL extract) in fermented milks prepared using commercial starter culture and some single strains of Lactobacillus. The results revealed that there were significant differences (P < 0.05) in the degree of proteolysis among fermented milks, depending on the type of starter culture. During the storage period, fermented milks containing commercial starter culture had the lowest proteolysis degree, while samples with Lb. helveticus showed the highest degree of proteolysis (P < 0.05). Differences between treatments could be attributed to the variation between starter cultures in proteinase-endopeptidase activity [37]. Strains of Lb. helveticus are known to have one of the most complex proteolytic

Mean values (±SD) with different small letters within the same column are significantly different; means with different capital letters within the same row are significantly different (P < 0.05).

systems among lactic acid bacteria [38]. On the first day the proteolysis degree was in the following order: Lb. helveticus > Lb. rhamnosus B-1445 = Lb. paracasei > Lb. rhamnosus B-442 > Lb. casei > commercial starter culture. This arrangement differed on the 7^th day and was as follows: Lb. helveticus > Lb. paracasei > Lb. casei > Lb. rhamnosus B-1445 = Lb. rhamnosus B-442 > commercial starter culture. This arrangement did not differ much on the 14^th day. The results also showed a significant decrease (P < 0.05) in the proteolysis degree in fermented milk containing Lb. helveticus with increased storage time. Fermented milks containing Lb. casei, Lb. paracasei or commercial starter culture exhibited a maximum degree of proteolysis on the 7^th day which decreased at the end of storage. The proteolysis degree of the samples prepared using Lb. rhamnosus B-1445 decreased on the 7^th day, then stabilized until the end of the storage. As for fermented milk containing Lb. rhamnosus B-442, there were no significant differences between the 1^st and 7^th days in the degree of proteolysis, which decreased markedly (P < 0.05) at day 14.

To determine the effect of lactobacillus strains on the degradation of camel milk proteins, samples of fermented milk at different times of storage were analyzed by SDS-PAGE (Figure 2). On the first day of storage, lower degree of proteolysis was observed in fermented milk containing commercial starter culture compared to the samples with single strains of lactobacillus. Meanwhile, Lb. helveticus showed higher proteolytic activity compared to other lactobacillus strains. On the 7^th day the degree of proteolysis increased in most samples, and higher proteolysis was observed with Lb. helveticus and Lb. casei samples. The lowest proteolysis was observed with samples containing commercial starter culture. On the 14^th day of storage the samples containing Lb. helveticus had the highest degree of proteolysis, while the lowest proteolysis was obtained with samples containing Lb. casei or Lb. paracasei. These results confirmed that the proteolytic

activity of most lactobacillus strains in this study increased with the increase in storage period and that the proteolytic activity of Lb. helveticus was the highest among the other lactobacillus strains. These results are consistent with those of El-Zahar et al. [39], who observed that all proteins gradually degraded during the cold storage of the yogurt. They also noted that αs₁-casein was higher degraded than β-casein during the cold storage. The proteolytic behavior of lactobacillus strains in this study in agree with that found by Li et al. [40], who reported that the intensity of κ-, β-, and α-casein bands in fermented milks were visually decreased during cold storage, meanwhile the electrophoretic bands referring to α-Lactalbumin and β-Lactoglobulin were unaffected in all treatments.

Data in Figure 3 illustrate the effect of commercial starter culture and some strains of Lactobacillus on the total phenolic concentration of fermented camel milks during cold storage. The results showed that on the 1^st day of storage the samples containing Lb. helveticus were the highest in TPC, while samples with commercial starter culture or Lb. casei were the lowest. Samples containing commercial starter culture, Lb. casei or Lb. paracasei showed an increase in TPC on day 7. Whereas, on day 14, the TPCs decreased significantly (P < 0.05) in all fermented milks except for samples containing Lb. rhamnosus B-1445. The order of TPC in fermented milk samples was as follows: Lb. helveticus > Lb. paracasei > Lb. casei > Lb. rhamnosus B-1445 > Lb. rhamnosus B-442 > commercial starter culture. The results also showed that TPC in fermented milk containing Lb. helveticus was the greatest (P < 0.05) throughout the storage compared to the other treatments. This result is related to the higher proteolytic activity of Lb. helveticus. In this study, there was a highly significant correlation (P < 0.0001) between the TPC and proteolysis degree. The decrease in TPC for fermented milk after 7 days of storage is in agreement with the results of Shori [35], who found that the TPC in plan camel yogurt was significantly decreased on day 7, and increased on day 14 of storage. Moreover, the proteolysis of milk proteins during storage may result in the release of phenolic amino acids and non-phenolic compounds which may interfere during TPC determination [41].

DPPH radical scavenging activity of fermented milks prepared using single strains of Lactobacillus as well as commercial starter culture was presented in Figure 4. As evident, there were significant differences (P < 0.05) in the scavenging activity values among the fermented milks. In addition, fermented milk containing commercial starter culture exhibited lower DPPH radical scavenging activity compared to other fermented milks during storage. On the 1^st day, samples

containing Lb. rhamnosus B-1445 or Lb. helveticus had the highest scavenging activity followed by samples with Lb. paracasei or Lb. casei, and then those containing Lb. rhamnosus B-442 or commercial starter. On the 7^th day, the highest activity was for samples containing Lb. rhamnosus B-442 followed by those containing Lb. paracasei, Lb. helveticus or Lb. rhamnosus B-1445 and then Lb. casei and finally the samples containing the commercial starter culture. On the 14^th day, the differences between samples were more pronounced compared to the 1^st and 7^th days of storage. The order of the samples in terms of activity was as follows: Lb. helveticus > Lb. paracasei > Lb. rhamnosus B-1445 > Lb. casei > Lb. rhamnosus B-442 > commercial starter. The results also revealed that the DPPH radical-scavenging activity for all fermented milks significantly decreased on day 7 of storage except for those samples containing Lb. rhamnosus B-442, which exhibited a significant increase (P < 0.05) in scavenging activity, while the scavenging activity of fermented milk containing Lb. paracasei was stable. At the end of storage period, there was a significant improvement (P < 0.05) in the scavenging activity for all fermented milk samples, particularly samples containing Lb. helveticus, Lb. paracasei and Lb. rhamnosus B-1445. In this study, no significant correlation (P > 0.05) was detected between the proteolysis degree and DPPH radical scavenging activity. This finding is consistent with the results of [42]. The absence of a relationship between the DPPH scavenging activity values and the degree of proteolysis indicates that the importance of the type and sequence position of functional amino acids in the peptides released for this activity [43] [44].

Overall, the antioxidant activity of fermented milk depends on various factors including the type and amount of the starter culture, the types of starter culture enzymes and the hydrolysis of proteins [45]. Further, the antioxidant properties of proteins and peptides count on many factors including the position of amino acids in their sequence as well as their physical structure, hydrophobicity and molecular weight [46].

Figure 5 presents the FCA of fermented milks prepared using single strains of lactobacillus during 14-day storage at 4˚C. The results showed that fermented milks differed significantly (P < 0.05) in FCA due to the type of starter culture. On the 1^st day, samples containing Lb. paracasei, Lb. rhamnosus B-442 or commercial starter culture recorded the highest FCA, while fermented milk containing Lb. helveticus had the lowest FCA. After 7 days of storage, the FCA of samples containing Lb. casei was the highest compared to other treatments, while fermented milk with Lb. helveticus were still the lowest in chelating ability. The FCA of fermented milks was in the following order: samples containing Lb. casei were higher than those containing Lb. paracasei or commercial starter, followed by samples with Lb. rhamnosus B-1445 or Lb. rhamnosus B-442, and then fermented milks with Lb. helveticus. By the end of the storage period, the arrangement of samples differed completely in their ability to chelate iron. The FCA of

fermented milks was in the following order: samples with Lb. rhamnosus B-1445 > Lb. casei > Lb. rhamnosus B-442 > Lb. helveticus > commercial starter culture > Lb. paracasei. The FCA of fermented milks could be attributed to the iron-binding peptides that are generated from milk proteins due to the proteolytic activity of the starter cultures. The hydrolysis of milk proteins by LAB to release peptides that have the potential to bind minerals has been reported [47] [48]. The functional groups of amino acid residues within derived peptides such as phosphate, carboxyl, hydroxyl and methyl as well as aromatic rings act as binding sites for transition metals ions [49]. In addition, The N- and C-terminals of peptides have been reported to contribute the metal chelating activity [50]. The variations in FCA between fermented milks during storage could be attributed to the structure of peptides that are generated as a result of the proteolytic activities of starter cultures [47]. In current study, no significant correlation (P > 0.05) was detected between the high degree of proteolysis and the FCA. A similar relationship between the hydrolysis degree and FCA was stated by [51].

The FCA decreased significantly (P < 0.05) for all fermented milks on the 7^th day, with the exception of samples containing Lb. paracasei, where the ferrous chelating ability increased. On the 14^th day of storage, a marked improvement in the FCA values was observed in fermented milks with Lb. helveticus and Lb. rhamnosus B-1445. While there was a continuous decrease in the FCA in the samples containing Lb. casei, Lb. paracasei and commercial starter culture. Regarding Lb. rhamnosus B-442 samples, the FCA was stable compared to day 7 of storage. A number of studies have found that the chelating capability of fermented milk samples increased with increased storage [45] [48].

Data in Figure 6 illustrate the FRAP values of fermented camel milks prepared using single strains of lactobacillus during cold storage. On the 1^st day of storage, the greatest FRAP value (0.322 mg GAE/mL) was found with Lb. helveticus samples. While, on the 7^th day the highest FRAP values (0.212, 0.211 and 0.206

mg GAE/mL) were observed with the samples containing Lb. paracasei, Lb. helveticus and Lb. casei, respectively. On the 14^th day the highest FRAP values (0.180 and 0.181 mg GAE/mL) were found with Lb. helveticus and Lb. rhamnosus B-1445 samples, respectively. On the other side, fermented milk samples containing Lb. rhamnosus B-442 or commercial starter culture showed the lowest FRAP values throughout storage.

The results also showed that on the 7^th day of storage, fermented milks containing Lb. casei showed a significant increase (P < 0.05) in FRAP values. In addition, FRAP values did not change in samples containing Lb. paracasei or commercial starter culture, while there was a decrease in these values in samples with Lb. helveticus, Lb. rhamnosus B-442 or Lb. rhamnosus B-1445 compared to the FRAP values on the 1^st day of storage. At the end of storage, all fermented milk samples showed a decrease in FRAP values. In this study, the correlation between the values of FRAP and proteolysis degree was highly significant (P < 0.0001). The change in the proteolytic activity of Lactobacillus strains during storage affected the release of peptides from milk protein and consequently impacted the FRAP values of fermented milks. It was reported that increased proteolysis in fermented milk caused increased generation of bioactive peptides, which ultimately resulted in increased antioxidant capacity [22]. Ayyash et al. [20] reported that the fermentation of camel milk with indigenous camel milk probiotic LAB strains (Lb. reuteri-KX881777, Lb. plantarum-KX881772, Lb. plantarum-KX881779) showed maximum antioxidant activity compared to bovine milk.