Microfiltration, a processing technology to have safe, sure buffalo dairy products with their natural quality

Pakistan is the third largest milk producer and second largest buffalo milk producer in the world. Buffalo milk is the second produced milk with a contribution of 12.7% in the world’s milk production. More than 92% of this milk is being contributed by India and Pakistan. This milk is richer in protein particularly casein i.e. 34-36 g.kg-1, equivalent to the total protein contents of cow milk. Casein represents 80% of the total protein of both milks and find as colloidal particles called casein micelles. In buffalo milk, casein micelles are bigger in size, more mineralized and less hydrated but with similar charge as compared to its counter part cow milk (Ahmad et al, 2008a and b). In spite of the important production of buffalo milk, the technological transformations of this milk into dairy products are very limited as compared to cow milk. Some cheeses or fermented milks like Cheddar, Mozarella, lassi and yoghurts are manufactured from this milk. Protein ingredients manufactured from buffalo milk such as caseinates associated with different minerals and purified whey proteins are absent on the dairy industrial market. One possible way to obtain these types of ingredients from buffalo milk is to use cross-flow microfiltration. This operation is widely used in the dairy industry for the fractionation of casein micelles and whey proteins from cow milk (Maubois, 1991; Saboya and Maubois, 2000; Nelson and Barbano, 2005). This process produces a retentate enriched in casein micelles with a purity expressed as the ratio [casein]/[total protein] higher than 0.90 (Fauquant et al, 1988; Pierre et al, 1992; Schuck et al, 1994). The technological behaviours of these purified casein micelles during acidification, rennet coagulation and heat treatment are also very similar to milk. For these reasons, these casein micelles purified by cross-flow microfiltration are considered as native. The objectives of this work were to apply cross-flow microfiltration (0.1 µm) firstly, to have bacteria free buffalo milk without any heat treatment, secondly to isolate and characterize the casein micelles from buffalo milk. For this purpose, cream and bacteria from milks were removed at 50°C through a cream separator and microfiltration (1.4 µm), respectively. Casein micelles were then separated and concentrated through cross-flow microfiltration (0.1 µm) followed by purification through four volume water and again concentrated through water removal. Purification was performed in three main steps: concentration of casein micelles, diafiltration of retentate and final concentration (Fig. 1). The products obtained at different steps of the process were analysed for pH, total solids, proteins fractions, minerals contents and as well for size, zeta potential and hydration of casein micelles. Figure 1: Technological scheme for the purification of casein micelles from buffalo and cow milk by cross-microfiltrations (pore size of 1.4 and 0.1 μm). Microbiological analyses of raw skimmed milk and bacteria free milk after the first microfiltration (1.4 µm) showed a great reduction of coliforms and FMAR in both milks (Table 1) so the first objective was well achieved. Table 1: Coliforms and FMAR counts Milk Coliforms FMAR Raw buffalo skimmed milk 3.3×101 7.5×101 Microfiltered buffalo milk <1 1.5×101 Raw cow skimmed milk 9.7×104 3.2×104 Microfiltered cow milk <1 6.5×101 As the second objective of this study was to purify casein micelles from buffalo milk by cross-flow microfiltration, so the results have been described and discussed for each step during the process. The biochemical compositions and physico-chemical characteristics of initial bacteria-free skimmed milks and the different retentates (intermediate, diafiltered and final) have been reported as under. Results obtained with buffalo milk were compared to cow milk and discussed. Bacteria-free skimmed milks: The contents in total protein and casein were higher for buffalo milk than that of cow milk. However, in both milks, casein content as compared to the total protein content corresponded to similar percentage i.e. 80 %. To know more about the differences in the composition between buffalo and cow milk, it was interesting to compare the different ratios. The ratios [total protein]/[total solid], [casein]/[total solid] and [casein]/[whey protein] were 0.41, 0.35 and 5.3 for buffalo milk and 0.35, 0.27 and 4.5 for cow milk, respectively. From these ratios, it is evident that the protein contents of buffalo milk were different than that of cow milk. The total and micellar calcium (Ca) and inorganic phosphate (Pi) concentrations were higher in buffalo milk than in cow milk as also observed by Ahmad et al. (2008a). The pH values of both milks were close to pH 6.7. The size of casein micelles from buffalo milk was higher, the zeta potential was similar and the micellar hydration was lower than that of cow milk and as described by Ahmad et al. (2008a). Intermediate retentates: The contents in total protein and casein were increased for both intermediate retentates in comparison with initial skimmed milks. A slight retention of whey proteins in the intermediate retentate for cow milk compared to buffalo milk was observed. The ratios [total protein]/[total solid] increased up to 0.54 and 0.57 for intermediate retentates of buffalo and cow milk, respectively. At the same time, the ratios [casein]/[whey proteins] increased from about 5 for both initial milks to about 10 for both intermediate retentates. These calculated ratios indicated that the purifications of casein micelles were in progress in a similar way for both milks. The total Ca and Pi concentrations increased by 1.7 and 1.6 times for intermediate retentates of buffalo milk, and by 2.3 and 2.1 times for intermediate retentates of cow milk, respectively in comparison with milks. As discussed with the non protein nitrogen compounds, minerals and ions H+ were also able to pass through the membrane and consequently their concentrations were the same in retentate and permeate. Diafiltered retentates: The contents in non casein nitrogen, non protein nitrogen, whey protein, Ca and Pi were strongly reduced. During the diafiltration, the major part of the soluble compounds present in the aqueous phase of retentates like lactose, minerals, whey proteins and small molecules were removed. The concentration of whey proteins was lower in diafiltered retentate of buffalo milk than that of cow milk. This difference indicated that the whey proteins of buffalo milk were more easily transmitted than cow milk. On the other hand, the contents in total solids of these diafiltered retentates were strongly reduced. During diafiltration, 50.3 and 54.5 g.kg-1 of total solid contents was removed from buffalo and cow milk, respectively. At the end of diafiltration, the total solid corresponded essentially to casein micelles containing minerals. These ratios indicated that diafiltration was particularly efficient to increase the level of purity of casein micelles for both milks. At the same time, the ratios [casein]/[whey proteins] were about 42.3 and 26.7 for diafiltered retentates of buffalo and cow milk, respectively suggesting a good and best elimination of the whey proteins in the diafiltered retentates of buffalo milk in comparison with those of cow milk. The pH values of both diafiltered retentates increased in a similar way by about 0.6 units. The reduction of the ionic strength and especially Pi and citrate from the aqueous phase due to the diafiltration explained theses increases in pH. Indeed, these ions contribute to the buffering capacity of milk and their removals during the diafiltration step led to an increase in pH. The micellar characteristics of both diafiltered retentates were affected in comparison with those determined for milks and intermediate retentates. The zeta potentials of casein micelles from both retentates were more negative after diafiltration. These were related to the increase in pH and to the reduction in the ionic concentration of the aqueous phase induced by diafiltration. Final retentates: The contents in total solid, total protein and casein were increased in comparison with the corresponding intermediate and diafiltered retentates. The ratios [total protein]/[total solid] were 0.86 and 0.89 for final retentates of buffalo and cow milk, respectively. For the cow milk, the ratio obtained in this work was in accordance with the published values of 0.84 (Pouliot et al., 1994), 0.86-0.88 (Pierre et al., 1992) and 0.90 (Schuck et al., 1994). The ratios [casein]/[total solid], [casein]/[total protein] and [casein]/[whey protein] were 0.84, 0.98 and 45 for final retentate of buffalo milk against 0.86, 0.96 and 27 for final retentate of cow milk. The last calculated ratio confirmed the slight retention of whey proteins for cow milk compared to buffalo milk. Two possible reasons can explain this difference of filtration. Firstly, the composition and structures of the whey proteins from buffalo milk are different and consequently their capacities to be microfiltered were also different. The other reason concerns the formation of a fouling layer at the membrane surface during filtration. These different calculated ratios showed that the level of purification of casein micelles of buffalo milk by cross-flow microfiltration was correct and satisfactory. 98-99% of Ca and Pi were associated to casein micelles in both final retentates. Micellar zeta potentials were always more negative as compared to casein micelles of milks because the aqueous phase became very poor in minerals. Conclusions: This study showed that the safe and microbial free buffalo milk can be obtained through cross-flow microfiltration without involving heat and without modifying the nature and nutritional quality of milk. So microfiltration can be considered as a processing technology to have safe, sure buffalo dairy products with their natural quality. It also showed that purification of casein micelles from buffalo milk was possible like cow milk without any major problem. The biochemical analyses of the final retentate of buffalo milk showed that it contained mainly casein and minerals and that the different compounds present in the aqueous phase were well removed. These ingredients can be used in different formulations for Ca and phosphate enrichments.