Vol. 6 (1997): 193-198. Optimization of centrifugal separation of a-lactalbumin and (3-lactoglobulin Tuomo Tupasela Agricultural Research Centre ofFinland, Food Research Institute, FIN-31600 Jokioinen, Finland, e-mail: tuomo.tupasela@mtt.fi Petri Koskela Department ofFood Technology, FIN-00014 University of Helsinki, Finland Eero Pahkala, Veikko Kankare Agricultural Research Centre ofFinland, Food Research Institute, FIN-31600 Jokioinen, Finland Whey proteins, which are mainly composed of P-lactoglobulin (P~lg) and a-lactalbumin (a-la), ac- count for about 20% of the proteins of bovine milk. In this study we investigated the effect of pH, dry matter content, concentration factor, heat treatment and centrifugation on the separation of a-la from P-lg using clarified whey as raw material, a-La precipitation was highest, 23.3%, when the dry mat- ter content ranged from 5.8% to 25.7%. The optimum pH of a-la precipitation depended on the dry matter content. The separation efficiency increased when the concentration factor and heat treatment time at 55°C increased. A longer centrifugation time and higher separation speed did not have a marked effect on the separation efficiency. Separation was more efficient with a higher centrifuga- tion speed at concentration levels 30 X and 60 X. The separation efficiency did not improve when the temperature was raised from 55°C to 65°C but it was better at a concentration level 120 X than at 60 X and 30 X, and also at concentration level 60 X than with 30 X. Key words: a-lactalbumin, P-lactoglobulin, separation fractionation, filtration, concentration ntroduction Whey proteins, which are mainly composed of P-lactoglobulin (p-lg), ~3.3 g/kg whey, and a- lactalbumin (a-la), ~1.0 g/kg whey, account for about 20% of the protein content ofbovine milk (Walstra and Jenness 1984). Many efforts have been made to separate these major components with a view to producing whey protein concen- trates with improved and specific functional properties (Amundson et al. 1982, Pearce 1983, Pearce 1987, Maubois et al. 1987, Rosenberg 1995). These efforts have involved gentle heat- ing of whey, followed by concentration of whey at acidic pH values to produce selective precip- itation of a-la or P-lg. Pearce (1983) reported that maximum pre- © Agricultural and Food Science in Finland Manuscript received March 1997 193 AGRICULTURAL AND FOOD SCIENCE IN FINLAND Tupasela, T. et al. Centrifugal separation of whey proteins cipitation of a-la in whey and optimum separa- tion of the proteins were obtained at about 65°C when the whey was kept at a pH range from 4.1 to 4.3 for several minutes. No clarifying pretreat- ment of whey was used. McDonough et al. (1974) and de Wit (1984) showed that the pres- ence of fat impairs the whipping properties of a whey protein concentrate (WPC). Lipids are also susceptible to oxidation and thus may contrib- ute to an off-flavour, which is best prevented by removal of lipids from whey (de Boer et al. 1977). In a clarification method introduced by Maubois et al. (1987) for the pretreatment meth- od of whey, the residual fat and lipid components were removed with microfiltration. Tupasela et al. (1994) studied the effect of clarification of whey on whey ultrafiltration. We discuss here the effect of clarification of whey on the separa- tion of a-la from (i-lg. The effects of pH, con- centration factor, dry matter content, heat treat- ment and centrifugation were studied. All exper- iments were conducted on pilot-plant or labora- tory scale. Material and methods Whey Fresh edam cheese whey was obtained from the Food Research Institute’s dairy plant, and the cheese milk was pasteurized at 74°C for 15 s. After the cheese had been made the whey was centrifuged and cooled to 2°C in a process tank. CaCI, solution (CaCl 2 x 2H,0 662.1 g/1 a.d. H,O), 720 ml per 100 I whey, was added to the whey and the pH of the whey was adjusted to 7.3 with 1 M NaOH solution using a process based on that ofFauquant et al. (1985). Immedi- ately after these treatments, the whey was heat- ed at 50°C for 8 min in a mixing tank provided with a heating/cooling jacket. After heat treat- ment the whey was cooled to 40°C with ice wa- ter. Whey processing Heat-treated whey was microfiltered with an APV CL 3/40 microfiltration (MF) unit equipped with 0.2 pm ceramic membranes (Ceraver, France) in two modules. Each module had a membrane area of 1.4 m 2. In the MF unit the whey circulation speed was 5 m/s and the tem- perature was kept at between 35 and 40°C with a heat exchange section in the circulation loop. The whey inlet pressure was 3 bar. The whey was microfiltered by recycling in a batch run. The microfiltration permeate was collected and the temperature was lowered to 20°C in a proc- ess tank by circulating ice water in the agitator. The MF unit was cleaned after each run with 1 % NaOH solution. The permeate was concentrated with a mem- brane ultrafiltration unit I (PCI Bro MK, UK; cut-off 9000) and unit II (Millipore, Pellicon Cassette, USA; cut-off 10 000 PTGC). The pH of the WPC was adjusted to the desired level with 2 M HCI, after which the WPC was heated in a water bath to 55°C or 65°C for 5, 15 or 30 min. This treatment caused precipitation of a-la. The precipitate was collected by centrifugation (Sor- vali RC-5B Superspeed Centrifuge using a SS- -34 rotor, Du Pont Instruments, Connecticut, USA), at 5000 rpm and 10 000 rpm (3020 g and 12 100 g, respectively). Analysis The total solids content of differently treated wheys was determined after drying the wheys for 16 h at 102°C using the modified method of IDF 4 (1958). The protein content was deter- mined by the Kjeldahl method with N conver- sion of 6.38 (TECATOR 1975). The pH was measured with a Knick Portamess 752 pH meter (Berlin, Germany). a-La and (3-lg were identi- fied and quantified by a FPLC Mono Q HR 5/5 chromatography column (Pharmacia) connected to a UV detector (280 nm) using the modified method of Humbrey and Newsome (1984). 194 AGRICULTURAL AND FOOD SCIENCE IN FINLAND Vol. 6 (1997): 193-198. Results and discussion Effect of whey dry matter content and pH on separation First we studied how ot-la precipitated when the pH range was from 3.2 to 4.5 at a dry matter content 12.5%. The a-la precipitation level was obtained when the (3-lg content in supernatant was determined. The results in Figure 1 show that the pH range from 3.4 to 3.8 is the most favourable for the precipitation of a-la. When the dry matter content of whey was 12.5% and the pH ranged from 3.2 to 4.5 the maximum a- la precipitation was achieved at a pH value of around 3.6. In the next trial we had ten different whey dry matter contents (from 5.8% to 25.7%) and a- la and (3-lg were separated at seven dif- ferent pH values (from 3.4 to 4.0). Higher pH values, 3.9 and 4.0, were still included in the experiment because the dry matter content of three samples were lower than earlier. Figure 2 shows that at the lowest dry matter content, 5.8%, a-la did not precipitate well enough. When the dry matter content was raised to 9.2%, more a- la was precipitated; maximum precipitation was achieved at a dry matter content of 23.3%. The pH optimum of a-la precipitation decreased as the dry matter content increased. Figure 3 sug- gests that, in the separation of (3-lg from a-la, the dry matter content had a greater effect than the protein content, because the dry matter con- tent line has a better positive correlation with the (3-lg content line than has the protein con- tent (%, protein in dry matter) line. Fig. 1.P-lg content (%) of supernatant. Dry matter content of whey 12.5%, heat treatment 55°C/30 min and centrifu- gation 10 000 rpm/20 min. Fig. 2. Effect of dry matter content and pH of whey on (3-lg contents (%) in supernatants. Dry matter contents are presented below the figure. 195 AGRICULTURAL AND FOOD SCIENCE IN FINLAND Tupasela, T et at. Centrifugal separation of whey proteins Effect of concentration factor, heat treatment and centrifugation speed and time on ot-la and (3-lg separation In this trial we used three different concentra- tion factors, 30 X, 60 X and 120 X. The dry matter contents (%) were 11.6, 14.5 and 19.0, respectively. The whey was processed as de- scribed above. The pH before treatments was adjusted to 3.6 with 2 M HCI. In trial 1 the con- centration factor was 30 X and the temperature treatments were 55°C/5 min, 55°C/15 min, 55°C/ 30 min and 65°C/30 min. Each heat-treated sam- ple was centrifuged at 5000 rpm/20 min and 10 000 rpm/20 min. In trial 2 the concentration factor was 60 X and the treatment was as in trial 1. In trial 3 the concentration factor was 120 X. The sample was treated as in trial 1 and 2, but was centrifuged also at 5000 rpm/10 min and 10 000 rpm/10 min. The results are shown in Table 1. From Table 1 it can be seen that the ratios (cc-la/p-lg X 100) range from 7 to 20. The lower the value, the purer was the supernatant with re- gard to P-lg. Note that when the concentration factor and heat treatment time at 55°C increased, the separation efficiency also increased. A long- er centrifugation time and greater separation speed did not have a marked effect on separa- tion efficiency. Separation efficiency improved with the higher centrifugation speed at concen- tration levels 30 X and 60 X but not when the temperature was raised from 55°C to 65°C. Sep- aration efficiency was better at a concentration level 120 X than at 60 X and 30 X, and also at concentration level 60 X than at 30 X. Conclusion In this study oc-la precipitation was highest at 23.3%. The optimum pH of oc-la precipitation depended on the dry matter content. Separation efficiency increased with an increase in concen- tration factor and heat treatment time at 55°C, and also with a higher centrifugation speed at concentration levels 30 X and 60 X. There was no improvement when the temperature was raised from 55°C to 65°C. Separation efficiency was better at concentration level 120 X than at 60 X and 30 X, and concentration level 60 X than at 30 X. In the light of the above results, and the in- dustrial potential for separating these major whey protein components from whey, in order to pro- Fig. 3. (3-lg contents (%) in supernatant versus whey dry matter content ( %) and protein content (%, protein in dry matter). 196 AGRICULTURAL AND FOOD SCIENCE IN FINLAND Voi 6 (1997): 193-198. Table I, Effect of concentration factor, heat treatment and centrifugation speed and time on tx-la and (1-lg separation. Concentration factor = 30 X, 60 X and 120 X; heat treatment = 55°C/5 min, 55°C/15 min, 55°C/30 min and 65°C/30 min; centrifugation speed and time = 5000 rpm/10 min, 5000 rpm/20 min, 10 000 rpm/10 min and 10 000 rpm/20 min. Supernatant a-la/p-lg X 100 ratio Heat treatment Concentration factor Centrifugation speed and time 30 X 60 X 120 X 55°C/smin 11 5000 rpm/10 min 18 12 10 5000 rpm/20 min 10 10 000 rpm/10 min 20 12 10 10 000 rpm/20 min 55°C/15min 8 5000 rpm/10 min 16 11 8 5000 rpm/20 min 8 10 000 rpm/10 min 12 10 8 10 000 rpm/20 min 55°C/30min 7 5000 rpm/10 min 14 9 7 5000 rpm/20 min 7 10 000 rpm/10 min 13 8 7 10 000 rpm/20 min 65°C/30min 9 5000 rpm/10 min 13 11 8 5000 rpm/20 min 9 10 000 rpm/10 min 13 8 8 10 000 rpm/20 min - = not determined duce whey protein isolates with specific func- (dry matter content e. 13%) and temperature tional properties, the most suitable combination treatment at 55°C/30 min. This combination would be a concentration factor of 30 X to 60 X should be tested on industrial scale. References Amundson, C.H., Watanawanichakorn, S. & Hill, C.G. 1982. Production of enriched protein fractions of p- lactoglobulin and ot-lactalbumin from cheese whey. Journal of Food Processing and Preservation 6: 55- 71. Boer, R., de Wit, J.N. & Hiddink, J. 1977. Processing of whey by means of membranes and some applica- tions of whey protein concentrate. Journal of the Society of Dairy Technology 30: 112-120. Fauquant, J., Vieco, E., Brule, G. & Maubois, J.-L. 1985. Clarification des lactosérums doux par agrégation thermocalcique de la matiére grasse résiduelle. Lait 65: 1-20. IDF 4.1958. Dry matter in cheese and processed cheese. Humbrey, R.S. & Newsome, L.J. 1984. High performance ion-exchange chromatography of the major bovine milk proteins. New Zealand Journal of Dairy Science and Technology\9: 197-204. Maubois, J.L., Pierre, A,, Fauquant, J. & Piot, M. 1987. Industrial fractionation of main whey proteins. IDF Bulletin 212: 154-159. McDonough, F.E., Hargrove, R.E., Mattingly, W.A., Posa- ti, L.R & Alhford, J.A. 1974. Composition and prop- erties of whey protein concentrates from ultrafiltra- tion. Journal of Dairy Science 57: 1438- 1443. Pearce, R.J. 1983. Thermal separation of p-lactoglobulin and a-lactalbumin in bovine Cheddar cheese whey. Australian Journal of Dairy Technology 38:144-148. Rosenberg, M, 1995. Current and future applications for membrane processes in the dairy industry. Trends in Food Science & Technology 6: 12-19. - 1987. Fractionation of whey proteins. IDF Bulletin 212: 197 AGRICULTURAL AND FOOD SCIENCE IN FINLAND Tupasela, T. et al. Centrifugal separation of whey proteins 150-153. TECATOR. 1975. Manual Kjeltec 11. Helsinborg, Sweden. Tupasela, T., Koskinen, H. & Antila, P. 1994. Whey pre- treatments before ultrafiltration. Agricultural Science in Finland 3: 473-479. Walstra, P. & Jenness, R. 1984. Dairy chemistry and phys- ics. J. Wiley & Sons, New York. p. 467. Wit, J.N. de, 1984. Functional properties of whey pro- teins in food systems. Netherlands Milk and Dairy Journal 38: 71-89. SELOSTUS oc-lactalbumiinin ja p-lactoglobuliinin sentrifugointierotuksen optimointi Tuomo Tupasela, Petri Koskela, Eero Pahkala ja Veikko Kankare Maatalouden tutkimuskeskus ja Helsingin yliopisto Heraproteiinit käsittävät noin 20 % maidon proteii- neista. Heraproteiinit koostuvat pääasiassa (i-lakto- globuliinista (P-lg) ja a-laktalbumiinista (a-la). Teh- dyissä kokeissa tutkimme pH:n, kuiva-ainepitoisuu- den, väkevyyden, lämpökäsittelyn ja sentrifugoinnin vaikutusta a-la:n ja (3-lg:n erottumiseen kirkastetus- ta juustoherasta. a-laktalbumiinin saostuminen onnistui parhaiten, kun kuiva-ainepitoisuus oli 23,3 %, kuiva-ainepitoi- suuden vaihdellessa 5,8 ja 25,7 %:n välillä. Kuiva- ainepitoisuus vaikutti a-la:n saostumiseen optimi pH- tasolla. Sentrifugointitulos parani, kun väkevyyttä ja aikaa nostettiin 55°C:n lämpökäsittelyssä. Pidempi sentrifugointiaika ja suurempi sentrifugointinopeus eivät vaikuttaneet paljon sentrifugointitulokseen. Konsentrointikertoimilla 30 X ja 60 X saatiin parempi sentrifugointitulos kun sentrifugointinopeutta kasva- tettiin. Sentrifugointitulos ei parantunut, kun käsit- telylämpötilaa nostettiin 55°C:sta 60°C:een. Konsen- trointikertoimella 120 X saatiin parempi sentrifu- gointitulos kuin konsentrointikertoimilla 30 X ja 60 X, ja konsentrointikertoimella 60 X saatiiin parempi sentrifugointitulos kuin konsentrointikertoimella 30 X. 198 AGRICULTURAL AND FOOD SCIENCE IN FINLAND