D A I R Y FOODS TECHNICAL NOTES Separation of Casein Micelles from Milk for Rapid Determination of Casein Concentration I R O B E R T N. C A R P E N T E R 2 and R O D N E Y J. B R O W N Department of Nutrition and Food Sciences Utah State University Logan 84322 ABSTRACT Size exclusion chromatography was used to separate casein micelles from whey protein in milk for rapid measurement of casein. Casein concentration was measured by size exclusion separation monitored by ultraviolet spectroscopy and by acid precipitation followed by Kjeldahl estimation of protein. The correlation coefficient between the two methods was .92. Effects of temperature, pH, and calcium on separation were evaluated. Addition of calcium was used to incorporate all casein into micelles before separation. INTRODUCTION Cheese producers have begun using yield formulas to calculate the price paid to producers for milk (6, 8). Both fat and casein percentages in milk must be known to predict accurately cheese yield. Casein concentration is estimated routinely as a percentage of total protein because protein percentage is obtained more easily and rapidly than casein percentage. Casein concentration as a percentage of total milk protein (casein number) is variable (3, 15, 20). Anything that influences percent casein, percent whey proteins, or percent noncasein nitrogen affects casein number, so accuracy of cheese yield predictions based on estimated casein is questionable. Casein micelles are spherical, highly hydrated, and sponge-like (5) protein particles suspended in milk. Ninety-five percent of casein micelles in milk are between 40 and 22 nm in diameter (4, 17). A small micelle of 250,000 daltons is Received June 12, 1984. 1Utah State Agricultural Experiment Station Journal Article No. 2984. Approved by the Director. 2Kraft Inc., Box T, Albany, MN 56307. 1985 J Dairy Sci 68:307--311 307 relatively large compared to whey proteins, which are from 15,000 to 70,000 daltons. Immunoglobulins are larger, 150,000 to 300,000 daltons, but exist in low concentrations (about 4% of total protein) in milk (10). Calcium is important in structural stability of casein micelles (2). Casein micelles dissociate as calcium ion activity is reduced. Small additions of calcium ions cause transfer of soluble casein to micelles without changing micelle radii, and addition of more calcium ions causes larger micelles to form (4). Casein usually is separated from whey proteins before measurements can be made. Casein then may be determined by protein measured in the casein fraction or the difference in percent protein between milk and whey. Size exclusion chromatography separates molecules in solution according to size (22) and has been used in the study of physiical and chemical properties and composition of casein miceIles (7, 11, 16, 18, 21). This study was to determine if size exclusion chromatography could be used to separate casein micelles from whey proteins for rapid casein determination. Casein micelles separated in this way remain in suspension and allow direct casein measurement b y various protein assays. An ultraviolet spectrophotometer operating at 280 nm wavelength was used to measure protein. Infrared spectroscopy would be ideally suited for these measurements and would make rapid casein testing possible on instruments now being used by the dairy industry. MATERIALS AND METHODS The apparatus for casein separation by size exclusion chromatography is in Figure 1. A pump and sample injector from a PerkinElmer series 2 chromatography system were used with a stainless steel chromatography column. Glycophase-coated controlled pore glass was chosen as chromatography column packing material because of its ability to endure 308 C A R P E N T E R AND BROWN Eluent Pump ~ Sample Column Computer Monitor I Figure 1. Flow diagram of equipment for separating casein micelles from whey protein and monitoring elution. faster flow rates and higher pressures than gels. Pore diameter and particle size were selected to separate casein micelles from whey proteins. Both 2,000 and 4,000 nm pore diameter, 37 to 74 /am diameter beads were evaluated. The 2,000 nm beads were chosen because of better separation of casein and whey protein. Separation was possible in columns as short as 15 cm; however, a 100-cm column was used in the experiments described in this paper. Inside diameter of the column is not a critical separation characteristic but was selected at .4 cm so small sample sizes (10 to 150/al) could be used. The column flow rate was held constant at 1 to 6 ml of 40 mM CaC12/min. Before measurement, the system was equilibrated by passing several milk samples through the column. A Beckman DU-8B spectrophotomer was used to monitor protein at 280 nm as it eluted from the column. The contribution of light scattering does not interfere with the usefulness of the readings. At 4-s intervals absorbances were sent from the spectrophotometer to a Tektronix 4052 computer, which produced an elution plot and stored data on tape to be evaluated later. The reference method for determining casein concentration in milk consisted of separation of casein by acid precipitation followed by Kjeldahl nitrogen determination of the precipitate (1, 19). Sample size and solutions of the separation step were reduced to 1/20th so that microKjeldahl procedures could be followed. A Journal of Dairy Science Vol. 68, No. 2, 1985 solution of 10% mercuric sulfate was used as catalyst. Boric acid solution containing methyl red/bromcresol green indicator and the ammonia distillate were titrated with .0258 34 HCI to a pink-grey end point. Percent casein was obtained by percent nitrogen, minus reagent blank, multiplied by 6.52 (12). Whole milk samples from individual cows had potassium dichromate tablets added at the farm as preservative. Samples were filtered through Whatman No. 1 filter paper and forced twice at 40°C through a small sample homogenizer. Each 50-ml sample was allowed to equilibrate for 24 h at 4°C. Some samples had .1 ml of 5 M CaClz added before equilibration. This could be added as a tablet at the farm along with the preservative tablet. Before injection, samples were held at 40°C for at least 2 h. A typical elution plot of milk produces two peaks (Figure 2). The computer was used to calculate the volume at which each peak eluted, maximum absorbance reading of each peak, and area under the curve according to Simpson's approximation. Area of the first peak was calculated from base line to a point midway between the two peaks. Area of the second peak was calculated from the same midpoint to a point after the second peak at which the curve returned to the baseline. All of these measures were stored on computer tape and later used in statistical analyses. Casein was obtained by precipitation from milk with 1 N HC1. The precipitate was redissolved in simulated milk ultrafiltrate (13). A solution of a-lactalbumin and fl-lactoglobulin also was prepared. Both of these solutions were injected into the column to determine their elution rates. Effluent containing casein and whey protein peaks collected from the column was analyzed by polyacrylamide gel electrophoresis following the procedure used by Heth and Swaisgood (11). The samples were dried in a vacuum oven at 50°C. Modified Poulik's buffer (50 pl) was added to each sample. Mercaptoethanol (100/al of 1:9 dilution) was added, and samples were refrigerated for 2 h. Dye marker, acrylamide solution, preserving, and staining solutions were prepared according to LKB Application Note 306 (9). Buffer solution and gel solutions were made according to Kiddy (14). The gel was prepared and applied to a flat bed electrophoresis unit at least 12 h before use. Samples DAIRY FOODS TECHNICAL NOTE 4.0 3.S 3.0 j 2.5 2.0 - dL 6 ,S I IX 10 VOLUME CmL~, 4.0 $.5 2.S ~ 2.0 i 1.0 .5 added, the eluent was distilled water. Samples were introduced randomly into the column, except that calcium-treated samples were run separately from untreated samples. Thirty six elution plots were obtained from which characteristics were calculated. Characters were evaluated by analysis of variance to determine each treatment effect. Casein percentage was estimated from triplicate elutions of eight milk samples from individual cows. Fat and protein percentages for each sample were obtained with a Multispec infrared instrument. Casein percentage was obtained by the standard acid precipitation and Kjeldahl nitrogen method. Measures from column elution and fat and protein percentages were tested for correlation with percent casein. Analysis of covariance and stepwise regression were performed, and a regression equation to estimate casein from elution measures was derived. RESULTS A N D DISCUSSION 0 - 309 ,S (! i i 6 I0 IE VOLUME On,L) Figure 2. Characteristic protein elution plots of nontreated (top) and calcium treated (bottom) milk samples. (10/21) were injected into gel slits and 20 m A of current was applied for 15 rain; then voltage was maintained at 200 V for about 4 h. Changes of the elution pattern caused by pH, temperature, and addition of calcium were evaluated. A factorial design experiment used a single commingled milk sample. Treatments were run through the column in duplicate. Temperature treatments were: milk injected cold at 4°C, milk held at 40°C for at least 2 h before injection, and milk heated to 72°C for 1 rain then cooled to 40 C. The pH treatments were milk at pH 6.58 (no treatment), pH 6.7 (100/21 1 N NaOH/50 ml), and pH 6.4 (50/21 1 N HC1/50 ml). Each temperature and pH combination also was run with calcium (100/21 CaCt2/50 ml milk) and without calcium. When calcium was added to milk, the eluent was a 40 mM CaC12 solution; when calcium was not A peak appeared at 5.1 ml and a slight rise in absorbance near 10 ml when casein was passed through the size exclusion column (Figure 3). The larger peak contained 97% of the total area under the elution curve and consisted of casein micelles. The small peak contained whey protein not separated by casein precipitation or nonmicellar casein. A solution of two whey proteins (~-lactalbumin and /3-1actoglobulin) also was eluted from the column (Figure 3). These proteins both eluted at 10.2 ml. The two peaks of an elution of milk are in the same places as these casein aggregate and whey protein peaks (Figures 2 and 3). Electrophoresis patterns of proteins in the first peak of elution with or without added calcium had dark casein bands and no whey protein bands. Electrophoresis of proteins in the whey protein peak from samples to which calcium had been added displayed clearly visible whey protein bands and faint casein bands. Calcium treatment had significant (~<.01) effects on elution volumes of casein and whey protein peaks, maximum absorbances of casein and whey protein peaks, areas of casein and whey protein peaks, ratio of casein peak maximum absorbance to whey protein peak maximum absorbance, and casein peak area as a Journal of Dairy Science Vol. 68, No. 2, 1985 310 CARPENTER A N D BROWN I.S .S - WHEY PRO'rEZNS I I~ .S I 18 IK I 18 l& VOLUME ( m L ) I.S CASEIN .S - I G .5 VOLUPIE C m L ) Figure 3. Protein elution curves of a-lactalbumin and t3-1actoglobulin (top) and casein (bottom). 90- 80- percentage of total area under the curve. Calcium treatment did n o t have a significant effect (a>.01) on the sum of maximum absorbances of both peaks or on the sum of the areas of both peaks. The casein peak eluted earlier, and the whey protein peak etuted later when calcium was added to the sample, Also, casein peaks increased in area and maximum absorbance, and whey protein peaks decreased in area and maximum absorbance. A comparison between casein peak areas as percentages of total areas under elution curves with respect to calcium addition and temperature treatments is in Figure 4. Calcium addition forced soluble casein into micetles so that temperature had less effect when calcium was added. Casein in the first peak as a percentage of total protein was between 78 and 80% when calcium was added, irrespective of temperature treatment. This is approximately the casein number of normal milk (3). The pH did not have a significant effect in any of the analyses of variance. This may be partly, due to the small pH range studied. Eight samples of milk from individual cows ranging in fat from 2.1 to 3.85% and in protein from 2.19 to 4,42% were run through the column. Elution plots were obtained, and characteristics of each plot were calculated. The NO CALCIUM CALCIUM ADDED 2.6- ~, 2.4- ~ ~-~\ \ - ~ %\\~, \ \ "-,\ \ \ ~\\-.~ i~ ~k kk kk k k kh 0 70 - \\ \\ \\ I ~\\~ I ...... \ \ \ \ \\ - - --~ z 2- i ~\\% ~\\~ ~\\~ %\\% 4 2.2- / \\ \\ ~ C,O ~ ! %\\'% 40 TEMPERATURE 7Z (C) Figure 4. Calcium addition and temperature treatment effects on percent area of first (casein) peak. Journal of Dairy Science Vol. 68, No. 2, 1985 // 1.8t 1.6 t6 t 1.8 i 2 i 2.2 I 2.4 ' - - I 2.6 ESIIMATEDCASEINcolumn Figure 5. Correlation of column and standard methods of casein estimation in milk. DAIRY FOODS TECHNICAL NOTE c o r r e l a t i o n c o e f f i c i e n t o f casein w i t h m a x i m u m a b s o r b a n c e o f t h e first p e a k was .77 a n d w i t h t h e area u n d e r t h e first p e a k was .65. C o r r e l a t i o n coefficients of casein w i t h m a x i m u m a b s o r b a n c e a n d area of t h e s e c o n d p e a k were .03 a n d .25. T h e r e l a t i o n s h i p b e t w e e n p e a k area a n d maxim u m a b s o r b a n c e o f t h e first peak w i t h p e r c e n t casein was d e m o n s t r a t e d b y analysis o f covariance. Stepwise regression i n d i c a t e d t h a t area a n d m a x i m u m a b s o r b a n c e of t h e first p e a k e s t i m a t e d casein w i t h a c o r r e l a t i o n c o e f f i c i e n t o f .92. E s t i m a t i o n o f casein b y this regression e q u a t i o n is p l o t t e d versus s t a n d a r d casein m e a s u r e m e n t s in Figure 5. CONCLUSIONS G l y c o p h a s e c o a t e d p o r o u s glass b e a d s w i t h 37 to 7 4 / 2 m particle size a n d 2 0 0 0 n m p o r e d i a m e t e r in a 1 0 0 b y .4 cm c o l u m n can b e u s e d to separate casein micelles f r o m w h e y p r o t e i n s b y size e x c l u s i o n c h r o m a t o g r a p h y . Best separat i o n o f t o t a l casein f r o m w h e y p r o t e i n in milk was a c h i e v e d b y a d d i t i o n o f 100 /21 o f 5 M CaC12/50-ml sample 24 h b e f o r e testing. This could b e a d d e d as a t a b l e t at t h e same t i m e as preservative. Casein c a n b e e s t i m a t e d f r o m m a x i m u m a b s o r b a n c e a n d u r e a o f t h e first peak e l u t e d w h e n m o n i t o r i n g at a w a v e l e n g t h o f 2 8 0 nm. S e p a r a t i n g casein micelles f r o m w h e y p r o t e i n s w i t h o u t c o a g u l a t i o n is a d v a n t a g e o u s for f u r t h e r t e s t i n g of t h e casein f r a c t i o n . Using t h e s e p a r a t i o n m e t h o d d e s c r i b e d in t h i s paper, casein m a y be m e a s u r e d d i r e c t l y b y i n f r a r e d milk testing i n s t r u m e n t s . ACKNOWLEDGMENTS T h e a u t h o r s t h a n k A. A. H e t h f o r assistance with the electrophoresis. REFERENCES 1 Association of Official Analytical Chemists. 1980. Official and tentative methods of analysis. 13th ed. Assoc. Of tic. Anal. Chem., Washington, DC. 2 Bingham, E. W., H. M. Farrell, Jr., and R. J. Carroll. 1972. Properties of dephosphorylated asl-casein. Biochemistry 11:2450. 3 Blake, R. W., I. B. Nmai, and R. L. Richter. 1980. Relationship between distribution of major milk proteins and milk yield. J. Dairy Sci. 63:141. 4 Bloomfield, V. A., C. V. Morr. 1973. Structure of casein micelles: physical methods. Neth. Milk Dairy J. 27:103. 5 Bloomfield, V. A., and R. J. Mead, Jr. 1975. Structure and stability of casein micelles. J. Dairy 311 Sci. 50:592. 6 Brown, R. J. 1981. Computerized cheese yield pricing of milk. 2nd Bienn. Marschall Int. Cheese Conf, Madison, WI. 7 Eckstrand, B., M. Larsson-Raznikiewicz, E. Brannang, and C. Swensson. 1981. Size distribution of casein micelles related to coagulation properties: A comparison between different breeds of cattle. Swed. J. Agric. Res. 11:57. 8 Ernstrom, C. A. 1980. A workable pricing system based on cheese yield. Utah State Univ. 4th Bienn. Cheese Ind. Conf., Logan. 9 Fehrnstrom, H., and V. Moberg. 1977. SDS and conventional polyacrylamide gel electrophoresis with LKB 2117 multiphor. Application note 306. LKB Producter AB, Bromma, Sweden. 10 Gordon, W. G., and E. B. Kalan. 1974. Proteins of milk. Pages 8 7 - 1 2 4 in Fundamentals of dairy chemistry. 2nd ed. B. H. Webb, A. H. Johnson, and J. A. Alford, ed. Avi Publ. Co., Inc., Westport, CT. 11 Heth, A. A., and H. E. Swaisgood. 1982. Examination of casein micelle structure by a method for reversible covalent immobilization. J. Dairy Sci. 65 : 2047. 12 Jenness, R. 1970. Protein composition of milk. Milk proteins. Vol. 1. H. A. McKenzie, ed. Academic Press, New York, NY. 13 Jenness, R., and J. Koops. 1962. Preparation of a salt solution which simulates milk ultrafiltrate. Neth. Milk Dairy J. 16:153. 14 Kiddy, C. A. 1974. Gel electrophoresis in vertical polyacrylamide beds. Pages 1 4 - 1 5 in Methods of gel electrophoresis of milk proteins. H. E. Swaisgood, ed. Dep. Food Sci., North Carolina State Univ., Raleigh. 15 Larson, B. L., G. D. Rolleri, and K. A. Kendall. 1956. Protein production in the bovine. Comparison of daily protein, fat, and milk production during the entire lactation period. J. Dairy Sci. 39:204. 16 McGann, T.C.A., R. D. Kearney, and W. J. Donnelly. 1979. Developments in column chromatography for the separation and characterization of casein micelles. J. Dairy Res. 46: 307. 17 McMahon, D. J., and R. J. Brown. 1984. Composition, structure and integrity of casein micelles: A review. J. Dairy Sci. 67:499. 18 Ono, T., S. Odagiri, and T. Takagi. 1983. Separation of submicelles from micellar casein by high performance liquid chromatography on a TSK-Gel G40OOSW column. J. Dairy Res. 50:37. 19 Rowland, S. J. 1938. The determination of the nitrogen distribution in milk. J. Dairy Res. 9:42. 20 Waite, R., J.C.D. White, and A. Robertson. 1956. Variations in the chemical composition of milk with particular reference to solids-not-fat 1. The effect of stage of lactation, seasOn on year and age of cow. J. Dairy Res. 23:65. 21 Yaguchi, M., and D. Rose. 1971. Chromatographic separation of milk proteins: a review. J. Dairy Sci. 54:1725. 22 Yau, W. W., J. J. Kirkland, and D. D. Bly. 1979. Modern size exclusion liquid chromatography: Practice of gel permeation and gel filtration chromatography. John Wiley & Sons, Inc., New York, NY. - - Journal of Dairy Science Vol. 68, No. 2, 1985