Isolation and Partial Characterization of Rat Casein Proteins ~ R. M. M e K E N Z l E and B. L. L A R S O N Department of Dairy Science University of Illinois Urbana 61601
ABSTRACT
Casein was isolated from rat milk by high speed centrifugation. Polyacrylamide disc gel electrophoresis of the whole casein yielded three major protein zones designated C.1, C.2, and C.3 in order of their decreasing electrophorefic mobility in the alkaline system. Zone 3 subsequently contained two possibly related bands, C.3.1 and C.3.2. The presence of phosphate in all four zones was indicated by staining and conformed by phosphorus-32 labeling studies. A glycoprotein character was indicated by all zones. Separation of the constituents of rat casein by diethylaminoethyl-cellulose ion exchange chromatography yielded the same four major protein entities. Three milk-specific phosphoproteins unique to rat whey eluted from such columns in the same general region as the casein constituents but appear to be otherwise unrelated to the four major components of micellar casein. Gel etectrophoresis in sodium dodecyl sulfate systems yielded apparent molecular weight estimates of approximately 24,000 for C.1, 38,000 for C.2, and 28,000 for C.3.1 and C.3.2. INTRODUCTION
Although there are many similarities, comparative analyses o f milks from different species have revealed large variations in the common milk constituents and widely differing chemical and physical properties of related
Received November 28, 1977. ~This work was supported under Contract US NO1-CB-23856 of the National Cancer Institute, National Institutes of Health, and Hatch Projects 35-320 and 35-351 of the Illinois Agricultural Experiment Station.
1978 J Dairy Sci 61:885-889
constituents (1, 4, 5, 6, 15). The latter is particularly true of that group of proteins which comprise the casein of milk. Casein has been characterized in virtually all milks by its precipitability at pH 4 to 5 and a high content of phosphorylated hydroxyamino acids with bound calcium. It exists in milk as a highmolecular-weight complex of several constituent proteins which differ markedly both within and between species (5, 8, 15). The bovine casein system has been studied most extensively; however, recent emphasis on mammary cancer has focused increasing attention on man and the use of rat mammary tissue and milk as a test system. In a comparative analysis of caseins in the milks of various species, Sloan et al. (14), using paper electrophoresis, found only one band in rat casein. Subsequently, Jenness (4) utilized starch gel electrophoresis and demonstrated several proteins with a compositional pattern dissimilar to that with caseins from other species. Recently, Rosen et al. (13) and Woodward (17) have reported studies on the separation of components from acid-precipitated rat casein with DEAE-cellulose ion exchange chromatography and gel electrophoresis. This report presents data on the isolation, fractionation, and partial characterization of rat milk casein. Other recent reports have concerned the a-lactalbumins (9) and whey phosphoprotein constituents (10) of rat milk. MATERIALS AND METHODS
Milk was obtained from rats of the Fischer 344 (or CDF) strain as in (9). Casein was obtained by one of two methods. High speed centrifuged casein was acquired by centrifugation of diluted skim milk at 40,000 × g for 90 to 120 rain at 4 C (9). Acid casein was obtained by adjusting the pH of the skim milk to 4.5 with 1 N HCI followed by centrifugation at 10,000 × g for 10 rain.
885
886
McKENZIE AND LARSON
Polyacrylamide Disc Gel Electrophoresis
Electrophoresis normally was on 6 mm x 75 mm, 7.5% polyacrylamide gels (Cyanogum41, Fischer Scientific Co., Chicago, IL) with the Tris-EDTA-borate buffer system described by Thompson et ai. (16). Casein samples were dissolved in a solution o f 20% sucrose, .3% mercaptoethanol, 7 M urea, and a minute amount of bromophenol blue. Coomassie blue was a general protein strain (2). Specific staining of phosphoproteins was accomplished by the procedure of Cutting and Roth (3). Staining with periodic acid-Schiff's reagent (PAS) was used to test for apparent glycoprotein character (11). When the polyacrylamide gels were analyzed for their 32p content, they were frozen, sliced into 1 mm sections, and counted as described (10). Molecular weights were determined on 10% polyacrylamide gels in the presence of sodium dodecyl sulfate (SDS) in a continuous phosphate buffer system as described by Maizel (7). Samples were dispersed in a solution of 1% SDS, .01 M sodium phosphate, 1% mercaptoethanol, and 10% glycerol. The apparent molecular weight values were confirmed by 13% gels and the SDS disc system also described by Maizel (7).
changes of .01 M sodium phosphate buffer, pH 7.0, and distilled water. Samples to be used for disc gel electrophoresis and DEAE-cellulose chromatography then were equilibrated against the appropriate buffers. The 32p contents of the various fractions were determined by scintillation counting as in (10).
RESULTS A N D DISCUSSION
The gel electrophoresis patterns of rat skim milk, high-speed casein, and whole whey are in Fig. 1. Three major zones, designated C.1, C.2, and C.3, in order of their decreasing electrophoretic mobilities, were observed in the highspeed casein preparations. Zone C.3 subsequently was composed of two barely resolvable zones that have been subclassified as C.3.1. and C.3.2. The supernatant whole whey contained small amounts of the casein components not removed in the centrifugation process. Casein samples isolated from the milks of several individual rats showed no variation in their electrophoretic patterns or in relative intensities of the major zones. Storage by freez-
i
2
CASEIN
SKIM MILK
DEAE-Cellulose Ion Exchange Chromatography
Whole casein was fractionated by chromatography on DEAE-cellulose ion exchange columns (Whatman DE-52, Whatman Corp., Clifton, NJ) equilibrated with 10 mM imidazole, 3.3 M urea, and 10 mM mercaptoethanol adjusted to pH 7.0 with HC1 according to Ribadeau-Dumas et al. (12). Casein samples were dissolved, equilibrated with column buffer by dialysis, and applied to a 2.8 x 30 cm column. A linear gradient (of 400 x 400 ml) from 0 to 300 mM NaC1 in column buffer was utilized to develop the column using a pumped flow of approximately 60 ml/h. 3 2 p Incorporation Studies
Casein labeled with 32p was produced in vivo in rats as detailed previously (10). The once sedimented, high speed casein pellet obtained in these experiments was resuspended in water and dialyzed successively against several Journal of Dairy Science Vol. 61, No. 7, 1978
3 WHOLE WHEY 7- HIGH MOLECULAR _[ WEIGHT COMPONENTS
C.3.1,32 C2 C.I
l SERUM ALBUMIN d.- LACTALBUMINS PHOSPHOPROTEINS P.I- P,4
M
® FIG. 1. Diagrammatic representation of polyacrylamide disc gel electrophoretic patterns of: 1) highspeed centrifuged rat casein, 2) rat skim milk and 3) centrifuged whole rat whey. The analyses were conducted under the conditions described by Thompson et al. (16) using urea and mercaptoethanol in a TrisEDTA-borate buffer system.
RAT CASEIN
ing at - 2 0 C was without notable effect on the results. Results for casein prepared by acid precipitation are not reported here other than some general comments. This casein contained the same three major protein zones as did the centrifuged casein but was contaminated more heavily with whey proteins. Several cycles of redissolving by neutralization and reprecipitation at acid pH were necessary to reduce these contaminants to trace amounts, procedures which were time-consuming and wasteful of product. Whole rat casein by high speed centrifugation was essentially a pure family of proteins. To evaluate its purity, a 30-ml sample of rat milk was processed with high speed centrifugation as described in Methods. The casein pellet was rinsed with, and then resuspended in, 30 ml of deionized water. After removal of a .1-ml aliquot, the casein was resedimented. This process was repeated until the casein pellet had been resuspended and sampled five times. Each .1 ml aliquot then was dissolved in 50 mM citrate buffer, pH 7.0. The absorbance of the casein solutions was determined at 280 nm, and each sample was adjusted to unit absorbancy with buffer. Equal volumes of each sample then were loaded onto polyacrylamide gels and etectrophoresed as described in Methods. The first washing step removed only minor contaminating whey bands in the once-sedimerited casein pellet. The subsequent washes resulted in a total apparent loss of about 20% of the A280 absorbing material. However, no discernable change in the electrophoresis patterns accompanied the loss in protein which was attributed to soluhilization of the casein micelle per se. The electrophoresis pattern obtained with samples of washed high speed casein was qualitatively the same whether stained with Coomassie blue or PAS stain for carbohydrate. The presence of phosphate in the three zones was demonstrated with a specific phosphate staining procedure (3). Thus, all of the major constituents of rat casein appeared to be both phosphoprotein and glycoprotein in character. Woodward and Messer (18) have reported the glycoprotein character of acid-precipitated whole rat casein. The phosphoprotein character of all of the casein protein components was confirmed by
887
studies of [32p] casein preparations obtained from rats injected in vivo with inorganic [32p1 phosphate (10). Fig. 2 depicts the distribution of the 32p in the rat casein relative to the Coomassie blue staining pattern on the same gel. Incubation of unlabeled rat milk with inorganic [32p]phosphate did not label any casein component (10). The elution pattern obtained by chromatography of this [3 2 P] casein preparation on DEAE-cellulose is in Fig. 3 with the peaks identified with respect to their electrophoretic character. Approximately 95% of the 32p eluted in association with the four major protein peaks. The distribution of 32p label among the four fractions was approximately 25% in C.1, 45% in C.2, 15% in C.3.1, and 10% in C.3.2. The minimum apparent molecular weights of the major casein protein peaks were determined by SDS disc gel electrophoresis. A SDS system using a continuous phosphate buffer and 10% gels yielded estimates of 24,000 for C.1, 38,000 for C.2, and 28,000 for C.3.1 and C.3.2. Essentially the same values (25,000, 40,000 and 28,000) were acquired with 13% gels and a discontinuous Tris-glycine-SDS system (7).
~
o I
-'7
• H
I
p
C.2
m
6
n
f
_o 5 x
C.5
E
~4
m
¢t N
D
2
i
I
L_ I
o Io
20 SLICE N0,
I
I
50
40
FIG. 2. Distribution of 32p relative to the Coornassie blue staining pattern of 32P_labeled rat casein analyzed by disc gel electrophoresis. Journal of Dairy Science Vol. 61, No. 7, 1978
888
McKENZIE AND LARSON I
t2°it 2
t
ABSTRACT
Casein was isolated from rat milk by high speed centrifugation. Polyacrylamide disc gel electrophoresis of the whole casein yielded three major protein zones designated C.1, C.2, and C.3 in order of their decreasing electrophorefic mobility in the alkaline system. Zone 3 subsequently contained two possibly related bands, C.3.1 and C.3.2. The presence of phosphate in all four zones was indicated by staining and conformed by phosphorus-32 labeling studies. A glycoprotein character was indicated by all zones. Separation of the constituents of rat casein by diethylaminoethyl-cellulose ion exchange chromatography yielded the same four major protein entities. Three milk-specific phosphoproteins unique to rat whey eluted from such columns in the same general region as the casein constituents but appear to be otherwise unrelated to the four major components of micellar casein. Gel etectrophoresis in sodium dodecyl sulfate systems yielded apparent molecular weight estimates of approximately 24,000 for C.1, 38,000 for C.2, and 28,000 for C.3.1 and C.3.2. INTRODUCTION
Although there are many similarities, comparative analyses o f milks from different species have revealed large variations in the common milk constituents and widely differing chemical and physical properties of related
Received November 28, 1977. ~This work was supported under Contract US NO1-CB-23856 of the National Cancer Institute, National Institutes of Health, and Hatch Projects 35-320 and 35-351 of the Illinois Agricultural Experiment Station.
1978 J Dairy Sci 61:885-889
constituents (1, 4, 5, 6, 15). The latter is particularly true of that group of proteins which comprise the casein of milk. Casein has been characterized in virtually all milks by its precipitability at pH 4 to 5 and a high content of phosphorylated hydroxyamino acids with bound calcium. It exists in milk as a highmolecular-weight complex of several constituent proteins which differ markedly both within and between species (5, 8, 15). The bovine casein system has been studied most extensively; however, recent emphasis on mammary cancer has focused increasing attention on man and the use of rat mammary tissue and milk as a test system. In a comparative analysis of caseins in the milks of various species, Sloan et al. (14), using paper electrophoresis, found only one band in rat casein. Subsequently, Jenness (4) utilized starch gel electrophoresis and demonstrated several proteins with a compositional pattern dissimilar to that with caseins from other species. Recently, Rosen et al. (13) and Woodward (17) have reported studies on the separation of components from acid-precipitated rat casein with DEAE-cellulose ion exchange chromatography and gel electrophoresis. This report presents data on the isolation, fractionation, and partial characterization of rat milk casein. Other recent reports have concerned the a-lactalbumins (9) and whey phosphoprotein constituents (10) of rat milk. MATERIALS AND METHODS
Milk was obtained from rats of the Fischer 344 (or CDF) strain as in (9). Casein was obtained by one of two methods. High speed centrifuged casein was acquired by centrifugation of diluted skim milk at 40,000 × g for 90 to 120 rain at 4 C (9). Acid casein was obtained by adjusting the pH of the skim milk to 4.5 with 1 N HCI followed by centrifugation at 10,000 × g for 10 rain.
885
886
McKENZIE AND LARSON
Polyacrylamide Disc Gel Electrophoresis
Electrophoresis normally was on 6 mm x 75 mm, 7.5% polyacrylamide gels (Cyanogum41, Fischer Scientific Co., Chicago, IL) with the Tris-EDTA-borate buffer system described by Thompson et ai. (16). Casein samples were dissolved in a solution o f 20% sucrose, .3% mercaptoethanol, 7 M urea, and a minute amount of bromophenol blue. Coomassie blue was a general protein strain (2). Specific staining of phosphoproteins was accomplished by the procedure of Cutting and Roth (3). Staining with periodic acid-Schiff's reagent (PAS) was used to test for apparent glycoprotein character (11). When the polyacrylamide gels were analyzed for their 32p content, they were frozen, sliced into 1 mm sections, and counted as described (10). Molecular weights were determined on 10% polyacrylamide gels in the presence of sodium dodecyl sulfate (SDS) in a continuous phosphate buffer system as described by Maizel (7). Samples were dispersed in a solution of 1% SDS, .01 M sodium phosphate, 1% mercaptoethanol, and 10% glycerol. The apparent molecular weight values were confirmed by 13% gels and the SDS disc system also described by Maizel (7).
changes of .01 M sodium phosphate buffer, pH 7.0, and distilled water. Samples to be used for disc gel electrophoresis and DEAE-cellulose chromatography then were equilibrated against the appropriate buffers. The 32p contents of the various fractions were determined by scintillation counting as in (10).
RESULTS A N D DISCUSSION
The gel electrophoresis patterns of rat skim milk, high-speed casein, and whole whey are in Fig. 1. Three major zones, designated C.1, C.2, and C.3, in order of their decreasing electrophoretic mobilities, were observed in the highspeed casein preparations. Zone C.3 subsequently was composed of two barely resolvable zones that have been subclassified as C.3.1. and C.3.2. The supernatant whole whey contained small amounts of the casein components not removed in the centrifugation process. Casein samples isolated from the milks of several individual rats showed no variation in their electrophoretic patterns or in relative intensities of the major zones. Storage by freez-
i
2
CASEIN
SKIM MILK
DEAE-Cellulose Ion Exchange Chromatography
Whole casein was fractionated by chromatography on DEAE-cellulose ion exchange columns (Whatman DE-52, Whatman Corp., Clifton, NJ) equilibrated with 10 mM imidazole, 3.3 M urea, and 10 mM mercaptoethanol adjusted to pH 7.0 with HC1 according to Ribadeau-Dumas et al. (12). Casein samples were dissolved, equilibrated with column buffer by dialysis, and applied to a 2.8 x 30 cm column. A linear gradient (of 400 x 400 ml) from 0 to 300 mM NaC1 in column buffer was utilized to develop the column using a pumped flow of approximately 60 ml/h. 3 2 p Incorporation Studies
Casein labeled with 32p was produced in vivo in rats as detailed previously (10). The once sedimented, high speed casein pellet obtained in these experiments was resuspended in water and dialyzed successively against several Journal of Dairy Science Vol. 61, No. 7, 1978
3 WHOLE WHEY 7- HIGH MOLECULAR _[ WEIGHT COMPONENTS
C.3.1,32 C2 C.I
l SERUM ALBUMIN d.- LACTALBUMINS PHOSPHOPROTEINS P.I- P,4
M
® FIG. 1. Diagrammatic representation of polyacrylamide disc gel electrophoretic patterns of: 1) highspeed centrifuged rat casein, 2) rat skim milk and 3) centrifuged whole rat whey. The analyses were conducted under the conditions described by Thompson et al. (16) using urea and mercaptoethanol in a TrisEDTA-borate buffer system.
RAT CASEIN
ing at - 2 0 C was without notable effect on the results. Results for casein prepared by acid precipitation are not reported here other than some general comments. This casein contained the same three major protein zones as did the centrifuged casein but was contaminated more heavily with whey proteins. Several cycles of redissolving by neutralization and reprecipitation at acid pH were necessary to reduce these contaminants to trace amounts, procedures which were time-consuming and wasteful of product. Whole rat casein by high speed centrifugation was essentially a pure family of proteins. To evaluate its purity, a 30-ml sample of rat milk was processed with high speed centrifugation as described in Methods. The casein pellet was rinsed with, and then resuspended in, 30 ml of deionized water. After removal of a .1-ml aliquot, the casein was resedimented. This process was repeated until the casein pellet had been resuspended and sampled five times. Each .1 ml aliquot then was dissolved in 50 mM citrate buffer, pH 7.0. The absorbance of the casein solutions was determined at 280 nm, and each sample was adjusted to unit absorbancy with buffer. Equal volumes of each sample then were loaded onto polyacrylamide gels and etectrophoresed as described in Methods. The first washing step removed only minor contaminating whey bands in the once-sedimerited casein pellet. The subsequent washes resulted in a total apparent loss of about 20% of the A280 absorbing material. However, no discernable change in the electrophoresis patterns accompanied the loss in protein which was attributed to soluhilization of the casein micelle per se. The electrophoresis pattern obtained with samples of washed high speed casein was qualitatively the same whether stained with Coomassie blue or PAS stain for carbohydrate. The presence of phosphate in the three zones was demonstrated with a specific phosphate staining procedure (3). Thus, all of the major constituents of rat casein appeared to be both phosphoprotein and glycoprotein in character. Woodward and Messer (18) have reported the glycoprotein character of acid-precipitated whole rat casein. The phosphoprotein character of all of the casein protein components was confirmed by
887
studies of [32p] casein preparations obtained from rats injected in vivo with inorganic [32p1 phosphate (10). Fig. 2 depicts the distribution of the 32p in the rat casein relative to the Coomassie blue staining pattern on the same gel. Incubation of unlabeled rat milk with inorganic [32p]phosphate did not label any casein component (10). The elution pattern obtained by chromatography of this [3 2 P] casein preparation on DEAE-cellulose is in Fig. 3 with the peaks identified with respect to their electrophoretic character. Approximately 95% of the 32p eluted in association with the four major protein peaks. The distribution of 32p label among the four fractions was approximately 25% in C.1, 45% in C.2, 15% in C.3.1, and 10% in C.3.2. The minimum apparent molecular weights of the major casein protein peaks were determined by SDS disc gel electrophoresis. A SDS system using a continuous phosphate buffer and 10% gels yielded estimates of 24,000 for C.1, 38,000 for C.2, and 28,000 for C.3.1 and C.3.2. Essentially the same values (25,000, 40,000 and 28,000) were acquired with 13% gels and a discontinuous Tris-glycine-SDS system (7).
~
o I
-'7
• H
I
p
C.2
m
6
n
f
_o 5 x
C.5
E
~4
m
¢t N
D
2
i
I
L_ I
o Io
20 SLICE N0,
I
I
50
40
FIG. 2. Distribution of 32p relative to the Coornassie blue staining pattern of 32P_labeled rat casein analyzed by disc gel electrophoresis. Journal of Dairy Science Vol. 61, No. 7, 1978
888
McKENZIE AND LARSON I
t2°it 2
t
