Proteolysis of Milk Proteins During Involution of the Bovine Mammary Gland1

Proteolysis of Milk Proteins During Involution of the Bovine Mammary Gland1 M. ASLAM and W. L. HURLEY2 Department of Animal Sciences, University of Illinois, Urbana 61801

ABSTRACT The role of proteolytic hydrolysis of milk proteins in the mammary gland during involution is unknown. The objectives of this study were to determine the activities of plasmin, plasminogen, and plasminogen activator in mammary gland secretions collected during involution and to identify peptides generated by proteolysis of casein and lactoferrin in those secretions. Mammary secretions were collected from Holstein cows on d 7, 14, and 21 of involution and on d 7 postcalving. Activities of plasmin, plasminogen, and plasminogen activator were determined on the defatted, filtered aqueous phase of mammary secretions. Activities of plasmin, plasminogen, and plasminogen activator were significantly higher on d 7, 14, and 21 of involution than were those on d 7 postcalving. Protein fragments resulting from hydrolysis were detected by SDS-PAGE in samples collected on d 7, 14, and 21 of involution, but few protein fragments were observed in samples collected on d 7 postcalving when plasmin activity was low. Immunoblot analysis showed that a number of peptides observed during involution were generated from as-casein (CN), bCN, k-CN, or lactoferrin. The appearance of peptides from proteins of mammary secretions during early involution was generally correlated with increased plasmin activity. Elevated plasmin activity during mammary involution may be primarily responsible for the observed concurrent hydrolysis of milk proteins in mammary secretions. ( Key words: involution, milk proteins, peptides, plasmin) Abbreviation key: Lf = lactoferrin, PA = plasminogen activator. INTRODUCTION A nonlactating period for the dairy cow is required for optimal milk production during the subsequent

Received July 22, 1996. Accepted March 17, 1997. 1Supported by Illinois Agricultural Experiment Station Projects 35-0315 and 35-0354. 2To whom correspondence should be addressed. 1997 J Dairy Sci 80:2004–2010

lactation ( 6 ) . The transition period between lactating and nonlactating states of the mammary gland is a period of active involution during which the mammary gland undergoes extensive ultrastructural changes and the secretions contained in the gland undergo dramatic compositional changes (13). Histological changes in bovine mammary tissue include appearance of large vacuoles in the secretory cells, reduced rough endoplasmic reticulum and Golgi apparatus, and increased stromal tissue between alveoli (12). Compositional changes in mammary secretions include a decrease in milk-specific proteins such as casein, a-LA, and b-LG ( 3 ) . Increased permeability of the mammary epithelium during early involution results in the transfer of serum proteins, such as serum albumin and immunoglobulins, into the gland (13). Concentrations of lactoferrin ( Lf) increase significantly in mammary gland secretions during the first weeks of involution (24). During lactation, milk contains a number of proteases such as plasmin, milk acid protease, and proteases that are derived from leukocytes in the milk (14, 20, 25). Components of the blood fibrinolytic system, such as plasminogen, plasminogen activator ( PA) , plasmin, plasmin inhibitor, and PA inhibitor, are present in milk (11, 22). Plasmin is a serine protease that is secreted as the zymogen, plasminogen. Plasminogen is converted into active plasmin by the action of PA. Most proteolytic activity found in milk is stimulated by plasmin (2, 9, 22). Components of the plasmin system are often associated with casein micelles but are also present in the serum and cream phases of milk (15, 20). Plasmin inhibitors are present in the serum phase of the milk as well (15). Plasminogen activators present in milk can be either tissue or urokinase types ( 7 ) . These proteolytic activities are particularly relevant to milk quality and processing characteristics. The plasmin system is thought to have a role in the mammary gland during involution. In the rodent mammary gland, PA converts plasminogen to plasmin during late lactation, which is associated with the onset of involution (19). Concentrations of plasminogen and plasmin increase in bovine milk as lactation progresses, and plasminogen activity increases further by d 3 after drying off (21). Politis et al. ( 2 2 )

2004

PROTEOLYSIS OF MILK PROTEIN DURING INVOLUTION

have suggested that the increased plasmin activity in late lactation may be involved in subsequent mammary gland involution. The relationship between increased proteolytic activity in mammary secretions during involution and the fate of milk proteins in secretions has not been determined. Profiles of proteins in mammary secretions change throughout the dry period (13). Hydrolytic fragments of milk proteins, particularly casein and Lf, have been identified in mammary secretions collected during involution of the bovine mammary gland ( 3 ) . Aslam et al. ( 3 ) have previously suggested that plasmin might be responsible for the hydrolysis of milk proteins during involution. Little information is available about the effects of the plasmin system on milk proteins in the mammary gland during involution or about the fate of milk proteins during involution. The objectives of this study were to determine the activities of plasmin, plasminogen, and PA in mammary secretions during involution of the bovine mammary gland and to identify hydrolyzed fragments of milk proteins resulting from proteolytic cleavage. The correlation between the presence of elevated enzyme activities and the appearance of peptides that are derived from milk protein suggests a cause and effect relationship. MATERIALS AND METHODS Materials The PhastSystem, PhastGel high density homogenous gels (code number 17–0679–0), and SDS buffer strips (code number 17–0516–0) were obtained from Pharmacia LKB Biotechnology (Piscataway, NJ). Goat anti-rabbit IgG alkaline phosphatase conjugate was from Bio-Rad Laboratories (Richmond, CA). Alkaline phosphatase substrates (5-bromo-4chloro-3-indolyl phosphate and nitroblue tetrazolium) were from BRL Laboratories (Gaithersburg, MD). Sigma Chemical Co. (St. Louis, MO) was the supplier of substrate S-2251 (H-D-valyl-L-leucyl-Llysine-P-nitroanilide dihydrochloride), plasminogen, and low molecular mass standards. Urokinase was purchased from American Diagnostica, Inc. (Greenwich, CT). Nunc-Immuno Plate MaxiSorp F96, flatbottomed microtiter plates were from Nunc (Roskilde, Denmark). Sample Collection and Preparation Mammary secretions were collected from Holstein cows on d 7, 14, and 21 after drying off ( n = 8/d) and on d 7 postcalving ( n = 5). Milk collected on d 7 postcalving was chosen to represent milk that was

2005

generally unaffected by proteolytic activity for comparison with mammary secretions during involution. Individual quarters were screened for bacterial contamination according to the procedure described by the National Mastitis Council (18). Samples were used from quarters that contained no bacteria. Cows had previously been given dry cow antibiotic treatment. Dry cows were pregnant at the time of sample collection, mean parity was 2.3 ± 0.3 lactations, mean DIM at drying off was 346 ± 15 d, and mean production at drying off was 14.8 ± 0.5 kg/d. Secretions were centrifuged at 10,000 × g for 25 min at 4°C, and the aqueous phase was collected. Secretions were passed through a series of filters from 3.3 to 0.45 mm and stored at –80°C. Enzyme activities were stable for at least 1 yr under these conditions. Identification of Hydrolyzed Fragments of Milk Proteins Proteins in samples were separated on SDS-PAGE using PhastGel high density and PhastGel SDS buffer strips to identify hydrolyzed fragments of milk proteins during involution. The PhastGel high density electrophoresis system consists of a homogeneous gel with a stacking gel zone of 7.5% T and 2% C and a continuous separating gel zone of 20% T and 2% C where T = total concentration of acrylamide, and C = amount of crosslinker in the monomer mixture. Molecular mass markers from 2.5 to 16 kDa were loaded onto each gel. Gels were stained by silver staining to visualize protein bands. Immunoblotting Proteins (10 mg per lane) were separated on SDSPAGE ( 1 6 ) and transferred to nitrocellulose membranes. Immunoblotting procedures were as described previously (30). Primary polyclonal antisera (generously provided by A. J. Guidry, Beltsville, MD) that were used for immunoblotting casein included rabbit anti-bovine as-CN, rabbit anti-bovine b-CN, and rabbit anti-bovine k-CN. Specificity of these antisera was confirmed by immunoblotting against milk caseins and by double diffusion (Ouchterlony) against milk proteins. Rabbit anti-bovine Lf was as described previously (24). The secondary antibody was goat anti-rabbit IgG conjugated with alkaline phosphatase. Determination of Plasmin, Plasminogen, and PA Activities Activities of plasmin, plasminogen, and PA were determined using a modification of the method described by Stelwagen et al. (28). Briefly, plasminoJournal of Dairy Science Vol. 80, No. 9, 1997

2006

ASLAM AND HURLEY

gen and PA activities were defined as the plasmin activity that was generated after the addition of 30 Plough units of urokinase and 50 mg/ml of plasminogen, respectively. Plasmin activity was measured without added urokinase or plasminogen. Activities of plasmin, plasminogen, and PA were determined in flat-bottomed 96-well microtiter plates. The reaction mixture consisted of 50 mM Tris buffer (pH 7.4), 0.6 mM substrate S-2251, and 10 ml of sample. Controls in this assay included buffer alone, buffer plus substrate S-2251, and buffer plus sample alone. All samples were run in duplicate. Absorbance at 405 nm was measured at 30-min intervals and was linear for up to 3 h. Rate of P-nitroaniline formation was calculated from the linear portion of the absorbance versus time curve. Plasminogen and PA activities were calculated by subtracting endogenous plasmin activity. One unit of plasmin, plasminogen, or PA activity was defined as the amount of enzyme that produced a change in absorbance of 0.001 at 405 nm in 1 min at 37°C at pH 7.4 (15). Interassay and intraassay variations were 10% for plasmin, plasminogen, and PA assays. Statistical Analysis Statistical analysis was performed using the general linear models procedure of SAS (26). Data were analyzed by using a completely randomized design, and least squares means were used to estimate the differences among samples collected on d 7, 14, and 21 of involution and on d 7 postcalving. Plasmin, plasminogen, and PA were the independent variables used, and one measurement was made on each cow. Significance was declared at P < 0.05. RESULTS Plasmin activity was higher ( P < 0.05) on d 7, 14, and 21 of involution than on d 7 postcalving (Figure 1). Plasmin activity was higher ( P < 0.05) on d 7 of involution than on d 21 of involution. Plasminogen and PA activities were also higher ( P < 0.05) on d 7, 14, and 21 of involution than on d 7 postcalving (Figure 1). Plasminogen and PA activities were not different ( P > 0.05) among d 7, 14, and 21 after drying off. Relative activities of plasminogen and PA were generally higher than plasmin activity on d 7, 14, and 21 after drying off and on d 7 postcalving. Intact bands of major milk proteins, such as casein, a-LA, and b-LG, were observed by SDS-PAGE in milk on d 7 postcalving (Figure 2). Peptide bands < 14 kDa were not observed in samples on d 7 postcalving. However, in samples from d 7, 14, and 21 of involution, several peptides with lower molecular masses ranging from approximately 13.5 to 2.5 kDa were Journal of Dairy Science Vol. 80, No. 9, 1997

Figure 1. Activities of plasmin, plasminogen, and plasminogen activator ( P A ) in mammary secretions collected on d 7 postcalving and on d 7, 14, and 21 of involution. The activity of each enzyme was different ( P < 0.05) between samples from d 7 postcalving ( n = 5 cows) and samples from d 7, 14, and 21 of involution ( n = 8 cows/ d). Plasmin activity was higher ( P < 0.05) on d 7 of involution than on d 21 of involution; other enzyme activities on d 7, 14, and 21 of involution were not different. Values are means and standard errors of the means. One unit is defined as the amount of enzyme that produced a change in absorbance of 0.001 at 405 nm in 1 min at 37°C at pH 7.4. Bars to the right of the dotted line represent days of involution.

PROTEOLYSIS OF MILK PROTEIN DURING INVOLUTION

observed in mammary secretions (Figure 2), which suggests that hydrolysis of milk proteins was occurring. In addition, the increased appearance of bands with larger molecular masses (> 14 kDa) on d 7, 14, and 21 of involution compared with those on d 7 postcalving suggests that some of the proteins were present in partially hydrolyzed form during that period of involution. Immunoblot analysis confirmed that some of the peptides observed on d 7, 14, and 21 of involution were generated from as-CN, b-CN, or k-CN (Figure 3 ) or from Lf (Figure 4). Only a few peptides were observed in samples collected on d 7 postcalving, and those peptides were generated from casein (Figure 3). No peptides originating from Lf were observed on d 7 postcalving (Figure 4). Immunoblot analysis for a-LA and b-LG (data not shown) did not show bands that represented hydrolyzed fragments of those proteins in samples collected during involution, which suggests that these major proteins from milk whey were generally resistant to proteolytic hydrolysis during involution. DISCUSSION The appearance of peptides from mammary secretion proteins during involution generally correlated with increased plasmin activity. Activities of plasmin, plasminogen, and PA were significantly elevated during active involution ( d 7, 14, and 21 of involution), which was concurrent with the appearance of discrete peptides resulting from hydrolysis of mammary secretion proteins. In contrast, relatively few peptides derived from milk proteins were observed on SDS gels or immunoblots in samples collected on d 7 postcalving, corresponding to a time when plasmin activity was low. Immunoblot analysis confirmed that peptides produced from milk proteins on d 7, 14, and 21 of involution originated from as-CN, b-CN, k-CN, and Lf. The presence of a range of fragments of these milk proteins throughout involution is consistent with the limited hydrolysis that occurred during this period. Even by d 21 of involution, many bands of partially hydrolyzed milk proteins were present. Both a-LA and b-LG were relatively resistant to proteolytic cleavage in mammary secretions during involution because no fragments of these proteins were detected. The fate of several of the milk proteins trapped in the mammary gland after cessation of milk removal included at least limited hydrolysis to discrete peptides. Total activities of plasminogen and PA were higher than the total activity of active plasmin. Neither plasminogen nor PA should directly hydrolyze milk pro-

2007

Figure 2. High density SDS-PAGE of protein and peptide bands in mammary secretions from d 7 postcalving ( A ) , d 7 of involution ( B ) , d 14 of involution ( C ) , and d 21 of involution ( D ) . Lanes 1 through 5 represent samples from five different cows. Intact casein, a-LA, and b-LG are indicated in gel A. Brackets indicate several peptides <14 kDa that were present on d 7, 14, and 21 of involution. Units of molecular mass markers are shown (right side).

teins. The low ratio of active plasmin to inactive plasminogen was consistent with the limited hydrolysis of proteins present in the gland. Although present in small quantities, each of the major milk proteins was present in mammary secretions throughout the dry period (3, 13). Interestingly, the extent of protein hydrolysis, indicated by the presence of peptides, appeared to increase as involution progressed. In contrast, plasmin activity, although higher than that in milk, declined between d 7 and 21 of involution. This decline may reflect an extended process of protein hydrolysis and an accumulation of protein fragments in the secretion during involution. At this time, there is no evidence of a major contribution to proteolysis by other nonplasmin activities. Furthermore, the incomplete hydrolysis of milk proteins and the continued presence of discrete peptide fragments, even as late as d 21 of involution, suggest that other proteases do not cause significant further hydrolysis throughout that time period. Previously published observations ( 3 ) have suggested that few peptides are present in mammary secretions during the latter half of the dry period relative to the involution phase. The ultimate fate of peptides that are generated from milk proteins during involution is unknown. Caseins are particularly susceptible to hydrolysis by plasmin. Several peptides produced from b-CN and Journal of Dairy Science Vol. 80, No. 9, 1997

2008

ASLAM AND HURLEY

as-CN were identified by immunoblotting during involution when plasmin activity was significantly elevated. A number of peptides are generated by the action of plasmin on b-CN in milk, including g1-CN, g2-CN, and g3-CN fragments (2, 8, 9, 27). Aslam et al. ( 3 ) previously identified a 13-kDa peptide in mammary secretions through d 21 of involution. The N-terminal sequence indicated that this peptide was probably generated by plasmin digestion of b-CN; this peptide was probably the g2-CN fragment. Similarly,

Figure 4. Immunoblot of lactoferrin ( L f ) in mammary secretions. Lanes are pooled secretions representing d 7 postcalving and d 7, 14, and 21 of involution, respectively. Arrowhead indicates intact Lf bands. Units of molecular mass markers are shown (right side).

Figure 3. Immunoblots of as-CN, b-CN, and k-CN in mammary secretions. Lanes are pooled secretions representing d 7 postcalving and d 7, 14, and 21 of involution, respectively. Arrowheads indicate intact casein bands. Molecular mass markers are shown (right side). Journal of Dairy Science Vol. 80, No. 9, 1997

plasmin digestion of as-CN has been shown to generate peptides with molecular masses ranging from 5.5 to 20.5 kDa (1,9). In the present study, most fragments of as-CN that were detectable on the immunoblots were >14 kDa, which is consistent with the incomplete hydrolysis in the gland during this period of involution. The concentration of as-CN peptides with lower molecular mass might have been below the detection limits of these methods. Immunoblot analysis showed several peptides that were produced from k-CN during involution in the present study. Although some studies ( 9 ) have reported that k-CN is resistant to hydrolysis by plasmin, others ( 2 ) reported that in vitro hydrolysis of kCN with plasmin generates discrete peptides. The differences in results might be attributable to differences in in vitro experimental digestion conditions between these studies. Kaminogawa et al. ( 1 4 ) reported that an enzyme referred to as milk acid protease hydrolyzes k-CN and produces a specific peptide called para-k-CN. Identification of k-CN fragments indicated that this protein was indeed hydrolyzed during involution in the mammary gland; however, the specific nature of the enzyme or enzymes that hydrolyze k-CN remains to be determined. No peptides from Lf were observed in milk samples from d 7 postcalving when lower plasmin activity was observed. However, as with observations of casein hydrolysis, a number of peptides were produced from the hydrolysis of Lf during involution. This finding is consistent with an earlier observation of a Lf peptide with a lower molecular mass that was found on d 21 of involution ( 3 ) . In the present study, higher plasmin activity in mammary secretions during involu-

PROTEOLYSIS OF MILK PROTEIN DURING INVOLUTION

tion might have contributed to the appearance of Lf peptides with lower molecular masses. In vitro incubation of Lf with plasmin resulted in the generation of several peptides (23), and Lf is susceptible to hydrolysis by other proteolytic enzymes, such as pepsin and trypsin (5, 17). There was no evidence of peptides from whey proteins such as a-LA and b-LG during involution when plasmin activity was elevated or on d 7 postcalving. These results are in agreement with previous studies that reported that a-LA and b-LG were resistant to hydrolysis by plasmin (31). In fact, native and denatured b-LG can inhibit the activity of plasmin ( 4 ) . Inhibition of plasmin activity by b-LG that was present in the mammary secretions during early involution might have contributed to the apparent limited hydrolysis of intact casein bands. Politis et al. ( 2 1 ) previously reported that plasminogen activity in mammary secretions increased during early involution ( d 3 after drying off) compared with lactation. Results from the present study demonstrate that plasmin and PA activities are also elevated during involution and that each of these activities remains elevated throughout the period of active involution compared with early lactation. It is interesting that plasmin activity declines during involution despite the continued presence of elevated plasminogen and PA. Rate of conversion of plasminogen to plasmin may decline as involution progresses, contributing to the incomplete hydrolysis of milk proteins. In contrast, an increased conversion of plasminogen to plasmin in part may be responsible for increased plasmin activity observed in the later stages of lactation (8, 22). Peptides generated in the mammary gland during involution may modulate the physiology of mammary tissue. For example, peptides produced from caseins have been shown to possess a number of biological functions such as antimutagenic, opioid, and immunomodulatory activities (10, 29). Pepsin digestion of bovine Lf produces a peptide, lactoferricin B, which possesses broad-spectrum antibacterial activity ( 5 ) . Although most of these peptides are produced by hydrolysis by proteases other than plasmin, endogenous peptides produced in the mammary gland by the action of plasmin deserve further attention to explore their contribution to the antimicrobial activity and other bioactivities of secretions in the involuting mammary gland. CONCLUSIONS This study provided information about the proteolytic activity in mammary secretions collected during

2009

involution of the bovine mammary gland and on d 7 postcalving. We concluded that only partial hydrolysis of milk proteins occurred in mammary secretions during involution and that intact milk proteins and discrete peptide fragments of those proteins continued to be present in mammary secretions up to at least d 21 of involution. Elevated plasmin activity during involution apparently was responsible for the increased hydrolysis of casein and Lf that occurred during involution. The absence of fragments of a-LA and b-LG suggested that they were relatively resistant to proteolytic cleavage during involution. Protease activities other than those of plasmin did not seem to play a major role in protein hydrolysis during involution. Further studies are necessary to determine the fate of peptides generated during involution and to determine whether peptides generated from milk proteins during involution have any biological activity in the gland. This information provides the basis for further study of the role of the plasmin system in mammary physiology during the nonlactating period. REFERENCES 1 Aimutis, W. R., and W. N. Eigel. 1982. Identification of l-casein as plasmin-derived fragments of bovine as1-casein. J. Dairy Sci. 65:175. 2 Andrew, A. T., and E. Alichanidis. 1983. Proteolysis of caseins and the proteose peptone fraction of bovine milk. J. Dairy Res. 50:275. 3 Aslam, M., R. Jimenez-Flores, H. Y. Kim, and W. L. Hurley. 1994. Two-dimensional electrophoretic analysis of bovine mammary gland secretions collected during the dry period. J. Dairy Sci. 77:1529. 4 Bastian, E. D., K. G. Hansen, and R. J. Brown. 1993. Inhibition of plasmin by b-lactoglobulin using casein as a synthetic substrate. J. Dairy Sci. 76:3354. 5 Bellamy, W., M. Takase, H. Wakabayashi, K. Kawase, and M. Tomita. 1992. Antibacterial spectrum of lactoferricin B, a potent bactericidal peptide derived from the N-terminal region of bovine lactoferrin. J. Appl. Bacteriol. 73:472. 6 Coppock, C. E., R. W. Everett, R. P. Natzke, and H. R. Ainsle. 1974. Effect of dry period length on the Holstein milk production and selected disorders at parturition. J. Dairy. Sci. 57:712. 7 Deharving, G., and S. S. Nelson. 1991. Partial purification and characterization of native plasminogen activators from bovine milk. J. Dairy Sci. 74:2060. 8 De Rham, O., and A. T. Andrews. 1982. The role of native milk proteinase and its zymogen during proteolysis in normal bovine milk. J. Dairy Res. 49:577. 9 Eigel, W. N., C. J. Hofmann, B.A.K. Chibber, J. M. Tomich, T. W. Keenan, and E. T. Mertz. 1979. Plasmin-mediated proteolysis of casein in bovine milk. Proc. Natl. Acad. Sci. USA 76: 2244. 10 Fiat, A. M., D. Migliore-Samour, P. Jolles, L. Drouet, C. Bal Dit Sollier, and J. Caen. 1993. Biologically active peptides from milk proteins with emphasis on two examples concerning antithrombic and immunomodulating activities. J. Dairy Sci. 76:301. 11 Grufferty, M. B., and P. F. Fox. 1988. Milk alkaline protease: review article. J. Dairy Res. 55:609. 12 Holst, B. D., W. L. Hurley, and D. R. Nelson. 1987. Involution of the bovine mammary gland: histological and ultrastructural changes. J. Dairy Sci. 70:935. Journal of Dairy Science Vol. 80, No. 9, 1997

2010

ASLAM AND HURLEY

13 Hurley, W. L. 1989. Mammary gland function during involution. J. Dairy Sci. 72:1637. 14 Kaminogawa, S., K. Yamauchi, S. Miyazawa, and Y. Koga. 1980. Degradation of casein components by acid protease of bovine milk. J. Dairy Sci. 63:701. 15 Korycka-Dahl, M., B. R. Dumas, N. Chene, and J. Martal. 1983. Plasmin activity in milk. J. Dairy Sci. 66:704. 16 Laemmli, U. K. 1970. Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature (Lond.) 227: 680. 17 Mata, L., H. Castillo, L. Sanchez, P. Puyol, and M. Clavo. 1994. Effect of trypsin on bovine lactoferrin and interaction between the fragments under different conditions. J. Dairy Res. 61:427. 18 National Mastitis Council. 1987. Laboratory and Field Handbook on Bovine Mastitis. W. D. Hoard and Sons, Fort Atkinson, WI. 19 Ossowki, L. D., D. Biegel, and E. Reich. 1979. Mammary plasminogen activators: correlation with involution, hormonal modulation and comparison between normal and neoplastic tissue. Cell 16:929. 20 Politis, I., D. M. Barbano, and R. C. Gorewit. 1992. Distribution of plasminogen and plasmin in fractions of bovine milk. J. Dairy Sci. 75:1402. 21 Politis, I., E. Block, and J. D. Turner. 1990. Effect of somatotrophin on the plasminogen and plasmin in the mammary gland: proposed mechanism of action for somatotrophin on the mammary gland. J. Dairy Sci. 73:1494.

Journal of Dairy Science Vol. 80, No. 9, 1997

22 Politis, I., E. Lachance, E. Block, and J. D. Turner. 1989. Plasmin and plasminogen in bovine milk: a relationship with involution. J. Dairy Sci. 72:900. 23 Rejman, J. J. 1988. Biochemical characterization of lactoferrin and a 39 kilodalton protein isolated from bovine mammary secretions during involution. Ph.D. Diss., Univ. Illinois, Urbana. 24 Rejman, J. J., W. L. Hurley, and J. M. Bahr. 1989. Enzymelinked immunosorbant assays of bovine lactoferrin and a 39-kilodalton protein found in mammary secretions during involution. J. Dairy Sci. 72:555. 25 Richardson, B. C. 1983. Variation of the concentration of plasmin and plasminogen in bovine milk with lactation. N.Z. J. Dairy Sci. Technol. 18:247. 26 SAS/STAT User’s Guide, Version 6, 4th Edition. 1990. SAS Inst., Inc., Cary, NC. 27 Schaar, J. 1985. Plasmin activity and proteose-peptone content of individual milks. J. Dairy Res. 52:369. 28 Stelwagen, K., I. Politis, J. H. White, B. Zavizion, C. G. Prosser, S. R. Davis, and V. C. Farr. 1994. Effect of milking frequency and somatotropin on the activity of plasminogen activator, plasminogen, and plasmin in bovine milk. J. Dairy Sci. 77:3577. 29 Van Boekel, M.A.J.S., C.N.J.M. Weerens, and A. Holstra. 1993. Antimutagenic effects of caseins and its digestion products. Food Chem. Toxicol. 10:731. 30 Ventling, B. L., and W. L. Hurley. 1989. Soy proteins in milk replacers identified by immunoblotting. J. Food. Sci. 54:766. 31 Yamauchi, K., and S. Kaminogawa. 1972. Decomposition of milk proteins by milk protease. Agric. Biol. Chem. 36:24.