Changes in the Lipid Composition of the Secretions of the Bovine Mammary Gland During the Dry Period JOEL BITMAN, D. L. WOOD, and ANTHONY V. CAPUCO Milk secretion and Mastitis Laboratory Uvestock and Poultry ScIences Institute Agricultural Research ServIce, USDA Beltsville, MD 20705
Abbreviation key: CE = cholesteryl esters, FA = fatty acids, FC = free cholesterol, PC = phosphatidyl choline, PE = phosphatidyl ethanolamine, PL = phospholipids, 8M = sphingomyelin.
ABSTRACT
To study initiation of milk fat synthesis, lipid composition of mammary secretions at --60, -40, and -10 d prepartum was studied in lactating and nonlactating Holstein cows. Eleven cows were dried off, and 13 cows were milked twice per day throughout the normal dry period. Total neutral lipid was similar in late lactation milk (--60 d) from lactating cows, 2.1 g/dl, and in milk from the dry group, 2.2 g/dl. Neutral lipids decreased to 1.3 and .9 gldl in quarters from dry cows at -40 and -10 d prepartum. In secretions from dry quarters, triglycerides were 97% of total lipids at -60 d and decreased to 85 and 91 % at -40 and -10 d, respectively. Conversely, FFA and monoglycerides increased during the dry period. Lipids associated with fat globule membrane components increased during the dry period. These increases were 10 times for cholesterol, 20 times for cholesteryl esters, and twice for phospholipids. In general, the content of fat globule core lipids (triglycerides) exhibited a pattern opposite that of membrane lipids (cholesterol, cholesteryl esters, phospholipids) during the prepartum period. Proportions of core lipids tended to decrease, whereas proportions of membrane lipids increased in prepartum mammary secretions. Lipid composition of prepartum secretions may be influenced by blood lipids, somatic cells, and alterations in mammary lipid synthesis. (Key words: involution, fatty acids, lipid classes, phospholipids)
Received June 5, 1991. Accepted October I, 1991. 1992 } Dairy Sci 75:435-442
INTRODUCTION
A dry period appears to be necessary in the dairy cow for optimal milk yield in the succeeding lactation. In cows, the mammary gland involutes as a result of the sudden cessation of milking at the beginning of the dry period. During the dry period before calving, many changes take place in the composition of the constituents of the mammary gland. A milky secretion can be expressed from the gland during the dry period, and, in numerous studies, the gross composition of these secretions has been studied extensively. High concentrations of sodium, chloride, lactoferrin, and immunoglobulins and low concentrations of lactose have been found in these secretions in goats (13) and in cows (15, 16,28). Enzyme concentrations have also been characterized in the secretions from the prepartum mammary gland (I, 16). Although lipids are the major energy components of milk, they have not been studied extensively during involution (17, 25). None of the studies of dry period has characterized mammary secretions with respect to the distribution of the various lipids into individual lipid classes. Thus, the proportions of free cholesterol (FC), cholesteryl esters (CE), diglycerides, FFA, monoglycerides, triglycerides, and phospholipids (PL) in these secretions is not known. Therefore, little is known about maturation of lipid secretory processes in the bovine mammary gland. We had three major objectives in the present study: 1) to investigate changes in the lipid composition of the secretions of the bovine mammary gland at 435
436
BITMAN ET AL.
different stages of the dry period (-60, -40, and -10 d parturition) in cows that were dried off 60 d prepartum or in cows that were milked continuously until parturition; 2) to study the nature of the lipids and detennine the distribution of neutral lipids and PL into classes; and 3) to study the fatty acid composition of the mammary secretions at different stages of the dry period. MATERIALS AND METHODS Animals, Treatments, and Sample Collection
Eleven multiparous Holstein cows were dried off 60 d before their expected calving dates. Because cows were slaughtered before parturition, we could not assess concordance between expected calving date and actual parturition. Thirteen cows, matched on the basis of previous milk production, were milked throughout the preparturient period. A small volume (5 to 20 ml) of mammary secretion was collected at -60, -40, and -10 d before expected parturition from one quarter of each cow. The samples of mammary secretion were stored at -20·C until analyzed. In preparation for lipid analysis, the samples were heated rapidly to 80·C and held at 80·C for 1.5 min to inactivate the lipases (29). Lipid was extracted from the milk sample with chloroformmethanol (2:1) by the Folch procedure (14). Lipid Analysis
Lipid classes in the total lipid extract were detennined by quantitative densitometry in situ (4,5) following separation by two-stage TLC (9) on 20- x 2o-cm Baker, laned., preadsorbeot silica gel plates (J. T. Baker Inc., Phillipsburg, NJ). The PL were separated into classes by TLC (8). Duplicate lCX>-f.1l aliquots were applied to the preadsoJ'bent zone of a 19-1ane, 20- x 2O-Cm Baker Si25O-PA (19C) plate. The plate was developed by continuous TLC for 90 min in a high performance TLC development tank: (Brinkmann Instruments Inc., Westbury, NY) containing 60 ml of chloroform:npropanol:ethyl acetate:methanol:.25% aqueous potassium chloride (25:25:25:13:9). Solvents were evaporated from the plate in a chamber under nitrogen for 30 min. The PL were visuJournal of Dairy Science Vol. 75, No.2, 1992
alized by dipping the plate for 3 s in 10% cupric sulfate in 8% phosphoric acid and charred by temperature-programmed heating in a gas chromatograph oven from 30 to 180·C at 30'C/min and held for 6 min at 180·C. Quantitative Densitometry
The charred TLC plates (either neutral lipid or PL) were quantitated by in situ spectrophotodensitometry using the Shimadzu CS930 Dual Wavelength TLC Scanner (Shimadzu Scientific Instruments, Inc., Columbia, MD). Plates were scanned in a linear mode at 370 DID with a tungsten lamp. Optical densities of sample zones were compared with least squares linear regression lines of individual lipid class standards to determine amount of each class present in the sample. Fatty Acids
Total fatty acids (FA) were transmethylated with boron trifluoride (7) and detennined by wide-bore capillary GLC. The fatty acid methyl esters were analyzed by a HewlettPackard 5840 gas chromatograph (HewlettPackard Co., Palo Alto, CA) equipped with a flame ionization detector on a 30-m x .S3-rom Ld. fused silica open tubular capillary column crosslinked with a l.S-J.lID film of DB 225 (J & W Scientific Inc., Folsom, CA). The column oven was temperature-programmed SO to 180·C at 20·C/min, 180 to 20S·C at 3·C/min, and held at 205·C for 15 min. Other GLC conditions were injection port temperature, 225·C; detector temperature, 250·C; carrier gas (He) flow rate, 10 ml/min; and auxilIary gas (He) flow rate, 40 ml/min. Peak amounts were detennined by use of an internal standard (methyl nonanoate) method, and report data were stored and analyzed by computer. Identities of fatty acid methyl esters and their relative responses were determined by gas chromatography of known quantitative standards. Because of the small volume of secretions available, total lipid was determined by TLC densitometry (7). Statistics
Statistical comparisons were made by analysis of variance. The general linear models
437
IJPIDS OF PREPARTIJM MAMMARY SECRETIONS
TABLE 1. Lipid class comJl9Sition of prepartum mammary secretions of cows dried off 60 d before or milked contiDDously until partmition. 1 Time before partmition
Dry group Composition
-60 d
-40 d
-10 d
2822& 2274 2241 33.4 98.3&
1259bB 135~ 13008 59.2 88.7b 11.38
981boB 957B
Lactating group
SEM
-60d
-40 d
-10 d
2891 2132 2100 32.4b 98.3& 1.7b
3393A 2472A 2434A
3012A 409 2231 A 253 2184A 252 46.2& 2.9 b 97.8 ,A .2 2.2a,B .2
SEM
(mg/dl) Total FA Total1ipid class Total NL Total PL % NL % PL
1.7b
382 451 905 B 449 52.4 12.2 94.3ab,B 3.7 5.7ab,A 3.7
38.3ab 98.4& 1.6b
a,~ with different lower case superscripts are significantly different between days witbin group (P < .05). A,BMeans with different capital supersai.pts are significantly different between the same days for different groups (P
< .OS). IFA
= Fatty acidS;
NL
= neutral
lipids, PL
= phospholipids.
procedure of SAS (SAS Institute, Inc., Cary, NC) was used for ANOVA, followed by least significant difference test of preplanned comparisons for mean separation when the ANOVA was significant (12). RESULTS AND DISCUSSION
Fat Content and Neutral Lipid Classes
In both the cows that were dried off and in the milked cows, the -6O-d sample represents a late lactation milk sample. Consequently, lipid data for secretions taken at this time were similar to those for milk secreted earlier in lactation, and total lipid content, based on summation of all FA, was 2.82 and 2.89 g/dl for the nonlactating and lactating groups, respectively (fable 1). The mammary secretioos of cows that continued to be milked maintained consistently high fat percentages during the prepartum period (Table 1). In the cows dried off prepartum, lipid content, based on total FA, decreased to 1.26 g/dl at -40 d and to .98 g/dl at -10 d (fable 1). Thus, by 20 d into the nonlactating period, it appears that the mammary secretions contained only about onethird of the lipid concentratioo of milk during late lactation. There is little quantitative informatioo in the literature on the changes in fat cootent of bovine prepartum secretions. Wheelock et al. (31) studied the yield and gross composition of
mammary secretions in the first 16 d of the dry period in six cows, a period corresponding to the -60- to -4O-d interval in our study (early involution). Although a decrease in volume of secretion and yield of fat was observed, they observed "no consistent pattern in the changes of fat concentratioo". However, examination of the individual graphs of fat for their six cows showed that fat percentage at 16 d into the dry period was lower in 5 of the 6 cows, decreasing from about 6 to 2%. Recently, Hurley (18) studied 80 pregnant Holstein cows during the first 30 d of involutioo to investigate relationships between secretion volume and coocentrations of components in the secretions. Total fat yield declined rapidly to about 22% of the initial level after 11 d of involution. However, because secretory volume also decreased, fat percentage remained constant at about 4% for the first 11 d of involution. Thereafter, fat yield declined more sharply than secretory volume. At 30 d of involution, fat yield was .2 g per quarter, and fat percentage was .6%. Rowland et al. (28) began milking eight Shorthorn cows 16 d before parturition (late involutioo) and observed that the coocentration of fat increased progressively toward that of milk. Hartmann (15) studied six cows during the latter part of the dry period, from 24 d before parturition to 5 d after parturition. The concentration of fat was low until 2 d before parturition and increased erratically to reach more stable coocentrations. Joumal of Dai1y Science Vol. 75, No.2, 1992
438
BITMAN ET AL. S A AS
6
Dry
A S AB~OO
S; AS
80 4
60
(ij
40
2
(ij
"5
20
f-
'5 (])
Cl
as well as about 50% of CI6, arise from de novo synthesis within the mammary gland (2). In contrast, the longer chain FA such as CI8:I are supplied from circulating lipids and arise from either dietary sources or from depot lipids (2). In the mammary secretions of the dry cows, the short- and medium-chain FA, which are synthesized in the mammary gland, showed a decrease from the proportions found in milk, as represented by the -6O-d sample (Figure 3, left panel). Thus, the FA with chain lengths of C4:0 to ClO:0, which constituted about 16% of total FA at --60 d, decreased in the mammary secretions of the dry cows to about one-half of that level by -10 d prepartum (Figure 3, left panel). The concentration of palmitic acid (C I 6:0) increased during the late prepartum period; C16:0 was the major FA compensating for the decrease observed in C4:0 to CIO:O FA occurring between the -60- and -10-d periods (Figure 3, left panel). The proportion of polyunsaturated FA (C I 8:2, C 18:3, C20:3, and ~O:4) also increased significantly during the dry period. In the lactating cows (Figure 3, right panel), the FA composition of the mammary secretions did not show a decrease in the FA syn-
PI _
-40d
PE -10d
Figure 2. Changes in phospholipid classes of bovine prepartum mammary secretions. Means with different letters above the column are significantly different (P < .05) between days within the group. Asterisks above columns denote a significant difference (P < .05) between dry and lactating secretions at that time. PC = Phosphatidyl choline; PE phosphatidyl ethanolamine; PI phosphatidyl inositol; PS phosphatidyl serine; SM sphingomyelin.
=
=
=
=
thesized de novo, indicating that FA synthesis remained proportional to overall milk yield during the entire prepartum period. As noted earlier (fable 1), total fat concentrations in the mammary secretions were also constant in the lactating cows. The proportions of longer chain FA (C I8 ) increased as the cows approached calving: C18:0 increased significantly, and a trend for higher proportions was observed for C 18:1 (Figure 3, right panel). There is very little information concerning the FA composition of prepartum mammary secretions. Parodi (25) reported on the FA composition of triglycerides from prepartum mammary gland secretion and early postpartum milk of a single Friesian cow. He found that prepartum mammary gland secretion differed greatly from normal milk in that it contained approximately twice the amount of Joumal of Dairy Science Vol. 75, No.2, 1992
440
BITMAN ET AL.
30): 1) active involution, 2) a period of steady state involution, and 3) reinitiation of lactogenesis. In terms of mammary gland function, the early dry period marks the period of active involution (comparisons of -60 vs. -40 d), whereas the later dry period is influenced substantially by pregnancy and the impending parturition. The changes in FA composition of the prepartum mammary secretions in the dry cows suggested that 1) the enzymatic systems involved in generation of the core lipid components were functioning at a lower level (less 30 20 30 40 total fat at -40 and -10 d, Table 1) and 2) Percentage of Total synthesis within the mammary gland of the =·60d =·40d _·10d short- and medium-chain FA was impaired (at Figure 3. Changes in fatty acid composition (carbon -40 and -10 d, Figure 3). In contrast, cows chain length:number of double bonds) of mammary secremilked continuously showed little change in tions of cows dried off 60 d prepartum (left panel) and of the composition of those FA synthesized cows milked continuously until parturition (right panel). within the gland. Thus, it appears that palmitic Means with different letters alongside bar are significantly different (P < .05) between days within the group. Asteracid and the short- and medium-chain FA (C4:0 isks alongside the bars denote a significant difference (P < to C10:0) readily reflect the milk fat synthetic .05) between dry and lactating secretions at that time. status of the mammary gland. This decrease in PUPA = Polyunsaturated fatty acids; SCFA = short- and de novo FA synthesis in the involuting mammedium-chain fatty acids. mary gland may reflect a decrease in the activity of acetyl-coenzyme A carboxylase (24). A comparison of the secretions from mamC 16:0 and larger than normal amounts of C I 4: however, in the mammary secretions of our cows. Brownhill et al. that are not being synthesized, the FA com(11) compared the FA composition of secre- position of the secretion does not reflect protions from goats milked infrequently or twice nounced differences, which implies that the daily until parturition. Goats milked regularly mammary gland is not inactive or nonfuncbefore parturition had significantly higher con- tional during involution and that biosynthesis centrations of triglycerides and the proportions of lipid continues, albeit at a very reduced of long-chain FA (CIS) in the total triglycer- level. The capability of mammary tissue to synides increased compared with the dry goats. Recently, Hurley et al. (17) studied changes in thesize milk fat has been examined during the FA composition of five Holstein cows during prepartum period. Mellenberger et al. (24) the first 31 d of involution. Concentrations of measured the composition of FA synthesized short- and medium-chain FA declined greatly from [14qacetate by mammaI)' tissue from during the first 3 d of involution with little cows at various times before and after parturichange thereafter. Stearic and oleic acids in- tion. The capacity to synthesize FA increased creased during the first 3 d of involution. dramatically at the time of parturition. Before parturition, the rate of FA synthesis was relaChanges In Lipid Components tively low, and the pattern of FA synthesized of Prepartum secretions was predominantly long-chain acids with small The nonlactating period has been consid- amounts of medium-chain FA, characterized ered to represent three successive stages (18, by Mellenberger et al. (24) as indicative of Dry
Lactating
Ioumal of Dairy Science Vol. 75, No.2, 1992
UPIDS OF PREPAR1lJM MAMMARY SECRETIONS
cellular lipid synthesis. After parturition, the rate of lipogenesis increased progressively, and the pattern of FA shifted toward increased percentages of short- and medium-chain FA (24). This was coocluded to be similar to the pattern of FA synthesized de novo and to be typical of functional secretory lactating mammary tissue. Kinsella and Heald (20) also studied bovine mammary tissue at two prepartum stages. They reported that tissue taken just 2 d prepartum was able to synthesize lipids typical of functional secretory tissue (20). Hurley (16) studied the composition of the aqueous phase of mammary secretions (protein, lactose, peroxidase, and glycosidic enzymes) to serve as a reference for transitions in the milksynthesizing function of the mammary gland during the nonlaetating period. Lactose concentrations decreased during early involution but increased rapidly just before parturition. The activities of several glycosidic enzymes exhibited large increases during the late dry period (16). fu mature bovine milk, the fat is packaged in globules surrounded by a lipoprotein membrane that contains .5% cholesterol (as a percentage of total lipid) and 1.0% PL (10). fu prepartum secretions of dry cows, FC, CE, and PI.. increased markedly to about 5% for total cholesterol (FC and CE) and 5 to 6% for PL. Although the increase in concentration of membrane constituents could suggest a different outer packaging of the core milk fat at late prepartum stages (Figure 2), there is a more likely explanation. The larger amounts of PL, FC, and CE might be present in prepartum secretions as residual fragments of epithelial cells, leucocytes, and red blood cells or might represent contamination with small amounts of plasma. Indeed. in a postabsorptive steady state, almost all plasma lipids are transported primarily in equal amounts in PL and in CE (10, 26). Thus, a small contamination with plasma would contribute lipids with this composition. The concentrations and types of s0matic cells present in the secretions of nonlactating bovine mammary glands have been characterized at various times during the dry period (19, 22, 23). Total somatic cells in the secretions remained low for the first 3 d after the last milking but increased 1 to 2 orders of magnitude by 7 to 8 d after cessation of milking (19, 22). These high levels were main-
441
tained until 10 d before parturition, when cell counts decreased slowly until calving (19, 23). These studies show that high concentrations of cells, including phagocytic neutrophils and macrophages, are present in secretions of the non1actating gland throughout most of Ibe dry period (19). It is possible that these cellular elements account for the observed increases in FC, CE, and PI.. in the mammary secretions. fu contrast to the scant literature on the lipid composition of prepartum mammary secretions, considerable research has been conducted on the protein, carbohydrate, and electrolyte composition of prepartum secretions in ruminants (21, 27). This has provided some insights into the mechanisms responsible for Ibese changes in composition of the mammary secretions. The studies by Linzell and Peaker (21) and Peaker (27) in the goat demonstrated Ibat dramatic alterations in permeability occur in mammary alveolar cells during the late prepartum and early postpartum periods. It was proposed that the changes in composition were brought about by transport through a paracellular pathway between the mammary epithelial cells. This involves an alteration in the permeability of the junctions between adjacent cells from leaky junctions during pregnancy to tight junctions near term. Prepartum secretions were found to contain more Na and CI, less K and lactose, and higher amounts of immunoglobulins Iban postpartum milk. Thus, the abrupt shifts that occurred in lipid components in the late prepartum period parallel abrupt shifts in a number of other mammary constituents (electrolytes, immunoglobulins, lactose, glucose, and proteins). Many of these changes may be brought about by permeability shifts occurring at this time of dynamic physiologic change. fucreased permeability through leaky junctions could provide an explanation for the increased presence of PC, CE, and PL. Thus, Ibe increased amounts of these membrane lipids could originate as blood contaminants, such as leucocytes and plasma lipids.
REFERENCES 1 Annison, B. F. 1983. Metabolite utilization by the nuninant mammary gland. Page 399 in Biochemistry of lactation, T. B. Mepham, cd. Elsevier Sci. Pub!. Co., Inc., New York, NY. 2 Bauman, D. E., and C. L. Davis. 1974. Biosynthesis of milk fat Page 31 in Lactation: a comprehensive IreaJoumal of Dairy Science Vol. 75, No.2, 1992
442
BITMAN ET AL.
tise. Vol. 2. B. L. Larson and V. R. Smith, ed. Academic Press, New York, NY. 3 Bilman, J., L. M Freed, M. C. Neville, D. L. Wood, P. Hamosh, and M Hamosh. 1986. Lipid composition of prepartum human mammary secretion and postpartum milk. J. Pediatr. Gastroenterol. Nutr. 5:608. 4 Bilman, J., and D. L. Wood. 1981. Quantitative densitometry in situ of lipids separated by thin layer chromatography. J. Liq. Chromatogr. 4:1023. 5 Bilman, J., and D. L. Wood. 1982. An improved copper reagent for quantitative densitometric thinlayer chromatography of lipids. J. Liq. Cbromatogr. 5: 1155. 6 Bilman, J., and D. L. Wood. 1990. Changes in milk fat phospholipids during lactation. J. Dairy Sci. 73: 1208. 7 Bilman, J., D. L. Wood, M Hamosh, P. Hamosh, and N. R. Mehta. 1983. Comparison of the lipid composition of breast milk from mothers of term and preterm infants. Am. J. Clin. Nutr. 38:300. 8 Bilman, J., D. L. Wood, N. R. Mehta, P. Hamosh, and M. Hamosh. 1984. Comparison of the phospholipid composition of breast milk from mothers of term and pretenn infants during lactation. Am. J. Clin. Nutr. 40: 1103. 9 Bilman, J., D. L. Wood, and J. M Ruth. 1981. Twostage, one-dimensional thin layer chromatographic method for separation of lipid classes. J. Liq. Chromatogr. 4:1007. 10 Bilman, J., D. L. Wood, H. F. Tyrrell, D. E. Bauman, C. J. Peel, A.C.G. Brown, and P. J. Reynolds. 1984. Blood and milk: lipid responses induced by growth hormone administration in lactating cows. J. Dairy Sci. 67:2873. 11 Brownhill, J., H. J. Stewart, and G. E. Thompson. 1985. Prepartum milking and the onset of secretion of milk fat in the goat. J. Physio1. 366:291. 12 Chew, V. 1977. Comparisons among treatment means in analysis of variance. USDA Agric. Res. Publ. ARSI H6, Beltsville, MD. 13 Fleet, I. R., J. A. Goode, M H. Hamon, M. S. Laurie, J. L. Limell, and M Peaker. 1975. Secretory activity of goat mammary glands during pregnancy and the onset of lactation. J. Physiol. 251:763. 14 Folch, J., M. Lees, and G. H. Sloane-Stanley. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J. BioI. Chem. 226: 497. 15 Hartmann, P. E., 1973. Changes in the composition and yield of the mammary secretion of cows during the initiation of lactation. J. Endocrinol. 59:231. 16 Hurley, W. L. 1987. Mammary function during the nonIactaling period: enzyme, lactose, protein concentrations and pH of mammary secretions. J. Dairy Sci. 70:20.
Journal of Dairy Science Vol. 75, No.2, 1992
17 Hurley, W. L., G. J. Warner, and R. R. Grummer. 1987. Changes in triglyceride fatty acid composition of mammary secretions during involution. J. Dairy Sci. 70:2406. 18 Hurley, W. L. 1989. Mammary gland function during involution. J. Dairy Sci. 72:1637. 19 Jensen, D. L., and R. J. Eberhart. 1981. Total and differential cell counts in secretions of the nonIactatjog bovine mammary gland. Am. J. Vet. Res. 42:743. 20 Kinsella, J. E., an1 C. W. Heald. 1972. Na 1_ 14C stearate and Na 2- 4C acetate metabolism and morphological analysis of late prepartum bovine mammary tissue. J. Dairy Sci. 55:1085. 21 Linzell, J. L., and M. Peaker. 1974. Changes in colostrum composition and in the permeability of the mammary epithelium at about the time of parturition in the goat. J. Physiol.243:129. 22 McDonald, J. 5., and A. J. Anderson. 1981. Total and differential somatic cell counts in secretions from noninfected bovine mammary glands: the early nonlactating period. Am. J. Vet. Res. 42:1360. 23 McDonald, J. S., and A. J. Anderson. 1981. Total and differential somatic cell counts in secretions from noninfected bovine mammary glands: the periperatum period. Am. J. Vet. Res. 42:1366. 24 Mellenberger, R. W., D. E. Bauman, and D. R. Nelson. 1973. Metabolic adaptations during lactogenesis. Fatty acid and lactose synthesis in cow mammary tissue. Biochem. J. 136:741. 25 Parodi, P. W. 1983. Positional distribution of fatty acids in triglycerides from prepartum mammary gland secretion and early postpartum milk 1. Dairy Sci. 66: 912. 26 Patton, S., and R. G. Jensen. 1976. Biomedical aspects of lactation. Pergamon Press, Elmsford, NY. 27 Peaker, M 1983. Secretion of ions and water. Page 285 in Biochemistry of lactation. T. B. Mepham, ed. Elsevier Sci. Pub!. Co., Inc. New York, NY. 28 Rowland, S. J., 1.H.B. Roy, H. J. Sears, and S. Y. Thompson. 1953. The effect of prepartum milking on the composition of the prepartum and postpartum secretions of the cow. J. Dairy Res. 20:16. 29 Schwartz, D. P., and O. W. Parks. 1974. The lipids of milk: deterioration. Page 220 in Fundamentals of dairy chemistry. B. H. Webb, A. H. Johnson, and J. A. Alford, ed. AVI Pub!. Co., Inc., Westport, CT. 30 Smith, K. L., and D. A. Todhunter. 1982. The physiology of mammary gland during the dry period and the relationship to infection. Page 87 in Proc. 21st Annu. Mtg. Natl. Mastitis Counc., Louisville, KY. 31 Wheelock, J. V., A. Smith, F. H. Dodd, and RLJ. Lyster. 1967. Changes in the quantity and composition of mammary gland secretion in the dry period between lactations. I. The beginning of the dry period. J. Dairy Res. 34:1.
Abbreviation key: CE = cholesteryl esters, FA = fatty acids, FC = free cholesterol, PC = phosphatidyl choline, PE = phosphatidyl ethanolamine, PL = phospholipids, 8M = sphingomyelin.
ABSTRACT
To study initiation of milk fat synthesis, lipid composition of mammary secretions at --60, -40, and -10 d prepartum was studied in lactating and nonlactating Holstein cows. Eleven cows were dried off, and 13 cows were milked twice per day throughout the normal dry period. Total neutral lipid was similar in late lactation milk (--60 d) from lactating cows, 2.1 g/dl, and in milk from the dry group, 2.2 g/dl. Neutral lipids decreased to 1.3 and .9 gldl in quarters from dry cows at -40 and -10 d prepartum. In secretions from dry quarters, triglycerides were 97% of total lipids at -60 d and decreased to 85 and 91 % at -40 and -10 d, respectively. Conversely, FFA and monoglycerides increased during the dry period. Lipids associated with fat globule membrane components increased during the dry period. These increases were 10 times for cholesterol, 20 times for cholesteryl esters, and twice for phospholipids. In general, the content of fat globule core lipids (triglycerides) exhibited a pattern opposite that of membrane lipids (cholesterol, cholesteryl esters, phospholipids) during the prepartum period. Proportions of core lipids tended to decrease, whereas proportions of membrane lipids increased in prepartum mammary secretions. Lipid composition of prepartum secretions may be influenced by blood lipids, somatic cells, and alterations in mammary lipid synthesis. (Key words: involution, fatty acids, lipid classes, phospholipids)
Received June 5, 1991. Accepted October I, 1991. 1992 } Dairy Sci 75:435-442
INTRODUCTION
A dry period appears to be necessary in the dairy cow for optimal milk yield in the succeeding lactation. In cows, the mammary gland involutes as a result of the sudden cessation of milking at the beginning of the dry period. During the dry period before calving, many changes take place in the composition of the constituents of the mammary gland. A milky secretion can be expressed from the gland during the dry period, and, in numerous studies, the gross composition of these secretions has been studied extensively. High concentrations of sodium, chloride, lactoferrin, and immunoglobulins and low concentrations of lactose have been found in these secretions in goats (13) and in cows (15, 16,28). Enzyme concentrations have also been characterized in the secretions from the prepartum mammary gland (I, 16). Although lipids are the major energy components of milk, they have not been studied extensively during involution (17, 25). None of the studies of dry period has characterized mammary secretions with respect to the distribution of the various lipids into individual lipid classes. Thus, the proportions of free cholesterol (FC), cholesteryl esters (CE), diglycerides, FFA, monoglycerides, triglycerides, and phospholipids (PL) in these secretions is not known. Therefore, little is known about maturation of lipid secretory processes in the bovine mammary gland. We had three major objectives in the present study: 1) to investigate changes in the lipid composition of the secretions of the bovine mammary gland at 435
436
BITMAN ET AL.
different stages of the dry period (-60, -40, and -10 d parturition) in cows that were dried off 60 d prepartum or in cows that were milked continuously until parturition; 2) to study the nature of the lipids and detennine the distribution of neutral lipids and PL into classes; and 3) to study the fatty acid composition of the mammary secretions at different stages of the dry period. MATERIALS AND METHODS Animals, Treatments, and Sample Collection
Eleven multiparous Holstein cows were dried off 60 d before their expected calving dates. Because cows were slaughtered before parturition, we could not assess concordance between expected calving date and actual parturition. Thirteen cows, matched on the basis of previous milk production, were milked throughout the preparturient period. A small volume (5 to 20 ml) of mammary secretion was collected at -60, -40, and -10 d before expected parturition from one quarter of each cow. The samples of mammary secretion were stored at -20·C until analyzed. In preparation for lipid analysis, the samples were heated rapidly to 80·C and held at 80·C for 1.5 min to inactivate the lipases (29). Lipid was extracted from the milk sample with chloroformmethanol (2:1) by the Folch procedure (14). Lipid Analysis
Lipid classes in the total lipid extract were detennined by quantitative densitometry in situ (4,5) following separation by two-stage TLC (9) on 20- x 2o-cm Baker, laned., preadsorbeot silica gel plates (J. T. Baker Inc., Phillipsburg, NJ). The PL were separated into classes by TLC (8). Duplicate lCX>-f.1l aliquots were applied to the preadsoJ'bent zone of a 19-1ane, 20- x 2O-Cm Baker Si25O-PA (19C) plate. The plate was developed by continuous TLC for 90 min in a high performance TLC development tank: (Brinkmann Instruments Inc., Westbury, NY) containing 60 ml of chloroform:npropanol:ethyl acetate:methanol:.25% aqueous potassium chloride (25:25:25:13:9). Solvents were evaporated from the plate in a chamber under nitrogen for 30 min. The PL were visuJournal of Dairy Science Vol. 75, No.2, 1992
alized by dipping the plate for 3 s in 10% cupric sulfate in 8% phosphoric acid and charred by temperature-programmed heating in a gas chromatograph oven from 30 to 180·C at 30'C/min and held for 6 min at 180·C. Quantitative Densitometry
The charred TLC plates (either neutral lipid or PL) were quantitated by in situ spectrophotodensitometry using the Shimadzu CS930 Dual Wavelength TLC Scanner (Shimadzu Scientific Instruments, Inc., Columbia, MD). Plates were scanned in a linear mode at 370 DID with a tungsten lamp. Optical densities of sample zones were compared with least squares linear regression lines of individual lipid class standards to determine amount of each class present in the sample. Fatty Acids
Total fatty acids (FA) were transmethylated with boron trifluoride (7) and detennined by wide-bore capillary GLC. The fatty acid methyl esters were analyzed by a HewlettPackard 5840 gas chromatograph (HewlettPackard Co., Palo Alto, CA) equipped with a flame ionization detector on a 30-m x .S3-rom Ld. fused silica open tubular capillary column crosslinked with a l.S-J.lID film of DB 225 (J & W Scientific Inc., Folsom, CA). The column oven was temperature-programmed SO to 180·C at 20·C/min, 180 to 20S·C at 3·C/min, and held at 205·C for 15 min. Other GLC conditions were injection port temperature, 225·C; detector temperature, 250·C; carrier gas (He) flow rate, 10 ml/min; and auxilIary gas (He) flow rate, 40 ml/min. Peak amounts were detennined by use of an internal standard (methyl nonanoate) method, and report data were stored and analyzed by computer. Identities of fatty acid methyl esters and their relative responses were determined by gas chromatography of known quantitative standards. Because of the small volume of secretions available, total lipid was determined by TLC densitometry (7). Statistics
Statistical comparisons were made by analysis of variance. The general linear models
437
IJPIDS OF PREPARTIJM MAMMARY SECRETIONS
TABLE 1. Lipid class comJl9Sition of prepartum mammary secretions of cows dried off 60 d before or milked contiDDously until partmition. 1 Time before partmition
Dry group Composition
-60 d
-40 d
-10 d
2822& 2274 2241 33.4 98.3&
1259bB 135~ 13008 59.2 88.7b 11.38
981boB 957B
Lactating group
SEM
-60d
-40 d
-10 d
2891 2132 2100 32.4b 98.3& 1.7b
3393A 2472A 2434A
3012A 409 2231 A 253 2184A 252 46.2& 2.9 b 97.8 ,A .2 2.2a,B .2
SEM
(mg/dl) Total FA Total1ipid class Total NL Total PL % NL % PL
1.7b
382 451 905 B 449 52.4 12.2 94.3ab,B 3.7 5.7ab,A 3.7
38.3ab 98.4& 1.6b
a,~ with different lower case superscripts are significantly different between days witbin group (P < .05). A,BMeans with different capital supersai.pts are significantly different between the same days for different groups (P
< .OS). IFA
= Fatty acidS;
NL
= neutral
lipids, PL
= phospholipids.
procedure of SAS (SAS Institute, Inc., Cary, NC) was used for ANOVA, followed by least significant difference test of preplanned comparisons for mean separation when the ANOVA was significant (12). RESULTS AND DISCUSSION
Fat Content and Neutral Lipid Classes
In both the cows that were dried off and in the milked cows, the -6O-d sample represents a late lactation milk sample. Consequently, lipid data for secretions taken at this time were similar to those for milk secreted earlier in lactation, and total lipid content, based on summation of all FA, was 2.82 and 2.89 g/dl for the nonlactating and lactating groups, respectively (fable 1). The mammary secretioos of cows that continued to be milked maintained consistently high fat percentages during the prepartum period (Table 1). In the cows dried off prepartum, lipid content, based on total FA, decreased to 1.26 g/dl at -40 d and to .98 g/dl at -10 d (fable 1). Thus, by 20 d into the nonlactating period, it appears that the mammary secretions contained only about onethird of the lipid concentratioo of milk during late lactation. There is little quantitative informatioo in the literature on the changes in fat cootent of bovine prepartum secretions. Wheelock et al. (31) studied the yield and gross composition of
mammary secretions in the first 16 d of the dry period in six cows, a period corresponding to the -60- to -4O-d interval in our study (early involution). Although a decrease in volume of secretion and yield of fat was observed, they observed "no consistent pattern in the changes of fat concentratioo". However, examination of the individual graphs of fat for their six cows showed that fat percentage at 16 d into the dry period was lower in 5 of the 6 cows, decreasing from about 6 to 2%. Recently, Hurley (18) studied 80 pregnant Holstein cows during the first 30 d of involutioo to investigate relationships between secretion volume and coocentrations of components in the secretions. Total fat yield declined rapidly to about 22% of the initial level after 11 d of involution. However, because secretory volume also decreased, fat percentage remained constant at about 4% for the first 11 d of involution. Thereafter, fat yield declined more sharply than secretory volume. At 30 d of involution, fat yield was .2 g per quarter, and fat percentage was .6%. Rowland et al. (28) began milking eight Shorthorn cows 16 d before parturition (late involutioo) and observed that the coocentration of fat increased progressively toward that of milk. Hartmann (15) studied six cows during the latter part of the dry period, from 24 d before parturition to 5 d after parturition. The concentration of fat was low until 2 d before parturition and increased erratically to reach more stable coocentrations. Joumal of Dai1y Science Vol. 75, No.2, 1992
438
BITMAN ET AL. S A AS
6
Dry
A S AB~OO
S; AS
80 4
60
(ij
40
2
(ij
"5
20
f-
'5 (])
Cl
as well as about 50% of CI6, arise from de novo synthesis within the mammary gland (2). In contrast, the longer chain FA such as CI8:I are supplied from circulating lipids and arise from either dietary sources or from depot lipids (2). In the mammary secretions of the dry cows, the short- and medium-chain FA, which are synthesized in the mammary gland, showed a decrease from the proportions found in milk, as represented by the -6O-d sample (Figure 3, left panel). Thus, the FA with chain lengths of C4:0 to ClO:0, which constituted about 16% of total FA at --60 d, decreased in the mammary secretions of the dry cows to about one-half of that level by -10 d prepartum (Figure 3, left panel). The concentration of palmitic acid (C I 6:0) increased during the late prepartum period; C16:0 was the major FA compensating for the decrease observed in C4:0 to CIO:O FA occurring between the -60- and -10-d periods (Figure 3, left panel). The proportion of polyunsaturated FA (C I 8:2, C 18:3, C20:3, and ~O:4) also increased significantly during the dry period. In the lactating cows (Figure 3, right panel), the FA composition of the mammary secretions did not show a decrease in the FA syn-
PI _
-40d
PE -10d
Figure 2. Changes in phospholipid classes of bovine prepartum mammary secretions. Means with different letters above the column are significantly different (P < .05) between days within the group. Asterisks above columns denote a significant difference (P < .05) between dry and lactating secretions at that time. PC = Phosphatidyl choline; PE phosphatidyl ethanolamine; PI phosphatidyl inositol; PS phosphatidyl serine; SM sphingomyelin.
=
=
=
=
thesized de novo, indicating that FA synthesis remained proportional to overall milk yield during the entire prepartum period. As noted earlier (fable 1), total fat concentrations in the mammary secretions were also constant in the lactating cows. The proportions of longer chain FA (C I8 ) increased as the cows approached calving: C18:0 increased significantly, and a trend for higher proportions was observed for C 18:1 (Figure 3, right panel). There is very little information concerning the FA composition of prepartum mammary secretions. Parodi (25) reported on the FA composition of triglycerides from prepartum mammary gland secretion and early postpartum milk of a single Friesian cow. He found that prepartum mammary gland secretion differed greatly from normal milk in that it contained approximately twice the amount of Joumal of Dairy Science Vol. 75, No.2, 1992
440
BITMAN ET AL.
30): 1) active involution, 2) a period of steady state involution, and 3) reinitiation of lactogenesis. In terms of mammary gland function, the early dry period marks the period of active involution (comparisons of -60 vs. -40 d), whereas the later dry period is influenced substantially by pregnancy and the impending parturition. The changes in FA composition of the prepartum mammary secretions in the dry cows suggested that 1) the enzymatic systems involved in generation of the core lipid components were functioning at a lower level (less 30 20 30 40 total fat at -40 and -10 d, Table 1) and 2) Percentage of Total synthesis within the mammary gland of the =·60d =·40d _·10d short- and medium-chain FA was impaired (at Figure 3. Changes in fatty acid composition (carbon -40 and -10 d, Figure 3). In contrast, cows chain length:number of double bonds) of mammary secremilked continuously showed little change in tions of cows dried off 60 d prepartum (left panel) and of the composition of those FA synthesized cows milked continuously until parturition (right panel). within the gland. Thus, it appears that palmitic Means with different letters alongside bar are significantly different (P < .05) between days within the group. Asteracid and the short- and medium-chain FA (C4:0 isks alongside the bars denote a significant difference (P < to C10:0) readily reflect the milk fat synthetic .05) between dry and lactating secretions at that time. status of the mammary gland. This decrease in PUPA = Polyunsaturated fatty acids; SCFA = short- and de novo FA synthesis in the involuting mammedium-chain fatty acids. mary gland may reflect a decrease in the activity of acetyl-coenzyme A carboxylase (24). A comparison of the secretions from mamC 16:0 and larger than normal amounts of C I 4: however, in the mammary secretions of our cows. Brownhill et al. that are not being synthesized, the FA com(11) compared the FA composition of secre- position of the secretion does not reflect protions from goats milked infrequently or twice nounced differences, which implies that the daily until parturition. Goats milked regularly mammary gland is not inactive or nonfuncbefore parturition had significantly higher con- tional during involution and that biosynthesis centrations of triglycerides and the proportions of lipid continues, albeit at a very reduced of long-chain FA (CIS) in the total triglycer- level. The capability of mammary tissue to synides increased compared with the dry goats. Recently, Hurley et al. (17) studied changes in thesize milk fat has been examined during the FA composition of five Holstein cows during prepartum period. Mellenberger et al. (24) the first 31 d of involution. Concentrations of measured the composition of FA synthesized short- and medium-chain FA declined greatly from [14qacetate by mammaI)' tissue from during the first 3 d of involution with little cows at various times before and after parturichange thereafter. Stearic and oleic acids in- tion. The capacity to synthesize FA increased creased during the first 3 d of involution. dramatically at the time of parturition. Before parturition, the rate of FA synthesis was relaChanges In Lipid Components tively low, and the pattern of FA synthesized of Prepartum secretions was predominantly long-chain acids with small The nonlactating period has been consid- amounts of medium-chain FA, characterized ered to represent three successive stages (18, by Mellenberger et al. (24) as indicative of Dry
Lactating
Ioumal of Dairy Science Vol. 75, No.2, 1992
UPIDS OF PREPAR1lJM MAMMARY SECRETIONS
cellular lipid synthesis. After parturition, the rate of lipogenesis increased progressively, and the pattern of FA shifted toward increased percentages of short- and medium-chain FA (24). This was coocluded to be similar to the pattern of FA synthesized de novo and to be typical of functional secretory lactating mammary tissue. Kinsella and Heald (20) also studied bovine mammary tissue at two prepartum stages. They reported that tissue taken just 2 d prepartum was able to synthesize lipids typical of functional secretory tissue (20). Hurley (16) studied the composition of the aqueous phase of mammary secretions (protein, lactose, peroxidase, and glycosidic enzymes) to serve as a reference for transitions in the milksynthesizing function of the mammary gland during the nonlaetating period. Lactose concentrations decreased during early involution but increased rapidly just before parturition. The activities of several glycosidic enzymes exhibited large increases during the late dry period (16). fu mature bovine milk, the fat is packaged in globules surrounded by a lipoprotein membrane that contains .5% cholesterol (as a percentage of total lipid) and 1.0% PL (10). fu prepartum secretions of dry cows, FC, CE, and PI.. increased markedly to about 5% for total cholesterol (FC and CE) and 5 to 6% for PL. Although the increase in concentration of membrane constituents could suggest a different outer packaging of the core milk fat at late prepartum stages (Figure 2), there is a more likely explanation. The larger amounts of PL, FC, and CE might be present in prepartum secretions as residual fragments of epithelial cells, leucocytes, and red blood cells or might represent contamination with small amounts of plasma. Indeed. in a postabsorptive steady state, almost all plasma lipids are transported primarily in equal amounts in PL and in CE (10, 26). Thus, a small contamination with plasma would contribute lipids with this composition. The concentrations and types of s0matic cells present in the secretions of nonlactating bovine mammary glands have been characterized at various times during the dry period (19, 22, 23). Total somatic cells in the secretions remained low for the first 3 d after the last milking but increased 1 to 2 orders of magnitude by 7 to 8 d after cessation of milking (19, 22). These high levels were main-
441
tained until 10 d before parturition, when cell counts decreased slowly until calving (19, 23). These studies show that high concentrations of cells, including phagocytic neutrophils and macrophages, are present in secretions of the non1actating gland throughout most of Ibe dry period (19). It is possible that these cellular elements account for the observed increases in FC, CE, and PI.. in the mammary secretions. fu contrast to the scant literature on the lipid composition of prepartum mammary secretions, considerable research has been conducted on the protein, carbohydrate, and electrolyte composition of prepartum secretions in ruminants (21, 27). This has provided some insights into the mechanisms responsible for Ibese changes in composition of the mammary secretions. The studies by Linzell and Peaker (21) and Peaker (27) in the goat demonstrated Ibat dramatic alterations in permeability occur in mammary alveolar cells during the late prepartum and early postpartum periods. It was proposed that the changes in composition were brought about by transport through a paracellular pathway between the mammary epithelial cells. This involves an alteration in the permeability of the junctions between adjacent cells from leaky junctions during pregnancy to tight junctions near term. Prepartum secretions were found to contain more Na and CI, less K and lactose, and higher amounts of immunoglobulins Iban postpartum milk. Thus, the abrupt shifts that occurred in lipid components in the late prepartum period parallel abrupt shifts in a number of other mammary constituents (electrolytes, immunoglobulins, lactose, glucose, and proteins). Many of these changes may be brought about by permeability shifts occurring at this time of dynamic physiologic change. fucreased permeability through leaky junctions could provide an explanation for the increased presence of PC, CE, and PL. Thus, Ibe increased amounts of these membrane lipids could originate as blood contaminants, such as leucocytes and plasma lipids.
REFERENCES 1 Annison, B. F. 1983. Metabolite utilization by the nuninant mammary gland. Page 399 in Biochemistry of lactation, T. B. Mepham, cd. Elsevier Sci. Pub!. Co., Inc., New York, NY. 2 Bauman, D. E., and C. L. Davis. 1974. Biosynthesis of milk fat Page 31 in Lactation: a comprehensive IreaJoumal of Dairy Science Vol. 75, No.2, 1992
442
BITMAN ET AL.
tise. Vol. 2. B. L. Larson and V. R. Smith, ed. Academic Press, New York, NY. 3 Bilman, J., L. M Freed, M. C. Neville, D. L. Wood, P. Hamosh, and M Hamosh. 1986. Lipid composition of prepartum human mammary secretion and postpartum milk. J. Pediatr. Gastroenterol. Nutr. 5:608. 4 Bilman, J., and D. L. Wood. 1981. Quantitative densitometry in situ of lipids separated by thin layer chromatography. J. Liq. Chromatogr. 4:1023. 5 Bilman, J., and D. L. Wood. 1982. An improved copper reagent for quantitative densitometric thinlayer chromatography of lipids. J. Liq. Cbromatogr. 5: 1155. 6 Bilman, J., and D. L. Wood. 1990. Changes in milk fat phospholipids during lactation. J. Dairy Sci. 73: 1208. 7 Bilman, J., D. L. Wood, M Hamosh, P. Hamosh, and N. R. Mehta. 1983. Comparison of the lipid composition of breast milk from mothers of term and preterm infants. Am. J. Clin. Nutr. 38:300. 8 Bilman, J., D. L. Wood, N. R. Mehta, P. Hamosh, and M. Hamosh. 1984. Comparison of the phospholipid composition of breast milk from mothers of term and pretenn infants during lactation. Am. J. Clin. Nutr. 40: 1103. 9 Bilman, J., D. L. Wood, and J. M Ruth. 1981. Twostage, one-dimensional thin layer chromatographic method for separation of lipid classes. J. Liq. Chromatogr. 4:1007. 10 Bilman, J., D. L. Wood, H. F. Tyrrell, D. E. Bauman, C. J. Peel, A.C.G. Brown, and P. J. Reynolds. 1984. Blood and milk: lipid responses induced by growth hormone administration in lactating cows. J. Dairy Sci. 67:2873. 11 Brownhill, J., H. J. Stewart, and G. E. Thompson. 1985. Prepartum milking and the onset of secretion of milk fat in the goat. J. Physio1. 366:291. 12 Chew, V. 1977. Comparisons among treatment means in analysis of variance. USDA Agric. Res. Publ. ARSI H6, Beltsville, MD. 13 Fleet, I. R., J. A. Goode, M H. Hamon, M. S. Laurie, J. L. Limell, and M Peaker. 1975. Secretory activity of goat mammary glands during pregnancy and the onset of lactation. J. Physiol. 251:763. 14 Folch, J., M. Lees, and G. H. Sloane-Stanley. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J. BioI. Chem. 226: 497. 15 Hartmann, P. E., 1973. Changes in the composition and yield of the mammary secretion of cows during the initiation of lactation. J. Endocrinol. 59:231. 16 Hurley, W. L. 1987. Mammary function during the nonIactaling period: enzyme, lactose, protein concentrations and pH of mammary secretions. J. Dairy Sci. 70:20.
Journal of Dairy Science Vol. 75, No.2, 1992
17 Hurley, W. L., G. J. Warner, and R. R. Grummer. 1987. Changes in triglyceride fatty acid composition of mammary secretions during involution. J. Dairy Sci. 70:2406. 18 Hurley, W. L. 1989. Mammary gland function during involution. J. Dairy Sci. 72:1637. 19 Jensen, D. L., and R. J. Eberhart. 1981. Total and differential cell counts in secretions of the nonIactatjog bovine mammary gland. Am. J. Vet. Res. 42:743. 20 Kinsella, J. E., an1 C. W. Heald. 1972. Na 1_ 14C stearate and Na 2- 4C acetate metabolism and morphological analysis of late prepartum bovine mammary tissue. J. Dairy Sci. 55:1085. 21 Linzell, J. L., and M. Peaker. 1974. Changes in colostrum composition and in the permeability of the mammary epithelium at about the time of parturition in the goat. J. Physiol.243:129. 22 McDonald, J. 5., and A. J. Anderson. 1981. Total and differential somatic cell counts in secretions from noninfected bovine mammary glands: the early nonlactating period. Am. J. Vet. Res. 42:1360. 23 McDonald, J. S., and A. J. Anderson. 1981. Total and differential somatic cell counts in secretions from noninfected bovine mammary glands: the periperatum period. Am. J. Vet. Res. 42:1366. 24 Mellenberger, R. W., D. E. Bauman, and D. R. Nelson. 1973. Metabolic adaptations during lactogenesis. Fatty acid and lactose synthesis in cow mammary tissue. Biochem. J. 136:741. 25 Parodi, P. W. 1983. Positional distribution of fatty acids in triglycerides from prepartum mammary gland secretion and early postpartum milk 1. Dairy Sci. 66: 912. 26 Patton, S., and R. G. Jensen. 1976. Biomedical aspects of lactation. Pergamon Press, Elmsford, NY. 27 Peaker, M 1983. Secretion of ions and water. Page 285 in Biochemistry of lactation. T. B. Mepham, ed. Elsevier Sci. Pub!. Co., Inc. New York, NY. 28 Rowland, S. J., 1.H.B. Roy, H. J. Sears, and S. Y. Thompson. 1953. The effect of prepartum milking on the composition of the prepartum and postpartum secretions of the cow. J. Dairy Res. 20:16. 29 Schwartz, D. P., and O. W. Parks. 1974. The lipids of milk: deterioration. Page 220 in Fundamentals of dairy chemistry. B. H. Webb, A. H. Johnson, and J. A. Alford, ed. AVI Pub!. Co., Inc., Westport, CT. 30 Smith, K. L., and D. A. Todhunter. 1982. The physiology of mammary gland during the dry period and the relationship to infection. Page 87 in Proc. 21st Annu. Mtg. Natl. Mastitis Counc., Louisville, KY. 31 Wheelock, J. V., A. Smith, F. H. Dodd, and RLJ. Lyster. 1967. Changes in the quantity and composition of mammary gland secretion in the dry period between lactations. I. The beginning of the dry period. J. Dairy Res. 34:1.
