- Email: [email protected]
Mechanisms of Secretion: Effects of Colchicine and Vincristine on Composition and Flow of Milk in the Goat
Mechanisms of Secretion: Effects of Colchicine and Vincristine on Composition and Flow of Milk in the Goat STUART PATTON Department of Food Science The PennsylvaniaState University University Park, PA 16802 ABSTRACT
regulatory steps. The plant alkaloid, colchicine, inhibits a number o f secretion processes, e.g., very low density lipoproteins by the liver (10, 14, 17), insulin by the pancreas (8), and neurotransmitters b y synaptosomes (9). The present research has evaluated this compound in the lactating m a m m a r y gland of the goat. Experiments with vincristine and adenine also are described. A preliminary report of this work has been published (11).
Colchicine, the plant alkaloid, produced a dramatic decrease in milk flow when infused into the udder of the goat. The c o m p o u n d (1 to 5 mg) dissolved in 5 ml of water was inserted into one side of the udder via the teat canal. Such treatments consistently caused a depression in milk yield from the infused side with maximum at 36 h and substantial reversal by 72 to 96 h. Milks from both the infused and uninfused sides o f the udder were essentially normal in composition (fat, protein, and lactose). However, globulins and riboflavin were elevated in milks from the infused side. The plant alkaloid, vincristine, produced effects on milk secretion similar to those of colchicine but at dosages roughly one-tenth the latter. The two substances had no effect on the amount of milk from the uninfused side of the udder. Experiments employing [carbon-14]colchicine revealed that less than 20% of the infused colchicine is secreted in the milk. Both the secretion of fat globules and the emptying of secretory vesicles by the lactating cell are inhibited by colchicine indicating that a portion of the cell population is turned off from secretion. Plant substances such as colchicine and vincristine may at times limit yields of milk, especially in grazing ruminants.
MATERIALS AND METHODS
INTRODUCTION Evidence that lactogenesis depends upon a concert of hormonal actions continues to accumulate (5, 21). However, crucial molecular events that cause milk to flow remain to be revealed. One approach to this problem concerns the use of antagonists to milk synthesis or secretion. Such substances may help define key Received November 28, 1975.
Saanen goats producing approximately 2 to 4 liters of milk a day were used in these experiments. They were maintained on a conventional hay-grain ration and were milked at 12-h intervals (0900 and 2100) during experimental periods. Substances evaluated for suppression of lactation were colchicine and adenine from Sigma Chemical Co. and vincristine sulfate from Eli Lilly and Co. Colchicine with 14C_labeling of the methoxyl group in ring C (New England Nuclear Corp.) was used in experiments to evaluate metabolism of the compound by the mammary gland. Five-milliliter quantities of aqueous solutions containing varied amounts of the compounds to be tested were infused into right or left halves of the udder. A cannula (1.5 m m inside diameter) was inserted through the opening in the teat canal to deliver the solutions. The latter were thoroughly massaged (up) into the tissue. Milkings from the two sides of the udder were measured for volume, and samples of the milkings were held at 2 to 4 C for analyses subsequently described. During periods in which flow of milk was suppressed, the milks were analyzed for fat, protein, and lactose contents to discern changes in milk synthetic or secretory mechanisms. Fat content was determined by Milko-Tester (turbidimetry) and milk protein by Pro-Milk Automatic Analyzer (dye-binding) (both instruments
1414
COLCHICINE AND MILK FLOW by Foss Electric Co., Hillerod, Denmark). The colorimetric method employing anthrone reagent (3) was used to quantitate lactose. The presence of other sugars in the milk samples was evaluated by the thinqayer chromato o graphic procedure of Gal (6). Three experiments involved infusion of [14C]colchicine. In the first two, 4 mg of colchicine with an activity of 5 x l 0 s cpm were employed; in the third, 5 mg with 5 x 106 cpm were used. Volumes and radioactivities of milkings were determined at 12-h intervals following infusions. For assay of radioactivity, 1 ml of milk dispersed in 15 ml of Aquasol (New England Nuclear Corp.) was analyzed by liquid scintillation spectrometry (Packard Tricarb Model 3330). To check the purity of the purchased [14C]colchicine and to evaluate its contribution to radioactivity in the milks, a thin-layer chromatographic procedure for colchicine was devised. The compound exhibits an Rf of approximately .85 on precoated silica gel plates (Merck) with the solvent system: chloroform, methanol, water, 28% ammonia, 130:70:8:.5 by volume. The radioactive preparation showed >90% of its activity coincident with the spot of authentic colchicine. To study the distribution of radioactivity in the milkings, skim milks were prepared by centrifuging whole milks at 2500 rpm (International Centrifuge, Model B) in 50-ml plastic tubes for 15 min. Skim milks were withdrawn
1415
from under the resulting compacted layers of fat globules with a hypodermic syringe. Whey free of casein micelles was derived by sedimenting the casein in fresh skim milk at 100,000 × g for 1 h at 4 C with a Sorvall centrifuge. Lipids were extracted from milk by the Roese-Gottlieb method (1) with minor modifications. The fatty acid composition of milk lipids was determined by gas chromatographic analysis of the methyl esters prepared by transesterification in methanol with 1% H2SO 4 (vol/vol)as catalyst. A Hewlett-Packard Model 5750 gas chromatograph employing flame ionization detection and a 3.2 mm by 2.15 m stainless steel column was used. The column was packed with 12% HIEFF-1BP on 80 to 100 mesh Gas Chrom P (Applied Science Laboratories). Polyacrylamide-gel separations of the milk proteins were achieved with a vertical electro'phoresis cell (E-C Apparatus Co.) as described by Thompson et al. (18). Riboflavin in milk samples was analyzed qualitatively by its spectral absorption between 350 and 500 n m and quantitatively at 444 nm with the aid of a Beckman Spectrophotometer (Model DU). The authentic compound was obtained from Eastman Kodak Co. R ESU LTS
The representative data of Table 1 and Fig. 1 demonstrate that intramammary infusions of colchicine suppress lactation in the infused half
TABLE 1. Effects of an intramammary infusiona of colchicine on the volume, fat, and protein content of milks from the infused and uninfused sides of the udder (goat). Infused side Milkings (day) 1 PM 2 AM 2 PM 3 AM 3 PM 4 AM 4 PM 5 AM
Volume (% of normal) b 76 39 33 39 55 77 84 88
Uninfused side
Fat (%)
Protein (%)
Volume (% of normal) b
Fat (%)
Protein (%)
3.29 3.26 3.44 4.98 5.24 4.07 3.65 3.61
2.93 3.61 4.43 4.78 4.01 3.20 3.11 3.03
100 110 104 94 93 100 101 102
3.46 3.58 3.75 4.02 3.74 4.03 4.31 4.19
2.95 2.73 2.87 2.98 3.08 3.01 3.14 3.07
aFive mg of colchicine in 5 ml of H~O infused into right half of goat's udder following the AM milking on day 1. bThe mean value of the three milks previous to infusion was used as the normal volume of milk. Journal of Dairy Science Vol. 59, No. 8
1416
PATTON
TABLE 2. Fatty acid composition a of milk lipids from the two sides of a goat's udder 24 and 36 h following an intramammary infusion of 5 mg of colchicine.
I00 _/ :E nO Z
80
,, 6 0 0
_.J 40 w
O
..1 20
0
I
I
I
I
I
I
2
3
4
5
COLCHIClNE
(MG)
FIG. 1. The relationship between the amount of colchicine infused into one half of a goat's udder on the milk yield. Data are maximum depressions in yields from 12-h milkings. of the goat's udder. As shown in Table 1 and reported previously (11), this inhibition substantially is reversed 72 to 96 h postinfusion. From Fig. 1 relative efficacy of the compound falls off with higher dosage, possibly due to differences in accessibility or saturation of drug-responsive sites. The appearance, odor, and flavor of milk from the infused side of the udder during the suppression period were normal. However, wheys from these milks were slightly more yellow-green in appearance due to elevated riboflavin content (see Discussion). Data on fat, protein, and carbohydrate contents of these milkings varied but appeared within ranges for the goat (7). The data of Table 1 suggest that some increase in fat and protein content of the milks occurred as the suppressing effect of the drug was reversed and accumulated fat and protein were released. Lactose was the only sugar detected, and its concentration was normal (4.6 to 5.0%) in milks from both infused and uninfused sides. Fatty acid compositions of the total milk lipids were compared for milks from the infused and uninfused sides at the time of m a x i m u m inhibition, 24 and 36 h following infusion (day 2, AM and PM milking,s, Table 1). While C16:0 is higher and total C18 lower (Table 2) in milks from the infused side, fatty Journal of Dairy Science Vol. 59, No. 8
Uninfused side
Infused side
Fatty acid
24 h
36 h
24 h
36 h
4:0 6:0 8:0 10:0 12:0 14:0 16:0 18:0 18:1 18:2
.8 2.4 2.6 10.5 3.7 12.9 33.2 11.5 20.1 2.3
.8 3.2 2.7 9.2 3.6 11.7 34.5 14.3 18.2 1.7
.8 2.2 2.0 8.8 2.9 10.3 43.4 9.4 18.2 2.1
.8 2.3 2.2 9.6 3.9 13.0 38.7 11.7 15.8 2.0
aAnalysis by gas chromatography of the methyl esters and quantitation of peak areas. Fatty acids are indicated by numbers of carbons, colon, and numbers of double bonds. acid composition for samples from both sides of the udder resemble those of ruminant milk fats. Gel electrophoretic analyses of the proteins in milk from the colchicine-infused and uninfused sides (day 2, PM, Table 1) yielded normal patterns of milk proteins (Fig. 2). However, the globulin band was more intense in milk from the infused side, and this sample also exhibited a unique fast moving component (possibly a phosphopeptide). Infusion of vincristine produced similar reversible suppression of milk in the goat to that with colchicine, but the effective dose was about one-tenth that for colchicine (Fig. 3). Fat and protein contents of milkings at times of maximum suppression by vincristine appeared normal with slightly higher values for the infused side (Table 3). Since adenine represents a product of energy-yielding reactions from adenosine triphosphate, the possibility that adenine accumulation in milk within the gland might produce feedback inhibition of milk flow was tested. A goat producing approximately 500 ml of milk per 12 h from each side of her udder was infused in one side with 150 mg of adenine dissolved in 5 ml of water. There was no significant change in milk volume (500 -+ 50 ml) from either side of the udder during a 36-h (3 milkings) observation. Although not measured
COLCHICINE AND MILK FLOW
1417
soo[400-
\/ /Ji
\/
\/
~ 300
0
20C
IMG
.2MG I 2 MILKING
C1
Wl
C2
W2
FIG. 2. Gel electrophoretic separations o f proteins in milk 36 h following i n t r a m a m m a r y infusion o f 5 m g of colchicine into one half of a goat's udder. C1 and C2 represents separations of caseins f r o m t h e uninfused and infused sides, respectively. W l a n d W2 are separations o f the w h e y proteins f r o m the u n i n f u s e d a n d infused sides. The upper a n d lower arrows p o i n t o u t unique c o m p o n e n t s in milk f r o m the infused side. This milk was o b t a i n e d at the time o f m a x i m u m suppression o f milk yield f r o m the infused side. Some c o n t a m i n a t i o n o f t h e w h e y protein runs with t h e major casein c o m p o n e n t is evident.
at precise 12-h intervals, subsequent milkings were of normal volumes from this animal. In two additional experiments infusion of 5 ml of water produced no significant effect on milk yield.
p
I 4
J
I 6
p
I 8
INTERVALS(DAYS)
FIG. 3. The effects of i n t r a m a m m a r y infusions o f vincristine on t h e yield of milk obtained at 12 h intervals f r o m the two sides of a goat's udder. The first infusion was .1 m g of the alkaloid into one side (e-e) following the complete milking at day .5. The second infusion was .2 m g into t h e o t h e r side (o-o) following the complete milking on day 4.0.
The two infusions each containing 4 mg of colchicine with 14C-activity of 5 × 10 s cpm showed significant radioactivity, i.e., counts > 2 × background in only the 12- and 24-h milkings. In both experiments, the total recovered activity did not exceed 1 to 2% of that infused. The third experiment involving 5 mg of colchicine with ] 4C-activity of 5 × 106 cpm led to an 18% recovery of activity in the first two postinfusion milkings. The relationship between yield of milk and 1 4 C-activity in the milkings is in Fig. 4. In this experiment, milk from the uninfused side showed an amount of activity (35 cpm) exceeding twice background in only
T A B L E 3. Effects o f i n t r a m a m m a r y infusions of vincristine o n percent fat and protein of milks f r o m infused and uninfused sides o f t h e u d d e r (goat) 36 h following t h e infusions a. Vincristine infused (rng)
Fat
Infused side Protein
Fat
Uninfused side Protein
.1 .2
3.97 4.10
3.08 3.68
3.51 3.90
2.83 3.08
aFor yields o f milk in this experiment, see Fig. 3. Journal o f Dairy Science Vol. 59, No. 8
1418
PATTON
500
x (2551) 500
400
/--*
~o je
~I
300
e~.~
200
I00
1
~X--x--~x---*--~-~ 2 3 4 MILKING INTERVALS (DAYS)
k---~ 5
o 6
FIG. 4. The relationship between yield and radioactivity for a series of milkings following intramammary infusion of [14el colchicine (5 mg containing 5 X l0 s cpm).
the second milking following infusion. Radioactivities in whole milk, skim milk, and caseinfree whey from the first milking of the infused side were 2614, 2686, and 3032 cpm indicating that the bulk of the activity in the milk was not associated with fat globules or casein micelles. Radioactivity extracted with the lipids of the skim milk proved to coincide precisely with colchicine on thin-layer chromatographic analysis. DISCUSSION
It generally is held that two principal mechanisms accomplish the secretion of milk: one in which fat droplets are expelled from the cell by envelopment in plasma membrane and the other involving emptying of secretory vesicles containing the skim milk phase (water, lactose, proteins, salts, etc.) through the plasma membrane. The rather modest knowledge of these mechanisms has been reviewed recently (12, 13). Colchicine and other agents including the vinca alkaloids (20) provide a further means of probing this matter. Colchicine and vincristine suppress secretion of milk when infused into the udder of the goat (Table 1 and Fig. 1 and 3). The composition of the milk issuing from the gland during the suppression period appears to be normal in fat, protein, and lactose composition suggesting that milk secretion is an integrated, all-or-none process, and that the infused alkaloids simply turn off cells to secretion. Our preliminary Journal of Dairy Science Vol. 59, No. 8
uhrastructural observations of colchicine-infused goat mammary tissue (B. Stemberger and S. Patton, unpublished) sustain this view. Colchicine produced engorgement of the lactating cells with secretory vesicles and fluid. This is consistent with the findings by Redman et al. (14) that the colchicine or vinblastine block to very low density lipoprotein secretion by the hepatocyte lies between normally filled secretory vesicles and their fusion with and emptying through the plasma membrane. This line of evidence makes failure of myoepithelial cells to respond to oxytocin an unlikely explanation of the arrested milk flow. Such a response would produce engorged lumens rather than bloated cells. While results herein support the idea that secretory vesicles contain the skim-milk phase of milk and that water of milk does not originate by simple diffusion from the lactating cell, the possibility of another secretory pathway from the blood is suggested by elevated globulins (Fig. 2) at the time milk flow was inhibited. An alternative interpretation is that these substances build up in the lactating cell when secretion is restricted. During this period, riboflavin achieved a level (4.0 mg/ liter) two to three times the normal (1.5 to 2.0 rag/liter) for goat milk. These circumstances of elevated globulins and riboflavin are suggestive of their much higher concentrations in colostrum. Researchers have emphasized the capacity of colchicine and vincristine to inhibit the assembly of microtubules as the basis for the various effects of the drugs on cellular processes (20). If microtubules are essential for milk secretion, it is conceivable that their formation determines the onset of lactation. However, involvement of microtubules may not account for all the observed actions of these drugs (14, 16). The tracer experiments reported here indicate that most (>80%) of the infused colchicine moved out of the milk pathways and that little if any colchicine was bound to the plasma (milk fat globule) membrane of the cell. Additional experiments involving less mass and higher specific activity of colchicine may reveal membrane binding. In the suppression of neuro-transmitter release by colchicine or vinblastine, Nicklas et al. (9) have postulated binding of the alkaloid to a myosin-like (membrane) protein that powers the secretion. They also noted
COLCHICINE AND MILK FLOW
inhibition of Mg2+ ATPase by the compounds. A coincidence is that this enzymatic activity, characteristic of many plasma membranes (4, 15, 19), also is implicit in myosin (2), one of the contractile protein components of muscle. Thus, the question of a role for contractile protein in exoeytosis (and membrane fusion) is raised. The effect of eolchicine and vincristine on milk flow is not to be confused with the antimitogenic action of these compounds. The mid-lactation animal has essentially a nonreplieating population of fully differentiated (lactating) eeUs. The effects of plant substances such as colchicine and vincristine on milk yield may have practical ramifications. Grazing of new and unusual pastures can produce economically serious drops in bovine milk production. Conceivably, milk yield and persistence of lactation are influenced by various natural inhibitors in mammalian food sources. The study of lactation by means of drugs and antimetabolites infused into the gland via the teat canal is a relatively undeveloped area of research. In this connection, the bifurcated gland of the goat appears to offer special utility in that the untreated side can be employed as an experimental control. ACKNOWL EDGM ENT
I thank John C. Weaver for assistance with fat and protein analyses and Marvin P. Thompson for gel electrophoretic analysis of the proteins. Use of facilities at the University of California, San Diego, through courtesy of Andrew A. Benson also is acknowledged gratefully. The research was supported in part by US Public Health Service grant HL 03632.
5 6
7
8 9
10
11 12
13
14
15
16
17
18 REFERENCES
1 Association of Official Agricultural Chemists. 1960. Page 190 in Official and tentative methods of analysis. 9th ed. 2 Bagshaw, C. R., D. R. Trentham, R. G. Wolcott, and P. D. Boyer. 1975. Oxygen exchange in the q,-phosphoryl group of protein-bound ATP during Mg2+-dependent adenosine triphosphatase activity of myosin. Proc. Natl. Acad. Sci. 72:2592. 3 Colvin, H. W., Jr., J. T. Attebery, and J. T. Ivy. 1961. Comparison of the anthrone reagent and a copper-reduction method for determining blood sugar in calves. J. Dairy Sci. 44:2081. 4 DePierre, J. W., and M. L. Karnovsky. 1974. Ecto enzymes of the guinea pig polymorphonuclear
19 20
21
1419
leucocyte. I. Evidence for an ectoadenosine monophosphatase, -adenosine triphosphatase and-p-nitrophenyl phosphatase. J. Biol. Chem. 249:7111. Djiane, J., C. Delouis, and R. Denamur. 1975. Lactogenesis in organ cultures of heifer mammary tissue. J. Endocrinol. 65:453. Gal, A. E. 1968. Separation and identification of monosaccharides from biological materials by t h i n - l a y e r chromatography. Anal. Biochem. 24:452. Jenness, R., and R. E. Sloan. 1970. The composition of milks of various species: A review. Dairy Sci. Abstr. 32:599. Lacy, P. E., and W. J. Malaisse. 1973. Microtubules and beta cell secretion. Recent Progr. Horm. Res. 29:198. Nicklas, W. J., S. Puszkin, and S. Berl. 1973. Effect of vinblastine and colchicine on uptake and release of putative transmitters by synaptosomes and on brain actomyosin-like protein. J. Neurochem. 20:109. Orci, L., Y. LeMarchand, A. Singh, F. Assimacopoulos-Jeannet, C. Rouillier, and B. Jeanrenaud. 1973. Role of microtubules in lipoprotein secretion by the liver. Nature 244: 30. Patton, S. 1974. Reversible suppression o f lactation by colchicine. FEBS Lett. 48:85. Patton, S., and R. G. Jensen. 1975. Lipid metabolism and membrane functions of the mammary gland. Prog. Chem. Fats Other Lipids 14:167. Patton, S., and T. W. Keenan. 1975. The milk fat globule membrane. Biochim. Biophys. Acta 415:273. Redman, C. M., D. Banerjee, K. Howell, and G. E. Palade. 1975. Colchicine inhibitors of plasma protein release from rat hepatocytes. J, Cell Biol. 66:42. Solyom, A., and E. G. Trams. 1972. Enzyme markers in characterization of isolated membranes. Enzyme 13: 329. Stadler, J., and W. W. Franke. 1974. Characterization of the colchicine binding of membrane fractions from rat and mouse liver. J. Cell Biol. 60: 297. Stein, O., and Y. Stein. 1973. Colchicine-induced inhibition of very low density lipoprotein release by rat liver in vivo. Biochim. Biophys. Acta 306:142. Thompson, M. P., C. A. Kiddy, J. O. Johnston, and R. M. Weinberg. 1964. Genetic polymorphism in caseins of cows' milk. II. Confirmation of the genetic control of/J-casein variation. J. Dairy Sci. 47:378. Trams, E. G., and C. J. Lauter. 1974. On the sidedness of plasma membrane enzymes. Biochim. Biophys. Acta 345:180. Wilson, L., J. R. Bamburg, S. B. Mizel, L. M. Grisham, and K. M. Creswell. 1974. Interaction of drugs with microtubule proteins. Fed. Proc. 33:158. Wood, B. G., L. L. Washburn, A. S. Mukherjee, and M. R. Bannerjee. 1975. Hormonal regulation of lobulo-alveolar growth, functional differentiation and regression of whole mouse mammary gland in organ culture. J. Endocrinol. 65:1. Journal of Dairy Science Voi. 59, No. 8
regulatory steps. The plant alkaloid, colchicine, inhibits a number o f secretion processes, e.g., very low density lipoproteins by the liver (10, 14, 17), insulin by the pancreas (8), and neurotransmitters b y synaptosomes (9). The present research has evaluated this compound in the lactating m a m m a r y gland of the goat. Experiments with vincristine and adenine also are described. A preliminary report of this work has been published (11).
Colchicine, the plant alkaloid, produced a dramatic decrease in milk flow when infused into the udder of the goat. The c o m p o u n d (1 to 5 mg) dissolved in 5 ml of water was inserted into one side of the udder via the teat canal. Such treatments consistently caused a depression in milk yield from the infused side with maximum at 36 h and substantial reversal by 72 to 96 h. Milks from both the infused and uninfused sides o f the udder were essentially normal in composition (fat, protein, and lactose). However, globulins and riboflavin were elevated in milks from the infused side. The plant alkaloid, vincristine, produced effects on milk secretion similar to those of colchicine but at dosages roughly one-tenth the latter. The two substances had no effect on the amount of milk from the uninfused side of the udder. Experiments employing [carbon-14]colchicine revealed that less than 20% of the infused colchicine is secreted in the milk. Both the secretion of fat globules and the emptying of secretory vesicles by the lactating cell are inhibited by colchicine indicating that a portion of the cell population is turned off from secretion. Plant substances such as colchicine and vincristine may at times limit yields of milk, especially in grazing ruminants.
MATERIALS AND METHODS
INTRODUCTION Evidence that lactogenesis depends upon a concert of hormonal actions continues to accumulate (5, 21). However, crucial molecular events that cause milk to flow remain to be revealed. One approach to this problem concerns the use of antagonists to milk synthesis or secretion. Such substances may help define key Received November 28, 1975.
Saanen goats producing approximately 2 to 4 liters of milk a day were used in these experiments. They were maintained on a conventional hay-grain ration and were milked at 12-h intervals (0900 and 2100) during experimental periods. Substances evaluated for suppression of lactation were colchicine and adenine from Sigma Chemical Co. and vincristine sulfate from Eli Lilly and Co. Colchicine with 14C_labeling of the methoxyl group in ring C (New England Nuclear Corp.) was used in experiments to evaluate metabolism of the compound by the mammary gland. Five-milliliter quantities of aqueous solutions containing varied amounts of the compounds to be tested were infused into right or left halves of the udder. A cannula (1.5 m m inside diameter) was inserted through the opening in the teat canal to deliver the solutions. The latter were thoroughly massaged (up) into the tissue. Milkings from the two sides of the udder were measured for volume, and samples of the milkings were held at 2 to 4 C for analyses subsequently described. During periods in which flow of milk was suppressed, the milks were analyzed for fat, protein, and lactose contents to discern changes in milk synthetic or secretory mechanisms. Fat content was determined by Milko-Tester (turbidimetry) and milk protein by Pro-Milk Automatic Analyzer (dye-binding) (both instruments
1414
COLCHICINE AND MILK FLOW by Foss Electric Co., Hillerod, Denmark). The colorimetric method employing anthrone reagent (3) was used to quantitate lactose. The presence of other sugars in the milk samples was evaluated by the thinqayer chromato o graphic procedure of Gal (6). Three experiments involved infusion of [14C]colchicine. In the first two, 4 mg of colchicine with an activity of 5 x l 0 s cpm were employed; in the third, 5 mg with 5 x 106 cpm were used. Volumes and radioactivities of milkings were determined at 12-h intervals following infusions. For assay of radioactivity, 1 ml of milk dispersed in 15 ml of Aquasol (New England Nuclear Corp.) was analyzed by liquid scintillation spectrometry (Packard Tricarb Model 3330). To check the purity of the purchased [14C]colchicine and to evaluate its contribution to radioactivity in the milks, a thin-layer chromatographic procedure for colchicine was devised. The compound exhibits an Rf of approximately .85 on precoated silica gel plates (Merck) with the solvent system: chloroform, methanol, water, 28% ammonia, 130:70:8:.5 by volume. The radioactive preparation showed >90% of its activity coincident with the spot of authentic colchicine. To study the distribution of radioactivity in the milkings, skim milks were prepared by centrifuging whole milks at 2500 rpm (International Centrifuge, Model B) in 50-ml plastic tubes for 15 min. Skim milks were withdrawn
1415
from under the resulting compacted layers of fat globules with a hypodermic syringe. Whey free of casein micelles was derived by sedimenting the casein in fresh skim milk at 100,000 × g for 1 h at 4 C with a Sorvall centrifuge. Lipids were extracted from milk by the Roese-Gottlieb method (1) with minor modifications. The fatty acid composition of milk lipids was determined by gas chromatographic analysis of the methyl esters prepared by transesterification in methanol with 1% H2SO 4 (vol/vol)as catalyst. A Hewlett-Packard Model 5750 gas chromatograph employing flame ionization detection and a 3.2 mm by 2.15 m stainless steel column was used. The column was packed with 12% HIEFF-1BP on 80 to 100 mesh Gas Chrom P (Applied Science Laboratories). Polyacrylamide-gel separations of the milk proteins were achieved with a vertical electro'phoresis cell (E-C Apparatus Co.) as described by Thompson et al. (18). Riboflavin in milk samples was analyzed qualitatively by its spectral absorption between 350 and 500 n m and quantitatively at 444 nm with the aid of a Beckman Spectrophotometer (Model DU). The authentic compound was obtained from Eastman Kodak Co. R ESU LTS
The representative data of Table 1 and Fig. 1 demonstrate that intramammary infusions of colchicine suppress lactation in the infused half
TABLE 1. Effects of an intramammary infusiona of colchicine on the volume, fat, and protein content of milks from the infused and uninfused sides of the udder (goat). Infused side Milkings (day) 1 PM 2 AM 2 PM 3 AM 3 PM 4 AM 4 PM 5 AM
Volume (% of normal) b 76 39 33 39 55 77 84 88
Uninfused side
Fat (%)
Protein (%)
Volume (% of normal) b
Fat (%)
Protein (%)
3.29 3.26 3.44 4.98 5.24 4.07 3.65 3.61
2.93 3.61 4.43 4.78 4.01 3.20 3.11 3.03
100 110 104 94 93 100 101 102
3.46 3.58 3.75 4.02 3.74 4.03 4.31 4.19
2.95 2.73 2.87 2.98 3.08 3.01 3.14 3.07
aFive mg of colchicine in 5 ml of H~O infused into right half of goat's udder following the AM milking on day 1. bThe mean value of the three milks previous to infusion was used as the normal volume of milk. Journal of Dairy Science Vol. 59, No. 8
1416
PATTON
TABLE 2. Fatty acid composition a of milk lipids from the two sides of a goat's udder 24 and 36 h following an intramammary infusion of 5 mg of colchicine.
I00 _/ :E nO Z
80
,, 6 0 0
_.J 40 w
O
..1 20
0
I
I
I
I
I
I
2
3
4
5
COLCHIClNE
(MG)
FIG. 1. The relationship between the amount of colchicine infused into one half of a goat's udder on the milk yield. Data are maximum depressions in yields from 12-h milkings. of the goat's udder. As shown in Table 1 and reported previously (11), this inhibition substantially is reversed 72 to 96 h postinfusion. From Fig. 1 relative efficacy of the compound falls off with higher dosage, possibly due to differences in accessibility or saturation of drug-responsive sites. The appearance, odor, and flavor of milk from the infused side of the udder during the suppression period were normal. However, wheys from these milks were slightly more yellow-green in appearance due to elevated riboflavin content (see Discussion). Data on fat, protein, and carbohydrate contents of these milkings varied but appeared within ranges for the goat (7). The data of Table 1 suggest that some increase in fat and protein content of the milks occurred as the suppressing effect of the drug was reversed and accumulated fat and protein were released. Lactose was the only sugar detected, and its concentration was normal (4.6 to 5.0%) in milks from both infused and uninfused sides. Fatty acid compositions of the total milk lipids were compared for milks from the infused and uninfused sides at the time of m a x i m u m inhibition, 24 and 36 h following infusion (day 2, AM and PM milking,s, Table 1). While C16:0 is higher and total C18 lower (Table 2) in milks from the infused side, fatty Journal of Dairy Science Vol. 59, No. 8
Uninfused side
Infused side
Fatty acid
24 h
36 h
24 h
36 h
4:0 6:0 8:0 10:0 12:0 14:0 16:0 18:0 18:1 18:2
.8 2.4 2.6 10.5 3.7 12.9 33.2 11.5 20.1 2.3
.8 3.2 2.7 9.2 3.6 11.7 34.5 14.3 18.2 1.7
.8 2.2 2.0 8.8 2.9 10.3 43.4 9.4 18.2 2.1
.8 2.3 2.2 9.6 3.9 13.0 38.7 11.7 15.8 2.0
aAnalysis by gas chromatography of the methyl esters and quantitation of peak areas. Fatty acids are indicated by numbers of carbons, colon, and numbers of double bonds. acid composition for samples from both sides of the udder resemble those of ruminant milk fats. Gel electrophoretic analyses of the proteins in milk from the colchicine-infused and uninfused sides (day 2, PM, Table 1) yielded normal patterns of milk proteins (Fig. 2). However, the globulin band was more intense in milk from the infused side, and this sample also exhibited a unique fast moving component (possibly a phosphopeptide). Infusion of vincristine produced similar reversible suppression of milk in the goat to that with colchicine, but the effective dose was about one-tenth that for colchicine (Fig. 3). Fat and protein contents of milkings at times of maximum suppression by vincristine appeared normal with slightly higher values for the infused side (Table 3). Since adenine represents a product of energy-yielding reactions from adenosine triphosphate, the possibility that adenine accumulation in milk within the gland might produce feedback inhibition of milk flow was tested. A goat producing approximately 500 ml of milk per 12 h from each side of her udder was infused in one side with 150 mg of adenine dissolved in 5 ml of water. There was no significant change in milk volume (500 -+ 50 ml) from either side of the udder during a 36-h (3 milkings) observation. Although not measured
COLCHICINE AND MILK FLOW
1417
soo[400-
\/ /Ji
\/
\/
~ 300
0
20C
IMG
.2MG I 2 MILKING
C1
Wl
C2
W2
FIG. 2. Gel electrophoretic separations o f proteins in milk 36 h following i n t r a m a m m a r y infusion o f 5 m g of colchicine into one half of a goat's udder. C1 and C2 represents separations of caseins f r o m t h e uninfused and infused sides, respectively. W l a n d W2 are separations o f the w h e y proteins f r o m the u n i n f u s e d a n d infused sides. The upper a n d lower arrows p o i n t o u t unique c o m p o n e n t s in milk f r o m the infused side. This milk was o b t a i n e d at the time o f m a x i m u m suppression o f milk yield f r o m the infused side. Some c o n t a m i n a t i o n o f t h e w h e y protein runs with t h e major casein c o m p o n e n t is evident.
at precise 12-h intervals, subsequent milkings were of normal volumes from this animal. In two additional experiments infusion of 5 ml of water produced no significant effect on milk yield.
p
I 4
J
I 6
p
I 8
INTERVALS(DAYS)
FIG. 3. The effects of i n t r a m a m m a r y infusions o f vincristine on t h e yield of milk obtained at 12 h intervals f r o m the two sides of a goat's udder. The first infusion was .1 m g of the alkaloid into one side (e-e) following the complete milking at day .5. The second infusion was .2 m g into t h e o t h e r side (o-o) following the complete milking on day 4.0.
The two infusions each containing 4 mg of colchicine with 14C-activity of 5 × 10 s cpm showed significant radioactivity, i.e., counts > 2 × background in only the 12- and 24-h milkings. In both experiments, the total recovered activity did not exceed 1 to 2% of that infused. The third experiment involving 5 mg of colchicine with ] 4C-activity of 5 × 106 cpm led to an 18% recovery of activity in the first two postinfusion milkings. The relationship between yield of milk and 1 4 C-activity in the milkings is in Fig. 4. In this experiment, milk from the uninfused side showed an amount of activity (35 cpm) exceeding twice background in only
T A B L E 3. Effects o f i n t r a m a m m a r y infusions of vincristine o n percent fat and protein of milks f r o m infused and uninfused sides o f t h e u d d e r (goat) 36 h following t h e infusions a. Vincristine infused (rng)
Fat
Infused side Protein
Fat
Uninfused side Protein
.1 .2
3.97 4.10
3.08 3.68
3.51 3.90
2.83 3.08
aFor yields o f milk in this experiment, see Fig. 3. Journal o f Dairy Science Vol. 59, No. 8
1418
PATTON
500
x (2551) 500
400
/--*
~o je
~I
300
e~.~
200
I00
1
~X--x--~x---*--~-~ 2 3 4 MILKING INTERVALS (DAYS)
k---~ 5
o 6
FIG. 4. The relationship between yield and radioactivity for a series of milkings following intramammary infusion of [14el colchicine (5 mg containing 5 X l0 s cpm).
the second milking following infusion. Radioactivities in whole milk, skim milk, and caseinfree whey from the first milking of the infused side were 2614, 2686, and 3032 cpm indicating that the bulk of the activity in the milk was not associated with fat globules or casein micelles. Radioactivity extracted with the lipids of the skim milk proved to coincide precisely with colchicine on thin-layer chromatographic analysis. DISCUSSION
It generally is held that two principal mechanisms accomplish the secretion of milk: one in which fat droplets are expelled from the cell by envelopment in plasma membrane and the other involving emptying of secretory vesicles containing the skim milk phase (water, lactose, proteins, salts, etc.) through the plasma membrane. The rather modest knowledge of these mechanisms has been reviewed recently (12, 13). Colchicine and other agents including the vinca alkaloids (20) provide a further means of probing this matter. Colchicine and vincristine suppress secretion of milk when infused into the udder of the goat (Table 1 and Fig. 1 and 3). The composition of the milk issuing from the gland during the suppression period appears to be normal in fat, protein, and lactose composition suggesting that milk secretion is an integrated, all-or-none process, and that the infused alkaloids simply turn off cells to secretion. Our preliminary Journal of Dairy Science Vol. 59, No. 8
uhrastructural observations of colchicine-infused goat mammary tissue (B. Stemberger and S. Patton, unpublished) sustain this view. Colchicine produced engorgement of the lactating cells with secretory vesicles and fluid. This is consistent with the findings by Redman et al. (14) that the colchicine or vinblastine block to very low density lipoprotein secretion by the hepatocyte lies between normally filled secretory vesicles and their fusion with and emptying through the plasma membrane. This line of evidence makes failure of myoepithelial cells to respond to oxytocin an unlikely explanation of the arrested milk flow. Such a response would produce engorged lumens rather than bloated cells. While results herein support the idea that secretory vesicles contain the skim-milk phase of milk and that water of milk does not originate by simple diffusion from the lactating cell, the possibility of another secretory pathway from the blood is suggested by elevated globulins (Fig. 2) at the time milk flow was inhibited. An alternative interpretation is that these substances build up in the lactating cell when secretion is restricted. During this period, riboflavin achieved a level (4.0 mg/ liter) two to three times the normal (1.5 to 2.0 rag/liter) for goat milk. These circumstances of elevated globulins and riboflavin are suggestive of their much higher concentrations in colostrum. Researchers have emphasized the capacity of colchicine and vincristine to inhibit the assembly of microtubules as the basis for the various effects of the drugs on cellular processes (20). If microtubules are essential for milk secretion, it is conceivable that their formation determines the onset of lactation. However, involvement of microtubules may not account for all the observed actions of these drugs (14, 16). The tracer experiments reported here indicate that most (>80%) of the infused colchicine moved out of the milk pathways and that little if any colchicine was bound to the plasma (milk fat globule) membrane of the cell. Additional experiments involving less mass and higher specific activity of colchicine may reveal membrane binding. In the suppression of neuro-transmitter release by colchicine or vinblastine, Nicklas et al. (9) have postulated binding of the alkaloid to a myosin-like (membrane) protein that powers the secretion. They also noted
COLCHICINE AND MILK FLOW
inhibition of Mg2+ ATPase by the compounds. A coincidence is that this enzymatic activity, characteristic of many plasma membranes (4, 15, 19), also is implicit in myosin (2), one of the contractile protein components of muscle. Thus, the question of a role for contractile protein in exoeytosis (and membrane fusion) is raised. The effect of eolchicine and vincristine on milk flow is not to be confused with the antimitogenic action of these compounds. The mid-lactation animal has essentially a nonreplieating population of fully differentiated (lactating) eeUs. The effects of plant substances such as colchicine and vincristine on milk yield may have practical ramifications. Grazing of new and unusual pastures can produce economically serious drops in bovine milk production. Conceivably, milk yield and persistence of lactation are influenced by various natural inhibitors in mammalian food sources. The study of lactation by means of drugs and antimetabolites infused into the gland via the teat canal is a relatively undeveloped area of research. In this connection, the bifurcated gland of the goat appears to offer special utility in that the untreated side can be employed as an experimental control. ACKNOWL EDGM ENT
I thank John C. Weaver for assistance with fat and protein analyses and Marvin P. Thompson for gel electrophoretic analysis of the proteins. Use of facilities at the University of California, San Diego, through courtesy of Andrew A. Benson also is acknowledged gratefully. The research was supported in part by US Public Health Service grant HL 03632.
5 6
7
8 9
10
11 12
13
14
15
16
17
18 REFERENCES
1 Association of Official Agricultural Chemists. 1960. Page 190 in Official and tentative methods of analysis. 9th ed. 2 Bagshaw, C. R., D. R. Trentham, R. G. Wolcott, and P. D. Boyer. 1975. Oxygen exchange in the q,-phosphoryl group of protein-bound ATP during Mg2+-dependent adenosine triphosphatase activity of myosin. Proc. Natl. Acad. Sci. 72:2592. 3 Colvin, H. W., Jr., J. T. Attebery, and J. T. Ivy. 1961. Comparison of the anthrone reagent and a copper-reduction method for determining blood sugar in calves. J. Dairy Sci. 44:2081. 4 DePierre, J. W., and M. L. Karnovsky. 1974. Ecto enzymes of the guinea pig polymorphonuclear
19 20
21
1419
leucocyte. I. Evidence for an ectoadenosine monophosphatase, -adenosine triphosphatase and-p-nitrophenyl phosphatase. J. Biol. Chem. 249:7111. Djiane, J., C. Delouis, and R. Denamur. 1975. Lactogenesis in organ cultures of heifer mammary tissue. J. Endocrinol. 65:453. Gal, A. E. 1968. Separation and identification of monosaccharides from biological materials by t h i n - l a y e r chromatography. Anal. Biochem. 24:452. Jenness, R., and R. E. Sloan. 1970. The composition of milks of various species: A review. Dairy Sci. Abstr. 32:599. Lacy, P. E., and W. J. Malaisse. 1973. Microtubules and beta cell secretion. Recent Progr. Horm. Res. 29:198. Nicklas, W. J., S. Puszkin, and S. Berl. 1973. Effect of vinblastine and colchicine on uptake and release of putative transmitters by synaptosomes and on brain actomyosin-like protein. J. Neurochem. 20:109. Orci, L., Y. LeMarchand, A. Singh, F. Assimacopoulos-Jeannet, C. Rouillier, and B. Jeanrenaud. 1973. Role of microtubules in lipoprotein secretion by the liver. Nature 244: 30. Patton, S. 1974. Reversible suppression o f lactation by colchicine. FEBS Lett. 48:85. Patton, S., and R. G. Jensen. 1975. Lipid metabolism and membrane functions of the mammary gland. Prog. Chem. Fats Other Lipids 14:167. Patton, S., and T. W. Keenan. 1975. The milk fat globule membrane. Biochim. Biophys. Acta 415:273. Redman, C. M., D. Banerjee, K. Howell, and G. E. Palade. 1975. Colchicine inhibitors of plasma protein release from rat hepatocytes. J, Cell Biol. 66:42. Solyom, A., and E. G. Trams. 1972. Enzyme markers in characterization of isolated membranes. Enzyme 13: 329. Stadler, J., and W. W. Franke. 1974. Characterization of the colchicine binding of membrane fractions from rat and mouse liver. J. Cell Biol. 60: 297. Stein, O., and Y. Stein. 1973. Colchicine-induced inhibition of very low density lipoprotein release by rat liver in vivo. Biochim. Biophys. Acta 306:142. Thompson, M. P., C. A. Kiddy, J. O. Johnston, and R. M. Weinberg. 1964. Genetic polymorphism in caseins of cows' milk. II. Confirmation of the genetic control of/J-casein variation. J. Dairy Sci. 47:378. Trams, E. G., and C. J. Lauter. 1974. On the sidedness of plasma membrane enzymes. Biochim. Biophys. Acta 345:180. Wilson, L., J. R. Bamburg, S. B. Mizel, L. M. Grisham, and K. M. Creswell. 1974. Interaction of drugs with microtubule proteins. Fed. Proc. 33:158. Wood, B. G., L. L. Washburn, A. S. Mukherjee, and M. R. Bannerjee. 1975. Hormonal regulation of lobulo-alveolar growth, functional differentiation and regression of whole mouse mammary gland in organ culture. J. Endocrinol. 65:1. Journal of Dairy Science Voi. 59, No. 8