M i l k Progesterone in P o s t p a r t u m and Pregnant Cows as a M o n i t o r of R e p r o d u c t i v e Status J. J. SCHIAVO, R. L. MATUSZCZAK, E. B. OLTENACU, and R. H. FOOTE Department of Animal Science Cornell University Ithaca, NY 14853 ABSTRACT
MATERIALS AND METHODS
Milk samples were analyzed for progesterone content by a petroleum ether extraction and competitive protein binding assay validated for milk. In one experiment, 11 cows were sampled twice daily for 24 days beginning with an observed estrus 15 to 45 days postpartum, and again 19, 21, 23, and 25 days after breeding. Progesterone values during the estrous cycle paralleled those for blood plasma but were slightly higher at estrus (1.49 ng/ml milk) and maximum (9 ng/ ml) on days 11 to 16 of the estrous cycle. After breeding, cows later diagnosed pregnant averaged 7.12 ng/ml while those later found to be nonpregnant averaged 2.36 ng/ml. All diagnoses of pregnancy were correct. In a separate experiment there was no difference between milk from front and rear quarters, but progesterone was highest in last milk, intermediate in composite milk, and lowest in first milk.
In the first experiment, composite milk samples were taken from weigh jars at morning and evening milkings 12 h apart from 11 Holstein-Friesian cows. Sampling began on the evening of an observed estrus 15 to 45 days postpartum and continued for the next 24 days. The milk was refrigerated immediately after collection and usually within a day was frozen until the time of assay. Samples of afternoon milk also were taken from these cows 19, 21, 23, and 25 days after artificial insemination. The general reproductive condition of each cow was determined by rectal palpation at this time, and pregnancy was checked 40 to 45 days after breeding. Since lipid content of first milk is different from that of last milk and since progesterone is associated primarily with the lipid fraction (7, 10), we performed a second experiment with four Holstein-Friesian cows. First milk from front and rear quarters, last milk from front and rear quarters, and composite milk samples were taken on six afternoons. All samples were refrigerated immediately and frozen shortly thereafter. Concentration of progesterone in the milk was determined by the competitive protein binding method as modified by Concannon (1). Approximately 1000 cpm of radioactive progesterone [progesterone "1,2,6,7, -3H(N), New England Nuclearl were added to each extraction vial as an internal standard for estimation of recovery. This amount of progesterone was negligible compared to concentrations in milk. Duplicate 1 ml samples of milk were extracted with 8 ml of petroleum ether for 30 rain on a horizontal shaker. The aqueous portion was frozen out in an ethanol dry ice bath and discarded. The organic fraction containing the progesterone was poured off and evaporated to dryness. To reduce lipid content, 2 ml of 70% methanol (AR) were added to the tubes which were then placed at - 2 0 C for 48 to 72 h. The neutral lipid moved into the water c o m p o n e n t
Reproductive efficiency of dairy herds is low (12, 15). Many causes have been identified but anestrus, or failure to detect estrus, is a major factor in all studies. Assays of steroid hormones in blood plasma have provided much information about the estrous cycle in the bovine. Progesterone, a key hormone in regulating the estrous cycle, has been measured in milk (3, 6, 10), which offers considerable advantage over blood plasma for field studies because of its ease of collection. Our experiments were designed to study effects of methods of obtaining milk samples on progesterone levels, and to determine if these could assist in characterizing the estrous cycle and in early detection of pregnancy.
Received December 26, 1974.
SCHIAVO ET AL.
TABLE 1. Recovery of known amounts of progesterone.
Amount recoveredb Amount addeda (ng/ml) 0 (pool) .25 .50 1.00 2.50 5.00 10.00
.03 .09 .10 .16 .11 .30 .21
... 280 180 124 110 109 92
(ng/ml) 1.46 .70 .90 1.24 2.75 5.45 9.29
aReplicates: n = 9. bcorrected for pool. and formed frozen droplets. One milliliter of the liquid was removed and extracted with 5 ml petroleum ether and 2 ml 2X distilled H20. The petroleum ether fraction was collected, dried, and assayed for progesterone. To validate the assay, inhibition curves were prepared by assay of increasing amounts of standard cold progesterone and extracts of increasing amounts of milk. These curves were parallel. Also, known amounts of cold progesterone were added to aliquots of a milk pool, which were extracted and assayed as usual. Although .25 and .50 ng consistently were overestimated, recovery of progesterone in the higher concentrations expected in milk agreed with the known added (Table 1). Nuti et al. (11) recently reported that milk samples purified by thin layer chromatography and assay by gas liquid chromatography for progesterone gave values not significantly different from samples assayed by radioimmunoassay. We expect, therefore, that most of the substance measured by the competitive protein binding assay is true progesterone.
Of all 11 cows sampled, 8 showed a "normal" estrous cycle. Figure 1 represents progesterone values from milk averaged over those 8 cows. This graph shows that on days 0 to 4 of the cycle progesterone values are at a baseline of about 1 ng/ml of milk. Staring on day 4, values begin to increase gradually and plateau at about 9 ng/ml on days 11 to 16. Subsequently, they begin to decline precipitously and are again at the baseline before onset of the next estrus. Individual cows, day of cycle, and their interaction had a highly significant (P.I). Of the three cows not classified "normal," two showed some cyclic activity but had low peak progesterone values of 2 to 3 ng/ml of milk at midcycle. The other cow cycled irregularly with estrus shown several times in the 24 days. Table 2 shows progesterone values for the 11 cows on days 19, 21, 23, and 25 after artificial insemination. We used these values to diagnose pregnancy. Cows diagnosed pregnant averaged 7.12 ng progesterone/ml while nonpregnant cows averaged 2.36 ng/ml. Our predictions agreed perfectly with the diagnosis of pregnancy by rectal palpation 7 wk after breeding. Results for fractioned milk are in Table 3. The amount of progesterone in milk was affected by day of cycle (P