- Email: [email protected]
Teat stimulation-induced oxytocin and catecholamine release in pregnant and lactating holstein heifers
DOMESTIC ANIMAL ENDOCRINOLOGY
Vol. 8(2):235-243, 1991
TEAT STIMULATION-INDUCED OXYTOCIN AND CATECHOLAMINE RELEASE IN PREGNANT AND LACTATING HOLSTEIN HEIFERS A.M. Lefcou# and R.M. Akers 2 U.S. Department of Agriculture Agriculture Research Service Livestock and Poultry Science Institute Beltsville, Maryland 20705 Received September7, 1990
ABSTRACT The effect of manual teat stimulation (milking paradigm) on release of oxytocin, epinephrine and norepinephrine was studied in (1) 15 heifers at 100, 150, 200 and 250 d of gestation and at 30 and 90 d of lactation (during machine milking) and (2) simultaneously in six heifers at <100 d and in six heifers at >200 d of gestation. Oxytocin responses to teat stimulation, including peak heights and area under the response curves, at 150, 200 and 250 d or at >200 d of gestation were similar and were significantly greater than responses at 100 d or at <100 d. Responses to milking were lower at 90 d compared to responses at 30 d. Catecholamines were measured only during gestation and were generally not affected by teat stimulation. Epinephrine levels were significantly higher at 200 and 250 d compared to levels at 100 and 150 d. Baseline oxytocin concentrations and responses to teat stimulation were greatest at 150 d of gestation when epinephrine levels were still low, suggesting that stimulatory mechanisms responsible for the release of oxytocin develop and/or are expressed prior to the development of inhibitory sympathetic mechanisms. For norepinephrine, linear analyses did not show significant responses to teat stimulation overall. However, elevated norepinephrine responses (>.2 pmol/ml) following teat stimulation were seen in 28 of 51 trims, and large oxytocin responses (>75 pg/ml/min) were seen predominantly only when norepinephrine responses were low (<=.2 pmol/ml). INTRODUCTION In lactating mammals, oxytocin concentrations increase in response to nipple stimulation and this increase is important for efficient milk removal (1,2,3,4). However, little is known about the development of this neuroendocrine reflex. Gorewit (5) found a small, but insignificant, release of oxytocin in 12 to 14 m old virgin cattle following 2 min of vigorous udder massage. In contrast, oxytocin is released in responses to udder massage for at least 30 d after milking is terminated (4). The neuroendocrine reflex responsible for the release of oxytocin in response to teat-stimulation has not been characterized for nulliparous pregnant heifers. In humans, oxytocin responds to breast stimulation during pregnancy but the magnitude of the response is less than in lactating women (6). In ewes, oxytocin increases in response to vaginal stimulation during lactation but not during pregnancy (7). The sympathetic nervous system plays an important role in regulating the efficiency of the milk ejection reflex, although it is not clear whether the primary interaction is centrally or peripherally mediated (1,8,9,10). Even if increased peripheral catecholamine levels do not directly inhibit the release of oxytocin in responses to teat-stimulation (8,9,10), increased peripheral release of catecholamines is probably indicative of an increase in sympathetic arousal. Increased sympathetic activity has been reported to inhibit release of oxytocin (1,3,11). Regardless, peripheral catecholamine levels and catecholamine responses to teatstimulation have not been characterized throughout pregnancy. Consequently, we studied oxytocin and catecholamine release after 30 s of udder wash and 4.5 rain of manual teat stimulation in nulliparous pregnant heifers at 100, 150, 200 and 250 d of gestation and the oxytocin responses of these same heifers to machine milking at 30 and Copydght © 1991 by Domendo, Inc.
235
0739-7240/91/$3.00
236
LEFCOURT AND AKERS
90 d postpartum. To address possible seasonal effects due to changes in ambient temperature or day length, oxytocin responses were measured simultaneously in two groups of cattle either at <100 d or at >200 d of gestation. A preliminary report has been published (12). MATERIALS AND M E T H O D S Blood Sampling. A polyvinyl cannula (0.48 m in length) was placed into a jugular vein of each animal in the direction of the heart, to a depth of 0.32 m, 1-2 d before the start of blood sampling. To accustom animals to the experimental protocol, blood (4 ml) was withdrawn and discarded at approximately 10 min intervals for 1-2 hr before the start of each trial. Blood samples (20 ml) were collected at -10, -5, -1,0, 1, 2, 3, 4, 5, 6, 8, 10, 15 and 30 min relative to the initiation of teat stimulation or milking. Integrated samples were taken during consecutive 1 min intervals (20 rnl/min) starting 1 rain before and ending 7 min after the onset of stimulation. During the week before blood sampling, lactating heifers were milked once each morning in a milking parlor with side-opening stalls, subsequently used on the day of experiment, and once each afternoon in accord with the usual routine with the remainder of the milking herd. This was to allow the animals to become accustomed to the surroundings and milking routine used on the day of sampling. For oxytocin, blood was collected on ice, stored at 4 C overnight and sera prepared by centrifugation the following morning. Sera samples were stored at -20 C until assayed as described by Gorewit (13). Antibody to oxytocin was purchased from Calbiochem; synthetic oxytocin (Calbiochem) was used as standard and was iodinated according to Bolt (14). Aliquots of a pooled serum sample were measured in each assay for calculation of intra- and inter-assay variation which were 5.9% and 3.4%, respectively. For measurement of catecholamines, samples taken at -10, -5, 1, 2, 4 and 8 min were collected using polystyrene syringes containing sodium heparin and sodium metabisulfite and stored on ice. Portions of each sample (10 ml) were cold (4 C) centrifuged and resulting plasma stored at -75 C until assayed for catecholamines. Thawed 1 ml samples of plasma were extracted into perchloric acid using activated alumina (15,16). Catecholamines were quantified using a high-pressure liquid chromatography assay with an ion-pairing column (Beckman) and electrochemical detection (17). Norepinephrine (Sigma) and epinephrine (Sigma) were used as standards. Butanephrine (kindly supplied by Sterling-Winthrop Research Institute, Rensselaer, New York) was used as an internal standard. Coefficients of variation for noradrenaline and adrenaline were 7 and 8%. The lower limit at which catecholamines could be quantified was about 0.4 pmol. With an extraction efficiency of 60%, this meant that concentrations below 0.7 pmol/ml could not be quantified although they were often detected. Most samples were run in duplicate. Unfortunately, due to failure of a freezer, catecholamine samples taken during lactation and for experiment 2 (below) could not be analyzed. Teat Stimulation. In an attempt to standardize the teat stimulation, all teat manipulation was done by one person. After a 30 s udder wash, one pair of front and rear teats was alternately pulled and released (to mimic handmilking) for 30 s. Teat stimulation for contralateral udder halves was alternated at 30 s intervals for a total of 4.5 min. In milking trials, the udder was washed for 30s and the milking machine was attached for 4.5 min. Experiment 1. Fifteen pregnant Holstein heifers, previously treated with prostaglandin F2~ to synchronize estrus (18), were used to determine the effect of stage of gestation and of lactation on oxytocin and catecholamine response to teat stimulation. The heifers were kept in a stanchion barn throughout most of gestation. Just before expected parturition the heifers were transferred to a maternity barn and, after calving, were maintained with the remainder of the lactating herd. Three heifers were eliminated from the experiment after parturition be-
237
HORMONAL RESPONSES TO TEAT STIMULATION
cause of calving difficulty or severe mastitis. Blood samples were taken at 100 (in December), 150, 200 and 250 (in May) d of gestation, and at 30 and 90 d after calving. Experiment 2. Twelve pregnant Holstein heifers (six < 100 and six >200 d of gestation) were used to determine whether responses in experiment l were due to stage of gestation or season. Four animals were tested on each of 3 consecutive days in October. For 2 weeks before the experiment started, the animals were housed in stanchions at the same location as the heifers in experiment 1. Statistical Analyses. Oxytocin, norepinephrine and epinephrine data were analyzed using a linear model (Procedure GLM,19). The model included factors for TIME of sample, STAGE of gestation or lactation, TIME by STAGE and, for experiment l, COW. In addition, average baseline values for each hormone were determined by averaging data by cow for times prior to stimulation. For oxytocin, the peak value, time to peak value, and area under the response curve (time weighted averages of values between 1 and 8 min) also were calculated. For catecholamines, the peak value and the average of values for l, 2 and 4 rain were computed. Measures calculated with individual baselines subtracted also were analyzed. Results with baselines subtracted were generally similar to those where baselines were not subtracted, and are shown only where differences exist. To examine relationships among measured variables, data were analyzed for correlations and for differences in frequencies of responses using Chi-square (20). RESULTS
Oxytoein: Experiment 1. Oxytocin responses to teat stimulation at 150, 200 and 250 d of gestation were similar and were significantly greater (P<.0 l) than responses at 100 d (Figure 1). This difference in responsiveness was also apparent in peak responses and area under response curves (Table 1). During lactation, oxytocin responses to milking were significantly (P < .05) reduced at 90 d compared to 30 d (Figure 2), as were peak responses and area under response curves. COW was a significant (P < .01) factor in all analyses.
400 -
/~~ Gestation I ~
Oxytocin
teat stimulation I~ I I
300350
I
m
E Q.
2oo2° l 150
------.._=_ ]
~-
w
=-111
v
250 days
Odays l
n"
200 days
U
100 days
- Standard Error of the Means
100 -10
= .... -5
I .... 0
= .... 5
i .... 10
I .... 15
= .... 20
I .... 25
= 30
Time (rain) Figure I. Leastsquaremeansof oxytocinconcentrationsin relationto teatstimulationduringgestationin 15 heifers.
238
LEFCOURT AND AKERS TABLE 1. OXYTOCINMEASURES RELATEDTO TEAT STIMULATIONDURING GESTATION,OR MILKINGDURINGLACTATION,IN ]5 HEIFERS.TO TEST SEASONAL EFFECTS, T w o GROUPS OF 6 HEIFERS EITHERAT <100 D OR AI >200 D OF GESTATIONWERE EVALUATEDSIMULTANEOUSININ A SINGLESEASON.
Stage o f Gestation
Oxytocin Peak Time of Peak (pg/ml) (rain)
Baseline (pg/ml)
100d 150 d 200d 250 d
176± 234 ± 197o+ 206 ±
Lactation 30d 90d
10 ~' l0 b I I"." 10 h
Area (pg/ml/min)
239 + 3T' 485 _+37 ~ 356+39 h 374 0+ 37"
3.6±.3 3.6 ± .3 3.1 ± . 4 3.8 _+ .3
201 334 271 275
± 17" ± 17 ~ 0+ 18 ~ _+ 17 h
251 ± 10 d 2 1 9 ± 12
4 0 4 ± 15" 3 5 3 ± 17'
3.6 ± . 3 3.3_+.3
3 1 6 ± I1" 2 6 9 ± 12'
131 ± 2 7 " 201 ± 27 h
210-+50 330 -+ 50
4.0_+.6 3.5 _+ .6
168±35" 267 _+ 35 b
Gestation <100d >200 d
"'b~Least square means are shown 0+SE and, tbr each stage, are different within a column stage when superscripts differ. ~JThe difference between 30 d and 90 d baselines was almost significant (P < .()6L
There was no apparent correlation between oxytocin concentrations and peripheral levels of either catecholamine. To examine whether oxytocin responses might be less when catecholamine levels were elevated, 2 by 2 contingency tables were constructed post hoc based on oxytocin and catecholamine levels. A significant (P<.01) relationship was found when areas under the oxytocin response curves with baselines subtracted were contrasted with average norepinephrine values with baselines subtracted (Table 2). Large oxytocin responses (>75 pg/ml/min) were seen predominantly only when norepinephrine responses were low (<=.2 pmol/ml).
400
Oxytocin t~on
milking 350
300 30 days 8
250
i I
200
90 days
150
T -
100
. . . .
-10
Standard Error of the Means
I
-5
. . . .
|
0
. . . .
i
5
. . . .
I
. . . .
10
Time
I
15
. . . .
i
20
. . . .
I
. . . .
25
(min)
Figure 2. Least square means of oxytocin concentrations in relation to milking during lactation in 12 heifers.
I
30
HORMONALRESPONSESTO TEAT STIMULATION
239
TABLE 2. OXYTOCIN, AREA UNDER RESPONSES CURVES WITH BASELINESSUBTRACTED,CONTRASTED IN A 2 BY 2 CONTINGENCYTABLE WITH AVERAGENOREPINEPHRINERESPONSES TO TEAT STIMULATION, ALSO WITH BASELINES SUBTRACTED. DATA INCLUDEMEASUREMENTS FROM 100 D, 150 D, 200 D AND 250 D OF GESTATIONFOR 15 HEIFERS.
Oxytocin (pg/ml/min) <=75 26 14
Norepinephrine >.2 (pmol/ml) <=.2 "Chi-square= 7.64, 1 degreeof freedomP<.01
>75 2 9
Oxytocin: Experiment 2. The oxytocin responses to teat stimulation were significantly (P<.05) greater in heifers at > 200 d gestation compared to heifers at <100 d (Figure 3; Table 1). Baseline oxytocin levels were lower in the < 100 d animals compared to the 100 d animals from experiment 1; however, other measures including area under response curves with baselines subtracted were similar. Responses at >200 d gestation fell in the middle of responses at 150 to 250 d gestation in experiment 1. Epinephrine: Experiment 1. Epinephrine data are biased by the low levels detected in early gestation and lack of assay sensitivity. Data points with values below detectable limits were excluded from analyses. Even so, epinephrine levels and baseline values were significantly (p < .05) higher at 2130 and 250 d gestation compared to 100 and 150 d (Figure 4; Table 3). Teat stimulation did not affect epinephrine levels. COW was a significant (P < .01) factor in all analyses. Due to failure of a freezer, catecholamine samples taken during lactation and for experiment 2 could not be analyzed.
40oI
Oxytocin
Gestation
within a season
350
300
E
teat stimulation
-
~~
~
- StandardError of the Means
250
200
150 w
100
. . . .
-10
I
-5
. . . .
I
0
. . . .
I
5
. . . .
I
. . . .
10
I
15
. . . .
I
20
....
I ....
25
I
30
Time (min) Figure 3. Least square means of oxytocinconcentrationsin relation to teat stimulationfor two groupsof 6 heifers at <100 d and at >200 d of gestation. To test seasonal effects, both groups were evaluated simultaneouslyin a single season.
240
LEFCOURT AND AKERS 2.0
Epinephrine Gestation teat stimulation
1.5 200 days m
E 0
E Q.
250 days 1.0
0.5
0.0
i
150 days
l
100 days
i
-10
-5
0
.I.
5
10
Time (min) Figure 4. Least square means of epinephrine concentrations in relation to teat stimulation during gestation in 15 heifers. No estimate is shown lbr -5 min at 150 d as the predicted value is negative. This negative value was due to lack of assay sensitivity and the small n u m b e r of samples in which epinephrine could be detected tor this time point.
Norepinephrine: Experiment 1. Norepinephrine values did not vary by stage of gestation or in response to teat stimulation; however, there was a tendency for levels to be elevated at 200 d (Figure 5; Table 3). COW was a significant (P < .01 ) factor in all analyses. Even though linear analyses did not show a significant responses to teat stimulation overall, examination of the data revealed that a number of animals showed evidence of a response. A tendency for a response can be seen even in the mean data at 100, 150 and 200 d gestation (Figure 5). When the data were analyzed using contingency tables (see above), elevated norepinephrine responses (>.2 pmol/ml) following teat stimulation were seen in 28 of 51 trials (Table 2). DISCUSSION Despite the importance of oxytocin to reproduction and lactation, little data exist concerning oxytocin values during pregnancy and almost nothing is known about the development of the neurohumoral reflex for the release of oxytocin in response to teat-stimulation. In one of
TABLE 3. CATECHOLAMINEMEASURES RELATEDTO TEAT STIMULATIONDURINGGESTATIONIN 15 HEIP~b.RS. Stage of
Epinephrine
Gestation
(pmol/ml) Peak
Average'
Baseline
.92 ± .29 ~'" .49 ± .32 ~' 1.56 -+ .29" 1.25 --- .34 "b
.86 ± .26 "b .24 _+ .35" I. 15 + .25 h 1.06 ± .29 "~
I. 18 _+ .28 I. 16 _+ .23 1.84 _+ .27 1.12 ± .24
100 d 150 d 200 d 250 d
Baseline .34 .07 1.29 1.14
_+ .32" ± .39 ~' ± .29 ~ ± .30 h
Norepinephrine (pmolhnl) Peak 1.86 + 1.77 + 2.89 ± 1.91 ±
.38 "b .33" .36 b .35 ~''~
~hLeast square means are shown ± SE and are different within a column when superscripts differ. ~Average of values at I. 2 and 4 rain tbllowing teat stimulation.
Average 1.66 ± 1.37 ± 2.16 ± 1.53 ±
.33 .28 .3 I .30
HORMONAL RESPONSES TO TEAT STIMULATION
3.0
241
Norepinephrine Gestation
2.5
T teat stimulation
T
~L=
2.0
E o 1.5
100 days
E Q.
150 days
1.0
0.5
0.0
I
-10
.
.
.
.
I
I
-5
0
.
.
.
.
I
I
5
10
Time (rain) Figure 5. Least square means of norepinephrine concentrations in relation to teat stimulation during gestation in 15 heifers.
two ewes, circulating oxytocin concentrations were elevated during the first 2 wk of pregnancy and then fell and remained constant even after parturition. In the other ewe, levels were somewhat elevated during pregnancy and peaked prior to parturition (21). In the current study, baseline oxytocin concentrations tended to be lower during early gestation and constant thereafter during gestation and postpartum. In humans, breast stimulation resulted in the release of oxytocin when tested during the third trimester of pregnancy. However, the magnitude of the responses was less when compared to responses of lactating women (6). In the current study, oxytocin levels increase in response to teat-stimulation at all times. However, the response at 100 d gestation was relatively small compared to later responses, and the response at 90 d postpartum was less than the response at 30 d. These findings are consistent with the lack of response of virgin heifers to udder massage (5); a transition from no response to full response obviously must occur and the results at 100 d and at <100 d probably are a reflection of this transition period. A decline in milking induced oxytocin release with advanced lactation has previously been reported (4). The difference in responsiveness to teat-stimulation during pregnancy for humans and cattle may be related to the relative rates of mammary development during pregnancy. In contrast with ruminant species, the initiation of milk synthesis in humans is delayed until the immediate postpartum period (22). There has been no systematic study of peripheral catecholamine levels during pregnancy. In this study, baseline norepinephrine values remained constant throughout pregnancy; there was a slight, not statistically significant, rise in baseline at 200 d gestation. Measured values were comparable to those reported for lactating cows (8,10). Baseline epinephrine concentrations at 100 and 150 d gestation were also comparable to previously reported values for lactating cows (8,10). However, concentrations at 200 and 250 d gestation were significantly el-
242
LEFCOURT AND AKERS
evated in comparison. This elevation could conceivable reflect a metabolic adaption to pregnancy and deserves further study. Peripheral catecholamines do not respond to teat-stimulation (milking) in lactating cows (8,10). However, statistically and physiological significant responses are often seen in lactating sheep in response to milking (23) and teat-stimulation (24). No data exist concerning catecholamine responses to teat-stimulation during gestation for any specie. As gestation in nulliparous mammals is a period of pronounced change for the mammary system, it is likely that sympathetic mechanisms involved in the regulation of lactation also undergo a period of transition. Hence, it is reasonable to examine catecholamine responses to teat-stimulation under these conditions. In this study, epinephrine did not respond to teat-stimulation and the tendency for a response in norepinephrine was not statistically significant. However, a significant relationship was found when areas under the oxytocin response curves with baselines subtracted were contrasted with average norepinephrine values with baselines subtracted. Elevated norepinephrine responses (>.2 pmol/ml) following teat stimulation were seen in 28 of 51 trials, and large oxytocin responses (>75 pg/ml/min) were seen predominantly only when norepinephrine responses were low (<=.2 pmol/ml). The relationship between elevated peripheral catecholamines and release of oxytocin in response to teat-stimulation is complex and controversial (1,3,8,10,11). One principle question is whether elevated peripheral concentrations of catecholamines directly inhibit oxytocin release or whether peripheral levels are merely indicative of a general increase in sympathetic activity. An general increase in sympathetic activity would most likely include activation of central mechanisms inhibitory to the release of oxytocin. In this study, considering the timing of release of both hormones, the relationship probably is not causal. The possible existence of a norepinephrine response to teat-stimulation during pregnancy in contrast to the lack of such a response during lactation may be due to one of two reasons. First, manual teatstimulation in pregnant heifers is a novel experience and the response may be extinguished with experience. Alternatively, the response may be due to the relative rates of development of central mechanisms involved in the control of oxytocin secretion during gestation in the nulliparous animal. The elevation in baseline epinephrine at 200 and 250 d of gestation supports the concept of development and change in central sympathetic control mechanisms. That oxytocin responses to teat-stimulation were greatest at 150 d gestation, when epinephrine levels were still low, suggests that stimulatory mechanisms responsible for the release of oxytocin develop and/or are expressed prior to the development of inhibitory sympathetic mechanisms. In conclusion, we have clearly shown that mechanisms for the release of oxytocin in response to teat-stimulation are developed during early gestation in nulliparous cows. Furthermore, our data strongly suggests that central sympathetic control mechanisms, perhaps related to the integrative control of oxytocin release, continue to develop throughout the gestational period.
ACKNOWLEDGEMENTS/FOOTNOTES LMilkSecretionand Mastitis Laboratory,Building 173, USDA.To whomcorrespondenceshouldbe addressed. ZCurrentAddress: VirginiaPolytechnicInstituteand State University,Departmentof DairyScience, Blacksburg, VA24061
REFERENCES 1. Lefcourt AM, Akers RM. is oxytocin really necessary tbr efficient milk removal in dairy cows? J Dairy Sci 66:2251-2259, 1983. 2. Mayer MH, Schams D, Worstorff H, Karg H. Oxytocin release during prestimulation and milking and its significance for milk removal in dairy cows. Vlamms Diergeneeskundig Tijdschrift 55:286296, 1986.
HORMONAL RESPONSESTO TEAT STIMULATION
243
3. Mena E Carmen C, Martinez-Escalera G, Pacheco P, Grosvenor CE. Integrative regulation of milk ejection. In: Oxytocin: Clinical and Laboratory Studies, Amico JA and Robinson AG (eds). Elsevier, New York, p. 179-199, 1985. 4. Wachs EA, Gorewit RC, Currie WB. Oxytocin concentrations of cattle in response to milking stimuli through lactation and mammary involution II. Domest Anim Endocrinol 1:14 l-154, 1984. 5. Gorewit RC. Physiological and pharmacological studies of oxytocin in virgin cattle IH. Domest Anim Endocrinol 1:155-165, 1984. 6. Amico JA, Finley BE. Breast stimulation in cycling women, pregnant women and a woman with induced lactation: Pattern of release of oxytocin, prolactin and luteinizing hormone. Clin Endocrino125:97-106, 1986. 7. Roberts JS, Share L. Oxytocin in plasma of pregnant, lactating and cycling ewes during vaginal stimulation. Endocrinology 83:272-278, 1968. 8. Blum JW, Schams D, Bruckmaier R. Catecholamines, oxytocin and milk removal in dairy cows. J Dairy Res 56:167-177, 1989. 9. Gorewit RC, Aromando MC. Mechanisms involved in the adrenalin-induced blockade of milk ejection in dairy cattle. Proc Soc Exp Biol Med 180:340-347, 1985. 10. Lefcourt AM, Akers RM. Small increases in peripheral noradrenaline inhibit the milk-ejection responses by means of a peripheral mechanism. J Endocrinol 100:337-344, 1984. I I. Lincoln DW, Paisley AC. Neuroendocrine control of milk ejection. J Reprod Fertil 65:571-586, 1982. 12. Lefcottrt AM, Akers RM. Oxytocin and catecholamine responses to teat stimulation in Holstein heifers. J Dairy Sci 68 [Suppl 1]:171, 1985. 13. Gorewit RC. Methods for determining oxytocin concentrations in unextracted sera: characterization in lactating cattle. Proc Soc Exp Biol Med 160:80-87, 1979. 14. Bolt DJ. Development of a homologous radioimmunoassay for ovine follicle stimulating hormone: studies after estrus, ovariectomy, estradiol and releasing hormone. J Anim Sci 53:730-741, 1981. 15. Anton A, Sayre D. A study of the factors affecting the aluminum oxide-trihydroxyindoleprocedure for analysis of catecholamines. J Pharm Exp Theraput 138:360-374, 1962. 16. Hjemdahl P, Daleskog M, Kahan T. Determination of plasma catecholamines by high performance liquid chromatography with electrochemical detection: comparison with a radioenzymatic method. Life Sci 25:131-138, 1979. 17. Hegstrand L, Eichelman B. Analysis of catecholamines in rat brain. Altex Chromatogram 3:1-2, 1979. 18. Manns JG, Hafs HD. Controlled breeding in cattle: a review. Can J Anita Sci 56:121-130, 1976. 19. SAS Institute Inc. SAS/STAT Guide for personal computers, Ver. 6, SAS Inst. Inc., Cary, NC, 1987. 20. Afifi AA, Azen SE Statistical Analysis: A Computer Oriented Approach, Academic Press, New York, NY, 1972. 21. Schams D, Lahlou-Kassi A. Circulating concentrations of oxytocin during pregnancy in ewes. Acta Endocrinol 106:277-281, 1984. 22. Vorherr H. Human lactation and breast feeding. In: Lactation: a Comprehensive Treatise, Larson BL (ed). Academic Press, New York, p. 181-280, 1978. 23. Barowitz T. Changes in blood catecholamine levels in sheep during machine milking. J Dairy Res 46:555-557, 1979. 24. Lefcourt AM, Mayer H, Paul G. Catecholamine response to manual teat stimulation prior to milking in Lacanne and Friesen dairy sheep. J Dairy Sci 73 [Suppl l]:201, 1990.
Vol. 8(2):235-243, 1991
TEAT STIMULATION-INDUCED OXYTOCIN AND CATECHOLAMINE RELEASE IN PREGNANT AND LACTATING HOLSTEIN HEIFERS A.M. Lefcou# and R.M. Akers 2 U.S. Department of Agriculture Agriculture Research Service Livestock and Poultry Science Institute Beltsville, Maryland 20705 Received September7, 1990
ABSTRACT The effect of manual teat stimulation (milking paradigm) on release of oxytocin, epinephrine and norepinephrine was studied in (1) 15 heifers at 100, 150, 200 and 250 d of gestation and at 30 and 90 d of lactation (during machine milking) and (2) simultaneously in six heifers at <100 d and in six heifers at >200 d of gestation. Oxytocin responses to teat stimulation, including peak heights and area under the response curves, at 150, 200 and 250 d or at >200 d of gestation were similar and were significantly greater than responses at 100 d or at <100 d. Responses to milking were lower at 90 d compared to responses at 30 d. Catecholamines were measured only during gestation and were generally not affected by teat stimulation. Epinephrine levels were significantly higher at 200 and 250 d compared to levels at 100 and 150 d. Baseline oxytocin concentrations and responses to teat stimulation were greatest at 150 d of gestation when epinephrine levels were still low, suggesting that stimulatory mechanisms responsible for the release of oxytocin develop and/or are expressed prior to the development of inhibitory sympathetic mechanisms. For norepinephrine, linear analyses did not show significant responses to teat stimulation overall. However, elevated norepinephrine responses (>.2 pmol/ml) following teat stimulation were seen in 28 of 51 trims, and large oxytocin responses (>75 pg/ml/min) were seen predominantly only when norepinephrine responses were low (<=.2 pmol/ml). INTRODUCTION In lactating mammals, oxytocin concentrations increase in response to nipple stimulation and this increase is important for efficient milk removal (1,2,3,4). However, little is known about the development of this neuroendocrine reflex. Gorewit (5) found a small, but insignificant, release of oxytocin in 12 to 14 m old virgin cattle following 2 min of vigorous udder massage. In contrast, oxytocin is released in responses to udder massage for at least 30 d after milking is terminated (4). The neuroendocrine reflex responsible for the release of oxytocin in response to teat-stimulation has not been characterized for nulliparous pregnant heifers. In humans, oxytocin responds to breast stimulation during pregnancy but the magnitude of the response is less than in lactating women (6). In ewes, oxytocin increases in response to vaginal stimulation during lactation but not during pregnancy (7). The sympathetic nervous system plays an important role in regulating the efficiency of the milk ejection reflex, although it is not clear whether the primary interaction is centrally or peripherally mediated (1,8,9,10). Even if increased peripheral catecholamine levels do not directly inhibit the release of oxytocin in responses to teat-stimulation (8,9,10), increased peripheral release of catecholamines is probably indicative of an increase in sympathetic arousal. Increased sympathetic activity has been reported to inhibit release of oxytocin (1,3,11). Regardless, peripheral catecholamine levels and catecholamine responses to teatstimulation have not been characterized throughout pregnancy. Consequently, we studied oxytocin and catecholamine release after 30 s of udder wash and 4.5 rain of manual teat stimulation in nulliparous pregnant heifers at 100, 150, 200 and 250 d of gestation and the oxytocin responses of these same heifers to machine milking at 30 and Copydght © 1991 by Domendo, Inc.
235
0739-7240/91/$3.00
236
LEFCOURT AND AKERS
90 d postpartum. To address possible seasonal effects due to changes in ambient temperature or day length, oxytocin responses were measured simultaneously in two groups of cattle either at <100 d or at >200 d of gestation. A preliminary report has been published (12). MATERIALS AND M E T H O D S Blood Sampling. A polyvinyl cannula (0.48 m in length) was placed into a jugular vein of each animal in the direction of the heart, to a depth of 0.32 m, 1-2 d before the start of blood sampling. To accustom animals to the experimental protocol, blood (4 ml) was withdrawn and discarded at approximately 10 min intervals for 1-2 hr before the start of each trial. Blood samples (20 ml) were collected at -10, -5, -1,0, 1, 2, 3, 4, 5, 6, 8, 10, 15 and 30 min relative to the initiation of teat stimulation or milking. Integrated samples were taken during consecutive 1 min intervals (20 rnl/min) starting 1 rain before and ending 7 min after the onset of stimulation. During the week before blood sampling, lactating heifers were milked once each morning in a milking parlor with side-opening stalls, subsequently used on the day of experiment, and once each afternoon in accord with the usual routine with the remainder of the milking herd. This was to allow the animals to become accustomed to the surroundings and milking routine used on the day of sampling. For oxytocin, blood was collected on ice, stored at 4 C overnight and sera prepared by centrifugation the following morning. Sera samples were stored at -20 C until assayed as described by Gorewit (13). Antibody to oxytocin was purchased from Calbiochem; synthetic oxytocin (Calbiochem) was used as standard and was iodinated according to Bolt (14). Aliquots of a pooled serum sample were measured in each assay for calculation of intra- and inter-assay variation which were 5.9% and 3.4%, respectively. For measurement of catecholamines, samples taken at -10, -5, 1, 2, 4 and 8 min were collected using polystyrene syringes containing sodium heparin and sodium metabisulfite and stored on ice. Portions of each sample (10 ml) were cold (4 C) centrifuged and resulting plasma stored at -75 C until assayed for catecholamines. Thawed 1 ml samples of plasma were extracted into perchloric acid using activated alumina (15,16). Catecholamines were quantified using a high-pressure liquid chromatography assay with an ion-pairing column (Beckman) and electrochemical detection (17). Norepinephrine (Sigma) and epinephrine (Sigma) were used as standards. Butanephrine (kindly supplied by Sterling-Winthrop Research Institute, Rensselaer, New York) was used as an internal standard. Coefficients of variation for noradrenaline and adrenaline were 7 and 8%. The lower limit at which catecholamines could be quantified was about 0.4 pmol. With an extraction efficiency of 60%, this meant that concentrations below 0.7 pmol/ml could not be quantified although they were often detected. Most samples were run in duplicate. Unfortunately, due to failure of a freezer, catecholamine samples taken during lactation and for experiment 2 (below) could not be analyzed. Teat Stimulation. In an attempt to standardize the teat stimulation, all teat manipulation was done by one person. After a 30 s udder wash, one pair of front and rear teats was alternately pulled and released (to mimic handmilking) for 30 s. Teat stimulation for contralateral udder halves was alternated at 30 s intervals for a total of 4.5 min. In milking trials, the udder was washed for 30s and the milking machine was attached for 4.5 min. Experiment 1. Fifteen pregnant Holstein heifers, previously treated with prostaglandin F2~ to synchronize estrus (18), were used to determine the effect of stage of gestation and of lactation on oxytocin and catecholamine response to teat stimulation. The heifers were kept in a stanchion barn throughout most of gestation. Just before expected parturition the heifers were transferred to a maternity barn and, after calving, were maintained with the remainder of the lactating herd. Three heifers were eliminated from the experiment after parturition be-
237
HORMONAL RESPONSES TO TEAT STIMULATION
cause of calving difficulty or severe mastitis. Blood samples were taken at 100 (in December), 150, 200 and 250 (in May) d of gestation, and at 30 and 90 d after calving. Experiment 2. Twelve pregnant Holstein heifers (six < 100 and six >200 d of gestation) were used to determine whether responses in experiment l were due to stage of gestation or season. Four animals were tested on each of 3 consecutive days in October. For 2 weeks before the experiment started, the animals were housed in stanchions at the same location as the heifers in experiment 1. Statistical Analyses. Oxytocin, norepinephrine and epinephrine data were analyzed using a linear model (Procedure GLM,19). The model included factors for TIME of sample, STAGE of gestation or lactation, TIME by STAGE and, for experiment l, COW. In addition, average baseline values for each hormone were determined by averaging data by cow for times prior to stimulation. For oxytocin, the peak value, time to peak value, and area under the response curve (time weighted averages of values between 1 and 8 min) also were calculated. For catecholamines, the peak value and the average of values for l, 2 and 4 rain were computed. Measures calculated with individual baselines subtracted also were analyzed. Results with baselines subtracted were generally similar to those where baselines were not subtracted, and are shown only where differences exist. To examine relationships among measured variables, data were analyzed for correlations and for differences in frequencies of responses using Chi-square (20). RESULTS
Oxytoein: Experiment 1. Oxytocin responses to teat stimulation at 150, 200 and 250 d of gestation were similar and were significantly greater (P<.0 l) than responses at 100 d (Figure 1). This difference in responsiveness was also apparent in peak responses and area under response curves (Table 1). During lactation, oxytocin responses to milking were significantly (P < .05) reduced at 90 d compared to 30 d (Figure 2), as were peak responses and area under response curves. COW was a significant (P < .01) factor in all analyses.
400 -
/~~ Gestation I ~
Oxytocin
teat stimulation I~ I I
300350
I
m
E Q.
2oo2° l 150
------.._=_ ]
~-
w
=-111
v
250 days
Odays l
n"
200 days
U
100 days
- Standard Error of the Means
100 -10
= .... -5
I .... 0
= .... 5
i .... 10
I .... 15
= .... 20
I .... 25
= 30
Time (rain) Figure I. Leastsquaremeansof oxytocinconcentrationsin relationto teatstimulationduringgestationin 15 heifers.
238
LEFCOURT AND AKERS TABLE 1. OXYTOCINMEASURES RELATEDTO TEAT STIMULATIONDURING GESTATION,OR MILKINGDURINGLACTATION,IN ]5 HEIFERS.TO TEST SEASONAL EFFECTS, T w o GROUPS OF 6 HEIFERS EITHERAT <100 D OR AI >200 D OF GESTATIONWERE EVALUATEDSIMULTANEOUSININ A SINGLESEASON.
Stage o f Gestation
Oxytocin Peak Time of Peak (pg/ml) (rain)
Baseline (pg/ml)
100d 150 d 200d 250 d
176± 234 ± 197o+ 206 ±
Lactation 30d 90d
10 ~' l0 b I I"." 10 h
Area (pg/ml/min)
239 + 3T' 485 _+37 ~ 356+39 h 374 0+ 37"
3.6±.3 3.6 ± .3 3.1 ± . 4 3.8 _+ .3
201 334 271 275
± 17" ± 17 ~ 0+ 18 ~ _+ 17 h
251 ± 10 d 2 1 9 ± 12
4 0 4 ± 15" 3 5 3 ± 17'
3.6 ± . 3 3.3_+.3
3 1 6 ± I1" 2 6 9 ± 12'
131 ± 2 7 " 201 ± 27 h
210-+50 330 -+ 50
4.0_+.6 3.5 _+ .6
168±35" 267 _+ 35 b
Gestation <100d >200 d
"'b~Least square means are shown 0+SE and, tbr each stage, are different within a column stage when superscripts differ. ~JThe difference between 30 d and 90 d baselines was almost significant (P < .()6L
There was no apparent correlation between oxytocin concentrations and peripheral levels of either catecholamine. To examine whether oxytocin responses might be less when catecholamine levels were elevated, 2 by 2 contingency tables were constructed post hoc based on oxytocin and catecholamine levels. A significant (P<.01) relationship was found when areas under the oxytocin response curves with baselines subtracted were contrasted with average norepinephrine values with baselines subtracted (Table 2). Large oxytocin responses (>75 pg/ml/min) were seen predominantly only when norepinephrine responses were low (<=.2 pmol/ml).
400
Oxytocin t~on
milking 350
300 30 days 8
250
i I
200
90 days
150
T -
100
. . . .
-10
Standard Error of the Means
I
-5
. . . .
|
0
. . . .
i
5
. . . .
I
. . . .
10
Time
I
15
. . . .
i
20
. . . .
I
. . . .
25
(min)
Figure 2. Least square means of oxytocin concentrations in relation to milking during lactation in 12 heifers.
I
30
HORMONALRESPONSESTO TEAT STIMULATION
239
TABLE 2. OXYTOCIN, AREA UNDER RESPONSES CURVES WITH BASELINESSUBTRACTED,CONTRASTED IN A 2 BY 2 CONTINGENCYTABLE WITH AVERAGENOREPINEPHRINERESPONSES TO TEAT STIMULATION, ALSO WITH BASELINES SUBTRACTED. DATA INCLUDEMEASUREMENTS FROM 100 D, 150 D, 200 D AND 250 D OF GESTATIONFOR 15 HEIFERS.
Oxytocin (pg/ml/min) <=75 26 14
Norepinephrine >.2 (pmol/ml) <=.2 "Chi-square= 7.64, 1 degreeof freedomP<.01
>75 2 9
Oxytocin: Experiment 2. The oxytocin responses to teat stimulation were significantly (P<.05) greater in heifers at > 200 d gestation compared to heifers at <100 d (Figure 3; Table 1). Baseline oxytocin levels were lower in the < 100 d animals compared to the 100 d animals from experiment 1; however, other measures including area under response curves with baselines subtracted were similar. Responses at >200 d gestation fell in the middle of responses at 150 to 250 d gestation in experiment 1. Epinephrine: Experiment 1. Epinephrine data are biased by the low levels detected in early gestation and lack of assay sensitivity. Data points with values below detectable limits were excluded from analyses. Even so, epinephrine levels and baseline values were significantly (p < .05) higher at 2130 and 250 d gestation compared to 100 and 150 d (Figure 4; Table 3). Teat stimulation did not affect epinephrine levels. COW was a significant (P < .01) factor in all analyses. Due to failure of a freezer, catecholamine samples taken during lactation and for experiment 2 could not be analyzed.
40oI
Oxytocin
Gestation
within a season
350
300
E
teat stimulation
-
~~
~
- StandardError of the Means
250
200
150 w
100
. . . .
-10
I
-5
. . . .
I
0
. . . .
I
5
. . . .
I
. . . .
10
I
15
. . . .
I
20
....
I ....
25
I
30
Time (min) Figure 3. Least square means of oxytocinconcentrationsin relation to teat stimulationfor two groupsof 6 heifers at <100 d and at >200 d of gestation. To test seasonal effects, both groups were evaluated simultaneouslyin a single season.
240
LEFCOURT AND AKERS 2.0
Epinephrine Gestation teat stimulation
1.5 200 days m
E 0
E Q.
250 days 1.0
0.5
0.0
i
150 days
l
100 days
i
-10
-5
0
.I.
5
10
Time (min) Figure 4. Least square means of epinephrine concentrations in relation to teat stimulation during gestation in 15 heifers. No estimate is shown lbr -5 min at 150 d as the predicted value is negative. This negative value was due to lack of assay sensitivity and the small n u m b e r of samples in which epinephrine could be detected tor this time point.
Norepinephrine: Experiment 1. Norepinephrine values did not vary by stage of gestation or in response to teat stimulation; however, there was a tendency for levels to be elevated at 200 d (Figure 5; Table 3). COW was a significant (P < .01 ) factor in all analyses. Even though linear analyses did not show a significant responses to teat stimulation overall, examination of the data revealed that a number of animals showed evidence of a response. A tendency for a response can be seen even in the mean data at 100, 150 and 200 d gestation (Figure 5). When the data were analyzed using contingency tables (see above), elevated norepinephrine responses (>.2 pmol/ml) following teat stimulation were seen in 28 of 51 trials (Table 2). DISCUSSION Despite the importance of oxytocin to reproduction and lactation, little data exist concerning oxytocin values during pregnancy and almost nothing is known about the development of the neurohumoral reflex for the release of oxytocin in response to teat-stimulation. In one of
TABLE 3. CATECHOLAMINEMEASURES RELATEDTO TEAT STIMULATIONDURINGGESTATIONIN 15 HEIP~b.RS. Stage of
Epinephrine
Gestation
(pmol/ml) Peak
Average'
Baseline
.92 ± .29 ~'" .49 ± .32 ~' 1.56 -+ .29" 1.25 --- .34 "b
.86 ± .26 "b .24 _+ .35" I. 15 + .25 h 1.06 ± .29 "~
I. 18 _+ .28 I. 16 _+ .23 1.84 _+ .27 1.12 ± .24
100 d 150 d 200 d 250 d
Baseline .34 .07 1.29 1.14
_+ .32" ± .39 ~' ± .29 ~ ± .30 h
Norepinephrine (pmolhnl) Peak 1.86 + 1.77 + 2.89 ± 1.91 ±
.38 "b .33" .36 b .35 ~''~
~hLeast square means are shown ± SE and are different within a column when superscripts differ. ~Average of values at I. 2 and 4 rain tbllowing teat stimulation.
Average 1.66 ± 1.37 ± 2.16 ± 1.53 ±
.33 .28 .3 I .30
HORMONAL RESPONSES TO TEAT STIMULATION
3.0
241
Norepinephrine Gestation
2.5
T teat stimulation
T
~L=
2.0
E o 1.5
100 days
E Q.
150 days
1.0
0.5
0.0
I
-10
.
.
.
.
I
I
-5
0
.
.
.
.
I
I
5
10
Time (rain) Figure 5. Least square means of norepinephrine concentrations in relation to teat stimulation during gestation in 15 heifers.
two ewes, circulating oxytocin concentrations were elevated during the first 2 wk of pregnancy and then fell and remained constant even after parturition. In the other ewe, levels were somewhat elevated during pregnancy and peaked prior to parturition (21). In the current study, baseline oxytocin concentrations tended to be lower during early gestation and constant thereafter during gestation and postpartum. In humans, breast stimulation resulted in the release of oxytocin when tested during the third trimester of pregnancy. However, the magnitude of the responses was less when compared to responses of lactating women (6). In the current study, oxytocin levels increase in response to teat-stimulation at all times. However, the response at 100 d gestation was relatively small compared to later responses, and the response at 90 d postpartum was less than the response at 30 d. These findings are consistent with the lack of response of virgin heifers to udder massage (5); a transition from no response to full response obviously must occur and the results at 100 d and at <100 d probably are a reflection of this transition period. A decline in milking induced oxytocin release with advanced lactation has previously been reported (4). The difference in responsiveness to teat-stimulation during pregnancy for humans and cattle may be related to the relative rates of mammary development during pregnancy. In contrast with ruminant species, the initiation of milk synthesis in humans is delayed until the immediate postpartum period (22). There has been no systematic study of peripheral catecholamine levels during pregnancy. In this study, baseline norepinephrine values remained constant throughout pregnancy; there was a slight, not statistically significant, rise in baseline at 200 d gestation. Measured values were comparable to those reported for lactating cows (8,10). Baseline epinephrine concentrations at 100 and 150 d gestation were also comparable to previously reported values for lactating cows (8,10). However, concentrations at 200 and 250 d gestation were significantly el-
242
LEFCOURT AND AKERS
evated in comparison. This elevation could conceivable reflect a metabolic adaption to pregnancy and deserves further study. Peripheral catecholamines do not respond to teat-stimulation (milking) in lactating cows (8,10). However, statistically and physiological significant responses are often seen in lactating sheep in response to milking (23) and teat-stimulation (24). No data exist concerning catecholamine responses to teat-stimulation during gestation for any specie. As gestation in nulliparous mammals is a period of pronounced change for the mammary system, it is likely that sympathetic mechanisms involved in the regulation of lactation also undergo a period of transition. Hence, it is reasonable to examine catecholamine responses to teat-stimulation under these conditions. In this study, epinephrine did not respond to teat-stimulation and the tendency for a response in norepinephrine was not statistically significant. However, a significant relationship was found when areas under the oxytocin response curves with baselines subtracted were contrasted with average norepinephrine values with baselines subtracted. Elevated norepinephrine responses (>.2 pmol/ml) following teat stimulation were seen in 28 of 51 trials, and large oxytocin responses (>75 pg/ml/min) were seen predominantly only when norepinephrine responses were low (<=.2 pmol/ml). The relationship between elevated peripheral catecholamines and release of oxytocin in response to teat-stimulation is complex and controversial (1,3,8,10,11). One principle question is whether elevated peripheral concentrations of catecholamines directly inhibit oxytocin release or whether peripheral levels are merely indicative of a general increase in sympathetic activity. An general increase in sympathetic activity would most likely include activation of central mechanisms inhibitory to the release of oxytocin. In this study, considering the timing of release of both hormones, the relationship probably is not causal. The possible existence of a norepinephrine response to teat-stimulation during pregnancy in contrast to the lack of such a response during lactation may be due to one of two reasons. First, manual teatstimulation in pregnant heifers is a novel experience and the response may be extinguished with experience. Alternatively, the response may be due to the relative rates of development of central mechanisms involved in the control of oxytocin secretion during gestation in the nulliparous animal. The elevation in baseline epinephrine at 200 and 250 d of gestation supports the concept of development and change in central sympathetic control mechanisms. That oxytocin responses to teat-stimulation were greatest at 150 d gestation, when epinephrine levels were still low, suggests that stimulatory mechanisms responsible for the release of oxytocin develop and/or are expressed prior to the development of inhibitory sympathetic mechanisms. In conclusion, we have clearly shown that mechanisms for the release of oxytocin in response to teat-stimulation are developed during early gestation in nulliparous cows. Furthermore, our data strongly suggests that central sympathetic control mechanisms, perhaps related to the integrative control of oxytocin release, continue to develop throughout the gestational period.
ACKNOWLEDGEMENTS/FOOTNOTES LMilkSecretionand Mastitis Laboratory,Building 173, USDA.To whomcorrespondenceshouldbe addressed. ZCurrentAddress: VirginiaPolytechnicInstituteand State University,Departmentof DairyScience, Blacksburg, VA24061
REFERENCES 1. Lefcourt AM, Akers RM. is oxytocin really necessary tbr efficient milk removal in dairy cows? J Dairy Sci 66:2251-2259, 1983. 2. Mayer MH, Schams D, Worstorff H, Karg H. Oxytocin release during prestimulation and milking and its significance for milk removal in dairy cows. Vlamms Diergeneeskundig Tijdschrift 55:286296, 1986.
HORMONAL RESPONSESTO TEAT STIMULATION
243
3. Mena E Carmen C, Martinez-Escalera G, Pacheco P, Grosvenor CE. Integrative regulation of milk ejection. In: Oxytocin: Clinical and Laboratory Studies, Amico JA and Robinson AG (eds). Elsevier, New York, p. 179-199, 1985. 4. Wachs EA, Gorewit RC, Currie WB. Oxytocin concentrations of cattle in response to milking stimuli through lactation and mammary involution II. Domest Anim Endocrinol 1:14 l-154, 1984. 5. Gorewit RC. Physiological and pharmacological studies of oxytocin in virgin cattle IH. Domest Anim Endocrinol 1:155-165, 1984. 6. Amico JA, Finley BE. Breast stimulation in cycling women, pregnant women and a woman with induced lactation: Pattern of release of oxytocin, prolactin and luteinizing hormone. Clin Endocrino125:97-106, 1986. 7. Roberts JS, Share L. Oxytocin in plasma of pregnant, lactating and cycling ewes during vaginal stimulation. Endocrinology 83:272-278, 1968. 8. Blum JW, Schams D, Bruckmaier R. Catecholamines, oxytocin and milk removal in dairy cows. J Dairy Res 56:167-177, 1989. 9. Gorewit RC, Aromando MC. Mechanisms involved in the adrenalin-induced blockade of milk ejection in dairy cattle. Proc Soc Exp Biol Med 180:340-347, 1985. 10. Lefcourt AM, Akers RM. Small increases in peripheral noradrenaline inhibit the milk-ejection responses by means of a peripheral mechanism. J Endocrinol 100:337-344, 1984. I I. Lincoln DW, Paisley AC. Neuroendocrine control of milk ejection. J Reprod Fertil 65:571-586, 1982. 12. Lefcottrt AM, Akers RM. Oxytocin and catecholamine responses to teat stimulation in Holstein heifers. J Dairy Sci 68 [Suppl 1]:171, 1985. 13. Gorewit RC. Methods for determining oxytocin concentrations in unextracted sera: characterization in lactating cattle. Proc Soc Exp Biol Med 160:80-87, 1979. 14. Bolt DJ. Development of a homologous radioimmunoassay for ovine follicle stimulating hormone: studies after estrus, ovariectomy, estradiol and releasing hormone. J Anim Sci 53:730-741, 1981. 15. Anton A, Sayre D. A study of the factors affecting the aluminum oxide-trihydroxyindoleprocedure for analysis of catecholamines. J Pharm Exp Theraput 138:360-374, 1962. 16. Hjemdahl P, Daleskog M, Kahan T. Determination of plasma catecholamines by high performance liquid chromatography with electrochemical detection: comparison with a radioenzymatic method. Life Sci 25:131-138, 1979. 17. Hegstrand L, Eichelman B. Analysis of catecholamines in rat brain. Altex Chromatogram 3:1-2, 1979. 18. Manns JG, Hafs HD. Controlled breeding in cattle: a review. Can J Anita Sci 56:121-130, 1976. 19. SAS Institute Inc. SAS/STAT Guide for personal computers, Ver. 6, SAS Inst. Inc., Cary, NC, 1987. 20. Afifi AA, Azen SE Statistical Analysis: A Computer Oriented Approach, Academic Press, New York, NY, 1972. 21. Schams D, Lahlou-Kassi A. Circulating concentrations of oxytocin during pregnancy in ewes. Acta Endocrinol 106:277-281, 1984. 22. Vorherr H. Human lactation and breast feeding. In: Lactation: a Comprehensive Treatise, Larson BL (ed). Academic Press, New York, p. 181-280, 1978. 23. Barowitz T. Changes in blood catecholamine levels in sheep during machine milking. J Dairy Res 46:555-557, 1979. 24. Lefcourt AM, Mayer H, Paul G. Catecholamine response to manual teat stimulation prior to milking in Lacanne and Friesen dairy sheep. J Dairy Sci 73 [Suppl l]:201, 1990.