| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
research-article |
Eutheria Foundation,3 Cross Plains, Wisconsin 53528
Animal Health and Biomedical Sciences,4 University of Wisconsin, Madison, Wisconsin 53706
ABSTRACT
Luteal blood flow was studied in heifers by transrectal color-Doppler ultrasound. Data were normalized to the decrease in plasma progesterone to <1 ng/ml (Day 0 or Hour 0). Blood flow in the corpus luteum (CL) was estimated by the percentage of CL area with color flow signals. Systemic prostaglandin F2alpha (PGF) treatment (25 mg; n = 4) resulted in a transient increase in CL blood flow during the initial portion of the induced decrease in progesterone. Intrauterine treatment (1 or 2 mg) was done to preclude hypothetical secondary effects of systemic treatment. Heifers were grouped into responders (luteolysis; n = 3) and nonresponders (n = 5). Blood flow increased transiently in both groups; induction of increased blood flow did not assure the occurrence of luteolysis. A transient increase in CL blood flow was not detected in association with spontaneous luteolysis when examinations were done every 12 h (n = 6) or 24 h (n = 10). The role of PGF pulses was studied by examinations every hour during a 12-h window each day during expected spontaneous luteolysis. At least one pulse of 13,14-dihydro-15-keto-PGF2alpha (PGFM) was identified in each of six heifers during the luteolytic period (Hours –48 to –1). Blood flow increased (P < 0.02) during the 3-h ascending portion of the PGFM pulse, remained elevated for 2 h after the PGFM peak, and then decreased (P < 0.03) to baseline. Results supported the hypothesis that CL blood flow increased and decreased with individual PGFM pulses during spontaneous luteolysis.
blood flow, cattle, corpus luteum function, female reproductive tract, luteolysis, mechanisms of hormone action, ovulatory cycle, prostaglandin F2
, steroid hormones
Secretion of prostaglandin F2 (PGF) by the uterus, augmented by intraluteal PGF production, terminates the luteal phase in cattle as in many other species [1, 2]. Studies involving placement of transducers on the ovarian artery and insertion and entrapment of radioactive microspheres of various diameters have shown that blood flow to the corpus luteum (CL) decreases dramatically in association with luteolysis [3]. It has been elusive as to whether the decreased luteal blood flow preceded or accompanied the luteolytic process [4]. However, rather than a decrease in luteal blood flow as an initial step in the luteolytic process, recent color-Doppler studies in cattle indicated that luteal blood flow initially and transiently increased during or prior to a decrease in plasma progesterone during exogenous PGF-induced luteolysis [5] and spontaneous luteolysis [6].
In the study with induced luteolysis [5], a single injection of a PGF analogue (cloprostenol) was used. The ratio of colored area from color-Doppler signals of blood flow on an image of the CL was used as a quantitative index of blood flow. Progesterone concentration decreased significantly between 0 h and 1 h after treatment, and blood flow transiently increased between 0 h and 0.5 h and then decreased after 2 h. Thus, blood flow increased while progesterone was decreasing. In the study with spontaneous luteolysis [6], blood sampling and blood flow determinations were made every 12 h. Blood flow increased between 16 and 17 days postestrus and then decreased, a PGF metabolite (13,14-dihydro-15-keto-PGF2; PGFM) was elevated at 17 and 18 days, and plasma progesterone began to decrease at 18 days. The authors concluded that luteal blood flow increased before progesterone decreased and proposed that the acute increase of intraluteal blood flow is a universal phenomenon prior to spontaneous luteolysis.
Contrasting with the reports of an association between a transient increase in blood flow and the initiation of luteolysis in cattle, recent color Doppler studies in horses did not find a similar blood flow phenomenon in association with either PGF-induced [7] or spontaneous [8] luteolysis. Treatment with PGF caused a progesterone increase within 5 min that remained elevated until 10 min and then decreased. A change in luteal blood flow did not occur until a decrease occurred at 24 h. Luteolysis, as indicated by decreasing progesterone, began well before the beginning of a detectable decrease in blood flow. The temporal association between changes in blood flow and progesterone concentrations during spontaneous luteolysis was studied at 24-h intervals. There was no indication that either an acute increase or decrease in luteal blood flow occurred prior to or at the beginning of the precipitous decrease in progesterone concentration in mares.
Intrauterine administration of PGF is effective in cattle for inducing luteolysis [9]; the minimal effective dose (1 or 2 mg [10]) is about one tenth of the systemic dose (15 mg [11]) when given into the uterine horn ipsilateral to the CL. The intrauterine effectiveness results from a unilateral utero-ovarian luteolytic pathway in cattle [12, 13]; the PGF is transferred into the ovarian artery after intrauterine administration [14] through a local venoarterial pathway between the utero-ovarian vein and the ovarian artery [15].
The main plasma metabolite of PGF is PGFM, and assay of PGFM is an indicator of PGF release into the circulation [16]. The half-life of PGFM is only about 8 min, but it is much longer than for PGF. A sampling interval of 1 h has been recommended for detection of PGFM pulses [16]. A series of PGFM pulses occurred during luteolysis, and the first pulse was followed immediately by a decrease in progesterone [17]. The series of PGFM pulses occurred over 2 to 3 days with a mean of 5 pulses, and each pulse had a duration of 1 to 5 h or a mean of 4 h [17, 18]. The pulsatile release of PGF from the CL as well as from the uterus during spontaneous luteolysis in the cow has been reported [19]. The hormonal regulation accounting for the stimulation of PGF pulses and the interval between pulses is complex and has been reviewed [20, 21]. In this regard, treatment with oxytocin on Day 17 in cows stimulates a surge of PGFM with characteristics that seem similar to a spontaneous PGFM pulse, except for a more rapid increase [22].
The rationale, objective, and sequence for the experiments in the present series were as follows: 1) The process of an increase in luteal blood flow prior to the progesterone decrease during spontaneous luteolysis that has been reported in cattle [6] was not detected in horses [8]; the objective of experiment 1 was to confirm the occurrence of the process in cattle. 2) The existence of the reported process during spontaneous luteolysis in cattle was not confirmed in experiment 1; the objective of experiment 2 was to confirm its reported [5] existence during induced luteolysis. 3) The process during induced luteolysis was confirmed in experiment 2, but the induced regression involved systemic treatment that may have produced increased blood flow by a secondary systemic effect; the objective of experiment 3 was to determine whether local intrauterine administration of a low dose of PGF would induce the process of an early blood flow increase in association with luteolysis. 4) Local intrauterine treatment also stimulated the process in experiment 3. Therefore, hourly sampling was done in experiment 4 to test the hypothesis that CL blood flow increased and decreased in close association with each individual PGFM pulse during spontaneous luteolysis.
Animals were handled in accordance with the United States Department of Agriculture guide for Care and Use of Agricultural Animals in Research. Holstein heifers, aged 17 to 20 mo, were used in the four experiments. Body condition for each heifer was high throughout the experiments. Heifers were selected with docile temperament and no apparent abnormalities of the reproductive tract, as determined by ultrasound examinations [23]. The heifers were acclimated to the handling procedures for 2 wk prior to experimentation. Heifers were sedated with a low dose of xylazine (14 mg) in experiment 1 but not in experiments 2, 3, and 4. In experiments 2 and 3, treatment was done 10 days after ovulation, which is well before expected luteolysis and conforms with the recently reported experiment [5].
A duplex B-mode (grayscale) and pulsed-wave color-Doppler ultrasound instrument (Aloka SSD 3500; Aloka America, Wallingford, CT) equipped with a linear-array, 7.5-MHz transducer was used for transrectal scanning. In color-Doppler mode, the extent and direction of blood flow in the vessels are indicated by color signals [24] and were used to display signals for blood flow in vessels of the CL. All Doppler scans were performed at a constant gain setting for color. A velocity setting of 10 cm/sec was used for experiments 1, 2, and 3, and a setting of 6 cm/sec was used for experiment 4. The lowest velocity setting available on the instrument was 6 cm/sec. The effectiveness of the settings on minimizing detection of venous flow in the CL in these studies is unknown. Real-time B-mode/color-Doppler images of the continuous scans were captured with an online digital videotaping system and stored for potential validation and confirmation purposes. Percentage of CL with color-Doppler signals for blood flow was estimated from the blood flow color displays of the real-time sequential two-dimensional planes of the entire CL, as described for mares [7]. The color flow signals at the periphery of the CL and within the CL were included in the percentage estimate. Blood flow was normalized to the examination in each heifer when the decreasing plasma progesterone concentrations reached <1 ng/ml (Day 0 or Hour 0).
Experiment 1. Spontaneous Luteolysis
Progesterone concentrations and percentage of CL with blood flow were determined daily in 10 heifers. Normalized data were evaluated from Days –7 to +2. In addition, the ovulation that preceded the experiment also was used as a reference, and data were evaluated from 8 to 18 days after ovulation; 18 days was the earliest day of the next ovulation. Ovulation was used as an additional reference in this experiment so that comparisons could be made to the reported results [6].
Validation of the procedure for estimating the percentage of CL with blood flow signals was done for the first five heifers in the experiment, and Days –5 to +2 were used. Selection of a still image from the film clips for pixel assessment was done without prior knowledge by the operator selecting the images of source or the results of the real-time evaluations. An image was selected that appeared to represent the largest cross-section of the CL with distinctive colored areas. The number of colored pixels in the captured image of the CL was determined as described [24] by an operator without knowledge of source or the real-time percentage estimates until the pixel evaluations for all images were completed. The percentage of colored pixels was calculated using the sum of the number of colored and noncolored pixels as the denominator. For comparison of day effects among end points, the actual values for progesterone concentration, percentage of CL with blood flow, and percentage of the selected CL image with colored pixels for each day were converted to the percentage of the maximum value for each heifer within each end point, as described [8].
Experiment 2. Systemically Induced Luteolysis
Heifers were treated with either physiologic saline or a PGF analogue (500 µg; cloprostenol, Estrumate; Schering-Plough Animal Health Corp., Union, NJ). Treatment was given i.m. 10 days after ovulation (n = 4 per group). The operator did not know the treatment group. The treatment product, day, and route were patterned after the previously reported study [5]. Blood flow examinations and blood sampling were done at 24 h before treatment, just before treatment, and at 0.5, 1, 2, 3, 4, 6, 8, 12, 24, and 48 h after treatment.
Experiment 3. Locally Induced Luteolysis
Natural PGF2 (dinoprost tromethamine, Lutalyse; Pharmacia & Upjohn Co., Kalamazoo, MI) was used in a single systemic or intrauterine treatment. Four groups of heifers (4 heifers per group) were used for systemic administration of 25 mg i.m. or intrauterine administration of 0, 1, or 2 mg. The operator did not know the treatment group. The intrauterine treatments were given in a volume of 0.4 ml physiologic saline inserted by the transcervical route into the middle of the horn ipsilateral to the CL. Blood samples and blood flow data were collected at 0, 2, 24, and 48 h after treatment. The area (cm2) of a cross-section of the CL for each examination was determined in B-mode from the maximum area averaged from two still images using the scanner's tracing function. Percentage change from the hour of treatment was used for progesterone, owing to significant differences among groups just before treatment.
In addition to obtaining color flow images, spectral Doppler examinations were done at 0, 2, and 24 h. For this purpose, a sample cursor or gate 1 mm wide was placed on the most prominent intraovarian color flow signal in the CL-containing ovary. The setting for the range of detection of flow velocity was adjusted for each examination to obtain the optimal spectral graph. A Doppler spectrum with three cardiac cycles was generated, and one of the cycles was used for spectral measurements of vascular indices and flow velocities. This was done a second time, and the mean of the two measurements was used in the statistical analyses. The angle of insonation between the ultrasound beams and direction of blood flow was unknown. However, the vascular indices of pulsatility index (PI) and resistance index (RI) are independent of angle of insonation [25]. True velocities cannot be determined without adjusting for angle of insonation, but relative peak systolic velocity (PSV) and time-averaged maximum velocity (TAMV) are useful for comparative purposes [25]. The formulas and meanings for the spectral end points are well established and are discussed [24, 25].
The experimental period extended from 14 days after ovulation until the day that CL blood flow decreased to <20% of CL area. Six heifers were used. A blood sample was collected and estimation of the percentage of CL area with blood flow was determined at 0800 and 2000 hours. Data obtained every 12 h were normalized to the period when progesterone first decreased to <1 ng/ml, regardless of whether <1 ng/ml occurred at 0800 or 2000 hours. Beginning at 15 days postovulation, blood samples and blood flow estimates were taken every hour for 12 h. The set of hourly examinations began at 0800 hours and ended at 2000 hours, yielding 13 hourly examinations. The set of hourly examinations was obtained each day; the last set was obtained on the day before blood flow first decreased to <20% of CL area at the 0800-hour examination.
PGFM data from each set of 13 hourly samples were evaluated for the presence of complete PGFM pulses. A pulse in PGFM was differentiated from variation resulting from extraneous factors, as described for detection of FSH pulses [26]. The coefficient of variation (CV) of the values composing the ascending and descending portions of the suspected pulse had to be at least three times higher than the mean intra-assay CV. In addition, a pulse had to include values from at least 4 h. Values for blood flow and progesterone concentration were normalized to the peak of each identified PGFM pulse.
PGFM pulses were assigned to a preluteolytic period (Hours –72 to –49), a luteolytic period (Hours –48 to –1), and a postluteolytic period (Hours 0 to 23). The luteolytic period was based, retrospectively, on the mean period of a progressive decrease in progesterone concentration before reaching <1 ng/ml. The PGFM pulse was assigned to the period containing the peak. When more than one pulse was identified for a set of hourly samples, the first pulse was used in the analyses.
Blood Samples and Hormone Assays
Blood samples were collected into heparinized tubes and centrifuged (2000 x g for 10 min), and plasma was decanted and stored (–20°C) until assay. Plasma progesterone concentrations were measured using a solid-phase radioimmunoassay kit containing antibody-coated tubes and 125I-labeled progesterone (Coat-A-Count Progesterone; Diagnostic Products Corp., Los Angeles, CA). The procedure has been described in detail for mare plasma in our laboratory [27] and was validated for assaying plasma concentrations of progesterone in bovine plasma. Serial volumes of a pool of diestrus bovine plasma (50–300 µl) were processed as for experimental samples and resulted in a displacement curve that was similar to the standard curve. The samples from experiments 1 and 2 were analyzed in one assay. The intraassay CV and sensitivity were 3.04% and 0.03 ng/ml, respectively. The intraassay CV and sensitivity for experiment 3 were 4.06% and 0.05 ng/ml, respectively; for experiment 4 they were 5.39% and 0.04 ng/ml, respectively.
Blood samples for PGFM assay (experiment 4) were collected and immediately placed in ice-cold water for 10 min before centrifuging and storing at –20°C. The plasma samples were assayed for PGFM by a modification of a radioimmunoassay procedure [22, 28] that was adapted and validated in our laboratory. The PGFM (catalog number D4143; Sigma Chemical Co., St. Louis, MO) standard solutions were made by serial dilution in assay buffer (0.05 M Tris-HCL, sodium azide, 1 g/l; pH 7.5). Standards (100 µl) were run in duplicate in concentrations of 10 000, 5000, 2500, 1000, 500, 250, 100, 50, and 25 pg/ml. The standard curve included 300 µl prostaglandin-free plasma, which was obtained from a heifer treated at a 12-h interval with two i.m. injections of a prostaglandin synthetase inhibitor (flunixin megalumine, Flumeglumine; 1 g per injection; Phoenix Pharmaceuticals Inc., St. Joseph, MO). Blood was collected 4 h after the second injection. Assay buffer (100 µl) was added to duplicates of the experimental plasma samples (300 µl), as was done for the standard curve. Rabbit anti-PGFM (J57) antiserum (100 µl; dilution 1:4000) was added, and the mixture was incubated for 30 min at room temperature. Tracer (tritiated PGFM; catalog number TRK517, batch 99; Amersham GE Healthcare, Chalfont St. Giles, UK) was diluted in assay buffer (40 µl in 10 ml) to contain approximately 20 000 cpm/100 µl. The tracer (100 µl) was added, and the tubes were vortexed and incubated at room temperature for 1 h and then at 4°C overnight. Cold polyethylene glycol (40% in distilled water; 750 µl) was added to each tube. The tubes were vortexed and centrifuged for 30 min at 1700 x g in a refrigerated centrifuge and decanted. The precipitate was resuspended in 750 µl assay buffer, and the vortexing, centrifuging, and decanting were repeated. The precipitate was resuspended in 1 ml assay buffer and vortexed vigorously for 3 min. The suspension was poured into polyethylene scintillation vials. Scintillation fluid (4 ml) was added to each vial and mixed before counting in a beta counter. Serial volumes of pools of diestrus and estrus bovine plasma (500 µl to 100 µl) were processed as for the experimental samples and resulted in displacement curves that were similar to the standard curve. Serial concentrations of PGFM (5000 pg to 78 pg) prepared in prostaglandin-free bovine plasma also gave a dose-dependent response similar to the standard curve. The intraassay and interassay CV values and the sensitivity for PGFM were 7.1%, 6.3%, and 33.5 pg/ml, respectively.
Data were examined for normality with the Kolmogorov-Smirnov test. Data that were not normally distributed were transformed to natural logarithms. Individual end points were analyzed for time effects (day or hour), and comparisons involving groups were analyzed for main effects (group, time) and the interaction. The mixed procedure of SAS (version 8.2; SAS Institute Inc.) was used with a repeated statement to account for autocorrelation between sequential measurements. Paired and unpaired Student t-tests were used to locate differences between times within an end point and among groups within a time, respectively, when an effect of time or an interaction was obtained. A probability of P 
0.05 indicated that a difference was significant, and a probability between P > 0.05 and P 0.1 indicated that significance was approached. Data are presented as the mean ± SEM.
Experiment 1. Spontaneous Luteolysis
The day effect was significant (P < 0.005) for both progesterone concentration and percentage of blood flow in the CL when the normalization was to either the day of ovulation or to the first decrease in progesterone to <1 ng/ml (Fig. 1). The first decrease for the latter normalization occurred between Days –2 and –1 for both progesterone (P < 0.004) and percentage of blood flow (P < 0.01).
|
In the validation study, the percentage of maximum value showed an interaction (P < 0.003) between day and the three end points (progesterone concentration, estimated percentage of blood flow, and percentage of colored pixels; Fig. 2). The interaction was primarily the result of lower (P < 0.05) percentage of maximum for progesterone on Days –1 and 0 than for the two types of blood flow measurements. There were no significant differences between estimated percentage of blood flow and percentage of colored pixels.
|
Experiment 2. Systemically Induced Luteolysis
The interaction of group (treated and control) and hour was significant for both progesterone (P < 0.0001) and percentage of CL blood flow (P < 0.001; Fig. 3). Within the control group the differences for hour were not significant for either end point, but within the treatment group they were different (P < 0.0001) for each end point. Progesterone decreased (P < 0.01) within 1 h after treatment with the PGF analogue, and percentage of blood flow increased (P < 0.005) within 0.5 h. An increase in progesterone between 2 and 3 h posttreatment did not approach significance.
|
Experiment 3. Locally Induced Luteolysis
Progesterone decreased to <1 ng/ml in all four heifers in the systemic group but in only one and two heifers treated with 1 mg and 2 mg, respectively, of PGF by the intrauterine route. The intrauterine-treated heifers were reassigned into a group with (n = 3) and without (n = 5) CL regression to <1 ng/ml (responsive and nonresponsive groups, respectively). In the nonresponsive group, progesterone reduction was nil or partial (0 to 37%) between 0 h and 48 h. In this regard, the length of the interovulatory interval was shortened (P < 0.0002) in the systemic group (14 ± 0.6 days) and the intrauterine-responsive group (15 ± 0.9 days), but not in the nonresponsive group (22 ± 1.3 days; controls, 20 ± 0.8 days).
The group-by-hour interaction was significant (P < 0.0001) for percentage change in progesterone, percentage of CL with blood flow, and CL area (Fig. 4). The negative percentage change in concentrations of progesterone at 24 h and 48 h posttreatment was greater (P < 0.05) in the systemic and intrauterine-responsive groups than in the other two groups. Percentage of CL with blood flow signals increased (P < 0.02) between 0 h and 2 h in all PGF-treated groups but not in the controls. For percentage of CL with blood flow and for CL area, the values at 24 h and 48 h posttreatment were less (P < 0.05) in the systemic and intrauterine-responsive groups than in the controls.
|
Spectral data were missing for one heifer in the intrauterine-responsive group, leaving only two heifers in the group. This group was included in the statistical evaluation of spectral data for the four groups but was not considered further in the assessments of individual groups. The effect of hour (P < 0.008) and the group-by-hour interaction (P < 0.05) were significant for the vascular indices (PI and RI), but only the main effects of group (P < 0.007) and hour (P < 0.02) were significant for the relative velocity end points (PSV and TAMV; Fig. 5). Within groups, excluding the intrauterine-responsive group, a posttreatment decrease (P < 0.04) in PI and RI occurred between 0 h and 2 h for the nonresponsive group and approached significance (P < 0.1) for the systemic group. The increase between 0 h and 2 h for PSV and TAMV was significant (P < 0.02) for the systemic and intrauterine-nonresponsive groups. None of the end points changed significantly between 0 h and 2 h in the control group. At 24 h, RI was higher (P < 0.05) and PSV and TAMV were lower (P < 0.05) in the systemic group than in the intrauterine-nonresponsive and control groups.
|
Progesterone concentration (P < 0.0001), percentage of CL area with blood flow signals (P < 0.0002), and PGFM concentration (P < 0.01) normalized to progesterone <1 ng/ml and evaluated at 12-h intervals all showed significant differences among hours (Fig. 6). Progesterone concentrations first decreased (P < 0.0007) between Hours –48 and –24 and continued to decrease to Hour 0. Percentage of CL with blood flow signals first decreased (P < 0.01) between Hours –24 and –12. Mean PGFM concentrations increased between Hours –60 and –24, reached maximum at Hour –12, and then decreased.
|
The total number of sets (12-h windows) of samples at hourly intervals obtained from the six heifers during the preluteolytic, luteolytic, and postluteolytic periods, respectively, was 3, 9, and 1, and the number of statistically identified PGFM pulses during the three periods was 2, 9, and 0, respectively. At least one pulse was obtained in each of the six heifers during the luteolytic period, and the first detected pulse for each heifer was used in the analysis. The remaining three pulses during the luteolytic period and the four pulses for the other periods were handled individually. During the luteolytic period, the percentage of CL area with blood flow normalized to the PGFM peak was different (P < 0.04) among hours. Blood flow increased (P < 0.02) during the 4 h ending at the peak, remained elevated for 2 h, and decreased (P < 0.03) to baseline between 2 h and 3 h after the peak (Fig. 7). Progesterone did not differ significantly during the 7 h associated with the PGFM pulses (not shown). Individual examples of the close relationship of PGFM to blood flow during a 12-h window are shown (Fig. 8), including an example in which an apparent nadir rather than a complete pulse was detected.
|
|
The estimation of percentage of CL area with blood flow signals during real-time scanning of the entire CL was a useful and convenient end point. The information was obtained in about 1 min of intrarectal manipulations of the transducer compared with about 5 min for the spectral-based end points (PI, RI, PSV, and TAMV). Owing to the shorter time, estimation of percentage of area with blood flow signals was presumably less stressful to the animal. The blood flow estimates were validated in the first five heifers by the close agreement with the counting of pixels in a still image. The comparisons of results between color flow and spectral end points in experiment 3 provided further confidence in the use of the percentage estimate of blood flow. The percentage technique has been validated for the equine CL as it was in experiment 1 for bovine CL and also by close agreement between two operators working independently [7, 8].
It was reported that progesterone reduction during spontaneous luteolysis in cattle was associated initially with an increase in luteal blood flow, based on sampling and examining every 12 h [6]. A subsequent study in mares did not find a similar phenomenon [8], and therefore experiment 1 was done to confirm the cattle results. The present study used heifers and the reported study used cows, but other protocol aspects were similar. Neither 24-h intervals between examinations at a blood velocity setting of 10 cm/sec (experiment 1) nor a 12-h interval at a setting of 6 cm/sec (experiment 4) confirmed the reported results. As expected, normalization to the time progesterone decreased to <1 ng/ml provided a more concise mean pattern of luteolysis than for the ovulation reference point; a significant decrease occurred in 1 day versus 2 days for the two approaches, respectively. That is, mean concentration decreased more gradually and more variably when ovulation and, presumably, the previous estrus, as used in the reported study, were the reference points. However, neither normalization to ovulation nor to progesterone <1 ng/ml in experiment 1 confirmed that an initial increase in blood flow was associated with luteolysis in cattle.
Because of the conflicting results between the reported and present data for spontaneous luteolysis, experiment 2 was done to confirm the report [5] that systemic administration of cloprostenol, a PGF analog, caused an increase in CL blood flow in association with the initiation of luteolysis in cattle; a subsequent study did not find a similar posttreatment event in mares [7]. The reported results in cattle were well confirmed. The two species differ in that uterine-induced luteolysis involves direct or local delivery of the luteolysin from the uterus to the ovary in cattle and apparent systemic delivery in horses [29]. Therefore, it seemed likely that cattle may be more influenced by presumptive secondary effects of a systemic injection of PGF, which could explain the species difference of an increase in luteal blood flow after a systemic injection of PGF in cattle but not in horses. In experiment 3, PGF was given locally by intrauterine administration of a low dose [10] in an attempt to preclude the hypothetical secondary effect of systemic treatment. However, the intrauterine route resulted in an increase in CL blood flow at the first posttreatment examination (2 h) that was not significantly different from the response to the systemic route. Furthermore, the intrauterine route was effective in inducing increased CL blood flow, regardless of whether the treatment induced luteolysis. That is, blood flow increased even in the five of eight intrauterine-treated heifers with no or only a partial decrease in progesterone. It appears that a PGF-induced increase in luteal blood flow did not assure the initiation of a cascade of events leading to a progesterone decrease to <1 ng/ml (luteolysis). That is, blood flow regulation in the CL seemed more sensitive than regression of the CL to the stimulation of exogenous PGF.
The finding of an initial stimulatory effect of intrauterine administration of PGF on estimated percentage of CL area with blood flow also was supported by the spectral-Doppler end points. Significant changes in values for the spectral end points did not occur in the controls. The vascular indices of PI and RI were altered as effectively by the intrauterine route as by the systemic route. The decrease in the indices indicated more vascular perfusion or decreased resistance to blood flow in the tissues downstream from the point of assessment [24, 25]. Between 2 h and 24 h, the vascular indices increased in the systemic group and apparently in the intrauterine group that responded with luteolysis. This result is attributable to luteal regression, as indicated by the decrease in CL area. The relative blood velocities (PSV and TAMV) increased between 0 h and 2 h in heifers treated by either the systemic or intrauterine routes, a result that is compatible with the increase in percentage blood flow in the CL. The continuation of the elevation in relative velocities in the intrauterine groups at 24 h, despite the associated increase in vascular resistance, was not expected and will require confirmation and specific study of the underlying mechanism.
The apparent enigma of increased CL blood flow from borderline luteolytic doses of PGF into the uterus and failure to find an increase during spontaneous luteolysis was studied by hourly blood sampling during 12-h windows each day, encompassing the expected luteolytic period. The hypothesis, which was supported, was that elevated blood flow during spontaneous luteolysis increases and decreases in close association with individual PGFM pulses. For the six identified and analyzed pulses during the luteolytic period, blood flow was elevated only for an average of about 4 h during PGFM pulses that occurred about every 12 h. Therefore, elevated blood flow involved only about one third of a 12-h interval and would not likely be detected statistically with sampling every 12 or 24 h and with the small number of heifers (10) used in the present and reported [6] studies.
Luteal blood flow increased during the ascending arm of the PGFM pulse, remained elevated for 2 h after the peak of the pulse, and then decreased. Extensive studies [6, 30–33] have indicated that an acute increase in luteal blood flow occurs together with increases in expression of endothelial nitric oxide synthase (eNOS) and secretion of endothelin-1 (ET-1) and angiotensin II (AngII) in the CL during the early stage of luteolysis in cattle. The eNOS synthesizes nitric oxide (NO), a vasodilator; ET-1 and AngII are vasoconstrictors. The acute increase in blood flow apparently is induced by NO, and the increased blood flow is a proposed trigger for the luteolytic cascade. The present studies concerned only the temporal relationships among PGFM, CL blood flow, and luteolysis. Cause and effect relationships were not studied, but the temporal results are compatible with a proposed role for NO in increasing luteal blood flow and a resultant triggering of luteolysis. However, the role of vasodilators in initiating the luteolytic process needs to be further studied with consideration of the present finding that increases in CL blood flow during the luteolytic period are temporally associated with individual PGFM pulses, including a decrease in flow after the PGFM peak. Inspection of blood flow results in individual heifers indicated that the increase and decrease in flow occurred in temporal association with each successive pulse; however, data were restricted by the 12-h windows and were too limited for studying critically the apparent repeatability for successive pulses. A decrease in blood flow after a PGFM peak and an apparent increase again during the ascending arm of a subsequent pulse was noted within two of three sets of hourly examinations with adequate information. The finding that some animals responded to intrauterine PGF treatment with an increase in luteal blood flow but not a decrease in progesterone could be related to a requirement of sequential PGF pulses and the associated blood flow responses before progesterone decreases. The pulse flow relationship during the preluteolytic period was not characterized. Another aspect that will require specific study with larger number of animals and pulses is the immediate temporal relationship between a pulse and hypothetical progesterone fluctuations.
In conclusion, an initial increase in CL blood flow occurred during luteolysis induced by systemic or intrauterine administration of PGF. However, a previous report [6] that increased CL blood flow is an initial component of spontaneous luteolysis was not confirmed when blood sampling and Doppler examinations were done every 12 h or 24 h. The novel finding in the present study was an increase in CL blood flow area during the ascending portion of a PGFM pulse, followed by a return to basal blood flow beginning 2 h after the peak of the PGFM pulse.
ACKNOWLEDGMENTS
The authors thank Dee Cooper for technical assistance, Pfizer Animal Health for a gift of Lutalyse, and W. W. Thatcher, University of Florida, for a gift of PGFM antiserum and advice on the PGFM assay.
FOOTNOTES
1Supported by the Eutheria Foundation projects P1-LS-06, P4-LS-06, P5-LS-06, and B1-RR-06. L.A.S. is supported by a CAPES Scholarship (Brazil) to the University of Florida. 
Correspondence: 2O.J. Ginther, Animal Health and Biomedical Sciences, 1656 Linden Dr., University of Wisconsin, Madison, WI 53706. FAX: 608 262 7420; e-mail: ginther{at}svm.vetmed.wisc.edu
Received: 25 September 2006.
First decision: 2 November 2006.
Accepted: 28 November 2006.
REFERENCES
injection in the cow. Biol Reprod 2002; 66:651–658 in cows: progesterone, estrogen, LH, estrus and ovulation. J Anim Sci 1974; 38:347–353 during the normal oestrous cycle and early pregnancy in heifers. Acta Endocrinol 1976; 82:134–149 as reflected by 15-keto-13, 14-dihydroprostaglandin F2 in the peripheral circulation during normal luteolysis in heifers. Prostaglandins 1976; 11:871–876[CrossRef][Medline] release episodes and oxytocin receptor devleopment during luteolysis in the cow. Anim Reprod Sci 2006; 93:328–336[CrossRef][Medline] release from uterus and corpus luteum during spontaneous luteolysis in the cow. Reproduction 2004; 128:189–195 during luteolysis in ruminants. Biol Reprod 1991; 45:655–663[Abstract] of cows in response to recombinant interferon-
. J Dairy Sci 1995; 78:1921–1931[Abstract], milk composition, and metabolic status of periparturient Holstein cows. J Dairy Sci 2004; 87:921–932 in inducing luteolysis and oestrus in the cow. Soc Reprod Fertil 2002; 124:311–315, endothelin-1, and angiotensin II during spontaneous luteolysis in the cow. Biol Reprod 2004; 71:1706–1711 with endothelin-1 and tumor necrosis factor- on the release of progesterone and oxytocin in ovine corpora lutea in vivo: a possible implication for a luteolytic cascade. Reproduction 2004; 127:117–124This article has been cited by other articles:
|
|
 |
|
  O.J. Ginther, R.R. Araujo, M.P. Palhao, B.L. Rodrigues, and M.A. Beg Necessity of Sequential Pulses of Prostaglandin F2alpha for Complete Physiologic Luteolysis in Cattle Biol Reprod, April 1, 2009; 80(4): 641 - 648. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Shirasuna, S. Watanabe, T. Asahi, M. P B Wijayagunawardane, K. Sasahara, C. Jiang, M. Matsui, M. Sasaki, T. Shimizu, J. S Davis, et al. Prostaglandin F2{alpha} increases endothelial nitric oxide synthase in the periphery of the bovine corpus luteum: the possible regulation of blood flow at an early stage of luteolysis Reproduction, April 1, 2008; 135(4): 527 - 539. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
