The effects of 3 gonadorelin products on luteinizing hormone release, ovulation, and follicular wave emergence in cattle

Abstract

The objective was to determine luteinizing hormone (LH) secretion and follicular dynamics in cattle following administration of 3 gonadorelin formulations that are commercially available in Canada. In experiment 1, nonlactating Holstein cows (n = 4 per group) were randomly assigned to receive 100 μg gonadorelin diacetate tetrahydrate, intramuscularly (C; Cystorelin, or FE; Fertagyl). Blood samples (for LH analysis) were collected 0, 1, 2, and 4 hours after treatment. In experiment 2, nonlactating Holstein cows (n = 10 per group) were randomly allocated to receive 100 μg gonadorelin, intramuscularly as follows: 2 mL of C; 1 mL of FE; or 2 mL of Factrel (FA, gonadorelin hydrochloride). Gonadorelin treatment was done on days 6 or 7 after ovulation and blood samples for LH analysis were collected at 0, 1, 2, 4, and 6 hours after treatment. Ovaries were examined by ultrasonography, twice daily, to detect ovulation. A replicate was conducted using only C (n = 10) or FE (n = 10); blood samples were collected at 0, 1, 2, 3, and 4 hours. In experiment 3, beef heifers (n = 10 per group) were randomly assigned to receive 1 of 3 GnRH gonadorelin treatments (as in the first phase of experiment 2) on days 6 or 7 after ovulation and blood samples were collected at 0, 0.5, 1, 1.5, 2, and 4 hours. In experiments 2 and 3, both mean and mean peak plasma LH concentrations were higher (P < 0.05) in cattle treated with C. The proportion of dominant follicles that ovulated was higher (P < 0.02) in Holstein cows treated with C than in those treated with FE or FA (18/19, 11/19, and 4/7, respectively), but there was no significant difference among the products in beef heifers (6/10, 6/10, and 4/10, respectively). No significant differences were found in the interval from treatment to the emergence of the next follicular wave. In summary, C induced a greater LH release and this resulted in a higher ovulatory rate in Holstein cows but not in beef heifers.

Introduction

Gonadotropin releasing hormone (GnRH) secretion induces acute release of luteinizing hormone (LH) and follicle stimulating hormone (FSH). The pulsatile pattern of LH secretion from the pituitary is determined by GnRH neurons in the hypothalamus and the preoptic area of the forebrain (1,2). During estrus, high estradiol concentrations stimulate GnRH release; this induces LH synthesis, secretion, and, ultimately, an LH surge that results in ovulation (3). Estradiol (1) and progesterone influence neurons involved in GnRH and LH release (reviewed in 2). In addition, LH receptor mRNA is expressed on granulosa cells when the dominant follicle is greater than 9 mm in diameter (4). These control mechanisms account for the variation in ovulatory response to a GnRH treatment given at different stages of the estrous cycle (5).

The design of GnRH agonists for clinical use has been based on alterations in the chemical structure of the native GnRH molecule to increase stability and capacity to bind plasma proteins and GnRH receptors (6). Some of the commercially available GnRH analogues and agonists are as follows: buserelin with a D-serine at position 6 and ethylamide at position 10; fertirelin acetate with ethylamide at position 10; and native GnRH in different forms (gonadorelin diacetate tetrahydrate and gonadorelin hydrochloride). All of these compounds have been used to manipulate ovarian function in cattle (7). Studies of the effects of exogenous GnRH on LH release and follicular and luteal dynamics have been reported (8,9,10,11,12,13,14). The objective of the present study was to determine LH secretion and follicular dynamics following administration of 3 gonadorelin formulations that are commercially available in Canada.

Material and methods

All cows were nonpregnant and all received GnRH by IM injection in the gluteal region. In experiment 1, serum, and in experiments 2 and 3, plasma were collected for endocrine analysis.

In experiment 1 (preliminary experiment), nonlactating Holstein cows (n = 8) at unknown stages of the estrous cycle were housed at the Western College of Veterinary Medicine, University of Saskatchewan, with ad libitum access to grass-alfalfa hay and water. Cows were randomly assigned to 2 groups to receive IM injections of 100 μg of gonadorelin diacetate tetrahydrate, either 2 mL of C (Cystorelin; Merial Canada, Victoriaville, Quebec) or 1 mL of FE (Fertagyl; Intervet Canada, Whitby, Ontario). Blood was collected for LH analysis by jugular venipuncture at 0, 1, 2, and 4 h after GnRH treatment.

In experiment 2, nonlactating Holstein cows (n = 50), weighing between 386 and 489 kg, were housed at the Western College of Veterinary Medicine, University of Saskatchewan, with ad libitum access to grass-alfalfa hay, water, and cobalt-iodized salt. Cows were given 500 μg cloprostenol (Estrumate; Schering-Plough Animal Health, Pointe Claire, Quebec), IM, to induce luteolysis. A transrectal ultrasound examination (Aloka SSD-500 with 7.5 MHz transrectal transducer; ISM, Edmonton, Alberta) was done once daily to detect ovulation (ovulation = day 0). On days 6 or 7, cows were randomly allocated to 1 of 3 treatment groups (n = 10 cows per treatment) to receive 100 μg gonadorelin diacetate tetrahydrate in 2 mL of C, 1 mL of FE, or 100 μg of gonadorelin hydrochloride in 2 mL (FA, Factrel; Wyeth-Ayerst Canada, Montreal, Quebec), IM. All GnRH treatments were given at time 0 and blood was collected by jugular venipuncture at 0, 1, 2, 4, and 6 h for estradiol, progesterone, LH, and FSH analyses. Ovaries were examined by transrectal ultrasonography, once daily, from days 0 to 6 and, then, twice daily until ovulation was detected (or to day 12 if ovulation was not detected). A replicate of this experiment was conducted using only C (n = 10) or FE (n = 10). Blood samples were collected at 0, 1, 2, 3, and 4 h after the GnRH challenge for estradiol, progesterone, LH, and FSH analyses. Ultrasonography was performed as described for the first phase.

In experiment 3, cross-bred beef heifers (n = 30; Angus, Hereford, Simmental, and Charolais, 350 to 400 kg) were housed outdoors in feedlot pens at the Goodale Research Farm, University of Saskatchewan, and fed barley silage and water. Heifers were given 500 μg cloprostenol, IM, to induce luteolysis, as in experiment 2, and were randomly allocated (n = 10 per group) to receive 1 of the 3 gonadorelin treatments (as used in the first phase of experiment 2) on days 6 or 7 after ovulation. Blood samples were collected at 0, 0.5, 1, 1.5, 2, and 4 h for estradiol, progesterone, and LH analyses. Ovarian examinations were conducted as described in experiment 2.

Blood processing

Blood samples for estradiol, progesterone, LH, and FSH concentrations were collected into heparinized tubes, kept at approximately 4°C, and centrifuged within 4 h after collection; plasma was stored frozen until assayed. Plasma LH concentrations were determined by double-antibody radioimmunoassay (14) and expressed in terms of National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)-bLH4. The sensitivity of the assay, assessed as the lowest concentration of LH capable of displacing labeled LH from the antibody, was 0.06 ng/mL. Intra- and interassay coefficients of variation were 16.9% and 4.8%, respectively, for reference sera with mean LH concentrations of 0.28 and 0.67 ng/mL. Peak LH concentrations were determined as the highest mean concentration of LH reached after treatment with GnRH, regardless of when peaks occurred in individual cows. Plasma concentrations of FSH were determined by using a liquid-phase antibody radioimmunoassay (15,16). The first antibody used was NIDDK anti-oFSH-l, and FSH concentrations were expressed in terms of U.S. Department of Agriculture (USDA)-bFSH-1. The sensitivity of the assay was 0.125 ng/mL. Intra- and interassay coefficients of variation were 10.7% and 6.5%, respectively, for reference sera with mean FSH concentrations of 0.97 and 1.98 ng/mL.

The progesterone and estradiol assays were conducted as described previously (15). Progesterone was extracted with 3 mL of hexane added to 200-μL aliquots of plasma. The sensitivity of the progesterone radioimmunoassay was 0.1 ng/mL. Intra- and interassay coefficients of variation were 12.0% and 13.8% (means 1.36 and 0.44 ng/mL, respectively). Estradiol was extracted with ethyl ether. The sensitivity of the assay was 0.5 pg/mL and the intra- and interassay coefficients of variation were 11.7% and 10.2% (means 11.6 and 23.2 pg/mL, respectively).

Data analysis

Statistical analyses of hormonal data were conducted with software (Statistical Analysis System, version 8.2; 2001, SAS Institute, Cary, North Carolina), whereas follicular data were analyzed using software (Statistix for Windows, Analytical Software, version 2.0; 1996, Tallahassee, Florida). Hormonal data were analyzed using the MIXED procedure for repeated measures, including the autoregressive covariance structure (main effects of treatment and time, and treatment-by-time interaction). Main effects and mean LH peak concentrations were compared by the Tukey test. One-way analysis of variance was used to compare intervals from treatment to ovulation and to wave emergence. The Mann-Whitney test was used to compare the interval from treatment to ovulation and, because data were not normally distributed in the replicate of experiment 2, the interval from treatment to follicular wave emergence. Variances for follicular data were compared by Bartlett's test. Ovulation rates were compared by chi-square or Fisher's exact test (depending on the number of animals per group). Only animals in which follicular wave emergence occurred between the day of GnRH treatment and 6 d later were included in the analysis of follicular data. Therefore, animals in which follicular wave emergence occurred before the day of treatment were removed from follicular data analysis. In experiment 3, Student's t-test was used to compare the day of emergence of the new follicular wave after gonadorelin treatment in heifers that ovulated versus those that did not.

The University of Saskatchewan Animal Care Committee has approved protocol number 19970080 for these experiments.

Results

Experiment 1

There was an effect of time (P < 0.0001), but no effect of treatment (P = 0.13) or treatment-by-time interaction (P = 0.16) on serum LH concentrations (Figure 1). Serum LH concentrations increased significantly (P < 0.0001) by 1 or 2 h after treatment and decreased to basal levels by 4 h. Mean concentrations of LH tended to be higher (P < 0.13) in the C group (3.0, s_¯x = 0.4 ng/mL) than in the FE group (2.0, s_¯x = 0.4 ng/mL).

Figure 1. Mean (and s_¯x) serum lutenizing hormone (LH) concentrations in nonlactating Holstein cows (treated at random stages of the estrous cycle) given 100 μg of gonadorelin in 2 mL (Cystorelin; C) or 1 mL (Fertagyl; FE) at hour 0 (experiment 1). There was an effect of time (P < 0.0001) on serum LH concentrations. Serum LH concentrations increased significantly (P < 0.01) by 1 and 2 h after treatment and decreased to basal levels by 4 h. Mean concentrations of LH tended to be higher (P < 0.2) in the C group (3.0, s_¯x = 0.4 ng/mL) than in the FE group (2.0, s_¯x = 0.4 ng/mL).

Experiment 2

Phase 1 — There was no difference in mean (and s_¯x) concentrations of estradiol (7.2, s_¯x = 1.0; 6.1, s_¯x = 2.0; 6.5, s_¯x = 0.9 pg/mL) or progesterone (1.3, s_¯x = 0.5; 1.8, s_¯x = 0.9; 2.0, s_¯x = 0.9 ng/mL) at the time of treatment for the C, FE, and FA groups, respectively (P = 0.12). There was an effect of time (P < 0.0001), a tendency for an effect of treatment (P = 0.08), and treatment-by-time interaction (P < 0.06) on mean plasma LH concentrations (Figure 2). In particular, peak LH concentrations were higher (P < 0.01) in the C group (6.6, s_¯x = 1.4 ng/mL) than in the FE group (4.7, s_¯x = 0.8 ng/mL) or the FA group (3.8, s_¯x = 0.5 ng/mL), which were not different. It is noteworthy that in all groups, 50% of cows had LH peaks at 1 h and 50% had LH peaks at 2 h after treatment.

Figure 2. Mean (and s_¯x) plasma LH concentrations (ng/mL) after treatment with 100 μg gonadorelin (Cystorelin [C], Fertagyl [FE], or Factrel [FA]; hour 0) in nonlactating Holstein cows on days 6 or 7 after ovulation (experiment 2; phase 1). There was a tendency (P = 0.11) for an effect of treatment, and there was an effect of time (P < 0.0001) and a treatment-by-time interaction (P < 0.05) on mean plasma LH concentrations. Peak concentrations of LH were higher (P < 0.01) in the C group than in the FE or FA groups (which were not different).

There was an effect of time (P < 0.001), but there was no effect of treatment (P = 0.2) or treatment-by-time interaction (P = 0.4) on mean plasma FSH concentrations. Mean FSH concentrations peaked at 1 h in the FE group (1.1, s_¯x = 0.1 ng/mL), and at 2 h in both the C (1.0, s_¯x = 0.1 ng/mL) and FA (1.1, s_¯x = 0.1 ng/mL) groups.

There was 1 cow in the C group, 1 cow in the FE group, and 3 cows in the FA group in which the diameter of the dominant follicle was < 9 mm at the time of treatment. These animals were excluded from analysis of follicular data. Another cow treated with FA was excluded from the analysis of day of follicular wave emergence, because the dominant follicle reached 27 mm in diameter by 6 d after treatment, without emergence of a new follicular wave. The mean (and s_¯x) diameter of the dominant follicle at the time of GnRH treatment was not different (P = 0.3) among groups and all exceeded 9 mm. Ovulatory rate tended to be higher (P = 0.1) in the C group (100%; n = 9) than in the FE (56%; n = 9) or FA (57%; n = 7) groups, which were not different. Mean (and s_¯x) day of emergence of the next follicle wave did not differ (P = 0.35) among groups (C, 1.7, s_¯x = 0.2 d; FE, 1.9, s_¯x = 0.3 d; FA, 2.3, s_¯x = 0.8 d).

Phase 2 — There was no significant difference in mean (and s_¯x) concentrations of estradiol (1.0, s_¯x = 0.9; 0.5, s_¯x = 0.2 pg/mL; P = 0.6) or progesterone (3.0, s_¯x = 0.2; 3.0, s_¯x = 0.2 ng/mL; P = 0.9) at the time of treatment in the C or FE groups, respectively. There was an effect of treatment (P < 0.007), time (P < 0.0001), and a treatment-by-time interaction (P < 0.008) on mean plasma LH concentrations (C, 4.1, s_¯x = 0.4 ng/mL; FE, 2.3, s_¯x = 0.4 ng/mL; Figure 3). Although the maximal difference (P < 0.0001) in mean LH concentrations between the C and FE groups occurred 2 h after treatment (Figure 3), the mean LH peak occurred 2 h after treatment in the C group (7.9, s_¯x = 1.2 ng/mL) and at 1 h after in the FE group (4.2, s_¯x = 0.5 ng/mL); 40% of animals in the C group and 60% of animals in the FE group had an LH peak at 1 h and the remainder had an LH peak at 2 h. There was an effect of time (P < 0.0001), but no effect of treatment (P = 0.08) or treatment-by-time interaction (P = 0.4) on plasma FSH concentrations.

Figure 3. Mean (and s_¯x) plasma luteinizing hormone (LH) concentrations (ng/mL) after treatment with 100 μg gonadorelin (Cystorelin [C] or Fertagyl [FE]; hour 0) in non lactating Holstein cows on days 6 or 7 after ovulation (experiment 2; phase 2). There was an effect of treatment (P < 0.01), time (P < 0.0001) and treatment-by-time interaction (P < 0.006) on mean plasma LH concentrations (C, 4.1, s_¯x = 0.4 ng/mL; FE, 2.3, s_¯x = 0.4 ng/mL). The maximal difference in LH concentrations between C and FE groups occurred 2 h after treatment (P < 0.0001).

There were no animals eliminated from this data set. The mean (and s_¯x) diameter of the dominant follicle at the time of GnRH treatment was 14.2, s_¯x = 0.5 mm and 14.1, s_¯x = 0.5 mm for C and FE groups, respectively (P = 0.9). The ovulatory response was 90% (9/10) in the C group and 60% (6/10) in the FE group (P = 0.2). The interval from GnRH treatment to ovulation was 37.3, s_¯x = 4.0 h in the C group, and 36.0, s_¯x = 0.0 h in the FE group (P = 0.4). The interval from treatment to emergence of a new follicular wave was 1.8, s_¯x = 0.1 and 2.2, s_¯x = 0.2 d for the C and FE groups, respectively (P = 0.08).

Phases 1 and 2 (combined) — As there was no effect of phase (P = 0.4) on any of the follicular end points (diameter of the dominant follicle at the time of treatment, interval from treatment to ovulation, interval from treatment to wave emergence), the follicular data were combined and the results are summarized in Table 1.

Table 1.

Experiment 3

There was no significant difference in mean (and s_¯x) concentrations of estradiol (2.7, s_¯x = 1.0; 2.4, s_¯x = 08; 2.5, s_¯x = 0.9 pg/mL; P = 0.4) or progesterone (3.6, s_¯x = 2.6; 4.0, s_¯x = 1.0; 2.7, s_¯x = 0.9 ng/mL; P = 0.9) at the time of treatment in the C, FE, and FA groups, respectively. There were effects of treatment (P < 0.002) and time (P < 0.0001), and a tendency for a treatment-by-time interaction (P = 0.1) on plasma LH concentrations (which peaked 1 h after treatment in all groups). Mean LH concentrations were higher in heifers treated with C than in those treated with FE or FA (P < 0.05); the maximal difference occurred 1.5 h after treatment (P < 0.05; Figure 4). In addition, peak concentrations of LH were higher (P < 0.05) in the C group than in the FE and FA groups, which were not different from each other. Mean LH concentrations reached basal concentrations by 4 h in all groups.

Figure 4. Mean (and s_¯x) plasma luteinizing hormone (LH) concentrations (ng/mL) after treatment with 100 μg of gonadorelin (C, FE, or FA) on days 6 or 7 after ovulation in beef heifers. There was an effect of treatment (P = 0.04), time (P < 0.0001), and a treatment-by-time interaction (P = 0.056) on plasma LH concentrations. Mean LH concentrations were higher at 1.5 h in heifers treated with C than in those treated with FE or FA (P < 0.001). In addition, peak concentrations of LH were higher (P < 0.05) in the C group than in the FE group with those in the FA group intermediate.

Dominant follicles of all heifers were ≥ 10 mm in diameter at the time of treatment. There was no effect of treatment on any of the follicular end points (Table 2). Although mean day of follicular wave emergence after GnRH treatment was not different among groups (P = 0.9), there was an effect of ovulation (P < 0.002) on the day of emergence of the new follicular wave. The mean (and s_¯x) interval from treatment to wave emergence was shorter (P < 0.05) and less variable (P < 0.0001) in heifers that ovulated (1.5, s_¯x = 0.1 d; n = 16) than in heifers that did not ovulate (2.7, s_¯x = 1.3 d; n = 14).

Table 2.

Discussion

In experiments 2 and 3, higher LH peak concentrations were induced by C than by the other 2 products. In addition, overall mean LH concentrations in the C groups were higher in experiment 2 (phase 2) and in experiment 3, and tended to be higher in experiments 1 and 2 (phase 1). Although there was some variability between experiments, LH responses were similar for FE- and FA-treated groups. Ovulatory responses to GnRH treatment tended to follow LH peak concentrations; the administration of 100 μg of C resulted in a significantly higher ovulatory response in Holstein cows (experiment 2), but differences between products were not significant in beef heifers.

Differences among products in the timing of the GnRH-induced LH surge may be attributed to differences in absorption. In Holstein cows, C induced an LH peak 2 h after treatment, whereas FE induced an LH peak at 1 h. In the beef heifers, maximal concentrations of LH were reached by 30 min in all groups. Although C and FE are considered the same gonadorelin compound (diacetate tetrahydrate), differences may be attributed to differences in volume (2 mL versus 1 mL for C and FE, respectively). Another possibility is that the vehicle, additives, or both may affect the rate of absorption. Although the vehicle used is unknown, it has been reported that vehicles such as polyethylene glycol protect polypeptides upon microencapsulation, increasing stability of the molecule and reducing the disintegration rate (17). The use of propylene glycol or a similar substance could alter the rate at which GnRH is absorbed (18), resulting in different intervals to the LH surge, its magnitude, and, possibly, ovulatory response.

In sheep, the initiation of the LH surge requires an abrupt increase in GnRH, but more GnRH support is required for maintenance of the LH surge (19). In the present experiments, the LH profiles for C were higher and appeared to be slightly more prolonged than those for either FE or FA. We speculate that treatment with C may have caused a higher and more prolonged elevation of plasma GnRH concentrations, resulting in higher LH release and, in turn, a higher ovulatory response. At the recommended dose (100 μg), C consistently induced sufficient LH release to cause ovulation of the dominant follicle present at the time of treatment in nonlactating Holstein cows. Although the LH profile in Holstein cows was somewhat different than that in beef heifers, baseline levels were achieved within 4 h in both Holstein cows and beef heifers, regardless of the product used.

Ovulatory response to GnRH in the cow has been shown to vary depending on whether treatment is administered during the luteal or the follicular stage of the cycle (5). A maximum response has been observed when progesterone is very low, that is, 12 h before the endogenous LH surge. The other factor affecting the ovulatory response to a GnRH treatment is the stage of development of the dominant follicle at the time of treatment. Dominant follicles acquire LH receptors at the time of selection, approximately 3 d after the emergence of a follicular wave, when the dominant follicle achieves a diameter of 9 mm (4). On days 6 or 7 after ovulation, the dominant follicle was expected to have reached a diameter of > 10 mm. Although all follicles were expected to have LH receptors at the time of treatment, the ovulatory response was variable. A greater ovulation rate has been obtained after a GnRH treatment at unknown stages of the estrous cycle in cows than in heifers (5). Difficulty in inducing ovulation of a mid-cycle dominant follicle with GnRH in heifers has been reported previously (5,13,21,22). In one study, treatment of beef heifers with 100 μg C at different stages of development of the dominant follicle of the first follicular wave resulted in ovulation in 56% of the animals (13). Similar results were achieved in the present study, even though heifers were treated at a stage of dominant follicle development expected to be responsive. Although C seemed to be most effective in inducing LH release in the beef heifers (experiment 3), there was no significant difference in the ovulatory response among the 3 different GnRH preparations, all of which appeared to induce a lower ovulatory response in beef heifers than in Holstein cows.

Although ovulation of the dominant follicle occurred in only 50% to 60% of the animals treated with FE or FA, synchronous follicular wave emergence occurred (overall mean of 2.1 d), suggesting that all products may be used effectively to synchronize follicular wave emergence in estrus synchronization programs. However, it is noteworthy that 2- or 3-wave cows would be expected to be near the time of emergence of the 2nd follicular wave on days 8 or 9 (1 to 2 d after treatment) (23). Therefore, spontaneous emergence of a follicular wave probably occurred in those animals that did not ovulate in response to treatment.

In beef heifers, there was no significant effect of treatment on the interval from treatment to wave emergence. However, the interval was shorter and less variable in heifers that ovulated than in those that did not. Elimination of the dominant follicle has been shown to result in the emergence of a new follicular wave (24,25,26). Therefore, GnRH is likely only to be effective in synchronizing emergence of a new follicular wave when treatment results in ovulation (13,22). However, it is also important to recognize that after the first injection of GnRH in Ovsynch-type programs for fixed-time artificial insemination, spontaneous follicular wave emergence may occur within 1 or 2 d of the expected time of wave emergence in many animals (13). Therefore, follicle wave emergence is likely to occur in some cases, even though ovulation in response to treatment did not occur. However, data reported herein, and previously (27,13), showing the LH response to GnRH treatment and the associated ovulatory response, suggest that the use of reduced dosages of GnRH (28) in the synchronization of follicular wave emergence in fixed-time breeding programs may not be advisable.