If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Effect of elevating luteinizing hormone action using low doses of human chorionic gonadotropin on double ovulation, follicle dynamics, and circulating follicle-stimulating hormone in lactating dairy cows
Department of Animal Sciences and Industry, Kansas State University, Manhattan 66506Department of Animal and Dairy Sciences, University of Wisconsin, Madison 53706Eutheria Foundation, Cross Plains, WI 53528
Department of Animal and Dairy Sciences, University of Wisconsin, Madison 53706Endocrinology and Reproductive Physiology Program, University of Wisconsin, Madison 53706
Eutheria Foundation, Cross Plains, WI 53528Department of Pathobiological Science, School of Veterinary Medicine, University of Wisconsin, Madison 53706
Double ovulation and twin pregnancy are undesirable traits in dairy cattle. Based on previous physiological observations, we tested the hypothesis that increased LH action [low-dose human chorionic gonadotropin (hCG)] before the expected time of diameter deviation would change circulating FSH concentrations, maximum size of the second largest (F2) and third largest (F3) follicles, and frequency of multiple ovulations in lactating dairy cows with minimal progesterone (P4) concentrations. In replicate 1, multiparous, nonbred lactating Holstein dairy cows (n = 18) had ovulation synchronized. On d 5 after ovulation, all cows had their corpus luteum regressed and were submitted to follicle (≥3 mm) aspiration 24 h later to induce emergence of a new follicular wave. Cows were then randomized to NoP4 (untreated) and NoP4+hCG (100 IU of hCG every 24 h for 4 d after follicle aspiration). Ultrasound evaluations and blood sample collections were performed every 12 h for 7 d after follicle aspiration. All cows were then treated with 200 μg of GnRH to induce ovulation. In replicate 2, cows (n = 16) were resubmitted to similar procedures (i.e., corpus luteum regression, follicle aspiration, randomization, ultrasound evaluations every 12 h, GnRH 7 d after aspiration). However, cows in replicate 2 received an intravaginal P4 device that had been previously used (∼18 d). Only cows with single (n = 15) and double (n = 16) ovulations were used in the analysis. No significant differences were detected for frequency of double ovulation, follicle sizes, and FSH concentrations across replicates (NoP4 vs. LowP4 and NoP4+hCG vs. LowP4+hCG), so data were combined. Double ovulation was 40% for control cows with no hCG (CONT) and 62.5% with hCG (hCG). Double ovulation increased as the maximum size of F2 increased: <9.5 mm and 9.5–11.5 mm (7.7%) and ≥11.5 mm (94.1%). The hCG group had more cows with F2 > 11.5 (69%) than with 9.5 ≥ F2 ≤ 11.5 (25%) and F2 < 9.5 (6%). In agreement, F2 and F3 maximum size were larger in the hCG group, but FSH concentrations were lower after F1 > 8.5 mm compared with CONT. In contrast, FSH concentrations were greater before deviation (F1 closest value to 8.5 mm) in cows with double ovulations than in those with single ovulations, regardless of hCG treatment. In addition, time from aspiration to deviation was shorter in cows with double rather than single ovulation and in cows treated with hCG as a result of faster F1, F2, and F3 growth rates before diameter deviation. In conclusion, greater FSH and follicle growth before deviation seems to be a primary driver of greater frequency of double ovulation in lactating cows with low circulating P4. Moreover, the increase in follicle growth before deviation and in the maximum size of F2 during hCG treatment suggests that increased LH may also have a role in stimulating double ovulation.
). Cattle are generally a monotocous species, however, under specific physiologic conditions, such as during lactation, they have an increased frequency of double ovulations (DOV; up to 40%), which has been associated with increased twin pregnancies (
). Although some economic gain has been reported for twin pregnancies, the losses generally outweigh the gains, with an average $135 loss for twin pregnancies compared with singletons in dairy cattle (
). Other explanations include a greater frequency of dystocia, freemartins, retained placenta, and calf mortality, as well as an increased number of days open and decreased fertility (
A retrospective study investigating the association of parity, breed, calving month and year, and previous parity milk yield and calving interval with twin births in US dairy cows.
) reported an average twinning rate of 6.4% for lactating cows (parity ≥2). Double ovulation is responsible for most twins, as the frequency of monozygotic twins was estimated to be only 5.5% of twin births and 0.33% of all births in lactating cows (
). Thus, it is clear that multiple ovulation and the resulting twin pregnancies are considered problematic, costly, and undesirable traits in dairy cattle (
Decreased circulating progesterone (P4) concentration is a critical factor leading to selection of multiple follicles rather than a single dominant follicle (DF). For example, decreased P4 has been associated with an approximately 300% increase in frequency of codominant deviations and twins in various studies of lactating dairy cows (
). Supporting these observations, manipulative studies that altered P4 in lactating dairy cows during the preovulatory follicular wave increased DOV from 12% in cows with high P4 (∼9 ng/mL) to 49% in cows with lower P4 (∼3 ng/mL;
). In nonlactating Holstein heifers, elevated P4 concentrations (such as those observed during the second follicular wave) have been associated with a smaller size of the largest (F1, 7.5 mm) and second largest (F2, <7 mm) follicles at diameter deviation compared with follicle sizes (F1, 8.5 mm; F2, >7 mm) in heifers with minimal P4 concentrations (
Hormonal mechanisms regulating follicular wave dynamics II: Progesterone decreases diameter at follicle selection regardless of whether circulating FSH or LH are decreased or elevated.
). In addition, an increased frequency of codominant follicles during the first follicular wave of lactating dairy cows was associated with increased daily LH concentrations (
Hormonal mechanisms regulating follicular wave dynamics II: Progesterone decreases diameter at follicle selection regardless of whether circulating FSH or LH are decreased or elevated.
) that increasing LH action [by treatments with low doses of human chorionic gonadotropin (hCG; 96 IU of hCG every 24 h) or porcine LH (1.25 mg every 12 h)] in the presence of high P4 concentrations would result in larger F1 and F2 diameters at the time of deviation in nonlactating Holstein heifers. In contrast to our hypothesis, treatment with LH did not increase the F2 growth rate, which was observed to decrease when F1 reached ∼7.5 mm. Further, we observed a tendency (P = 0.07) for a decreased F2 growth rate in heifers treated with hCG or porcine LH, contrary to the theory that inadequate LH is the reason for the smaller size of F1 and F2 at deviation in a high P4 environment. Further, this result suggests that LH action may produce the opposite effect than hypothesized (i.e., reduced F2 growth rate). Nonetheless, studies that directly manipulate LH action to evaluate the effects on follicle growth and DOV in lactating dairy cows have not yet been performed.
Thus, the current study tested the hypothesis that increased LH action before the expected time of diameter deviation alters the circulating FSH concentrations, maximum size of the F2, and frequency of codominant follicles and DOV in lactating dairy cows with minimal P4 concentrations. This 2-sided hypothesis was based on 2 conflicting observations. First, greater LH was observed in lactating cows with multiple rather than single ovulations (SOV) during the first follicular wave, potentially supporting the theory that increasing LH increases F2 size and the likelihood of codominant follicles and DOV (
). Alternatively, the treatment of heifers with low doses of hCG decreased the size of F2 at diameter deviation, consistent with the opposite theory that LH action decreases the likelihood of codominance and DOV (
Hormonal mechanisms regulating follicular wave dynamics II: Progesterone decreases diameter at follicle selection regardless of whether circulating FSH or LH are decreased or elevated.
Selection of fewer dominant follicles in Trio carriers given GnRH antagonist and luteinizing hormone action replaced by nonpulsatile human chorionic gonadotropin.
Hormonal mechanisms regulating follicular wave dynamics II: Progesterone decreases diameter at follicle selection regardless of whether circulating FSH or LH are decreased or elevated.
). This approach was based on the high affinity and specificity of hCG for the LH receptor and the relatively long half-life of hCG, allowing continuously elevated LH action (
The study was conducted at the Dairy Cattle Center of the University of Wisconsin-Madison in the northern temperate zone during the winter–spring season in 2019. Multiparous, nonbred lactating Holstein dairy cows (n = 18) at 371 ± 25 DIM with no apparent abnormalities in their reproductive tracts, as evaluated by ultrasound, were used in this study. Cows were kept in tie-stall barns with ad libitum access to water, fed TMR, and milked twice daily, which allowed collection of data and treatment of cows at specific times. All cows received the same TMR to meet or exceed the nutrient requirements for a lactating Holstein cow producing 50 kg of milk per day with 3.5% fat and 3.1% true protein when DMI was 24 kg/d (
All management was performed according to the United States Department of Agriculture Guide for Care and Use of Agricultural Animals in Research. Additionally, the experimental protocol used in this study (A005727) was approved by the Animal Care and Use Committee of the University of Wisconsin-Madison.
Experimental Design and Treatments
Figure 1 summarizes the experimental procedures and groups. All cows (n = 18) were synchronized to be on d 5.5 after ovulation and subsequently received 2 doses of 500 μg i.m. of cloprostenol sodium (Estroplan, Parnell Pharmaceuticals) 24 h apart to regress the corpus luteum. On d 6.5 after ovulation, follicles (≥3 mm) were aspirated transvaginally to induce the emergence of a new follicular wave (
). Epidural anesthesia was performed before follicle aspiration to reduce animal pain and distress. Re-aspiration of follicles was performed 12 h later if any refilled antral follicles were observed. The study was conducted in 2 replicates. In replicate 1, cows did not receive any P4 treatment and were randomly assigned to control (CONT; untreated) or low-dose hCG [100 IU of hCG (50 IU/mL; Chorulon, Merck Sharp & Dohme Corp.) every 24 h for 4 d, starting 12 h after aspiration]. Seven days after aspiration, all cows received an ovulatory GnRH dose [200 mg GnRH i.m. (gonadorelin acetate; Gonabreed, Parnell Pharmaceuticals)]. Ovulation was confirmed by the disappearance of follicles between 2 consecutive 12-h scans. On d 5.5 after ovulation, 16 cows had the previous procedures repeated (cloprostenol administration, follicle aspirations, and randomizations) to conduct replicate 2. In replicate 2, all procedures were similar to replicate 1 except that all cows received (on the day of follicle aspiration) an intravaginal P4 device [controlled internal drug release device (CIDR) delivering 1.38 g of P4, Eazi-Breed CIDR, Zoetis] previously used for ∼18 d. This treatment was intended to maintain low circulating P4 concentrations. The CIDR was removed 7 d after aspiration, when cows received the ovulatory GnRH dose. Thus, replicate 1 had 2 experimental groups, NoP4 (n = 8) and NoP4+hCG (n = 8), and replicate 2 had 2 similar experimental groups, LowP4 (n = 7) and LowP4+hCG (n = 8). Before data collection, we decided that NoP4 and LowP4 would be combined (control, n = 15) and NoP4+hCG and LowP4+hCG would be combined (hCG, n = 16) if no major differences were found.
Figure 1Schematic of experimental design. Presynchronized cows on d 5.5 after ovulation (OV) received 2 doses of 500 μg i.m. of cloprostenol sodium (PGF) 24 h apart to regress the corpus luteum. On d 6.5 after OV, follicle (≥4 mm) content aspiration (Foll Asp) was performed to induce the emergence of a new follicular wave. Re-aspiration of follicles was performed 12 h later. The study was conducted in 2 replicates. In replicate 1, cows lacked progesterone (P4) treatment, whereas in replicate 2 all cows received a used (∼18 d) controlled internal drug release device (CIDR) on the day of follicle aspiration. In both replicates, cows were randomly assigned to the groups CONT (control, untreated) and hCG [100 IU of human chorionic gonadotropin (50 IU/mL) every 24 h from d 6.5 to d 10.5 after OV]. On d 13, 200 μg GnRH was administered i.m. (gonadorelin acetate) and cows were checked for OV. Four groups were generated from replicates 1 and 2: NoP4, NoP4+hCG, LowP4, and LowP4+hCG. In further analysis, 2 groups were compared: CONT (combined NoP4 + LowP4) vs. hCG (combined NoP4+hCG + LowP4+hCG). Inverted open triangles indicate the hCG treatments on d 0.5, 1.5, 2.5, 3.5, and 4.5 after follicle aspiration. US = ultrasound.
A sample size of 17 cows per group was calculated using Power Procedure from SAS software with α = 0.05 and power = 0.8. This was based on expected frequencies of 45% versus 5% for DOV in control versus hCG.
Ultrasound, Blood Sample Collections, and Hormonal Assays
For 7 d after the day of follicle aspiration, ultrasound evaluations and blood sample collections were performed every 12 h. Ultrasound evaluations were performed transrectally with a Mindray M5-Vet machine (Mindray North America) equipped with a multifrequency linear transducer set at 5.0 MHz. At each evaluation, individual videos were obtained by slowly scanning each ovary from the lateral to the medial direction or vice versa. Videos were evaluated frame by frame to aid in the identification and measurement of individual follicles (≥4 mm) based on 2 perpendicular measurements (height and width) at the apparent maximal area of the antrum (
). The disappearance of the preovulatory DF confirmed ovulation.
Blood samples were collected in EDTA evacuated tubes by coccygeal venipuncture at 12-h intervals before each ultrasound examination. Samples were centrifuged at 2,000 × g for 20 min, and plasma was subsequently transferred into 5-mL Eppendorf tubes, frozen, and stored at −20°C until FSH was assayed. Circulating FSH concentrations were determined for all 12-h samples in duplicates using a validated RIA technique (
). The intraassay coefficient of variation was 3.1%, and the mean assay sensitivity was 0.08 ng/mL.
Circulating P4 concentrations were determined in singlets for the samples collected on d 2 and 6 after ablation. These 2 d represent the emergence and common growth phase of the follicles (d 2) and the dominance phase of the induced follicular wave (d 6) based on the data in Figure 2. Circulating P4 concentrations were determined using a solid-phase RIA kit containing antibody-coated tubes and 125I-labeled P4 (ImmuChem Coated Tube Progesterone 125I RIA Kit, MP Biomedicals) as previously described (
). The P4 intraassay coefficient of variation and sensitivity were 8.02% and 0.021 ng/mL, respectively.
Figure 2Mean ± SEM for the diameter (mm) of the largest or future dominant follicle (F1), second largest follicle (F2; on occasion, ovulatory), and third largest follicle (F3) in addition to the FSH concentrations (ng/mL) relative to the day of follicle content aspiration. Comparisons were performed between the low and no progesterone (LowP4 and NoP4, respectively) groups (panels A, B, C, and D) and between the LowP4+human chorionic gonadotropin (hCG) and NoP4+hCG groups (panels E, F, G, and H). The number of cows per group (n) and the probabilities for the main effect of group (G), main effect of day (D), and interaction of group by day (GD) are indicated in the figure.
Follicles were assigned to F1, F2, and F3 terminology depending on their maximum size ranking (first, second, and third largest follicle) and regardless of the dominance status. For example, heifer A had a single DF termed F1, whereas the largest subordinate follicle was termed F2. Heifer B had 2 DF, and thus the largest DF was termed F1, whereas the second largest DF was termed F2. Measurement of follicles and terminology assignment were conducted by a technician blinded to the treatments.
An initial comparison (from the day of ablation until 6.5 d later) was made between the replicates to determine if the 2 replicates could be combined for subsequent comparisons. The comparisons were made between the groups that did not receive hCG in replicates 1 and 2 (i.e., NoP4 vs. LowP4 groups) and separately between the groups that received hCG (i.e., NoP4+hCG vs. LowP4+hCG group; Figure 2). Although 1 interaction was significant for F2, no differences were found between groups within any of the d 0 to 6.5 measurements; thus, data were combined from the 2 replicates for all subsequent analyses.
For the combined analyses, data were normalized to time when F1 was closest to 8.5 mm (d 0) and compared before and after diameter deviation separately. To determine the effect of hCG treatment, 3 comparisons were performed: CONT (combined NoP4 and LowP4) versus hCG (combined NoP4+hCG and LowP4+hCG) using all cows (Figure 3), using only cows that had a SOV (CONT vs. hCG), or only cows that had DOV (CONT vs. hCG). To determine factors associated with DOV, comparisons were made between the SOV and DOV groups using all cows (Figure 4), regardless of whether or not they received hCG; using only CONT cows (SOV vs. DOV; data not shown); or using only cows that received hCG (SOV+hCG vs. DOV+hCG; data not shown).
Figure 3Mean ± SEM for the diameter (mm) of the largest or future dominant follicle (F1), second largest follicle (F2; on occasion, ovulatory), and third largest follicle (F3) in addition to the FSH concentrations (ng/mL) relative to the day that F1 was closest to 8.5 mm for the control (CONT) versus the human chorionic gonadotropin (hCG) group. The number of cows per group (n) and the probabilities for the significant main effect of group (G), main effect of day (D), and interaction of group by day (GD) are indicated in the figure. Between the 2 groups: *significant difference (P < 0.05), #tendency for a difference (0.05 < P ≤ 1).
Figure 4Mean ± SEM for the diameter (mm) of the largest or future dominant follicle (F1), second largest follicle (F2; on occasion, ovulatory), and third largest follicle (F3) in addition to the FSH concentrations (ng/mL) relative to the day that F1 was closest to 8.5 mm for cows with single (SOV) versus double ovulation (DOV) regardless of the hCG treatment (with and without). The number of cows per group (n) and the probabilities for the significant main effect of group (G), main effect of day (D), and interaction of group by day (GD) are indicated in the figure. *Significant difference (P < 0.05) between the 2 groups.
The days from follicle aspiration to when the F1 was closest to 8.5 mm in diameter and the growth rates of F1, F2, and F3 were compared between SOV and DOV and between CONT and hCG groups (Table 1). Each follicle had its own diameters on d −2 subtracted from its diameter on d 0 (F1 closest to 8.5 mm) to obtain the follicle growth from d −2 to 0. Likewise, follicle diameter on d 0 was subtracted from the follicle diameter on d 2 to obtain the growth rate from d 0 to 2. The data were also classified and compared according to the maximum F2 size: cows with F2 < 9.5 mm versus cows with 9.5 mm ≥ F2 < 11.5 mm versus cows with F2 ≥ 11.5 mm (Table 2).
Table 1Comparison of cows with single (SOV) versus double ovulation (DOV) in all cows and for effect of human chorionic gonadotropin (hCG) treatment (CONT vs. hCG) in all cows, or only SOV or DOV cows for days from follicle aspiration to F1 of ~8.5 mm and growth rate
For all analyses, each cow within a replicate was considered one experimental unit.
Comparisons of continuous data with no repeated measurements (days from ablation to F1 closest to 8.5 mm, and growth rates of F1, F2, and F3) were performed by t-test. Frequency analyses were performed with SAS (version 9.4, SAS Institute Inc.). Specifically, the frequencies of single and multiple DF were compared using Fisher's exact test obtained by the PROC FREQ with the chisq exact option. Distribution frequencies within the same category (e.g., cows that did or did not receive hCG and had F2 < 9.5 mm, 9.5 mm ≥ F2 < 11.5 mm, or F2 ≥ 11.5 mm) were compared using chi-squared goodness of fit with PROC FREQ to determine if an observed frequency distribution differed from a theoretical frequency distribution of 50%.
Statistical analysis for continuous data with repeated measurements was performed with linear mixed models [ANOVA using PROC MIXED (SAS Version 9.4; SAS Institute)]. The cow was treated as a random effect to account for autocorrelation between sequential measurements. Continuous data that were not normally distributed based on the Shapiro-Wilk test (P < 0.05) were transformed to natural logarithms or ranks. The rank transformation was used when logarithms did not resolve a lack of normality. For each continuous variable (F1, F2, and F3 diameters and FSH), a linear model was run (before and after deviation, separately) with the dependent variable (F1, F2, and F3 diameters, FSH) and the fixed effects of group, day, and interaction of group by day. For P4 concentrations, a single model was run using d 2 and 6 after follicle aspiration and with the dependent variable (P4) and fixed effects of group, day, and interaction of group by day. When the main effect of group, day, or the interaction group by day was significant, Tukey's honestly significant difference test was used to compare groups over the days or within a day. A probability of P ≤ 0.05 was considered a significant difference, whereas P > 0.05 to P ≤ 0.1 was considered a tendency to significance. Data are presented as raw mean ± standard error of the mean unless indicated otherwise.
RESULTS
Two cows that failed to ovulate in replicate 1 (one from NoP4 and one from NoP4+hCG) were not used in replicate 2 and were removed from the analysis in addition to 1 cow with triple ovulation (replicate 2, LowP4). Thus, only cows with SOV (n = 15) and DOV (n = 16) were used in the analyses. A comparison between the LowP4 and NoP4 groups resulted in a similar (P = 1.0) frequency of DOV [3/7 vs. 3/8 (40%)].
As expected, the insertion of a used CIDR generated minimal but still slightly greater (P < 0.0001) P4 concentrations in replicate 2 (0.62 ± 0.06 ng/mL) than in replicate 1 (0.09 ± 0.02 ng/mL). Moreover, within replicate 1, P4 was low on d 2 (0.07 ± 0.02 ng/mL) and d 6 (0.12 ± 0.04 ng/mL) after follicle aspiration, and the effects of group (P = 0.7), day (P = 0.5), and interaction of group by day (P = 0.6) were not significant. Similarly, in replicate 2, P4 concentrations were low on d 2 (0.65 ± 0.06 ng/mL) and d 6 (0.59 ± 0.10 ng/mL) after follicle aspiration, and the effects of group (P = 0.4), day (P = 0.2), and interaction of group by day (P = 0.5) were not significant. Moreover, for F1, F2, and F3 diameters and FSH concentrations (Figure 2A, B, C, D), the day but not the group effect was significant, and the interaction of group by day was not significant during the 7 d after follicle aspiration. Similarly, a comparison between the LowP4+hCG and NoP4+hCG groups resulted in a similar (P = 0.6) frequency of DOV (6/8 vs. 4/8). For the F1 and F3 diameters and FSH concentrations (Figure 2E, G, H), the day but not the group effect or the interaction of group by day was significant during the 7 d after follicle aspiration. For the F2, the day effect and the interaction of group by day were significant, although no significant differences were detected within a day. Thus, the treatment groups from the 2 replicates were combined for subsequent analyses (CONT vs. hCG and SOV vs. DOV).
The comparison between the CONT and hCG groups revealed an approximately 50% increase (P = 0.19) in the DOV group [6/15 (40%) vs. 10/16 (62.5%)]. For analyses, data were normalized to F1 closest value to 8.5 mm (Figure 3). For F1, the day and the group effect but not the interaction of group by day were significant from d −2 to 0 (F1 closest value to 8.5 mm), whereas from d 0.5 to 3, only the main effect of day was significant. For F2 and F3, the group and the day effect were significant or tended to significance from d −2 to 3. Additionally, the interaction of group by day was significant from d 0.5 to 3 for both F2 and F3. For F2, the interaction was primarily from the larger F2 size in the hCG group than in the CONT group during d 1 to 3. For the F3, the interaction was significant both before and after deviation and similarly was primarily related to the larger F3 in the hCG group than the CONT group from d 0 to 3. For FSH concentrations, the day effect and the interaction of group by day were significant from d −2 to 0. This interaction was likely related to the numerically greater FSH in the hCG group than in the CONT group from d −2 to −0.5 (all differences P > 0.1) and to the subsequent lower FSH concentrations in the hCG group than in the CONT group on d 0. In contrast, the day effect and the group effect were significant from d 0.5 to 3, but the interaction was not significant. The group effect was from lower averaged FSH concentrations from d 0.5 to 3 in the hCG group (0.19 ± 0.01 ng/mL) compared with the CONT group (0.22 ± 0.01 ng/mL).
A comparison between SOV versus SOV+hCG normalized to the F1 closest value to 8.5 mm (not shown) resulted in a significant (P < 0.0001) day effect for F1 from day −2 to 3. For the F2, the day effect was significant (P < 0.0001) from d −2 to 3, whereas the group effect and the interaction of group by day approached significance (P < 0.07) from d 0.5 to 3. The interaction was primarily due to the larger (P < 0.0001) F2 size in the SOV+hCG group on d 2 and 3. Additionally, for the F3 diameter, the day effect was significant (P < 0.0001) from d −2 to 0, whereas the group effect was significant (P = 0.005) from d 0.5 to 3 due to the larger F3 size averaged over the days in the SOV+hCG group (7.8 ± 0.1 mm) compared with the SOV group (6.6 ± 0.1 mm). For FSH, only the day effect was significant (P < 0.0001), whereas the interaction of group by day approached significance (P = 0.07) from d −2 to 0. The interaction was primarily from lower FSH concentrations on d 0 in the SOV+hCG group.
A comparison between DOV and DOV+hCG normalized to F1 closest value to 8.5 mm (not shown) resulted in a tendency (P = 0.06) for a group effect from d −2 to 0 and in a significant (P < 0.0001) day effect from d −2 to 3 for F1. The group effect was from the smaller F1 diameter averaged over the days in the DOV+hCG group relative to the DOV group. Moreover, the group by day interaction was significant (P = 0.003) from d 0.5 to 3, primarily from the greater F1 size on d 0.5 for the DOV group than the DOV+hCG group. For the F2, the group effect tended to significance (P = 0.09) from d −2 to 0 and the day effect was significant (P < 0.0001) from d −2 to 3. The group effect was from the smaller F2 diameter averaged over the days in the DOV+hCG group compared with the DOV group. For the F3, the day effect was significant (P < 0.0001) from d −2 to 0, whereas the group effect tended to significance (P < 0.1) from d 0.5 to 3 because of the greater F3 size averaged over days in the DOV+hCG group than the DOV group. For FSH, the day effect was significant (P < 0.001) from d −2 to 3, whereas the group effect was significant (P = 0.02) from d 0.5 to 3 because of lower FSH concentrations averaged over the days in the DOV+hCG than DOV group.
The comparisons of SOV vs. DOV groups without hCG and SOV+hCG vs. DOV+hCG were also performed but are not shown because the results were analogous to those obtained for the comparison of SOV versus DOV regardless of hCG treatment (Figure 4). Moreover, combining all cows for SOV versus DOV comparison allowed for a more powerful analysis (15 vs. 16 cows instead of 9 vs. 6 or 6 vs. 10). For the F1, SOV versus DOV (Figure 4) resulted in a significant effect of group and interaction of group by day from d −2 to 0. The interaction was primarily from greater F1 size in the SOV than DOV groups during d −2 to −0.5. In contrast, an analysis of the F1 growth rate revealed that the F1 in the DOV group grew faster from d −2 to 0. In addition, time from aspiration to F1 of ∼8.5 mm was shorter in the DOV group than in the SOV group (Table 1). For the F2, the main effect of day was significant from d −2 to 3, whereas the group effect and the interaction of group by day were significant from d 0.5 to 3. The interaction was primarily from greater F2 size in the DOV group than in the SOV group on d 1 to 3, as expected. For the F3, the day effect and the interaction of group by day were significant from d −2 to 0, although no differences were detected between the groups for any specific day comparison. For FSH, the group effect was significant from d −2 to 0 with greater FSH concentrations at all times before d 0 in the DOV group than in the SOV group, and for FSH averaged over the days in the DOV group (0.38 ± 0.01 ng/mL) than in the SOV group (0.30 ± 0.01 ng/mL). Moreover, the day effect was significant from d −2 to 3 due to declining FSH concentrations over the days.
Table 1 (left columns) illustrates the time from aspiration to F1 reaching the closest diameter to 8.5 mm. Briefly, when all cows were considered (n = 31), the DOV group reached 8.5 mm in a significantly shorter time compared with the SOV group. Likewise, treatment with hCG decreased the time from follicle aspiration to 8.5 mm in analyses of all cows, SOV cows only, and DOV cows only. The shorter time from aspiration to deviation indicated a faster follicle growth rate (Table 1, right column), supported by a greater follicle growth rate (d −2 to d 0; F1 ∼8.5 mm) for DOV than SOV cows for F1 (+0.5 mm/d), F2 (+0.3 mm/d), and F3 (+0.5 mm/d). In addition, treatment with hCG increased the growth rate of F1, F2, and F3 compared with CONT in analyses of all cows and SOV cows only, but not DOV cows only.
Finally, a comparison of cows according to the maximum F2 size (Table 2) showed no statistical differences for the frequency of cows with F2 < 9.5 mm, 9.5 to 11.5 mm, or >11.5 mm within the CONT group. In contrast, the hCG group had differences (P < 0.05) between the categories with greater frequency of cows with F2 size >11.5 mm, intermediate frequency of cows with F2 = 9.5 to 11.5 mm, and a lower frequency of cows with F2 < 9.5 mm. In a comparison of CONT and hCG cows within the F2 categories, the hCG group had fewer cows with F2 < 9.5 mm than the CONT group, but no differences were detected in the other F2 categories. Although no differences were observed in the F1 maximum size or the day that the F1 maximum size was reached, the F2 maximum size and the day that F2 reached the maximum size increased along with the F2 size category. Moreover, ovulation of F2 (i.e., DOV) was similar in the F2 < 9.5 mm and F2 = 9.5 to 11.5 mm categories (7.7%), but was ∼12-fold greater in the F2 > 11.5 mm category (94.1%).
DISCUSSION
Given that the presence of multiple ovulations is the most direct physiologic basis for twinning in dairy cattle, understanding the causes of multiple ovulations or identifying strategies to prevent them could provide substantial economic value to dairy producers. Numerous physiologic factors have been associated with the risk of multiple ovulations and twinning in cattle, such as genetics, season, parity, and milk production, with most reviews of this topic emphasizing the potential endocrine basis for many of these risk factors (
). Decreased circulating concentrations of P4 and increased circulating concentrations of estradiol, LH, and FSH have all been observed in lactating cows with multiple compared with single DF, and these factors have been speculated to be key parts of the endocrine basis for the high DOV rate (
). Our current study focused on the role of elevating circulating LH-like activity in a continuous, low-dose manner using hCG as a potential means to decrease the likelihood of DOV in lactating dairy cows, given the smaller follicle size at deviation in heifers treated with hCG. This approach left open the possibility that increasing circulating LH activity could potentially increase DOV, given the previous associations of DOV with increased circulating LH. Three main hypotheses were considered: (1) increased DOV is related to increased circulating FSH concentrations, (2) hCG is related to changes in the size of the F2 help explain the effects of increased LH action on follicle and hormonal dynamics, and (3) increased LH action alters codominance and DOV in a manner that implicates LH as either an underlying cause of DOV in lactating dairy cows or a strategy to prevent it.
A key observation from this study was the clear elevation in circulating FSH concentrations during the 2 d before deviation in cows that had DOV compared with cows that had SOV (Figure 4), regardless of whether they were treated with hCG. This increase in FSH in cows with codominance is consistent with observations in previous studies done during the first follicular wave (
) and supports the theory that greater circulating FSH leading up to and encompassing the expected time of diameter deviation (F1 ∼8.5 mm) is an important factor in selection of multiple DF in lactating dairy cows (
). Secretion of FSH occurs by 2 pathways. The constitutive FSH secretion pathway is regulated by inhibitory ovarian factors such as inhibin and estradiol, whereas the pulsatile FSH secretion pathway is directly stimulated by pulses of GnRH (
Neuroendocrine control of follicle-stimulating hormone (FSH) secretion. I. Direct evidence for separate episodic and basal components of FSH secretion.
Gonadotropin-releasing hormone, estradiol, and inhibin regulation of follicle-stimulating hormone and luteinizing hormone surges: Implications for follicle emergence and selection in heifers.
). It seems likely that the elevation in FSH in DOV cows is at least partly due to greater constitutive FSH secretion, but greater pulsatile FSH secretion is also likely to be present owing to the greater GnRH/LH pulse frequency that would be expected in the low P4 environment used in this study. In a recent study, we observed the importance of FSH/LH pulses in selection of multiple DF in carriers of the Trio allele, a mutation that elevates SMAD6 and causes a high ovulation rate (
Selection of fewer dominant follicles in Trio carriers given GnRH antagonist and luteinizing hormone action replaced by nonpulsatile human chorionic gonadotropin.
). In that study, inhibition of LH/FSH pulses by using a GnRH antagonist combined with low-dose hCG treatments to ensure growth of the DF after deviation resulted in a decrease in the number of selected DF in carriers of the Trio allele from ∼3.5 in controls to ∼1 in treated heifers that had pulsatile LH/FSH ablated with acyline. It is possible that the elevation in FSH in DOV compared with SOV cows during the 2 d before and encompassing deviation may also have been related to elevated GnRH-induced FSH pulses; however, in previous experiments, treatment with acyline did not reduce the predeviation FSH surge (
Relationships between FSH patterns and follicular dynamics and the temporal associations among hormones in natural and GnRH-induced gonadotropin surges in heifers.
Selection of fewer dominant follicles in Trio carriers given GnRH antagonist and luteinizing hormone action replaced by nonpulsatile human chorionic gonadotropin.
). Regardless of the pathway involved in increased FSH secretion, constitutive or GnRH-induced, it seems likely that the elevation in FSH near deviation is driving the selection of 2 DF in cows that have DOV.
Treatment with hCG appeared to directly inhibit secretion of FSH, but only after selection of the DF (diameter deviation) began to occur. Figure 3 presents evidence for this statement, with reduced FSH on the day of deviation and a continuation of the reduced FSH (group effect of hCG) after deviation. Of importance, hCG treatment did not reduce FSH before deviation. This inhibitory effect of hCG on circulating FSH after deviation was also observed in the comparison of the DOV versus DOV+hCG cows. In SOV cows, an inhibitory effect of hCG was also observed at the time of deviation, but it did not reach significance (P = 0.198) after deviation. This inhibitory effect of hCG treatment on circulating FSH after deviation is consistent with our previous results (
). The observation that this inhibitory effect was only observed after selection of a DF (8.5 mm) is consistent with the observation that when the largest follicle (F1) reaches a diameter of ∼8.5 mm, its granulosa cells acquire LH receptors and shift its dependency from FSH to mainly LH (
). In cows selecting a single DF, nadir FSH concentrations are reached near deviation, and thus, the second largest follicle (F2) decreases its growth rate due to lack of FSH (
Hormonal mechanisms regulating follicular wave dynamics II: Progesterone decreases diameter at follicle selection regardless of whether circulating FSH or LH are decreased or elevated.
). Thus, we speculate that in both studies, the LH action of the low-dose hCG treatments combined with its long half-life resulted in greater production of inhibitors of FSH secretion, probably estradiol, although other follicular factors cannot be excluded at this time (
Alterations in intrafollicular regulatory factors and apoptosis during selection of follicles in the first follicular wave of the bovine estrous cycle.
). This effect of greater circulating LH activity reducing nadir FSH concentration is also consistent with the greater nadir FSH in heifers that undergo deviation in elevated P4 (low LH) versus low P4 (high LH) environments (
The second major hypothesis that was tested in this study was that increasing LH action, by treatment with low doses of hCG, would alter the size of F2 and potentially F3, indicating a role for LH in the selection of single and multiple DF. In our previous studies, we observed a decrease in F2 size when heifers in either low or high P4 categories (wave 1 or induced wave 2, respectively) were treated every 12 or 24 h with low doses of hCG from the day of emergence until F1 reached at least 10 mm (
Hormonal mechanisms regulating follicular wave dynamics II: Progesterone decreases diameter at follicle selection regardless of whether circulating FSH or LH are decreased or elevated.
). In contrast, the results of the present study using lactating dairy cows in a low P4 environment, treatment with hCG had the opposite effect, increasing size of the F2 and F3. This outcome can be clearly observed in Figure 3, with a larger size of F2 and F3 observed near or after the time of diameter deviation (F1, 8.5 mm). In addition, this effect of hCG to increase the size of F2 and F3 was clearly observed when only SOV cows were evaluated, although not in cows with DOV. Low-dose hCG treatment also increased the growth rate of F1, F2, and F3 prior to F1 reaching 8.5 and decreased the time from follicle aspiration to F1 reaching 8.5 mm in analyses of all cows or SOV or DOV cows only (Table 1). These results indicate that hCG may be having effects on the follicle before the expected time of acquisition of LH receptors in the granulosa cells of the DF (∼8.5 mm). One speculation is that hCG may act on the thecal cells to increase androgen production and that androgens stimulate FSH receptors in granulosa cells, as previously observed during in vitro experiments (
). The lack of an effect of hCG on F1 growth rate after selection of DF may be related to the high numbers of LH receptors on granulosa cells that could already be maximizing DF growth rate (
Hormonal mechanisms regulating follicular wave dynamics II: Progesterone decreases diameter at follicle selection regardless of whether circulating FSH or LH are decreased or elevated.
) but increased the size of F2 in the lactating cows used in the present study. The reduction in circulating FSH after deviation was consistent in both studies, but the opposite effect of hCG on size of the F2 is puzzling. Double ovulation is low in heifers and greatly reduced by a high P4 environment, both conditions that were present in the previous but not the present experiment. In the present experiment, LH pulses would be expected to be rapid due to low circulating P4 and the high GnRH/LH pulsatility that seems to be inherent to lactating dairy cows (
). Future research is clearly needed to determine whether the disparity in response to increased LH action in these 2 experiments is due to physiologic differences between lactating dairy cows and heifers, or alternatively, differences between high and low P4 environments.
Finally, we expected that increased LH action, provided by continuous hCG treatment, would alter the likelihood of codominance and DOV. One possibility that we considered was that hCG would stimulate production of FSH inhibitory factors from the granulosa cells of F1, leading to reduced circulating FSH, and greater inhibition of growth of F2. Our experimental design did not allow us to adequately test this hypothesis because, although DOV was increased more than 50% in the hCG-treated cows, we did not have sufficient statistical power to detect a difference. Thus, we can only indicate that DOV was not decreased by hCG treatment, as we might have anticipated from our previous results in heifers (
Hormonal mechanisms regulating follicular wave dynamics II: Progesterone decreases diameter at follicle selection regardless of whether circulating FSH or LH are decreased or elevated.
). Instead, our results are more consistent with the concept that increasing LH action may stimulate growth of F2 and increase selection of codominant follicles and DOV, although a definitive test of this theory will require a larger study. Thus, our results support a role for increased LH in the high frequency of DOV in high-producing lactating dairy cows but are not definitive.
One final speculation is possible on the basis for DOV in lactating dairy cows after detailed analysis of the results of this experiment. The elevation in FSH before deviation in cows with DOV compared with SOV made us speculate that perhaps follicles in DOV cows were growing more rapidly and reached 8.5 mm earlier than in SOV cows, while circulating FSH remained elevated. This speculation was consistent with the smaller F1 size at 2 d before deviation and the greater F1 growth rate. Indeed, time from aspiration to deviation was shorter for cows that had DOV (3.3 ± 0.01 d) than SOV (3.8 ± 0.2 d), but time from aspiration to emergence was similar between the 2 groups (1.0 ± 0.1 d). Interestingly, cows with DOV also had greater growth rates of the F2 and F3 than the CONT cows, suggesting that the higher FSH stimulated the growth rate of all follicles. Surprisingly, hCG treatment also stimulated growth of F1, F2, and F3 before deviation perhaps by stimulating FSH receptors in the granulosa cells through stimulation of androgens by thecal cells (
). Thus, we speculate that greater FSH action, due to greater FSH concentrations (DOV vs. SOV) or greater FSH responsiveness (CONT vs. hCG), and perhaps other factors in lactating dairy cows may be driving faster growth rate of follicles before deviation, leading to earlier attainment of key follicle differentiation stages such as follicle selection, before final suppression of FSH. For example, an increase in IGF-1, androgens, or other factors in DOV lactating dairy cows may facilitate increased follicle growth, perhaps through expression of FSH receptors (
) or other pathways. The shorter time from follicle aspiration to expected deviation (8.5 mm) and the greater growth rate during this time for F1, F2, and F3 in DOV versus SOV cows are novel observations and may be important to consider in future studies focused on understanding mechanisms determining selection of multiple DF.
Although the results of this study did not result in a practical approach to decreasing DOV and twinning in lactating dairy cows, it did provide physiologic information that may focus future studies in this important area. Future studies may focus on reducing DOV through increasing circulating P4 during preovulatory follicle development using protocols such as Double-Ovsynch (
Manual rupture versus transvaginal ultrasound-guided aspiration of allanto-amniotic fluid in multiple pregnancies: A clinical approach to embryo reduction in dairy cattle.
). In regard to follicle aspiration, we have recently demonstrated that follicle aspiration can be used to reduce number of DF and ensure ovulation of a single DF in each ovary to increase bilateral twinning rates in beef cattle that are carriers for the high fecundity Trio allele (
). Future hormonal methods, based on a thorough understanding of the endocrine basis for DOV, may be developed to practically optimize reproductive efficiency through manipulation of ovulation rate.
In summary, maintaining minimal P4 concentrations in lactating dairy cows resulted in a high frequency of DOV, regardless of whether cows were treated with hCG to increase LH action. Double ovulation in this experimental setting was accompanied by and potentially driven by greater circulating FSH before deviation, shorter time from aspiration to F1 reaching 8.5 mm, and greater F1, F2, and F3 growth rates before deviation in comparison with SOV. Additionally, cows receiving hCG also had greater F1, F2, and F3 growth rates before diameter deviation compared with cows that did not receive hCG. In contrast, treatment with hCG was associated with decreased FSH but only after deviation, perhaps due to secretion of increased FSH inhibitors from granulosa cells after acquisition of LH receptors.
In conclusion, greater FSH and follicle growth before deviation seems to be a primary driver of greater frequency of DOV in lactating cows with low circulating P4. However, the increase in follicle growth before deviation and in maximum size of F2 during hCG treatment suggests that increased LH may also have a role in stimulating DOV and decreasing FSH after F1 > 8.5 mm. Thus, a practical hormonal method for reducing DOV in lactating dairy cows was not produced from this research. However, a greater understanding of DOV was provided, particularly concerning the role of elevated FSH, LH, and more rapid follicle growth before deviation in DOV compared with SOV, which seemed to be stimulated by LH action. This may allow future development of physiologically rational methods for reducing DOV.
ACKNOWLEDGMENTS
Funding was provided by the Eutheria Foundation; Wisconsin Experiment Station as Hatch Project WIS02013 to MCW; USDA-National Institute of Food and Agriculture (NIFA) project 2018-67015-27612 to MCW; and USDA-NIFA project 2022-67015-36372 to MCW and VGL. The authors thank Meghan Connelly and the students of the 2019 DYSCI534 (Reproductive Management of Dairy Cattle) class (Department of Animal and Dairy Sciences, University of Wisconsin-Madison) for their technical assistance, including Cassidy Dabbs, Zachary Endres, Alexandra Hafey, Caleb Hamm, Emily Hutterer, Zachary Lensmire, Megan Mezera, Michael Moede, Natalie Schmidt, Katy Vacula, and Tyler Vande Wettering. The authors have not stated any conflicts of interest.
REFERENCES
Adams G.P.
Matteri R.L.
Kastelic J.P.
Ko J.C.
Ginther O.J.
Association between surges of follicle-stimulating hormone and the emergence of the follicular wave in heifers.
Manual rupture versus transvaginal ultrasound-guided aspiration of allanto-amniotic fluid in multiple pregnancies: A clinical approach to embryo reduction in dairy cattle.
Alterations in intrafollicular regulatory factors and apoptosis during selection of follicles in the first follicular wave of the bovine estrous cycle.
Selection of fewer dominant follicles in Trio carriers given GnRH antagonist and luteinizing hormone action replaced by nonpulsatile human chorionic gonadotropin.
Hormonal mechanisms regulating follicular wave dynamics II: Progesterone decreases diameter at follicle selection regardless of whether circulating FSH or LH are decreased or elevated.
Gonadotropin-releasing hormone, estradiol, and inhibin regulation of follicle-stimulating hormone and luteinizing hormone surges: Implications for follicle emergence and selection in heifers.
Relationships between FSH patterns and follicular dynamics and the temporal associations among hormones in natural and GnRH-induced gonadotropin surges in heifers.
Neuroendocrine control of follicle-stimulating hormone (FSH) secretion. I. Direct evidence for separate episodic and basal components of FSH secretion.
A retrospective study investigating the association of parity, breed, calving month and year, and previous parity milk yield and calving interval with twin births in US dairy cows.
00362-9&id=fx1.jpg){kind=link}
00362-9&id=gr1.jpg){kind=link}
00362-9&id=gr2.jpg){kind=link}
00362-9&id=gr3.jpg){kind=link}
00362-9&id=gr4.jpg){kind=link}