Theriogenology xxx (2014) 1–7
Contents lists available at ScienceDirect
Theriogenology journal homepage: www.theriojournal.com
Intraovarian effect of dominant follicle and corpus luteum on number of follicles during a follicular wave in heifers O.J. Ginther a, b, *, M.M. Hoffman a a b
Eutheria Foundation, Cross Plains, Wisconsin, USA Department of Pathobiological Sciences, School of Veterinary Medicine, University of Wisconsin–Madison, Madison, Wisconsin, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 22 November 2013 Received in revised form 13 March 2014 Accepted 18 March 2014
The intraovarian relationships among dominant follicle (DF), corpus luteum (CL), and number of follicles between Days 0 to 5 (Day 0 ¼ ovulation) in wave 1 (n ¼ 65 waves) and Days 9 to 13 in wave 2 (n ¼ 62) were analyzed in separate experiments in Bos taurus heifers. Ovaries were grouped into intraovarian patterns of DF–CL, DF alone, CL alone, and neither DF nor CL. In wave 1, the pattern frequencies of DF–CL or neither DF nor CL (34% each) were greater (P < 0.0004) than for DF alone or CL alone (16% each). The number of growing follicles 5.0 mm, was greater (P < 0.0001) in ovaries with the DF, even when the DF was removed from the tally (P < 0.03). In a factorial analysis of wave 1, there was a positive main effect of DF (3.9 0.2 vs. 2.2 0.2 follicles; P < 0.0001), but the main effect of CL and the interaction of DF and CL were not significant. In a factorial analysis of wave 2, there were more (P < 0.0001) follicles greater than 6 mm in ovaries with a DF when the DF was included and an approaching difference (P < 0.09) when the DF was excluded. The main effect of CL and the interaction of DF and CL were not significant. The hypothesis that both the DF and CL have a positive intraovarian effect on number of follicles in waves 1 and 2 was only partly supported; the DF, but not the CL, had an effect in the factorial analyses. Previous reports in cattle and sheep of a positive intraovarian effect of CL on number of follicles are questionable in that location of the DF was not considered. Ó 2014 Elsevier Inc. All rights reserved.
Keywords: Corpus luteum Dominant follicle Follicular waves Number of follicles Right and left ovaries
1. Introduction A wave of follicular activity develops about every 9 days in cattle of the Bos taurus breeds, resulting in two or three waves during the interovulatory interval (IOI) [1]. Wave 1 (first wave of the IOI) is anovulatory. Wave 2 in two-wave and three-wave IOIs is ovulatory and anovulatory, respectively. Wave 1 emerges with 4 or 5 mm follicles on the day of ovulation. When the largest follicle of the wave reaches about 8.5 mm on about 2.5 days after wave emergence, deviation occurs and is characterized by continued growth rate of the future dominant follicle and reduction in growth rate and regression of the remaining
* Corresponding author. Tel.: þ1 608 798 3777; fax: þ1 608 798 3722. E-mail address: [email protected] (O.J. Ginther). 0093-691X/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.theriogenology.2014.03.013
follicles (subordinates). Dominancy is established when the follicle reaches 10 mm, as indicated by capacity for ovulation [2]. The interval from ovulation to deviation seems similar between Bos taurus and Bos indicus breeds, but the diameter of the largest follicle at deviation is much smaller (about 6.0 mm [3]). The follicle and hormone characteristics of a wave have been studied most extensively for wave 1. The relationships between the ovary containing the dominant follicle (DF) versus the developing corpus luteum (CL) and the differential and combined intraovarian effects of the DF and CL on the number of follicles in the wave 1 have not been reported, despite the considerable knowledge on follicle and hormone dynamics [1]. Studies in ewes first demonstrated an intraovarian stimulatory effect of the CL on follicles. In unilaterally ovulating ewes, follicles are larger during the mid-estrous
2
O.J. Ginther, M.M. Hoffman / Theriogenology xxx (2014) 1–7
cycle in the ovary that contains the CL (ipsilateral relationship) than when the CL and follicles are in opposite ovaries (contralateral relationship) [4]. Four days after destruction of follicles in both ovaries, follicles are more developed in the ipsilateral ovary [4,5]. However, results of studies in cattle on a similar intraovarian relationship seem to be contradictory. More medium-sized follicles are present in the ovary ipsilateral to the CL than in the contralateral ovary at all stages of the estrous cycle [6]. Transrectal ultrasonic imaging is being used to study follicle diameters throughout the estrous cycle [7]. More follicles 2 or 3 mm are located ipsilateral to the CL than contralateral [8]. In contrast, no relationship was found between the location of the CL and the number of antral follicles on selected days of the estrous cycle [9]. Apparent intraovarian effects in cattle are further complicated by differences between the right and left ovaries. Greater follicular activity has been reported for the right ovary [10,11]. There are more follicles 4 to 6 mm and greater than13 mm in the right ovary, regardless of the location of the CL [8]. A study of spatial relationships between the DF and CL in cattle found no intraovarian relationships between the ovary containing the CL and the ovary containing the DF or between the location of the CL and the characteristics of the DF [12]. Another study found more follicular waves with the DF ipsilateral than contralateral to the CL, using all follicular waves of the estrous cycle [13]. The intraovarian effect of the CL on the number of growing follicles in cattle during a follicular wave needs clarification, due to contradictory reports. In addition, the intraovarian effect of the DF on number of follicles apparently has not been considered. It is expected that a large number of waves will be needed to assess and unravel the expected intricacies of the intraovarian effects of each structure (DF and CL) on the number of follicles. This seems especially so in that the two structures may be in the same ovary, opposite ovaries, or neither ovary, and each may be in the right or left ovary. The current studies in heifers considered the number of growing follicles during waves 1 and 2 in ovaries that contained both the DF and CL, DF alone, CL alone, or neither a DF nor CL. Both a separate one-way analysis for each factor (DF, CL, side, wave) and a factorial analysis were used so that the results could be compared with those of the previous studies. The hypothesis was that both the DF and CL have a positive intraovarian effect on the number of follicles in waves 1 and 2. Differential effects of the right and left ovaries were also considered.
determined by transrectal ultrasound examinations [14], and an IOI was not used if two ovulations occurred. Heifers were handled in accordance with the US Department of Agriculture Guide for Care and Use of Agricultural Animals in Research. The ovaries were examined by transrectal ultrasonography each day from Days 0 to 5 in experiment 1 (wave one) and Days 9 to 13 in experiment 2 (wave two). The diameter of each growing follicle 5 mm (experiment one) and 6 mm (experiment two) was determined. The identity of each follicle was maintained from day-to-day as described [14]. The number of growing follicles per wave in each ovary and the location of the DF and CL were recorded. The largest follicle on Day 5 in experiment 1 and Day 13 in experiment 2 was used to identify the ovary with the DF [1]. The ovaries were grouped into four intraovarian patterns based on the ovary containing the DF and CL as follows: DF–CL, DF, CL, and neither a DF nor a CL. The number of ovaries was compared among the four intraovarian patterns to determine if the frequencies of the patterns differed from equality and if the side of the ovary (right vs. left) affected the frequencies. The effect of intraovarian pattern, side of ovary, and each structure (DF and CL) on the number of growing follicles in each wave and day of maximum number of follicles were determined. Experiments 1 and 2 were done during different IOIs by different operators and were analyzed separately. Within each experiment, data that were not normally distributed were transformed to natural logarithms or ranks. The statistical analyses used SAS PROC MIXED (version 9.3; SAS Institute Inc., Cary, NC, USA). When a factor or interaction was significant or approached significance, the PDMIX800 macro in SAS [15] was used to assign letter groupings to indicate differences among groups when more than two groups were involved. Least significant difference was also used to compare follicle numbers among ovaries. Chi-square tests of independence were used for frequency data. Chi-square goodness-of-fit was used to determine if the frequency of intraovarian patterns within a wave differed from equality. A probability of P 0.05 indicated 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, unless otherwise indicated.
2. Materials and methods
3.1. Experiment 1 (wave 1)
Holstein dairy heifers (a Bos taurus breed) aged 18 to 24 months were used for a total of 65 and 62 IOIs in experiments 1 and 2, respectively. The heifers were kept under natural light in an open shelter in the northern temperate zone and were provided ad libitum access to water, trace-mineralized salt, and primarily grass hay. The estrous cycles were natural in that they were not preceded by induced luteolysis, induced ovulation, or synchronization of estrus or ovulation. The heifers were not treated and were not bred. The day of ovulation (Day 0) was
The number or frequency of ovaries in wave 1 with the intraovarian patterns of both a dominant follicle (DF) and CL, only the DF, only the CL, and neither the DF nor CL is shown (Fig. 1). The frequency of ovaries containing both structures (34%) or neither structure (34%) was significantly greater than the number that contained only the DF (16%) or only the CL (16%). The effect of location of the ovary on right or left side on the frequency of the four intraovarian patterns was not significant and did not approach significance.
3. Results
O.J. Ginther, M.M. Hoffman / Theriogenology xxx (2014) 1–7
3
Fig. 1. Intraovarian patterns in waves 1 and 2 of the locations of the dominant follicle (DF) and CL and side of the ovary (LO, left ovary; RO, right ovary). The number of ovaries containing both the DF and CL, only the DF, only the CL, and neither the DF nor the CL are shown. In wave 1, but not in wave 2, the number of ovaries with both structures or neither structure was greater than the number with only a DF or only a CL.
The day of maximum number of follicles 5.0 mm in wave 1 was Day 0.94 0.07, and mean diameter of the dominant follicle on Day 5 was 13.6 0.2 mm. Number of growing follicles in an ovary on Days 0 to 5 was not normally distributed, even after log transformation, and analyses were done by using rank. The number of follicular waves in two-wave versus three-wave IOIs did not have an effect on number of follicles in wave 1, and number of waves was not used as a factor in the analyses. There was a main effect of intraovarian pattern (P < 0.0001) and a main effect of side (right or left ovary; P < 0.03) on the number of growing follicles (Fig. 2). The effect of pattern averaged over the two sides was from greater (P < 0.05) number of follicles in the DF–CL or DF patterns than in the CL pattern or the pattern of neither DF nor CL as shown. The side effect was from more follicles in the right ovary (3.4 0.3 follicles) than in the left ovary (2.7 0.2). The number of follicles/wave was greater (P < 0.0001) combined for the two patterns with a DF (3.8 0.2 follicles/wave) than in combined two patterns without a DF (2.2 0.2). After excluding one follicle to negate the contribution of the DF, the number of follicles remained greater (P < 0.03) in the two patterns with a DF (2.8 0.2 follicles/wave) than in the other two patterns (2.2 0.2). A separate one-way analysis for each factor (DF, CL, and side) indicated significantly more follicles in the DF ovary
versus non–DF ovary, CL ovary versus non–CL ovary, and right versus left ovary both before and after removing the DF from the tally (Table 1). When data were analyzed by a three-way factorial, the main effect of DF remained significant, the main effect of CL did not approach significance, and the main effect of side was significant from more follicles in the right ovary (Table 1). Neither the two-way interactions nor the three-way interaction were significant or approached significance. Based on the sum of squares, 60% of the variation was attributable to the factorial model and the remainder to the animals. The unique main contribution of each factor was 55%, 5%, and 12% for DF, CL, and side, respectively. The remainder (28%) involved the nonsignificant two-way interactions and the three-way interaction. 3.2. Experiment 2 (wave 2) The diameter of the future dominant follicle in wave 2 on Day 9 was greater (P < 0.03) in three-wave IOIs (7.2 0.3 mm) than in two-wave IOIs (6.4 0.3 mm) but was not different significantly on Day 13 (12.0 0.3 mm vs. 12.4 0.3 mm). The day of maximum number of follicles in wave 2 was Day 10.4 0.2 and the number of follicles in the wave was 8.9 0.9. The frequency among the four intraovarian patterns in wave 2 did not differ from equality when two-wave and three-wave IOIs were combined
4
O.J. Ginther, M.M. Hoffman / Theriogenology xxx (2014) 1–7
effect of side (P < 0.1) combined for two-wave and threewave IOIs; there were more follicles in the right ovary (4.6 0.3) than in the left ovary (3.9 0.3). There was a main effect of intraovarian pattern for two-wave IOIs (P < 0.01) and for three-wave IOIs (P < 0.03) (Fig. 2). The effect of pattern for each of two-wave and three-wave IOIs was from more (P < 0.05) follicles averaged over sides in each of the DF–CL and DF patterns than in the CL pattern or the pattern with neither DF nor CL. An exception was a greater average number of follicles in the pattern with neither DF nor CL in three-wave IOIs. The interaction of pattern and side did not approach significance. The only suggestions of a greater number of follicles in the right ovary than the left ovary were in the DF–CL pattern in twowave IOIs (P < 0.06) and in the pattern with neither a DF nor CL in three-wave IOIs (P < 0.09). One-way analysis of each factor in wave 2 indicated that there were more follicles in ovaries with a DF and in ovaries on the right side (Table 2). In addition, wave 2 in threewave IOIs had more follicles than two-wave IOIs. Therefore, number of waves was used as a factor in addition to the presence or absence of a DF, presence or absence of a CL, and right or left side, resulting in a four-way factorial analysis. The four-way analysis indicated significantly more follicles in the DF ovary (Table 2). After excluding the DF from the follicle tally, a greater number of follicles in the DF ovary approached significance. None of the other main effects (CL, side, wave) were significant, but a greater number of follicles in three-wave IOIs approached significance before and after the DF was excluded from the follicle tally. None of the interactions of the four factors (two-way, three-way, or four-way) were significant, but the three-way interaction of DF, CL, and side approached significance (P < 0.07) when the DF was either included or excluded. 4. Discussion
Fig. 2. Mean SEM number of follicles in waves 1 (n ¼ 65 IOIs) and 2 (n ¼ 62 IOIs) in the left and right ovaries for four intraovarian patterns of location of the dominant follicle (DF) and CL. There was a main effect of pattern (P < 0.0001) for each of the three panels and side (P < 0.03) for the top panel. The interaction of pattern and side was not significant for any panel. ab ¼ means with a different letter within a pattern are different between right and left ovaries (P < 0.05). xy ¼ for each panel, means with a different letter between patterns averaged for left and right ovaries are different (P < 0.05).
(Fig. 1). When partitioned into two-wave and three-wave IOIs for the DF–CL and DF patterns (Fig. 3), the frequency of the DF–CL pattern was not different between wave 2 of two-wave IOIs (60%) and wave 1 (68%). The frequency of the DF–CL pattern versus the DF pattern in wave 2 of threewave IOIs (49%) approached being less (P < 0.06) than in wave 1 (68%). The frequencies of each of the four patterns in three-wave IOIs were close to 25% (range, 24.3%–25.6%), compared with 20.0% to 30.0% in two-wave IOIs. The number of follicles was normally distributed. Analysis of the number of follicles among intraovarian patterns and between sides indicated an approaching main
The DF–CL intraovarian pattern (DF and CL in same ovary) can also be expressed as an ipsilateral relationship between DF and CL; a DF pattern and a CL pattern occur during a contralateral relationship. There was no indication of an involvement of left versus right ovaries in the frequency of patterns. A greater frequency of the ipsilateral than contralateral relationship during wave 1 was not significant in a reported study [16] but was significant in the current study. The greater ipsilateral relationship between DF and CL for wave 1 did not occur for wave 2 and seemed dependent on the number of waves per IOI. When partitioned into two-wave and three-wave IOIs, the differences in frequency among intraovarian patterns in wave 2 of twowave IOIs (ovulatory wave) was in the direction of greater frequency for the ipsilateral relationship, but the differences were not significant in the current study and in a previous study [16]. In wave 2 of three-wave IOIs, frequencies were not different among the four intraovarian patterns. However, in a reported survey [16] with more three-wave IOIs (n ¼ 133), the anovulatory wave 2 had significantly more ipsilateral relationships (60%) between the DF and CL than contralateral relationships (40%). An apparent conclusion from the current and reported results is that the CL has a positive intraovarian effect on location
O.J. Ginther, M.M. Hoffman / Theriogenology xxx (2014) 1–7
5
Table 1 Wave 1. Mean SEM number of growing follicles during wave 1 for the indicated factors and for the relationship of each ovary to each factor. Experiment 1. Factor
Relationship of each factor for each ovary
Dominant follicle (DF) All follicles 5.0 mm DF excluded CL All follicles 5.0 mm DF excluded Side All follicles 5.0 mm DF excluded
Present 3.9 0.2 2.9 0.2 Present 3.6 0.2 2.9 0.2 Right ovary 3.4 0.3 2.9 0.2
Absent 2.2 0.2 2.2 0.2 Absent 2.5 0.2 2.2 0.2 Left ovary 2.7 0.2 2.2 0.2
Probabilitya One-way
Three-way
Contents lists available at ScienceDirect
Theriogenology journal homepage: www.theriojournal.com
Intraovarian effect of dominant follicle and corpus luteum on number of follicles during a follicular wave in heifers O.J. Ginther a, b, *, M.M. Hoffman a a b
Eutheria Foundation, Cross Plains, Wisconsin, USA Department of Pathobiological Sciences, School of Veterinary Medicine, University of Wisconsin–Madison, Madison, Wisconsin, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 22 November 2013 Received in revised form 13 March 2014 Accepted 18 March 2014
The intraovarian relationships among dominant follicle (DF), corpus luteum (CL), and number of follicles between Days 0 to 5 (Day 0 ¼ ovulation) in wave 1 (n ¼ 65 waves) and Days 9 to 13 in wave 2 (n ¼ 62) were analyzed in separate experiments in Bos taurus heifers. Ovaries were grouped into intraovarian patterns of DF–CL, DF alone, CL alone, and neither DF nor CL. In wave 1, the pattern frequencies of DF–CL or neither DF nor CL (34% each) were greater (P < 0.0004) than for DF alone or CL alone (16% each). The number of growing follicles 5.0 mm, was greater (P < 0.0001) in ovaries with the DF, even when the DF was removed from the tally (P < 0.03). In a factorial analysis of wave 1, there was a positive main effect of DF (3.9 0.2 vs. 2.2 0.2 follicles; P < 0.0001), but the main effect of CL and the interaction of DF and CL were not significant. In a factorial analysis of wave 2, there were more (P < 0.0001) follicles greater than 6 mm in ovaries with a DF when the DF was included and an approaching difference (P < 0.09) when the DF was excluded. The main effect of CL and the interaction of DF and CL were not significant. The hypothesis that both the DF and CL have a positive intraovarian effect on number of follicles in waves 1 and 2 was only partly supported; the DF, but not the CL, had an effect in the factorial analyses. Previous reports in cattle and sheep of a positive intraovarian effect of CL on number of follicles are questionable in that location of the DF was not considered. Ó 2014 Elsevier Inc. All rights reserved.
Keywords: Corpus luteum Dominant follicle Follicular waves Number of follicles Right and left ovaries
1. Introduction A wave of follicular activity develops about every 9 days in cattle of the Bos taurus breeds, resulting in two or three waves during the interovulatory interval (IOI) [1]. Wave 1 (first wave of the IOI) is anovulatory. Wave 2 in two-wave and three-wave IOIs is ovulatory and anovulatory, respectively. Wave 1 emerges with 4 or 5 mm follicles on the day of ovulation. When the largest follicle of the wave reaches about 8.5 mm on about 2.5 days after wave emergence, deviation occurs and is characterized by continued growth rate of the future dominant follicle and reduction in growth rate and regression of the remaining
* Corresponding author. Tel.: þ1 608 798 3777; fax: þ1 608 798 3722. E-mail address: [email protected] (O.J. Ginther). 0093-691X/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.theriogenology.2014.03.013
follicles (subordinates). Dominancy is established when the follicle reaches 10 mm, as indicated by capacity for ovulation [2]. The interval from ovulation to deviation seems similar between Bos taurus and Bos indicus breeds, but the diameter of the largest follicle at deviation is much smaller (about 6.0 mm [3]). The follicle and hormone characteristics of a wave have been studied most extensively for wave 1. The relationships between the ovary containing the dominant follicle (DF) versus the developing corpus luteum (CL) and the differential and combined intraovarian effects of the DF and CL on the number of follicles in the wave 1 have not been reported, despite the considerable knowledge on follicle and hormone dynamics [1]. Studies in ewes first demonstrated an intraovarian stimulatory effect of the CL on follicles. In unilaterally ovulating ewes, follicles are larger during the mid-estrous
2
O.J. Ginther, M.M. Hoffman / Theriogenology xxx (2014) 1–7
cycle in the ovary that contains the CL (ipsilateral relationship) than when the CL and follicles are in opposite ovaries (contralateral relationship) [4]. Four days after destruction of follicles in both ovaries, follicles are more developed in the ipsilateral ovary [4,5]. However, results of studies in cattle on a similar intraovarian relationship seem to be contradictory. More medium-sized follicles are present in the ovary ipsilateral to the CL than in the contralateral ovary at all stages of the estrous cycle [6]. Transrectal ultrasonic imaging is being used to study follicle diameters throughout the estrous cycle [7]. More follicles 2 or 3 mm are located ipsilateral to the CL than contralateral [8]. In contrast, no relationship was found between the location of the CL and the number of antral follicles on selected days of the estrous cycle [9]. Apparent intraovarian effects in cattle are further complicated by differences between the right and left ovaries. Greater follicular activity has been reported for the right ovary [10,11]. There are more follicles 4 to 6 mm and greater than13 mm in the right ovary, regardless of the location of the CL [8]. A study of spatial relationships between the DF and CL in cattle found no intraovarian relationships between the ovary containing the CL and the ovary containing the DF or between the location of the CL and the characteristics of the DF [12]. Another study found more follicular waves with the DF ipsilateral than contralateral to the CL, using all follicular waves of the estrous cycle [13]. The intraovarian effect of the CL on the number of growing follicles in cattle during a follicular wave needs clarification, due to contradictory reports. In addition, the intraovarian effect of the DF on number of follicles apparently has not been considered. It is expected that a large number of waves will be needed to assess and unravel the expected intricacies of the intraovarian effects of each structure (DF and CL) on the number of follicles. This seems especially so in that the two structures may be in the same ovary, opposite ovaries, or neither ovary, and each may be in the right or left ovary. The current studies in heifers considered the number of growing follicles during waves 1 and 2 in ovaries that contained both the DF and CL, DF alone, CL alone, or neither a DF nor CL. Both a separate one-way analysis for each factor (DF, CL, side, wave) and a factorial analysis were used so that the results could be compared with those of the previous studies. The hypothesis was that both the DF and CL have a positive intraovarian effect on the number of follicles in waves 1 and 2. Differential effects of the right and left ovaries were also considered.
determined by transrectal ultrasound examinations [14], and an IOI was not used if two ovulations occurred. Heifers were handled in accordance with the US Department of Agriculture Guide for Care and Use of Agricultural Animals in Research. The ovaries were examined by transrectal ultrasonography each day from Days 0 to 5 in experiment 1 (wave one) and Days 9 to 13 in experiment 2 (wave two). The diameter of each growing follicle 5 mm (experiment one) and 6 mm (experiment two) was determined. The identity of each follicle was maintained from day-to-day as described [14]. The number of growing follicles per wave in each ovary and the location of the DF and CL were recorded. The largest follicle on Day 5 in experiment 1 and Day 13 in experiment 2 was used to identify the ovary with the DF [1]. The ovaries were grouped into four intraovarian patterns based on the ovary containing the DF and CL as follows: DF–CL, DF, CL, and neither a DF nor a CL. The number of ovaries was compared among the four intraovarian patterns to determine if the frequencies of the patterns differed from equality and if the side of the ovary (right vs. left) affected the frequencies. The effect of intraovarian pattern, side of ovary, and each structure (DF and CL) on the number of growing follicles in each wave and day of maximum number of follicles were determined. Experiments 1 and 2 were done during different IOIs by different operators and were analyzed separately. Within each experiment, data that were not normally distributed were transformed to natural logarithms or ranks. The statistical analyses used SAS PROC MIXED (version 9.3; SAS Institute Inc., Cary, NC, USA). When a factor or interaction was significant or approached significance, the PDMIX800 macro in SAS [15] was used to assign letter groupings to indicate differences among groups when more than two groups were involved. Least significant difference was also used to compare follicle numbers among ovaries. Chi-square tests of independence were used for frequency data. Chi-square goodness-of-fit was used to determine if the frequency of intraovarian patterns within a wave differed from equality. A probability of P 0.05 indicated 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, unless otherwise indicated.
2. Materials and methods
3.1. Experiment 1 (wave 1)
Holstein dairy heifers (a Bos taurus breed) aged 18 to 24 months were used for a total of 65 and 62 IOIs in experiments 1 and 2, respectively. The heifers were kept under natural light in an open shelter in the northern temperate zone and were provided ad libitum access to water, trace-mineralized salt, and primarily grass hay. The estrous cycles were natural in that they were not preceded by induced luteolysis, induced ovulation, or synchronization of estrus or ovulation. The heifers were not treated and were not bred. The day of ovulation (Day 0) was
The number or frequency of ovaries in wave 1 with the intraovarian patterns of both a dominant follicle (DF) and CL, only the DF, only the CL, and neither the DF nor CL is shown (Fig. 1). The frequency of ovaries containing both structures (34%) or neither structure (34%) was significantly greater than the number that contained only the DF (16%) or only the CL (16%). The effect of location of the ovary on right or left side on the frequency of the four intraovarian patterns was not significant and did not approach significance.
3. Results
O.J. Ginther, M.M. Hoffman / Theriogenology xxx (2014) 1–7
3
Fig. 1. Intraovarian patterns in waves 1 and 2 of the locations of the dominant follicle (DF) and CL and side of the ovary (LO, left ovary; RO, right ovary). The number of ovaries containing both the DF and CL, only the DF, only the CL, and neither the DF nor the CL are shown. In wave 1, but not in wave 2, the number of ovaries with both structures or neither structure was greater than the number with only a DF or only a CL.
The day of maximum number of follicles 5.0 mm in wave 1 was Day 0.94 0.07, and mean diameter of the dominant follicle on Day 5 was 13.6 0.2 mm. Number of growing follicles in an ovary on Days 0 to 5 was not normally distributed, even after log transformation, and analyses were done by using rank. The number of follicular waves in two-wave versus three-wave IOIs did not have an effect on number of follicles in wave 1, and number of waves was not used as a factor in the analyses. There was a main effect of intraovarian pattern (P < 0.0001) and a main effect of side (right or left ovary; P < 0.03) on the number of growing follicles (Fig. 2). The effect of pattern averaged over the two sides was from greater (P < 0.05) number of follicles in the DF–CL or DF patterns than in the CL pattern or the pattern of neither DF nor CL as shown. The side effect was from more follicles in the right ovary (3.4 0.3 follicles) than in the left ovary (2.7 0.2). The number of follicles/wave was greater (P < 0.0001) combined for the two patterns with a DF (3.8 0.2 follicles/wave) than in combined two patterns without a DF (2.2 0.2). After excluding one follicle to negate the contribution of the DF, the number of follicles remained greater (P < 0.03) in the two patterns with a DF (2.8 0.2 follicles/wave) than in the other two patterns (2.2 0.2). A separate one-way analysis for each factor (DF, CL, and side) indicated significantly more follicles in the DF ovary
versus non–DF ovary, CL ovary versus non–CL ovary, and right versus left ovary both before and after removing the DF from the tally (Table 1). When data were analyzed by a three-way factorial, the main effect of DF remained significant, the main effect of CL did not approach significance, and the main effect of side was significant from more follicles in the right ovary (Table 1). Neither the two-way interactions nor the three-way interaction were significant or approached significance. Based on the sum of squares, 60% of the variation was attributable to the factorial model and the remainder to the animals. The unique main contribution of each factor was 55%, 5%, and 12% for DF, CL, and side, respectively. The remainder (28%) involved the nonsignificant two-way interactions and the three-way interaction. 3.2. Experiment 2 (wave 2) The diameter of the future dominant follicle in wave 2 on Day 9 was greater (P < 0.03) in three-wave IOIs (7.2 0.3 mm) than in two-wave IOIs (6.4 0.3 mm) but was not different significantly on Day 13 (12.0 0.3 mm vs. 12.4 0.3 mm). The day of maximum number of follicles in wave 2 was Day 10.4 0.2 and the number of follicles in the wave was 8.9 0.9. The frequency among the four intraovarian patterns in wave 2 did not differ from equality when two-wave and three-wave IOIs were combined
4
O.J. Ginther, M.M. Hoffman / Theriogenology xxx (2014) 1–7
effect of side (P < 0.1) combined for two-wave and threewave IOIs; there were more follicles in the right ovary (4.6 0.3) than in the left ovary (3.9 0.3). There was a main effect of intraovarian pattern for two-wave IOIs (P < 0.01) and for three-wave IOIs (P < 0.03) (Fig. 2). The effect of pattern for each of two-wave and three-wave IOIs was from more (P < 0.05) follicles averaged over sides in each of the DF–CL and DF patterns than in the CL pattern or the pattern with neither DF nor CL. An exception was a greater average number of follicles in the pattern with neither DF nor CL in three-wave IOIs. The interaction of pattern and side did not approach significance. The only suggestions of a greater number of follicles in the right ovary than the left ovary were in the DF–CL pattern in twowave IOIs (P < 0.06) and in the pattern with neither a DF nor CL in three-wave IOIs (P < 0.09). One-way analysis of each factor in wave 2 indicated that there were more follicles in ovaries with a DF and in ovaries on the right side (Table 2). In addition, wave 2 in threewave IOIs had more follicles than two-wave IOIs. Therefore, number of waves was used as a factor in addition to the presence or absence of a DF, presence or absence of a CL, and right or left side, resulting in a four-way factorial analysis. The four-way analysis indicated significantly more follicles in the DF ovary (Table 2). After excluding the DF from the follicle tally, a greater number of follicles in the DF ovary approached significance. None of the other main effects (CL, side, wave) were significant, but a greater number of follicles in three-wave IOIs approached significance before and after the DF was excluded from the follicle tally. None of the interactions of the four factors (two-way, three-way, or four-way) were significant, but the three-way interaction of DF, CL, and side approached significance (P < 0.07) when the DF was either included or excluded. 4. Discussion
Fig. 2. Mean SEM number of follicles in waves 1 (n ¼ 65 IOIs) and 2 (n ¼ 62 IOIs) in the left and right ovaries for four intraovarian patterns of location of the dominant follicle (DF) and CL. There was a main effect of pattern (P < 0.0001) for each of the three panels and side (P < 0.03) for the top panel. The interaction of pattern and side was not significant for any panel. ab ¼ means with a different letter within a pattern are different between right and left ovaries (P < 0.05). xy ¼ for each panel, means with a different letter between patterns averaged for left and right ovaries are different (P < 0.05).
(Fig. 1). When partitioned into two-wave and three-wave IOIs for the DF–CL and DF patterns (Fig. 3), the frequency of the DF–CL pattern was not different between wave 2 of two-wave IOIs (60%) and wave 1 (68%). The frequency of the DF–CL pattern versus the DF pattern in wave 2 of threewave IOIs (49%) approached being less (P < 0.06) than in wave 1 (68%). The frequencies of each of the four patterns in three-wave IOIs were close to 25% (range, 24.3%–25.6%), compared with 20.0% to 30.0% in two-wave IOIs. The number of follicles was normally distributed. Analysis of the number of follicles among intraovarian patterns and between sides indicated an approaching main
The DF–CL intraovarian pattern (DF and CL in same ovary) can also be expressed as an ipsilateral relationship between DF and CL; a DF pattern and a CL pattern occur during a contralateral relationship. There was no indication of an involvement of left versus right ovaries in the frequency of patterns. A greater frequency of the ipsilateral than contralateral relationship during wave 1 was not significant in a reported study [16] but was significant in the current study. The greater ipsilateral relationship between DF and CL for wave 1 did not occur for wave 2 and seemed dependent on the number of waves per IOI. When partitioned into two-wave and three-wave IOIs, the differences in frequency among intraovarian patterns in wave 2 of twowave IOIs (ovulatory wave) was in the direction of greater frequency for the ipsilateral relationship, but the differences were not significant in the current study and in a previous study [16]. In wave 2 of three-wave IOIs, frequencies were not different among the four intraovarian patterns. However, in a reported survey [16] with more three-wave IOIs (n ¼ 133), the anovulatory wave 2 had significantly more ipsilateral relationships (60%) between the DF and CL than contralateral relationships (40%). An apparent conclusion from the current and reported results is that the CL has a positive intraovarian effect on location
O.J. Ginther, M.M. Hoffman / Theriogenology xxx (2014) 1–7
5
Table 1 Wave 1. Mean SEM number of growing follicles during wave 1 for the indicated factors and for the relationship of each ovary to each factor. Experiment 1. Factor
Relationship of each factor for each ovary
Dominant follicle (DF) All follicles 5.0 mm DF excluded CL All follicles 5.0 mm DF excluded Side All follicles 5.0 mm DF excluded
Present 3.9 0.2 2.9 0.2 Present 3.6 0.2 2.9 0.2 Right ovary 3.4 0.3 2.9 0.2
Absent 2.2 0.2 2.2 0.2 Absent 2.5 0.2 2.2 0.2 Left ovary 2.7 0.2 2.2 0.2
Probabilitya One-way
Three-way
