Effect of Heat Stress on Follicular Development during the Estrous Cycle in Lactating Dairy Cattle'

Transcription

1 BIOLOGY OF RPRODUCTION 5, (1995) ffect of Heat Stress on Follicular Development during the strous Cycle in Lactating Dairy Cattle' D. WOLFNSON, ' 3 W.W. THATCHR, L. BADINGA, J.D. SAVI, R. MIDAN, 3 BJ. LW, 3 R BRAW-TAL, 5 and A. BRMAN 3 Department of Animal Science, 3 Faculty of Agriculture, The Hebrew University, Rehovot 71, Israel Dairy and Poultry Science Department, University of Florida, Gainesville, Florida 311 Institute of Animal Science, 5 Agricultural Research Organization, Bet-Dagan 55, Israel ABSTRACT In this study we examined, in two experiments, patterns of follicular development and dominance under conditions of heat stress. strous cycles were programmed to include two follicular waves (wave 1 and ). On Day 1 of the estrous cycle (Day = estrus), cows were assigned randomly to cooled (C; n = ) or heat-stressed (H; n = ) groups. In experiment 1, on Day 1 prostaglandin (PG) F, was injected and a controlled intravaginal drug release device (1.9 g progesterone) was inserted (this was removed on Day 17). In experiment, PGF, was injected on Day 1. Ovarian structures were examined daily by ultrasonography, and blood samples were collected at each scanning. Cycle lengths were and 17 days in experiments 1 and, respectively. Mean maximal body temperatures were higher (p <.1) in H (.3 C) than in C (3. C) cows. In experiment 1, the rate of increase in number of large follicles (- 1 mm) was greater in H than in C cows (p <.1), resulting in 53% more large follicles in H cows during wave 1; this was associated with a lower (p <.5) number of mediumsized (-9 mm) follicles between Days 7 and 1 of the cycle. Heat stress hastened (p <.) the decrease in size of the firstwave dominant follicle and hastened (p <.1) the emergence of the second dominant (preovulatory) follicle by days. In H cows, lack of decline (p <.) was noted in the number of medium-sized follicles during wave. In experiment, as in experiment 1, the rate of increase in the number of large follicles was greater in H than in C cows (p <.5), resulting in 5% more large follicles in H cows during wave 1. The preovulatory follicle emerged earlier (p <.5) in H cows, and greater variance (p <.5) indicated that in half of H cows the emergence of dominance occurred days earlier. Heat stress suppressed (p <.1) plasma concentrations of estradiol during the second half of wave 1 and tended to reduce (p <.9) plasma concentration of inhibin. Results indicate that heat stress appears to impair follicular development and to alter the dominance of the first-wave dominant follicle and the preovulatory follicle in cattle. INTRODUCTION Low summer fertility in cattle induced by heat stress is a multifactorial problem, because hyperthermia of various organs and tissues results in diverse functional alterations and impairments [1-3]. In addition, the level to which tissue temperature is increased during summer thermal stress modulates these responses []. Recent studies have indicated that ovarian follicles are susceptible to thermal stress [5]. The first-wave dominant follicle in heat-stressed, lactating cows was found to be smaller in diameter and to contain less fluid than that of controls on Day of the cycle. Ultrasonographic records during the first 7 days of the cycle showed that follicular dominance had been altered, as indicated by the lack of decrease in the number of mediumsized follicles and the size of the second largest, subordinate follicle [5]. Furthermore, heat stress reduced granulosa cell viability, aromatase activity, and androstenedione production in theca cells from Day 7 first-wave dominant follicles []. A chronic effect of heat stress also was evidenced by lower concentrations of estradiol ( ) in plasma and fol- Accepted January, Received April, 199. 'Supported by BARD grant IS Correspondence. FAX: licular fluid in late summer, as compared to concentrations in cows in early summer that had experienced this stress for a shorter period [5]. The pattern of follicular development also was affected by the status of energy balance in lactating cows [7, ]. Follicular dynamics were altered by a negative energy balance and by factors affecting energy balance, such as stage of lactation, level of milk production, intake of energy-rich nutrients such as calcium salts of long-chain fatty acids, and injection of bovine somatotropin. The number of large follicles and the concentration of during the preovulatory period differed between postpartum lactating and nonlactating cows. Dietary fats stimulated follicular growth when fed to increase the energy balance or when fed in a ration that was isocaloric with the control diet. Alterations in follicular development during heat stress in lactating cows may interact with other components of the reproductive system. For example, impairment of preovulatory follicle function may affect subsequent CL function and progesterone (P ) secretion, resulting in potential alterations of oviductal or uterine environments that may affect the developing embryo [9]. The reduced secretion of P from luteal cells of CL collected in summer versus winter [1] may be related to impaired function of the preovulatory follicle. Also, conception rates in lactating dairy cows Downloaded from by guest on 1 November 1 11

2 FOLLICULAR DVLOPMNT IN HAT-STRSSD COWS 117 are related inversely to ambient temperature during the follicular phase [11]. This may reflect effects of heat stress on the ovulatory oocyte [1, 13], as well as on functions of follicular granulosa and theca cells. The present study extends characterization of follicular development under conditions of heat stress, from the first 7 days of the cycle [5] to the entire estrous cycle. It examines plasma concentrations of and inhibin, and the patterns of the first and second (preovulatory) follicular waves, during periods of recruitment, emergence, growth, and dominance of follicles. This evaluation was performed in two separate experiments, at early and late stages of lactation. MATRIALS AND MTHODS Two experiments were conducted during the summer months, one in Florida (experiment 1) and the other in Israel (experiment ), with cyclic lactating Holstein cows. xperiment 1 Animals. xperiment 1 was carried out on 1 cows in late lactation (13.5 mo after calving; mean yield of 1 kg milk/day). Cows were fed a complete mixed ration containing 1.5% protein and 1.5 Mcal/kg dry matter. strous cycles were synchronized by a progestin and prostaglandin (PG) F,. An s.c Norgestomet ear implant ( mg, Synchromate-B; Sanofi Animal Health Inc., Overland Park, KS) was implanted for 9 days, and 5 mg of PGF,, (Lutalyse; Upjohn Co., Kalamazoo, MI) was injected i.m. 7 days after insertion of the implant. Cows expressing estrous behavior within h after implant removal were included in the subsequent experiment. Heat stress management. Before the experiments, cows were kept under a shade structure equipped with a sprinkling and ventilation cooling system [1,15]. On Day 1 (Day = day of estrus), cows were assigned randomly to either cooled (C; n = in each experiment) or heat-stressed (H; n = in each experiment) groups. Cows assigned to the C group remained under the shade structure at all times. Cows assigned to the H group were exposed to direct solar radiation from h to approximately 15 h daily, and thereafter moved to a shade structure without a cooling system, until the next morning. The exact time each cow was removed from the heat stress depended on the degree of hyperthermia and the severity of the thermal stress on a particular day. Body temperatures were monitored in the morning (7 h), afternoon (15 h), and evening (19 h) hours, to ensure maintenance of the desired level of hyperthermia (.5 C) in H cows, and normothermia (c 39. C) in C cows. xperimentalprotocol. An estrous cycle with only two follicular waves was programmed. This prevented development of a third follicular wave and enhanced uniformity among animals to improve experimental sensitivity. We followed an experimental system for the control of follicular waves that was used successfully in the past [7]. During the first 1 days of the cycle, coinciding with the first follicular wave and its turnover to the second follicular wave, no treatment was applied. On Day 1 of the cycle, PGF, was injected, and a controlled intravaginal drug-releasing device (CIDR, asy-breed; AHI Plastics, Hamilton, New Zealand), containing 1.9 g P, was inserted. The CIDR was removed 5 days later, on Day 17. The absence of an active CL and the low P concentration in plasma maintained by the CIDR prevented early ovulation and enabled development of the preovulatory follicle from the second-wave dominant follicle until estrus (Day ). All cows ovulated after estrus. Data related to wave 1 and dominant and subordinate follicles from one C cow that ovulated the firstwave dominant follicle were excluded from the analysis. Data collection. Ovarian follicular development was monitored by ultrasonography using an quisonics instrument (LS 1; Tokyo Keiki Co., Tokyo, Japan) equipped with a 7.5-MHz transrectal linear transducer. Positions and sizes of the follicles ( 3 mm) and CL in the ovaries were traced at each scanning, and the exact location of the follicles was recorded. This allowed an accurate identification of individual follicles on successive days. Follicular responses examined included size of the dominant and subordinate (second largest) follicles in each follicular wave, size of the CL, and number of small (3-5 mm), medium (-9 mm), and large ( 1 mm) follicles. Data were obtained daily, from Day 1 until estrus. Blood samples were collected daily and centrifuged, and plasma was stored at - C until assayed for P. xperiment xperimental design. xperiment was performed in Israel after completion of experiment 1 in Florida. This experiment extended investigations to higher-producing dairy cows in early stages of lactation; this situation more typically represents production systems when cows are normally inseminated. As in experiment 1, an estrous cycle with two follicular waves was programmed to enhance uniformity among animals and to improve experimental sensitivity. However, a single dose of PGF, was injected on Day 1 of the cycle to regress the CL and induce ovulation of the second-wave dominant follicle (in all cows). This alternative approach is more practical than the one used in experiment 1 but allowed less time for the second (preovulatory) dominant follicle to develop until estrus. Cows in experiment were not treated with PGF,, until Day 1 of the cycle. Therefore, the periods between Day 1 and Day 1 in both experiments were comparable: in each case the period comprised the first follicular wave (wave 1) and the emergence of the second follicular wave (wave ). Regulation of wave differed in experiments 1 and. Animals and data collection. All experimental details described above for experiment 1 regarding animals, heat stress management, experimental protocol, and data col- Downloaded from by guest on 1 November 1

3 11 e - WOLFNSON T AL. ment (model SSD-1DXII; Aloka Co., Tokyo, Japan), equipped with 5.-MHz transducer operated in single-frame mode. Data were obtained daily, from Day 1 until Day of estrus (except on Days,, and ). Blood samples were collected daily for P, and inhibin determinations. I FIG. 1. Plasma concentrations (mean SM) of P in cooled (n =, circles) and heat-stressed (n =, squares) cows in experiment 1 during programmed estrous cycles. A single dose of PGF was injected on Day 1 and a CIDR device was inserted on that day for 5 days. lection were similar for experiment, except for the following: experiment was carried out on 1 cows in early lactation ( mo after calving; mean yield of 3 kg milk/ day); cows were fed a complete mixed ration containing 1.% protein and 1. Mcal/kg dry matter; for estrous synchronization, 5 pxg Cloprostenol (a PGF,. analogue, strumate; Coopers, Burgwedel, Germany) was injected i.m. (instead of Lutalyse), and a CIDR was inserted (instead of the Norgestomet ear implant in experiment 1) for 9 days; PGF,, was injected on Day 1 of the experimental cycle, and cows manifested estrus 3 days later. This procedure provided a shorter growth period for the preovulatory follicle than in experiment 1. This latter change resulted in a shorter estrous cycle (17 days). Ovarian follicular development was monitored with an Aloka ultrasound instru-._ o o O, I I 11 I I I FIG.. Growth pattern in experiment 1 of dominant (open squares, solid squares) and subordinate (open circles, solid circles) follicles from first and second follicular waves, respectively, of all cows In = 11). I I Hormone Analyses Concentrations of P in plasma were analyzed by singleantibody RIAs described previously, for experiment 1 [1] and experiment [, 17]. Assay sensitivities were.1 and.1 ng/ml in experiments 1 and, respectively. Intra- and interassay CVs were, respectively, 1.5% and.% in experiment 1, and 7% and 9.3% in experiment. Plasma concentrations of (experiment ) were analyzed by a previously described and validated single-antibody RIA [1, 19]. Antibody (1:7 dilution; Biomakor, Rehovot, Israel) crossreacted with estrone (15%), testosterone (<.1%), P (<.1%), estriol (1.3%), and estradiol-17a (1.%). Assay sensitivity was. pg/ml. Intra- and interassay CVs were 1.% and 17.%, respectively. Plasma concentrations of immunoreactive inhibin (experiment ) were measured by RIA using a kit supplied as a gift by the NIAMDD (Bethesda, MD). The assay procedure has been described recently []. Purified 31-kDa inhibin (1 ng = 7 U of World Health Organization [Geneva, Switzerland] porcine inhibin standard /9) served as the standard. The antiserum exhibited significant cross-reactivity (%) with the pro a inhibin subunit []. Sensitivity of the assay was. ng/ml. Intraand interassay CVs were 5% and 1%, respectively. Data Analysis Sizes of the dominant and subordinate follicles of waves 1 and, number of follicles classified by size, and concentrations of the hormones were analyzed according to the General Linear Model procedure of the Statistical Analysis System (SAS Institute Inc., Cary, NC). The statistical model included effects of treatment (C vs. H groups), cow (within treatment), day of estrous cycle, and treatment-by-day interaction. Data were analyzed separately for experiments 1 and. Data were analyzed for the whole estrous cycle or separately for waves 1 and. Tests of homogeneity of regression curves, and error variance tests were performed. xperiment 1 RSULTS Body temperatures and plasma P. xposure to solar radiation increased (p <.1) maximal body temperatures of cows in the H group to a level typical of hyperthermia experienced commonly during the summer by lactating cows. Mean maximal body temperatures at 15 h were.3 +. and C for H and C groups, respectively. After approximately h in the shade, H cows were still hyper- Downloaded from by guest on 1 November 1

4 FOLLICULAR DVLOPMNT IN HAT-STRSSD COWS Uo LL r'll...,i; "~.. I A a i uay Uol yc;ec FIG. 3. Growth patterns of first- and second-wave dominant follicles, respectively, in cooled (open circles, open squares; n = 5) and heat-stressed (solid circles, solid squares; n = ) cows of experiment 1 during estrous cycle. a. -I a.u thermic, with an average rectal temperature of C, reflecting a slow rate of body cooling after exposure to solar radiation. Plasma P concentrations throughout the estrous cycle (Fig. 1) did not differ between H and C cows. Concentrations of P declined gradually after PGF a injection and CIDR insertion on Day 1. P dropped further to its lowest level on Day 17 after CIDR removal on Day 17. Follicular dynamics. The growth pattern of wave 1 and dominant and subordinate follicles for all cows are presented in Figure. The wave 1 dominant follicle peaked in size on Day 9 and was mm larger in diameter than the preovulatory (second dominant) follicle on day of estrus. The wave dominant follicle stopped growing around Day 17, and a 3-day period of slow growth (static phase) was evident, until the day of estrus (Day ). The subordinate follicles completed their growth at Day 5 and Day 1 for waves 1 and, respectively. Heat stress altered the growth patterns of the first- and second-wave dominant follicles (Fig. 3). The average size of the first-wave dominant follicle was similar in C and H cows; however, its size decreased earlier in H cows than in C cows (Day 1 vs. Day 17, treatment-by-day interaction, p <.). The size of the second-wave dominant follicle in H cows increased earlier (treatment-by-day interaction, p <.1), suggesting the earlier emergence of the secondwave preovulatory follicle. For example, the size of the preovulatory follicle on Days and 1 was 5 and 7 mm in H cows, and only 3 and mm in C cows, respectively. Size of the CL was not affected by heat stress (data not shown). Heat stress altered the frequency of follicles in various size classes (Fig. ). Heat-stressed cows tended (p <.1) to have fewer small follicles (3-5 mm) than C cows (-%) during wave (lower panel). A treatment-by-day interaca am FIG.. Number of small (3-5 mm, lower panel), medium (-9 mm, middle panel), and large ( 1 mm, upper panel) follicles in cooled (n =, open circles) and heat-stressed (n =, solid squares) cows during estrous cycle in experiment 1. tion was detected (p <.5, middle panel) for the number of medium-sized (-9 mm) follicles in wave 1. Heat-stressed and C cows had a comparable number of medium follicles during the first half (Days -) of wave 1. However, the number of medium-sized follicles was lower in H cows between Days 7 and 1 of the cycle. During the second follicular wave, there were fewer medium-sized follicles in the H cows, and a distinct decline detected in the C group was not detected in the H group (p <.). Associated with the smaller number of medium-sized follicles developed during the spontaneous first wave in H cows was an increase in the number of large follicles ( 1 mm; Fig., upper panel), representing a decrease in first-wave follicular dominance of H cows. The rate of increase in number of large follicles was greater in H than in C cows ( >.1.3 follicles per day; p <.1). As a consequence, 53% (.3 > 1.5) more large follicles were present in H than in C cows during Days -1 of the estrous cycle. xperiment Body temperatures and hormone concentrations. Similar to the effects in experiment 1, exposure to solar Downloaded from by guest on 1 November 1

5 111 WOLFNSON T AL. I- C o O, FIG. 5. Plasma concentrations (mean + SM) of P in cooled (n =, open circles) and heat-stressed (n =, solid squares) cows in experiment during programmed estrous cycles. Single dose of PGF, was injected on Day 1. radiation increased (p <.1) maximal body temperatures of H cows to.3 +. and C in C cows. Plasma P concentrations (Fig. 5) were not different for H and C cows; in both C and H cows P dropped abruptly after PGF. injection on Day 1. A treatment-by-day interaction was detected (p <.1) for plasma concentrations (Fig., upper panel) during wave 1. Concentrations of were higher during the first days of the cycle and lower between Days and in the H group, as compared to the C group. Con- ' o,-. C FIG.. Plasma concentrations of (upper panel) and 31-kDa immunoreactive inhibin (lower panel) in cooled (n =, open circles) and heatstressed (n = ; solid squares) cows during programmed estrous cycle in experiment. _. a-q LL FIG. 7. Growth pattern in experiment of dominant (open squares, solid squares) and subordinate (open circles, solid circles) follicles from first and second follicular waves, respectively, of all cows (n = 1). centrations of plasma inhibin varied with stage of the cycle (day effect,p <.1; Fig., lower panel), but did not differ according to treatment (.3.7 and ng/ ml in C and H groups, respectively; p <.9). Among the controls (n = ), cows had appreciably high inhibin concentrations, whereas concentrations in the remaining cows were comparable to the values recorded in the H cows. Follicular dynamics. The growth patterns of wave 1 and dominant and subordinate follicles for all cows (C and H) in experiment are presented in Figure 7. The first dominant follicle peaked in size on Days 9-1 (similar to experiment 1). The second-wave dominant follicle grew linearly until the day of estrus (Day 17). As in experiment 1, completion of growth of subordinate follicles occurred on Days 5 and 1 for waves 1 and, respectively. In experiment, unlike experiment 1, the growth pattern of the first dominant follicle was not affected by heat stress (Fig. ). However, as in experiment 1, heat stress increased the size of the second-wave dominant follicle earlier (treatment-by-day interaction, p <.5); the size of the preovulatory follicle on Day was 5 mm in H cows and only 3 mm in C cows. Further analysis indicated that the variances were significantly different between groups (H > C, p <.5) for some discrete responses, such as the day individual second dominant follicles became medium-sized (5-9 mm) or the day a second dominant follicle emerged mm or more larger than its subordinate (second-largest) follicle (emergence of dominance). For emergence of dominance, for example, variances were.17 and 1.9 for C and H cows, respectively. The larger variance in H cows reflects 3 cows with emergence of dominance on Days 7, 7, and, and 3 cows with emergence on Day 1, 1, and 15, indicating markedly earlier emergence (-5 days) induced by heat stress in half of the H cows in experiment Downloaded from by guest on 1 November 1

6 FOLLICULAR DVLOPMNT IN HAT-STRSSD COWS 1111._ - w co '.5 -. o T a U r% - -a _ '^ U uay ot CyCle FIG.. Growth patterns of first- and second-wave dominant follicles, respectively, in cooled (open circles; open squares; n = ) and heat-stressed (solid circles; solid squares; n = ) cows of experiment during estrous cycle.. As in experiment 1, size of the CL was not affected by heat stress (data not shown). The effect of heat stress on frequency of follicles in various size classes is presented in Figure 9. As in experiment 1, H cows tended to have (p >.1) fewer (-3%) small follicles than did C cows (lower panel). Unlike the significant effects of heat stress on medium follicles found in experiment 1, the number of medium-sized follicles in experiment (middle panel) was not affected by heat stress. However, as in experiment 1, heat stress increased the number of large follicles during wave 1 (Fig. 9, upper panel). As in experiment 1, the rate of increase in number of large follicles was greater in H than in C cows (.1 +. >. +. follicles per day;p <.1). As a consequence, 5% more (1.5 > 1.) large follicles were present in H than in C cows during Days -11 of the cycle. DISCUSSION This study comprehensively described follicular development throughout the entire estrous cycle of heat-stressed cows at two different stages of lactation. It includes the development of the first and second follicular waves and in particular the patterns of growth and regression of the firstwave dominant follicle, and the emergence and growth of the second dominant (preovulatory) follicle. Changes in the patterns of follicular development recorded in experiments 1 and during wave 1 and during the turnover period to wave indicate that dominance of the first-wave follicle was altered by heat stress. These changes included an enhancement in number of large follicles (Figs. and 9) and earlier emergence of the wave dominant follicle (Figs. 3 and ), for both experiments. This corresponded well with the depressed plasma found in experiment (Fig. ). a co , i FIG. 9. Number of small (3-5 mm, lower panel), medium (-9 mm, middle panel), and large ( 1 mm, upper panel) follicles in cooled (n =, open circles) and heat-stressed (n =, solid squares) cows during estrous cycle, in experiment. Heat stress induced a marked increase in the number of large follicles during wave 1, resulting in a 5% increase in the number of large follicles during the second half of wave 1 in both experiments 1 and (Figs. and 9; upper panels). This response may have resulted from heat stressinduced reduction in dominance of the large follicle, permitting growth of one more large follicle. The increase in the number of large follicles during heat stress probably accounts for the decrease in medium-sized follicles in H cows during wave 1 (Fig. ). This relationship was found in experiment 1 and not in experiment. In experiment, movement of medium follicles into the large class of follicles and a potentially greater replenishment of mediumsized follicles (heat-stressed decrease in number of small follicles, Fig. 9, lower panel) would suggest a greater reduction in follicular dominance. This difference in follicular dynamics may be related to the potential difference in energy balance and/or stage of lactation of the cows between the two experiments [7]. nergy balance has recently been shown to affect the pattern of follicular development during wave 1 [1]. Although energy balance was not estimated in the present studies, cows in experiment 1, which were in late lactation, had a positive energy balance, whereas the high milk-yielding cows in experiment during early Downloaded from by guest on 1 November 1

7 111 WOLFNSON T AL. lactation were in a state of either negative or less positive energy balance. The latter state is usually amplified during heat stress, when feed intake is suppressed. The second indication of heat stress-induced attenuation in dominance of the first-wave dominant follicle is the earlier emergence of the second-wave dominant/preovulatory follicle. This phenomenon was observed in both experiments. In experiment 1, the heat stress induced early emergence by days, as is reflected clearly in the difference between C and H curves in Figure 3. In experiment, a significantly high variation among H cows was detected regarding the day individual second dominant follicles became medium-sized, or the day each dominant follicle emerged mm or larger than its subordinate follicle. According to these criteria, a marked -day difference was recorded between H cows whose preovulatory follicle emerged earlier and those that did not. Since time of estrus and ovulation were not altered by the thermal state, the earlier emergence (by - days) of preovulatory follicles of heatstressed cows may result in ovulation of older follicles. Bovine follicles usually remain viable for only a few days before onset of atresia []. Possible involvement of ovulation of aged follicles during heat stress in the low summer fertility syndrome needs further investigation. The average size of the first-wave dominant follicle was not affected by heat stress in either experiment. This contrasts to the smaller diameter and less fluid found in firstwave dominant follicles of heat-stressed cows on Day of the cycle [5]. It is worth noting, however, that follicle size is not a good indicator of functional follicular dominance [3]. Heat stress seemed to depress not only dominance of the first dominant follicle but also that of the second-wave dominant follicle. This was evidenced, in experiment 1 only, by a lack of decline in the number of medium-sized follicles in H cows during most of the period of wave development (Days 13-19). This contrasts with the significant and consistent decline noted in C cows from Day 13 to the day of estrus (Day ; Fig. ). The dominant follicle exerts a transient inhibitory influence on growth of cohort subordinate follicles [3,], and the decrease in number of medium-sized follicles is considered a sensitive indicator of follicular dominance. A pronounced and significant decline in the number of medium-sized follicles was noted during the preovulatory period (Days 1-17, Fig. ) of wave in experiment for both C and H cows. The lack of alteration in dominance of the preovulatory follicle due to heat stress may be due to experimental management of second-wave follicle. Injection of PGF,, induced CL regression, and the subsequent follicular phase occurred earlier (Day 1) in the cycle, causing the wave dominant follicle to approach estrus (Day 17) and ovulation at an earlier stage (Fig. 5). This is in contrast to the P + PGF, management of experiment 1 (Fig. 1), which gave the preovulatory follicle a longer period to develop, including a longer static phase [5], resulting in estrus on Day. Since the heat stress-induced alterations in dominance were not consistent during the preovulatory phase of wave between experiments, it is possible that alterations of dominance due to heat stress occur in the presence of a functional CL but not during the preovulatory period. Induced ovulation of younger preovulatory follicles under conditions of thermal stress to improve summer fertility is currently being assessed in fertility studies. stradiol and inhibin secretion by the largest follicle within a wave is thought to mediate, in part, the inhibitory effect of the dominant follicle on subordinate follicles []. In the present study, heat-stressed animals had a significantly lower concentration of during Days - of wave 1, when the largest follicle exerts its dominance. This reduction may serve as an additional indication of altered follicular function and dominance under conditions of thermal stress. Heat stress tended to lower plasma concentrations of 31-kDa immunoreactive inhibin (Fig. ). This difference was not statistically significant because of large animal variation. For that reason, and also because of the high cross-reaction of the antiserum with the pro a subunit [7], results should be interpreted with caution. The range of 31-kDa inhibin found injapanese brown cattle [] during the 1 days before and after the LH peak (.-.3 ng/ml) was similar to the concentration found in C cows in the present study (.3 ng/ ml). Inhibin and are involved in regulation of pituitary FSH secretion []. Alteration of FSH secretion by heat stress may affect follicular dynamics. The patterns of inhibin and FSH secretion under heat stress warrant further investigation. In conclusion, follicular dynamics are altered and follicular dominance is depressed by heat stress. This was supported by significant alterations under heat stress in the number of large follicles and an associated decrease in number of medium-sized follicles, decline in plasma concentration of ; earlier emergence of the preovulatory follicle, and lack of decline in the number of medium-sized follicles during the follicular phase. The altered follicular development, which was evident in early and late stages of lactation, may be associated with depressed summer fertility. ACKNOWLDGMNTS The authors thank the NIAMDD for its generous provision of reagents for inhibin RIA, and M. Maman and Y. Graber for their technical assistance. RFRNCS 1. Roman-Ponce H, Thatcher WW, Wilcox CJ. Hormonal interrelationships and physiological responses of lactating dairy cows to shade management system in a subtropical environment. Theriogenology 19; 1: Wolfenson D, Flamenbaum, Berman A Hyperthermia and body energy store effects on fertility and ovarian function in dairy cows. J Dairy Sci 19; 71: Putney DJ, Malayer JR, Gross TS, Thatcher WW, Hansen PJ, Drost M. Heat-stress induced alterations in the synthesis and secretion of proteins and prostaglandins Downloaded from by guest on 1 November 1

8 FOLLICULAR DVLOPMNT IN HAT-STRSSD COWS 1113 by cultured bovine conceptuses and uterine endometrium. Biol Reprod 19; 39: Putney DJ, Drost M, Thatcher WW. mbryonic development in superovulated dairy cattle exposed to elevated ambient temperatures between day 1 to 7 postinsemination. Theriogenology 19; 3: Badinga L, Thatcher WW, Diaz T, Drost M, Wolfenson D. ffect of environmental heat stress on follicular steroidogenesis and development in lactating Holstein cows. Theriogenology 1993; 39: Lew BJ, Wolfenson D, Meidan R Heat stress affects steroid content of follicular fluid and steroid production by granulosa and theca cells in the bovine dominant follicle. In: Program of the annual meeting of the Israel ndocrine Society; 1993; Tel-Aviv. Abstract Lucy MC, Savio JD, Badinga L, De La Sota RL, Thatcher WW. Factors that affect ovarian follicular dynamics in cattle. J Anim Sci 199; 7: Lucy MC, Staples CR, Michel FM, Thatcher WW. ffect of feeding calcium soaps to early postpartum dairy cows on plasma prostaglandin F., luteinizing hormone, and follicular growth. J Dairy Sci 1991; 7: Breuel KF, Lewis P, Schrick FN, Lishman AW, Inskeep K, Butcher RL. Factors affecting fertility in the postpartum cow: role of the oocyte and follicle in conception rate. Biol Reprod 1993; : Wolfenson D, Luft O, Berman A, Meidan R ffects of season, incubation temperature and cell age on progesterone and prostaglandin F, production in bovine luteal cells. Anim Reprod Sci 1993; 3: Ingraham RH, Stanley RW, Wagner WC. Relationship of temperature and humidity to conception rate of Holstein cows in Hawaii. J Dairy Sci 197; 59: Baumgartner AP, Chrisman CL. Ovum morphology after hyperthermic stress during meiotic maturation and ovulation in mouse. J Reprod Fertil 191; 1: Putney DJ, Mullins S, Thatcher WW, Drost M, Gross TS. mbryonic development in superovulated dairy cattle exposed to elevated ambient temperatures between the onset of estrus and insemination. Anim Reprod Sci 199; 19: Berman A, Wolfenson D. nvironmental modifications to improve production and fertility. In: Van Horn HH, Wilcox CJ (eds.), Large Dairy Herd Management. Champaign: ADSA; 199: Strickland JT, Bucklin RA, Nordstedt RA, Beede DK, Bray DR Sprinkler and fan cooling system for dairy cows in hot, humid climates. Appl ngin Agric 199; 5: Knickerbocker J, Thatcher WW, Bazer FW, Drost M, Barron DH, Fincher KB, Roberts RM. Proteins secreted by day 1 to 1 conceptuses extend corpus luteum function in cows. J Reprod Fertil 19; 77: Meidan R, Girsh, Bloom O, Aberdam. In vitro differentiation of bovine theca and granulosa cells into small and large luteal-like cells: morphological and functional characteristics. Biol Reprod 199; 3: Badinga L, Driancourt MA, Savio JD, Wolfenson D, Drost M, De La Sota RI, Thatcher WW. ndocrine and ovarian responses associated with the first-wave dominant follicle in cattle. Biol Reprod 199; : Tortonese DJ, Lewis P, Papkoff H, Inskeep K. Roles of the dominant follicle and the pattern of estradiol in induction of preovulatory surge of LH and FSH in prepubertal heifers by pulsatile low doses of LH. J Reprod Fertil 199; 9: Braw-Tal R, Bor A, Gootwine. Plasma immunoreactive inhibin and FSH in prepubertal Assaf and Booroola-Assaf ewe lambs. Domest Anim ndocrinol 1993; 1: Lucy MC, De La Sota RL, Staples CR, Thatcher WW. Ovarian follicular populations in lactating dairy cows treated with recombinant bovine somatotropin (sometribove) or saline and fed diets differing in fat content and energy. J Dairy Sci 1993; 7: Ireland DT. Control of follicular growth and development. J Reprod Fertil Suppl 197; 3: Fortune J, Sirois J, Turzillo AM, Lavoir M. Follicle selection in domestic ruminants. J Reprod Fertil Suppl 1991; 3: Ginther OJ, Knopf L, Kastelic JP. Temporal association among events in cattle during oestrous cycles with two and three follicular waves. J Reprod Fertil 199; 7: Savio JD, Thatcher WW, Badinga L, De La Sota RL, Wolfenson D. Regulation of dominant follicle turnover during the oestrous cycle in cows. J Reprod Fertil 1993; 97: Findlay JK An update on the roles of inhibin, activin, and follistatin as local regulators of folliculogenesis. Biol Reprod 1993; : Knight PG, Beard AJ, Wrathall JH, Castillo RJ. vidence that the bovine ovary secretes large amounts of monomeric inhibin alpha subunit and its isolation from bovine follicular fluid. J Mol ndocrinol 199; :19-.. Kaneko H, Watanabe G, Taya K, Sasamoto S. Changes in peripheral levels of bioactive and immunoreactive inhibin, estradiol-17, progesterone, luteinizing hormone, and follicle-stimulating hormone associated with development in cows induced to superovulate with equine chorionic gonadotropin. Biol Reprod 199; 7:7-. Downloaded from by guest on 1 November 1