H. Cardenas and W.F. Pope
Department of Animal Sciences
The effects of 0 (vehicle), 1, 10, or 100 mg of testosterone when administered on days 17 and 18 (day 0 = first day of the estrous cycle) were examined in 82 gilts regarding the number of preovulatory follicles by day 19 or the subsequent number of corpora lutea (CL) and blastocysts by day 11. The mean number of preovulatory follicles increased (P < 0.01) in a dose-related manner. Likewise, gilts that received either 1, 10, or 100 mg of testosterone had more (P < 0.05) CL than gilts treated with vehicle. The mean number of blastocysts was greater (P = 0.06) in gilts receiving 1 mg of testosterone but decreased (P < 0.05) in gilts treated with 10 mg of testosterone as compared to gilts receiving vehicle. Similarly, recovery rates of blastocysts decreased (P < 0.05) in gilts treated with 10 or 100 mg of testosterone relative to gilts administered 0 or 1 mg of testosterone. These results indicated that, although the 10 and 100 mg dosages of testosterone were detrimental to blastocysts survival, the 1 mg dosage increased the number of CL and day 11 blastocysts.
Testosterone is used as a substrate for follicles to synthesize estradiol. Estradiol, in turn, is required by follicles to remain healthy (nonatretic) and grow. For example, the concentration of estradiol in follicular fluid of atretic (unhealthy) follicles is lower than those classified as nonatretic.
Although cause and effect relationships remain unknown, the low amounts of estradiol in follicles might result from either decreased aromatase activity, low availability of substrate, or both. The role of androgens as a limiting substrate for estradiol synthesis in follicles has been suggested in other species. However, no evidence exists to demonstrate that testosterone treatment affects follicle health and ovulation rate in swine.
The objectives of the present experiments were to determine whether administration of testosterone during the follicular phase of the estrous cycle alters the number of preovulatory follicles, number of CL, and day 11 blastocysts in gilts.
General. A homogeneous group of crossbred Landrace (1/4) x Yorkshire (1/4) x Duroc (1/2) gilts, weighing from 140 to 160 kg, were used in this study. Estrus was detected twice daily (0830 and 2000) in the presence of intact boars. The first day of estrus was considered the day 0 of the estrous cycle. Only gilts that previously exhibited estrous cycles of 19 to 21.5 days were included in the experiments.
Experiment 1. Gilts (n = 5 per group) were administered intramuscular injections of 0 (vehicle, control), 1, 10, or 100 mg of testosterone on days 17 and 18 of the estrous cycle. Gilts were ovariectomized on day 19, and the number and stage of atresia of the surface follicles > 3 mm in diameter were determined. Follicular diameter was measured to the nearest mm at the surface of the ovary by use of a caliper. Follicles were classified as medium (3 to 5 mm) or large (6 mm).
Experiment 2. Gilts (n = 13) were treated with testosterone as before and observed for the subsequent display of estrus. Gilts were bred twice to different boars during the first day of estrus and then ovariohysterectomized on day 11. Blastocysts were recovered by flushing the uterus with physiological saline. Blastocysts were counted and their maximum diameter and stage of morphological development recorded. Recovery rate of blastocysts was calculated by dividing the number of blastocysts by the number of corpora lutea and multiplying by 100.
Statistical Analysis. Data regarding follicles, CL, blastocysts, and estrous cycle length were analyzed by analysis of variance for a completely randomized design (Steel and Torrie, 1980). Means were compared to one another by using orthogonal contrasts. These analyses were performed using SYSTAT (Wilkinson, 1990). Pregnancy rates (percentage of gilts from which blastocysts were recovered) were analyzed by Chi-square tests of independence (FREQ procedure; SAS, 1988). The diameters of blastocysts were subjected to one-way analysis of variance with subsamples (Steel and Torrie, 1980) using the GLM procedure of SAS (SAS, 1988). The model included the effects of dose and animals within dose.
Testosterone treatment influenced (P < 0.01) the number of preovulatory follicles on day 19. Gilts treated with 10 or 100 mg of testosterone had more (P < 0.05) preovulatory follicles than those treated with vehicle; however, gilts that received 1 mg of testosterone or vehicle did not differ (Table 1).
| Table 1. Mean number of larger ( > 6 mm) follicles on day 19 of the estrous cycle in gilts treated with vehicle or testosterone. | ||||
| Group1 | Total | Nonatretic
(Preovulatory) |
Atretic | % Atretic |
| 0 | 20.2 | 14.0a | 6.2 | 30.1 |
| 1 | 20.2 | 15.8ab | 4.4 | 21.4 |
| 10 | 23.4 | 18.0bc | 5.4 | 22.2 |
| 100 | 25.0 | 20.6c | 4.4 | 16.2 |
| SEM2 | 1.9 | 1.3 | 1.4 | 3.2 |
| 1 Dosages of testosterone (mg/day).
2 n = 5. abc Means without common superscripts are different (P < 0.05). | ||||
Testosterone treatment did not influence the total number (nonatretic plus atretic) of follicles or the number and percentage of atretic follicles within the classifications of large (Table 1) and medium (Table 2).
| Table 2. Mean number of medium (3 to 5 mm) follicles on day 19 of the estrous cycle in gilts treated with vehicle or testosterone. | ||||
| Group1 | Total | Nonatretic
(Preovulatory) |
Atretic | % Atretic |
| 0 | 14.0 | 1.2 | 12.8 | 92.7 |
| 1 | 16.4 | 4.4 | 12.0 | 80.9 |
| 10 | 15.8 | 1.8 | 14.0 | 91.3 |
| 100 | 14.8 | 1.8 | 13.0 | 89.8 |
| SEM2 | 4.8 | 1.9 | 3.6 | 6.8 |
| 1 Dosages of testosterone (mg/day).
2 n = 5. Means are not different. | ||||
Gilts treated with 1, 10, or 100 mg of testosterone had more (P < 0.05) CL than gilts receiving vehicle (Table 3). The increase in CL per gilt ranged from 2.3 (1 mg dose) to 3.5 (100 mg dose) relative to the mean of gilts treated with vehicle. The number of blastocysts was greater (P = 0.06) in gilts that received 1 mg of testosterone but decreased (P < 0.05) in gilts treated with 10 mg of testosterone as compared to gilts receiving vehicle (Table 3).
Recovery rates of blastocysts of gilts treated with vehicle or 1 mg of testosterone were higher (P < 0.05) than those of gilts administered 10 or 100 mg of testosterone (Table 3). Blastocyst diameter was not altered by testosterone treatment (Table 3), and all were spherical in morphology. Pregnancy rates of gilts receiving vehicle, 1, or 10 mg of testosterone were 92.3, 100.0, and 92.3%, respectively, and were higher (P = 0.007) than that of gilts receiving 100 mg of testosterone (53.8%). Duration of the estrous cycle of gilts treated with 10 mg of testosterone was shorter (P < 0.05) than those of gilts treated with vehicle or 1 mg of testosterone and was not different from that of gilts treated with 100 mg of testosterone (Table 3).
The results of this experimentation demonstrated for the first time that administration of testosterone during the follicular phase increased the number of preovulatory follicles and resulting CL in swine.
A heterogeneous group of 40 to 50 follicles, 2 to 8 mm in diameter, is present by the end of recruitment (day 16 of the estrous cycle) in swine. The fate of most of these follicles is atresia, and only a small proportion are selected to proceed to ovulation by still unknown mechanisms. Testosterone in swine is utilized as substrate for the synthesis of estradiol by ovarian follicles. The synthesis of estradiol increases steadily as follicles grow and mature to the preovulatory stage. Administration of testosterone on days 17 and 18 in the present study was coincidental with this period of rapid increase in estradiol synthesis. Although this experiment was not designed to examine this phenomenon, it is possible that during this period of high steroidogenic activity and demand for estrogens the supply of androgens may become critically deficient for some follicles.
| Table 3. Reproductive characteristics of gilts treated with vehicle or testosterone.1 | |||||
| Group2 | Number of CL3 | Number of blastocysts4 | Recovery rate (%)4 | Blastocyst size (mm)4,5 | Estrous cycle6 |
| 0 | 14.4a | 12.3 + 1.0a | 84.7 + 4.8a | 5.4 + 0.8 | 20.2a |
| 1 | 16.7b | 15.0 + 1.0c | 89.9 + 4.6a | 5.4 + 0.7 | 20.1a |
| 10 | 17.1b | 8.9 + 1.0b | 52.4 + 4.8b | 4.4 + 1.1 | 19.4b |
| 100 | 17.9b | 9.6 + 1.4ab | 49.6 + 6.3b | 3.2 + 1.4 | 19.7ab |
| 1 Values are means + SEM.
2 Dosages of testosterone (mg/day). 3 SEM = 0.7, n = 13. 4 n = 12, 13, 12, and 7 for the number of pregnant gilts per group receiving 0, 1, 10, and 100 mg of testosterone, respectively. 5 Least squares means + standard error of the least square mean. 6 Length of the estrous cycle (days) during which testosterone was injected on days 17 and 18, SEM = 0.2, n = 13. abc Means without common superscripts within each column are different (P < 0.05) except for number of blastocysts (P < 0.06). | |||||
Other actions of testosterone may include modulation of follicular progesterone synthesis and aromatase activity. Androgen receptors being present in the ovaries of the rat and primates also might be mediators of some of the paracrine actions of androgens. None of these endocrine actions can be excluded at this time and warrant further study. Androgens, however, do not appear to play a direct role on gonadotropin secretion in females.
Unlike the number of CL, embryonic survival and pregnancy rates were affected in a negative manner by the higher doses of testosterone. These effects of testosterone have not been reported. It is unknown whether the decrease in recovery rate of blastocysts was due to a decrease in the rate of fertilization, embryonic mortality prior to day 11, or both. High doses of testosterone propionate produced degenerative changes in oocytes and preantral follicles of estrogen-stimulated hypophysectomized rats. These effects were not evident at the low dosages of testosterone. It also has been reported that testosterone may interfere with the actions of estradiol in the ovary by decreasing the number of available estrogen receptors. Changes in estrus and mating behavior were not detected in the present study; however, gilts that received 10 mg of testosterone had estrous cycles that were approximately 1 day shorter than those of control gilts but still within the range of normal duration. In conclusion, exposure to 1 mg of testosterone during 2 days of the follicular phase increased the number of CL and number of recovered blastocysts in gilts. Dosages of testosterone higher than 10 mg decreased fertility. Administration of a low dose of testosterone during the follicular phase has two potential considerations. First, testosterone might prove to be a mechanism to increase litter size in swine. Second, the novelty of this finding will open new avenues of investigation on the factors that regulate ovulation rate. Many obvious questions remain regarding the timing and mechanisms of these effects of testosterone and are presently under investigation in our laboratory.
SAS. 1988. SAS/STAT User's Guide (Release 6.03). SAS Inst. Inc., Cary, NC.
Steel, R.G.D., and J.H. Torrie. 1980. Principles and Procedures of Statistics: A Biometrical Approach (2nd Ed.). McGraw-Hill Inc., New York.
Wilkinson, L. 1990. SYSTAT: The System for Statistics. SYSTAT, Inc., Evanston, IL.