Introduction

The objective of superovulation in cattle is to maximize the number of fertilized and transferable embryos with a high probability of producing pregnancies.¹ However, wide ranges in superovulatory responses and embryo yield have been reported in reviews of commercial embryo transfer records. In 2048 beef donor collections, a mean of 11.5 ova/embryos with 6.2 transferable embryos was recorded.² Variability was great, both in the superovulatory response and embryo quality; 24% of the collections did not yield viable embryos, 64% produced fewer than average numbers of transferable embryos, and 30% yielded 70% of the embryos. Embryo recovery from 987 dairy cows yielded slightly fewer ova/embryos but there was similar variability.³ The high degree of unpredictability in superovulatory response creates problems that affect both the efficiency and profitability of embryo transfer programs.⁴

Variability in ovarian response has been related to differences in superstimulatory treatments, such as gonadotropin preparation, batch and total dose of gonadotropins, duration and timing of treatments, and the use of additional hormones. Additional factors, which may be more important, are inherent to the animal and its environment. These may include nutritional status, reproductive history, age, season, breed, effects of repeated superstimulation, and ovarian status at the time of treatment. While considerable progress has been made in the understanding of bovine reproductive physiology, factors inherent to the donor animal are only partially understood.

Gonadotropins and superovulation

Factors affecting superovulatory response associated with the administration of exogenous gonadotropins include source, batch, and biological activity.⁵ Three different types of gonadotropin preparations have been used to induce superovulation in the cow: gonadotropins from extracts of porcine or other domestic animal pituitaries, equine chorionic gonadotropin (eCG), and human menopausal gonadotropin.^5,6 Human menopausal gonadotropin did not offer any advantages in cattle⁷ and has not been used commercially; therefore it will not be discussed further.

Pituitary extracts contain follicle-stimulating hormone (FSH). The biological half-life of FSH in the cow has been estimated to be 5 hours or less,⁸ so it must be injected twice daily to successfully induce superovulation.^9,10 The usual treatment regimen is twice-daily intramuscular treatments with FSH for 4 or 5 days, with a total dose of 28–50 mg (Armour) of a crude pituitary extract, or 260–400 mg NIH-FSH-PI of a partially purified pituitary extract (Folltropin®-V, Bioniche Animal Health Inc., Belleville, Ontario, Canada). At 48 or 72 hours after initiation of treatment, prostaglandin (PG)F_2α is administered to induce luteolysis. Estrus occurs in 36–48 hours, with ovulations beginning 24–36 hours later.^11,12

Equine chorionic gonadotropin is a complex glycoprotein with both FSH and luteinizing hormone (LH) activity.¹³ It has been shown to have a half-life of 40 hours in the cow and persists for up to 10 days in the bovine circulation; thus it is normally injected intramuscularly once followed by a PGF_2α injection 48 hours later.¹⁴ The long half-life of eCG causes continued ovarian stimulation, unovulated follicles, abnormal endocrine profiles, and reduced embryo quality.^15–18 These problems have been largely overcome by the intravenous injection of antibodies to eCG at the time of the first insemination, 12–18 hours after the onset of estrus.^14,19 Recommended doses of eCG range from 1500 to 3000 IU, with 2500 IU by intramuscular injection commonly chosen.

Monniaux et al.¹⁰ treated groups of cows with either 2500 IU eCG or 50 mg (Armour) pituitary FSH and observed that ovulation rate and the percentage of cows with more than three transferable embryos was slightly higher with FSH than eCG. Although these results were in agreement with those of Elsden et al.,²⁰ others have found no differences between pituitary extracts containing FSH and eCG.^21,22 Endocrine studies have revealed that eCG-treated animals more frequently had abnormal profiles of LH and progesterone,^15,23 which were associated with reductions in both ovulation and fertilization rates²⁴ as compared with FSH-treated cows.

Although folliculogenesis in mammals requires both FSH and LH, there is considerable variability in FSH and LH content of crude gonadotropin preparations.⁵ Radioreceptor assays and in vitro bioassays have revealed variability in both the FSH and LH activity of eCG, not only among pregnant mares but also within the same mare at different times during gestation.¹³ The effects of the FSH/LH ratio of eCG on superovulatory responses have been examined and there was a positive correlation between the ratio of FSH/LH activity and superovulatory response. Lower ratios of FSH/LH activity reduced ovulatory response in immature rats and LH added to eCG reduced superovulatory response in cows.^5,13

Purified pituitary extracts with low LH contamination have been reported to improve superovulatory response in cattle. Chupin et al.²⁵ superstimulated three groups of dairy cows with an equivalent amount of 450 μg pure porcine (p)FSH and varying amounts of LH and showed that the mean ovulation rate and the number of total and transferable embryos increased as the dose of LH decreased. It has been suggested that high levels of LH during superstimulation cause premature activation of the oocyte.¹⁶ Several experiments with an LH-reduced pituitary extract²⁶ utilizing several different total doses, ranging from 100 to 900 mg of NIH-FSH-PI, revealed no evidence of detrimental effects of dose on embryo quality.^6,27 On the other hand, doubling the dose of crude pituitary extracts containing both FSH and LH resulted in significantly reduced fertilization rates and percentages of transferable embryos.⁶ Collectively, data support the hypothesis that the detrimental effects of high doses of pituitary gonadotropins on ova/embryo quality is due to an excess of LH.

Although it is generally believed that some LH is required for successful superovulation, endogenous LH levels may be adequate. Looney et al.²⁸ reported that recombinantly produced bovine (b)FSH induced high superovulatory responses without the addition of exogenous LH. These data suggest that LH is not needed in superstimulatory preparations and that embryo quality may be superior with pure FSH. The very high fertilization rates and transferable embryo rates in the absence of exogenous LH suggest that administration of LH in superstimulation protocols, at any dose, may be detrimental to embryo quality. Further, these results and more recent results with recombinant bFSH²⁹ indicate that future progress lies in the use of recombinantly derived gonadotropins.

The effects of LH on superovulatory response has also been demonstrated in an experiment involving Brahman-cross (Bos indicus) heifers superstimulated with 400 mg NIH-FSH-PI containing 100%, 16% or 2% LH.³⁰ Although the more purified preparations in this experiment caused higher superovulatory responses, there were obvious seasonal effects; responses with the most purified and intermediate preparations were superior to the least purified preparation during summer months, but only the most purified preparation was highly efficacious during winter months. These results would appear to contradict the findings of Page et al.³¹ who reported that superovulatory response and embryo quality in Holstein heifers was not affected by LH levels in cool weather, but that during heat stress a more purified preparation yielded more corpora lutea and significantly more fertilized ova and transferable embryos. It becomes apparent that stress is the common factor; Bos taurus breeds likely find summer heat stressful, whereas Bos indicus breeds likely find winter temperatures stressful. In either case, the more purified extracts resulted in greater superovulatory responses during conditions of environmental stress.