Genomic heritability and genome-wide association analysis of anti-Müllerian hormone in Holstein dairy heifers.

Anti-Müllerian hormone (AMH) is an ovarian growth factor that plays an important role in regulation of ovarian follicle growth. The objectives of this study were to estimate the genomic heritability of AMH and identify genomic regions associated with AMH production in a genome-wide association (GWA) analysis. Concentrations of AMH were determined in 2,905 dairy Holstein heifers genotyped using the Zoetis medium density panel (Zoetis Inclusions, Kalamazoo, MI) with 54,519 single nucleotide polymorphism (SNP) markers remaining after standard genotype quality control edits. A linear mixed model was used to model the random effects of sampling day and genomics on the logarithm of AMH. The genomic heritability (± standard error of the mean) of AMH was estimated to be 0.36 ± 0.03. Our GWA analysis inferred significant associations between AMH and 11 SNP markers on chromosome 11 and 1 SNP marker on chromosome 20. Annotated genes with significant associations were identified using the Ensembl genome database (version 88) of the cow genome (version UMD 3.1; https://www.ensembl.org/biomart). Gene set enrichment analysis revealed that 2 gene ontology (GO) terms were significantly enriched in the list of candidate genes: G-protein coupled receptor signaling pathway (GO:0007186) and the detection of chemical stimulus involved in sensory perception (GO:0050907). The estimated high heritability and previously established associations between AMH and ovarian follicular reserve, fertility, longevity, and superovulatory response in cattle implies that AMH could be used as a biomarker for genetic improvement of reproductive potential.

Concentration of anti-Müllerian hormone in dairy heifers is positively associated with productive herd life.

Reliable biomarkers predictive of productive herd life (time in herd after birth of first calf) have heretofore not been discovered in dairy cattle. However, circulating concentrations of anti-Müllerian hormone (AMH) are positively associated with number of follicles or antral follicle count (AFC), ovarian function, and fertility, and approximately 25% of cows have a relatively low AFC and low AMH concentrations. The present study tested the hypothesis that heifers with the lowest AMH concentrations have suboptimal fertility and are removed from a herd for poor reproductive performance at a greater rate, and therefore have a shorter productive herd life compared with age-matched herdmates with higher AMH. To test this hypothesis, 11- to 15-mo-old Holstein heifers (n=281) were subjected to a single measurement of AMH. All heifers not removed from the herd had the opportunity to complete 2 lactations and start their third lactation after calving. During this time, performance and health parameters for each individual were recorded daily by herd managers. Results showed that the quartile of heifers with the lowest AMH concentration also had, on average, a shorter productive herd life (by 196 d), a reduced survival rate after birth of the first calf, the lowest level of milk production (first lactation), the lowest total percentage of cows pregnant (across all lactations), the highest culling rates (first and second lactations and overall), and the highest culling rate for poor reproduction (first lactation) compared with age-matched herdmates with higher AMH. We concluded that a single determination of AMH concentration in young adult dairy heifers may be a simple diagnostic method to predict herd longevity, and AMH may be a useful phenotypic marker to improve longevity of dairy cows.

Effects of maternal environment during gestation on ovarian folliculogenesis and consequences for fertility in bovine offspring.

Mammals such as cattle, swine, sheep and humans are born with a highly variable number of ovarian follicles and oocytes in the ovaries that dwindle during ageing and are never replenished. This variation in the ovarian reserve is reflected in the numbers of antral follicles in the ovaries at all ages after birth. As numbers of follicles in ovaries are determined during gestation, the role of maternal nutrition and health during gestation (at time of ovarian development in their foetuses) has been investigated as factors that may impact oogonia proliferation and thus follicle numbers post-natally. These studies have found that both nutrition and health impact numbers of follicles in their offspring. The idea that numbers of follicles and oocytes in ovaries impact fertility is a long-held belief in reproductive biology. This has recently been tested in cattle, and it has been shown that cows with a relatively high number of antral follicles in ovaries have higher pregnancy rates, shorter calving to conception intervals and fewer artificial inseminations during the breeding season compared with cows with a lower number of follicles, and similarly, heifers with many follicles had higher pregnancy rates than those with fewer follicles. Studies summarized in this review highlight the importance of the maternal environment during gestation in determining the size of the ovarian reserve in their offspring and also the contribution of the ovarian reserve to subsequent fertility in cattle.

Does size matter in females? An overview of the impact of the high variation in the ovarian reserve on ovarian function and fertility, utility of anti-Müllerian hormone as a diagnostic marker for fertility and causes of variation in the ovarian reserve in cattle.

The mechanism whereby the inherently high variation in ovary size and the total number of high-quality oocytes in ovaries (ovarian reserve) impact on ovarian function and fertility, diagnostics to measure the size of the ovarian reserve and the factors that cause variation in the ovarian reserve are unknown. Our results show that cattle can be phenotyped reliably based on the number of antral follicles growing during follicular waves (antral follicle count, AFC). Young adult cattle with a consistently low v. a high AFC have smaller gonads, a markedly diminished ovarian reserve and many other phenotypic characteristics usually associated with ovarian aging and infertility. A powerful new approach based on a single measurement of serum concentration of anti-Müllerian hormone (AMH) is described to test the longstanding hypothesis that the size of the ovarian reserve is positively associated with fertility. Also, new evidence shows that maternal environment has a critical role in regulation of the high variation in the ovarian reserve and perhaps fertility in offspring. These results support the conclusion that the inherently high variation in the ovarian reserve, potentially caused by alterations in the maternal environment, has a negative impact on ovarian function that may result in suboptimal fertility in young adult cattle, and a single AMH measurement can be used reliably in future studies to determine if fertility is suboptimal in young adult cattle with low circulating AMH concentrations and a correspondingly diminished ovarian reserve.

Evidence that high variation in antral follicle count during follicular waves is linked to alterations in ovarian androgen production in cattle.

Androgens have an important role in ovarian follicular growth and function, but circulating androgen concentrations are also associated with ovarian dysfunction, cardiovascular disease, and metabolic disorders in women. The extent and causes of the variation in androgen production in individuals, however, are unknown. Because thecal cells of follicles synthesize androstenedione and testosterone, variation in production of these androgens is hypothesized to be directly related to the inherently high variation in number of healthy growing follicles in ovaries of individuals. To test this hypothesis, we determined whether thecal CYP17A1 mRNA (codes for a cytochrome P450 enzyme involved in androgen synthesis), LH-induced thecal androstenedione production, androstenedione concentrations in follicular fluid, and circulating testosterone concentrations were lower in cattle with relatively low versus high number of follicles growing during follicular waves and whether ovariectomy reduced serum testosterone concentrations. Results demonstrated that cattle with a low follicle number had lower (P<0.05) abundance of CYP17A1 mRNA in thecal cells, reduced (P<0.01) capacity of thecal cells to produce androstenedione in response to LH, lower (P<0.01) androstenedione concentrations in ovulatory follicles, and lower (P<0.02) circulating testosterone concentrations during estrous cycles compared with animals with high follicle number. Also, serum testosterone in cattle with low or high follicle number was reduced by 63 and 70%, respectively, following ovariectomy. In conclusion, circulating androgen concentrations are lower in cattle with low versus high number of follicles growing during follicular waves, possibly because of a reduced responsiveness of thecal cells to LH.

Inherent capacity of the pituitary gland to produce gonadotropins is not influenced by the number of ovarian follicles > or = 3 mm in diameter in cattle.

We hypothesised that higher serum FSH concentrations in cattle with low v. high follicle numbers during follicular waves are caused by a different capacity of the pituitary gland to produce gonadotropins. Dairy cows with high (> or = 30; n = 5) and low (< or = 15; n = 5) follicle numbers were selected and serum concentrations of oestradiol and FSH during an oestrous cycle were measured. Cows were ovariectomised at oestrus and bled frequently up to 8 days after ovariectomy. After 33 days, cows were injected with gonadotropin-releasing hormone (GnRH) and bled intensively up to 8 h after GnRH injection. One day later, animals were injected with follicular fluid (FF) from bovine follicles and were bled intensively up to 2 days after the first injection. Serum concentrations of FSH and LH were measured. After 2 days, cows were killed and their pituitary glands collected. Prior to ovariectomy, serum oestradiol concentrations were similar between groups, whereas FSH concentrations were higher in cattle with low v. high numbers of follicles. No differences were detected in serum gonadotropin concentrations after ovariectomy, GnRH injection or FF challenge between groups. The results indicate that the inherent capacity of the pituitary gland to secrete gonadotropins does not differ between cattle with high v. low numbers of follicles during follicular waves.

Development of the "waveless" bovine model.

Development of a bovine model without ovarian follicular waves (waveless) and transient increases in gonadotropin secretion during estrous cycles may lead to new methods to more consistently regulate ovulatory follicle growth thereby improving efficiency of embryo transfer. We hypothesized that the GnRH antagonist acyline would inhibit gonadotropin secretion thereby blocking follicular waves, ovarian function and ovulation during estrous cycles of cattle. To test this hypothesis, beef heifers (n = 5 per group) were treated twice daily with vehicle (control) or 25 or 50 µg/kg acyline beginning 12 h after GnRH-induced ovulation and ending 21 days later. Each animal was subjected to ovarian ultrasonography for 25 days to monitor number and growth of follicles ≥3 mm in diameter and growth of the corpus luteum (CL). Blood samples were taken at various intervals to determine circulating concentrations of FSH, LH, anti-Müllerian hormone (AMH, marker of small follicle growth) and progesterone. Results show that acyline treatment decreased or blocked: circulating concentrations of LH, transient FSH spikes associated with emergence of follicular waves, preovulatory FSH/LH surges, follicular waves, dominant follicle development, CL growth, and progesterone production. In contrast, the largest acyline dose increased AMH concentrations. In conclusion, long-term acyline treatment blocks follicular waves but not growth of preantral and small antral follicles (≤ 3 mm). Future studies will determine if the waveless bovine model, which has enhanced development of preantral and small antral follicles, can be used to develop new methods to improve predictability of response of cattle to superovulation.

Evidence for a negative intrafollicular role for inhibin in regulation of estradiol production by granulosa cells.

Intrafollicular concentrations of inhibin A and estradiol vary inversely during development of dominant follicles in cattle. Thus, we hypothesized that inhibin has a negative autocrine or paracrine effect on estradiol production by granulosa cells. To examine this hypothesis, a homologous model system was used to test the effects of bovine antibovine inhibin antibodies, bovine inhibin, and a peptide fragment of bovine inhibin (bINH) on capacity of granulosa cells isolated from individual estrogen-active or -inactive dominant or subordinate follicles to produce estradiol during short-term (18 h) serum-free culture. Immunoblot analysis of media demonstrated that granulosa cells basally produce different molecular weight forms of inhibin, similar to those in bovine follicular fluid. Immunoneutralization of endogenous inhibin in culture with different doses (12.5-1000 microg) of highly purified bovine antibovine inhibin antibodies increased estradiol production 2- to 15-fold, compared with controls. Preadsorption of the anti-inhibin antibodies with bINH precursors or bovine pro-alpha(C) suppressed the capacity of anti-inhibin antibodies to enhance estradiol production by granulosa cells, compared with controls. Treatment of granulosa cells with an immunoaffinity-purified preparation of bINH suppressed basal estradiol production 60%, compared with controls. In contrast, treatment of granulosa cells with the bINH peptide increased estradiol production 14-fold, compared with controls. Based on these results, we concluded that both anti-inhibin antibodies and bINH blocked the suppressive local effects of basally produced inhibin on estradiol production during culture of granulosa cells and that inhibin has a negative autocrine or paracrine effect on the in vitro capacity of granulosa cells isolated from dominant or subordinate follicles to produce estradiol.

Evidence for autocrine or paracrine roles of alpha2-macroglobulin in regulation of estradiol production by granulosa cells and development of dominant follicles.

alpha(2)-Macroglobulin (alpha(2)-M) inhibits proteinases and modulates the actions of growth factors and cytokines. Despite the key roles proteinases, growth factors, and cytokines have in folliculogenesis, the role of alpha(2)-M in follicular development is unknown. Our objectives were to: 1) determine whether granulosa cells produce alpha(2)-M and have alpha(2)-M receptors, 2) examine the effect of alpha(2)-M on estradiol production by granulosa cells, 3) establish whether amounts of alpha(2)-M and alpha(2)-M receptors were altered during dominant nonovulatory follicle development, and 4) examine alpha(2)-M's mechanism of action. The results demonstrated that bovine granulosa cells contain 5.2- and 15-kb mRNAs and 720- and 500-kDa proteins that correspond, respectively, to sizes of mRNAs and proteins for alpha(2)-M and the alpha(2)-M receptor. Treatment of granulosa cells with alpha(2)-M resulted in a specific dose-responsive increase in estradiol production. Cell viability, cell number, and the amount of aromatase in granulosa cells were not altered by alpha(2)-M. Treatment of granulosa cells with factors that bind alpha(2)-M or its receptor did not mimic alpha(2)-M action. Although intrafollicular amounts of alpha(2)-M remained unchanged, amounts of alpha(2)-M receptor in granulosa cells were strongly inversely associated with concentrations of estradiol in dominant and subordinate follicles. Based on these results, we concluded that alpha(2)-M may have autocrine or paracrine roles in granulosa cells potentially important for regulation of estradiol production and development of dominant follicles.

Causes and consequences of the variation in the number of ovarian follicles in cattle.

In cattle we have noted that the antral follicle count (AFC, follicles > or = 3 mm in diameter) varies greatly among animals (from 5 to 50), is repeatable within animals, and is highly correlated with the total number of healthy follicles in ovaries. Also, animals with low AFC have higher serum concentrations of FSH and LH, but lower concentrations of Anti-Mullerian Hormone, progesterone and androgens than animals with high AFC. We have investigated the effect of maternal environment during gestation on their offspring AFC by restricting maternal nutrition to 60% of maintenance requirements (compared with 100% in controls) during the first third of gestation. Calves born to nutritionally restricted mothers had 60% lower AFC compared with calves born to mothers fed control diets. In other studies we have evidence to indicate that fertility may be compromised in animals with low AFC due to effects on oocytes, progesterone and the endometrium compared with animals with high AFC. To examine this directly we assessed AFC in post-partum dairy cows and found that cows with a high AFC had higher pregnancy rates, shorter calving to conception intervals and received fewer services during the breeding season compared with cows with a low AFC. In addition, the high variation in follicle numbers in adults may not only be reflective of reproductive disorders and suboptimal fertility, but there is evidence to indicate that it may be associated with alterations in the function of other non-reproductive systems (e.g. cardiovascular) that may have profound effects on the animal's health and welfare.

Evidence that high variation in ovarian reserves of healthy young adults has a negative impact on the corpus luteum and endometrium during estrous cycles in cattle.

Low progesterone concentrations and diminished ovarian reserves (total number of healthy oocytes) during reproductive cycles are linked to infertility in single-ovulating species like cattle. However, the extent and mechanisms whereby the inherently high variation in ovarian reserves may negatively affect progesterone production are unknown. Cattle were chosen to address these questions because the size of their ovarian reserves can be predicted based on an antral follicle count (AFC) during follicular waves. The present study determined if progesterone concentrations, differentiation and function of the corpus luteum (CL), and endometrial thickness differed during reproductive cycles of age-matched healthy young adult cattle with low versus high AFC during follicular waves. The results showed that, despite enhanced LH secretion, progesterone concentrations were lower during estrous cycles for animals with low versus high AFC. Animals with low versus high AFC also had a decreased basal, LH-, and 25-hydroxycholesterol-induced capacity of luteal and granulosal cells to produce progesterone, reduced amounts of STAR and mRNAs for STAR and LH receptor in the CL, and no change in endometrial thickness during estrous cycles. Taken together, these results 1) supported the conclusion that high variation in ovarian reserves of young adults is associated with alterations in differentiation and function of the CL and 2) provided insight into the potential factors that may cause suboptimal luteal function (e.g., heightened LH secretion and desensitization of the LH receptor, diminished LH responsiveness, diminished STAR, inherent deficiency in capacity of granulosal cells to undergo luteinization) and infertility (e.g., low progesterone, poor endometrial growth) in individuals with diminished ovarian reserves.

Antral follicle count reliably predicts number of morphologically healthy oocytes and follicles in ovaries of young adult cattle.

Methods to predict numbers of healthy oocytes in the ovaries of young adults could have important diagnostic relevance in family planning and animal agriculture. We have observed that peak antral follicle count (AFC) determined by serial ovarian ultrasonography during follicular waves is very highly reproducible within individual young adult cattle, despite 7-fold variation among animals. Herein, we tested the hypothesis that AFC is positively associated with the number of morphologically healthy oocytes and follicles in ovaries and with serum concentrations of anti-Müllerian hormone (AMH), an indirect marker for number of healthy follicles and oocytes in ovaries. In the present study, age-matched young adult cattle (12-18 mo old) were subjected to serial ultrasonography to identify animals with a consistently high (> or =25 follicles that were > or =3 mm in diameter) or low (< or =15 follicles) AFC during follicular waves. Differences in serum AMH concentrations, ovary weight, and number of morphologically healthy and atretic follicles and oocytes were determined. The phenotypic classifications of cattle based on AFC during follicular waves or AMH concentrations both predict reliably the relative number of morphologically healthy follicles and oocytes in ovaries of age-matched young adult cattle.

Follicle numbers are highly repeatable within individual animals but are inversely correlated with FSH concentrations and the proportion of good-quality embryos after ovarian stimulation in cattle.

BACKGROUND: The significance of the high variation in numbers of follicles produced during reproductive cycles in humans and cattle is unknown. METHODS: We selected beef heifers with high (> or =25) or low (< or =15) numbers of ovarian follicles and determined the association with alterations in FSH and estradiol concentrations, as well as responsiveness to superstimulation and embryo quality. The variation in follicle numbers was also compared with oocyte quality in natural cycles using IVF and abattoir-sourced bovine ovaries. RESULTS: Results show that: (i) FSH was lower (P < 0.03) in animals with high compared with low follicle numbers per follicle wave; (ii) after superovulation, in the high versus low follicle number group, the number of oocytes/embryos recovered after insemination (10.6 +/- 2.7 versus 4.7 +/- 0.7) and the number of transferable embryos (5.4 +/- 1.3 versus 3.8 +/- 0.8) per animal were greater (P < 0.05), whereas the proportion of transferable embryos (50.7% versus 79.8%) was lower (P < 0.05); (iii) in unstimulated animals, the numbers of high-quality oocytes harvested and in-vitro fertilized oocytes developing into blastocysts were up to 4-fold greater (P < 0.05) for ovaries with high versus low numbers of follicles, but the proportions of oocytes developing into blastocysts were similar in the two groups. CONCLUSION: Phenotypic classification based on numbers of follicles may be useful to improve superovulation procedures. The lower proportion of transferable embryos following superovulation of ovaries with high numbers of follicles is probably not the result of differences in the quality of oocytes before superovulation.

Molecular evidence that growth of dominant follicles involves a reduction in follicle-stimulating hormone dependence and an increase in luteinizing hormone dependence in cattle.

The bovine dominant follicle (DF) model was used to identify molecular mechanisms potentially involved in initial growth of DF during the low FSH milieu of ovarian follicular waves. Follicular fluid and RNA from granulosa and theca cells were harvested from 10 individual DF obtained between 2 and 5.5 days after emergence of the first follicular wave of the estrous cycle. Follicular fluid was subjected to RIA to determine estradiol (E) and progesterone (P) concentrations and RNA to cDNA microarray analysis and (or) quantitative real-time PCR. Results showed that DF growth was associated with a decrease in intrafollicular E:P ratio and in mRNA for the FSH receptor, estrogen receptor 2 (ER beta), inhibin alpha, activin A receptor type I, and a proliferation (cyclin D2) and two proapoptotic factors (apoptosis regulatory protein Siva, Fas [TNFRSF6]-associated via death domain) in granulosa cells. In contrast, mRNAs for the LH receptor in granulosa cells and for two antiapoptotic factors (TGFB1-induced antiapoptotic factor 1, LAG1 longevity assurance homolog 4 [Saccharomyces cerevisiae]) and one proapoptotic factor (tumor necrosis factor [ligand] superfamily, member 8) were increased in theca cells. We conclude that the bovine DF provides a unique model to identify novel genes potentially involved in survival and apoptosis of follicular cells and, importantly, to determine the FSH-, estradiol-, and LH-target genes regulating its growth and function. Results provide new molecular evidence for the hypothesis that DF experience a reduction in FSH dependence but acquire increased LH dependence as they grow during the low FSH milieu of follicular waves.

Effects of oestradiol and progesterone on secretion of gonadotrophins and health of first wave follicles during the oestrous cycle of beef heifers.

Antral follicle development in cattle is initially FSH dependent and then LH dependent. The aim of the present study was to determine the effects of oestradiol- and progesterone-induced suppression of FSH and LH on growth and differentiation of first wave follicles. Cyclic heifers (n = 45, n = 6-10 per group) received the following i.m. injections or treatments beginning 30 h after oestrus: (i) saline (controls); (ii) 0.75 mg oestradiol benzoate (ODB); (iii) insertion of a progesterone-releasing intravaginal device (PRID) for 42 h (progesterone); (iv) 0.75 mg oestradiol benzoate plus PRID (ODB plus progesterone); (v) 0.75 mg ODB plus injection of 1 mg Ovagen(TM) at 33, 39 and 45 h after onset of oestrus (ODB plus FSH). In Expt 1, follicle development was monitored by ovarian ultrasonography once a day. In Expt 2, heifers were ovariectomized. Emergence of the first follicle wave and dominant follicle selection were delayed in ODB plus progesterone-treated heifers compared with controls. Interval to nadir FSH concentration was shorter in ODB-, progesterone- and ODB plus progesterone-treated heifers compared with controls. Frequency of LH pulses was unaffected in ODB- or ODB plus FSH-treated heifers, decreased in progesterone-treated heifers and further decreased in ODB plus progesterone-treated heifers. Intrafollicular oestradiol concentrations were lower in the largest follicle from ODB plus progesterone-treated heifers compared with control (66 h) heifers, but follicle diameter and concentrations of insulin-like growth factor binding proteins (IGFBPs) and inhibin forms were unaffected. Treatment with ODB decreased follicular oestradiol concentration in smaller follicles in the cohort. It is concluded that growing cohort follicles are uniformly responsive to increased FSH concentration but differentially responsive to suppressed FSH and LH release, which is consistent with an LH-mediated survival advantage of the largest follicle in the cohort before cessation of the growth of remaining follicles in the cohort occurs.

Identification of genes involved in apoptosis and dominant follicle development during follicular waves in cattle.

We hypothesize that granulosa and theca cells from growing dominant follicles, with relatively high intrafollicular concentrations of estradiol, have a greater expression of genes involved in inhibiting apoptosis pathways and lower expression of genes involved in apoptosis pathways than growing subordinate follicles with lower estradiol concentrations. Using the well-characterized bovine dominant follicle model, we collected granulosa and theca cells from individual dominant and the largest subordinate follicle 3 days after initiation of a follicular wave in four animals. Based on ultrasound analysis, both follicle types were in the growth phase at the time of ovariectomy. However, dominant follicles were larger (9.8 +/- 1.0 versus 7.6 +/- 0.6 mm in diameter, P < 0.05) and had greater intrafollicular concentrations of estradiol (132.2 +/-3 8.5 versus 24.1 +/- 12.1 ng/ml, P < 0.05), compared with the largest subordinate follicles. We used bovine cDNA microarrays, which contained a total of 1400 genes, including a subset of 53 genes known to be involved in apoptosis pathways, to determine which apoptosis and marker genes from each of the four dominant versus subordinate follicles were potentially differentially expressed. Using a low stringency-screening criterion, 22 genes were identified. Quantitative real-time polymerase chain reaction confirmed that 16 of these genes were differentially expressed. Our novel results demonstrate that the high intrafollicular concentrations of estradiol in growing dominant follicles were positively associated with enhanced expression of mRNAs in granulosa cells for aromatase, LH receptor, estradiol receptor beta, DICE-1, and MCL-1, compared with granulosa cells from subordinate follicles (all survival-associated genes). In contrast, the relatively low intrafollicular concentrations of estradiol in growing subordinate follicles were positively associated with enhanced expression of mRNAs in granulosa cells for beta glycan, cyclo-oxygenase-1, tumor necrosis factor alpha, caspase-activated DNase, and DRAK-2, and in theca cells for beta glycan, caspase 13, P58(IPK), Apaf-1, BTG-3, and TS-BCLL, compared with granulosa or theca cells from dominant follicles (genes that are all associated with cell death and/or apoptosis). We suggest that that these genes may be candidate estradiol target genes and that they may be early markers for the final stages of follicle differentiation or initiation of apoptosis and thus selection of dominant follicles during follicular waves.