Effects of Maternal Environment During Gestation on Ovarian Folliculogenesis and Consequences for Fertility in Bovine Offspring

Follicle formation during foetal development

In ruminants, follicle formation occurs during foetal life, starting with migration of primordial germ cells to the genital ridge followed by differentiation into oogonia (Pepling 2006), which is completed by the first or second month of foetal life (Smitz and Cortvrindt 2002; Binelli and Murphy 2010). After arrival of the primordial germ cells into the gonadal ridge, the germ cells continue to proliferate and differentiate into primordial follicles, comprised of an oocyte surrounded by squamous granulosa cells (primordial follicles are arrested in meiosis) (Webb and Campbell 2007). In cattle, differentiation of primordial follicles starts between 90 and 140 days of foetal life (Erickson 1966b; Russe 1983; Tanaka et al. 2001; Yang and Fortune 2008). In addition, it is well established that the population of oogonia in the developing bovine ovary reaches an estimated maximum number of 2.1 million (Smitz and Cortvrindt 2002), which rapidly declines during mid- to late-foetal life owing to apoptosis until the population of follicles is reduced at birth to approximately 130 000 with a high variation among individuals (Erickson 1966b; Aerts and Bols 2010).

David Barker [hence the Barker hypothesis (Barker 1992) or more recently the thrifty phenotype hypothesis (Hales and Barker 2001)] suggests that environmental influences early in human foetal life are reflected in impaired growth, development and metabolism, leading to increased risk for diseases in adulthood. The hypothesis proposes that some diseases originate through foetal adaptations to malnutrition that permanently alter body function. This hypothesis is supported by animal models (McMillen and Robinson 2005). Most studies have examined the link between malnutrition with cardiovascular disease, obesity and diabetes (McMillen and Robinson 2005; Boo and Harding 2006), but only a few studies have examined the impact of maternal environment on reproduction. Hence, it is reasonable to speculate that the maternal environment during gestation, at the time of ovarian development in their foetuses, may impact oogonia proliferation and thus follicle numbers post-natally.

Evidence that maternal disease during gestation negatively impacts AFC in their offspring

We have recently examined the effect of disease during pregnancy in cattle on ovarian development and function of their offspring. Somatic cells normally present in milk of dairy cows are predominantly white blood cells. When somatic cells are counted, results are referred to as the somatic cell count (SCC). The major factor affecting SCC in milk is a mammary gland infection. Somatic cell count in milk samples collected from individual dairy cows at various intervals (e.g. monthly) during lactation is commercially determined and recorded in dairy herds. Consequently, high SCC in milk (e.g., >200 000) is useful as an index for potential previous or current udder infections/inflammation that may have been caused by mastitis or injury (Caraviello et al. 2005). In a preliminary study, we have recently shown that dairy cows with a high number of SCC measurements >200 000 during pregnancy produce daughters with relatively low circulating concentrations of anti-Müllerian hormone (AMH) as adults. Anti-Müllerian hormone is a glycoprotein produced exclusively by granulosa cells of healthy growing follicles (La Marca and Volpe 2006), and circulating AMH concentrations are not only static during reproductive cycles, but also highly correlated with AFC and size of the ovarian reserve in cattle and other species (see review Ireland et al. 2010). In a recent follow-up study with larger animal numbers, we subjected two groups of 12-month-old Holstein heifers to two injections of prostaglandin F_2α (PG) spaced 11 days apart to synchronize oestrous cycles. A single blood sample was removed from the tail vein of each animal 96 h after the last PG injection, and serum AMH concentration was determined using our previously validated AMH assay (intra- and interassay cv ≤ 10%; Ireland et al. 2008). One group of heifers were from dams (n = 25 animals) that had 4 or 5 SCC measurements >200 000 during pregnancy, while the other group of heifers were from dams (n = 20) with 0 or 1 SCC measurement >200 000 during pregnancy. Like our previous preliminary study (Ireland et al. 2010), results showed that the heifers from dams with chronically high SCC had AMH concentrations nearly 50% lower (0.13 ± 0.03 vs 0.075 ± 0.01 ng/ml, p < 0.05) compared with heifers from dams with zero or a low SCC (Fig. 3). Taken together, these results imply that chronic infections during pregnancy of dairy cows diminish the size of the ovarian reserve (and lower AMH) and potentially future reproductive performance in their offspring.

Evidence that maternal nutrition during gestation negatively impacts AFC in their offspring

Most studies examining the thrifty phenotype hypothesis investigate the consequences of nutritional challenges imposed on the dam on health of their offspring as young adults. However, we tested the hypothesis that restriction of maternal nutrition during gestation, at the time of ovarian development in their foetuses, negatively impacts proliferation of primordial germ cells (to be consistent with Introduction) and thus follicle numbers post-natally. We tested this hypothesis by restricting nutrition of cows to 0.6 of their maintenance energy requirements shortly before conception to the end of the first trimester of pregnancy. The first trimester of pregnancy coincides with the peak in number of follicles and oocytes in foetal ovaries (Erickson 1966a). Results show that, although weight of offspring was unaltered at birth, AFC and circulating AMH concentrations in calves born to nutritionally restricted mothers were 60% lower compared with calves born to control mothers (Mossa et al. 2009) (Table 1). Because AFC and AMH concentrations are positively correlated with size of the ovarian reserve (Ireland et al. 2008), these findings imply that maternal nutrition as well as chronic infections during pregnancy may have a previously unrecognized yet important role in regulation of size of the ovarian reserve and perhaps fertility of cattle.

Effects of genotype on AFC

Another intriguing observation of the numbers of follicles in cattle is a comparison of Bos indicis vs Bos taurus breeds of cattle. While there are small differences in the characteristics of oestrous cycles and follicle waves between these types of cattle (Sartori et al. 2010), it is clear that there are different number of follicles in follicle waves. At the onset of each follicular wave, there are on average 18 (Ireland et al. 2007) to 24 (Ginther et al. 1996) small antral follicles detected in the ovaries of B. taurus cattle, whereas there appears to be at least double this in B. indicus cattle (Gimenes et al. 2009), with reports of average counts ranging from 40 (Alvarez et al. 2000) to 50 (Buratini et al. 2000; Sartori et al. 2010) follicles.

While these differences in antral follicle numbers between breeds have been suggested to be due to differences in circulating concentrations of insulin and IGF-I (Sartori et al. 2010; Sartori and Barros 2011), it is also possible that they arise owing to differences in the maternal environment during gestation affecting numbers of follicles in the ovarian reserve. In addition, it is interesting to note that while B. indicus cattle have more follicles than B. Taurus breeds, there is some evidence to support the notion that prenatal maternal nutrition also negatively impacts the ovarian reserve in offspring of B. indicus breeds of cattle (Sullivan et al. 2009), as shown in our studies (Mossa et al. 2009).

AFC and superovulation

Responsiveness to superovulation and reduced success of IVF and embryo transfer are well established to be negatively associated with ageing and to correspond to a reduction in number of follicles and oocytes in ovaries in single-ovulating species such as cattle (Kawamata 1994; Cushman et al. 1999; Taneja et al. 2000; Singh et al. 2004) and humans (Beckers et al. 2002; Styer and Toth 2011). Our results show that young adult cattle with low vs high AFC respond poorly to superovulation and have a reduced number of transferable embryos (Ireland et al. 2007).

AFC, progesterone concentrations and endometrial growth

Progesterone is a key physiological marker of corpus luteum function and thus ovarian function, which also has a well-established positive role in uterine function, embryo development and fertility in single-ovulating species such as cattle (Diskin and Morris 2008). In support of the crucial role for progesterone in successful reproduction, low circulating progesterone concentrations are associated with high rates of embryo mortality in cattle (Diskin and Morris 2008) and women (Daily et al. 1994; Elson et al. 2003) and luteal phase defects and infertility in women (Pritts and Atwood 2002). Therefore, we asked whether the high variation in AFC was associated with CL function and circulating progesterone concentrations during oestrous cycles. The combined results for 63 different oestrous cycles for 47 animals illustrated that young adult beef cattle and late lactation dairy cows with a low AFC had progesterone concentrations consistently 30–50% lower than cattle with a high AFC and also that this was reproducible from one oestrous cycle to the next (Jimenez-Krassel et al. 2009). Whether the difference in circulating progesterone concentrations observed for heifers with a low vs a high AFC also impacted uterine development was then examined, and it was found that young adult cattle with a low AFC had chronically much lower circulating progesterone concentrations throughout the early to mid-luteal phase and poor endometrial development during the early luteal phase of oestrous cycles compared with cattle with a high AFC (Jimenez-Krassel et al. 2009). These observations, coupled with many reports of the association between low serum progesterone concentrations with embryo mortality in cattle (Diskin and Morris 2008) and between reduced endometrial thickness and infertility in women (Raga et al. 1999; Basir et al. 2002), imply that rates of embryo mortality may be higher in young adult cattle with a low vs a high AFC and correspondingly reduced quantity of follicles and oocytes in ovaries.

AFC and intrafollicular markers of oocyte quality

We have also investigated markers of follicle, cumulus and oocyte cell function from animals with high vs low AFC and conclude that the high variation in follicle and oocyte numbers is also associated with physiologically significant alterations in follicular differentiation and in turn oocyte quality (Ireland et al. 2009). In particular, intrafollicular estradiol concentrations are approximately twofold higher in animals with a low vs a high AFC (Ireland et al. 2009), and this may have detrimental effects on oocyte maturation and developmental competence in cattle with low follicle numbers because of high physiological concentrations of estradiol block maturation of bovine oocytes in vitro (Beker-van Woudenberg et al. 2004). In addition, we have observed that young adult cattle with a low vs a high AFC have much higher expression of cathepsin mRNA in cumulus cells (Ireland et al. 2009), which is a marker for poor oocyte quality in cattle (Bettegowda et al. 2008).