Different types of maternal nutritional status have been identified to program adult diseases [
2]. And ample evidences demonstrated that maternal exposure to HFD during gestation and lactation has harmful effects on the offspring [
20,
21]. In the present study, we focused on the influence of maternal exposure to HFD during gestation only to mimic the excessive gestational weight gain and investigated reproductive potential of the female offspring at PND4 and PND30. We found that early follicular development at neonatal and prepuberty follicular development in the offspring was impaired by maternal HFD during gestation. The exact mechanism remains unclear, however, it appears that kisspeptin/GPR54 system may be involved.
Effects of HFD during gestation on body weight
The developing organism has inherent developmental plasticity which is capable of adapting to various environments, including unfavorable environment in uterus, and the response will ultimately determine the adult phenotype [
2]. In the current study, we observed that birth weight of female offspring was not different between the two groups which proved intrauterine adversity might not affect the intrauterine growth (Fig.
1 a). At lactation, all of the offspring in two groups showed significant weight gain, performed lively and the surface hair was thick and shiny which indicated that sufficient nutrition had been supplied for pups. While the rats subsequently grew up to PND70, the HFD group showed accelerated growth but did not reach significant statistical difference (0.05 <
p < 0.1 at PND 56, 63, 70,
n = 3). The results of body weight were controversial, as some pups born to maternal HFD were lighter at birthweight and displayed accelerated growth [
4] while some studies were consistent with our results [
22]. Although there was no obvious difference in body weight, it is clear that intrauterine factors play an important role in reproductive system as there are changes in phenotypes found in the other studies [
2].
HFD during gestation impairs the initial recruitment of follicles and decreases the progression from primary to secondary follicles
At PND4 in rats, oocytes become completely surrounded by a single layer of flattened granulosa cells forming primordial follicles and some primordial follicles begin to differentiate into primary follicles and subsequently into secondary follicles [
23]. Neonates born to HFD mothers showed comparable numbers of oocytes with NCD group, indicating that the initial pool of primordial follicles was not reduced. The ratio of primordial follicles, primary follicles and atretic follicles in HFD group displayed no statistical significance compared with control group. However, independent of the number of primordial follicles, the decreased ratio of secondary follicles in HFD group indicated that there was a reduction in the recruitment of follicles (Fig.
2 g). We hypothesized that this change might be accompanied by changes in the expression of kisspeptin and FSHR. We found that kisspeptin was mainly expressed in oocyte and FSHR was in granulosa cells. However, the results showed there was no significant difference between NCD group and HFD group in the staining intensity (Fig.
2 h). Fernandois et al. used the mini-osmotic pump containing kp-10 to achieve partial infusion of the ovary in 5 or 9 month old rats for 28 days and evaluated the long-term effect of kisspeptin on ovarian follicular development [
24]. They postulated that kisspeptin could down-regulating FSHR in ovary and up-regulating circulating AMH to attenuate the initial follicle recruitment from primary follicles to secondary follicles. On the contrary, Tsoulis et al. indicated that rats born from mothers of HFD during pregnancy had more primordial follicles converted into growing follicle with considerable down-regulation of AMH and its receptor, AMHRII [
20]. For the discrepancy, rats from different background (SD rats vs Wistar rats) and percentage of fat in food (60% vs 45%) might be responsible.
HFD during gestation accelerating growth of follicles may cause advanced puberty and estrous cyclicity irregular in the offspring
At PND30, the average ovaries quotiety increased significantly in the HFD group when compared with the control group (Fig.
1 b), and this may be attributed to the increasing antral and preovulatory follicles with less amount of secondary follicles (Fig.
3 i). Increased follicular development was accompanied by increased expression of follicular growth-related genes (Fig.
3 k). The type of antral follicles and preovulatory follicles is the primary source for the increasing levels of estrogen before puberty [
10]. Consistently, we observed that HFD group increased the expression of estrogen synthesis related genes (Fig.
3 m). We also found the expression of ovulation-related genes was down-regulated (Fig.
3 l), suggesting there might be an ovulation dysfunction in HFD group. However, we cannot determine the effect of HFD during gestation on ovulation from morphology, because there was no ovulation (corpora luteum) in prepuberty. This phenomenon could be the result of a compensatory response to the chronic HFD condition during gestation, but the driving factor or internal mechanism in ovary is still unknown.
Kisspeptin/GPR54 were expressed in oocytes, granulosa cells and theca cells (Fig.
3 e-h). In the HFD group, increased expression of kisspeptin in granulosa cells and theca cells might be responsible for altered gene expression of steroidogenesis synthesis related genes. A report documented that kisspeptin could stimulate progesterone secretion directly with no effect on the secretion of estrogen in rat luteal cells [
25]. On the other hand, we also demonstrated that kp-10 could increase the viability of primary granulosa cells (Fig.
6 e). Regarding oocytes, we demonstrated kp-10 could directly enlarged the size of oocytes in vitro (Fig.
6 f). And in some other species, administration of kisspeptin could also promote the maturation of oocytes in the cumulus-oophorus complex [
26,
27]. However, in vitro culture system, we cannot separate the interaction functions of various cells as granulosa cells play an important role in the development of oocytes. Since both granulosa cells and oocytes could express kisspeptin/GPR54 in rats, it is easily confused whether the kisspeptin-induced oocyte maturation is mediated by the granulosa cells or the oocyte itself. Fernandois et al. observed a phenomenon that ovaries infused with a low dose of kisspeptin at 5 or 9 month old rats had an increased number of preovulatory follicles and corpora lutea with a fewer number of antral follicles [
24]. In gene inactivated models, both
kiss1−/− and
kiss1r−/− mice displayed significantly reduced ovarian weight and size, which might be resulted from the absence of large follicles [
28,
29]. Furthermore, the
kiss1r haploinsufficient (
kiss1r+/−) mice exhibited progressively decreased number of preantral follicles after puberty with no significant differences observed before puberty compared with wild-type [
30]. Interestingly, the gonadotropin levels were not significantly different between the wild-type and
kiss1r+/− mice as demonstrated in another study [
31].
It has been demonstrated that the central kisspeptin/GPR54 system is an essential gatekeeper of puberty onset [
32‐
36]. The hypothalamic
kiss1 neurons were involved in mediating the positive feedback and negative feedback effects of estrogen [
37]. The increasing level of estrogen before puberty produces the positive feedback on central kisspeptin and luteinizing hormone (LH), and leads to puberty onset [
10]. In the present study, we found that the puberty onset was earlier in female offspring of dam fed HFD through pregnancy and our result was consistent with previous studies [
38]. We can reasonably assume that the role of ovarian kisspeptin in puberty onset might be through promoting follicular development along with increasing the levels of estrogen. A study reported that direct ovarian infusion with p234 (kisspeptin antagonist) in prepubertal rats resulted in delayed vaginal opening, thus indicating that ovarian kisspeptin could affect ovarian function and participate in the onset of puberty [
39]. However, we cannot discard that the hypothalamus can be affected directly by the maternal HFD during gestation. Unfortunately, we did not detect the expression of
kiss1 mRNA in hypothalamus. In addition to advanced puberty onset, maternal HFD during gestation also disrupted estrous cyclicity in female offspring. Particularly, the pups were more likely to display estrous cycles characterized by prolonged proestrous or estrous (Fig.
4 a-b) and Connor et al. suggested that prolonged estrous may be associated with premature ovarian failure [
4]. This change might be owing to accelerated follicle development accompanied by increased estrogen or abnormal development in the reproductive axis. Whether maternal HFD during gestation could contribute to premature ovarian failure in adulthood is uncertain, and the phenotype of adulthood needs further investigation.
Perinatal environment, including lactation, is also a very important programming window considered to play a critical role in the development of the offspring [
1]. Though the breast milk is mainly affected by the lactation diet, it is inevitable that gestational diet would also have certain influence on the composition [
40]. Therefore, it is important to study the composition of nutrients and hormones in breast milk, which was what we neglected when designing the experiment. Although there are some limitations in the present study, we believe that these will not change our main findings and speculations.