Fertilization represents the union of two terminally differentiated gametes to form a single embryo capable of developing into a unique individual. Gamete fusion is only the beginning of the process of embryo development, but it sets in motion a series of events collectively termed “egg activation” that turn the unified gametes into a totipotent embryo capable of becoming a healthy individual. The key event of egg activation that triggers development in all species is a rise in the cytoplasmic calcium level (Kashir et al., 2013). What is unusual about mammalian eggs is that following the initial calcium rise, they go on to have a series of oscillations in cytoplasmic calcium levels that persist for several hours after fertilization. These calcium oscillations drive downstream events of egg activation including exocytosis of cortical granules, activation of calcium/calmodulin-dependent protein kinase II-gamma, resumption of the cell cycle, and pronucleus formation, all of which are essential for initiating proper embryo development (Stein et al., 2020). Persistent calcium oscillations can be triggered in the absence of sperm using the divalent cation strontium, resulting in parthenogenetic egg activation mediated by the transient receptor potential channel TRPV3 (Fraser, 1987; Carvacho et al., 2013). In the mouse, an inappropriate pattern of calcium oscillations following fertilization is associated with reductions in implantation efficiency, reduced development to term, and abnormalities in offspring growth (Ozil and Huneau, 2001; Ducibella et al., 2002; Ozil et al., 2005, 2006).

The pattern of calcium oscillations at fertilization is modulated by the amount of PLCζ released by the sperm but also by many factors intrinsic to the fertilized egg (Stein et al., 2020). Calcium oscillations depend on the amount of calcium in endoplasmic reticulum (ER) stores as well as egg factors that regulate how quickly calcium is released from the ER, cleared from the cytoplasm, and then pumped back into the ER such that sufficient calcium stores are available for the next calcium release event. Calcium homeostasis is regulated by the activities of calcium pumps and ion channels (Berridge et al., 2003). Sarco-endoplasmic reticulum calcium-ATPases pump calcium back into the ER, and plasma membrane calcium-ATPase pumps clear calcium from the cytoplasm by extruding it across the plasma membrane. These pumps depend on mitochondrial production of ATP. Calcium influx channels support calcium entry into the cytoplasm down a concentration gradient from the extracellular milieu; calcium influx is necessary for complete refilling of ER stores and continuation of calcium oscillations (Igusa and Miyazaki, 1983; Kline and Kline, 1992). Together, these activities allow the egg to tolerate the release of large amounts of calcium from the ER and to refill ER stores once the calcium release event has been completed.

All the molecular components needed to support calcium homeostasis, egg activation, and preimplantation embryo development, such as stored proteins, nucleic acids, and energy substrates, are generated during the oocyte growth phase. In a normal estrous cycle, ovarian follicles are recruited into the growing pool independent of gonadotropin hormones until the secondary follicle stage, when the oocyte is surrounded by several layers of granulosa cells (Strauss and Williams, 2017). Follicle stimulating hormone (FSH), secreted from the pituitary, initiates further follicle development to the antral stage, when fluid begins to collect between the granulosa cells. FSH and luteinizing hormone (LH) together promote continued survival and development of antral follicles toward the preovulatory stage (Chun et al., 1996; Sullivan et al., 1999). When FSH levels decline in response to estradiol-mediated negative feedback on the pituitary, antral follicles instead will undergo atresia. A select few “dominant” follicles continue to grow despite declining FSH levels. These follicles are either more sensitive to the available FSH or have developed sufficient responsiveness to LH to continue their developmental trajectory (Sullivan et al., 1999). As a result, a species-specific number of preovulatory follicles forms during each estrous cycle.

Successful superovulation was first reported almost 100 years ago, when whole pituitary tissue from donor animals was injected intramuscularly into rats and mice and resulted in either ovaries containing very large numbers of normal-appearing follicles or oviducts containing large numbers of ovulated eggs (Smith and Engle, 1927). Although the exact procedure has been modified extensively since that time, superovulation is commonly utilized to maximize the number of ovulated eggs available for either experimental purposes or for clinical use within assisted reproduction cycles. Superovulation works by providing high amounts of exogenous gonadotropin hormone so that non-dominant late secondary and early antral follicles, which are sensitive to gonadotropin stimulation and would otherwise be destined to undergo atresia, are instead maintained in the growing follicle pool and form preovulatory follicles. The preovulatory follicles can be allowed to ovulate spontaneously or induced to ovulate by administration of chorionic gonadotropin, which binds to the LH receptor and begins the ovulatory signaling cascade.

There is evidence in mice, cows, sheep, and humans that superovulation has a detrimental impact on the quality of the resulting ovulated eggs or embryos (Moor et al., 1985; Hyttel et al., 1986; Yun et al., 1989; Assey et al., 1994; Blondin et al., 1996; Van der Auwera, 2001; Baart et al., 2007; Fortier et al., 2008; Market-Velker et al., 2010; Lee et al., 2017). These findings could be explained by many factors, including accelerated oocyte growth, rescue of abnormal oocytes from atresia, and recruitment for ovulation of non-mature oocytes, all of which could lead to eggs that lack the maternally derived stores needed for developmental competence. Because calcium oscillatory patterns depend heavily on maternal components that regulate calcium homeostasis, we hypothesized that superovulation results in eggs that have abnormal patterns of calcium oscillations following parthenogenetic activation or fertilization. Small differences in calcium oscillatory behavior between spontaneously ovulated and superovulated eggs were observed following parthenogenetic activation, but there were no differences following in vitro fertilization (IVF). These findings indicate that calcium responses to sperm-induced egg activation are sufficiently robust to overcome subtle deficits in egg competence associated with superovulation.

We used the inbred mouse line, C57BL/6J, to test whether there were differences in calcium oscillatory patterns between eggs from naturally cycling (NC) and superovulated (SOV) mice. This choice was driven in large part because knockout facilities utilize this strain for the generation of the majority of their animal models. Furthermore, many transgenic and knockout lines are uniquely available on this background strain. As anticipated, far more eggs were obtained from SOV mice (median = 60.5 eggs) than from NC mice (median = 8 eggs; Mann–Whitney U = 41, n_NC = 20 n_SOV = 22, p < 0.0001; two-tailed).

We first tested SOV and NC eggs for differences in calcium oscillatory patterns following strontium-induced parthenogenetic activation, which prevents any potential variability that could be introduced by differences in individual sperm during IVF. SOV eggs took slightly longer to begin oscillating and the first transient was about 50% longer (Figures 1A–C), indicating that there were differences in strontium sensitivity and in cytoplasmic calcium handling in the presence of strontium. A detailed analysis of additional parameters including the percentage of eggs that continued to display calcium oscillations (oscillation persistence), oscillation frequency, and the area under the curve of calcium signal revealed no significant differences between the two groups (Figures 1D–F).

A similar analysis was performed of the calcium oscillatory responses of SOV and NC eggs to sperm-induced egg activation. There were no obvious differences in the patterns of calcium oscillations, the time until the first transient, or the duration of the first transient (Figures 2A–C). Likewise, there were no significant differences in calcium oscillation persistence, frequency, or total calcium signal (Figures 2D–F). These findings suggest that superovulation does not impact fertilization-induced calcium oscillatory patterns to any measurable degree.

Here we tested in a highly controlled fashion whether superovulation causes abnormalities in calcium signaling following strontium-induced parthenogenetic activation or IVF in the mouse. We found that strontium-induced oscillations were slightly different in superovulated eggs, with a small delay in initiation and longer total duration of the first transient. Because TRPV3 mediates strontium-induced egg activation, differences in the amount of TRPV3 on the egg surface could explain the altered time to the first transient (Carvacho et al., 2013). In addition, differences in the ability of the egg to clear large amounts of calcium from the cytoplasmic compartment in the presence of strontium could explain the prolonged first transient. Our findings are similar to previous observations that eggs from aged mice differ in some of their calcium oscillatory responses to strontium relative to those in eggs from young mice, whereas these differences are not observed following more physiological egg activation methods (microinjection of either sperm heads or cRNA encoding the PLCζ protein) (Haverfield et al., 2016).

Despite our prediction to the contrary, the data clearly showed that in a commonly utilized inbred strain, C57BL/6J, superovulation according to a standard protocol does not alter the calcium oscillatory patterns at fertilization. It is possible that alternate superovulation protocols or repetitive superovulation could produce a different result. Superovulation maximizes the numbers of ovulated eggs that can be obtained for study from each mouse, reducing the numbers of animals that must be euthanized for experimental purposes by about 10-fold based on our data. Therefore, this technique is routinely performed to obtain oocytes and early embryos in laboratories and animal facilities. Our findings demonstrating normal fertilization-induced calcium oscillatory patterns despite superovulation is reassuring regarding interpretation of experiments examining factors that influence calcium signals given that these types of experiments are almost always performed with eggs obtained by superovulation. Our data support the use of superovulation in future studies of calcium signaling at fertilization, but suggest caution in the interpretation of results obtained in experiments utilizing chemical methods to activate superovulated eggs.

Despite being a widespread practice, superovulation is associated with abnormalities in several phenotypic features in oocytes and ovulated eggs. For example, an increased incidence of meiosis arrest prior to metaphase II, asynchrony between nuclear and cytoplasmic maturation, abnormalities in subcellular structure, and abnormal patterns of protein synthesis have all been observed in various mammalian species following superovulation (Moor et al., 1985; Callesen et al., 1986; Hyttel et al., 1986; Yun et al., 1989; Assey et al., 1994). Given that the hormone injections only rescue late secondary and antral follicles, these differences are clearly a consequence of inadequate preparation for oocyte maturation during later stages of follicle development. Although the oocyte is transcriptionally inactive after reaching the antral stage, translation continues to occur, creating additional maternal components to support maturation, and embryo development. In addition, mRNAs and long noncoding RNAs are transferred to the oocyte from the surrounding cumulus cells, at least some of which are translated and may contribute to oocyte competence to undergo maturation (Macaulay et al., 2014, 2016). Superovulation could accelerate late follicle development enough that there is insufficient time for the accumulation of specific maternal components that could explain the altered sensitivity to strontium in SOV and NC eggs. However, the absence of differences in fertilization-induced calcium dynamics suggests that adequate amounts of the maternal components required for physiological calcium homeostasis are already present prior to superovulation or, alternatively, that these components are not affected in a detrimental fashion by the superovulation protocol.

Unlike oocyte maturation defects, which occur prior to fertilization, some superovulation-associated abnormalities are observed following fertilization. Reduced competence to develop to or beyond the 16-cell stage, both in vivo and in vitro, abnormalities in acquisition and maintenance of methylation marks on imprinted genes and fetal growth retardation all are associated with superovulation protocols in cows and/or mice (Blondin et al., 1996; Van der Auwera, 2001; Fortier et al., 2008; Market-Velker et al., 2010). Differences in embryo development likely result from a lack of accumulation of other maternal components in the final period of follicle growth to the periovulatory stage that are not related to the calcium signaling toolkit. Future studies will be needed to determine the identity of such components or to develop superovulation protocols that do not disrupt preovulatory follicle development.