Background
Oocytes secrete paracrine factors such as growth differentiation factor 9 and bone morphogenetic protein 15, which affect the gene expression of cumulus cells, thereby leading to cumulus expansion [
1,
2]. Cumulus expansion involves hyaluronan accumulation in the intercellular space of cumulus cells [
3], and the structural integrity of the cumulus cell extracellular matrix (ECM) is essential for oocyte maturation. Several cumulus proteins linked to ECM hyaluronan, including pentraxin-3 (PTX3) [
4,
5], are required for cumulus integrity, thereby ensuring cumulus expansion and oocyte maturation. In addition, nutrients and metabolites are transported through gap junctions between the oocyte and cumulus cells [
6]. Taken together, bidirectional intercellular communication between oocytes and their surrounding cumulus cells is essential for the development of an egg that is competent to undergo fertilization and embryo development [
3,
7,
8].
Cumulus cells gene expression may reflect oocyte maturation and competence. Numerous studies have been conducted to profile cumulus gene expression, identify gene markers, and predict oocyte or embryo competence [
9‐
11]. The expression of potential marker genes in cumulus cells has been suggested to be associated with oocyte competence [
10‐
14], embryo quality [
9,
15‐
17], pregnancy outcome [
9,
18,
19], and live birth [
20].
GJA1, also known as connexin 43, is the major gap junction protein that is expressed in cumulus granulosa cells, participating in connexons with other cumulus granulosa cells or the oocyte. GJA4, corresponding to connexin 37, appears to be the only connexin synthesized by the oocyte; it forms a heterologous gap junction with granulosa cells [
6]. Loss-of-function of
GJA4 interferes with the development of antral follicles [
21,
22] and can be rescued by replacement with
GJA1 [
23].
GJA1 has been identified as a gene marker for oocyte fertilization potential and embryo quality [
16,
17].
PTX3 gene expression in cumulus cells was also associated with oocyte/embryo competence and was a potentially reliable predictor of embryo developmental competence [
14].
Proteases play an important role in the proteolysis process and are essential for tissue remodeling and functions of the ovary in which extensive tissue remodeling during folliculogenesis and ovulation take place.
SERPINE2 mRNA levels in granulosa cells have been suggested to be a potential pregnancy biomarker [
18]. Higher
SERPINE2 expression levels were detected in cumulus cells of human immature oocytes than in those of mature oocytes [
24,
25].
PRSS35 belongs to the trypsin class of serine proteases. This protease is exclusively expressed in the mouse ovary, suggesting that
Prss35 is involved in ovarian functions [
26].
In the present study, we focused on the four cumulus-expressed genes GJA1, PRSS35, PTX3, and SERPINE2 and evaluated whether their expression levels correlate with oocyte maturation, fertilization, and embryo development in humans.
Discussion
GJA1-mediated gap junctional communication regulates oocyte meiosis resumption, and lower levels of GJA1 in cumulus cells are beneficial for oocyte maturation [
32]. Our data showed that there was significantly lower expression of
GJA1 in cumulus cells of mature oocytes than in those of immature oocytes. Feuerstein et al. reported a similar result [
12]; however, no difference was found in a study with a relatively smaller sample size [
16].
GJA1 mRNA levels in cumulus cells showed no significant differences between fertilized and unfertilized oocytes (Fig.
3), in line with the results reported by Hasegawa et al. [
16].
GJA1 expression in cumulus cells surrounding the mature oocytes did not show any difference in developing embryos with good or poor morphology (Fig.
4), as reported previously [
12]; however, significantly lower expression was also reported for embryos with good morphology [
16]. Intriguingly, while
GJA1 expression in cumulus cells enclosing oocytes achieving blastocyst development showed significantly lower levels than that in cumulus cells enclosing oocytes unable to develop to blastocyst stage [
12], its expression in cumulus cells was also reported to be significantly higher in pregnant patients [
17].
SERPINE2 mRNA levels in cumulus cells are associated with oocyte maturation, with a significantly lower level in mature oocytes [
24,
25]. This correlation was verified in our study (Fig.
2) using a larger sample cohort. In a mouse animal model, higher levels of SERPINE2 were demonstrated to impair cumulus expansion and oocyte maturation [
25]. Although SERPINE2 showed a trend toward a higher expression in cumulus cells of oocytes developing into good embryos than in cumulus cells of oocytes developing into poor embryos (Fig.
4), the difference was not significant between the two groups. Nevertheless,
SERPINE2 mRNA levels were significantly higher in granulosa cells collected from follicles that resulted in pregnancy [
18].
Our data revealed that
PRSS35 mRNA levels are associated with oocyte fertilization potential (Fig.
2). This is the first report demonstrating a relationship between human
PRSS35 expression and oocyte competence.
Prss35-null mice have no defects in female fertility, suggesting that the gene is dispensable for murine fertility and embryo development [
33]. However, this remains unclear in case of humans. Diao et al. did not detect any compensatory upregulation of other proteases reported in the uterus; however, the expression of other protease-related genes cannot be ruled out [
33]. Alternatively, other compensatory effects may exist in
Prss35 knockout mice as PRSS35 may not possess any serine protease activity because the amino acids in its protease catalytic active site are changed. A study employing knockout mice lacking specific proteases that cause some unexpected results in the ovulatory process because of redundancy and overlapping functions has also been conducted previously [
34].
While the expression of
PTX3 has been reported to be significantly lower in cumulus cells from immature oocytes than in those from mature oocytes [
15], our data showed no significant difference (Fig.
2c). Pooled cumulus cells from fertilized oocytes were found to have 3- to 12-fold increases in
PTX3 levels than those from unfertilized oocytes [
14]; however, the cumulus cells collected from individual COC in our study showed no significant difference. Congruent with previous reports [
10,
20], we found that cumulus
PTX3 mRNA levels were not associated with the embryo morphological grade.
Selecting embryos with good morphology for transplanting has proved to be correlated with a good pregnancy outcome [
35]. Our results showed that expression levels of
GJA1,
PRSS35,
PTX3, and
SERPINE2 did not differ significantly between cumulus cells of embryos with good and poor morphology, indicating that these genes may not be used as marker genes for predicting early embryo development.
Comparison of cumulus gene expression of pregnant and nonpregnant embryo, which is the transfer of a single embryo, would be a good strategy for revealing good marker genes. Using the single embryo transfer (SET) strategy, several studies have proposed marker genes for pregnancy outcome [
20,
36‐
38].
This study has some limitations: the controlled ovarian hyperstimulation procedure produces more mature oocytes than immature oocytes, and the ICSI treatment results in few unfertilized oocytes; thus, the sample size of immature oocyte or unfertilized oocyte is always small in this type of study, as seen previously [
10,
12,
15,
16]. In addition, dead and arrested embryos are also found. Although cumulus cells that developed into good or poor embryos showed significantly higher
GJA1 mRNA levels than those that developed into arrested embryos (data not shown), the sample size was relatively less because of the unavailability of arrested embryos. Furthermore, this kind of study often produces contradictory results (e.g.,
GJA1 and
PTX3 cases mentioned above). The cumulus expression of
EFNB2,
RGS2, and
VCAN, which are proposed as biomarkers of pregnancy [
19,
20,
36‐
38], was also inconsistent as revealed in a recently published report [
39].
The selection of embryos with higher developmental potential has been one of the major challenges in assisted reproductive technology (ART). Because cumulus cells are discarded in ART, the cumulus samples have the advantage of being noninvasively collected. Although cumulus gene expression may represent a promising method compared with the currently used morphology-based method, more investigations are warranted. Thus, a prospective larger cohort study or the use of SET cumulus samples remains necessary to clarify the effectiveness.
In summary, GJA1 and SERPINE2 represent gene markers potentially associated with oocyte maturation, and PRSS35 may be correlated with oocyte fertilization ability. GJA1, PRSS35, PTX3, and SERPINE2 may not be considered as markers for predicting early embryo development.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
SHL participated in the design of the study and drafted the original manuscript. MHL and YMH contributed to sample collection and data analysis. CHL and LYY conducted qPCR experiments and contributed to data analysis. YJC helped in sample collection. RKKL conceived the study and participated in project design and coordination. All authors read and approved the final manuscript.