New markers of human cumulus oophorus cells cultured in vitro – transcriptomic profile

Cumulus cells (CCs) are differ significantly from granulosa cells (GCs), resting on the basal lamina of ovarian follicle. CCs are in close physical contact with the oocyte, forming a cumulus-oocyte complex (COC). The oocyte controls the differentiation and expansion of CCs, which in turn are involved in the metabolism of pyruvate and glucose consumed during energy production in the oocyte [1,2]. Due to this proximity, CCs are influenced by regulatory factors produced by the oocyte. These regulatory factors include primarily: growth differentiation factor 9 (GDF9), bone morphogenetic protein 15 (BMP15) and fibroblast growth factor 8 (FGF8). They are primarily responsible for suppressing expression of the LH receptor and genes responsible for steroid production. Unlike the GCs, which “remain” in the ruptured follicle, CCs are “ejected” from the ruptured follicle during ovulation [3]. CCs also produce many inflammatory factors and cytokines that are released during ovulation. Cumulus cell-oocyte complex (COC) released from the ovary is therefore a kind of a microenvironment [4, 5, 6]. Increased concentration of gonadotropins in the female body causes increased production of hyaluronic acid by CCs, which expands the spaces between these cells. Correct ovulation requires the production of prostaglandins. GDF9 (a member of the TGF-beta superfamily) plays a major role in the induction of Ptgs2 expression in CCs through increased luteinizing hormone (LH) concentration. Lack of GDF9 blocks the development and growth of follicles, thus leading to infertility [7].

In most mammal species, including humans, CCs cells surround the oocyte at conception. It is believed that CCs interact with the oocyte and sperm, and thus are involved in promoting the fertilization process and oocyte developmental competence [8,9]. It has been proven many years ago that CCs are involved in maintaining myocyte retention in the oocyte. Oocytes removed from the follicle, lacking CCs, resume meiotic division, which should be completed only after fertilization [10]. In addition, CCs facilitate the capture of COC by ciliated oviduct cells [11]. Glycodelin-C, a derivative of glycodelin, is isolated from the matrix of CCs cells. This substance stimulates the binding of spermatozoa to the zona pellucida of oocyte [9]. It has also been reported that the rate of CCs associated with morphologically abnormal oocytes or immature oocytes is much higher than in CCs surrounding morphologically normal, mature oocytes. The increase in CC apoptosis has a negative impact on a number of processes related to the correct fertilization and development of blastocyst and reduces the effectiveness of in vitro fertilization (IVF) [12,13].

The expression of genes associated with the proper function of CCs results from a number of factors produced by the oocyte, but also the environment of the mature ovarian follicle, and all processes occurring during follicle growth in the ovarian microenvironment [14].

CCs also produce antioxidant compounds that reduce the level of oxidative stress caused by reactive oxygen species (ROS). These compounds, mainly superoxide dismutase (SOD) [15] and Glutathione transferase S theta 1 (GSTT1) protect the oocyte against oxidative stress [16,17].

Knowledge about the expression of individual genes, as well as the presence of individual components of the intracellular metabolic pathways, or the CCs apoptosis index can become a source of valuable molecular markers determining the developmental competence of oocytes. It is known that the expression of amphiregulin (AREG) and epiregulin (EREG) (epidermal growth factor (EGF) -like factors) depends on the dose of LH administered during ovarian stimulation. Therefore, it is suggested that AREG and EREG are part of the signal transduction pathway, which leads to the release of the oocyte from the mature ovarian follicle and the luteinization process in women [18]. Recent studies described of new properties of CCs. CCs produce a large amount of hyaluronan, which targets CD44 (marker of cancer cells). Culture of pancreatic cancer cells in a medium conditioned with CCs activated pro-apoptotic genes in these cancer cells [19]. The mechanisms of communication between the oocyte and CCs cells through gap connections are well known. It is known, however, that this communication also takes place by means of paracrine signaling. Less known methods of communication are the transfer of non-coding RNAs through exosomes from the cumulus to the oocyte [20]. Coenzyme Q10 (CoQ) is another factor that has a significant impact on the quality of oocyte and CCs cells. This relationship is important for the proper functioning of mitochondria, not only in the oocyte but also in CCs cells. A decrease in mitochondrial activity is associated with aging of oocytes. Therefore, reduced CoQ production in older women translates into reduced fertility and a higher risk of birth defects for embryos [21].

Understanding the molecular mechanisms and regulation of gene expression in CCs may reflect processes in the oocyte. Changes occurring at the molecular level may be a signpost for identifying oocyte quality, oocyte acquisition of developmental competences, and quality of obtained blastocyst. The main purpose of the research was to identify the potential molecular markers responsible for cell junction organization, migration, differentiation, morphogenesis and motility.

We used Human Genome U219 Array Strip for the microarray gene expression analysis of human cumulus oophorus cells. This method allowed us to study the gene expression of 22,480 transcripts at 1, 7, 15 and 30 days of in vitro cumulus oophorus cell culture. We selected genes with more than 2- fold changes and corrected p-values less than 0.05 for downstream analysis. A total of 4773 differentially expressed genes (DEGs) were identified according to the above criteria. We started the microarray gene expression analysis with subjecting the list of DEGs to DAVID software, which showed that the genes can be assigned to 775 GO BP, 33 GO MF and 125 GO CC gene ontology terms. This paper focused on the genes involved cellular morphogenesis, junction and migration. The DAVID software indicated the following GO BP terms, which cover the above processes: “cell junction organization”, “cell migration”, “cell morphogenesis involved in differentiation”, “cell morphogenesis” and “cell motility”. The 150 genes involved in those processes were clustered using hierarchical clustering and presented as heatmaps (Fig. 1). It is worth mentioning that 144 genes were upregulated, which is the greater part of the list of genes used for hierarchical clustering. The downregulated genes are: SLC7A8- solute carrier family 7 (amino acid transporter light chain, L system), member 8; DFNB31- deafness, autosomal recessive 31; COL1A1- collagen, type I, alpha 1; CDC42SE1- CDC42 small effector 1; TGFBR3- transforming growth factor, beta receptor III; HMGB1- high mobility group box 1. The direction of expression change (upregulation or downregulation) was maintained in cumulus oophorus cell culture in subsequent points of analysis (after 7, 15 and 30 days of in vitro culture). The 10 most significantly upregulated and all of the downregulated genes, their symbols, fold changes and corrected p- values are shown in table 1.

Figure 1

In the next part of analysis, we focused on the z-scores, which tell us whether the molecular function is more likely to be decreased (negative value) or increased (positive value). The z-scores were presented as segments of inner circles in the figure 2. As can be seen from the figure, expression of most genes was increased (green dots) in all ontological groups. The z-scores of above-mentioned GO BP terms had positive values, so the processes described by these GO BP terms were upregulated. The expression pattern did not change at any of the analyzed time points. Considering the above, the subsequent analysis was based only on 7D/1D comparison.

Figure 2

In the next section, we checked the interaction between selected ontological groups. One of the most visually appealing way of presenting such interaction is dendrogram (Fig. 3). Clusters contain functionally related genes based on their expression pattern. The middle circle represents a logarithm of fold change (logFC) of differentially expressed genes assigned to the studied GO terms. The GO terms are shown as the outer ring. The genes whose expression is downregulated form clusters marked by blue part of the middle circle and analogously, red indicates upregulated genes. Clusters of the same color over the entire width of the outer circle represent genes that are unique for a specific GO term. Clusters of different colors on the cross section of outer circle show sets of genes which are likely to be functionally related. The dendrogram showed that many genes belong simultaneously to “cell motility” and “cell migration” or to “cell morphogenesis” and “cell morphogenesis involved in differentiation”. The genes that are unique for a specific GO term belong mainly to “cell morphogenesis”.

Figure 3

In the gene ontology database, single genes may belong to many ontological terms. For this reason, we used plots with visualization of logFC values and relationship between genes and selected GO BP terms (Fig. 4 ). The relationship was also presented as a heatmap (Fig. 5). The strongest upregulated genes from examined GO BP terms included, among others: ANXA3- annexin A3, KIAA1199, HTR2B- 5-hydroxytryptamine (serotonin) receptor 2B, G protein-coupled, VCAM1- vascular cell adhesion molecule 1 and DKK1-dickkopf WNT signaling pathway inhibitor 1.

Figure 4

Figure 5

In the next part of analysis, we focused on the interaction between proteins encoded by DEGs belonging to studied GO BP terms. Firstly, we used STRING software for the interaction prediction. The number of genes used to create STRING interaction network was limited to 50 most changed DEGs for readability (Fig. 6).

Figure 6

Finally, we used ReactomeFIViz app for investigation of functional interactions between proteins en coded by DEGs belonging to selected GO BP terms. Among the most significantly enriched functional interaction networks were FI networks for “Cell migration” and “Positive regulation of cell migration” (Figs 7 and 8).

Figure 7

Figure 8

It is known that CCs are necessary in the process of acquiring developmental competence by the oocyte, including enabling the resumption of meiosis and the transition to the meiosis metaphase II [25]. Thanks to the gap connections, it is possible to transport molecules between CCs cells and the oocyte [26]. Proper oocyte maturation without CCs is practically impossible, and the effectiveness of fertilization of such an oocyte and obtaining a correct blastocyst drastically decreases [27]. Furthermore, metabolomic studies of the spent culture medium obtained show that CCs are secreted into the external environment and allow oocyte maturation [28,29]. Due to the closeness and interaction of CCs and oocyte, they have become an interesting research model in embryology.

In the presented studies, during the transcriptome analysis of CCs cells maintained in long-term primary in vitro culture, groups of genes were selected, responsible primarily for processes associated with “cell junction organization”, “cell morphogenesis involved in differentiation”, cell morphogenesis”, “cell motility”, “cell migration”. For the purposes of the presented research, these selected ontological groups can be conventionally divided into two group: 1) “cell motility”, “cell migration” - here, genes are primarily responsible for the movement of cells from one place to another; 2) “cell junction organization”, “cell morphogenesis involved in differentiation”, “cell morphogenesis “ are responsible for organizing the components of connections between two cells but also between the cell and the extracellular matrix. In addition, a second group of selected ontological groups describes the genes responsible for cell movement from one site to the destination directed by molecular information. Moreover, genes involved in shaping cells and changing their form, shape and size that occur when non-specialized cells acquire the specialized structural features of a cell population characteristic of an organ.

The presented research results indicate that the strongest upregulated genes from examined GO BP terms included, among others: DKK1, ANXA3, KIAA1199, VCAM1, HTR2B. Only 6 genes demonstrated downregulation relative to the reference value, which was the gene expression shown on day 1 of primary culture in vitro.

The factor conditioning correct embryo implantation is activation of canonical WNT signaling. The WNT signaling pathway is regulated by steroids [30,31]. WNT activation in the embryo may or may not affect the correct implantation of the embryo, but is responsible for regulating pluripotency [32]. DKK1 is an antagonist of the WNT signaling pathway [33], as well as the protein produced by endometrial cells during the menstrual cycle, early pregnancy or estrus in other mammals [34,35]. It is believed, that the proper development of the embryo and the proper implantation depend on a number of regulatory factors secreted by the mother’s reproductive system [36]. DDK1 is also potentially involved in mother-embryo communication [37]. Studies on bovine embryos indicate that DKK1 facilitates TE and hypoblast differentiation. DKK1 plays one of the key roles in the proper functioning of the reproductive system of most mammals, including humans [37, 38, 39, 40, 41]. In the light of our research and existing knowledge about the role of CCs in communication with the oocyte, high DKK1 expression may also affect the quality of obtained blastocyst and its implantability [42]. Studies suggest that adding CCs cells to the embryo environment improves embryo quality and pregnancy rates.[43]. Perhaps the high expression of this gene in CCs cells may also affect the acquisition of oocyte developmental competences and then having sufficiently high pluripotent properties. Therefore, our research suggests that the presence of CCs cells in the oocyte environment during fertilization may result in improved in vitro fertilization efficiency.

The genes also responsible for the proper development of blastocyst include the ANXA3 gene. The protein encoded by this gene is responsible for binding and maintaining primarily cell membranes. ANX are proteins found in a variety of tissues, from the respiratory and urinary epithelium to the presence in peripheral nerves. Proteins from the anexin family due to their expression in individual tissues most likely take part in the processes of equalization of individual tissues, and are also partly responsible for their physiological properties [44]. The presented studies suggest an increase in ANXA3 gene expression in the following days of primary culture in vitro in the absence of oocyte. To date, the presence of the ANXA3 gene has been reported in the morula and blastocyst stages during preimplantation development [45]. The ANXA3 protein is part of human neutrophils and promotes tight connections between membranes of individual neutrophils, which causes their aggregation [46, 47, 48].

As mentioned in the introduction, hyaluronan (HA), or hyaluronic acid, plays an important role in the reproductive system. It is produced primarily by the oocyte, embryo and other elements of the reproductive system, depending on the species of the mammal (fallopian tube, uterus, cervix) [49,50]. In addition, HA is produced by CCs and GCs cells [51,52]. The KIAA1199 (CEMIP; HYBID) gene showed a significant change in expression. The protein encoded by this gene plays a key role in depolymerization of HA molecules. Microarray analysis by Abe et al. it identified the expression of this gene in the cochlea, as well as in the inner ear [53]. Other studies claim that this gene is responsible for promoting the process of apoptosis, cancer progression. In addition, he is responsible for regulating cell proliferation, adhesion, motility and epithelial-mesenchymal transition cells of cancer. [54].

The gene closely related to cell adhesion and extracellular matrix modeling is VCAM1. It shows increased expression in GCs and CCs cells [55]. High expansion of the VCAM1 gene has also been demonstrated in mouse granular cells after dehydroepiandrosterone (DHEA) androgenization. The result of the increase in VCAM1 expression was an exacerbation of the symptoms of polycystic ovarian syndrome in mice (PCOS). It is suggested that VCAM1 expression knockout is one therapeutic option for PCOS [56]. Expression of the VCAM1 protein was not detected in mouse embryos prior to implantation. However, it is known that VCAM1 is present in oocytes and early human embryos [57,58]. The presented research results suggest that CCs cells in in-vitro culture medium show significant increase in expression in the following days of the in vitro primary culture. Perhaps the lack of interaction with the oocyte determines this growth.

Only 6 genes showed a decrease in expression relative to the control. In the presented article we focused on two genes the most downregulated (TGFBR3 and HMGB1).

In the presented studies, it seems surprising that the expression of TGFBR3 gene shows a decrease. This gene codes for the receptor protein for TGF-beta. The transforming growth factor family (TGF-beta) are multifunctional molecules involved in a number of processes primarily associated with tissue development, cell differentiation process, cell migration and adhesion, as well as the production of extracellular matrix. Receptor types I, II and III are involved in the action of TGF-beta. TGFBR3 is a TGF-beta binding glycoprotein that can have a membrane-bound form as well as a dissolved form [59]. This receptor may also exist as an inhibin co-receptor [60]. TGFBR3 expression in CCs cells during long-term culture decreases. This may be associated with a lack of communication with the oocyte and a change in the natural environment for CCs cells. The expression of TGFBR3 decreases during the development of breast cancer [61]. In studies on the effect of age on the quality of oocytes, TGFBR3 was found to be overexpressed in a group of women over 37 years of age [2]. It turns out that changing the environment to extracorporeal can have a great impact on obtaining oocyte developmental competences, as well as the development of the embryo and its proper implantation after transfer to the mother’s uterus [62]. This factor can have a significant impact on the success and effectiveness of the in vitro fertilization procedure. The decrease in TGFBR3 expression is associated with reduced oocyte developmental competence [63]. On the other hand, studies in mice prove that the Tgfbr3 knockout was associated with their high fertility and increased folliculogenesis [64].

HMGB1 is another gene that showed a decrease in expression. Melatonin has shown increased expression of this gene during oocyte maturation [65]. High expression of HMBG1 reduces blastocyst demand and thus increases the chances of embryo survival [66,67]. Moreover, studies show that there is a correlation with the amount of HMGB1 protein in follicular fluid (FF) and pregnancy and endometrial thickness. There are scientific reports confirming the fact that the protein encoded by the present gene can be an important indicator of the success of in vitro fertilization [68]. The protein encoded by the HMGB1 gene is responsible for DNA binding and participates in transcription, DNA replication and repair. The decrease in expression of the presented gene may be the result of a change in the CCs cell culture environment. For transfer to in vitro conditions, we may suggest that these cells have lost their physiological properties [69].

The presented research results allowed to identify groups of genes responsible for processes related to “cell junction organization”, “cell morphogenesis involved in differentiation”, “cell morphogenesis”, “cell motility”, and “cell migration”, as well as to define interactions between individual genes. The results suggest that the most upregulated genes are primarily responsible for processes related to communication with the oocyte as well as the proper development of the embryo and its implantation. The obtained results indicate that in in vitro cultured CCs do not lose their properties or, on the contrary, their gene expression decreases due to the loss of contact with the oocyte.