Interactions of bovine oocytes with follicular elements with respect to lipid metabolism

Sample collection

Material was collected from ovaries of slaughterhouse origin. Each sample analyzed in this experiment represented an individual 5–8 mm follicle and comprised four follicular components: CCs, granulosa cells (GC), the oocyte and the FF. The following analyses were performed: lipid droplets staining (in oocytes), messenger RNA (mRNA) expression of seven selected genes (in cumulus and granulosa cells) and fatty acid composition (in follicular fluid).

COCs were recovered from individual follicles. Each follicle was aspirated separately with a 1 mL syringe and needle (1.2 mm diameter). FF aspirated from a single follicle was transferred to a Petri dish. After removing COC from each FF sample, follicular fluid was centrifuged to obtain a pellet with granulosa cells and supernatant with FF – both frozen in liquid nitrogen. With regard to COCs, all the CCs surrounding the individual oocytes were removed by vigorous pipetting in 150 μL of 0.2% phosphate-buffered saline / polyvinylpyrrolidone (PBS/PVP) solution and frozen in liquid nitrogen. The remaining oocytes were subjected to LD staining.

LD staining, fluorescent imaging

The procedure involved individual oocytes and was performed as described by Aardema et al. (2011) with some modifications. A denuded oocyte was fixed in 4% (v/v) paraformaldehyde (PF) for 1 h at 37°C, was washed and stored in 1% PF at 4°C for a maximum of 1 week. Next, the oocyte was washed twice in PBS with 0.3% (w/v) PVP, was permeabilized for 30 min in PBS with 0.2% (w/v) Triton and washed in PBS. Lipid droplets were then stained with a specific neutral lipid stain BODIPY 493/503 (Molecular Probes, Eugene, Oregon, USA) in PBS (20 μg/mL, 1 h) and the oocyte was washed three times in PBS with PVP. Further, each oocyte was individually mounted on a concave glass slide in 40 μL drops containing antifade solution with 4′,6-diamidino-2-phenylindole (Vectashield mounting medium; Vector Laboratories, Burlingame, CA, USA).

Fluorescent signals were visualized using a Zeiss Axiovert 200 M laser scanning confocal microscope (Carl Zeiss Microscopy GmbH, Jena, Germany). The stack of slices taken every 10 μm from the equator to one of the poles was saved and the number of LDs was counted on each slice.

RNA extraction, complementary DNA synthesis

Total RNA was extracted with the High Pure miRNA Isolation Kit (Roche, Diagnostics, Indiana, USA) according to the manufacturer's protocol. Further synthesis of complementary DNA (cDMA) was carried out with a maximum of 2 μg of RNA, using Transcriptor High Fidelity cDNA Synthesis Kit (Roche). The cDNA samples were stored at −20°C. Although the final cDNA concentration was not measured, 1 μL of cDNA was used in each quantitative PCR reaction.

Quantitative gene expression analysis – real-time PCR

The analysis was performed on a Roche Light Cycler 2.0 system. The primer sequences used in the experiment are summarized in Table 1. 18S rRNA and GAPDH genes were used as reference genes. The reactions were carried out in 20 μL capillaries with Light Cycler FastStart DNA Master SYBR Green I (Roche Diagnostics). Product specificity was confirmed by melting analysis. The PCR protocol included an initial step of cDNA denaturation at 95°C (10 min), followed by 45 cycles of 94°C (10 s), 57°C (5 s) and 72°C (10 s). The melting protocol included the initial step of denaturation at 95°C and heating starting from the annealing temperature of the analyzed gene up to 94°C, holding each temperature for 0.1 s while monitoring fluorescence. Each cDNA sample was analyzed twice.

Fatty acids composition of the FF

Fatty acid composition of the FF was analyzed by gas chromatography according to the procedure described by Cieslak et al. (2009) and adapted to suit the FF analysis by Warzych et al. (2011).

Statistical analysis

Distributions of the analyzed parameters were skewed, therefore, a log-transformation was applied in order to normalize the data before final analyses. The magnitude of the association and the direction of the relationship between the analyzed variables was assessed as Pearson's correlation coefficient.

More LDs – more nearly all FA in FF

FF is a product of both the blood plasma constituents and of the secretory activity of granulosa and theca cells (Fortune 1994). Composition of the FF has a direct impact on the follicular cells and further on the oocyte. Among many FF components, fatty acids emerge as promising factors, which were described to influence oocyte growth and maturation (e.g. Leroy et al. 2005; Marei et al. 2009, 2010, 2012; Aardema et al. 2011; Carro et al. 2013). Since bidirectional signaling between CCs and oocytes has been shown (Hussein et al. 2006), there is an obvious expectation to find relations between fatty acid composition of FF and lipid metabolism within the oocyte.

The present study reveals a positive correlation of LD numbers in the oocyte and the level of the majority of FAs detected in FF. Studies showed that intracellular lipid content is correlated to developmental competence of oocytes (Paczkowski et al. 2014). Further, others previously reported that LDs play a protective role, storing higher doses of fatty acids, which could cause some negative effects on oocyte metabolism (Carro et al. 2013). Also oocytes of good quality (those which successfully went through the process of growth, development, fertilization and embryo development) had higher cytoplasmic lipid content (Jeong et al. 2009). Considering the published data along with the results of the present study, it can be concluded that the more FA in follicular fluid is stored, the more LDs are formed. It may suggest an active element of a protection system for oocytes, where excess of lipids originating from FF are stored in higher numbers of lipid droplets. Such a system may support the defense mechanism against lipotoxicity, along with accumulation of excess of lipids from FF in cumulus cells, which has been previously described by Aardema et al. (2013).

LD numbers in oocytes versus gene expression in cumulus and granulosa cells

Energy substrates such as glucose or fatty acids are either produced by the oocyte or supplied via the cumulus cells from FF (in vivo) or culture medium (in vitro). Cumulus cells penetrate the zona pellucida through the gap junctions and contact the oolema. This communication system allows bidirectional signaling, which is essential for many significant changes within the oocyte, such as growth and maturation.

In the present experiment a set of genes was analyzed with regard to the mRNA content both in cumulus and granulosa cells. Since all follicular elements originated from individual ovarian follicles, the correlation between gene expression in cells and LD numbers in corresponding oocytes could be analyzed. Unexpectedly, the LD number was not correlated with mRNA expression of any of the analyzed genes regulating FA metabolism. Only GLUT1 (glucose transporter) mRNA levels both in cumulus and granulosa cells showed negative correlation to LD numbers in oocytes. Studies showed that glucose consumption and fatty acid metabolism appear to be related in the mouse (Paczkowski et al. 2014) and bovine (Van Hoeck et al. 2011) COCs. The ‘glucose-fatty acid cycle’ concept was proposed, which described the inter-regulation of glucose and fatty acid metabolism to maintain homeostasis. Glucose provision promotes glucose oxidation and glucose and lipid storage, and inhibits fatty acid oxidation. Provision of free fatty acids promotes fatty acid oxidation and storage, inhibits glucose oxidation and may promote glucose storage if glycogen reserves are incomplete (Randle 1998). The increased ability of the COC to utilize fatty acids as energy substrates may decrease the need for the COC to metabolize glucose, or allow glucose to be stored or participate in other essential pathways. Thus, an appropriate balance between glucose and fatty acid metabolism should be observed in oocytes. The results from the present study may indicate high numbers of LDs in oocytes as a hallmark of high FA metabolism within the oocyte. Therefore, in order to sustain the homeostasis, lower glucose metabolism is observed in follicular cells. Also, previously Paczkowski et al. (2014) observed that the inhibition of fatty acid oxidation during in vitro maturation of mouse oocytes resulted in an increase of mRNA content of GLUT1 gene both in oocytes and cumulus cells. This data confirms the significance of lipid and glucose metabolism within the follicle.

An oocyte enclosed by CCs is able to incorporate FAs from its environment (Aardema et al. 2011; Carro et al. 2013). However, in our experiment LD numbers were not correlated with relative abundance (RA) of genes regulating fatty acids metabolism in CC and granulosa cells. Others have shown that CCs are able to incorporate and store FAs by forming LDs as a protection system from SFAs (Aardema et al. 2013). Further, CCs themselves exert lipogenic and lipolytic activity (Auclair et al. 2013). It was also proven that inhibition of fatty acids oxidation in bovine CCs arrests meiotic maturation of enclosed oocytes (Brisard et al. 2014). However, the question arises, what origin are the lipids transferred by follicular cells to the oocyte? Are the cells transferring their own lipids or lipids originating from the FF? Since our experiment did not reveal any significant data on gene expression in follicular cells with regard to lipid metabolism within the oocyte, we suggest that FAs accumulated in the oocyte are of FF origin rather than of follicular cells. However, this interesting subject needs to be thoroughly investigated.

Gene expression in cumulus and granulosa cells versus fatty acids concentration in FF

Data from the present experiment revealed correlations between FA content in FF and expression of several genes (SCD, FADS2, ELOVL5, GLUT8) in granulosa cells, but only SCD gene in CCs. It may be suggested that granulosa cells participate in follicular lipid metabolism more actively than CCs. This statement may be supported by the fact that granulosa cell proliferation and steroid production are both required to sustain follicular growth (Drummond & Findlay 1999). Also, high concentrations of C16.0, C18.1 and C18.0 FA inhibit granulosa cell proliferation (Vanholder et al. 2005). Further, Uzbekova et al. (2015) described distinct lipid profiles between cumulus, granulosa and theca cells, confirmed also by different expression of lipid metabolism-related genes, thus substantial differences in the metabolism of different cell types within the follicle are expected.

The present experiment reveals significant positive correlations between GLUT8 gene expression and concentration of several individual FAs (C16.0, C18.0, C18.1cis9) as well as groups of FA (total FA, ΣC18.1trans, ΣC18.1cis, SFA, UFA, MUFA, PUFA, n6). GLUT8 protein is a member of the class III of facilitative glucose transporters (SLC2As) with high affinity for glucose; however, it may transport also other hexoses. Studies showed that GLUT8^−/− gene knockout mice exhibited abnormalities potentially resulting in a poorer quality oocytes. The authors observed lower gaining of weight on a high-fat diet, which could be attributed to a lack of body fat stores (Adastra et al. 2012). Thus we suggest GLUT8 as a gene of interest in terms of oocyte quality and its lipid metabolism.

In the growing follicle, the oocyte is enclosed by somatic cells, which have a substantial impact in oocyte quality determination. The follicles are well described trails in the bidirectional communication between follicular fluid and oocytes. Thus, one could assume existence of defined relations between FF composition and gene expression in cumulus/granulosa cells. The results of the present study do not support this assumption due to a lack of convincing data linking the mRNA expression of genes regulating FA metabolism in follicular cells with FA composition of the FF. Although some single fatty acids concentrations were related to SCD, FADS2 or ELOVL5 genes, in our opinion the obtained results do not give any clear answer on the questions how cumulus or granulosa cells might be involved in FA composition in follicular fluid. The literature describes that FADS2 gene product catalyzes the initial step in the enzymatic cascade of the synthesis of PUFA, SCD is a microsomal rate-limiting enzyme involved in the desaturation of stearic acid to oleic acid, whereas ELOVL5 is implicated in the elongation of long-chain PUFA. If we assume that FF is the source of substrates for enzymatic changes controlled by FADS2, SCD and ELOVL5, the expression of these genes may not directly reflect the concentration of fatty acids in FF. Basically, in our opinion it confirms how complex the lipid metabolism of the ovarian follicle is and it indicates the need to develop this subject in further experiments.

In summary, the results of our study indicate significant correlation between oocyte LD numbers and FA composition in FF. However, gene expression studies on follicular cell gene expression studies are contradictory and do not reflect metabolic pathways emerging from the oocyte to the FF. Nevertheless, the obtained data exhibits differences between cumulus and granulosa cells with regard to gene expression as well as indicates the importance of glucose metabolism in relation to lipid metabolism, which contributes to better understanding of energy metabolism conversion between oocytes and the maternal environment.