A growing number of studies describe that an excess amount of NEFAs disrupts the function of GCs during metabolic diseases. One of the apparent effects of elevated NEFA levels is the induction of dramatic morphological alterations in GCs. In addition to the significant impact on the expression of critical genes and hormone production, Yenuganti et al. 2016 [
33] and Sharma et al. 2019 [
30] reported the formation of foam cell-like structures in cultured bovine GCs upon treatment with increased levels of NEFAs such as OA, PA, and SA; the formation was possibly the result of excessive lipid droplet accumulation. The altered morphology of GCs has been attributed to the active uptake of FAs by GCs via the CD36 translocase system. CD36 is an 88-kDa transmembrane glycoprotein with lipid-based ligand binding for low-density lipoproteins (LDL), high-density lipoproteins (HDL), very low-density lipoproteins (VLDL), fibrillar β-amyloid, and collagen molecules. Expression of CD36 induces the accumulation of conjugated linoleic acids
(CLA) in human macrophages and 15P-1 cell lines (testicular cells) of mice [
34,
35]. Similarly, CD36 protein expression is increased upon feeding ALA enriched diets to rats promoting the transport of lipids into their resting skeletal muscles [
36]. This information suggests that CD36 is actively involved in the uptake of FAs in mammalian cells. Upon cellular uptake, different FAs appeared to elicit different responses, depending on the degree of saturation/unsaturation of FAs.
Saturated fatty acids
Accumulating evidence indicates that SFAs have adverse effects on GC function. PA and SA have been found to induce apoptosis in cultured human primary GCs and in human KGN cells in a dose-dependent manner [
37]. Particularly, at a concentration of 300 μM, PA was found to cause a significant decline in the expression of the anti-apoptotic protein BCL2 (B-cell lymphoma 2) and an increase in the expression of BAX (BCL2 associated X protein), a pro-apoptotic protein, in human primary GCs. Distinct DNA fragmentation in PA (300 μM) treated GCs confirmed the apoptotic effects of SFAs. However, PA had no apoptotic effects at a concentration of 100 μM, whereas SA could still pose a marginal yet significant decrease in cell viability at 100 μM, which became very severe as the concentration of SFAs was increased. This indicates that elevated concentrations (> 100 μM) of SFAs are cytotoxic to human GCs. Similar negative effects of SFAs have been reported in mice, cows, and pigs [
30,
38]. Shibahara et al. 2020 [
39] have recently reconfirmed the proapoptotic effects of SFAs as they observed induced expression of caspase-3 and C/EBP homologous
protein (CHOP) and decreasing Akt phosphorylation in porcine GCs. These effects were observed in association with reduced cell viability and increased ceramide accumulation.
FAs are well-known ligands for peroxisomal proliferator-activated receptors (PPARs) in mammalian cells. In an interesting experiment, Mu et al. 2001 [
37] induced PPAR signaling with a synthetic FA analog, fenofibrate. The results revealed that fenofibrate could not induce apoptosis in human GCs, suggesting that FA induced apoptosis may not be the result of direct interactions of FAs with PPAR in GCs. Interestingly, cellular apoptosis seems to be mediated by metabolites of SFAs as shown in chicken and human GCs. PA-induced apoptosis of GCs is accompanied by increased expression of genes coding for various enzymes of the FA and lipid metabolism, including carnitine palmitoyl transferase-1 (CPT1), serine palmitoyltransferase (SPT), acyl CoA oxidase (ACO), and sphingomyelinase (SMASE) [
40]. Therefore, the effects of different chemical inhibitors that could inhibit these enzymes were tested to identify the mechanisms involved in FA induced cell death in GCs. Surprisingly, independent supplementation with triacsin C (fatty acyl CoA synthase inhibitor), imipramine (sphingomyalinase inhibitor), fumonisin B1 (ceramide synthesis inhibitor) and pyrrolidine dithiocarbamate (free radicle scavenger) rescued PA-induced cell death in chicken GCs [
40]. Similarly, triacsin-C supplementation was found to significantly decrease both PA and SA induced cell death in human GCs [
37]. However, treatment with fumonisin B1 and the nitric oxide synthase inhibitor aminoguanidine could not inhibit PA-induced apoptosis in human GCs. These above studies suggest that especially acylated forms of FAs, which are precursor forms of FA oxidation, are detrimental for cell health. Further validation of this phenomenon was provided when human GCs were treated with acyl derivates of different SFAs and UFAs. Similar to the effects of PA and SA, palmitoyl CoA and stearoyl CoA significantly decreased cell viability in a dose-dependent manner [
37]. All these observations indicate that excessive β-oxidation of SFAs could be a potential inducer of apoptosis and cell death, possibly by generating an excessive amount of ROS in GCs of different species.
However, PA and SA could not elicit early apoptotic effects as shown by an Annexin V assay in bovine GCs which might be due to increased transcription of FSH signaling induced genes such as cytochrome P450 family 19 subfamily A member 1 (
CYP19A1), follicle stimulating hormone receptor (
FSHR), luteinizing hormone/choriogonadotropin receptor (
LHCGR) as well as enhanced synthesis of E2 by PA and SA [
6,
30]. It has been shown that higher E2 levels protect the cells from Fas ligand-mediated apoptosis and induce proliferation by increasing the percentage of cells entering the S phase of the cell cycle [
41]. Similar upregulation of steroidogenesis has been reported in bovine adrenal cells exposed to SFAs [
42]. However, Vanholder et al. 2005 [
6] showed in bovine GCs that both PA and SA could significantly increase the number of dead cells under in vitro conditions. These contradictory observations can be very likely explained by the opposing effects of SFAs and E2 on apoptosis. On the one hand, apoptosis can be promoted by PA and SA treatment, but on the other hand, these detrimental effects of SFAs can be overridden by increased E2 production in E2 active GC. Therefore, the E2 active state of GCs must be acknowledged while discussing the effects of FA.
SFAs were also found to have detrimental effects on bovine cumulus cells, which are distinct from mural GCs and have a relatively lower steroidogenic ability. In vitro exposure of bovine cumulus-oocyte complexes (COCs) to PA and SA induces endoplasmic reticulum (ER) stress, mitochondrial damage and apoptosis in cumulus cells [
43,
44]. ER stress in cumulus cells has been determined by measuring the gene expression of
ATF4 (activating transcription factor 4) and
HSPA5 (heat shock protein family A (Hsp70) member 5) [
43]. The expression of
ATF4 and
HSPA5 genes was also up-regulated in mouse COCs exposed to the lipid-rich follicular fluid compared to lipid-deprived follicular fluid [
26].
IGF-1 signaling plays a key role in GC steroidogenesis and proliferation [
45]. More recently, it was observed that PA inhibits glucose uptake and induces insulin resistance in KGN cells by inhibiting IGF-1 induced Akt phosphorylation while increasing c-Jun N-terminal kinase (JNK) phosphorylation [
46] (Fig.
2). Furthermore, the inhibition of JNK phosphorylation reversed the PA-induced downregulation of Akt phosphorylation. These results are in line with earlier observations made by Walsh et al. 2012 [
47] that NEB like conditions such as acute dietary restriction, which elevate NEFA levels, could affect the IGF-1 and gonadotropin signaling associated gene expression in GCs, leading to an anovulatory phenotype in animals.
Overall, it is apparent that the metabolites of SFAs could induce apoptosis in GCs of different species, which may be ameliorated by inhibitors of FA metabolism and antioxidant molecules. In any case, the E2 active phenotype of GCs may play a key role in the apoptotic response of GCs to elevated SFAs.
Unsaturated fatty acids
Similar to SFAs, UFAs are also known inducers of fatty acid transporters,
CD36, and solute carrier family 27 member 1 in GCs, thus promoting their active uptake [
33]. However, in contrast, UFAs show mostly opposite biological effects on GC function compared to SFAs. It has been shown that UFAs can counteract the cytotoxic effects of SFAs by stimulating triacyl glyceride formation, which leads to a reduction of SFA concentrations available for oxidative processes, the byproducts of which are primarily cytotoxic [
40,
48]. However, MUFAs such as OA have been shown to cause a significant decline in the proliferation of in vitro cultured GCs of both human and bovine origin [
6,
28,
37]. On the other hand, Sharma et al. 2019 [
30] reported upregulation of the proliferation marker cyclin D2 (CCND2) in bovine GCs by SFAs such as PA and SA. These effects clearly suggest that different, and perhaps opposite signaling mechanisms are influenced by SFAs and UFAs. This can be partially explained by data from rat skeletal muscle cells [
49], which showed that OA could induce Akt phosphorylation and also reverse the PA-induced dephosphorylation of Akt by activating the PI3K (phosphoinositide 3-kinase-3-kinase) pathway in skeletal muscle cells. However, the downstream pathways of Akt such as forkhead box protein O1 (FOXO1),
glycogen synthase kinase 3 (GSK3) and mammalian target of rapamycin (mTOR), which could hold the key to explaining these observed effects, have not yet been studied with respect to UFAs.
OA and LA have been found to elicit only a marginal apoptotic response at a concentration of 300 μM, but unlike SFAs, they have no effect at lower concentrations. AA (arachidonic acid; 20:4) could not induce any apoptotic response within its physiological range (1 to 10 μM) in human GCs [
37]. Furthermore, Valckx et al. 2014 [
12] revealed that OA could reduce the expression of
BAX and induce
BCL2 and
Gadd45b (growth arrest and DNA damage-inducible beta) expression in human GCs. A recent report revealed that AA could promote the survival of bovine GCs at lower concentrations (50 μM) but had detrimental effects at higher concentrations (200 μM).
OA downregulates the mRNA expression of different genes associated with FSH and LH signaling in bovine GCs such as steroidogenic acute regulatory protein (
STAR),
CYP19A1, FSHR, LHCGR, cytochrome P450 family 11 sub-family A member 1 (
CYP11A1), hydroxy-delta-5-steroid dehydrogenase, 3 beta- and steroid delta-isomerase 1 (
HSD3B1),
CCND2 and proliferating cell nuclear antigen (
PCNA), which are induced by SFAs [
30]. Estrogen receptor 2 (
ESR2), follistatin (
FST), and
PPARG are among the other genes that were down-regulated by OA in these cells [
50]. However, the protein abundance of the above genes has yet to be determined. Similar to OA, CLA, an isomer of LA has been found to affect FSH and IGF-1 induced steroidogenesis by down-regulating the transcription of
CYP19A1, insulin-like growth factor 1 receptor (
IGFR1) and
GATA4 genes [
51]. These observed effects were attributed to decreased levels of Akt phosphorylation and increased protein abundance of PPARG and phosphatase and tensin homolog (PTEN) by CLA in buffalo GCs [
51]. It was reported that PPARG activation decreased E2 production by inducing the ubiquitination of cyclin D1 and estrogen receptor α, and that it inhibited the expression of
CYP19A1 through nuclear factor-kappaB (NFkB) activation [
52,
53]. Nevertheless, decreased E2 levels by both CLA and OA suggest that UFAs are detrimental to female steroid production during metabolic stress conditions. However, the opposite regulatory effects on Akt phosphorylation by OA and CLA is very intriguing for future studies to understand the importance of Akt signaling concerning different FAs.
It seems that UFAs play an important role in PCOS pathophysiology, whose etiology is largely attributed to high levels of androgens [
54]. Elevated circulatory concentrations of OA were correlated with the adverse pregnancy outcomes in obese women with PCOS undergoing controlled ovarian hyperstimulation [
14]. Huang et al. 2018 [
55] showed that PCOS affects the metabolism of UFAs in rats. In particular, metabolites of AA generated by the cyclooxygenase enzyme were significantly increased in the ovaries of PCOS rats. These AA derived metabolites could modulate the ovarian cycle and induce luteolysis. Niu et al. 2014 [
14] showed that levels of OA in the follicular fluid were significantly higher in obese PCOS patients compared to obese women, indicating a role of OA in PCOS. More critical cues on the role of OA can be derived from the regulation of forkhead transcriptional regulator (FOXL2) expression. OA has been shown to have a dose dependent effect on both mRNA and protein expression of FOXL2 in FSH treated cultured bovine GCs [
33] (Fig.
2). FOXL2 is vital for the biogenesis of female steroids and is required to prevent the trans-differentiation of the ovary into testis, thus preventing androgen production, as shown in mice [
56]. Uhlenhaut et al. 2009 [
57] showed that the deletion of FOXL2 immediately increases the expression of testes-specific SRY-box transcription factor 9 (SOX9) in the mouse ovary and as a consequence, leads to comparable testosterone production as in male littermates. In agreement with this earlier report, OA supplementation promoted SOX9 expression in bovine GCs and inhibited E2 biosynthesis [
50]. It has been shown that serum anti Mullerian hormone (AMH) levels were increased in patients with PCOS [
58]. However, no correlation was observed with respect to circulatory NEFAs and AMH in cows and humans [
59,
60]. In any case, future experiments are needed to identify the effects of different NEFAs on AMH production by GCs.
A few reports have shown the effects of UFAs on progesterone (P4) production under serum-supplemented conditions. In goats, OA and LA, but not ALA, could induce P4 production by increasing the phosphorylation of the mitogen-activated protein kinase, ERK1/2 [
61]. Supplementation with the ERK1/2 inhibitor U0126 decreased the OA and LA induced P4 production. However, these results indicate that OA and LA may induce premature luteinization of GCs. Zhang et al. 2019 [
62] recently reported effects of AA on bovine GCs. It is clear from their data that AA induces ERK and Akt phosphorylation similar to OA and LA. Furthermore, the expression of different marker genes such as
CYP19A1, FSHR, HSD3B1 and
STAR, and the production of E2 were decreased by AA in a dose-dependent manner. Interestingly, AA dramatically increased the production of P4, which is quite unexpected considering the downregulation of
HSD3B1. This indicates that the correlation of
HSD3B1 gene expression and P4 progesterone production may not be stringent at the transcriptional level [
62]. On the other hand, unlike in goats, OA supplementation did not affect P4 production in bovine GCs cultured with 10% fetal calf serum (FCS) [
28]. These opposite effects might be partly attributed to species differences and to the addition of serum in the culture media, whereas GCs inside the ovarian follicle have no direct contact with blood serum due to the non-vascularized GCs layer. It is well known that IGF-1 and FSH signaling plays a vital role in GCs proliferation via the protein kinase A (PKA) and phosphoinositide 3-kinase (PI3K) pathways. It is reported in goats that OA could affect the IGF-1 but not the FSH-induced proliferation of GCs under serum-supplemented conditions.
Overall, unlike for SFAs, it is not completely clear whether UFA metabolites can induce apoptosis. However, multiple reports state that OA, CLA, and AA can down-regulate E2 production and perhaps also the proliferation of GCs. Furthermore, MUFAs could play a role in the etiology of PCOS as they could promote androgen biosynthesis. Studies are yet to be performed to clarify the role of other UFAs such as ALA on GC function.