Macrophages are the most dominant ovarian leukocytes that can be found in the theca layers of growing follicles and in human FF [20]. Over the last few decades of investigation, it is becoming increasingly evident that macrophages are phenotypically and functionally heterogeneous. In fact, macrophages display high degrees of plasticity in their roles when exposed to various environments. M1 macrophages are polarized by Th1 cytokines, such as IFN-γ, or by recognition of LPS, producing high levels of pro-inflammatory cytokines (TNF-α and IL-1β). Functionally, the pro-inflammatory M1 macrophages are implicated in the clearance of pathogens in the course of infection by generating reactive oxygen/nitrogen species (ROS/RNS), which can induce tissue damage. In contrast, M2 macrophages are polarized by Th2 cytokines, such as IL-4 and/or IL-13, secreting pro-resolving mediators (CCL18, CCL22, and SPMs). Functionally, pro-resolving M2 macrophages show the ability to phagocytose and promote tissue remodelling and wound healing. The heterogeneous and dynamic nature of macrophages in the ovaries implies their proactive involvement in ovarian homeostasis and hormonal control [21].
Considering that M1 and M2 macrophages are associated with inflammatory processes, macrophages polarization may be an important aspect of their interaction with hGCs. Thus, the first aim of this study was to obtain M1 and M2 macrophages phenotypes from the THP-1 human monocyte cell line. The THP-1 is an immortalized cell line isolated from the peripheral blood of a 1-year-old male patient with monocytic leukemia. Resting THP-1 cells retain most of the inflammatory monocytes signalling pathways and exhibit the ability to differentiate into macrophages when stimulated with PMA. For these reasons, this cell line was chosen as an appropriate cell model for obtaining macrophages derived from human monocyte [22].
However, although the THP-1 cell line is widely recognized as a suitable model for assessing macrophage functions and responses to foreign stimuli in vitro, there is currently no standardized protocol for the differentiation of THP-1 monocyte into macrophages using PMA. The lack of a standard protocol has a significant impact on the interpretation of results and comparison of studies. The differentiation protocol used in this study was based on the one previously published by our research group. We initiated THP-1 differentiation by exposing the cells to PMA (25 ng/mL) for 48 h, followed by a 24 h resting period [18]. Differentiation of monocytes into resting M0 was confirmed by observation that macrophage-like cells adhered to tissue culture flask, adopting a stellate morphology. After that, resting M0 were primed for 24 h with fresh medium supplemented with IFN-γ (20 ng/mL) + LPS (10 pg/mL) to differentiate into M1 and by IL-13 (20 ng/mL) for M2. Transcriptional markers previously used to characterize distinct subsets of macrophages were identified in the literature and used in the present study [22, 23]. As stated earlier, a distinctive hallmark of M1 polarization is the high production of pro-inflammatory cytokines, while M2 polarization regulates the production of pro-resolving mediators. Thus, M1 and M2 macrophages have been shown to be distinguished by distinct chemokine arrays. Studies conducted in ex vivo human systems reported that IL-1β and TNF-α genes were expressed in the polarized M1macrophages [24]. Our results showed a very similar array of up-regulated genes in polarized M1macrophages, suggesting a successful differentiation. In turn, the chemokines CCL18 and CCL22 are categorized according to the M2 state [23, 24]. Our findings show that polarized M2 macrophages expressed all the chemokines mentioned above (Fig. 1).
Inflammatory conditions can severely disrupt normal ovarian function and oocyte quality. In vitro models that adequately simulate the ovarian microenvironment may provide a helpful tool to study the mechanisms by which pro-inflammatory M1 and pro-resolving M2 macrophages may affect the hGCs functions. At this point, we established an indirect co-culture model between CMs of either M1 or M2 macrophages and GCs to mimic the ovary in inflamed states. Co-culture models have long been widely used to study fundamental cellular interactions of any kind. These systems are highly relevant for drug discovery as they provide a more representative in vivo model of human tissue than animal models [25]. Indirect co-culture models incorporate a physical barrier between cell types, allowing only signalling through the cell secretome (CS). The CM is commonly employed, where the medium is used first to culture one cell type and then transferred to the second cell type. The CM is constituted by the CS of the first cell type that contains soluble factors and can affect the behaviour of the second cell type in a positive and/or negative way [26].
In the co-culture model adopted in this study, we decided to use only primary human GCs isolated from women who had male factor as a cause of infertility. Exclusion criteria for women with endometriosis or PCOS, for example, were based on eliminating cells obtained from inflammatory-associated pathologies and, therefore, investigating only the contribution of macrophages to physiological ovarian inflammation. An immortalized hGCs line could be helpful in studying many of the processes that are implicated in human follicle development [27]. We chose to employ the COV434 cell line due to its ability to synthesize estradiol from FSH stimulation [28].
Regarding cell viability, we observed that M1 and M2 macrophages affected the viability of both hGCs and COV434 cells, when compared to the resting macrophage (M0), which was used as a control. However, no morphological changes were observed in both cells. As seen in Fig. 2, when hGCs and COV434 cells are exposed to CM of M1 macrophages, a significant decrease in cell viability occurs, suggesting that the pro-inflammatory environment may be influencing cells’ viability. However, it is possible to notice a greater significant difference in the decrease of hGCs viability (p < 0.05) in comparison to COV434 (p < 0.001), which in part can be explained by the fact that hGCs are originated from a previous inflammatory environment, such as ovaries. Furthermore, it is known that pro-inflammatory M1 macrophages can produce ROS/RNS, which can induce apoptosis in hGCs [29]. However, this observation requires further investigation. Thus, the pro-inflammatory environment can affect the ovarian oxidative balance. Regarding cell viability when hGCs and COV434 are exposed to the CM of M2 macrophages, it can be observed that it is affected in a similar statistical proportion for both cells (p < 0.0001) when compared to the M0 control, suggesting that the anti-inflammatory environment provided by M2 macrophages can, to some extent, contribute to cell proliferation.
In the present study, co-culture between CM of polarized M2 macrophages and hGCs or COV434 cells for 72 h induced a significant increase in mRNA and protein levels of COX-2. However, no significant increase was observed when cells were exposed to CM of polarized M1 macrophages (Figs. 4a and 5a). Narki et al. [30] previously reported that the expression of COX-2 in GCs isolated from women undergoing ART was induced by the pro-inflammatory cytokine IL-1β, contrasting our results. From these findings, it is possible to suggest that the induction of COX-2 expression by the CM of M2 may increase the production of pro-resolving mediators by hGCs and COV434 cells, leading to the resolution of inflammation.
Feldam et al. [31] demonstrated for the first time the expression of 5-, 12-, and 15-LOX in GCs isolated from women undergoing ART through the characterization of specific products derived from the metabolism of arachidonic acid (AA), such as 5-, 12-, and 15-hydroxyeicosatetraenoic acids (HETEs). In a recent study, Zhang et al. [8] reported that the pro-resolving mediator resolvin E1 (RvE1) improved oocyte quality by decreasing apoptosis rate of CCs and increasing cell viability and proliferation. Considering that COX-2 and LOXs, both involved in RvE1 production, when stimulated can produce pro-inflammatory and pro-resolving mediators, we also investigated their expression in hGCs and COV434 cells by CMs of M1 and M2. From our data, it is possible to observe that neither CM of M1 nor that of M2 were effective in inducing 5-LOX expression in hGCs, both in terms of RNA and protein levels, when compared to the positive control (PC, placenta homogenate) (Fig. 4b). Regarding COV434 cells, a significant increase in 5-LOX mRNA levels is observed when exposed to CMs of M1 and M2. However, the same was not observed according to the results of western blotting, where a significant increase in 5-LOX protein expression was observed only when the cells were exposed to the CM of M2 (Fig. 5b). Concerning 12-LOX expression, a significant increase in its expression is only observed when hGCs are exposed to the CM of M1 (Fig. 4c). However, 12-LOX expression is not observed when COV434 cells are exposed to CMs of M1 and M2 (Fig. 5c). These results can be explained, at least to some extent, by the fact that the hGCs are originated from an inflammatory environment per se. Thus, increased expression of 12-LOX may be a compensatory way in which cells produce pro-resolving mediators to overcome inflammation. Finally, CMs of M1 and M2, despite 72 h of co-culture with hGCs, were not able to significantly affect the 15-LOX expression when compared to control (M0) (Fig. 4d). This result contrasts with the one published by Liao et al. [32], in which the authors demonstrated that 15-LOX expression was up-regulated in GCs isolated from women with PCOS. In regard to COV434 cells, a significant increase in 15-LOX expression is observed when these cells are also exposed to CMs of M1 and M2 (Fig. 5d).