Serum epinephrine and norepinephrine were significantly elevated in women with infertility and pregnancy failure
Given the strong association between mental stress and infertility, we attempted to test whether the serum levels of stress-related hormones differed in infertile women. Since epinephrine and norepinephrine are crucial stress hormones [22, 23], we tested the levels of these hormones in patients with and without infertility. The levels were found to be significantly higher in women with infertility than in the controls (Fig. 1A, 1B). In patients with infertility problem who undergo the embryo transfer, the serum epinephrine and norepinephrine levels of those with failed embryo implantation were higher than those with pregnancy; and the difference in epinephrine level was particularly significant (Fig. 1C, 1D). Since all the patients were transferred with good quality embryos, human endometrial receptivity dysfunction may a main cause for pregnant failure.
Analysis of the transcriptome data from 7endometrial samples with RIF and 6control samples revealed that 2511 mRNAs were differentially expressed (1260 down-regulated and 1251 up-regulated, log2-fold change>=0.0 and P value≤0.05 as the cut‑off criteria) (Fig. 1E). The detailed information for differential mRNA is presented in Data S1. These results demonstrated that the gene profiles of the endometrium with RIF were significantly different from those of control. Next, we applied GO analysis to predict the functions of aberrantly expressed genes. Of these DEGs, 378 mRNAs were down-regulated, while 240 mRNAs were up-regulated, with |log2‑fold change|≥1.0 and P value≤ 0.05 as the cut‑off criteria. The biological processes of GO are shown in Figure 1F. The up-regulated genes were mainly involved in translation and cell proliferation, and the down-regulated genes were associated with immune regulation and inflammatory response. Immune and inflammatory signals have been confirmed to be important for embryo implantation and early pregnancy [24, 25](Fig. 1F).
The adrenergic receptor ADRA2C was highly expressed in decidualized HESCs
It is estimated that embryo quality accounts for one-third of implantation failures, while unsatisfactory endometrial receptivity and changes in embryo–endometrial dialogue are responsible for the remaining two-thirds [26, 27]. We then assessed whether adrenergic receptor signaling is involved in human endometrial receptivity. We generated heatmaps to reveal the transcriptomic patterns of adrenergic receptors (Fig. 2A), and found that only ADRA2C was highly expressed in the mid-secretory phase. However, when compared with the control group, the transcription level and protein level of ADRA2C in RIF was significantly reduced (Fig. 2A-B). We further checked the expression of ADRA2C in the human endometrium and found that it was significantly enhanced during the endometrial secretory phase (Fig. 2C-E). While investigating the cellular localization of ADRA2C in human endometrium, as shown in Fig. 2C, staining for ADRA2C was low or undetectable in proliferative specimens and positive in the secretory stage. We found that ADRA2C was highly expressed in glandular epithelial cells, luminal epithelial cells and stromal cells in the receptive endometrium.
Since the ADRA2C receptor was highly expressed in stroma cells and decidualization of HESCs is crucial for the establishment of endometrial receptivity, we utilized an immortalized human endometrial stromal cell line, which can be induced to decidualization to employ the function of adrenergic signal for endometrium [15]. We screened the expression profiles of different adrenergic receptors using RT-PCR during decidualization, and only ADRA2C exhibited an enhanced expression pattern during the in vitro decidualization (Fig. 2F-I). Moreover, we detected phenylethanolamine N-methyltransferase (PNMT) and tyrosine hydroxylase (TH), which are key rate-limiting enzymes for the synthesis of adrenergic ligands[28, 29]. The results signified that these adrenergic receptors were present in the human endometrium, however, HESCs could not synthesize epinephrine ligands per se(Supplemental Fig. S1A-B).
Adrenergic Signaling Is Involved In The Decidualization Of Hescs
To elucidate the role of ADRA2C in the decidualization of HESCs, we examined the expression level of classical decidualization markers, including prolactin (PRL) and insulin-like growth factor-binding protein 1 (IGFBP-1), in different concentrations of epinephrine-treated HESCs. We found that only 5.0 µM epinephrine effectively promoted decidualization (Fig. 3A-3B) and that this promotion was time dependent (Fig. 3C-3D).In addition, 10µM adrenaline loss the promoting effects compared with 5µM (Fig. 3A-B, Fig. 3E-F). Since ADRA2C was weakly expressed in undifferentiated HESCs, we used 0–10.0 µM epinephrine for cell growth analysis to evaluate its effect on HESCs culture in vitro. As shown in Supplemental Fig. S2, different concentrations of epinephrine display little effects on the proliferation activity of HESCs. Therefore, we used the exposure dose of 5 µM epinephrine for further studies.
Adrenergic signaling inhibits AKT activity to increase Forkhead box O1 (FOXO1) expression and nuclear localization
The observation that appropriate level of adrenergic signal promotes the HESC decidualization motivate us to explore the underlying mechanism for ADRA2C regulating the normal process of human endometrial stromal differentiation. ADRA2C belongs to G-protein-coupled receptors (GPCRs), and its common downstream activation pathways include PKA, ERK, and AKT [30]. Western blotting was conducted to detect the expression level of these signal molecules. Only the AKT signaling pathway was affected by epinephrine stimulation, and the phosphorylation level of p-AKT (Ser473) was reduced by 5 µM epinephrine stimulation, but not in 10µM group (Fig. 4A). It was also noticed that the ADRA2C receptor as down-regulated in the high (10 µM) epinephrine treated group (Fig. 4A). A wealth of data indicated that AKT pathway activity was involved in decreased FOXO1 protein and blunted the decidual response [31, 32]. Furthermore, as expected, the level of FOXO1 protein increased with the decrease in AKT phosphorylation level (Fig. 4A and 4B). We also assayed other transcription factors such as HOXA10 and STAT3, which were key mediators in decidualization [33–35]. Interestingly, the protein level of factor HOXA10 is increased by epinephrine stimulation but seem not to be affected by epinephrine concentration and the protein level of ADRA2C (Fig. 4B). We found that the change of FOXO1 was affected by epinephrine concentration; moreover, these changes were affected by AKT phosphorylation signal regulation (Fig. 4A and 4B). In addition, 10.0µM epinephrine cannot promote the decidualization as the 5.0µM, consistent with the inability to increase the FOXO1 level. Based on these observations, it seems that the physiological level of epinephrine is benefit for decidualization through its receptor ADRA2C, and aberrantly elevated epinephrine loss of its promoting effects for decidualization. During decidualization, FOXO1 is responsible for transcribing the IGFBP-1 and PRL genes [36, 37]. FOXO1 might be compartmentalized in both the cytoplasm and nucleus. Nonetheless, it must be located in the nucleus to perform transcriptional activity. Hence, we next used western blotting and immune fluorescence experiments to detect the distribution of FOXO1 in the cytoplasm and nucleus of HESCs. Indeed, the nuclear localization of FOXO1 was increased during HESC differentiation under the induction of 5.0µM epinephrine, but not 10.0 µM (Fig. 4C-E). Collectively, these results suggested that ADRA2C activation was an important process in assisting the decidualization of HESCs, and high concentrations of epinephrine restrain the process of HESCs decidualization.
Aberrant Adra2c Signaling Hampered Hescs Decidualization
To verify whether the clearance of the accumulated ADRA2C would reverse the effects of assisting in HESCs decidualization, we employed a small interference RNA (siRNA) technique to knockdown the expression of ADRA2C in HESCs. As shown in Fig. 5A-B and Supplemental Fig. S3A, during the decidualization process, ADRA2C-siRNA largely reduced ADRA2C expression at both the mRNA and protein levels in the HESCs. This decreased expression can partially eliminate the synergistic effect of adrenergic signals on the decidualization of HESCs. Real-time quantitative PCR was used to evaluate the decidualization markers IGFBP-1 and PRL (Fig. 5C-D). Consistent with this observation, knockdown of ADRA2C expression also largely enhanced the activity of AKT signaling. This increase lowered the level of FOXO1 protein, which attenuated the effects of adrenergic signaling to promote the decidualization of HESCs (Fig. 5E). Then, we tested whether ADRA2C knockdown would reduce the nuclear localization of FOXO1. As shown in Fig. 5F and Supplemental Fig. S3B, when compared with the control group, the level of FOXO1 protein and the localization of FOXO1 in the nucleus of HESCs were both reduced after ADRA2C knockdown. These findings further strengthen the view that ADRA2C signaling and HESCs decidualization were functionally associated. Overall, these experiments suggest that epinephrine could promote HESC decidualization by inhibiting the AKT activity to increase FOXO1 expression and nuclear localization via ADRA2C (Fig. 5G).
Altered expression of ADRA2C and FOXO1 in the endometrium during the implantation window from RIF patients
Our previous study found that when compared with the endometrial proliferation phase, the level of ADRA2C protein in the endometrial secretory phase was increased. Moreover, we found that an appropriate dose of epinephrine could promote the differentiation of HESCs. Interestingly, we discovered that high concentrations of epinephrine did not promote the differentiation of HESCs in vitro, but down-regulate the expression of its receptor ADRA2C (Fig. 4B). Since the epinephrine level of the RIF group was higher than that of the control group, we checked whether endometrium from the mid-luteal phase of the menstrual cycle displayed the similar alternation as observed in vitro. We identified that ADRA2C was involved in significantly differentially down-regulated genes, as displayed in the volcano map (Fig. 6A). Next, we used pathway analyses to identify the significant pathways associated with the DEGs according to Kyoto Encyclopedia of Genes and Genomes (KEGG) (Fig. 6B). KEGG pathway analysis revealed that Akt signaling was involved in the affected pathway among the up-regulated genes (Fig. 6B). In addition, WB data showed that the protein level of ADRA2C in the mid-luteal phase of the menstrual cycle from RIF patients was lower than that of controls (Fig. 6C), and expression and distribution patterns of ADRA2C and FOXO1 in the endometrium of RIF patients were more heterogeneous compared with those in controls (Fig. 6D-E). We also noted that the expression patterns of ADRA2C, FOXO1, and PR, which were considered to be the genes related to endometrial receptivity [37, 38], were more consistent in terms of the expression level and spatial location in the normal individuals (Fig. S4A-B). Therefore, consistent with these findings, we hypothesized that serum epinephrine levels may alter the endometrial receptivity by affecting the expression of adrenal receptors and thus the downstream signal cascade involved AKT-FOXO1 in the endometrium.
High epinephrine down regulated the uterine adrenergic receptor and compromised the function of decidual cells in mice
To further verify the detrimental effects of high level of adrenergic ligand on decidualization and early pregnancy in vivo, a mouse stress model was utilized (Fig. 7A). Stress-induced mice had increased serum epinephrine levels (Fig. 7B-C), significantly decreased mating rate (Fig. 7D), and poor embryonic development (Fig. 7E-F). As shown in Fig. 7F, the arrow points out growth-restricted implantation sites on D6 and D8 of stressed mouse models. Our previous studies had confirmed that mouse decidual cells in early pregnancy express Adra1b, Adra2b, and Adrb2 receptors [15]. We speculated that these impairments of decidualization may be related to abnormal adrenergic receptor signaling. Subsequently, through in situ and quantitative PCR detection, we found that the expressions of these adrenergic receptors in the decidual cells of the stressed mice were significantly reduced (Fig. 7K-O). Moreover, exogenous epinephrine decreased adrenergic receptor expression in uterine decidual cells during early pregnancy (Fig. 7K-N). We use the artificially induced decidualization mouse model (Fig. 7H) and found that exogenous epinephrine exposure can impair the decidual differentiation of endometrial stromal cells (Fig. 7I-J). This indicated that the significantly decreased expression of these receptors was not caused by abnormal embryonic development. These interesting findings suggest that elevated epinephrine in vivo impaired adrenergic receptor signaling, disrupted the decidualization of uterine stromal cells leading to early pregnancy loss.