Abstract

1. Introduction

Progestin-only long-acting reversible contraceptives (pLARCs), e.g., injectable form of depo-medroxyprogesterone acetate (DMPA), subdermal implants releasing etonogestrel (ETO), and intrauterine devices releasing levonorgestrel (LNG), are safe and highly effective in averting unwanted pregnancies, particularly among adolescents, and preterm births by prolonging inter-pregnancy intervals [1,2]. The use of pLARCs is preferred in women in whom estrogen-containing formulations are contraindicated, e.g., during lactation or with an estrogen-dependent tumor [3,4]. Their safety, affordability, and prolonged efficacy make pLARCs particularly ideal for use in countries with limited healthcare access [5]. However, abnormal uterine bleeding (AUB) in pLARC users is the major discontinuation reason [6,7].

Progesterone withdrawal with uniform reductions in endometrial hemostasis and spiral arterial vasospasm is the cause of menstrual bleeding, while pLARC-induced AUB occurs intermittently and focally from irregularly distributed hyperdilated, superficial, and fragile microvessels [8,9]. Our prior studies in pLARC-administered women or guinea pigs showed that pLARCs reduced uterine blood flow, causing endometrial hypoxia and generating reactive oxygen species [10,11,12], which cause AUB directly by damaging endometrial microvessels and indirectly by inducing aberrant angiogenesis [13]. The microscopic evaluation of endometria from pLARC users revealed that only bleeding sites displayed abnormal microvessels enmeshed in a compromised extracellular matrix, which was contiguous with decidualized human endometrial stromal cells (HESCs). Decidualized HESCs express elevated levels of tissue factor (encoded by F3 gene), the primary initiator of hemostasis via Factor-VIIa- and Factor-Xa/Va-generated thrombin [14,15]. Thus, pLARC-induced vascular damage generates excess thrombin by delivering circulating Factor VII to decidualized HESC-expressed tissue factor. Thrombin is well documented to prevent bleeding by activating platelets and generating fibrin. However, thrombin also displays distinct cellular effects by binding to protease-activated receptors. Thrombin: (1) paradoxically promotes endothelial permeability and focal hemorrhage [16,17,18]; (2) induces aberrant angiogenesis and inflammation by generating vascular endothelial growth factor (VEGF) [19] and interleukin–8 (IL–8) levels, respectively, in HESCs [20]; (3) enhances the expression of HESC-derived matrix metalloproteinase-1 and -3 [21,22]. Therefore, the combined actions of thrombin-mediated angiogenic, proteolytic, and inflammatory processes markedly contribute to pLARC-induced AUB [20,23].

The zinc finger and BTB domain-containing 16 (ZBTB16) protein, also known as the promyelocytic leukemia zinc finger, contains several zinc finger domains, a BTB domain, and three RD2 subdomains [24,25]. The zinc finger domains at the C-terminus bind either directly to a specific sequence in the genome or interact with other transcription factors, thereby controlling the transcriptional specificity of ZBTB16 [25]. In contrast, the BTB domain at the N-terminus mediates the formation of homo- and/or hetero-dimers with other ZBTB proteins. Therefore, the BTB domain plays a critical role in the stability, localization, and transcriptional activity of ZBTB16 [25] and is also involved in chromatin remodeling by binding to different corepressors and histone modification enzymes [26,27]. ZBTB16 is expressed by several tissues/cell types, including neuronal, muscle, hematopoietic, respiratory, and reproductive cells [28], and regulates many biological processes, including stem cell renewal, proliferation, cell cycle regulation, differentiation, and apoptosis [28,29,30,31]. ZBTB16 exerts these actions as a transcription factor and/or chromatin remodeler to repress and/or activate gene expression in a cell-type-specific or stimulation-dependent manner [32,33,34]. Previous studies showed that ZBTB16 is induced by glucocorticoid and progesterone in HESCs [35] and plays a crucial role in HESC decidualization by regulating the expression of the early growth response 1 gene [36].

Although several mechanisms are involved in the pathogenesis of AUB [13,37,38,39], a gap exists as to precisely how pLARCs trigger AUB by altering endometrial cellular and/or molecular responses. Our previous whole-genome microarray analysis identified ZBTB16 as one of the most-highly induced genes by both medroxyprogesterone acetate (MPA) and ETO in primary cultured HESCs [40]. Thus, we hypothesized that pLARC-induced ZBTB16 expression contributes to HESC decidualization, whereas prolonged enhancement of ZBTB16 levels triggers an inflammatory milieu by inducing pro-inflammatory gene expression and tissue-factor-mediated thrombin generation in decidualized HESCs. This hypothesis was tested by investigating the in situ regulation of ZBTB16 levels in endometrial sections obtained from pLARC-administrated women or guinea pigs. Subsequently, we evaluated in vitro pLARC-mediated regulation of ZBTB16 in HESCs and human endometrial endothelial cells (HEECs), as well as the functional role of ZBTB16 in cultured HESCs to gain insight into the underlying mechanism of pLARC-induced inflammation and AUB.

2. Results

2.1. Increased Endometrial ZBTB16 Levels in Women Receiving DMPA Treatment

The immunohistochemical analysis of paired endometrial sections (n = 7) obtained from women pre- and post-DMPA administration revealed that ZBTB16 is expressed by endometrial stromal, glandular epithelial, and endothelial cells (Figure 1A). Moreover, strong ZBTB16 immunoreactivity was observed in post-DMPA-administered endometrial stromal and glandular cells, as well as superficial microvascular endothelial cells, whereas pre-DMPA endometria displayed significantly less ZBTB16 immunostaining, which was primarily localized to stromal cells (Figure 1A). HSCORE analysis (Figure 1B) confirmed significantly higher ZBTB16 levels in endometrial stromal cells (mean ± SEM; 185.4 ± 9.2 vs. 92.1 ± 13.8; p < 0.001), glandular cells (181.3 ± 20.4 vs. 47.7 ± 14.7; p < 0.001), and endothelial cells (185.3 ± 19.9 vs. 45.4 ± 16.3; p < 0.001) in post-DMPA vs. pre-DMPA administration.

Open in a separate window

Figure 1

Endometrial stromal and glandular cells display enhanced ZBTB16 immunoreactivity in women administered depo-medroxyprogesterone acetate (DMPA). (A) Representative ZBTB16 immunostaining (brown) in paraffin sections from paired endometria obtained from pre- and 3 months post-DMPA administration. Original magnification ×40. Enhanced ZBTB16 immunoreactivity was observed in post-DMPA endometrial compared to pre-DMPA. A strong ZBTB16 immunoreactivity is seen in post-DPMA vascular endothelium (arrows) in the inset micrographs. Original magnification ×100. (B) HSCOREs for ZBTB16 immunoreactivity confirmed significantly higher ZBTB16 immunoreactivity in both endometrial stromal (185.4 ± 9.2 vs. 92.1 ± 13.8) and glandular cells (181.3 ± 20.4 vs. 47.7 ± 14.7), as well as endothelial cells (185.3 ± 19.9 vs. 45.4 ± 16.3; p < 0.001) in post- vs. pre-DMPA endometria. Bars represent the mean ± SEM; n = 7/each; *** p < 0.001 vs. pre-DMPA analyzed by the t-test.

2.2. Elevated Endometrial ZBTB16 Levels in Guinea Pigs Treated with MPA

ZBTB16 immunostaining analysis of endometrial sections of guinea pigs treated 21 days with placebo (control; n = 5), or estradiol (E₂; n = 4), or MPA (n = 5), or E₂ + MPA (n = 6) indicated that, compared to placebo (63.6 ± 14.1): (1) neither E₂ (48.3 ± 11.6; p = 0.56) nor MPA (105.0 ± 11.0; p = 0.11) significantly altered ZBTB16 levels; and (2) E₂ + MPA significantly enhanced ZBTB16 immunoreactivity (128.8 ± 22.5; p < 0.05; Figure 2A,B) in endometrial stromal cells. In contrast, in endometrial glandular epithelial cells, significantly higher ZBTB16 expression was observed in MPA-treated endometria (153.3 ± 15.6; p < 0.05), but not following E₂ + MPA (74.1 ± 18.5) or E₂ treatment (67.0 ± 29.8) vs. the control (90.4 ± 18.6; Figure 2A,B). These results indicated differential regulation of ZBTB16 by steroids in stromal vs. epithelial cells in the guinea pig endometrium.

Open in a separate window

Figure 2

MPA administration induces endometrial ZBTB16 expression in guinea pigs. (A) Representative images of ZBTB16 immunostaining in endometria obtained from ovariectomized guinea pigs treated 21 days with placebo as the control (Cont; n = 5), or estradiol (E₂; n = 4), or medroxyprogesterone acetate (MPA; n = 5), or E₂ + MPA (n = 6). Original magnification: ×20. The insert represents negative control staining. (B) HSCORE analysis of ZBTB16 immunoreactivity in stromal and glandular cells of endometria. Bars represent the mean ± SEM; * p < 0.05 vs. the control or E₂ treatment in stromal cells, and * p < 0.05 vs. the control or E₂ treatment in glandular epithelial cells analyzed by one-way ANOVA followed by the Holm–Sidak method.

2.3. Increased ZBTB16 Levels during In Vitro Decidualization of Primary Cultured HESCs

We next evaluated ZBTB16 mRNA levels during in vitro decidualization of HESCs treated with 10⁻⁸ M E₂ + 10⁻⁷ M MPA + 5 × 10⁻⁵ M cyclic AMP (i.e., EMC) for 3 or 6 days. EMC treatment significantly increased ZBTB16 mRNA levels in HESCs by ~200- and 400-fold at 3 and 6 days of decidualization, respectively, vs. Day 0 of induction (Figure 3A; p < 0.001). In addition to verifying decidualization-related morphological changes of stromal fibroblasts to epithelioid-like decidual cells, the well-documented decidualization markers, prolactin (PRL) and insulin-like growth factor binding protein 1 (IGFBP1) levels [41], were also measured by qPCR to confirm in vitro decidualization. Compared to their levels on Day 0, EMC treatment significantly induced both PRL and IGFBP1 mRNA levels by up to 2-fold on Day 3 with a further increase observed for both PRL (by 10-fold) and IGFBP1 (by 28-fold) during HESC decidualization on Day 6 (Figure S1; p < 0.05).

Open in a separate window

Figure 3

Increased ZBTB16 expression during decidualization of cultured HESCs. (A) ZBTB16 mRNA levels in HESCs treated with 10⁻⁸ M E₂ + 10⁻⁷ MPA + 5 × 10⁻⁵ cyclic AMP (EMC) for 0, 3, or 6 days by qPCR. The data represent the fold change as the mean ± SEM; n = 5/each; *** p < 0.001 vs. Day 0 analyzed by one-way ANOVA followed by the Student–Newman–Keuls method. (B) MPA mediated upregulation of ZBTB16 mRNA levels in HESCs treated with vehicle (control; Cont) or E₂ or MPA or cAMP for 6 days. Bars represent mean ± SEM; n = 4/each; * p < 0.05 vs. Cont or E₂ or cAMP analyzed by one way ANOVA followed by Student-Newman-Keuls method.

To determine the individual contribution of the EMC components to HESC ZBTB16 expression, parallel experiments were performed in HESCs treated with either the control, E₂, cAMP, or MPA for 6 days. Compared to the control, neither E₂ nor cAMP treatment induced ZBTB16 mRNA levels, whereas MPA treatment significantly enhanced ZBTB16 levels by ~200-fold (Figure 3B; p < 0.05), confirming MPA as a primary EMC component that upregulates ZBTB16 mRNA in HESCs.

2.4. In Vitro Comparison of Induction of ZBTB16 Levels by Different pLARCs

ZBTB16 mRNA and protein levels were measured by qPCR and immunoblotting, respectively, in HESCs treated with 10⁻⁸ M E₂ as the control ± various progestins used in different pLARC formulations including 10⁻⁷ M ETO, LNG, and the mixed progestin/glucocorticoid agonist, MPA, as well as a pure glucocorticoid, dexamethasone (DEX), and a pure progestin, Organon 2058 (ORG), for 7 days. This analysis revealed that all treatments significantly increased ZBTB16 mRNA levels compared to the E₂ treatment (Figure 4A; p < 0.05). Among these treatments, DEX was the most-effective (~280 fold), whereas ORG was the least-effective (~8-fold) inducer of ZBTB16 levels in HESCs. However, based on the fold increases of ZBTB16 levels by individual progestins used in pLARCs, MPA was the strongest inducer, increasing ZBTB16 mRNA levels by 150-fold, followed by ETO and LNG treatment, which increased ZBTB16 by 47- and 15-fold, respectively (Figure 4A; p < 0.05). Immunoblot analysis confirmed parallel changes in ZBTB16 protein levels. Thus, compared to the E₂-treated control, each treatment significantly elevated immunoreactive ZBTBP16 levels in HESCs, with DEX-treated HESCs displaying the greatest induction, followed by MPA, then ETO, LNG, and ORG by ~17, 9, 6, and 4-fold, respectively (Figure 4B; p < 0.05).

Open in a separate window

Figure 4

pLARCs induce ZBTB16 mRNA and protein levels in cultured HESCs. ZBTB16 mRNA (A) and protein (B) levels were analyzed by qPCR and immunoblotting, respectively, in HESCs treated with 10⁻⁸ M E₂ ± 10⁻⁷ M ORG, ETO, LNG, MPA, or DEX for 7 days. Bars represent the mean ± SEM; (A) n = 4 for mRNA fold change; * p < 0.05 vs. E₂ alone; and (B) n = 3 for protein levels after normalization to β-actin; * p < 0.05 vs. E₂ alone analyzed by one-way ANOVA followed by the Student–Newman–Keuls method. E₂: estradiol, ORG: Organon 2058, ETO: etonogestrel, LNG: levonorgestrel, MPA: medroxyprogesterone acetate, DEX: dexamethasone. (C) NR3C1, ZBTB16, and FKBP5 mRNA levels in HESCs transfected with either the nonspecific control (Cont) or NR3C1-specific siRNA and treated with EMC for 3 days. Bars represent the mean ± SEM; n = 3; *** p < 0.001 vs. control siRNA by the t-test.

To determine if pLARC-mediated ZBTB16 upregulation is primarily glucocorticoid-receptor-dependent, HESCs were transfected with either scramble, nonspecific control siRNA, or the glucocorticoid receptor encoded by NR3C1 gene-specific siRNA, then treated with EMC for 3 days. After confirming that the NR3C1 mRNA levels were significantly downregulated by the NR3C1 siRNA compared to control siRNA transfection (0.09 ± 0.01 vs. 1.02 ± 0.002; p < 0.001), NR3C1-siRNA-transfected HECSs displayed significantly reduced ZBTB16 expression compared to the control (0.08 ± 0.01 vs. 1.03 ± 0.003; p < 0.001). The reduction in glucocorticoid receptor activity was confirmed by measuring the levels of FKBP5, which is a well-documented glucocorticoid-receptor-regulated gene (0.27 ± 0.04 vs. 1.01 ± 0.01; p < 0.001; Figure 4C). Thus, pLARC-mediated ZBTB16 upregulation is primarily glucocorticoid-receptor-dependent.

In addition to HESCs, cultured HEECs were also treated with E₂ alone ± ORG, or ETO, or LNG, or MPA, or DEX to further confirm glucocorticoid-receptor-mediated ZBTB16 upregulation in endothelial cells since HEECs express the glucocorticoid receptor [42], but not the progesterone receptor [43,44]. The evaluation of ZBTB16 levels in HEECs by qPCR showed that only MPA and DEX significantly induced ZBTB16 mRNA levels by ~28- and 80-fold, respectively, compared to the E₂-treated control group (Figure S2; p < 0.05), whereas ETO- or LNG-induced changes did not attain statistical significance, and ORG as a pure progestin did not alter ZBTB16 expression in HEEC cultures (Figure S2).

2.5. ZBTB16 Overexpression Increases Expression of Decidualization Markers in HESCs

To investigate the functional role of ZBTB16 on pLARC-induced decidualization, HESCs were transiently transfected with either the ZBTB16 expression vector or an empty vector as the control. Following transfection, the cultures were incubated with or without EMC for 3 days. As shown in Figure S3, the qPCR results confirmed that, compared to control vector transfection, the ZBTB16 levels were significantly increased in ZBTB16-transfected HESCs. However, EMC did not significantly increase ZBTB16 in transfected cells (p = 0.19; Figure S3A).

The decidualization markers’, PRL and IGFBP1, levels were also measured in the control or ZBTB16-transfected HESCs. The PRL and IGFBP1 levels were 1.9- and 3.4-fold higher in ZBTB16-vector-transfected cells vs. control-vector-transfected cells in the absence of EMC. In control-transfected cells, EMC treatment also significantly increased both PRL and IGFBP1 levels by 3.4- or 2.2-fold, respectively (Figure 5A). EMC treatment further enhanced PRL (6.8 ± 2.2 vs. 3.4 ± 1.1; p < 0.05) and IGFBP1 levels (6.0 ± 1.2 vs. 2.2 ± 0.4; p < 0.05) in ZBTB16-transfected vs. control-transfected HESCs (Figure 5A).

Open in a separate window

Figure 5

ZBTB16 overexpression induces decidualization markers, as well as F3, IL-8, and PTGS2 levels in HESCs. (A) Decidualization markers’, PRL and IGFBP1, mRNA levels in ZBTB16-vector (ZBTB16-v) or control-vector (Cont-v)-transfected HESCs treated with or without 10⁻⁸ M estradiol + 10⁻⁷ M medroxyprogesterone acetate + 5 × 10⁻⁵ M cAMP (EMC) for 3 days. Bars represent the mean ± SEM; n = 4; *** p < 0.05 vs. control-v-EMC; ⁺ p < 0.05 vs. ZBTB16-v + EMC. (B) Tissue factor (F3) mRNA levels in control- (Cont-v) or ZBTB16-vector (ZBTB16-v)-transfected HESCs treated with or without EMC for 3 days ± 1 U/mL of thrombin (THR) for 6 h. Bars represent the mean ± SEM; n = 4; * p < 0.05 vs. Cont-v or ZBTB16-v; ⁺ p < 0.05 vs. Cont-v; ^# p < 0.05 vs. Cont-v + THR. (C) Interleukin 8 (IL-8) and prostaglandin endoperoxide synthase 2 (PTGS2 aka cyclooxygenase 2; COX2) mRNA levels in the control- or ZBTB16-vector-transfected HESCs treated with EMC for 3 days ± 1 U/mL THR for 6 h. Bars represent the mean ± SEM; n = 4; * p < 0.05 vs. Cont-v or ZBTB16-v; ⁺ p < 0.05 vs. Cont-v; ^# p < 0.05 vs. Cont-v + THR. The data were analyzed by one-way ANOVA followed by the Student–Newman–Keuls method.

3. Discussion

A variety of contraceptive agents including progestin-only or combined oral contraceptive pills, DMPA injections, subdermal implants, or intrauterine devices are used by women to prevent unintended pregnancies during their reproductive lifespan [45]. AUB accompanying pLARCs is a common side effect and the most-cited cause for discontinuation of these otherwise safe and effective agents. Each year, more than one-million unintended pregnancies occur because of the discontinuation or misuse of these contraceptives [46]. The bleeding sites of pLARC-treated endometria display decidualized stomal cells, as well as irregularly distributed hyperdilated thin-walled superficial vessels associated with deficiency in both basement membranes and perivascular smooth muscle cells or pericytes [8,9,39]. Clotting factor leakage from these impaired vessels interacts with the decidualized stromal cell tissue factor induced by pLARCs, thereby generating excess thrombin, which cause dysfunctional regulation of cellular pathways and/or mediators that play crucial roles in endometrial hemostasis, angiogenesis, and inflammation (Figure 6). In support of these findings, we previously showed that excess thrombin disrupts newly formed vessels [47]. The current study revealed that ZBTB16 may indirectly contribute to pLARC-associated AUB by inducing endometrial pro-inflammatory molecules IL8 and PTGS2, as well as decidualization-associated F3 levels. The integration of our in situ findings from endometria of human DMPA users and MPA-treated guinea pigs with in vitro results from ZBTB16 overexpressing HESCs supports the potential role of enhanced ZBTB16 levels in pLARC-associated AUB.

Open in a separate window

Figure 6

Schematic demonstration of the role of elevated ZBTB16 expression in the pathogenesis of pLARC-induced AUB. Administration of progestin-only, long-acting, reversible contraception (pLARCs) reduce uterine blood flow, which causes local hypoxia [8,9] and induces HESC decidualization, which increases tissue factor [15,23] and ZBTB16 levels. Increased ZBTB16 levels induce excess tissue factor expression, which generates thrombin. The resulting excess thrombin binds to protease-activated receptors (PARs), which increase the expression of several inflammatory factors and angiogenic factors, such as VEGF or IL-8 [19,20,48]. These factors promote excess angiogenesis, which results in abnormal uterine bleeding. pLARC-induced excess ZBTB16 exacerbates thrombin-induced endometrial angiogenesis and inflammation by increasing IL8 and PTGS2 levels in HESCs.

Previous studies demonstrated that ZBTB16 is widely expressed in various tissues and displays diverse biological functions such as cell growth, self-renewal, and differentiation through the regulation of tissue-specific target genes [25,28]. In addition to the documented cellular and molecular roles of ZBTB16, in the female reproductive system, Fahnenstich et al. [35] initially reported elevated ZBTB16 expression in endometrial stromal and myometrial cells during the secretary vs. proliferative phase of the menstrual cycle, thereby revealing hormonal regulation of ZBTB16 in the human endometrium. Subsequently, Szwarc et al. [49] reported that ZBTB16 plays a crucial role during HESC decidualization by acting as a potential mediator of progesterone receptor actions. The current study expanded the potential role of ZBTB16 in the endometrium by investigating whether ZBTB16 contributes to pLARC-induced tissue-factor-mediated thrombin generation and the inflammatory milieu. Therefore, we initially evaluated ZBTB16 expression in pre- and post-DMPA-administered women and guinea pigs as a highly relevant model [9,11]. Consequentially, the results provided the first evidence that DMPA administration induces ZBTB16 expression in situ in endometrial stromal, glandular epithelial, as well as endothelial cells in both women and guinea pigs.

Decidualization represents the cellular transformation of fibroblast-like endometrial stromal cells into epithelial-like decidual cells and is essential for the establishment of a successful pregnancy [50]. Decidualization results from complex interactions of several key factors including steroid hormones, growth factors, signaling molecules, transcription factors, etc., [50,51,52]. Here, we confirmed that both morphologically and functionally differentiated decidual cells display significantly elevated ZBTB16 levels during in vitro decidualization, as previously shown by Kommagani et al. [36]. Our investigation of the individual effect of in vitro decidualization stimuli by using E₂, or MPA, or cAMP alone identified MPA (progesterone/glucocorticoid) as the responsible agent for the robust increase in ZBTB16 levels in decidualized HESCs, suggesting that ZBTB16 is both a progestin- and glucocorticoid-induced transcription factor. This result was consistent with previous studies demonstrating ZBTB16 upregulation in a variety of cell types by progesterone [35], corticosteroids [53], testosterone [54], and aldosterone [55]. Taken together, these results indicate that ZBTB16 is a steroid-responsive transcription factor.

Our current results obtained from in the situ analysis of human and guinea pig endometria, as well as in vitro HESC decidualization led us to analyze ZBTB16 expression in HESCs treated with various progestins including ORG, ETO, LNG, and MPA, which are used in pLARC formulations or DEX (a pure glucocorticoid). Although all these compounds elevated ZBTB16 levels, robust induction was observed in HESCs following either DEX or MPA exposure, confirming that glucocorticoids display a stronger effect on the upregulation of ZBTB16 mRNA and protein levels than progestins. Among progestins, ETO displayed the greatest induction in ZBTB16 levels, indicating that ETO exhibits both glucocorticoid and progestogenic effects [56]. Depending on their derivative molecules, several synthetic progestins display mixed progestin–glucocorticoid actions by binding to the progesterone and glucocorticoid receptors in target cells [57]. Previously, Kommagani et al. [36] reported that progestin-mediated ZBTB16 induction in HESCs requires a progesterone receptor. However, the results of the current study indicated that this effect is primarily mediated by the glucocorticoid receptor rather than the progesterone receptor because the MPA-induced ZBTB16 levels were significantly diminished in glucocorticoid-receptor-silenced HESCs. This was further supported by a moderate induction of ZBTB16 by pure progestin ORG.

Our current study also confirmed MPA-mediated ZBTB16 upregulation in HEECs, which express the glucocorticoid, but not the progesterone receptor [42,43,44]. Our findings that ZBTB16 levels were significantly upregulated in HEECs following either DEX or MPA exposure, but not by ORG, ETO, or LNG treatment, further support glucocorticoid-receptor-mediated regulation of ZBTB16 expression. Several studies reported that increased levels of ZBTB16 expression induce apoptosis and inhibit VEGF-induced proliferation, thereby acting as an anti-angiogenic factor in human umbilical cord and corneal endothelial cell cultures [58,59,60]. Thus, pLARC-induced endothelial ZBTB16 expression may impair endometrial angiogenesis to contribute to the mechanism of pLARC-induced AUB.

The current study provided the first demonstration that ZBTB16 overexpression in cultured HESCs without decidual stimuli (EMC treatment) significantly increases the generation of tissue factor, which is specifically enhanced at bleeding sites vs. non-bleeding sites in the endometrium of pLARC users [14]. Moreover, thrombin-treated HESCs display increased levels of tissue factor, which is further elevated in ZBTB16-overexpressed HESCs, suggesting the mechanism whereby tissue factor elicits excess local thrombin generation, which stimulates VEGF expression, thereby exacerbating abnormal endometrial angiogenesis observed in the endometrium of pLARC users [19]. Moreover, the endometrium of pLARC users contains a higher number of neutrophils and NK cells [61,62] and displays elevated IL-8 levels [20]. In support, our results revealed that thrombin further increases IL8 and COX2 (PTGS2) levels in ZBTB16 overexpressed HESCs, indicating that elevated ZBTB16 levels promote thrombin-induced inflammation in pLARC users (Figure 6) and providing the first demonstration of the combined inflammatory effect of thrombin and ZBTB16 in HESCs. Previous studies reported that thrombin induces the expression of several cytokines and chemokines in a variety of cell types [48,63,64,65]. Moreover, ZBTB16 has been shown to display a wide range of cell-type-specific functions including cell proliferation, cell cycle regulation, apoptosis, stem cell renewal, and differentiation [66,67,68,69,70,71,72]. These wide-ranging ZBTB16-induced effects result from either an activator or repressor effect of ZBTB16 binding to either DNA-specific consensus sequences or other transcription factors, resulting in upregulation or downregulation of gene expression. Mao et al. [70] showed that ZBTB16 indirectly regulates the secretion of several cytokines such as IL-4, -13, -17, or IFNγ by binding to multiple T-helper-specific transcription factors in NKT cells. ZBTB16 can interfere with glucocorticoid-mediated transcription [71,73]. In the current study, the amplification of thrombin-induced IL8 and PTGS2 expression in ZBTB16-overexpressed HESCs supported the function of ZBTB16 to interfere with glucocorticoid-receptor-mediated suppression of inflammatory gene expression. However, further studies are required to verify this effect in HESCs. In summary, ZBTB16 displays diverse functions as a transcription factor and/or chromatin remodeler. Thus, it activates and/or represses target gene expression depending on physiological or pathological conditions. The current study uncovered the novel roles played by ZBTB16 by revealing that: (1) pLARC-induced ZBTB16 expression contributes to HESC decidualization; (2) increased ZBTB16 expression in the endometrium of pLARC-treated women is primarily glucocorticoid receptor mediated; (3) elevated ZBTB16 levels induce tissue factor, which elicits excess local thrombin generation, thereby contributing to AUB; and (4) thrombin treatment in ZBTB16-overexpressed HESCs further increases PTGS2 and IL-8 levels, thereby promoting endometrial inflammation in pLARC users. The limitation of our study is that most of our results were obtained from in vitro findings. Therefore, further in situ study is needed to test the role of ZBTB16 in pLARC-induced inflammation and AUB in higher primates.

4. Materials and Methods

Supplementary Materials

The Supporting Information can be downloaded at: https://www.mdpi.com/article/10.3390/ijms241310532/s1.

Funding Statement

This study was funded, in part, by NIH/NICHD 2 RO1 HD 033937 to C.J.L.

Author Contributions

O.G.-K., U.A.K., S.A., F.S. and C.J.L. designed the research studies, as well as wrote and edited the manuscript. U.A.K. and O.G.-K. conducted the experiments and acquired and analyzed the data. O.G.-K. performed the qPCR and transfection experiments. S.A., N.S., A.O. and K.L. performed the cell culture, immunohistochemistry, and immunoblotting experiments and acquired and analyzed data. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

All paraffin-embedded samples were de-identified after receiving written informed consent at New York University under Institutional Review Board approval (#H6023) and later approval of the University of South Florida (#Pro00019480). The animal studies were performed under the approval of the Yale Institutional Animal Care and Use Committee (IACUC#2006–11002).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Data supporting the results of the current study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that no conflicts of interest exist.

Footnotes

References