Solving the Puzzle: What Is the Role of Progestogens in Neovascularization?

Ovarian sex steroids can modulate new vessel formation and development, and the clarification of the underlying mechanism will provide insight into neovascularization-related physiological changes and pathological conditions. Unlike estrogen, which mainly promotes neovascularization through activating classic post-receptor signaling pathways, progesterone (P4) regulates a variety of downstream factors with angiogenic or antiangiogenic effects, exerting various influences on neovascularization. Furthermore, diverse progestins, the synthetic progesterone receptor (PR) agonists structurally related to P4, have been used in numerous studies, which could contribute to unequal actions. As a result, there have been many conflicting observations in the past, making it difficult for researchers to define the exact role of progestogens (PR agonists including naturally occurring P4 and synthetic progestins). This review summarizes available evidence for progestogen-mediated neovascularization under physiological and pathological circumstances, and attempts to elaborate their functional characteristics and regulatory patterns from a comprehensive perspective.


Introduction
Neovascularization is composed of two conceptions: angiogenesis and vasculogenesis. New capillaries that originate from existing capillaries are called angiogenesis, while vasculogenesis indicates new vessels initially derived from endothelial progenitor cells (EPCs). Although angiogenesis has been proven to be a major part of neovascularization in most situations, these two conceptions coexist and complement each other [1]. Under normal conditions, the vasculature generally remains quiescent, and neovascularization generally occurs in specific physiological and pathological processes, such as the menstrual cycle and tumorigenesis [1,2]. When tissue injury or tumor-associated angiogenic dysregulation happens, angiogenesis can be effectively activated by hypoxia, ischemic stimulus, angiogenic factors and chemoattractant signals, and activated endothelial cells (ECs) from vascular areas acquire proliferative and migrative activities and form angiogenesis sprouting, resulting in vascular morphogenesis from disruptive sites of the basement membrane [3]. Meanwhile, EPCs leave the bone marrow and home to tissue-remodeling sites through the peripheral circulation, and directly incorporate into new vessels, differentiate into mature ECs and produce molecular signals to influence preexisting ECs (Figure 1). Although only a minor player in most instances, vasculogenesis can provide complementary powers especially when there are insufficient conditions for local angiogenesis, contributing to primitive vessel formation [4].
Ovarian sex steroids, mainly estradiol (E2) and progesterone (P4), regulate neovascularization in diverse settings, especially in the endometrium and tumors derived from sex hormone target organs [5,6]. Cholesterol is the common precursor for steroid hormones, and it can be synthesized into P4 and E2 under the actions of enzymes from steroid-generating organs. Based on their lipophilic nature, ovarian sex steroids generally bind to receptors located within the cytoplasm and influence the homing, mobilization P4 mainly functions by crossing the cell membrane and binding to the nuclear progesterone receptor (PR). The P4-PR complex transfers to the nucleus and binds to the special sites of DNA, regulating gene expression. There is an interaction between P4, E2 and their receptors ( Figure 2). On the one hand, the estrogen receptor (ER) promotes PR expression at the transcriptional level [7]. On the other hand, PR-A and PR-B, two major PR isoforms involved in P4-mediated actions, exert distinct effects on reprograming E2 signaling; PR-A represses almost 70% of ER-induced transcriptional genes through inhibiting ER genomic recruitment, while PR-B primarily redistributes ER chromatin binding [8,9]. The crosstalk among ER and PR is well-validated. PR, especially PR-B, combines with ER through ER interaction domains to form the PR/ER complex in the presence or absence of P4, which results in a PR/ER crosstalk that can alter gene expression and downstream signaling. Two main crosstalk modes have been reported; the PR/ER complex binds to progesterone response elements (PREs) or other estrogen response elements (EREs), and thus conveys an alteration or enhancement to the original transcriptomes and signaling [10,11]. The tight interaction between PR and ER makes P4 act as a major modifier in E2 effects, and P4 treatment could attenuate ER expression and control ER-mediated transcriptional regulation in line with P4 actions, exhibiting an antagonism on E2 [12][13][14].
In addition to naturally occurring P4, a variety of synthetic progestins are widely used as PR agonists in in vitro or in vivo PR signaling studies, because synthetic progestins possess P4-like structures and effects, as well as great receptor binding affinities and metabolisms [15]. These PR agonists, including naturally occurring P4 and progestins, are classified as progestogens [15]. E2 generally promote neovascularization via binding to ER [16]; however, conflicting observations have been reported in studies on progestogenmediated neovascularization. Further, E2 combinations can convey a multifaced interaction in neovascularization; progestogens interfere with the expression of E2-induced angiogenic factors and counteract E2 actions in some cases, while the functional synergy between E2 and progestogens has also been confirmed in other studies. Outcomes from diverse experimental models and conditions are yet to clarify the exact role of progestogens.
There is no doubt that progestogens play an indispensable part in neovascularization. The aim of this review is to discuss their influence on neovascularization, and to summarize available evidence on the role of progestogens in physiological and pathological neovascularization, thus yielding an accurate understanding on their functional characteristics and regulatory patterns.

Downstream Factors Involved in Progestogen-Mediated Neovascularization
The activation of EPCs is of great significance in neovascularization, with EPCs serving as one of the primary sources of ECs constructing the endothelial lining of blood vessels [17][18][19]. Both EPCs and ECs possess the ability to proliferate, adhere and migrate, under the stimulation of pertinent angiogenic factors. Progestogens regulate the expression of cytokines and growth factors in a receptor-dependent manner; through genomic regulation activated by the transcriptional activity of nuclear receptors, or through nongenomic regulation via the second messengers mediated by membrane receptors [20,21]. Therefore, clarifying these downstream factors is a prerequisite for an accurate understanding of progestogen-mediated neovascularization.

Vascular Endothelial Growth Factor (VEGF)
VEGF is a crucial regulator for physiological and pathological neovascularization [22]. Shweiki et al. found that VEGF mRNA tends to be expressed in steroidogenic or steroidresponsive cell types from the female reproductive system, and they first proposed hormones that can regulate VEGF expression to influence the natural angiogenic process [23]. These assumptions were supported by later studies. Progestogens that are combined with E2 induce VEGF expression, leading to EC proliferation in the human endometrial tissue [24,25]. P4 has been confirmed to stimulate VEGF expression, leading to endometrial EC proliferation and capillary-like tube formation [26][27][28][29][30][31]. These observations support the vital role of VEGF in progestogen-mediated neovascularization. In addition to in the endometrium, progestogens promote VEGF expression in skin flaps [32], ovarian tissue [33,34], bovine granulosa cells [35,36], embryonic lung cells [37], granulomas [38] and tumor tissues [39][40][41][42]. However, Keck et al. reported an opposite result, founding that P4 has no impact on the proliferation rate of human umbilical vein endothelial cells (HUVECs), a macrovascular cell model used for endothelial studies in vitro [43]. An anti-angiogenic role of progestogens has been reported in several studies. Okada et al. found that high concentrations of progestins decrease E2-induced VEGF in human endometrial stromal cells [44]. Similar conclusions have been drawn in rat models. Through reducing VEGF expression, P4 effectively ameliorates abnormal angiogenesis in diabetic renal disease and prevents tissue plasminogen activator-induced hemorrhagic transformation in stroke brain tissues [45][46][47]. Although VEGF has been proven to be involved in the downstream of progestogens signaling, conflicting observations suggest that diverse regulatory mechanisms are involved.
PR comprises the main bridge in progestogen-mediated neovascularization. Kim et al. showed that the P4-PR axis is implicated in VEGF physiological changes in the decidual stromal cells, and that PR expression coincides with VEGF in decidual angiogenesis, reporting a rise in expression during pregnancy and a fall during the postpartum period [48]. Hyder et al. confirmed that P4 upregulates VEGF mRNA in a PR-dependent manner [49]. The P4-PR axis is vital to the activation of EPCs. Yu et al. demonstrated that P4-mediated PR improves VEGF expression in EPCs, and thus enhances cell mobilization and tube formation, leading to increases in vessel density and the recovery of neurological function in traumatic brain injury [20,50]. In a previous study, the authors found that P4 regulates EPCs viability through activating PI3K/Akt signaling [51], a validated upstream pathway of VEGF, implying a nongenomic regulation activated by P4. A study found that PR-binding sites were in the promoter regions of the VEGF gene, and that P4-mediated PR regulates VEGF mRNA, confirming that VEGF is directly regulated by the P4-PR axis at the transcriptional level [52]. Meanwhile, some evidence suggests that PR agonist dienogest attenuates VEGF expression in human endometriotic epithelial cell lines [53]. This negative regulation has been confirmed in HUVECs: P4 activates cSrc, a non-receptor tyrosine kinase, to promote the formation of the cSrc-PR complex, which represses downstream RhoA/ROCK signaling and inhibits the proliferation and migration of HUVECs; this effect is abrogated by blocking PR [54]. Obviously, PR is involved in P4-induced VEGF, but its ultimate effect remains unclear. In addition, a study found that P4 promotes the formation of tube-like structures and sprouts of endometrial ECs independent of classic genomic actions by PR [55]. Angio-miRNA might be an important regulator, and Salmasi et al. revealed that P4 treatment promotes the expression of miR-16-5p and miR-17-5p, which bind to the untranslated regions or coding sequence regions of VEGF, and eventually enhances VEGF expression and endometrial EC proliferation [29,30]. Further, membrane progesterone receptors are also implicated in VEGF expression. P4-mediated progesterone membrane component 1 (PGRMC1) induces VEGF expression via PKC signaling, which might promote neovascularization within the retina [56]. Neubauer et al. made a similar conclusion, and they found that P4-mediated PGRMC1 enhances VEGF expression and proliferation of breast cancer cells [57]. However, a study found that the knockout of PGRMC1 and PGRMC2 conveys a repression on luteal vascularization in a mice model, although increasing VEGF mRNA levels were observed in the ovaries and corpora lutea, with the authors proposing that this effect mainly originated from reduced levels of lipocalin 2, resulting in the decreasing expression of matrix metallopeptidase 9 and the activity of VEGF [58].
In addition, P4 influences VEGF angiogenic activity by affecting the levels of vascular endothelial growth factor receptors (VEGFR): VEGFR-1 and VEGFR-2. With rising levels of ovarian sex steroids during pregnancy, VEGF and its receptors are overexpressed in stromal histiocytes and ECs in pregnant tumors of the gingiva [59]. A similar change has been observed in the menstrual cycle; compared to the proliferative phase, a fold increase in VEGFR-1 and vascular density appears in the endometrium during the secretory phase, a stage characterized by high serum P4 concentrations [60], implying a positive correlation between P4 and VEGF receptors. Mandana et al. found that P4 enhances VEGFR levels to ameliorate antiangiogenesis after ovarian stimulation [27]. Subsequent studies confirmed that P4-mediated PR leads to stable increases in VEGF and VEGFR-2, leading to a heightened density of endometrial blood vessels [61], and that P4 antagonist represses VEGFR-2 expression in pregnant uterus tissue [62]. A similar conclusion has been made in studies on breast cancer; the expression of VEGF and VEGFR-2 rises under P4 stimulation, leading to the proliferation of tumor epithelial cells [63]. However, several studies have provided opposite conclusions, finding that P4 withdrawal leads to increases in VEGF, VEGFR-1 and VEGFR-2 in the superficial zone of the macaque or human endometrium, revealing an inhibitory effect on VEGF and its receptors [64,65].

Basic Fibroblast Growth Factor (bFGF)
As a potent angiogenic factor, bFGF binding to its receptor modulates angiogenesis in autocrine and paracrine manners [66]. Chromatin immunoprecipitation (ChIP) and ChIP-sequencing analysis has indicated several PR binding sites in the promoter region of bFGF, and that P4-mediated PR-B upregulates bFGF mRNA, confirming bFGF as the downstream gene in PR signaling [52]. Evidence supports P4 s positive regulation of bFGF. Yuan et al. found increasing blood vessel density and bFGF in granulomas during pregnancy, and that P4 significantly improves the concentration of bFGF in the human monocyte-like U937 cell under the stimulation of lipopolysaccharide [38]. However, medroxyprogesterone acetate (MPA), a synthetic progestin, suppresses bFGF expression and secretion in well-differentiated uterine endometrial tumor cells, while it has no effect on poorly differentiated uterine endometrial tumor cells [67]. In agreement with this finding, P4 and synthetic progestins can inhibit the transcription of bFGF in endometrial tissues, leading to a decreasing vessel density [68].At the same time, P4 exerts an antagonistic effect on E2-induced bFGF [40,69], triggering decreases in bFGF in Ishikawa cells.

Platelet-Derived Endothelial Cell Growth Factor (PD-ECGF)
The increasing expression of PD-ECGF, a strong angiogenic enzyme, parallels the raised concentration of P4 during the late secretory phase in the endometrium, while remains at a normal and stable level during the proliferative phase [70]. A study revealed the physiological concentration of P4 only induces a moderate expression of PD-ECGF in stomal cells, while when combined with transforming growth factor-β1 (TGF-β1), it can induce significant fold changes in epithelial cells, even though P4 or TGF-β1 alone has no influence on PD-ECGF levels [71]. A similar outcome has been observed in uterine endometrial cancer; P4 and its metabolite can lead to a significant increase in PD-ECGF [72]. However, Zavarhei et al. made a controversial conclusion, founding that P4 could suppress PD-ECGF via PR in colorectal cancers [73].

Angiopoietin (Ang)
Ang-1 and Ang-2 are well-studied growth factors involved in angiogenesis; Ang-1 ensures the stability and integrity of the vessel, and Ang-2 influences the matrix contact, EC activity and the permeability of the vessel. Progestogens regulate the two Ang isoforms to maintain the steady function and normal structure of the female reproductive system. Park et al. demonstrated that P4-mediated PR enhances the expression of Ang-2 in human uterine endothelial cells, leading to corresponding vascular remodeling during pregnancy [74]. A paper reported that P4 priming enhances the production of Ang-1 and Ang-2 in large preovulatory follicles, which is important for luteal angiogenesis and normal function [34]. Krikun et al. revealed the mechanism behind abnormal uterine bleeding in long-term progestin-only contraceptives users. MPA promotes Ang-1 in endometrial stromal cells, while it stimulates Ang-2 in endometrial ECs, and progestin-induced local hypoxia diminishes MPA-induced Ang-1, and thus impairs the balance of Ang-1 and Ang-2, resulting in abnormal blood vessels structures [75].

Hypoxia Inducible Factor 1α (HIF1α)
Hypoxia is a contributing factor in progestogen-mediated neovascularization. Longterm progestin-only contraceptives result in endometrial local hypoxia and reperfusion, which damages the progestin-mediated dynamic balance in neovascularization, leading to aberrant angiogenesis, such as abnormally enlarged and fragile microvessels [76]. The interaction between P4 and HIF1α, a hypoxia-induced transcription factor with a proangiogenic function, has been reported in previous studies. HIF1α fluctuates in a menstrual stagespecific manner during corpus luteum development, and HIF1α-mediated VEGF interacts with luteal function, influencing the concentration of P4 and angiogenic status [77]. A microarray analysis for differentiating endometrial stromal cells revealed that P4-mediated PR activates pathways related to blood vessel development and responses to hypoxia, and that VEGF and HIFI1α are important regulators in this process [52], implying a correlation among hypoxia, neovascularization and the P4-PR axis. These results have been confirmed in studies of the bovine granulosa cells and the ewe uterine, which have found that P4 can induce the expression of HIF-1α mRNA and downstream factor VEGF [35,78]. Although publications support the interaction between progestogens and HIF-1α, whether there is a feedback regulation is still unclear.

Nitric Oxide Synthase (NOS)
Previous studies have reported that ECs originating from females are prone to higher endothelial NOS expression, a vital factor in regulating capillary outgrowth and vascular development [79,80], implying a correlation between female sex hormones and endothelial NOS. P4 has been confirmed to enhance endothelial NOS expression through promoting the expression of specificity protein-1, which could form a complex with PR-A and then bind to the promoter region of endothelial NOS [81,82]. Pablo et al. found that P4 promotes nitric oxide (NO) synthesis, EC viability and tube formation, whereas MPA exerts an opposite effect. Irreversible NOS inhibitor significantly reduces P4-induced NO production, and blocking PR partially represses both P4-and MPA-mediated actions; the same inhibitory effect was observed after the treatment of MAPK and PI3K inhibitor [31,83], implying NOS and its related signaling pathways are involved in vessel development. It is not unusual for studies to find discrepancy between the biologic effects of P4 and MPA, with Simoncini et al. finding P4-mediated PR promotes the production of NO and endothelial NOS via ERK/Akt phosphorylation, while MPA has an opposite effect. They further verified that P4 and PR antagonist convey no effect on E2-induced NO and endothelial NOS expression, while MPA significantly interferes with E2 actions, suggesting that MPA exerts an inhibitory effect independent of PR [83,84]. A later study confirmed that blocking the glucocorticoid receptor can attenuate the inhibitory effect of MPA [85]. Therefore, progestins might activate nongenomic regulation or crosstalk with other steroid hormone receptors, and thus alter downstream signaling and gene transcription [86,87]. A more recent study confirmed this activation of nongenomic regulation; P4-mediated membrane progestin receptors (mPRs), which can mediate a rapid nongenomic signaling, were found to significantly increase endothelial NOS activity and phosphorylation in HUVECs, accompanied by an activation of the PI3K/Akt and MAPK pathways, ultimately influencing the vascular status and function [88].

Other Downstream Factors
Despite limited evidence, several angiogenic factors are likely to serve a downstream role in progestogen-mediated neovascularization. Platelet-derived growth factor-A (PDGF-A) is known to potentiate VEGF production and release [89], and a paper reported that P4 treatment leads to increased expression of PDGF-A in breast cancer MCF7 and T47D cells [90]. Thrombospondin-1 (TSP-1), an antiangiogenic glycoprotein, is stimulated by P4 in the endometrium, leading to inhibition of the migration of ECs and vessel formation [91]. A consistent conclusion has been made with regard to breast cancer and Ishikawa cells, and progestins have been found to promote TSP-1 expression in a PR-dependent manner [92].
In general, numerous effective factors are involved in progestogen-mediated neovascularization, suggesting that it is a co-regulated process influenced by multiple pertinent angiogenic factors. The conflicting regulation of progestogens on most factors implies their action is probably susceptible to some intrinsic and extrinsic factors.

Progestogen-Mediated Physiological Neovascularization
P4 is one of the major drivers of physiological changes in the endometrium. Dynamic neovascularization occurs in cyclic changes in the endometrium [93]. The menstrual cycle leads to corresponding changes in P4 concentration, influencing vasculogenesis and angiogenesis in uterus tissue [5,94]. During the early proliferative phase, the serum P4 concentration is relatively low (lower than 1 ng/mL). After ovulation, the endometrium goes into the secretory phase, which is characterized by a high microvascular density, providing a basis for fertilization and implantation, and serum P4 peaks in the midsecretory phase (10-35 ng/mL) [8,95]. Neovascularization periodically ranges from the proliferative phase to the secretory phase [96,97].
The vasculature growth and VEGF expression exhibits period features corresponding to P4 concentration throughout the menstrual cycle, and a significant increase occurs in the late proliferative and secretory phases, in which P4 levels rapidly reaches their peak [28,98,99]. Meanwhile, increases in EPCs have been observed in peripheral circulation during the secretory phase, and they then migrate to the endometrium [100,101], implying a regulatory correlation between serum P4 and EPC mobilization. Under the influence of ovarian hormones, EPCs acquire an improved angiogenic potential, and P4-mediated PR improves the proliferative ability of EPCs in a dose-dependent manner [102]. Vessel remodeling, a type of endometrial neovascularization characterized by a high average microvessel length and low blood vessel junction, is prominent from the late proliferative to mid-secretory phase [103], accompanied by the proliferation of vascular smooth muscle cells and coverage of mural cells, which contribute to the maturation of endometrial vascular tissues [104][105][106]. A study reported that vessel maturation is primarily related to P4 while E2 has no impact on this process, implying that P4 rather than E2 plays a pivotal role in this process [26]. Later studies confirmed that P4 facilitates vessel development and maturation in the endometrium [107,108]. As a prerequisite in decidual angiogenesis, P4 upregulates VEGF and NOS in the endometrium, and thus elevates blood perfusion and embryo receptivity, contributing to suitable conditions for embryo implantation [25,27,48,[109][110][111][112]. Blocking VEGFR-2 can decrease P4-induced decidual angiogenesis during early pregnancy [113]. Meanwhile, embryo implantation and placentation potentiate endometrial angiogenesis by facilitating VEGF generation [114], and a low expression of VEGF in decidual cells and the epithelium of endometrial glands is associated with spontaneous abortion [115]. In a way, the relationship between neovascularization and embryo implantation is based on P4 and its downstream factor. PR knockout renders the uterus unable to undergo decidual transformation even under an artificial decidual stimulus [116]. Blocking P4/PR interferes with the development of capillary vessels in the postovulatory endometrium, leading to the regression of spiral arteries and abnormal veins [117,118]. Although it is a wellestablished antagonistic feature of P4 to E2-mediated actions [44,119], P4 combined with E2 further enhances human uterine microvascular endothelial cell invasion and sprouting angiogenesis [120].
A significant portion of studies support the pro-angiogenic role of progestins, but several studies indicated that P4 and progestins play an antiangiogenic role in endometrial vascular development. Rashidi et al. demonstrated that P4 attenuates EC proliferation induced by gonadotropins, leading to an inhibitory effect on endometrial angiogenesis [121], and a consistent pattern has been found in ectopic endometrial lesions [68]. Synthetic progestins have a similar influence, and several studies have revealed that the levonorgestrel-releasing subdermal contraceptive has an anti-angiogenic effect, and that it significantly inhibits the proliferation rate of endometrial ECs, which results in reduced endometrial angiogenic activity [122,123]. Further, a sustained peak expression of cyclin E and A has been observed in this process, resulting in arresting of the EC cycle in G1 [124]. Hsu et al. demonstrated that P4-mediated PR can upregulate p21 and p27 in a p53-dependent manner, and that it inhibits DNA synthesis of HUVECs, leading to G0/G1 arrest and a decrease in cell proliferation and capillary-like tube formation [125,126]. This antiangiogenic effect has been observed in the setting of E2 priming, and P4 attenuates E2-induced angiogenesis by inhibiting ER expression, with the PR-A acting as a trans-repressor for ER [13]. A paper reported that P4 attenuates E2-mediated angiogenesis in an ER-independent manner [127], meaning that multiple regulatory patterns are involved in this process.

Progestogen-Mediated Pathological Neovascularization
Neovascularization conveys a more efficient vascular supply for solid tumors, and it has proven to be an important mechanism supporting tumoral aggressiveness [128]. Progestogens are mainly implicated in the neovascularization of female sex hormonedependent cancers, including endometrial cancer and breast cancer.

Endometrial Cancer
Compared to the healthy population, lower P4 concentration tends to appear in endometrial cancer patients [129]. P4 inhibits E2-driven growth, and progestogen-mediated PR can prevent endometrial hyperplasia and well-differentiated endometrial cancer from E2-induced growth and invasiveness [130]. Progestin therapy is an effective anti-tumor method for endometrial cancer cases, which can inhibit tumor recurrence and progression in nearly 80% of early stage patients [131]. Publications confirmed the mechanism underlying the inhibitory effect of progestins on endometrial cancer is associated with angiogenesis. After human uterine microvascular ECs were incubated with the conditioned media from Ishikawa cells treated by P4, decreased invasion and tube formation were observed, and these changes were abrogated by blocking PR, implying that the tumor microenvironment functions as a mediator in progestogen actions. Meanwhile, inhibiting Akt signaling could potentiate the antiangiogenic effect [132]. Studies have found that P4 treatment alone cannot alter VEGF in Ishikawa cells [133], but P4, MPA and 17 a-hydroxyprogesterone effectively suppress E2-induced VEGF [134], confirming that progestogens exert antagonistic impacts on E2-induced angiogenesis in endometrial cancer. However, a study reported that progestins exert a synergistic effect with E2 actions. P4 and its metabolite, 17alpha-hydroxyprogesterone, could increase PD-ECGF expression; a fold change in PD-ECGF was observed when it was combined with E2, but this effect could not be induced by MPA, suggesting a discrepant regulation of P4 and progestin on PD-ECGF [72]. Further, a cell-specific regulatory pattern was observed; P4 and MPA suppress E2-induced bFGF secretion in well-differentiated uterine endometrial cancers cells, while they cannot suppress bFGF in poorly differentiated cells [67,69], which accords with the protective effect of progestogens on well-differentiated endometrial cancer.

Breast Cancer
Although serum P4 is generally in the normal range, the activation of its receptors exerts indispensable influence on breast cancer progression [135]. Progestogens that bind to PR promote breast cancer growth. ER upregulates PR expression, and PR is thus a well-validated biomarker for predicting patient response to ER-targeted therapies. How-ever, PR-targeted therapies have failed to progress significantly in clinical therapy for breast cancer, due to confusion regarding the effects of progestogens, which exhibit both stimulatory and inhibitory effects on angiogenesis [9,136]. Hyder et al. first reported that P4 stimulates VEGF expression via PR in human breast cancer cells, and that specificity protein-1 and MAPK signaling are involved in transcriptional and posttranscriptional regulation [49,133,137]. Genomic profile analysis revealed that the expression of VEGF and P4-mediated pathways are both activated in breast cancer [138]. This result is consistent with the results of next-generation sequencing, which have confirmed that VEGF acts as a downstream gene in PR signaling [139]. Later studies confirmed that P4 or synthetic progestins activate angiogenic pathways associated with VEGF, and this effect is preferentially mediated by PR-B and inhibited by blocking PR [140][141][142]. Botelho et al. proposed that P4-mediated VEGF primarily promotes angiogenesis in the early stages of tumorigenesis, which leads to a high aggressiveness [39]. In view of the higher P4 concentrations in young people, this finding can partially explain the more advanced and aggressive nature of breast cancer in adolescents and young adults (age < 40 years) [143]. In addition to VEGF, P4 can induce PDGF-A expression in the breast cancer MCF7 cell [90]. However, a study proposed that PR is not significantly associated with microvessel density in breast cancer [6], while Vameşu et al. confirmed that a significant correlation between high microvessel density and PR negative status exists [144]. P4 and synthetic progestins binding to PR potentiate the expression of anti-angiogenic factor TSP-1 in breast cancer cells [92], implying that P4-PR axis plays an antiangiogenic role in this process. This antiangiogenic effect was validated in a rat model: MPA and fluorinating MPA significantly inhibit angiogenesis, leading to the low growth of mammary carcinomas [145].
In addition, several studies have reported that progestogens influence angiogenesis in other cancer types. A study showed that P4, as well as E2, can stimulate VEGF secretion in non-small cell lung cancer cells [146]. An in vitro study confirmed that P4 and MPA promote A549 to secrete VEGF in order to enhance vascular endothelial cell proliferation, and tumor-associated angiogenesis reaches the maximum when combined with E2 [42]. Similar to its activity in lung cancer, P4-mediated PR has been found to promote VEGF expression in human astrocytoma cell lines, and steroid receptor coactivator-1, which participates in PR downstream gene transcription, exerts a transcriptional modulation in this process [41].

Discussion
As outlined above, progestogens exert various effects on neovascularization, but conflicting observations in previous studies mean an accurate understanding of their functional characteristics and regulatory patterns remains elusive. We presumed several intrinsic and extrinsic factors that lead to this situation (Figure 3).
First of all, different concentrations of progestogens contribute to contrary outcomes, mainly in in vitro studies. Several studies have found a consistent phenomenon: low concentrations of P4 promote neovascularization, while high concentrations inhibit the original effect [20,35]. The level of PR probably is the cause, and PR expression is influenced by sex hormones; E2 directly regulates the transcription of PR, and P4 exerts a dosedependent effect on PR expression [20]. The spatiotemporal characteristics of PR expression have been found in endometrial cells; PR rises differently in the epithelium and stroma during peri-implantation, while it is restrictively expressed in stroma cells under changes in ovarian sex hormones after implantation, which means sex hormones can cause dynamic changes in PR [147]. We noted that the anti-angiogenic effect of P4 tends to occur in in vitro studies using a concentration higher than 10 −8 mol/L (Table 1). Yu et al. confirmed that progestogens exhibit a dose-specific regulation for PR distribution and expression, and that P4 promotes PR expression in low concentrations while it repairs PR expression in high concentrations, which could explain the concentration-dependent influence of progestogens, to a certain extent [20].   P4: 10 −7 mol/L MPA: 10 −9 to 10 −7 mol/L NET: 10 −9 to 10 −7 mol/L LNG: 10 −9 to 10 −7 mol/L DNG:10 −9 to 10 −7 mol/L [149] MPA  Granulosa cells from large follicles VEGFA mRNA ↑ 300 ng/mL (9.5×10 −7 mol/L) [35] P4 Ovary bovine granulosa cells (cows) VEGF120 ↑ (10ng/mL) HIF1α ↑ (10ng/mL) VEGFR2 ↑ (10ng/mL) VEGF120 ↓ (100ng/mL) HIF1α ↓ (100ng/mL) VEGFR2 ↓ (100ng/mL) 10 ng/mL (3.18 ×10 −8 mol/L) 100 ng/mL (3.18×10 −7 mol/L) Moreover, cell and tissue specificity are other important factors. As shown in Table 1, progestogens boost VEGF expression in diverse breast cancer cell lines, but they exert discrepant effects on different types of human endometrial cells. MPA increases VEGF in epithelial cells, while it reduces VEGF in fibroblast cells [149], implying a cell-specific regulation. Manifold expression patterns of PR isoforms in these cell types could be the root reason for this [150,151], due to the fact that PR isoforms regulate distinct genes profiles. Only 25% of differential expression genes are regulated by both PR-A and PR-B, and PR-mediated responses depend on the coordinated actions of each PR isoform, meaning that different molecular signaling pathways are activated by PR isoforms, which leads to a distinct influence under progestogen treatment [52]. A study confirmed that PR-B prominently enhances VEGF expression, while PR-A inhibits PR-B-mediated actions [142], and thus PR isoforms convey different regulations dependent on the PR-A/PR-B ratio [152]. In addition, conflicting observations are prone to occur in endometrial epithelial cells and stromal cells, and progestogens exert contrary effects even on the same cell type. Heryanto et al. confirmed that there are different proliferative responses to sex hormones among endometrial ECs and epithelial and stromal cells [153]. Further, the procedure of separating epithelial and stromal cells from the normal endometrium could contribute to the suboptimal purity of isolated cells, resulting in an experimental bias. Researchers have proposed that controversial outcomes might be derived from the mistaken identification of endometrial cell types due to technological defects [26]. Regardless, we have noted inconsistent observations between HUVECs and human endometrial endothelial cells [120], suggesting that inevitable cell-specificity occurs even in the well-validated endothelial vascular cell model, and another study confirmed different expression patterns of PR in endothelial cells from different tissues [124]. In contrast to in vitro studies, studies in vivo have reported relatively unanimous observations ( Table 2); progestogens promote uterine neovascularization, while they suppress angiogenesis in endometriotic tissues, and this difference can be explained by the patchy expression pattern of PR isoforms in endometriosis [154,155]. In general, tissue and cell specificity make some results controversial.  In addition, E2 priming and its combination with E2 bring changes to original regulatory patterns. Certainly, some amount of the changes come from E2-mediated signaling, which activates cognate nuclear receptor and non-genomic pathways, leading to altered proangiogenic profiles [86]. Further, more than 80% of genes regulated by E2 are regulated by progestogens, while progestogen treatment can alter more than half of E2-induced transcriptomes [14], suggesting that PR dominantly conveys an antagonist to E2-related gene expression and signaling. However, progestogen-mediated PR not only exerts a partial antagonist for E2 [10,157], but a synergy has been observed in many studies (Table 3). One explanation is that progestogen mediates synergistic actions via nongenomic regulation such as mPRα or PGRMC1 signalings, circumventing PR, especially in cell lines with weak expression of PR or without PR [158]. Several studies have reported a positive correlation between PGRMC1 and ER [159,160]. PGRMC1 overexpression promotes E2 biosynthesis and expression of ER-α and its downstream genes [161]. E2 can activate the downstream signaling of PGRMC1, and ER-α knockout downregulates PGRMC1 levels while silencing PGRMC1 inhibits ER-α expression, suggesting a crosstalk and synergistic effect between ER-α and PGRMC1 [162]. An alternative explanation is that E2 potentiates the effect of progestogens, and that even the effect is opposite to its own. Publications have confirmed that E2 upregulates PR, thus enhancing P4 s effect [142,150,163], while the selective PR antagonist inhibits E2-mediated actions, confirming the functional synergy between PR and ER [14]. A publication confirmed that special genes are expressed only under ER and PR coaction, binding to their own sites of promoter regions, and this regulatory mode partially accounts for the synergistic effect [164]. In this regard, the concentration of E2 acts as a contributing factor. Studies have demonstrated that different concentrations of E2 convey contrary influences on the action of progestins [153,165]. Moreover, distinct regulatory modes between E2 and P4 could result in a conflict; P4 tends to exert a late and sustained effect on the expression of VEGF, whereas the effect of E2 is early and transient [20,61,98]. Therefore, observations in different periods can vary, even though they have only been reported in studies on normal cells and tissues. In general, E2 and its receptors make progestogen-mediated neovascularization more complicated [166,167].    Finally, progestogen types can cause conflicts. Due to the different structures among progestogens, altering molecular biofunctions makes natural P4 and synthetic progestins work in individual patterns. Many papers have revealed drug-specific influences: P4 mainly upregulates VEGF in the basal layers of epithelial cells, while the effect of MPA is primarily exerted on stromal and epithelial cells [156]; P4 modulates vascular remodeling by inducing VEGF and endothelial NOS, while the regulation of MPA mainly depends on platelet activation [31,83]. Progestins were proved to have a different regulatory intensity for the same angiogenic factor [148]. Furthermore, the pharmacological effect of PR modulators and P4 antagonists has been reported to work in a cell-specific manner [149]. A study confirmed that P4 and progestins activate or repress unique genomic and proteomic profiles [168], which explains some puzzling findings in previous studies, such as that P4 uniquely affects angiogenesis while MPA affects cell migration [86]. Therefore, some conflicting outcomes are likely due to the use of different progestogen types.
Certainly, the spatial and temporal factors make progestogen-induced actions variable, and thus some observations might deviate from the truth. The question of how to preclude these influencing factors is particularly important for future studies. Progestogens activate the signaling molecules to modulate neovascularization through binding to their receptors, and we noted that ligand-independent activation is widely validated in steroid hormone receptors. In contrast to the classical steroid hormone-dependent manner, ligandindependent activation can be triggered by the steroid hormone receptors themselves, growth factors and various receptors [169][170][171]. The type of activation conveys similar downstream consequences with activation by steroid hormones, and an increased receptor concentration enhances the ligand affinity and potency of post-receptor signaling [172,173]. The P4 receptors, whether they are PR, mPRα or PGRMC1, have been confirmed to be activated in the absence of ligands [58,174,175]. Therefore, gene transfection technology can generate a model with receptor knockdown or overexpression, which provides a feasible and effective method to explore the corresponding change in post-receptor signaling without influences from progestogen concentrations and types. Given the complexity of progestogen-mediated neovascularization, in vitro studies need to be proofed with evidence from in vivo studies. Moreover, there are few direct studies on neovascularization during P4 changes in pathological conditions associated with placenta dysfunction and neuroendocrine neoplasms, and this is a direction worthy of investigation. Furthermore, future studies can explore the role of progestogens in subdivisions of neovascularization, such as sprouts, intussusception and elongation, which might provide novel insight into the mechanisms of progestogen actions.
Author Contributions: Conceptualization, Q.C. and Z.X.; Literature search, Z.X. and J.X.; Writingoriginal draft preparation, Z.X.; Writing-review and editing, Z.X. and J.X. All authors have read and agreed to the published version of the manuscript.
Funding: This research is funded by The Natural Science Youth Foundation of Hunan Province (No. 2021JJ41036).

Conflicts of Interest:
The authors declare that there are no conflict of interest.