Therapeutic Strategies and Potential Actions of Female Sex Steroid Hormones and Their Receptors in Colon Cancer Based on Preclinical Studies

Several epidemiological studies have reported that the use of female sex steroid hormones could reduce the risk of colon cancer (CRC). This review summarizes the available data related to estradiol (E2) and progesterone (P4) single and dual treatments in CRC male and female in vitro and in vivo models, mainly from preclinical studies, alongside their potential molecular mechanisms. Most of the studies showed that E2 exogenous treatment and/or reactivation of its beta receptor (ERβ) significantly inhibited cell proliferation, induced cell cycle arrest, and promoted apoptosis by modulating several molecular pathways. Likewise, the inhibition of ERα receptors produced similar antitumorigenic actions, both in vivo and in vitro, suggesting that E2 could have dual opposing roles in CRC that are dependent on the expression profile of its nuclear receptors. The available studies on P4 are scarce, and the results revealed that in vitro and in vivo treatments with natural and synthetic progesterone were also associated with promising tumoricidal actions. Nevertheless, the combination of E2 with P4 showed enhanced anticancer activities compared with their monotherapy protocols in male–female cell lines and animals. Collectively, the studies suggested that the female sex steroid hormones could provide a novel and effective therapeutic strategy against CRC.

Although considerable improvements in the therapeutic strategies against CRC have been made, the rates of metastasis, as well as relapse, are increasing due to presentation at late stages of malignancy, development of chemoresistance, and occurrence of severe side effects during treatment [14,15]. Chemoresistance is the primary reason for treatment failure and occurs in more than 90% of CRC patients with metastatic tumors [14]. Current researchers are still searching for alternative therapies for CRC to overcome these clinical problems and/or to increase survival rates and quality of life. Till now, CRC remains one of the leading causes of cancer-associated morbidity and mortality in males and females worldwide [16]. Nevertheless, the incidence rate of CRC is lower in premenopausal females and linked with better 5-year survival rates compared with age-matched males [5,[17][18][19][20][21][22][23].
Indeed, oophorectomy and the early suppression of female sex hormones have been revealed to increase the risk of CRC by 30% [24]. Other studies also reported that premenopausal CRC patients had higher 5-year survival rates than males of the same age or postmenopausal women; further proposing that female sex steroid hormones could improve prognosis [22,25]. In this regard, age is considered to be a major risk factor for CRC, with an estimated 90% of cases and 94% of deaths occurring in individuals older than 50 years [19,22,23]. Moreover, a meta-analysis of 13 retrospective cohort studies and 1 randomized controlled trial in 2017 revealed that survival in CRC patients was genderdependent [26]. Another recent German population-based cohort study that included 185,967 patients also demonstrated significantly better overall and recurrence-free survival in females than males [27]. Hence, the differences related to the patient's gender and age of CRC onset could be associated with abnormal alterations in systemic and/or colonic levels of ovarian sex steroids (i.e., estrogens and progesterone) [18][19][20][21][22][23]28,29].

Female Sex Steroid Hormones
Sex steroid hormones are derived from cholesterol and involve estrogens (e.g., 17βestradiol, estriol, and estrone), progestogens (e.g., progesterone), and androgens (e.g., testosterone) [22]. While sex hormones are mainly produced by female ovarian and male testicular tissues, there are other tissues, including adipose tissue, adrenal glands, skin, brain, and bone, which also contribute to the production of sex hormones [22,30,31]. In plasma, sex hormones are primarily bound to albumin and sex-hormone-binding globulin (SHBG), whilst only 1-3% of those hormones are free and biologically active [22]. Both genders produce all types of sex hormones but in differential amounts, and their biological effects depend on their systemic concentrations, expression of their nuclear receptors, and their specific interactions with target organs [22]. Commonly, estrogens and progesterone serum concentrations are higher in females than males and play several roles in female physiology, including the menstrual cycle and pregnancy in premenopausal females, whereas the production of both hormones from the ovaries declines to extremely low levels in postmenopausal females; postmenopausal levels correspond to male estrogen and progesterone levels [32][33][34]. Based on the current epidemiological data, estrogens and/or progesterone could play a role in preventing CRC (Section 1.2).
Progestogens are the first products of cholesterol side cleavage in gonadal tissues during sex hormones synthesis [35][36][37], and the most potent family member is progesterone (P4) [35][36][37]. On the other hand, there are three natural estrogens, known as estrone (E1), 17β-estradiol (E2), and estriol (E3) [22]. The most potent member of the estrogen family which is abundant in humans is E2 [19,22,23], which is chiefly generated by reproductive tissues, such as ovarian follicles, placenta, and corpus luteum [23]. The actions of E2 and progesterone are dependent on binding to their specific nuclear receptors (ERs and PRs, respectively) [33,38], which act as transcription factors to stimulate genomic and non-genomic actions that often converge at the adjacent responsive genes [33,38,39].
ERs are classified into alpha (ERα) and beta (ERβ) subtypes, and both contribute to the initiation, development, and progression of hormone-dependent neoplasms [38]. Moreover, E2 could produce opposing actions in cancer biology that are dependent on the type of activated ER [19,23]. In further detail, E2 induces ERα-dependent cell cycle progression, cell proliferation, inflammation, and carcinogenesis [19,23,38,40]. In contrast, the activation of ERβ by E2 provokes cell cycle arrest, modulates the expression of many of the ERαregulated genes, and induces multiple anticancer activities [19,23,38,40]. Similarly, PR has two isoforms-the N-terminal-truncated A (PRA) and the full-length B (PRB) [33]-and the activation of both isoforms by P4 causes a structural change following segregation and dimerization of heat shock proteins, which bind to specific DNA promoter sites as P4 response elements, causing transcriptional activation or repression [33].

E2 and P4 in CRC: Epidemiological Findings
Numerous epidemiological [29,41,42] and randomized controlled trials (RCTs) [29,[43][44][45][46][47][48] studies have found that exogenous treatment with oral contraceptives (OC) and hormone replacement therapy (HRT) were linked with protective roles against CRC development. Under this theme, women using OC had a 10-20% lower risk of developing CRC [41]. Concurrently, others have demonstrated that menopausal females using HRT had significantly lower rates of CRC, by almost 50-63%, compared with age-matched females who were not using HRT; these results imply that female sex steroids hormones could protect against CRC [49][50][51][52][53][54][55]. Likewise, the use of HRT for a few years prior to CRC diagnosis in females was associated with improved survival rates [56][57][58]. Furthermore, several meta-analysis studies revealed that HRT reduced the risk of developing CRC in microsatellite-stable states [42], as well as improving mortality rates in female CRC patients [59,60]. In addition, Labadie et al. (2020) [61] has also confirmed the strong inverse correlation between HRT use and overall CRC risk in postmenopausal females, suggesting efficient chemopreventive actions for HRT by inhibiting adenoma-carcinoma in CRC. In contrast, a few epidemiological studies have failed to detect associations between the HRT and CRC incidence in postmenopausal females [62][63][64][65].
The effects of combined estrogen and progesterone therapy and CRC development in postmenopausal females was also studied, and data from the Women's Health Initiative revealed that dual hormonal therapy in postmenopausal females was associated with a reduction in CRC incidence by 38% [53]. Similarly, Chlebowski et al. (2004) [28] has demonstrated that combining conjugated equine estrogens at 0.625 mg/day and medroxyprogesterone acetate (MPA) at 2.5 mg/day decreased the incidence of CRC by 40% in postmenopausal women aged between 50 and 79 years. In contrast, treatment with conjugated equine estrogens alone showed no effects on CRC incidence [28], suggesting that estrogen monotherapy has limited anticancer activity, whereas the addition of progesterone could have more potent chemopreventive actions against CRC [28,66]. Although Labadie et al. (2020) [61] reported that estrogen, both alone and in combination with progestin, reduced the risk of CRC, they proposed that the protective roles were mediated by E2, not progestin, in CRC.
Hence, a better understanding of the roles of female sex steroid hormones and their related receptors in CRC could be a promising approach for the prevention, early diagnosis, and/or treatment development of CRC, thus improving treatment and survival outcomes, as well as the quality of life in patients diagnosed with CRC.

Review Aim
This review, therefore, aimed to summarize the currently available preclinical experiments that focused on the potential beneficial effects of female sex steroid hormones (i.e., estrogen and progesterone) and their related receptors in the treatment of colon cancer cell lines/animal models; additionally, this review aimed to determine the molecular mechanisms underlying their tumoricidal actions against colon neoplasia in both genders.

Method
A literature search was carried out in PubMed for obtaining information related to the therapeutic strategies and potential actions of female sex steroid hormones (estrogen and progesterone, alone and in combination) and their related receptors in colon cancer based on preclinical studies. The keywords for the literature search were the following: colon/colorectal cancer, female sex steroid hormones, therapeutic potential effect/molecular mechanisms of estrogen, estradiol, estrogen receptor, progesterone, progesterone receptor, endogenous expression of estrogen receptor, endogenous expression of progesterone receptor, in vitro human male and female CRC cell lines, and in vivo CRC animal models.

Endogenous Expression of ERα/β in Colonic Tissues and Cells
Earlier studies demonstrated that ER has endogenously low expression in normal colonic and malignant colonic tissues, but without determining the receptor isoforms [67][68][69][70]. Consequent studies then noted that ERβ is the most abundantly expressed receptor in normal colonic tissues [71][72][73]. However, both ERα and β were detected in CRC clinical samples, animal models, and in various colon cancer cell lines, with controversial results regarding their roles in colon carcinogenesis. In detail, in vivo studies revealed that deletion of any of the ER subtypes in mice promoted CRC development and progression, suggesting that both receptors are essential for maintaining the integrity of colonic tissue [74]. In contrast, the ERα subtype is highly expressed in clinical malignant colonic specimens, proposing that the receptor expression status could be a potentially sensitive and specific prognostic marker [75]. Cavallini et al. (2005) [76] has also found that estrogen-receptor-related receptor-α (ERRα) was highly expressed in tumorous tissues compared with their corresponding adjacent normal colon tissues. Others also demonstrated that ERα produced different activities during colon carcinogenesis, which could be dependent on the existence of several receptor splice forms, among which the ERα46 induced antiproliferative and apoptotic actions, whilst the shorter variant, ERα36, was promoted CRC progression [77]. The authors have also suggested that ERα46 receptor isoform could represent a potential target for CRC treatment [77].
Remarkably, cancerous colon tissues had reduced expression of ERβ subtype compared with normal colon mucosa [71,[78][79][80][81]; additionally, the receptor status correlated negatively with poor differentiation [79], advanced Dukes' staging [18], and depth of invasion [82]. Therefore, these studies suggested that estrogens may induce antitumorigenic effects through β, not α, receptors in colonic tissue [71,72,80,[83][84][85][86]. In agreement, Topi et al. also found that high ERβ expression in female CRC patients associated independently with improved prognosis [87] and lower metastases [88] compared with patients with low ERβ expression. Thus, ERβ seems to have chemopreventive actions against CRC development and progression, which may provide a ground for developing alternative diagnostic tools and novel approaches for the prevention and/or treatment of CRC.
Additionally, ERβ was naturally expressed in nine immortalized human male (Table 1) and three female ( Table 2)
On the other hand, Tanaka et al. (2008) [95] has reported the expression of PRB but not PRA receptors in normal and neoplastic colonic tissues, as well as in human female and male colon cancer cell lines (HT29 and HCT116). Other investigators showed that PR is highly expressed in colorectal tumor tissues relative to normal colonic tissues, suggesting that PR could play a role in colon carcinogenesis [98,99]. Interestingly, the most recent work by Zhang et al. (2021) [96] showed PR expression in CRC patients correlated inversely with tumor's size, differentiation, vascular invasion, and advanced clinical stages [100]. The authors also reported that low P4 levels and PR expression in CRC malignant tissues are linked with poor prognosis, suggesting that the hormone could inhibit cancer progression via a PR-mediated mechanism [96].

Preclinical Studies of Female Sex Steroid Hormones and Their Receptors in CRC
Most of the available preclinical studies on the potential therapeutic effects of female sex steroid hormones and/or their related receptors in CRC are in vitro experiments, and few of them confirmed their results using in vivo animal models of CRC. Notably, the potential therapeutic roles of E2 and/or P4 were mainly investigated in several human male CRC cell lines (Table 1), with few using human female colon cancer cell lines ( Table 2). The ages of male patients related to male derivative cell lines ranged between 48 and 73 years (Table 1), and the female patients' age of female derivative cell lines was 44 years (Table 2).

DLD-1 Human Male Colon Cancer Cell Lines
DLD-1 cells (Table 1) endogenously express ERβ1 isoform and Erα [90,94]. Fiorelli et al. (1999) [90] initially showed that treating DLD-1 cells for 48 h with E2 (10 nM) inhibited the cells proliferation and increased progesterone-specific binding to its receptor (Table 3). E2 treatment at (10 nM) for 24 h or 30 h in DLD-1 cells also inhibited the cells proliferation [40,100], induced cell cycle arrest at the sub-G 1 phase [100], and triggered apoptosis by increasing cysteine-aspartic-acid-specific protease 3 (caspase-3) production and poly ADP-ribose polymerase (PARP) cleavage [40,91,100,101] (Table 3). Furthermore, the data demonstrated that the actions of E2 in DLD-1 cells were ERβ-dependent, which activated the p38 mitogen-activated protein kinase (p38/MAPK) pathway [40,91,100,101] (Table 3). In contrast, Fiocchetti et al. (2015) [102] also showed that E2 treatment at 10 −8 M concentration for 24 h upregulated neuroglobin (NGB), which later reallocated into mitochondria, thus increasing oxidative stress, suppressing the release of cytochrome c into the cytosol, and inhibiting cell apoptosis (Table 3). A different therapeutic strategy was applied by Bielecki et al. (2011) [103], who investigated the effects of soy isoflavones (genistein, daidzein, and glycitein) at 25 µg/mL for 24 h in DLD-1 cells with overexpression or knocking down of ERβ gene. The soy isoflavones induced cell cycle arrest at the G 2 /M phase by decreasing the expression of cyclin A and cyclin-dependent kinase 4 (CDK4) and increasing the expression of cyclindependent kinase inhibitors as p21 (Waf1/Cip1) in ERβ-transfected cells [103]. In addition, the soy isoflavones also suppressed the expression of proliferating cell nuclear antigen (PCNA), extracellular-signal-regulated kinase (ERK1/2), alpha serine/threonine protein kinase (AKT), and nuclear factor κB (NF-κB) [103]. Remarkably, the anticancer actions of soy isoflavones were inhibited in DLD-1 cells, lacking ERβ expression, suggesting that the receptor plays a crucial role in mediating the antiproliferative effects of isoflavones in CRC [103]. On the other hand, silencing the expression of ERα effectively inhibited proliferation and colony formation in DLD-1 cells and reduced the tumorigenic capacity in a xenograft mice model [104]. Mechanistically, the deletion of ERα in DLD-1 cells arrested the G 1 phase of the cell cycle by decreasing the cyclin-dependent kinase 2 (CDK2) activity and hyperphosphorylated state of the retinoblastoma (Rb) protein alongside increasing the expression of p21 Cip1 and p27 Kip1 [104].

HCT116 Human Male Colon Cancer Cell Lines
The HCT116 cells (Table 1) endogenously express ERβ 2-5 isoforms, whilst being negative for ERα expression [73,90,94]. The initial studies in 1999 showed an antiproliferative effect for E2 in HCT116 [90,97], while more recent studies found that treating HCT116 cells with E2 at 1-500 nM for 96 h [94] or at 20 nM for 48 h [95] had no effect on cell proliferation ( Table 4). The contradictory results between these studies could be due to the different doses of treatment, cell culture conditions, or methods of proliferation assessment [90,94,95,97].
In contrast, a study by Tu et al. (2012) [106] also found that transfected HCT116 cells with ERβ alone caused differential effects on proliferation, apoptosis, invasion, and metastasis (Table 4). However, raloxifene treatment at 10 µM for 48 h in ERβ-transfected HCT116 cells synergistically inhibited cell proliferation, induced apoptosis, reduced invasion, and halted cell migration compared with ERβ alone and/or raloxifene alone [106]. The authors confirmed their findings in a CRC xenograft model [106]. Raloxifene is an ER modulator that either exhibits estrogenic or antiestrogenic effects depending on tissue types [106].
Recent work by Topi et al. (2020) [88] has also demonstrated that an ERβ-selective agonist (ERB-041) induced anticancer activities in HCT116 cells by inhibiting cell proliferation, provoking apoptosis through the action of caspase-3, and decreasing cell migration (Table 4). Mechanistically, this action occurred by increasing the expression of ERβ and decreasing the nuclear expression of β-catenin, cysteinyl leukotriene receptor 2 (CysT 2 R), and 15-hydroxyprostandlin dehydrogenase (15-PGDH), which consequently reduced the expression levels of cyclin D1 and c-MYC [88]. The authors also used a zebrafish xenograft model, and treatment with ERB-041 alone for 48 h reduced distant metastasis compared with the untreated control group [88].   Symbols: ↑-increase; ↓-decrease; #-no effect; grey shading indicates in vivo experiment.
Another study by Bernatchez et al. (2013) [104] reported marked in vitro inhibitions in cell proliferation and colony formation following silencing the expression of ERα subtype in HCT116, as well as significant reductions in tumorigenic capacity using s xenograft mouse model injected with HCT116 cells (Table 4). The ERα-depleted HCT116 cells also showed cell cycle arrest at G 1 and S phases following decreases in CDK2 activity and hyperphosphorylated state of the Rb protein with concurrent increases in p21 Cip1 and p27 Kip1 proteins [104]. ERα inhibition also reduced the expression of essential genes related to glycolysis, tricarboxylic acid cycle, lipid synthesis, glucose incorporation, and glucosemediated lipogenesis [104]. In agreement, Zhou et al. (2018) [107] reported that suppression of ERRα in HCT116 cells by the FDA-approved ERRα activity inhibitor simvastatin, at 10 µM for 48 h, inhibited cell proliferation and colony formation by reducing the expression of c-MYC and cyclin D1 proteins (Table 4). In addition, the researchers proclaimed that simvastatin also enhanced the cytotoxicity of trametinib, an MEK inhibitor (50 nM), when used in combination for 48 h [107]. Consequently, they also showed that the combination of both drugs had synergistically inhibited cell proliferation and survival in a xenograft mouse model following subcutaneously implanting HCT116 cells transfected with ERRα genes [107]. Altogether, the findings of earlier studies suggest that ERRα could have oncogenic activities in CRC cells and suppressing its expression alone or combined with treatment with MEK inhibitor could represent promising efficient therapeutic strategies against CRC [107].

SW480 and SW620 Human Male Colon Cancer Cell Lines
The SW480 cells were derived from a primary tumor collected from a 51-year-old male patient diagnosed with Dukes' type B CRC, whereas the SW620 cells were generated from lymph node metastases (Dukes' type C) from the same patient of SW480 cells [89] (Table 1). Regarding the endogenous ERs expression, SW480 cells express ERβ and lack ERα [94]. In contrast, the SW620 cells express both ERα and ERβ [73]. The current studies found that 10 nM E2 treatment for 24 h following insertion of ERβ in the SW480 cells inhibited cells proliferation [97], arrested cell cycle at G 0 /G 1 phase [97,105] and induced anti-inflammatory response via reducing the IL-6 expression [105] (Table 5). Mechanistically, inhibition of proliferation and/or cell cycle arrest was induced by decreases in MYC [96,108], MYB, and PROX1 [105] oncogenes; additionally, inhibition caused increases in p21 (Waf1/Cip1) and p27 Kip1 expression [97]. In 2013, Edvardsson et al. [108] also demonstrated that the reexpression of ERβ with E2 treatment regulated the microRNA (miRNA) pool in SW480 cells, and ERβ-induced downregulation of MYC oncogene strongly suppressed the oncogenic miR-17-92 cluster and miR-200a/b, which then promoted cell death upon DNA damage in ERβ-expressing SW480 cells [108]. Furthermore, they also determined that the suppression of miR-200a/b decreased the expression levels of E-cadherin, which then impaired cell migration in the ERβ-expressing SW480 cells [108]. These studies demonstrated that exogenously re-expressed ERβ in combination with E2 treatment could act as an anticancer strategy in SW480 cells [97,105,108]. On the other hand, inserting the ERβ gene in the SW620 metastatic CRC cells showed no effects on MYC and miRNAs [108], suggesting that E2 treatment following ERβ activation could be more beneficial for preventing and/or treatment of early stage, but not advanced stages, colon cancers.
In contrast, Nguyen-Vu et al. (2016) [109] reported no effects following E2 treatment in the SW480 (Table 5) and SW620 (Table 6) cells, suggesting the E2-induced anticancer activities could be dependent on the expression status of ERβ in these cell lines [109]. Moreover, the authors found that ERβ transfection in the SW480 and SW620 cells increased the levels of miR-205 and suppressed the expression of PROX1 oncogene, which subsequently inhibited cell growth, arrested cell cycle at the G 0 phase, and increased cell adhesion [109] Interestingly, ERβ re-expression along with high expression levels of miR-205 also reduced invasion and/or metastatic potential, particularly in SW620 metastatic colon cancer cells [109] (Table 6).   Symbols: ↑-increase; ↓-decrease; #-no effect.
Another promising treatment strategy has also been suggested by Zhou et al. (2018) [107], who transfected the SW480 cells while with ERRα isoform followed by simvastatin treatment (ERRα activity inhibitor) at 10 µM for 48 h; which effectively reduced cell proliferation and colony formation via decreasing the expression of c-MYC and cyclin D1 [107]. Simvastatin also enhanced the cytotoxicity of trametinib (MEK inhibitor) at 50 nM when used in combination for 48 h in the SW480 cells [107] (Table 5). Collectively, there is a compelling need to explore the potential therapeutic effects of E2 and its receptors in the SW620 metastatic cells, since only a few studies are available (Table 6).

LoVo Human Male Colon Cancer Cell Lines
LoVo cells (  [73] found that LoVo cells expressed with both ERβ6 and Erα. Among the current studies (Table 7), only one study showed that E2 treatment had no effects on the proliferation of LoVo cells [94]. The remaining works found that E2 treatment at 10 −8 -10 −6 M doses [110] or 10 nM [90] for 48 h efficiently reduced the proliferation of LoVo cells (Table 7). Lointier et al. (1992) [110] showed that progesterone and androgen hormones were less effective than E2 treatment. Consequently, a study by Hsu et al. explored the functions of the E2 and its receptors and determined the potential molecular targets of the E2-ERβ complex in LoVo cells (Table 7). Following transfecting LoVo cells with ERβ [111] or ERα [112], treatment with E2 at 10 -8 M for 12-16 h effectively inhibited cell proliferation, arrested the cell cycle at G 1 and S phases via increasing the expression of p21 (Waf1/Cip1) and p27 Kip1 , decreased β-catenin, cyclin D1/E and Rb gene and protein expression, and induced apoptosis and DNA fragmentation through activating the caspases-8, -9, and/or 3 [111,112] (Table 7). Remarkably, the anticancer action of ERβ with E2 was mediated through upregulating the expression of p53 [111], whereas ERα with E2 was propagated by increasing the expression of TNF-α signaling [112] in LoVo cells (Table 7).

HCT8 Human Male Colon Cancer Cell Lines
HCT8 cells (Table 1) endogenously express ERβ 1-5 isoforms and lack ERα [90]. Earlier work showed that treating the HCT8 cells with E2 at 10 nM for 48 h had no effects on cell proliferation [90] (Table 8). In contrast, Martineti et al. (2005) [117] reported that E2 at 10 nM for 24 h following ERβ transfection in HCT8 cells inhibited cell proliferation, and induced cell cycle at G 1 and S phases via reducing cyclin E and increasing p21 Waf1/Cip1 expression (Table 8). An explanation for the contradictory results reported could be the different treatment conditions and overexpression of ERβ by gene insertion.   [116] Symbols: ↑-increase; ↓-decrease; #-no effect. Symbols: ↑-increase; ↓-decrease; #-no effect.

HT29 Human Female Colon Cancer Cell Lines
HT29 cells (Table 2) were reported to express ERβ1 isoform, but not ERα [72,94]. In contrast, another study showed that HT29 cells expressed both estrogen receptors (ERβ and ERα) [73]. Earlier studies disclosed that E2 treatments at 1-500 nM for 96 h [94] or at 20 nM for 48 h [95] did not inhibit proliferation in HT29 cells (Table 10). Although the reasons behind the lack of antiproliferative effect in HT29 cells are not entirely understood, a possible explication could be due to the lower ERβ gene expression in these cells compared with HCT116 and SW480 male CRC cells lines [94]. In addition, the expression of ERα in HT29 cells could underly the proliferative effects of E2 [73], since the receptor contributes significant roles to cell growth. In support to the previous suggestion, treating human female CRC cell lines (HT29 and COLO320) with the ERα inhibitor tamoxifen significantly inhibiting proliferation [94].
Although Hartman et al. [97] also reported that ERβ transfection in HT29 cells followed by 10 nM E2 treatment for 24 h did not inhibit cell proliferation, it induced cell cycle arrest at G 1 phase by decreasing the expression of c-MYC, F-box, and p45 Skp2 proteins, with concurrent increases in p21 (Waf1/Cip1) and p27 Kip1 expression. Likewise, more recent studies that utilized a similar treatment strategy in HT29 cells showed cell cycle arrest with reduced expression of MYC, MYB, and PROX1 oncogenes, as well as the proinflammatory cytokine IL-6 [105]. Moreover, Nguyen-Vu et al. (2016) [109] found that the transfection of HT29 cells with ERβ directly increased the levels of miR-205 and suppressed the expression of PROX1 oncogene, which subsequently inhibited cell proliferation, induced G 0 phase arrest, increased cell adhesion, and reduced cell invasion and metastasis. Recently, Topi et al. (2020) [88] used a zebrafish xenograft model that was injected with HT29 cells with ERβselective agonist (ERB-041) monotherapy for 48 h, and their results showed reductions in distant metastasis compared with the control group (Table 10) [88].
Other research has transfected the HT29 cells with ERα46 form followed by treatment with E2 at 10 −8 mol/L for 48 h, which subsequently inhibited the cells proliferation, arrested the cell cycle at the G 0 /G 1 phase, and induced apoptosis [77]. Indeed, ERα could produce different actions in the colon carcinogenesis, dependent on the existence of several splice forms of the receptor; among these, ERα46 has been shown to induce antiproliferative and apoptotic activities, while the shorter variant, ERα36, was associated with tumor progression in the colon [77]. Furthermore, the HT29 cells expressed low levels of ERα46 mRNA as determined by RT-qPCR [77]. Clinically, the expression of ERα46 mRNA levels is low in human colorectal tumor tissues compared with the normal adjacent colorectal tissues [77]. Symbols: ↑-increase; ↓-decrease.  Symbols: ↑-increase; ↓-decrease; #-no effect. Grey shading indicates in vivo experiment. Bernatchez et al. (2013) [104] used another strategy and silenced the expression of ERRα in HT29 cells, showing that it could delay the cells' transition from the G 0 /G 1 phase to the S phase via decreasing the expression of CDK2 activity and hyperphosphorylated state of the Rb protein, along with increasing the expression of p21 (Waf1/Cip1) and p27 Kip1 . They suggested that silencing the expression of ERRα could be a promising therapeutic strategy for female CRC cells [104]. In fact, ERRα has appeared as a transcriptional metabolic regulator, promoting many processes linked with cancer development and progression [120]. High expression of ERRα is correlated with poor prognosis for the cancers of the ovary, breast, prostate, endometrium, and colon [120]. Overall, the available data suggest and re-expression or hyperactivation of ERβ and/or ERα46, whereas the inhibition/blocking of ERRα could provide promising effective therapeutic strategies for treating CRC female patients.

The Effect of Progesterone and/or Its Receptors
The available data on the roles of natural progesterone (P4) and synthetic progestins and/or their receptors in colon carcinogenesis is limited compared with estrogen and its receptors. Although synthetic progestins are bioidentical to endogenous P4, they differ in their chemical structures [121]. One of the most used synthetic progestins in OC and HRT is medroxyprogesterone acetate (MPA) [28,121,122]. Although physiologic MPA has high affinity to progesterone receptors, the drug could also bind other steroid receptors, including glucocorticoids, androgen, and mineralocorticoid receptors [123]. MPA has been successfully applied for treating advanced breast, prostate, and endometrial cancers [122]. The effects of MPA and/or P4 were studied in several male CRC cell lines (HCT116, COLO205, DLD-1, LoVo, SW480, SW620) and a female colon cancer cell line (HT29), as well as in a few in vivo experimental models (Table 11).
Regarding the expression of PR in CRC cell lines, most male cells (HCT116, DLD-1, SW480) were negative for endogenous PR, whereas the SW620 and LoVo metastatic cells showed differential expression profiles of PR (PR − /low PR/PR + ) ( Table 1). On the other hand, PR was found to be naturally expressed by the male COLO205 and female HT29 cells (Tables 1 and 2). Tanaka et al. (2008) [95] treated in vitro human HT29 female and HC116 male colon cancer cell lines with MPA at 20 nM for 48 h following transfecting the PR gene, and the results revealed potent antiproliferative effects and cell cycle arrest at the G 0 /G 1 phase by decreasing the expression of cyclin E and increasing the expression of p21 WAF1/CIP1 (Table 11). In contrast, natural P4 treatment showed no effects on cell proliferation in PR-transfected HT29 and HCT116 cells [95] (Table 11).
Moreover, another two studies suggested that the anticancer activity of P4 could be enhanced by increasing endogenous expression of PR in colon cancer cells [93,124]. In this regard, the researchers showed that decreased growth in PR + COLO205 male [93] and PR + HT29 female [124] colon cancer cells following treatment with natural progesterone. P4 also induced apoptosis via increasing caspase-8 activity [124] (Table 11). The contradictory effects of MPA and P4 in CRC treatment could be due to the chemical differences, which could produce different actions at the cellular level, binding potencies to PR in colon cancers, treatment conditions, and/or different types of colon cancer cell lines.
On the other hand, Heijmans et al. (2011) [97] found that both MPA and P4 monotherapies had no effects on cell proliferation in HT29, HCT116, SW480, DLD-1, and LoVo CRC cell lines (Table 11). This study also failed to detect significant effects for MPA treatment (25 mg/day; 90 days) by subcutaneous implants on colon tumorigenesis in female rodents [96], which agrees with another recent in vivo study demonstrated no anticancer effects for P4 single treatment in Sprague-Dawley female rats with CRC [125] (Table 11).
In contrast, another in vivo study by Meijer et al. (2018) [126] has shown that MPA monotherapy reduced colon tumorigenesis in a postmenopausal mouse model, but not in fertile female mice, suggesting that synthetic progestin could be an effective chemopreventive only in postmenopausal women. Consistently, more recent research also demonstrated that P4, as monotherapy-inhibited cell proliferation, arrested the cell cycle at the G 2 /M phase, and induced apoptosis in LoVo and SW620 metastatic male colon cancer cells; in addition, it was found to inhibit tumor growth in a xenograft CRC model with implanted SW620 cells [96]. Furthermore, apoptosis was induced by decreasing the expression of BCL2 and increasing the expression of BAX, which consequently activated caspase-3 in the treated cells [96]. In addition, the P4 therapy was associated with increased activity of the JNK pathway via upregulating the expression of antiproliferative and DNA-damage-inducible protein α (GADD45α) in colon cancers [96] (Table 11).
Collectively, the results of the different studies (Table 11) demonstrated the tumoricidal efficacy of MPA and/or P4 in CRC, which could support the use of the natural or synthetic forms of the hormone in the treatment of this common malignancy.

The Combination Effect of Estrogen and Progesterone
Experimental studies on the combinatory effects of estrogen and progesterone in CRC are very limited, and only one in vitro [95] and another in vivo study are available at the present [125] (Table 12). Tanaka et al. (2008) [95] found that synthetic progestins, such as MPA at 20 nM both as monotherapy and in combination with E2, inhibited cell proliferation of human female (HT29) and male (HCT116) colon cancer cells, whereas single hormonal treatments showed no anticancer effects in either cell line [95]. Likewise, the in vivo study by Sasso et al. (2019) [125] found that individual treatments with E2 or P4 had no effects on cell proliferation and apoptosis in colonic tumors induced in Sprague-Dawley female rats [125]. In contrast, the combination of E2 (60 µg/kg) and P4 (10 mg/kg) induced enhanced reductions in cell proliferation and promoted apoptosis by increasing the expression of cleaved caspase-8, caspase-3, and PARP in malignant tissues [125]. The combination treatment was also associated with higher expression of ERβ in colonic tumors [126], suggesting that the anticancer actions of E2 and P4 co-therapy could be dependent on the activation of the ERβ receptor. Table 11. Summary of studies that measured the effects of synthetic progesterone (medroxyprogesterone (MPA))/natural progesterone (P4) and/or its related receptor in CRC in vitro cells/in vivo models in chronological order.    [125] Symbols: ↑-increase; #-no effect. Grey shading indicates in vivo experiment.

Key Findings of Preclinical Studies
The current knowledge about the therapeutic strategies and potential actions of female sex steroid hormones (estrogen and progesterone) in CRC is greatly based on preclinical studies. Therefore, this review can provide a groundwork that will encourage more detailed research at the clinical level, considering the most promising therapeutic strategies of female sex steroids in the treatment of either males or females with CRC.
The most promising therapeutic strategies were E2 exogenous treatment and reactivation of the ERβ subtype, either alone or in combination, or inhibition of the ERα subtype alone (Tables 3-10). These strategies have been effectively exerted multi anticancer effects including antiproliferative, proapoptotic, antiangiogenetic, and/or antimigratory effects, both in in vitro human colon cancer cell lines (mainly for male cells that derived from advanced stages of CRC) and in vivo CRC models. Remarkably, E2 treatment could produce its anticancer actions by mediating the ERβ in CRC cells. The anticancer activities of those strategies have occurred by regulating different genes and/or protein expression, including tumor suppressor genes such as p53 and Rb; oncogenes such as MYC, MYB, and PROX1; cell cycle regulators such as p21, p27, and cyclin A/D/E; caspase cascades such as caspase-9 and -3; migration-related factors such as uPA, uPA, MMP-2/9; and signaling pathways such as p38/MAPK, AKT, ERK1/2, JNK1/2, which are commonly linked with onset and progression of CRC. The in vitro experiments were limited in female colon cancer cell lines, which were less responsive to some therapeutic strategies, particularly E2 monotherapy, than in vitro male CRC cells. This could be due to the overexpression of the ERα subtype.
The other possible therapeutic strategy was the progesterone exogenous treatment (natural or synthetic progesterone) (Table 11), which was also associated with promising tumoricidal actions mediated via PR, while the available studies on progesterone are still limited. In addition, there were some inconsistent findings between the effect of natural progesterone (P4) and synthetic progesterone (MPA), and this could be due to the different nature of progesterone compounds that were used, different patterns of PR expression, or different types of CRC cell lines/models. Nevertheless, the combination of E2 with P4 (Table 12) showed enhanced anticancer activities compared with their monotherapies in male-female cell lines and animals, which could be also another promising therapeutic approach for CRC treatment, but the studies were very limited, and the action mechanisms remain to be further elucidated. Together, the studies suggested that female sex steroid hormones could provide a novel and effective therapeutic method against CRC.

Limitations of Female Sex Steroid Hormones in Cancer Treatments
Female sex steroid hormones can exert pro-or antitumorigenic actions depending on the origin of the cancer type, the endogenous expression patterns of their receptors in the tissue, and the stage of cancer [127,128]. In this regard, the actions of E2 in CRC have shown to be receptor-dependent actions; ERα exerts oncogenic actions, whilst ERβ exerts antitumorigenic actions [129]. In this context, E2 therapy promoted cell cycle progression and cell survival via ERα-mediated mechanisms, while inducing apoptosis via activating caspase-3 by ERβ in DLD1 CRC cells [130]. Moreover, ERα and ERβ presented opposing effects on cell cycle progression in HeLa human cervical cancer cells via increasing and decreasing the gene expression of cyclin D1, respectively [131]. Paterni et al. (2014) [127] reported that selective stimulation of ERβ could be as effective as a preventive and/or therapeutic approach against some types of cancers, such as CRC. Because estrogens have been shown to inhibit proliferation and induce apoptosis in CRC, these protumorigenic actions are believed to be mediated through the ERβ, which is a predominant estrogen receptor subtype in the colon [73,132]. The expression of ERβ is decreased during colon carcinogenesis, as discussed in Section 2.1 [38,71,79,128].
Notably, genetic studies have also found that both ERβ and PR pathways interacted together in CRC, and PR antitumorigenic actions were also dependent on the activities of ERβ in malignant tissues [133]. Furthermore, concomitant E2 and P4 dual therapy markedly increased the protein expression of ERβ, PR, and caspase-3 alongside cell apoptosis, whereas it downregulated cell proliferation markers, compared with single hormonal therapies in an ovariectomized female rat model of CRC [125]. However, none of the prior studies measured the potential antitumorigenic effects of E2 and P4 sequential treatment in CRC, although this protocol mimics normal endocrinology in reproductive-age females. Most current studies suggested that E2 and/or P4 could be alternative therapeutic strategies against CRC, and their effectiveness could be dependent on the expression profiles of ERs and PR in colon cancer tissues [125,[133][134][135].
Indeed, the molecular pathways mediated via ERs and PR interact together, and the activation of one receptor could influence the activities of the other in neoplastic diseases [136]. In this context also, the activation of ERβ [134] and PR [135] inhibited the growth of breast cancer cells by suppressing the ERα-induced oncogenic actions. In contrast to colon cancer, the estrogens (either endogenous or exogenous) have been shown to induce cell proliferation and tumor development in the breast and uterus, and these protumorigenic effects are supposed to be mediated via the ERα, which is predominant in these tissues [132]. In fact, female sex steroid hormones and their receptors play a vital role in breast cancer development and progression [137]. Almost 70% of breast cancers are hormone-receptor-positive cancers (ER + and/or PR + ), which are correlated to cancer cell growth, progression, and spread [137]. The PR modulates ERα action and an upregulated target gene of ER, and its expression is dependent on estrogen in breast cancer [137]. Thus, ERα-antagonists have currently proved to be effective agents for reducing/inhibiting cell proliferation in hormone-sensitive breast cancer [137].
Normally, ERα is found mainly in the mammary gland, uterus, ovary (thecal cells), prostate (stroma), male reproductive organs (testes and epididymis), bone, liver, and adipose tissue [127]. While ERβ is present mainly in the colon, ovary (granulosa cells), prostate (epithelium), bladder, adipose tissue, and immune system [127], the alpha subtype has a more prominent role on the mammary gland and uterus, and induces cell proliferation in tissues including breast and uterus. The beta subtype generally counteracts the ERαinduced cell hyperproliferation in breast and uterus tissues [127]. The actions of estrogens are modulated by the ERs subtypes (ERα and ERβ), and estrogen-mediated ERs subtypes can exert beneficial effects in some types of cancers, such as colon cancer; however, they can have adverse and severe effects in breast and uterus tissues [127]. Thus, more profound knowledge of the various biological activities of ERα or ERβ in different tissues would help in the decision of which receptor subtype is the best to use for the management or cure of cancers [127]. Deli et al. (2019) [138]. have also classified the oncologic risk (recurrence and progression) of hormonal replacement therapy (HRT) in cancers for females, as follows: (a) HRT is advantageous for colon cancer, endometrial cancer type I, cervical adenocarcinoma, hepatocellular cancer, hematologic malignancies, and local cutaneous malignant melanoma; (b) HRT is neutral and thus it should not be denied for BRCA 1/2 mutation carriers without cancer, endometrial cancer type II, uterine carcinosarcoma, and adenosarcoma, certain types of ovarian cancer, cervical cancer, kidney cancer, pancreatic cancer, and thyroid cancer; (c) HRT is relatively contraindicated, and thorough, individualized decision making is essential for certain types of ovarian tumors, brain tumors, leiomyosarcoma, advanced metastatic malignant melanoma, lung cancer, gastric cancer, and bladder cancer; (d) HRT is disadvantageous and harmful for hormone-dependent cancers, such as breast cancer, endometrial stroma sarcoma, hormone-receptor-positive (ER + /PR + ) gastric and bladder cancer, meningioma, and glioma [138].
Overall, it is essential to consider the hormonal factors during cancer occurrence before using any hormonal therapy, as well as evaluate the exposure patterns to endogenous sex hormones, such as age at menarche and menopause, years of periods/fertility, number of pregnancies and births, duration of breastfeeding, and other related factors; iin addition, it is important to evaluate the exposure to exogenous hormones, such as HRT, OC, or fertility treatments [128,138].

Conclusions
The roles of female sex steroid hormones (estrogen and progesterone) and their nuclear receptors in CRC represent an attractive avenue for developing novel efficient alternative hormonal therapies. Most of the currently available data suggest that E2 could induce cancer progression via ERα whilst promoting ERβ-mediated anticancer activities. Possible approaches may involve overexpression and/or selectively stimulating ERβ, with and without selective inhibition of ERα activities in male and female CRC patients. Similarly, natural and synthetic progesterone showed promising tumoricidal actions that were mediated via PR in male and female experimental models. Nevertheless, the combination of E2 with P4 appears to be a more effective strategy, relative to individual hormonal therapy, both in vitro and in vivo. Hence, additional research is needed to explore the therapeutic values of E2 and P4 combined therapy, with and without androgen deprivation therapy, in colon cancer. Acknowledgments: The author would like to thank the Deanship of Scientific Research at Umm Al-Qura University for supporting this work by grant code 19-MED-1-03-0007. In addition, the author would like to thank Bassem Refaat (barefaat@uqu.edu.sa) from Umm Al-Qura University of Faculty of Applied Medical Sciences for the supervision during the preparation of the review and reviewing the paper at the final stage.

Conflicts of Interest:
The author declares no conflict of interest.