Sex steroids and colon cancer development: what do we know?Epidemiology

Epidemiological data indicate that colorectal cancer (CRC) is the third most common cancer worldwide and the second leading cause of mortality. Risk factors for CRC include dietary habits, inflammatory bowel diseases, metabolic disorders, alcohol consumption, and tobacco smoking [5]. Interestingly, men have a greater incidence of CRC than women do, suggesting that sex is a contributing factor. This greater incidence can be attributed to various causes, categorized into sex and sexual differences, including disparities in behavior and physiological conditions. Moreover, sexual dimorphism is consistently observed across different ethnicities and geographical locations, indicating that intrinsic factors play a significant role in explaining this phenomenon. One of the key intrinsic factors that differs between the sexes is the concentration of sex steroids, although not the only factor. Supporting this idea, women between 18 and 44 years old have a higher survival rate than men of the same age or older women (over 50 years old) [6]. Interestingly, women have a greater risk of develop tumors in proximal colon than men. Also, the incidence of CRC in proximal colon increases with age; in men, the effect of age is less significant [7]. These data suggest that estrogens protect against CRC in younger women against the develop of left-sided colon neoplasia. In line, the exposition to estrogens by hormonal replacement therapy or pregnant reduce the risk of microsatellite instability, which is more frequently in left-site compared with right-sided colon tumors [8].

Compared with healthy individuals, CRC patients exhibit dysregulation of serum sex steroid levels, although the exact role of sex steroids has not yet been determined. For instance, postmenopausal women with CRC have higher levels of estradiol (E2) and estrone (E1) and a greater ratio of testosterone (T) to E2 than control patients do; however, this pattern is not observed in men [9, 10]. In normal and neoplastic colon tissue, the expression of estrogen receptor beta (ER-β) is predominating respect with ER-α and their expression is loss in advance stages of CRC disease. Also, the loss of expression is associated with poorer survival rate in both sexes [11]. These data suggest that estrogens protect against CRC due their interaction with ER- β. In line, postmenopausal women with hormonal replacement therapy have a lower risk of develop CRC in ER-β positive cases but not in ER-β negative [12]. Also, in preclinical studies with animals deficient of ER- β increase the number and size of colon tumors [13]. On another hand, the higher expression of ER-α is associated with poor prognosis (less overall survival, tumor differentiation, tumor invasion, lymph node status and Dukes stag) [14, 15]. When ER- β is expressed, E2 reduce the ER- α protein levels which could explain the predominance of ER- β in normal colon epithelium and that in neoplastic tissue there is a reduction of ER- α [16].

Moreover, men exhibit an increase in the level of ER-α [17]. Conversely, the expression of aromatase, an enzyme responsible for converting T to E2, is greater in neoplastic tissue in both men and women, and this expression is associated with the proliferation index, estrogen concentration, and lower survival rate [10]. Consistently, intratumoral levels of E2 are greater in male patients with CRC than in healthy controls [18].

The relationship between androgens and CRC has been poorly studied. Contrary to expectations, the circulating level of T is inversely associated with overall survival and mortality in CRC patients but only in men [19, 20]. A reduction in androgen levels is associated with fewer CAG repeats in the androgen receptor (AR) gene, which subsequently increases transcriptional activity [21]. AR is overexpressed in colon tumor tissue and is associated with tumor size, differentiation, and distant metastasis [22]. In contrast, this receptor is not expressed in the nonneoplastic mucosa. Interestingly, in postmenopausal women, a few CAG repeats in the AR gene and a CA repeat in the ER-β gene are associated with high serum androgen levels, suggesting that these polymorphisms have a stimulatory effect on T production in women [23]. Although further research is needed, deregulation of sex steroid concentrations, expression of sex steroid receptors, and enzymes involved in sex steroid metabolism have been observed in CRC patients (Table 1).

Table 1 Epidemiological, in vivo and in vitro studies about the role of sex steroids in colorectal cancer development. Arrow up indicates stimulation and arrow down inhibition

Therefore, these genes may be potential markers for CRC, which is an important step in understanding the dimorphic nature of this neoplasia. This finding allows for the proposal of more targeted treatments based on sex. These associations suggest that sex steroids play a role in the development of CRC and that their effects differ between men and women. Thus, to elucidate the mechanisms through which sex steroids influence the development of CRC, it is necessary to analyze both in vivo and in vitro studies.

In vivo

According to epidemiological data, different animal models, such as ICR (outbred mouse), C57BL6 (endogamous mouse), and PIRC (rats naturally susceptible to developing adenomas), develop more and larger tumors than their female counterparts. Various analyses have been conducted to elucidate whether sex steroids play a role in this susceptibility and to determine their specific effects. The azoxymethane (AOM)/dextran sodium sulfate (DSS) model of colitis-associated cancer is a consistent, reproducible, and relatively inexpensive initiation-promotion model that utilizes chemical induction of DNA damage followed by repeated cycles of colitis [41]. Compared with control female, ovarietomized (OVX) female ICR mice treated with AOM/DSS developed more tumors but of equal size, and reconstitution with E2 inhibited this effect [42]. It is important to note that tumor size is important in colorectal cancer (CRC) because it is associated with cancer stage, tumor aggressiveness, and distant metastasis [43]. Despite this, OVX animals exhibited greater tissue damage than control animals, which was not observed in the reconstituted group, and only animals treated with E2 presented lower concentrations of IL-6, Cox2, and TNF-α [42]. In contrast, in PIRC rats, neither ovariectomy nor reconstitution with E2 influenced the number of adenomas [26]. These findings suggest that E2 participates in the maintenance of colon architecture but has a minor role in the pathophysiology of CRC in females.

Interestingly, E2 appears to have different effects on males and females. Male ICR mice had more tumors than male mice treated with E2 and females. Additionally, treatment with this hormone significantly reduce the transcription of genes involved in tumor progression pathways, such as NF-κB, NRF2, and NLRP3 25. NRF2 is a crucial in the response to oxidative stress under normal conditions and participate in the colorectal cancer progression. In AOM/DSS model NRF2 knockout mouse increase the incidence in tumor formation that is associated with an elevated oxidant factors such as COX-2 [44]. In vitro studies shown that the activation of NRF2 inhibits the activation of NF-κB increasing apoptosis in colon cancer cell lines [45]. Interestingly, the expression of the genes of these pathways (NF-κB, NRF2, and NLRP3) is different between males and females only when treatment with AOM and DSS is administered and not basally. In early stages of induction of tumors (2 weeks) males have higher levels of NF-κB related genes (iNOS, COX-2, IL-6 and TNF-α) than females and lower levels of NRF2 and some related genes of antioxidant factors (NQ-O1). This suggest that early inflammation help to the susceptibility in males. When E2 is administrated in males, the levels of NF-κB related genes diminish and the expression of NRF2 related genes increase it [25]. These results indicate that E2 reduces inflammation from the early stages of colorectal tumor development. On another hand, in advance stages of the induction of tumors (10 weeks) males have higher levels of NRF2 and its related genes than males treated with E2 and females. These suggest a double role of these pathways in the pathogenesis of CRC [25]. This is in line with other studies that suggest that the activation of NRF2 specifically in tumors cells increase their tumorigenicity, survival, growth and chemoresistance [46]. In the case of NLRP3, males exhibit greater expression of NLPR3-related genes (IL-1β and IL-18) than males treated with E2 and females. IL-1β and IL-18 trigger a chronic inflammatory process promoting the formation of tumors [47]. The treatment with E2 in males results in a downregulate expression of NLPR3 and its effectors (IL-1β and IL-18) [25].

Treatment with T has different effects on androgen levels. Orchiectomy (ORX) in C57BL6 mice drastically reduces the number and size of induced tumors. Intraperitoneal propionate of T slightly increased the number of tumors in males and the size only in females, but it did not completely reverse the effect caused by orchiectomy. At the molecular level, factors associated with the NFκB and NRF2 pathways, such as iNOS, COX-2, and NRF2, are expressed at lower levels in ORX mice and females than in control males [37].

On the other hand, in the AOM/DSS-treated BALB/c mouse group and in the APC+/- mouse group (unspecifed sex), mice subcutaneously injected with a complex of T-albumin (the target membrane androgen receptor) developed significantly fewer tumors. These animals had lower levels of p-Akt and p-Bad than did the control animals. Additionally, treated mice exhibit more apoptotic cells than healthy controls [38, 39]. The choice of study model and the analysis by sex are important factors to consider when elucidating the role of T; moreover, its metabolism could be considered since T can be converted into E2 or DHT. Interestingly, in ORX PIRC rats reconstituted with DHT, the protective effect of gonadectomy is inhibited [26]. These findings suggest that castration of animals protects against tumor induction due to the bioavailability of DHT.

Consistent with the idea that T could have a protective effect, a lower concentration of T in CRC patients is associated with lower survival and mortality. There are two possible explanations for this difference. First, T is converted to E2, which has a protective effect. The second explanation is that its interaction with the membrane androgen receptor triggers an antitumoural role, which is supported by the in vivo studies analyzed in this section. There is clear modulation of tumor growth by sex steroids (Table 1).

However, additional data are needed to draw definitive conclusions. While estrogen appears to participate in the maintenance of colon architecture and plays a minor role in CRC pathophysiology in females, it has a more drastic protective effect in males. On the other hand, the role of androgens in CRC is complex and dependent on the model and analysis by sex. Orchiectomy provides protection against tumor induction, and T seems to exert a protective effect mediated by the membrane androgen receptor or its conversion to E2.

In vitro

E2 reduces the viability of many colon cancer cell lines and alters important processes, such as migration, motility, and apoptosis. For example, this hormone reduces the viability of colon cancer cells by activating P53, which subsequently upregulate the levels of p21 and p27, that consequently inhibits cyclin D1 gene that reduce proliferation [30]. The reduction in viability of DLD-1 cells caused by incubation with E2 was inhibited by palmitoylation inhibitors such as 2-Br-palmitate. Palmitoylation is the process of biding a protein to and fatty acid and this led the attachment of proteins (such as estrogen receptors) to the plasma membrane. 2-Br-palmitate inhibit fatty acid CoA ligase and other enzymes that reduce the levels of intracellular palmitoyl-CoA [48]. ER-β activates p38 downstream pathway when is binding to the membrane, however If palmitoylation is inhibited, this process does not occur. This finding suggested that the effect of E2 is mediated through a nongenomic pathway involving the activation of the P38/MAPK pathway [31]. Additionally, E2 treatment decreases cellular migration by reducing MMP-2 and MMP-9 levels and inhibiting the JNK 1/2/PGE2 pathway, which is involved in cellular motility [32]. In recent years, researches with a metabolite of E2, 2-methoxyestradiol, indicates thar regulate several biology processes such as proliferation and shown anticancer properties [49]. In this case, this metabolite is capable of increasing apoptosis colon cancer cells by upregulating proapoptotic factors such as P53, Bax, p21, and caspases 3 and 9 and decreasing antiapoptotic factors such as c-Myb and bcl-2 33–35. Moreover, blocking the p38 signal reduces the expression of ER-β, indicating that rapid (nongenomic) is necessary for this slow effect (genomic effect). Experiments with microarrays have shown that ER-β plays an important role in the regulation of transcription factors such as MYC, MYB, RUNX2, and PROX1, which are involved in cell viability, proliferation, apoptosis, and differentiation. This modulation potentially triggers an antitumoral cascade [36].

Androgens, particularly T, also regulate processes in colon cancer cells. Treatment with this hormone stimulates the reorganization of actin filaments, and interestingly, flutamide does not inhibit this effect. Flutamide inhibits the bind of androgens just with cytoplasmatic androgen receptors, suggesting that the pathway is mediated through the membrane androgen receptor [50]. First, T increases FAK protein levels, which triggers a signaling cascade that stimulates actin reorganization. FAK overexpression regulates survival, cellular adhesion, motility, and proliferation, and in cancer, it is associated with advanced stages and metastasis [51]. On the other hand, cytoskeletal reorganization is a marker of early apoptosis in a very complex process [52]. Just an example, the reorganization actin-membrane linker protein ezrin activate CD95 that activate apoptosis, disruption of microfilaments by cytochalasin D induce the translocation of proapoptotic factors, among other [53]. Similarly, T increases apoptosis in CaCo2 cells by reducing PI3K/Rac1, which activates the kinase JKN, stimulating the transcription of proapoptotic genes involved in intrinsic and extrinsic apoptosis [40, 54]. Finally, reorganization of the cytoskeleton reduces the migration and invasion of colon cancer cells. This effect is not influenced by the inhibition of aromatase, suggesting that the effect is directly mediated by T and not by the conversion of T to E2 50; since, this enzyme metabolizes T to E2 and E2 is a final metabolite in the steroidogenesis.

As mentioned above, there are close interactions among the three macrosystems present in the colon. In this section, we analyze certain aspects of the regulatory effect of sex steroids on the development of colorectal cancer. However, most related studies have focused only on analyzing the effects of certain factors on the immune system and have not conducted deeper examinations. Furthermore, these studies failed to explore the effects on the enteric nervous system, which is crucial for obtaining a complete understanding of this neuroimmunoendocrine interaction. Consequently, in the subsequent sections, we will examine the impact of estrogens and androgens on cells of the immune system, the enteric nervous system, and the intestinal microbiota, all of which maintain a close relationship with these three macrosystems.

The role of inflammation in the development of colorectal cancer: a general overview

Inflammatory bowel disease (IBD) has an increased risk of develop CRC. The cumulative risk is dependent of the years with the disease 1%, 3%, and 7% at 10, 20, and 30 years, respectively, while, sporadic is the most common type of CRC [55]. However, some mutation and especially the need for a chronic inflammatory process are similar in both types. In this review, we focus on the research focus in process that occur in IBD patients and in the AOM and DSS animal model, which is more like CRC associated with colitis than to sporadic CRC [56].

Figure 1A represent the early stages of CRC that is characteristic by chronic inflammation where cells of the innate and adaptive immune response participate and promotes the development of tumors growth.

Fig. 1

Pathogenesis of colorectal cancer associated with ulcerative colitis (early stages of CRC). Innate cells and adaptative immune cells increase their production of soluble factors which can trigger a chronic inflammation process. If this state is maintained, the possibility of developing cancer is greater due to an increase in the mutation rate and the loss of intestinal homeostasis. One of the pathway involved in this process is NF-kB that contributes increasing the proliferation of intestinal cells, modulating the production of proinflammatory cytokines and stimulating the expression of receptors such as TLRs and MCH II (A). On another hand, in advance stages of cancer, immune cells change to an immunosuppressor phenotype. Both tumor and immune cells secrete factors that hinder the effective elimination of transformed cells (B). abbreviations: ROS, reactive species of oxygen. COX2, cyclooxygenase 2. NF-kB, nuclear factor kappa-light-chain-enhancer of activated B cells. IL-(6,10, 1β), Interleukin. TGF-β, Transforming growth factor beta. TNF-α, tumor necrosis factor alpha. MHC, Major Histocompatibility Complex. TLR, Toll-like receptors. VEGF, Vascular endothelial growth factor. MMPs, metalloproteinases. This figure was created with BioRender.com

The first cells involved early in pathogenesis are myeloid cells, primarily neutrophils and macrophages, as well as natural killers that increase the production of reactive oxygen and nitrogen species (ROS and NOS), cytotoxic granules, and proinflammatory cytokines [57]. Chronic states of inflammation, such as those observed in patients with IBD, contribute to tissue damage, loss of the epithelial membrane, ulcer formation, and an increased mutation rate, potentially leading to tumor formation. IBD patients exhibit high infiltration of myeloid cells that perpetuate the chronic inflammatory process. The activation of NF-κB is one of the principal mechanisms influencing the proinflammatory environment, as it regulates the release of proinflammatory molecules such as TNF-α, IL-1β, IL-6, iNOS, and ROS, which are elevated in the serum of patients with ulcerative colitis. TNF-α promotes the survival of epithelial cells, the release of protumoral cytokines, and direct disruption of the epithelial barrier (Fig. 1A).

Moreover, the inhibition of TNF receptors reduces tumor formation and the infiltration of neutrophils and macrophages in mice with tumors induced by azoxymethane (AOM) and dextran sulfate sodium (DSS). IL-6 and IL-1β promote proliferation and survival by activating the transcription factor STAT-3 while downregulating the key protein P53, leading to an increased mutation rate. Animal models have shown that IL-6 enhances tumor growth associated with colitis, and this effect can be inhibited by blocking IL-6. ROS and iNOS cause tissue damage and direct DNA damage and stimulate the secretion of proinflammatory cytokines, thereby promoting inflammation and increasing the mutation rate [58] (Fig. 1A).

Adaptive immune cells also participate in the progression of ulcerative colitis toward colon tumors. Dendritic cells connect the innate immune response with the adaptive immune response. In the basal state of the epithelium, these cells exhibit an immunosuppressive phenotype and primarily secrete anti-inflammatory cytokines such as IL-10. Inflammatory bowel disease (IBD) patients exhibit increased infiltration of DCs and increased expression of TLRs, which help activate NF-κB and subsequently promote an inflammatory microenvironment [58]. Conversely, DCs stimulate the migration and differentiation of lymphocytes. IBD patients exhibit increased infiltration of activated lymphocytes. CD8 + lymphocytes directly kill tumor cells by secreting cytotoxic granules and proinflammatory cytokines (TNF-α and IFN-γ). The inhibition of costimulatory molecules that activate these cells (CD80) in AOM/DSS models increases tumor formation [57] (Fig. 1A).

On the other hand, CD4 + lymphocytes play a determinant role in regulating the growth or elimination of transformed cells; however, their function depends on their phenotype. TH1 cells assist the innate immune response in killing neoplastic cells. The Th2 response inhibits the Th1 response by releasing IL-13, IL-21, and IL-25. IL-13 affects epithelial membrane integrity, inducing apoptosis in epithelial cells and preventing their regeneration. The increase in TH2 cells depends on the IL-33 concentration, which is elevated in IBD patients. TH17 cell infiltration is increased in IBD patients and promotes a proinflammatory microenvironment by directly releasing and/or stimulating the production of cytokines such as TNF-α, IL-6, IL-17, IL-22, and IL-21. Blocking IL-21 reduces tumor formation and the production of proinflammatory cytokines (Fig. 1A).

Among Tregs, the phenotype CD4 + Foxp3 + cells inhibit tumor growth in AOM/DSS models, IL17 + Foxp3 + CD4 + T cells are present in higher concentrations in IBD patients, while Foxp3 + RORγt + T cells potentially aid in tumor progression through the secretion of IL-17, favoring an inflammatory microenvironment, and have a greater capacity to inhibit the cytotoxic response of CD8 + lymphocytes, thereby preventing the elimination of transformed cells [57,58,59] (Fig. 1A).

The microenvironment in colon carcinoma differs from that in dysplasia in IBD patients. Both tumor and immune cells secrete factors that hinder the effective elimination of transformed cells. These cells primarily maintain an immunosuppressed state through the synthesis of anti-inflammatory cytokines (TGF-β and IL-10), express inhibitory receptors such as CTL-4 and PD-1 and secrete factors that modulate angiogenesis and disrupt the extracellular matrix, such as VEGF and MMPs. These factors enable cells to invade other tissues and eventually reach other organs, primarily the liver and lungs [60] Fig. 1B.

General mechanisms of sex steroids in the immune system

Estradiol (E2), testosterone (T), and dihydrotestosterone (DHT) are steroid hormones that have both genomic and nongenomic effects on immune cells. These hormones can bind to specific nuclear receptors to regulate gene transcription and protein synthesis. On the other hand, the nongenomic effects of these hormones occur through their interaction with cell membrane receptors, triggering rapid signaling cascades.

In the Fig. 2, we represent the genomic and non-genomic effect of E2, T and DHT and their general influence on immune response. In general, E2 can have a dual role depending of the dose, the tissue and the specific cell target. While, androgens have, in general, an immunosuppressor growth that could be favoring the growth of tumor (Fig. 2).

Fig. 2

General action mechanism of sex steroids. These molecules can exert their effects on cells through binding to cytoplasmic receptors (genomic effect). These receptors translocate to the nucleus and binds to EREs or AREs. The other mechanism is through membrane receptors (non-genomic effect) that activates signaling cascades such as PI3K/AKT, MAPKs and ERK. Both mechanisms modulate process involved on inflammation such as survival, cytokine production, chemokine segregation. The effect of estrogens depends principally on the dose and the type of receptor expressed (alpha or beta) (A). On another hand, in general, androgens exert and immunosuppressor phenotype (B). abbreviations: PI3K: phosphoinositide 3-kinases. MAPKs: mitogen-activated protein kinases. EREs: estrogen response elements. AREs: androgen response elements. ERK: extracellular signal-regulated kinase. AKT: alpha serine/threonine-protein kinase. SRC: proto-oncogene tyrosine-protein kinase. E2: estradiol. T: testosterone. This figure was created with BioRender.com

E2 exerts its genomic effects by binding to estrogen receptors (ER-α and ER-β), which are expressed by several types of immune cells, including T cells, B cells, dendritic cells, natural killer cells, monocytes, neutrophils, and macrophages. When estradiol binds to ERs, it dimerizes and translocate to the nucleus, where it binds to estrogen response elements on target gene promoters and regulates transcription. For example, estradiol upregulates the expression of genes involved in survival (cd22, shp-1, bcl-2, and vcam-1), cytokine production (TNF-α, IL-6, IL-1β, and IL-10), and chemokine segregation (CINC-1, CINC-2β, and CINC-3), among others. Additionally, E2 can regulate the expression of transcription factors that modulate the expression of immune-related genes, such as NF-κB [61, 62] (Fig. 2).

T and DHT exert their genomic effects by binding to androgen receptors (AR), which are expressed by several types of immune cells, including T cells, B cells, dendritic cells, monocytes, neutrophils, and macrophages. Upon binding to testosterone or DHT, as described above, ARs dimerize and translocate to the nucleus, where they bind to androgen response elements on target gene promoters and regulate transcription. For example, T upregulates the expression of genes involved in anti-inflammatory cytokine production (IL-10 and TGF-β) and downregulates the expression of activated molecules (MHC-1 and CD86) and immunoglobulins [63] (Fig. 2).

In addition to their genomic effects, E2, T, and DHT also exert nongenomic effects on immune cells through their interaction with cell membrane receptors, such as G protein-coupled receptors. E2 can activate the MAPK and PI3K/Akt signaling pathways through its interaction with GPCRs, which in turn can regulate T-cell proliferation, survival, and cytokine production [