Traditionally, EC is divided into two subtypes based on its histological characteristics (60). Type I EC is endometrioid EC (EEC), accounting for ~80% of all EC cases, whereas the remaining histological patterns, such as serous and clear cell EC, are classified as type II EC.

Researchers have detected significantly higher expression of SIRT1 in EC cells (ECCs) compared with that in normal endometrial cells, indicating that SIRT1 may contribute to the onset of EC (61,62). However, Bartosch et al (63) investigated SIRT1 mRNA and protein expression by quantitative PCR and immunohistochemistry, respectively, in 76 patients with EC and 30 non-EC subjects, and found that SIRT1 mRNA was underexpressed and SIRT1 protein was overexpressed in ECs. Additionally, significantly lower levels of SIRT1 mRNA were found in type II ECs compared with type I ECs, but no significant associations were found between the SIRT1 protein level and the histological subtype, grade, stage, or lymphovascular invasion. These results suggest that the effects of SIRT1 on the pathogenesis of EC are complex. Although the mechanism through which SIRT1 promotes EC remains unclear, there are significant correlations between SIRT1 expression and risk factors for EC.

EECs are estrogen-dependent, as the exposure to a continuous estrogen overload unopposed by progesterone may lead to endometrial hyperplasia, which may evolve into complex atypical hyperplasia and, eventually, progress into cancer (6,64). The ER pathway is the traditional pathway for 17β-estradiol (E2) (65), which is activated by the binding of E2 to ERα; the E2-ER complexes bind to the promoters of target genes in the nucleus to regulate transcription with the help of coactivators, such as P300, PPARγ and PPARγ coactivator 1α, leading to cancer onset and progression (66–68). The ER pathway is also linked to other signaling pathways, such as the EGFR, PI3K and ERK pathways (69,70). In addition to participating in the ER pathway, E2 can also bind to G-protein coupled receptors (GPCRs) to exert similar effects quickly through a non-classical pathway when ER is not expressed (71).

SIRT1 affects the expression of ERα and the secretion of E2, resulting in the development and maintenance of estrogen-related cancer: E2-ERα complexes promote the expression of SIRT1, which is consistent with the observation that SIRT1 is overexpressed in most ERα-positive breast cancer samples (72). In turn, SIRT1 increases the ERα level to enhance the carcinogenic effect of estrogen, forming a positive feedback loop. Similarly, the SIRT1 inhibitor reduces the levels of ERα and interferes with cellular proliferation (73), and E2-GPCR complexes have a parallel function (74). Most importantly, SIRT1 directly facilitates the secretion of E2 by proliferating the expression and function of aromatase (75). However, several studies have indicated that SIRT1 may suppress the function of ERα (34,76,77).

Other risk factors for ECCs are obesity, diabetes and hypertension (6), and the roles of SIRT1 in these diseases have been described in the literature. SIRT1 has been shown to induce mitochondrial-related gene expression and help maintain a healthy weight in mice (78). In animals fed a high-fat, high-sugar diet, SIRT1 improves insulin sensitivity, maintains a normal glucose level in the plasma, and reduces the complications of diabetes (79,80). Insulin secretion is an ATP-dependent process (81), while SIRT1 can increase the level of ATP by inhibiting UCP2 to promote insulin secretion and activate NeuroD and MafA, two transcription factors that promote insulin synthesis (82,83). In addition, SIRT1 can reduce the levels of low-density lipoprotein and cholesterol, increase high-density lipoprotein, obstruct lipid oxidation, platelet aggregation and vascular smooth muscle cell proliferation, and protect cardiovascular cells (84).

Molecular studies have identified PTEN mutation as the most common gene mutation driving EEC, and the PI3K/AKT pathway is known to be affected by PTEN (6). SIRT1 has been shown to inhibit the function of the PTEN protein by deacetylation, thereby activating the PI3K/AKT pathway (85), indicating that SIRT1 may directly promote the development of precancerous lesions.

Type II ECs are usually estrogen-independent, and age is the main risk factor (6). SIRT1 increases the lifespan and resists aging in disparate species, as it is involved in the morphology and function of telomerase (86). p53 acts as a cancer-driving gene in type II ECs (6), while SIRT1 can limit the expression and function of p53.

Close to 100% of cervical cancer cases are caused by high-risk HPV, with HPV16 and HPV18 being responsible for 70% of the cases (5). HPV is a small double-stranded DNA virus encoding eight proteins (early stage: E1, E2 and E4-E7; late-stage: L1 and L2), among which E6 and E7 are responsible for progression from cervical intraepithelial neoplasia to cancer (87,88).

SIRT1 is overexpressed in squamous intraepithelial neoplasia and squamous cell carcinoma (SCC), and its expression is increased in advanced illness (89,90). Extensive variation in SIRT1 expression in SCC from low to high suggests that SIRT1 may be an indicator of early lesions (90). Possible causes for the facilitating effect of SIRT1 on cervical cancer are shown in Fig. 2.

SIRT1 cooperates with the E1 and E2 proteins to ensure the quantity and high fidelity of HPV genome replication (91,92). When infected with HPV, cervical epithelial cells undergo DNA damage, leading to the recruitment of DDR-associated proteins at the E1-E2 binding position in the nucleus and the initiation of HR-dependent replication (5,87,88). SIRT1 is a component of the E1-E2 replication complex and a necessary molecule for initiating replication (91). DDR also stimulates the transcription of SIRT1, which deacetylates Nbs1 and then activates ATM; these processes are followed by the recruitment of the MRN complex and the enhancement of replication. Furthermore, deacetylated Werner helicase displays a greater ability to work with E1 and E2, guaranteeing the fidelity of viral replication (92).

As the downstream target of E7, SIRT1 protects HPV-infected cells from aging and apoptosis (93,94). E6 lowers the level of p53 and further inhibits cell apoptosis, while the expression of retinoblastoma protein (e.g., pRb) is affected by E7; these proteins contribute to cell cycle regulation. Both p53 and pRb are essential factors for the survival and proliferation of cervical cancer cells (87,88). E7 upregulates SIRT1, and the latter reduces apoptosis by deacetylating target proteins, such as FOXO3 and p53. By contrast, SIRT1 has been shown to be significantly downregulated after E7 knockout, and specific knockout of SIRT1 has been shown to induce tumor cell death (93,94).

SIRT1 may also help infected cells escape innate antiviral immune responses in mammals (95). The overexpression of SIRT1 in tumor cells suppresses the transcription of absent in melanoma 2 (AIM2), a key factor of antiviral immunity, by NF-κB. When SIRT1 is suppressed, the levels of AIM2 and associated inflammasome genes increase and lead to infected cell death through the immune response. Of note, SITR1-knockout cells release vesicles containing AIM2 and inflammasomes, which can be transferred to adjacent cells and cause cell death, demonstrating that the role of SIRT1 is not confined to a single cell.

The function of SIRT1 in epithelial OC displays significant heterogeneity. Differentially expressed genes and differential expression network analyses of the ArrayExpress database have revealed that SIRT1 is a hub gene and pathogenic gene for OC (96). Studies based on the Hendrix, GEPIA and Bonome databases show the same expression pattern of SIRT1 (97). However, Kanda et al (98) reported that SIRT1 is upregulated in epithelial OC, and peptide analysis of the plasma verified that the level of SIRT1 precursor is higher in patients with OC compared with that in healthy participants (99).

This heterogeneity may be attributed to tumor components. SIRT1 expression is significantly lower in cancer cells compared with that in stromal cells (100). In tumor cells, SIRT1 deacetylates and inhibits high mobility group box 1, a multifunctional chromosome-related protein that promotes the occurrence and development of OC, which results in the inhibition of metastasis and angiogenesis (101).

In addition, different pathological types of OC may also express varying levels of SIRT1. A study demonstrated that SIRT1 was significantly increased in serous OC (SOC), but its level was not associated with disease stage or grade, suggesting that SIRT1 may facilitate tumorigenesis rather than progression (98). However, another study demonstrated that SOC exhibited the same level of SIRT1 as normal tissues, and that this pattern was due to the high mutation frequency of p53 in SOC, which limits the effect of p53 deacetylation on tumor promotion (102). In addition, SIRT1 may be associated with malignant transformation in endometriosis, given its high expression in endometriosis, endometrioid OC and clear cell OC (102). Ovarian sex cord-stromal tumors may also be associated with SIRT1. In granulosa cell tumors, SIRT1 deacylates and inhibits FOXL2, causing cell cycle arrest and DDR, which promotes tumorigenesis, whereas nicotinamide, a SIRT suppressor, inhibits tumor cell proliferation (103). Furthermore, SIRT1 has been shown to be overexpressed in yolk sac cell tumors, which promotes tumor cell proliferation and differentiation and maintains the germ cell state, which may cause recurrence, whereas SIRT1 inhibitors have been shown to inhibit differentiation and invasiveness (104).

SIRT1 overexpression promotes EC progression, which is inhibited by specific SIRT1 inhibitors (61). As a synthetic inhibitor of histone deacetylase, MHY2256 can significantly decrease SIRT1 levels, thereby increasing the levels and activities of p53, p21 and other cell cycle proteins, which induce autophagy and apoptosis (105). The cancer-promoting effect of SIRT1 may also stem from metabolic changes. SIRT1 can increase the expression of SREBP-1, a nuclear lipogenic transcription factor, and promote the biosynthesis of fatty acids and triglycerides (62). However, miR-29c can bind to the 3′ untranslated region of SIRT1 and inhibit its expression, causing cell death (106); by contrast, OGFRP1, a long non-coding (lnc)RNA, has the capacity to reduce the production of miR-124-3p and further increase the level of SIRT1 in cancer cells, thereby promoting invasion and metastasis of ECs (107).

SIRT1 promotes the metastasis of cervical cancer. A decrease in miR-29 in malignant cells can reduce SIRT1 transcriptional suppression; the resulting increased levels of SIRT1 can promote EMT by increasing the expression and activity of MMP9 and enhancing cell migration and invasion (108,109). The lncRNA TUG1, which is upregulated in cancer cells and may act as a molecular sponge, can bind to miR-138-5p and amplify SIRT1, leading to cell proliferation via the Wnt/β-catenin pathway (110), whereas the overexpression of miR-138-5p can inhibit the transcription of SIRT1 and further arrest cells in the G0/G1 phase, thereby inducing apoptosis (111). However, it has been shown that, when pyruvate augments the binding between the histone gene promoter and SIRT1, heterochromatin density and genome imbalance suppress cell proliferation (112).

SIRT1 plays a dual role in OC metastasis. SIRT1 may promote tumor metastasis. The deacetylation of cortactin mediated by SIRT1 increases the expression of vimentin, a mesenchymal cell marker, and reduces that of E-cadherin, an epithelial cell marker, thereby promoting EMT and tumor cell migration (113,114). SIRT1 expression has been shown to be correlated with ER expression and to impact the effect of E2-ER complexes on EMT (115). Amyotrophic lateral sclerosis, hypoxia and 15d-PGJ2 induce apoptosis and reduce the migration of cells due to the reduced levels of SIRT1 (116–118). However, other research has indicated that higher levels of SIRT1 in OC suppress invasion and metastasis by increasing the epithelial phenotype (119). SIRT1 deacetylates and activates KLF4, which binds to the CLDN5 promoter to maintain the characteristics of epithelial cells (120). SIRT1 can also lower the EMT-inducing zinc finger E-box binding homeobox 1 by deacetylating hypoxia-inducible factor (HIF)1α (121). Rikkunshito, a Japanese herbal medicine, can decrease EMT by stimulating the expression of SIRT1 (98).

SIRT1 also has opposing roles in OC cell proliferation. SIRT1 reduces the levels of reactive oxygen species, promoting tumor cell proliferation and reducing apoptosis (122). miR-494 and artesunate accelerate cell death by downregulating SIRT1 (123,124). Additionally, 15-deoxy-Δ12,14-prostaglandin J2, a SIRT1 inhibitor, and its derivatives exert anticancer effects by blocking the cell cycle and triggering apoptosis and autophagy (125). Calorie restriction and metformin have been shown to increase the expression of SIRT1 in a mouse model and cause significant decreases in the tumor load in the peritoneum, liver, kidney, spleen and intestine; furthermore, growth factors, such as insulin-like growth factor-1, insulin, interleukin-1, monocyte chemoattractant protein-1 and vascular endothelial growth factor, in the plasma and ascitic fluid have been found to decrease the expression of SIRT1 (126,127).

Chemotherapy is a crucial method for the postoperative adjuvant therapy of EC, and SIRT1 is closely associated with platinum and paclitaxel resistance (61). Progesterone therapy is another common adjuvant treatment; it is used mainly as conservative treatment of patients who wish to preserve their fertility and as adjuvant treatment for patients with advanced disease (6). Researchers have found that SIRT1 is overexpressed in medroxyprogesterone acetate (MPA)-resistant tumors and that the level of progesterone receptor (PR) expression is decreased in those tumors (128). It has been found that, after SIRT1 is knocked out, the levels of PR are restored and tumor sensitivity to MPA is recovered, indicating that SIRT1 participates in progesterone resistance.

SIRT1 has been shown to increase multidrug resistance-associated protein (MRP) expression on the cell membrane, inducing chemotherapy resistance in paclitaxel-resistant tumor cells (129). In addition, SIRT1 is partly responsible for the transformation of chemoresistance, as it transfers deacetylated FOXO1 and ATP-binding cassette transporter p-glycoprotein and MRP to neighboring cells (130). Moreover, SIRT1 induces p53-dependent chemoresistance by p53 deacetylation; for example, the response to doxorubicin was shown to be restored in SIRT1-knockout cells (131).

Surgery after two to three cycles of neoadjuvant chemotherapy (NACT) is an alternative treatment for patients with locally advanced cervical cancer whose tumors are too large to operate or are at an inoperable stage (3). Currently, clinicians lack predictive indicators for evaluating NACT efficiency; therefore, disease progression during NACT may occur in some patients. SIRT1 may be used as such a predictor. Teramae et al (132) divided patients with locally advanced cervical cancer who had received NACT into two groups based on the level of SIRT1; the weighted score of the low group was ≤4, whereas that of the high group was ≥6. They found that the patients in the low-SIRT1 group significantly benefited from NACT, indicating that an elevated SIRT1 level is a negative indicator of NACT.

SIRT1 is highly expressed in SOC patients with chemoresistance. Its high expression in these patients is due to its stimulating effect on MRP (133,134). Cancer stem cells (CSCs) are a possible source of drug resistance and a cause of relapse in malignant tumors. Researchers have found that the percentage of EpCAM⁺CD45⁺ (CSC potential surface marker) cells in ascites is increased in patients with high-SIRT1 expression, and that hypoxia increases SIRT1 levels by affecting HIF1α and NF-κB, thereby maintaining the stemness characteristics of CSCs (135). Following exposure to cytotoxic drugs, intracellular SIRT1 levels have been shown to be upregulated and to further improve the transcription of stemness-associated genes, leading to acquired chemotherapy resistance (114).

Recently, PARPi, which cause defects in single-strand break repair, have become the most promising targeted drugs for OC and were recommended by the National Comprehensive Cancer Network guidelines as first-line maintenance therapy (136). PARPi kill tumor cells when HR defects exist, of which the BRCA mutation is the most common, also known as ‘synthetic lethality’ (Fig. 1) (137). Moreover, the damage caused by the continuous binding of PARP-PARPi to DNA is far more lethal than the reduction in PARP protein; therefore, the amount and function of PARP proteins are crucial for PARPi efficacy (138). SIRT1 may suppress the synthesis and function of PARP (139) and impair the therapeutic effect of PARPi (Fig. 1). As both PARP and SIRT1 require NAD⁺, they have a competitive relationship. SIRT1 can post-translationally modify PARP and affect its function; when overexpressed, SIRT1 can deacetylate PARP-1 and inhibit its function; SIRT1 can also specifically bind to the promoter of PARP and inhibit its transcription. In addition, SIRT1 promotes HR repair and impedes PARPi (Fig. 1). Notably, the loss of the BRCA1 function increases NAD⁺ synthesis and SIRT1 activity, suggesting that SIRT1 may partially compensate for the HR repair effect when BRCA1 is defective (140).

There is abundant evidence suggesting that SIRT1 overexpression is responsible for poor prognosis in EC (61), cervical cancer (95,108,110) and OC (97,98,102). After HPV infection of cervical epithelial cells, increased SIRT1 expression is closely associated with poor prognosis. Similarly, decreases in miR-29a and miR-138-5p levels lead to an increase in SIRT1 levels and poor prognosis. High levels of SIRT1 in OC, particularly SOC, are associated with reductions in the progression-free and overall survival of patients with OC.