Letrozole promotes the expression of integrin αvβ3 and HOXA10 in endometrium of endometriosis

ABSTRACT

Endometriosis is a common estrogen-dependent chronic inflammatory disease that leads to infertility in women of reproductive age. Perhaps infertility reflects the reduced expression of integrin αvβ3 and HOXA10 in endometriosis. Previous studies have shown that administration of letrozole, a non-steroidal aromatase inhibitor for cancer treatment, increased the clinical pregnancy rate in women with endometriosis, but the mechanisms remain to be determined. In this communication, a rat model of endometriosis was established. Animals were treated with letrozole at 2ug/kg of body weight, intragastric administration for 15 consecutive days. Letrozole increased the expression of αvβ3 and HOXA10 in the endometriosis model and endometrial receptivity.

Abbreviations: WOI: window of implantation; RGD: Arg-Gly-Asp; HOX: homeobox; E2: estradiol; SPF: specific pathogen-free.

KEYWORDS:

• Letrozole
• endometriosis
• endometrial receptivity
• integrin αvβ3
• HOXA10

Introduction

Endometriosis is an estrogen-dependent inflammatory disease defined by the presence of endometriotic gland and stroma outside of the uterine cavity (Vallve-Juanico et al. 2019). It affects approximately 10% of reproductive age women, with up to 50% experiencing infertility (May et al. 2010). Although the underlying mechanisms that lead to endometriosis-induced infertility remain unclear, several causes have been proposed, including pelvic microenvironment, ovarian dysfunction, and altered endometrial receptivity (de Ziegler et al. 2010).

Endometrial receptivity is a complex process that provides the embryo with the opportunity to attach, invade, and develop, relying on various mechanisms originating from the endometrium (Lessey and Young 2019). Embryo implantation occurs only during the period of endometrial receptivity, also known as the window of implantation (WOI), which occurs between 7 and 10 after ovulation in humans (Wilcox et al. 1999), and 5.5 days after mating in rats (Lecce et al. 2011). A variety of molecules, including αvβ3 and HOXA10, have been identified as endometrial receptivity markers (Achache and Revel 2006).

Integrin αvβ3 is a cell-surface receptor for extracellular matrix proteins that play an important role in embryo implantation (Lessey et al. 1992; Lessey, Castelbaum et al. 1994). Its expression shows periodic changes, and the peak time is consistent with the implantation window (Tabibzadeh 1992; Lessey et al. 1994a). The expression of integrin αvβ3 is decreased in women with infertility and endometriosis (Lessey et al. 1994a). Estrogen treatment can down-regulate the expression of integrin αvβ3 (Somkuti et al. 1997). HOXA10, = upstream gene integrin αvβ3 (Zhu et al. 2013), is also highly expressed during the implantation window and is a marker of endometrial receptivity (Gui et al. 1999). Estrogen can stimulate the expression of HOXA10, both in the myometrium and the endometrium (Cermik et al. 2001). In endometriosis, calpain7 can down-regulate the expression of HOXA10 via the PEST sequence, and the expression of integrin αvβ3 is also reduced (Yan et al. 2018). In addition, the expression of HOXA10 is decreased in endometriosis due to promoter methylation (Lu et al. 2013).

Estrogen dominance and progesterone resistance caused by inflammation can interfere with endometrial receptivity (Lessey and Kim 2017). Letrozole, an aromatase inhibitor, can specifically inactivate aromatase to inhibit estrogen formation. Studies have shown that the combination of letrozole and gonadotropins in IVF leads to higher pregnancy and live birth rates than only using high-dose gonadotropins (Lazer et al. 2014). Previous studies have shown that letrozole can treat endometriosis and significantly improve pregnancy rates in women with endometriosis during an IVF cycle (Miller et al. 2012; Ferrero et al. 2018), but the mechanisms remain to be elucidated.

We tested if letrozole may affect endometrial receptivity in endometriosis by regulating the expression of HOXA10 and integrin αvβ3 using a rat model of endometriosis. We demonstrated that letrozole increases the expression of αvβ3 and HOXA10 and thereby improves endometrial receptivity.

Results and discussion

Establishing an endometriosis model

The experimental and control groups are outlined in Table 1. As shown in Figure 1A, a trimmed segment of endometrial tissue was sutured to the inner side of the abdominal wall. We noted that an ectopic mass in the abdominal wall adhered to the surrounding connective tissues, and neovascularization developed in the surface of the mass (Figure 1B, C), which showed that a model of endometriosis was established. The cyst was 5–10 mm in diameter and contained a clear or purplish-red fluid. Figure 1D shows the vaginal plug, which is an important marker of the endometrium entering the window of implantation (WOI). Hematoxylin-eosin staining analysis, showed the growth of glands in the ectopic mass (Figure 1E, F), modeling endometriosis. The glandular epithelial cells were columnar or cubic, and inflammatory cells and secretions existed in some glandular lumen.

Figure 1. Establishment and results of endometriosis model. (A) The left uterine horn was ligated and excised. The endometrium was exposed along its antimesenteric axis, and 5*5 mm sections were cut. The trimmed endometrial piece was sutured to the inner side of the abdominal wall. (B) Ectopic mass in the abdominal cavity of rats in the endometriosis group. (C) Ectopic mass in the abdominal cavity of rats in the letrozole group. (D) Vaginal plug, a marker of mating between male and female rats. Mating can promote ovulation in female mice, allowing the endometrium to enter the window of implantation. (E) The gland of ectopic mass in the endometriosis group (Hematoxylin-eosin staining, 200×). (F) The gland of ectopic mass in the letrozole group (Hematoxylin-eosin staining, 200×).

Table 1. Experimental groups.

	Control	Endometriosis	Letrozole
Number of rats	12	15	16
Animal model	No endometrial implanting	Endometrial implanting	Endometrial implanting
Medication	Saline	Saline	Letrozole

Control group: the endometriosis model was not established, saline was administered intragastrically.

Endometriosis group: the endometriosis model was established, saline was administered intragastrically.

Letrozole group: the endometriosis model was established, letrozole was administered intragastrically.

Integrin αvβ3 and endometrial receptivity

Integrins are cell-surface receptors for extracellular matrix proteins involved in diverse cellular processes, including embryogenesis, wound healing, the immune response, and cancer (Albelda and Buck 1990; Hynes 1992). Integrin αvβ3, localized in the endometrial luminal and glandular epithelium, is one of the best-characterized endometrial biomarkers related to embryo implantation (Lessey and Castelbaum 2002). During blastocyst apposition in embryo implantation, trophoblast cells adhere to the endometrial epithelium. Integrin αvβ3 recognizes and binds to the Arg-Gly-Asp (RGD) amino acid sequence in osteopontin and may participate in cell-cell interactions. Both trophoblast cells and endometrial epithelial cells maintain the same integrin and osteopontin (Vanderpuye et al. 1991; Damsky et al. 1993), and the expression of integrin αvβ3 persists into pregnancy with expansion to the decidua (Lessey et al. 1992, 1994a). It is plausible that integrin αvβ3 plays an independent role in the interaction between trophoblastic cells and endometrium. The expression of integrin αvβ3 can be down-regulated by estrogen (Somkuti et al. 1997).

The expression of integrin αvβ3 mRNA in endometriosis was lower than its expression in the control group (p = 0.002, Figure 2B). After applying letrozole, the expression of integrin αvβ3 mRNA was elevated in the endometrium (p = 0.008, Figure 2B). The expression of αvβ3 protein was also reduced in endometriosis (p = 0.018, Figure 2C, D). After letrozole treatment, the expression of integrin αvβ3 protein in endometrium was elevated (p = 0.002, Figure 2C, D) as summarized in Table 2.

Figure 2. HOXA10 and Integrin αvβ3 expression in endometrium. (A) As compared with control group, the mRNA expression of HOXA10 was reduced in endometriosis (p < 0.0001). After the application of letrozole, the expression of HOXA10 was elevated in endometrium (p = 0.010). (B) The expression of integrin αvβ3 mRNA in endometriosis was lower as compared to its expression in control group ((p = 0.002). After the application of letrozole, the expression of integrin αvβ3 mRNA was elevated in endometrium (p = 0.008). (C and D) The expression of αvβ3 protein was also reduced in endometriosis ((p = 0.018). After letrozole treatment, the expression of integrin αvβ3 protein in endometrium was elevated ((p = 0.002). GAPDH (Glyceraldehyde-3-phosphate dehydrogenase) was used as a loading control. (**p < 0.01, *p < 0.05).

Table 2. The expression of HOXA10 and Integrin αvβ3, and the concentration of serum E2.

	Control	Endometriosis	Letrozole
HOXA10 mRNA (2^‑∆∆Ct)	1.00 ± 0.30	0.55 ± 0.16	0.75 ± 0.23
Integrin αvβ3 mRNA (2^‑∆∆Ct)	1.00 ± 0.43	0.54 ± 0.25	0.80 ± 0.23
Integrin αvβ3/GAPDH (Relative Expression)	0.66 ± 0.30	0.42 ± 0.20	0.66 ± 0.18
E2 concentration (pg/ml)	5.48 ± 1.15	6.63 ± 3.14	4.13 ± 2.64

HOXA10 mRNA: β-actin as internal reference factor. The expressions of HOXA10 mRNA are 1.00 ± 0.30 in control group, 0.55 ± 0.16 in endometriosis group, 0.75 ± 0.23 in letrozole group.

Integrin αvβ3 mRNA: β-actin as internal reference factor. The expressions of Integrin αvβ3 mRNA are 1.00 ± 0.43 in control group, 0.54 ± 0.25 in endometriosis group, 0.80 ± 0.23 in letrozole group.

Integrin αvβ3/GAPDH: GAPDH is used as internal reference factor to detect protein relative expression. They are 0.66 ± 0.30 in control group, 0.42 ± 0.20 in endometriosis group, 0.66 ± 0.18 in letrozole group.

E2 concentration: The concentration of E2 is 5.48 ± 1.15 in control group, 6.63 ± 3.14 in endometriosis group, 4.13 ± 2.64 in letrozole group.

HOXA10 and endometrial receptivity

HOXA10, a member of the homeobox (HOX) family, is expressed in the nucleus of endometrial glands and stroma of the endometrium (Gui et al. 1999). Its expression depends on the stage of the menstrual cycle, dramatically increasing at the time of implantation (Taylor et al. 1998; Gui et al. 1999). The expression of HOXA10 is altered in the endometrium of women with infertility, such as patients with endometriosis and polycystic ovary syndrome (Cermik et al. 2003). Previous studies have shown that HOXA10 regulates the expression of ITGB3, which encoding integrin αvβ3, affects endometrial receptivity. HOXA10 activates reporter gene expression via a 41-bp β3-Integrin subunit element containing two of the four HOXA10 binding sites (Daftary et al. 2002). In addition, HOXA10 can alter embryo viability by regulating tryptophan 2,3-dioxygenase through serotonin signaling (Doherty et al. 2011). The elevated expression of HOXA10 promotes the expression of integrin αvβ3 and enhances embryo vitality, promoting embryo implantation.

As shown in Figure 2, the expression of HOXA10 was reduced in endometriosis but elevated after letrozole application. As compared with control group, the mRNA expression of HOXA10 was reduced in endometriosis (p < 0.0001, Figure 2A). After the application of letrozole, the expression of HOXA10 was elevated in endometrium (p = 0.010, Figure 2A).

Endometriosis impairs endometrial receptivity

Infertility caused by endometriosis affects reproductive age women and is an estrogen-dependent disease. The upregulation of estrogen in endometriosis may be due to increased aromatase expression and impaired breakdown of estradiol due to decreased 17 beta-hydroxysteroid (Bulun et al. 2002). We also found Serum E2 (estradiol) concentration increased in the endometriosis model. As shown in Figure 3, the standard curve (correlation coefficient was 0.998) was drawn according to the standard substance of gradient concentration and its absorption value. The E2 concentration was 5.48 ± 1.15 pg/ml in the control group, 6.63 ± 3.14 pg/ml in the endometriosis group. E2 concentration in the endometriosis group was slightly higher than that in the control group (P > 0.05), but the difference was not statistically significant, which may be due to small sample size or experimental error.

Figure 3. Serum E2 concentration of rats was detected by ELISA. (A) is the standard curve, the correlation is 0.998. (B) Compared with the control group, estrogen concentration in the endometriosis group was slightly increased (P = 0.247), but the difference was not statistically significant. Estrogen concentration decreased after letrozole administration (P = 0.053), but the difference was not statistically significant.

Estrogen dominance and progesterone resistance can impair endometrial receptivity (Lessey and Kim 2017), this may reflect that estrogen can inhibit the expression of integrin αvβ3. This reduced endometrial receptivity may be due to a shift away from normal progesterone action and excessive estrogen activity (Aghajanova et al. 2010). Previous studies have shown that the expression of integrin αvβ3 was reduced in endometriosis (Lessey et al. 1994a), and HOXA10 was also downregulated (Wu et al. 2005). Our results also confirm these conclusions.

The effects of letrozole

The effect of letrozole on the fertility of women varies in current reports. Some investigators have shown that letrozole can boost fertility by increasing the number of follicles (Kilic-Okman et al. 2003). Others have shown that letrozole can increase pregnancy rate by elevating the expression of αvβ3 (Miller et al. 2012). Possible explanations are that the research methods and the dosage regimen in studies may be different. Unlike previous studies, we established a model of endometriosis to begin to investigate the effect of low-dose letrozole on endometrial receptivity and endometriosis. We found that letrozole could promote the expression of integrin αvβ3 and HOXA10 in endometrium of endometriosis and may improve endometrial receptivity.

Letrozole is an aromatase inhibitor, which can competitively bind to the heme of the cytochrome P450 subunit, thereby reducing estrogen levels. Aromatase is a rate-limiting enzyme in the biological synthesis of androgen to estrogen, and elevated aromatase levels are associated with IVF implantation failure (Brosens et al. 2004). As shown in Figure 3, we found that the serum E2 concentration was 4.13 ± 2.64 pg/ml in letrozole group. As compared with endometriosis group, E2 the concentration decreased after letrozole administration (P = 0.053), and the difference was not statistically significant, which may reflect the small sample size and/or associated error. As we showed, low-dose letrozole improved the expression of integrin αvβ3 perhaps by lowering the estrogen level. In addition, low dose letrozole increased the expression of HOXA10. In the past few years, many studies have demonstrated favorable results of letrozole in assisted reproductive technology, with no short – or long-term side effects (Palomba 2015; Tatsumi et al. 2017). In summary, letrozole can promote the expression of integrin αvβ3 and HOXA10 in the endometrium during endometriosis. It should be noted that the specific regulatory mechanisms will require a larger randomized set of in vitro and in vivo experiments.

Materials and methods

Animals

Forty-three Sprague-Dawley rats at 8 weeks old, weighing 240 ± 20 g, were used for the study. The study was conducted in Animal Laboratory Center of Shandong Provincial Hospital. Animals were raised in a specific pathogen-free (SPF) facility with a temperature-controlled (22°C-26°C) and humidity-controlled (40%-60%) environment, under a 12-h light/dark cycle, and ad libitum access to food and water. The experimental groups are shown in Table 1.

Modeling and tissue collection

Step 1: Establishment of the endometriosis model

All animals were anesthetized through intraperitoneal administration of 2% pentobarbital sodium (Sigma, USA) at a dose of 70 mg/kg. Endometriosis was surgically induced using the method described in the previous literature with a minor modification according to our previous experience (Yildirim et al. 2010). Using aseptic techniques, the left uterine horn was ligated and excised, placed in sterile saline. The endometrium was exposed along its antimesenteric axis, and 5*5 mm sections were cut. The trimmed endometrial piece was sutured to the inner side of the abdominal wall. Abdominal layers were closed anatomically. The control group also received laparotomy, only ligation of the left uterine horn, without implantation of abdominal wall ectopic endometrium.

Step 2: Drug intervention

After the surgery, the rats were raised for 28 days and during this time they did not receive medication. After 28 days, they were randomly divided into two groups, the endometriosis group and the letrozole group. Letrozole group was administered intragastrically letrozole (MCE, USA), 2ug/kg of body weight daily in 2 mL of normal saline for 15 days. Referring to the research on reproductive toxicity in letrozole’s drug label, we selected this dose which can inhibit estrogen dominance in endometriosis and maintain a certain estrogen concentration. The control group and endometriosis group were administered normal saline.

Step 3: Sample collection

Female rats were caged with male rats and vaginal plug were checked every morning and evening. All the male rats had undergone vasectomy. The mating behavior can stimulate ovulation in female rats and make the endometrium enter the window of implantation. The day of vaginal plug discovery (confirming insemination) was designated as Day 1. The rats were anesthetized on Day 5.5, when the endometrium was in the implantation window (Lecce et al. 2011). The ectopic mass and the right endometrium of the rats were removed. Blood was collected from the inferior vena cava and serum was collected.

Hematoxylin-eosin staining

The ectopic tissue was washed and fixed with 4% paraformaldehyde. After dehydration and paraffin-embedding, samples were cut into 4um thick sections and placed on glass slides. After 5 mins of dewaxing, hydration, hematoxylin staining, the sample were stained with eosin stained for 2 minutes. The stained sections were scanned by Digital Tissue Section Scanner (TissueGnostics, magnification, ×200).

Total RNA extraction and real-time PCR analysis

Total RNA was extracted from fresh endometrium using TRIzol reagents (Accurate, Hunan, China). Extracted RNA was reverse-transcribed by Evo M-MLV RT Premix (Accurate, Hunan, China) to obtain cDNA. RT-PCR of mRNA were performed on the Light Cycler 480 System (Roche, Basel, Switzerland) with SYBR® Green Pro Taq HS Premix (Accurate, Hunan, China). The qPCR amplification conditions were as follows: 40 cycles of 95°C for 5 sec, 60°C for 30 sec. Quantification of the target gene expression in the samples was calculated and adjusted to the quantitative expression of β‑actin. Relative gene expression was calculated using the 2‑∆∆Ct method. The primer for HOXA10 was 5ʹ-AGTGCTGGGCTGTGTTTAATCA-3ʹ and 5ʹ-CACGCTTCCATACTCTGCTCTT-3ʹ. The primer for integrin αvβ3 was 5ʹ-TAGAAGAGCCTGAGTGTCCTAAG-3ʹ and 5ʹ-TTCCAGATGAGCAGAGTAGCA-3ʹ. The primer for β-Actin was 5ʹ- GGAGATTACTGCCCTGGCTCCTA-3ʹ and 5ʹ- GACTCATCGTACTCCTGCTTGCTG-3ʹ.

Tissue protein extraction and Western blot analysis

Fresh endometrium samples were lysed in RIPA lysis buffer (Beyotime, Shanghai, China) with protease inhibitors (MCE, USA) to extract whole‑cell proteins. Protein levels were measured using the BCA Protein Concentration Determination kit (Solarbio, Beijing, China). Total proteins (20ug) were run on a 12.5% polyacrylamide gel and separated through electrophoresis, then transferred onto polyvinylidene fluoride membranes (Millipore Corporation, Billerica, USA). Membranes were blocked with 5% fat‑free milk followed by incubation with 0.5% integrin β3 (NB600-1342; Novus, Colorado, USA) and 0.1% GAPDH (Glyceraldehyde-3-phosphate dehydrogenase) (10,494; Proteintech, Wuhan, China) overnight at 4°C. Membranes were then incubated with 0.5‰ Anti-mouse IgG Antibody (7076; CST, USA) or 0.2‰ Goat Anti‑rabbit IgG Antibody (ZB‑2301; Zhongshan Golden Bridge Biotechnology, Beijing, China) in blocking buffer, respectively. Protein bands were detected by using an ECL detection kit (BD Biosciences, Franklin Lakes, NJ).

Enzyme linked immunosorbent assay (ELISA)

Rat E2 ELISA Kit (E-EL-0152 c, Elabscience, Shanghai, China) was used in the experiment. Standard and sample were first added, followed by rapid addition of biotinylated antibody. After incubation and washing, the enzyme conjugate was added. Incubated and washed again, added substrate reagent and stop solution respectively. The optical density of each hole was measured immediately with the enzyme plate instrument (450 nm).

Statistical analysis

Data were analyzed using SPSS 22.0. The statistical significance of the difference between groups were determined by independent T-test. The value of p < 0.05 was considered statistically significant.

Ethics approval

The experimental procedures were approved by the Committee for Animal Research. All animal studies strictly conformed to the animal experiment guidelines of the Committee for Humane Care.

Authors’ contributions

Designed the study, performed experiments, and wrote major parts of the manuscript, Figures and Tables: JZ. Helped with study design, supervised all the experiments and the progress of the study, and wrote parts of the manuscript: ML. Guided manuscript writing and wrote parts of the manuscript: CL, HZ. Performed some of the experiments: LW, RL. All authors read and approved the final manuscript.

Acknowledgments

The authors are grateful to all study participants, gynecologists, pathologists, animal laboratory researchers and central laboratory researchers at Shandong Provincial Hospital affiliated to Shandong University.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The study was supported by grants NO.2017YFC1001202 from the National Key Research and Development Program of China, NO.81671433 (MJL) from the National Natural Science Foundation of China, and NO.2014WS0079 from the Science and Technology Development Plan Project of the Medical and Health of Shandong Province, China.