INTRODUCTION

Cytoskeletal reorganization is an ongoing process when cells adhere, move or invade extracellular substrates. In order for cells to adhere, they employ focal adhesions (^{Fabry et al., 2011}). Focal adhesion (FA) is a specialized structure formed where bundles of actin filaments are anchored to transmembrane receptors of the integrin family through a complex of adaptor and signaling proteins enabling cells to adhere, spread and migrate (^{Gilmore & Burridge, 1996}). The binding of ligands to integrins on the extracellular side promotes recruitment of various intracellular proteins to the cytoplasmic tails of integrins that mechanically link them to the actin cytoskeleton (^{Hirata et al., 2014}). These signaling events culminate in reorganization of the actin cytoskeleton; a prerequisite for changes in cell shape and motility, gene expression and pregnancy (^{Lele et al., 2008}; Fabry et al.; ^{Kaneko et al., 2011}). Moreover, the implantation of blastocysts into the endometrial stroma cells is regulated by the Rho GTPases RhoA, Rac 1 and Cdc42 (^{Grewal et al., 2010}).

Other studies have provided evidence to suggest that the binding of vinculin to actin and talin can be regulated by phosphatidyl-inositol-4-5-bisphosphate (Gilmore & Burridge) and inhibited by acidic phospholipids. Clearly, the structural links between actin filaments and integrins are regulated in at least some cell types, and it would not be unreasonable to hypothesize that agents known to alter cell behavior may do so by affecting the expression or function of actin binding proteins such as vinculin, talin, paxillin and integrins that are recruited to focal adhesions (^{Mwakikunga et al., 2011}). Focal adhesions are dynamically assembled and disassembled by cells, including uterine luminal epithelial cells (^{Hosie et al., 2008}; ^{Kaneko et al., 2008}, 2011), and these mechanisms are likely governed by the modulation of cross-talk between their constituent proteins, which influences their gene expression and morphology.

Gonadotropin injections are medically dangerous and invasive, and their compliance is not easy. They are also associated with multiple pregnancies (^{Bruna-Catalán et al., 2011}). Similarly, polycystic ovarian syndrome patients are resistant to chlomiphine citrate (^{Casper & Mitwally, 2011}). Aromatase inhibitors (AIs) such as letrozole and anastrozole, which are used to treat breast cancer, have been suggested as alternatives to gonadotropins and chlomiphine citrate due to their easy compliance and oral administration (^{Lee & Ledger, 2011}). The use of letrozole as a first line therapy for women with anovulation is not fully conclusive and still remains deba(^{Kamath & George, 2011}). Anastrozole is clinically effective in ovulation induction, but it has not been well researched (^{Tredway & Schertz, 2011}). Investigating and understanding the effects of anastrozole on the endometrium at the time of implantation in vivo will enable manipulation of uterine receptivity to control fertility and to improve the outcome of assisted reproductive procedures (^{Paria et al., 2001}; ^{Hosie et al., 2003}; ^{Karaer et al., 2005}).

Since blastocyst adhesion onto the endometrium is at the center of the implantation process (^{Enders et al., 1986}; ^{Murphy, 2004}), this means that focal adhesion proteins are employed. The expression and distribution of the focal adhesion proteins in the rat uterine epithelial cells are associated with uterine receptivity (^{Kaneko et al., 2008}). In other words, the dynamics of focal adhesions are crucial at the time of implantation. Disassembly of the basal focal adhesions along the uterine epithelial cells makes cells less adherent to the underlying basement membrane, and this facilitates their removal, enabling embryonic trophoblast cells to invade the endometrial decidual cells beneath. Since previous studies demonstrated that focal adhesion proteins are more likely regulated in uterine epithelia to change cell behavior (Kaneko et al., 2009), and that super-ovulatory drugs can alter expression of key molecules in the uterine epithelial cells during implantation (^{Hosie et al., 2003}), anastrozole may act on focal adhesion proteins during implantation. To date, anastrozole regulation of focal adhesion proteins in the uterine epithelial cells during implantation has not been established.

The aim of this study was to determine the gene and protein expression levels and localization of the focal adhesion proteins vinculin and integrin β5 in the rat uterine epithelial cells during implantation in vivo.

MATERIAL AND METHOD

Animals. The study was approved by the University of the Witwatersrand Animal Ethics Committee with a clearance certificate number: 2012/11/03.

Thirty five female virgin inbred Wistar rats were used in the preliminary dose response study (^{Mwakikunga & Hosie, 2016}) to determine the optimal dose of anastrozole required to superovulate rats in the main study. On day 6 pregnant rats were used in the preliminary study (Mwakikunga & Hosie). A chlomiphene citrate (CC) treatment group was also used as a control to compare with the anastrozole treatment groups. Rats were housed in plastic cages at 21 ºC under a controlled 12 hour light-dark cycle. They were provided with water and food ad libitum (Dukes et al., 1996). Vaginal smears were obtained from all female rats to confirm that they had regular cycles for two cycles before drug administration (^{Singletary et al., 2005}; ^{Jaramillo et al., 2012}). Thirty five mature female rats, 12 to 14 weeks old and weighing 200-250 g, were randomly divided into 7 groups of 5 rats each. Vaginal smear was done in the late afternoon (^{Kaneko et al., 2008}) and rats in pro-oestrus were treated with the drugs or placebo and then caged overnight with males of proven fertility. The morning of finding the vaginal plug or presence of spermatozoa in the smear was designated as day 1 of pregnancy (Kaneko et al., 2008).

Rationale for the doses. A daily dose of 1mg/kg anastrozole (Sigma-Aldrich Co., St. Louis, MO, USA) for 5 days is the standard recommended dose to achieve ovulation in humans (^{Franik et al., 2014}). A 12-14 week mature female Wistar rat weighs between 200-250 g and has a 4-day estrous cycle, so a single dose of anastrozole administered at pro-estrus was deemed appropriate. ^{Kilic-Okman et al. (2003}) employed a similar regime using letrozole. Earlier studies have suggested that the standard 1 mg/kg anastrozole of body weight dose is too low for optimal follicle recruitment and ovulation, so higher doses are recommended (^{Al-Omari et al., 2004}; ^{Casper & Mitwally, 2012}). A single 25 mg/kg anastrozole dose has been successful in inducing ovulation in mice and is associated with favorable embryo development (Karaer et al.). However, in our preliminary work, a drug concentration study with anastrozole was carried out to determine the optimal dose to be used to superovulate and achieve pregnancy in the Wistar rats (Mwakikunga & Hosie). This was found to be dose-specific, and the 15 mg/kg anastrozole dose superovulated and achieved pregnancy the most; therefore, this dose was used in the subsequent experiments in this study. Dose rates in the preliminary work were started at 1 mg/kg. The chlomiphene citrate dose of 1.25 mg/kg (Sigma-Aldrich Co., St. Louis, MO, USA) was adopted from earlier studies (Hosie et al., 2003) and used as a comparison.

Immunofluorescence and confocal microscopy

Tissue processing and sectioning. All histological glass slides that were used in this study were gelatin-coated (^{Rajamohamedsait & Sigurdsson, 2012}) in order to stick the tissue sections to the slides during the staining and washing process. Using standard procedures, samples for immunofluorescence and confocal microscopy were fixed in 10 % buffered formalin for 48 hours (Karaer et al.), then placed in an automatic tissue processor (Shandon Citadel 1000, Labotek, South Africa) in which samples were incubated in a series of 70 %, 95 %, 95 %, 95 %, 100 %, 100 %, 100 % ethanol, then in chloroform and finally in paraffin wax. Samples were then embedded in paraffin wax. Histological sections (5 mm) were cut using a Leica 2035 Biocut microtome (Leica, Nussloch, Germany) fitted with a disposable blade and then dewaxed overnight by placing them in an oven at 60 ºC, then immersed in histoclear for 5 minutes and then repeated in fresh histoclear for 5 minutes, rehydrated through a graded series of ethanol (100 %, 100 %, 95 %, 80 %, 70 % and 60 %) for 30 seconds each, then washed in running water for 5 minutes (^{Mohan et al., 2008}; Rajamohamedsait & Sigurdsson, 2012).

Immunofluorescence. The immunolocalization and double labeling protocol was adopted and modified from previous studies (Mohan et al.; ^{Kaneko et al., 2011}; Mwakikunga, et al.). De-waxed sections of uterus from each group were washed 3 times, 2 min each with PBS. This was followed by a 10 min incubation in 30 % H2O2 (Sigma-Aldrich Co., St. Louis, MO, USA) in methanol to block endogenous peroxidase activity (Mwakikunga et al.), and then washed 3 times, 2 min each with PBS. Sections were then permeabilized for 30 minutes in 0.1 % Triton-X 100 (Sigma-Aldrich Co., St. Louis, MO, USA) in PBS, and incubated for 30 minutes in blocking solution 5 % (v/v) normal goat serum (Sigma-Aldrich Co., St. Louis, MO, USA) in PBS. All primary and secondary antibodies were diluted in blocking solution (1:100). Sections were incubated with mouse monoclonal anti-Vinculin, rabbit polyclonal anti-Integrin (Abcam, Cambridge, MA, USA) for 24 hours at 4ºC (double staining: vinculin with integrin β5)

(Mohan et al.; ^{Kaneko et al., 2011}). After washing with PBS 3 times 5 min each, sections were incubated with fluorescein isothiocyanate (FITC) conjugated to AffiniPure Goat antimouse IgG secondary antibody (Abcam) and rodamine conjugated Goat anti-rabbit IgG secondary (Abcam) at a dilution of 1:100 [10] for 30 min in the dark followed by 3 PBS washes of 5 min each. Sections were also counterstained with DAPI to label nuclei. Expression and localization of vinculin and integrin b5 were examined under the Zeiss LSM 780 confocal microscope (Carl Zeiss, Jena, Germany) and images were acquired using the Zeiss LSM software (Carl Zeiss, Jena, Germany). The focal adhesion protein expression levels and localization in the confocal images were scored using a scoring method modified from previous studies (^{Englund et al., 2001}) as shown in Table II. JMP10 software (SAS Institute, Cary, NC, USA) was used for the statistical comparisons of means of focal adhesion protein expression of vinculin and integrin b5 among treatment groups using a one way ANOVA followed by a Tukey- Kramer post hoc analysis. Differences were considered statistically significant when p ≤ 0.05.

Table II Description of the scoring system for the focal adhesion protein expression and localization in the confocal images among treatment groups adopted from previous studies (Englund et al., 2001; Mwakikunga et al., 2011). 

Real time quantitative polymerase chain reaction (qPCR). In this study, the qPCR was performed for the quantitation of gene expression of the focal adhesion proteins vinculin and integrin b5 in the uterine epithelial cells from day 1 and day 6 pregnant rats of all treatment regimes. The house keeping genes b-actin, 18Sr RNA and Lactate dehydrogenase A (Ldha) were used as reference genes (^{Al-Bader & Al-Sarraf, 2005}; ^{Hong et al., 2006}; ^{Li et al., 2014}).

The RNA from the uterine epithelial cells was prepared with the GeneJET RNA purification kit (Thermo Scientific Inc., 2011), and genomic DNA was removed according to the manufacturer’s instructions (Thermo Scientific Inc., 2011). For qPCR, amplification was performed in a 7500 real-time PCR cycler (Applied Biosystems, Foster City, CA) and revealed with a QuantiFast SYBR Green PCR kit (ThermoFisher Scientific Inc., 2011). The amplification cycling conditions were as follows: 95 ºC for 10min, (95 ºC for 15 s, 60 ºC for 1min in 40 cycles), 95 ºC for 15 s, 60 ºC for 1min, 95 ºC for 30 s, 60 ºC for 15 s. Table III shows the primer sequences for the real-time qPCR.

Table III Description of the real time PCR primer sequence characteristics for the genes of interest and reference genes used in the study. 

SYBR green qPCR data analysis. The 2^-∆∆CT method was used to determine the gene expression fold change relative to the control (calibrator sample) following drug treatment regime (^{Schmittgen & Livak, 2008}). The data (threshold cycle or ct values) were first normalized using the three reference genes (β-actin, 18S rRNA and lactate dehydrogenase A); then the fold change was calculated (Schmittgen & Livak, 2008). JMP10 software was used to conduct statistical analyses. A one way ANOVA, followed by a Tukey-Kramer post hoc analysis, was performed to compare the means of normalized relative quantities (NRQ) between treatment groups. Differences were considered statistically significant when p ≤ 0.05.

RESULTS

Vinculin and integrin β5 localization. Vinculin and integrin β5 are co-localized at the base of the uterine epithelium at day 1 of pregnancy whereas at day 6, they disassemble from the basal focal adhesions and colocalize and significantly increase their expression apically (p ≤ 0.0001) as noted in Figs. 1, 2, 3, 4 and Table IV. Additionally, vinculin is expressed in the perinuclear region at day 1 and relocates apically at day 6 of pregnancy as seen in Figs. 1, 2, 3, 4. Moreover, there is a significant difference (p ≤0.0001) in the expression levels of both vinculin and integrin β5 between untreated (control) and chlomiphene citrate (CC) treated rats, anastrozole and CC treated rats at both day 1 and day 6 as seen in Figs. 1, 2, 3, 4. Although there is no significant difference in the expression levels between untreated and anastrozole treated rats at day 1 and day 6 (p £ 0.05), the means for vinculin and integrin b5 expression are higher in the anastrozole treated groups. The non-immune controls (not shown) were used to validate and determine non-specific binding of the secondary antibody on other targets within the epithelium; there is no non-specific staining.

Fig. 1 Vinculin and intergrin β5 expression in the surface luminal uterine epithelial cells of day 1 and day 6 pregnant rats. There is a significant difference in the expression of vinculin and integrin β5 between anastrozole day 1 and day 6 (p≤0.0001), untreated (control) day 1 and day 6 (p≤0.0001), untreated day 6 and CC day 6 (p≤0.0001). No significant difference in vinculin and integrin β5 expression is noted between chlomiphene citrate day 1 and chlomiphene citrate day 6 (p≤0.05), untreated day 6 and anastrozole day 6 (p ≤0.05). Generally, chlomiphene citrate (CC) seems to decrease their expression while anastrozole seems to enhance their expression as pregnancy progresses. Note: numerical scale on the left represents the scoring system average scores as described previously in Table II. 

Fig. 2 Vinculin and intergrin β5 uterine epithelial localization. Vinculin and integrin β5 are colocalized at the base of the uterine epithelium at day 1 of pregnancy whereas at day 6, they disassemble from the basal focal adhesions and co-localize and significantly increase their expression apically (p≤0.0001) as also noted in Figs. 3 and 4. Note: vinculin is also expressed in the perinuclear region at day 1 and relocates apically at day 6 of pregnancy in the untreated and anastrozole treated rats as seen in Figs. 3 and 4. Note: numerical scale on the left represents the scoring system average scores as described previously in Table II (1, basal; 2, apical; 3, lateral; 4, basolateral; 5, basal and apical). 

Fig. 3 Micrograph showing vinculin and integrin β5 expression and co-localization in luminal uterine epithelial cells from day 1 of pregnancy in rats. (A), (B) and (C) show nuclei stained with DAPI (blue). (D) and (E) A distinct band of vinculin (green, FITC) expression at the base of the uterine epithelium (yellow arrow) is noted in both untreated and anastrozole treated rats. (F) The basal expression of vinculin in chlomiphene citrate (CC) treated rat is not as high as in (D) and (E). (G) and (H) A distinct band of integrin β5 (red, rodamine) expression at the base of the uterine epithelium (yellow arrow) is noted in both untreated and anastrozole treated rats. (I) The basal expression of integrin β5 in CC treated rat is not as high as in (G) and (H). (J), (K) and (L) Vinculin and integrin β5 co-localize basally on day 1 of pregnancy. All images are representative of staining from the 5 rats in each of the treatment regimes. 

Fig. 4 Micrograph showing vinculin and integrin β5 expression and co-localization in luminal uterine epithelial cells from implantation sites of day 6 pregnant rats. (A), (B) and (C) show nuclei stained with DAPI (blue); orange arrow shows the embryo. (D) and (E) Vinculin (green, fluorescein isothiocyanate (FITC) disassemble from the base of the epithelium (yellow arrows) and become highly expressed apically (white arrows) in both untreated and anastrozole treated rats. (F) The apical expression of vinculin in chlomiphene citrate (CC) treated rat is down-regulated. (G) and (H) Integrin β5 (red, rodamine) disassemble from the base of the epithelium (yellow arrows) and become highly expressed apically (white arrows) in both untreated and anastrozole treated rats. (I) The apical expression of integrin β5 in CC treated rat is down-regulated. (J), (K) and (L) Vinculin and integrin β5 co-localize apically on day 6 of pregnancy. All images are representative of staining from the 5 rats in each of the treatment regimes. 

Table IV A summary of vinculin and integrin β5 expression and localization. 

Relative quantification (RQ) analysis of vinculin and integrin β5 gene expression in the uterine epithelial cells from day 1 and day 6 pregnant rats. At day 1 of pregnancy, vinculin gene expression in uterine luminal epithelial cells is fairly similar (1.2-fold increase) in the anastrozole treated rats relative to the calibrator sample (day 1 untreated rats). This is also true for integrin β5 gene expression (1.01-fold increase) (p<0.05). Vinculin and integrin β5 gene expression in uterine luminal epithelial cells, however, is down-regulated (0.77-fold decrease) and (0.79-fold decrease) respectively with chlomiphene citrate (CC) treatment at day 1 of pregnancy relative to the calibrator sample as shown in Tables V and VI and Figure 5. Interestingly, like the immunofluorescence experiments, at day 6 of pregnancy vinculin and integrin β5 gene expression are significantly up-regulated (more than 1.5-fold increase) (p ≤ 0.0001) in uterine luminal epithelial cells in the implantation and non-implantation sites in the anastrozole treated group relative to the calibrator sample (Tables V and VI and Fig. 5). A significant increase in the integrin β5 gene expression is also noted in uterine luminal epithelial cells of day 6 implantation sites in the untreated group (more than 1.5-fold increase) (p ≤ 0.0001) while it has remained similar (1.04-fold increase) (p≤0.05) in the non-implantation sites relative to the calibrator sample (Tables V and VI and Fig. 5). However, vinculin gene expression in uterine luminal epithelial cells is down-regulated with CC treatment at day 6 in implantation sites (0.96-fold decrease) and non-implantation sites (0.93-fold decrease) relative to the calibrator sample. This is also true for integrin β5 gene expression (0.86-fold decrease and 0.42-fold decrease) respectively. In general, there is an increase in vinculin and integrin β5 gene expression at day 6 of pregnancy in uterine luminal epithelial cells in untreated and anastrozole treated groups, as reflected in the immunofluorescence experiments.

Fig. 5 Graphical representation of vinculin and intergrin β5 gene expression in the surface luminal uterine epithelial cells of day 1 and day 6 pregnant rats. There is a significant difference in the gene expression of vinculin and integrin β5 between anastrozole day 1 and day 6 (p≤0.0001), untreated (control) day 1 and day 6 (p≤0.0001), untreated day 6 and CC day 6 (p≤0.0001). No significant difference is noted between CC day 1 and CC day 6 (p≤0.05), untreated day 6 and anastrozole day 6 (p≤0.05). Note: Generally, chlomiphene citrate (CC) seems to decrease their expression while anastrozole seems to enhance their expression as pregnancy progresses.’ 

DISCUSSION

In this study, the focal adhesion proteins vinculin and integrin β5 are co-localized at the base of the uterine epithelium at day 1 of pregnancy whereas at day 6, they disassemble from the basal focal adhesions and co-localize and significantly increase their expression apically (p≤0.0001). Additionally, vinculin is also expressed in the perinuclear region at day 1 and relocates apically at day 6 of pregnancy. Moreover, there is a significant difference in the protein expression levels of both vinculin and integrin β5 between untreated (control) and chlomiphene citrate (CC) treated rats(p≤0.0001), anastrozole and chlomiphene citrate (CC) treated rats at day 6 (p≤0.0001) suggesting the interpretation that CC seems to decrease their expression. Although there is no significant difference in the expression levels between untreated and anastrozole treated rats at day 1 and day 6 (p≤0.05), the means for vinculin and integrin β5 protein expression are higher in the anastrozole treated groups suggesting perhaps, that anastrozole seems to enhance their expression in order to assist in the implantation process of the embryo.

Moreover, vinculin and integrin β5 gene expression in uterine luminal epithelial cells at day 1 of pregnancy agrees with the protein expression. Vinculin gene expression is fairly similar (1.2-fold increase) (p≤0.05) in the anastrozole treated rats relative to the calibrator sample (day 1 untreated rats). This is also true for integrin β5 gene expression (1.01-fold increase) (p≤0.05) suggesting the interpretation that anastrozole seems not to retard vinculin and integrin β5 gene expression in the process of endometrial readiness for implantation. Vinculin and integrin β5 gene expression in uterine luminal epithelial cells, however, is down-regulated (0.77-fold decrease) and (0.79-fold decrease) respectively with chlomiphene citrate (CC) treatment at day 1 of pregnancy relative to the calibrator sample, which may suggest that CC perhaps decreases their expression. Interestingly, like the immunofluorescence experiments, at day 6 of pregnancy, vinculin and integrin β5 gene expression are significantly up-regulated (more than 1.5-fold increase) (p≤0.0001) in uterine luminal epithelial cells in the implantation and nonimplantation sites in the anastrozole treated group relative to the calibrator sample. A significant increase in the integrin β5 gene expression is also noted in uterine luminal epithelial cells of day 6 implantation sites in the untreated group (more than 1.5-fold increase) (p≤0.0001) while it has remained similar (1.04-fold increase) (p≤0.05) in the nonimplantation sites relative to the calibrator sample suggesting the interpretation that anastrozole appears to enhance their expression to probably assist in blastocyst implantation. However, vinculin gene expression in uterine luminal epithelial cells is down-regulated with chlomiphene citrate (CC) treatment at day 6 in implantation sites (0.96- fold decrease) relative to the calibrator sample. This is also true for integrin β5 gene expression (0.86-fold decrease),

which may suggest that chlomiphene citrate (CC) may lag endometrial receptivity. In general, there is an increase in vinculin and integrin β5 gene expression at day 6 of pregnancy in uterine luminal epithelial cells in untreated and anastrozole treated groups, as reflected in the immunofluorescence experiments, in order to perhaps assist in implantation.

Vinculin is one of the core and best characterized focal adhesion (FA) proteins (^{Humphries et al., 2007}; ^{Carisey et al., 2013}). Tyrosine phosphorylation of vinculin modifies focal adhesion dynamics and cell tractions (^{Küpper et al., 2010}). Vinculin regulates the recruitment and release of other core focal adhesion proteins in a force dependent manner and interacts with the talin-integrin complex and thus drives the recruitment and release of core FA components (Humphries et al.; Carisey et al.). Earlier studies demonstrated that lack of cytoskeletal connecting proteins like vinculin, talin and focal adhesion kinase (FAK) reduces the overall cellular binding strength (Carisey et al.). As mentioned earlier, focal adhesions disassemble from the base of uterine luminal epithelial cells at the time of implantation to facilitate their removal so that the implanting blastocyst can invade into the underlying endometrial decidual cells during early pregnancy in the rat (^{Kaneko et al., 2009}). Kaneko et al. (2009) showed a major core FA protein talin distributional change between different hormone regimes. Talin is highly concentrated along the basal cell surface of uterine luminal epithelial cells in response to estrogen treatment which is observed at day 1 of pregnancy. However, this prominent staining of talin is absent in response to progesterone alone or progesterone in combination with estrogen, which is also observed at the time of implantation (Kaneko et al., 2009). This supports one of the important findings of the present study as noted in the immunofluorescence experiments in which FA proteins vinculin and integrin β5 disassemble from the base of the uterine epithelium at day 1 of pregnancy and co-localize and increase their expression apically at day 6 of pregnancy to facilitate the implantation process.

Of particular significance, the disassembly of vinculin and integrin β5 from the basal focal adhesions and their apical expression increase at day 6 of pregnancy in the anastrozole treated rats means that anastrozole may favor implantation. Moreover, integrin aVb3 is required for endometrial receptivity (^{Zhang et al., 2011}). Other studies demonstrated that human preimplantation embryos constitutively express integrin b3 and integrin β5 subunit proteins (^{Bloor et al., 2002}). Put together, this means that integrins play a role in the implantation process and could be regulated by agents that are known to alter cell behavior such as anastrozole and chlomiphene citrate (CC).

On the other hand, the decrease in the apical vinculin and integrin β5 expression with chlomiphene citrate (CC) treatment at day 6 in the current study may suggest the interpretation that CC might negatively affect the implantation process in this regard. This agrees with earlier studies that also observed a decrease in uterine integrin b3 expression in chlomiphene citrate (CC) treated patients during the window of implantation, suggesting that chlomiphene citrate (CC) might affect the expression of uterine receptivity markers (^{Palomino et al., 2005}). On the contrary, other studies noted that chlomiphene citrate (CC) does not affect the secretion of integrin α3, integrin aV and integrin b1 during the implantation window in patients with unexplained infertility (^{Lacin et al., 2001}). All in all, chlomiphene citrate (CC) appears to negatively affect the implantation process by down-regulating the expression of key implantation markers such as vinculin and integrin β5 and thus being associated with low pregnancy rates.

Additionally, other studies done on the effects of letrozole and chlomiphene citrate (CC) on the expression of integrin α_Vβ₃ during implantation in rats demonstrated that the expression of integrin α_Vβ₃ in the CC group is significantly lower than in the letrozole and saline groups (^{Bao et al., 2009}) suggesting that CC suppresses uterine receptivity more than letrozole (Bao et al.). Similar observations are made by previous studies in which the expression of the molecular markers of endometrial receptivity, integrin α_Vβ₃ and glycodelin, are decreased in endometrial biopsy specimens from women with PCOS when ovulation is induced with clomiphene citrate (^{Gonzalez et al., 2001}; ^{Jakubowicz et al., 2001}). Interestingly, these observations agree with the results of the current study in which integrin β5 expression in the uterine luminal epithelial cells of CC treated rats is significantly lower than anastrozole and untreated groups during implantation (p≤0.0001). However, in the present study, integrin β5 disassembles from the basal focal adhesions and later get localized at the apical plasma membrane of the rat uterine luminal epithelial cells at the time of implantation as also noted by (^{Kaneko et al., 2011}) in their studies with integrin b3. Moreover, integrin β5, b3 and aV are apically distributed in uterine epithelial cells during implantation (^{Aplin & Kimber, 2004}) suggesting their crucial role in the implantation process. Similarly, ^{Creus et al. (2003}) have shown that the human endometrial maturity is characterized by integrin aVb3 expression and pinopod formation. Therefore, this clearly shows that integrins (including integrins β5) are pivotal with regard to blastocyst-uterus interaction during early pregnancy. Viewed in its totality, anastrozole seems to enhance the focal adhesion protein vinculin and integrin β5 expression at day 6 suggesting the interpretation that anastrozole is implantation friendly as also evidenced by its high implantation rates (Mwakikunga & Hosie).

CONCLUSIONS

The immunofluorescence and gene expression results for the uterine receptivity markers FA proteins vinculin and integrins β5 appear to demonstrate that anastrozole is implantation friendly. It is also important to note that anastrozole is a competitive inhibitor of aromatase in the conversion of androgens to estrogens (E2); therefore, the 15 mg/kg anastrozole dose, which is effective in causing ovulation, may allow just enough of the intermediates and E2 to be present with regard to the hormone profiles for ovulation and successful implantation of the embryo. Essentially, it may appear that there is blocking of E2 and increasing of androgens in anastrozole treated rats compared to controls. Progesterone (P) is probably also high as E2 is low. In chlomiphene citrate (CC) treated rats, estrogen receptors might be blocked, which increases E2. The effect of anastrozole in the level of E2 and P may change the E2/P ratio at day 6 of pregnancy, suggesting its effect on endometrial maturation and therefore improving receptivity. If this is true, the advantage of anastrozole therapy over chlomiphene citrate (CC) may be the lower E2/P ratio and increased levels of FSH and LH. Therefore, measuring parameters such as the intermediates, estrogens and androgens in future studies may also be of use in determining hormonal profiles for implantation. Anastrozole may be a suitable replacement or alternative drug used for hyperovulation or just to promote fertility in patients, particularly those suffering from polycystic ovary syndrome and those that are unresponsive to chlomiphene citrate (CC).