Cyclic changes and hormonal regulation of annexin IV mRNA and protein in human endometrium

Abstract

Annexin IV (ANXA4) belongs to a ubiquitous family of Ca²⁺-dependent phospholipid-binding proteins. ANXA4 has been shown to be involved in a range of physiological functions including ion channel regulation, exocytosis and Ca²⁺-dependent signal transduction. The aims of this study were to fully characterize ANXA4 mRNA and protein in human endometrium during the menstrual cycle and to investigate the hormonal regulation of ANXA4. ANXA4 mRNA expression was quantified by real-time PCR in fresh endometrial tissue from cycling women, and protein expression was analysed by immunohistochemistry and western blotting. Hormonal regulation of ANXA4 transcription and translation was investigated using an endometrial explant system. ANXA4 mRNA was significantly up-regulated during mid-secretory (MS) and late-secretory (LS) phases compared with proliferative phases during the menstrual cycle. ANXA4 protein was localized to glandular and luminal epithelium and was present in high levels throughout the menstrual cycle except during early-secretory (ES) phase, when it was significantly reduced. Our data also show that, in proliferative explants, progesterone significantly increased the ANXA4 mRNA and protein after 48h in culture. Estrogen did not have any significant effects. This is the first study to show that ANXA4 transcription and translation are regulated by progesterone and suggests that ANXA4 may be important in regulating ion and water transport across the endometrial epithelium.

Introduction

Human endometrium undergoes a complex series of cyclic changes each month under the overall control of estrogen and progesterone. Although the use of chronological and histological dating has improved the understanding of anatomical changes within the endometrium during the menstrual cycle, most of the molecular pathways that are responsible for these changes remain unknown. The fundamental knowledge of the molecular changes governing the overall regulation of the endometrium is essential for the understanding of the biological processes occurring in this tissue, such as proliferation, implantation and menstruation. To identify novel genes and new molecular pathways that are involved in these complex processes, we profiled the human endometrium during the menstrual cycle at the transcriptional level (Ponnampalam et al., 2004). This study identified many previously unreported genes that showed evidence for a role in endometrial biology.

Many microarray studies of human endometrium have shown that annexin IV (ANXA4) mRNA is significantly up-regulated during the secretory phase of the menstrual cycle compared with the proliferative phase (Kao et al., 2002; Riesewijk et al., 2003; Ponnampalam et al., 2004; Mirkin et al., 2005). ANXA4 belongs to a ubiquitous family of structurally related proteins capable of binding anionic phospholipid membranes in a calcium-dependent manner. Annexins have been shown to be involved in a wide range of physiological activities, such as inhibition of coagulation, inhibition of phospholipase A2 activity, Ca²⁺ ion channel activity, interactions with membranous and cytoskeletal elements, vesicular transport and exocytosis and endocytosis. One of the least studied members of the annexin family is ANXA4, which shares 50–60% sequence homology with the relatively well-studied annexin V. ANXA4 is found at high levels in kidney, trachea, lung, intestine and stomach. It is thought to be a marker for polarized epithelial cells. Although the biological roles for ANXA4 remain largely unclear, it has been implicated in the regulation of calcium-activated epithelial chloride channels (Kaetzel et al., 1989; Chan et al., 1994) and shown to have anti-inflammatory properties (Katoh, 2000; Gotoh et al., 2005). ANXA4 has also been shown to play a role in kidney organogenesis in Xenopus laevis, where ablation of the gene product results in abnormal development of pronephric tubules (Seville et al., 2002). Hill et al. (2003) showed that ANXA4 can regulate passive membrane permeability to water and protons and can alter physical properties of the membranes by associating with them.

Structurally, all members of the annexin family share a common conserved core of 4–8 annexin repeats consisting of 70 amino acids. Each annexin has a unique amino-terminal believed to be responsible for specific functions of each family member. ANXA4 has a short amino-terminal of 12 amino acids, which is susceptible to phosphorylation by protein kinase C. ANXA4 is a 36-kDa protein that can aggregate on the inner leaflet of cellular membranes. Soluble monomers of ANXA4 trimerize in the cytoplasm bind to the membrane and then assemble into higher order structures at the membrane interface creating a crystallization cascade in 2D across a large cross-sectional area of membrane (Zanotti et al., 1998). In vitro cross-linking studies demonstrate that trimer, hexamer and higher aggregates of annexin form in the presence of Ca²⁺ and phospholipid-containing vesicles (Concha et al., 1992). Phosphorylation by protein kinase C causes the release of the N-terminal of ANXA4 and inhibits ANXA4’s ability to aggregate on the membrane (Kaetzel et al., 2001). ANXA4 has also been found in the cytoplasm (Zimmermann et al., 2004) and nucleus (Raynal et al., 1996), and it can be secreted (Masuda et al., 2004). Intranuclear ANXA4 has been shown to translocate to the cytoplasm because of an increase in intracellular calcium (Mohiti et al., 1995; Barwise and Walker, 1996; Raynal et al., 1996), and during Fas-induced cell death (Gerner et al., 2000), however, the functional role of nuclear ANXA4 is currently not known.

The aims of this study were to examine the mRNA and protein profile of ANXA4 in human endometrium during the menstrual cycle and to investigate the changes in ANXA4 mRNA and protein expression in response to ovarian steroids in an in vitro endometrial explant system. On the basis of previous literature (Kao et al., 2002; Riesewijk et al., 2003; Ponnampalam et al., 2004; Mirkin et al., 2005), we hypothesized that ANXA4 expression is regulated by progesterone in human endometrium.

Materials and methods

Tissue collection

Ethical approval for the study was obtained from Southern Health Human Research and Ethics Committee B. Endometrial biopsies were obtained from normal cycling women after informed consent. All the endometrial samples used in the study were classified as normal by routine histopathology. Subjects ranged in age between 18 and 47 years and had not used hormonal contraception in three months before tissue collection. A small portion of the tissue was sent for histopathology to evaluate the cycle stage. An experienced pathologist later reconfirmed the cycle stage of each tissue sample. For the evaluation of mRNA (n = 64) and protein profile (n = 50), the menstrual cycle stages of 60 samples were divided into seven groups by histopathology, based on well-established criteria (Noyes et al., 1950): early proliferative (EP), mid proliferative (MP), late proliferative (LP), early secretory (ES), mid secretory (MS), late secretory (LS) and menstrual (M). Endometrial tissues were either snap frozen on dry ice immediately after collection and stored at −80°C until RNA and protein extraction or fixed with 10% formalin for immunohistochemistry.

The methods for the endometrial explant system were as described in Cornet et al. (2002) with slight modifications. Endometrial biopsies were collected in ice-cold, phenol red-free HEPES-buffered Dulbecco modified Eagle medium/Hams F-12 (DMEM/F-12; Invitrogen, Australia) with 1% antibiotic–antimycotic solution (final concentrations: 100µg/ml of penicillin G sodium, 100µg/ml of streptomycin sulphate, 0.25µg/ml of amphotericin B; Invitrogen). A small portion of the tissue was sent for histopathology to evaluate the cycle stage, which was recorded as either proliferative or secretory. Endometrial samples were cut into pieces of ∼1mm³ with a sterile surgical blade and placed in tissue culture inserts (Millipore, Bedford, MA, USA; six explants/12mm insert). DMEM medium, free of serum and phenol red (supplemented with 1% antibiotic–antimycotic solution), was placed in the lower chamber (300µl in 12-mm inserts). The medium was either hormone free or supplemented with 10nM estrogen (E; Sigma-Aldrich, Australia), 100nM progesterone (P; Sigma-Aldrich) or a combination of both (E + P). At the end of the culture (after 24 or 48h), the explants were stored at −80°C until RNA and protein extraction.

RNA extraction and RT

Total RNA was extracted using Trizol® reagent (Invitrogen). Curettings were homogenized (800µl Trizol® + 200µg of glycogen) and incubated at room temperature for 5min. After the addition of chloroform (0.2×volume of Trizol®), samples were incubated for another 3min at room temperature and centrifuged for 15min at 12 000g (4°C), and the aqueous phase RNA was separated from the DNA/protein fraction, mixed with 400µl of isopropanol and incubated at room temperature for 20min. This was followed by centrifugation for 10min at 12 000g (4°C). The RNA pellet was washed twice with 75% ethanol followed by centrifugation for 6min at 10 000g (4°C). RNA was resuspended in RNase-free water. To eliminate potential genomic DNA contamination, the RNA samples were treated with RNase-free DNase (M6101, Promega, USA) according to manufacturer’s instructions. DNase treatment was performed for 30min at 37°C followed by 10-min inactivation at 65°C. After the DNase treatment, RNA was purified by storing overnight at −20°C in two volumes of 100% ethanol and 0.1 volume of 3M sodium acetate. RNA was centrifuged at 12 000g for 40min at 4°C, the pellet was washed twice with 70% ethanol and the RNA was resuspended in RNase-free water. The quantification and the estimation of purity were derived by measuring the absorbance of each RNA sample at 260 and 280nm. All RNA samples were stored at −80°C until use. RT was performed using Superscript III Reverse Transcriptase (Invitrogen), according to the manufacturer’s instructions. RNA, 0.5µg, from each sample was used.

Real-time quantitative RT–PCR

A Roche Light Cycler and LC fast start DNA master SYBR green kit were used to perform real-time PCR, according to the manufacturer’s instructions. Primer concentrations were 0.5µmol/l. Each set of primers was optimized for annealing temperature and extension times. The primer sequences used and the protocols are summarized in Table I. Relative mRNA expression was determined by measurement against a specific cDNA standard. 18S rRNA was used as a housekeeping gene to normalize all results.

Immunohistochemistry

Formalin-fixed specimens were dewaxed and dehydrated. Endogenous peroxidase activity was blocked before immunohistochemistry by incubation in 3% H₂O₂ in methanol for 10min. After two washes in phosphate-buffered saline (PBS), slides underwent antigen retrieval using 10mmol/l citrate buffer with incubation in a microwave oven for 15min (700W). Slides were allowed to cool in citrate buffer for 20min and were subsequently washed in PBS. Tissue sections were then blocked with protein blocking solution for 10min (protein blocking agent (PBA), Immunon Shandon, PA, USA) to prevent non-specific binding, followed by 1-h incubation with ANXA4 goat polyclonal antibody at room temperature [4µg/ml in 1% bovine serum albumin in PBS (BSA/PBS), raised against the N-terminus (sc-1930), Santa Cruz Biotechnology, Santa Cruz, CA, USA]. Negative control was goat immunoglobulin G (IgG) at 4µg/ml. The primary antibody step was followed by a biotinylated secondary antibody and streptavidin-peroxidase steps at room temperature for 15min each, using reagents of the LSAB + Kit (Dako, Carpentaria, CA, USA). DAB (Sigma) was applied for 5min at room temperature as a chromogen. The intensity of ANXA4 immunostaining in the endometrium was semi-quantified using a graded scale: 0, no staining; 1, weak staining; 2, moderate staining and 3, strong staining.

Protein extraction

Total protein was extracted from endometrial tissues/explants using Trizol® reagent according to manufacturer’s instructions. Following RNA/DNA removal, the protein pellet was washed and dissolved in 1% sodium dodecyl sulphate (SDS). Protein quantification was performed by BCA Protein Assay kit, following manufacturer’s instructions (Pierce Biotech, Rockford, USA).

Western blotting

Ten micrograms of protein from total endometrial tissues was subjected to SDS–polyacrylamide gel electrophoresis (PAGE). The protein samples were mixed with 2× SDS loading buffer with or without 2-mercapto ethanol, and samples were heated at 95°C for 5min (for reducing conditions) or were warmed at 40°C for 15min (for non-reducing conditions). The SDS–PAGE was run on a 10% agarose gel at 100V for 1h. The proteins resolved in the gel were electrophorectically transferred overnight to nitrocellulose membrane (BioRad, USA). The transferred membrane was treated with blocking solution for 1h (SuperBlock; Pierce Biotech). The membrane was then incubated with 1:1000 dilution of the ANXA4 antibody (sc-1930; Santa Cruz Biotechnology) for 1h. This was followed by washing and incubation with horse-radish peroxidase (HRP)-conjugated secondary antibody and finally developed using Supersignal® West Dura Extended Duration Substrate (Pierce Biotech), according to manufacturer’s instructions. The developed films were scanned and assessed by densitometry (Quantity One Software, BioRad). Pre-stained SDS–PAGE standard protein markers (BioRad) were used to calibrate the molecular mass. β-Actin was used as the loading control. The same nitrocellulose filter was treated with Restore Western Blot Stripping Buffer (Pierce Biotech) to remove the anti-ANXA4 antibody and was subjected to western blotting (described as above) using a mouse monoclonal antibody for β-actin (1:4000 dilution; Sigma-Aldrich).

Statistical analysis

Statistical tests were performed using the SPSS 12 statistical analysis package and GraphPad Prism software (version 4.00, GraphPad Software, San Diego, CA, USA). For mRNA and protein profile of ANXA4 during the menstrual cycle, one-way analysis of variance (ANOVA) was used. For the hormonal regulation data, the fold change of the hormone-treated samples over the controls within each experiment was calculated. The effects of treatments on fold change in mRNA expression were analysed using the non-parametric repeated ANOVA (Friedman test) with Dunn correction. P ≤ 0.05 was considered significant.

Results

ANXA4 mRNA expression and protein localization during the menstrual cycle

Temporal expression of ANXA4 mRNA during the menstrual cycle is shown in Figure 1. ANXA4 mRNA is significantly up-regulated during the MS and LS stages of the menstrual cycle compared with all proliferative stages. These results confirm and extend previous observations (Ponnampalam et al., 2004) by significantly increasing the sample number and by utilizing conventional histopathological rather than molecular classification of the menstrual cycle.

ANXA4 protein was localized to the glandular (g) and luminal epithelium (le) in the human endometrium (Figure 2). The staining was heterogenous among glands, but the overall intensity of the staining was high throughout MS and LS stages, M phase and EP, MP and LP phases of the cycle and was significantly reduced during the ES phase (Figures 2 and 3). Glands of EP and MP endometria showed predominant membrane staining (Figure 2A–C), whereas the glands of LS endometria showed nuclear, membrane and cytoplasmic staining (Figure 2G–I).

Cyclic changes in the intensity of ANXA4 in luminal epithelium followed the same pattern as the glands during the menstrual cycle. Endometrium from EP and MP stages showed luminal epithelium with strong but diffuse staining, including membrane staining, of many individual cells (Figure 2K). In contrast, luminal epithelium of LS endometrium showed very strong apical staining (Figure 2L). The strong staining of the nuclei seen in many glandular cells of LS endometrium was not apparent in luminal epithelium. Replacing the ANXA4 antibody with an equivalent amount of control goat IgG resulted in the complete absence of positive immunoreactivity (Figure 2M).

ANXA4 protein expression by western blotting

Figure 4 shows ANXA4 protein expression by western blotting throughout the menstrual cycle, under both reducing and non-reducing conditions. Under reducing conditions, all endometrial samples had a distinct band at 36kDa. LS samples also had an additional band below 6kDa (Figure 4B). Under non-reducing conditions, a very distinct higher molecular weight band between 102and 107kDa was present in high levels during proliferative, MS and LS and menstrual stages (Figure 4A). The 36-kDa band was still present in some MS and LS samples, while a 6-kDa band can be seen in both LS and M samples (Figure 4A). Both the higher molecular weight band and the 36-kDa band were fainter in ES samples under non-reducing and reducing conditions. The other faint bands observed in the blot may be because of background or non-specific staining and were not consistent between samples.

Steroid hormone regulation of ANXA4 transcription after 24 and 48h

Expression of ANXA4 mRNA in explants from both proliferative and secretory phases did not change in response to hormonal treatment after 24h (Figure 5A and B). After 48h, there was a significant increase (P ≤ 0.05) in ANXA4 mRNA levels in explants of proliferative endometria treated with progesterone compared with the control (Figure 5C). ANXA4 mRNA levels did not change significantly in explants of proliferative endometria that were treated with estrogen and progesterone or estrogen alone. Explants from secretory endometria did not show any significant responses to hormonal treatments (Figure 5D).

Steroid hormonal regulation of ANXA4 protein expression after 48 h

The effects of steroid hormones on ANXA4 translation were examined by western blotting under non-reducing conditions, in proliferative endometria after 48h of hormone treatment. Although there was a trend towards an increase in the progesterone-treated explants (P = 0.067), the 36-kDa ANXA4 monomer did not vary significantly between the treatment groups (Figure 6A). The higher molecular weight band (102–107kDa; Figure 6B) and the <6-kDa (Figure 6C) band were significantly higher in proliferative endometria treated with progesterone compared with the control (Figure 6).

Discussion

In this study, we report the temporal expression of ANXA4 mRNA and protein during the menstrual cycle and the hormonal regulation of its transcription and translation. In accordance with the previous studies (Kao et al., 2002; Riesewijk et al., 2003; Ponnampalam et al., 2004; Mirkin et al., 2005), we found that ANXA4 mRNA was significantly up-regulated during the MS and LS stages of the menstrual cycle; however, ANXA4 protein expression was high throughout the cycle apart from the ES stage, when it was significantly reduced. Protein localization was restricted to endometrial glands and luminal epithelium. The study also found that ANXA4 transcription and translation can be induced in proliferative phase endometrial explants after 48-h exposure to progesterone in vitro.

It is interesting to note that the temporal expression of ANXA4 protein during the menstrual cycle does not correlate with the ANXA4 mRNA expression. Discordant patterns of expression between mRNA and protein of many annexins during the cell cycle have been reported previously (Raynal et al., 1997), suggesting that annexins could be regulated post-transcriptionally. ANXA4 protein is reported as having a half-life of nearly 4 days (Raynal et al., 1997), which could also contribute to the observed differences between mRNA and protein levels. While only speculation at this stage, it is possible that both post-transcriptional and post-translational events could contribute to the discordant patterns of mRNA and protein expression seen in this study. For example, differences in the 5′-UTR structure have been shown to affect mRNA processing, localization, stability and translation efficiency (Child et al., 1999; Hanfrey et al., 2002; Lee et al., 2002; Li et al., 2003), and different isoforms of the same gene may also be regulated differently post-transcription. Complexities in the regulation of transcription and translation that could also contribute to discrepancies between mRNA and protein levels have been reported in the mouse, where ANXA4 gene have multiple transcripts which code for similar proteins that are differentially expressed in different tissues (Li et al., 2003). Transcript differences originate through the first exons of the 5′ non-coding regions (5′-UTR) due to alternative promoter usage. Given that human and mouse ANXA4 genes both have similar structures with 13 exons, of which 12 are coding, it is possible that similar complex regulatory mechanisms occur in the human. If such additional transcripts exist, it is also possible that the primers used in this study did not amplify all of them.

Because ANXA4 mRNA is significantly increased during the MS and LS stages when progesterone levels are high in the endometrium, it is not surprising that ANXA4 mRNA and protein were significantly up-regulated in vitro by progesterone acting on endometrial explants. Mirkin et al. (2005) reported that the ANXA4 gene has multiple potential estrogen responsive element (ERE)-like sequences, but has no progesterone responsive element (PRE)-like sequences. This group therefore suggested that ANXA4 is a likely candidate for direct regulation by the estrogen receptor. Our findings contradict this suggestion. It should also be noted that the potential ERE-like sequences found in ANXA4 were not tested experimentally. During the menstrual cycle, circulating progesterone levels gradually increase during the ES stage and do not peak until the MS stage, which may explain why the rise in ANXA4 does not occur until the MS stage. In addition, the in vitro response to progesterone takes 48h, which means that even if in vivo progesterone levels are high enough to influence ANXA4 transcription in the second half of the ES stage, an increase in mRNA would only become apparent in the MS stage.

ANXA4 is a 36-kDa protein that has the capacity to form higher molecular weight aggregates at the membrane interface (Zanotti et al., 1998). Presumably, the higher molecular weight band (102–107kDa) seen under non-reducing conditions (Figure 4) in high levels throughout the menstrual cycle (except during ES stage) comprises these aggregates. ANXA4 has a short N-terminal of 12 amino acids, which is cleaved when phosphorylated by protein kinase C (Kaetzel et al., 2001). It is likely that the lower molecular weight band (<6kDa) corresponds to this cleaved N-terminal. The antibody we used in this study is raised against the N-terminal of ANXA4 protein; thus, the cleaved N-terminal may still contain the epitope recognized by the antibody. Our immunohistochemistry findings support the western blotting data, as the membrane staining is seen at all stages of the cycle except during the ES stage. Our hormonal regulation data imply that progesterone not only increases the membrane aggregates of ANXA4 (102–107kDa) but also increases the phosphorylation of ANXA4 (as evidenced by an increase in the intensity of the <6-kDa band).

Although the biological functions of ANXA4 in human endometrium have yet to be elucidated, results from previous studies have shown that ANXA4 is important in kidney function and that membrane-associated ANXA4 protein is important in regulation of ion channels (Chan et al., 1994; Kaetzel et al., 1994), exocytosis (Sohma et al., 2001), Ca²⁺-regulated signal transduction (Raynal et al., 1996) and membrane permeability (Hill et al., 2003). ANXA4 can alter the biophysical properties of model membranes and consequently reduce water and proton permeability (Hill et al., 2003). It is most abundantly expressed in the kidneys of many species, the organ which plays the primary role in maintaining the ion and water homeostasis of the body. ANXA4 has also been found in close association with membranes of many secretory and absorptive epithelia (Kaetzel et al., 1989; Massey et al., 1991). Regulation of the fluid environment in the uterus may be critical for many reproductive events, including sperm and embryo transport and implantation. Although cyclic changes in the volume and composition of uterine fluid are known to occur, little is known of the molecular regulation of these changes. On the basis of our own and previous studies, it seems more likely that ANXA4 may play an important role in regulating ion and water movement across the endometrial epithelium.

Previous studies have shown that estrogen stimulates membrane permeability, whereas progesterone inhibits it (Gorodeski, 1998; Jablonski et al., 2003). However, endometrial glands also have absorptive function, which can be stimulated by progesterone (Naftalin et al., 2002). In the rat uterus, progesterone treatment induces net fluid absorption, while estrogen causes net fluid secretion (Salleh et al., 2005). Fluid movement across epithelia follows the movement of solutes, particularly Na⁺ and Cl⁻ ions, which are essential for electrolyte and fluid absorption and secretion, respectively (Chan et al., 2002). Evidence suggests that amiloride-sensitive endometrial Na⁺ channels (ENaC) are stimulated by progesterone (Salleh et al., 2005). ENaC inhibits both cAMP-activated (Chan et al., 2000) and Ca²⁺-activated (Wang et al., 2001) chloride conductance, which in turn will result in increased fluid absorption. Although ANXA4 has been shown to have no effect on cAMP-regulated Cl⁻ channel, it inhibits Ca²⁺ and calmodulin-dependent protein kinase II (CAMKII) activated Cl⁻ conductance (Chan et al., 1994). Taken together, these results suggest that one mechanism by which progesterone regulates ion and water transport in endometrial epithelium is by increasing the expression of ENaC and ANXA4.

In the human, estrogen levels gradually increase during the LP phase and reach maximum levels just before ovulation. Because estrogen induces both fluid secretion and membrane permeability and ANXA4 counteracts these functions, the significant reduction in ANXA4 protein during the ES phase of the menstrual cycle may be an indirect or downstream effect of the rise in estrogen levels during the LP phase. Whether estrogen facilitates ANXA4 protein degradation remains to be elucidated.

The first signs of apoptosis in the endometrium are seen during MS and LS stages of the menstrual cycle (von Rango et al., 1998). ANXA4 has also been shown to play a role in the early phases of apoptosis (Sohma et al., 2001; Herzog et al., 2004), with translocation from the nucleus to the cytosol during this process (Gerner et al., 2000). It is thus possible that cytosolic ANXA4 during the LS stage may be associated with early phases of apoptosis before menstruation.

In conclusion, this study demonstrates the mRNA and protein profile of ANXA4 in human endometrium throughout the menstrual cycle and provides the first evidence that both are up-regulated by progesterone in vitro. Although possible biological functions of ANXA4 are discussed, its functional role in endometrium requires further investigation.

Acknowledgements

We thank Fiona Lederman and Jacqui Donoghue for their technical assistance with immunohistochemistry and western blotting, respectively. The endometrial tissue samples were collected by research nurses, Nancy Taylor and Nikki Sam. We are grateful to Dr Beatrice Susil for histopathological evaluation of all the endometrial tissue samples. We gratefully acknowledge Dr Mark Lawrence, Prof David Healy, Dr Beverley Vollenhoven, Dr Elizabeth Farrell and other gynaecologists at Southern Health for their help in obtaining the tissue samples. P.A.W.R is a Principal Research Fellow of National Health and Medical Research Council of Australia (NHMRC) (grant no. 143805).

References