Epigenetic regulation may be involved in modulation of gene expression during the normal cyclic changes of the human endometrium. We investigated expression of DNA methyltransferases (DNMTs) in endometrium during the menstrual cycle and the influence of sex steroid hormones on DNMT in endometrial stromal cells (ESC) in culture.
Expression of DNMT1, DNMT3a and DNMT3b was assessed by immunohistochemistry and real-time RT–PCR in endometrial tissue (n = 42 women). ESC (n = 3 women) were cultured with estradiol and medroxyprogesterone acetate (E + MPA) for 17 days, and DNMT mRNA levels were measured by real-time RT–PCR.
Nuclei of both epithelial and stromal cells immunostained for DNMT1, DNMT3a and DNMT3b during each phase of the menstrual cycle. Tissue levels of DNMT1 and DNMT3a mRNA were significantly lower in the mid-secretory phase than in the proliferative phase (P < 0.01). For DNMT3b, the change in mRNA levels showed a similar trend to that for DNMT3a. In ESC culture, DNMT3a and DNMT3b mRNA levels were significantly decreased by E + MPA treatment (P < 0.01 and P < 0.05, respectively) at Day 8 and Day 17.
DNMT mRNAs declined in the human endometrium during the secretory phase, and E + MPA down-regulated DNMT3a and DNMT3b mRNAs in ESC in culture. These results suggest that DNMTs have regulatory functions in gene expression that is associated with decidualization.
The human endometrium, which mainly consists of endometrial epithelial cells and endometrial stromal cells (ESC), has cyclic changes in morphology and in function depending on female sex steroid hormone exposure. These cells actively proliferate under estrogen exposure during the proliferative phase and thereafter differentiate under progesterone exposure during the secretory phase. A number of genes are involved in proliferation, differentiation and tissue breakdown in the endometrium under the influence of female sex steroid hormones (Sugino et al., 2002a, 2004; Ace and Okulicz, 2004; Ponnampalam et al., 2004). A great number of genes are up-regulated or down-regulated in the human endometrium during decidualization, which occurs around the time of embryo implantation (Popovici et al., 2000; Kao et al., 2002; Okada et al., 2003; Riesewijk et al., 2003; Ace and Okulicz, 2004; Mirkin et al., 2005), suggesting the presence of complex mechanisms of gene expression. However, little is known about the molecular mechanisms involved in the regulation of gene expression in the human endometrium.
In the last decades, it has become clear that epigenetic regulation, including DNA methylation and histone modification, plays a key role in transcriptional regulation. DNA methylation occurs at cytosines within CpG dinucleotides that are clustered frequently in regions of ∼1–2 kb in length, called CpG islands, in or near the promoter and first exon regions of genes (Esteller et al., 2002; Jones and Baylin, 2002). DNA methylation at the CpG dinucleotides is a post-replication event catalyzed by DNA methyltransferase (DNMT) (Smith, 1994) that adds a methyl group to the cytosine ring to form methyl cytosine, which establishes normal methylation patterns during embryogenesis and reproduces these patterns during replication of adult cells (Li et al., 1993; Razin and Kafri, 1994). DNA methylation is an important mechanism of epigenetic gene regulation, and is involved in genomic imprinting, X chromosomal inactivation, aging, mutagenesis and regulation of tissue-specific gene expression during development and adult life (Li et al., 1993; Razin and Kafri, 1994; Ohgane et al., 1998; Imamura et al., 2001; Li, 2002; Shiota and Yanagimachi, 2002). Aberrant methylation of CpG islands, located in the 5′-promoter region of genes, is commonly associated with transcriptional inactivation (Nan et al., 1998). Such inactivation is well known in various human cancers, especially in tumor suppressor genes (Ushijima and Okochi-Takada, 2005).
Several DNMTs exist. DNMT1 functions as a ‘maintenance’ DNMT in mammalian cells and is therefore responsible for accurately replicating genomic DNA methylation patterns during cell division (Liu et al., 1998). On the other hand, DNMT3a and DNMT3b are thought to catalyze de novo methylation of DNA (Hsieh, 1999). Recent research also shows that DNMT1, DNMT3a and DNMT3b co-operatively maintain DNA methylation (Ting et al., 2004).
Recently, aberrant expression of DNMTs was observed in endometriosis, which is a non-cancerous ectopic growth of the human endometrium (Wu et al., 2007). Aberrant DNA methylation of the promoter region is involved in aberrant gene expression of steroidogenic factor-1 and estrogen receptor in endometriosis (Xue et al., 2007a, b; Utsunomiya et al., 2008). Furthermore, in the eutopic endometrium of women with endometriosis, reduced expression of HOXA10, which is a transcription factor and plays an important role in uterine receptivity, was found to be due to DNA methylation of the promoter region (Wu et al., 2005). In addition, aromatase expression in ESC is under epigenetic regulation (Izawa et al., 2008). Histone acetylation is involved in differentiation of ESC and endometrial epithelial cells (Sakai et al., 2003; Uchida et al., 2005). These reports led us to hypothesize that epigenetic regulation is involved in the normal cyclic changes of the human endometrium. To test this hypothesis, we investigated changes in the expression of DNMTs in the normal endometrium during the menstrual cycle and the influence of female sex steroid hormones on DNMT expression in ESC.
Materials and Methods
This study was reviewed and approved by the Institutional Review Board of Yamaguchi University Graduate School of Medicine. Informed consent was obtained from the women before collection of any samples for this study.
Endometrial tissues were collected from hysterectomy specimens or biopsies for histological dating of the endometrium in 42 women with regular menstrual cycles (aged 22–50 years, median 36.9 years). All of the women received no steroid medications. Endometria were dated according to the histological criteria by Noyes et al. (1950) and were classified into four different groups: proliferative phase (days 6–14, n = 14), early secretory phase (days 15–18, n = 10), mid-secretory phase (days 19–23, n = 11) and late secretory phase (days 24–28, n = 7). Endometrial samples were snap-frozen in liquid nitrogen and stored at −80°C until RNA isolation. For immunohistochemistry, the tissue specimens were fixed in 10% buffered formalin and embedded in paraffin.
Immunohistochemistry for DNMTs in the endometrium was performed on 4 µm thick paraffin sections mounted on silane-coated glass slides (Dako, Glostrup, Denmark) using anti-DNMT1 monoclonal antibody (IMG-261 mouse; Imgenex, San Diego, CA, USA), anti-DNMT3a polyclonal antibody (RB1852 rabbit; Abgent, San Diego, CA, USA) or anti-DNMT3b polyclonal antibody (RB1906 rabbit; Imgenex), as reported previously (Sugino et al., 1996, 2002b). Briefly, the sections were deparaffinized in xylene and dehydrated through a graded series of ethanol. For antigen retrieval, the sections were autoclaved at 121°C for 15 min. Endogenous peroxidase activities and non-specific binding were then blocked with 1% H2O2 and 10% normal rabbit serum (Nichirei, Tokyo, Japan) for DNMT1 or 10% normal goat serum (Nichirei) for DNMT3a and DNMT3b, respectively. The sections were then incubated with the primary antibody diluted 1:50 overnight at 4°C. Parallel control sections were incubated with normal mouse serum or normal rabbit serum (Dako) instead of specific primary antibodies. Biotinylated antimouse antibody (Nichirei) for DNMT1 or biotinylated antirabbit antibody (Nichirei) for DNMT3a and DNMT3b was used as the secondary antibody. After the sections were rinsed in phosphate-buffered saline (PBS), they were incubated in streptavidin–peroxidase complex (Nichirei) for 5 min, rinsed in PBS and then visualized with diaminobenzidine and counterstained with hematoxylin. Dark brown nuclear staining indicated a positive reaction. The histological sections were independently evaluated by three observers, and relative intensities of the signals were estimated at + (weakly positive) to +++ (strongly positive).
For ESC culture, endometrial tissues that were histologically diagnosed as being in the late proliferative phase were used. Tissue samples were obtained from three individuals, and cells from an individual were cultured in triplicate. ESC were isolated as reported previously (Sugino et al., 2000, 2002c). Endometrial tissues were washed with phenol red-free Dulbecco's modified Eagle's medium (DMEM) (Invitrogen, Paisley, UK) containing 4 mmol/l glutamine (Invitrogen), 50 µg/ml streptomycin (Invitrogen) and 50 IU/ml penicillin (Invitrogen), and minced into small pieces of <1 mm3. After the enzymatic digestion of minced tissues with 0.2% collagenase (Sigma, St Louis, MO, USA) in a shaking water bath for 2 h at 37°C, ESC were separated by filtration through a 70 µm nylon mesh. The filtrates were washed three times, and the number of viable cells was counted by trypan blue dye exclusion. The homogeneity of the stromal cell preparation (98%) was verified by immunocytochemistry using an antibody against vimentin, a specific marker of stromal cells. ESC were seeded at 105 cells/cm2 in 75 cm2 tissue culture flasks and grown until confluence in phenol red-free DMEM containing glutamine, antibiotics and 10% dextran-coated charcoal-stripped fetal calf serum (FCS) (Biological Industries, Kibbutz Beit Haemek, Israel) at 37°C, 95% air and 5% CO2. If necessary, cells were subcultured in 75 cm2 tissue culture flasks after the first passage until confluence. For treatments, cells were subcultured into 25 cm2 tissue culture flasks (second or third passage), and the cell culture medium was changed to the treatment medium at 80% confluence.
To examine the effect of estrogen and progesterone on DNMTs mRNA levels in ESC, cells were incubated with treatment medium (phenol red-free DMEM supplemented with glutamine, antibiotics and 2% stripped FCS) containing a combination of estradiol (10−8 M) (Sigma) and medroxyprogesterone acetate (MPA, 10−6 M) (Sigma) for 17 days at 37°C, in 95% air and 5% CO2. The concentrations of estradiol and MPA, and the period of incubation were based on our previous reports (Sugino et al., 2000, 2002d). The medium was changed every other day. Decidualization was confirmed by morphology and mRNA expression of insulin-like growth factor-binding protein-1 (IGFBP-1), which is a specific marker of decidualization (Giudice et al., 1992; Sugino et al., 2000). Total RNA was isolated from cultured cells, and RT–PCR for DNMTs was performed as described below, with a duplicate PCR for each culture.
Total RNA was extracted from endometrial tissues and cultured cells using Isogen (Wako, Osaka, Japan), and real-time RT–PCR was performed as reported previously (Asada et al., 2008). RT reactions were performed with ExScript RT reagent kit (TAKARA, Kyoto, Japan) according to the manufacturer's protocol. Briefly, 2 µg of total RNA was incubated with 4 µl of 5× ExCript buffer, 1 µl of dNTP mixture (10 mM each), 1 µl of Random primers (50 µM), 0.5 µl of ExCript reverse transcriptase (200 U/µl) and 0.5 µl of RNase inhibitor (40 U/µl) in 20 µl of reaction mixture at 42°C for 15 min, after which the reverse transcriptase was inactivated by heating the samples at 95°C for 2 min. The complementary DNA (cDNA) was immediately used for PCR. All PCRs were performed using SYBR Premix Ex Taq (TAKARA) and a LightCycler (Roche Applied Science, Basel, Switzerland). Briefly, 2 µl of aliquots containing cDNA were amplified in a total volume of 20 µl containing 4 µl of a 5× SYBR PreMix Ex Taq and 0.2 µM each primer. For internal controls, TATA box-binding protein (TBP) cDNA or glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was also amplified. According to the previous report (Girault et al., 2003), the following primers were used: DNMT1 (forward, 5′-TACCTGGACGACCCTGACCTC-3′, reverse, 5′-CGTTGGCATCAAAGATGGACA-3′) (product size: 103 bp); DNMT3a (forward, 5′-TATTGATGAGCGCACAAGAGAGC-3′, reverse, 5′-GGGTGTTCCAGGGTAACATTGAG-3′) (111 bp); DNMT3b (forward, 5′-GGCAAGTTCTCCGAGGTCTCTG-3′, reverse, 5′-TGGTACATGGCTTTTCGATAGGA-3′) (113 bp); TBP (forward, 5′-TGCACAGGAGCCAAGAGTGAA-3′, reverse, 5′-CACATCACAGCTCCCCACCA-3′) (132 bp); IGFBP-1 (forward, 5′-CGAAGGCTCTCCATGTCACCA-3′, reverse, 5′-TGTCTCCTGTGCCTTGGCTAAAC-3′) (98 bp) and GAPDH (forward, 5′-AGGTGAAGGTCGGAGTCA-3′, reverse, 5′-GGTCATTGATGGCAACAA-3′) (99 bp). All samples were run in duplicate. For appropriate negative controls, the RNA template was replaced with nuclease-free water in each run. Melting curves of the products were obtained after cycling by a stepwise increase of temperature from 55 to 95°C. At the end of 40 cycles, reaction products were separated electrophoretically on an agarose gel and stained with ethidium bromide for visual confirmation of the PCR products.
Statistical analysis was carried out using the Statistical Package for the Social Sciences for windows 13.0. To evaluate whether tissue mRNA levels significantly vary during the menstrual cycle, the Tukey honest significant difference test was used. For ESC cultures, differences in mRNA levels were determined using Duncan's new multiple range test. P < 0.05 was considered to be significant.
Nuclei of both epithelial cells and stromal cells in tissue sections immunostained for DNMT1, DNMT3a and DNMT3b during each phase of the menstrual cycle. Representative results from the late proliferative phase are shown in Fig. 1. The staining intensities did not vary among the menstrual phases.
Changes in DNMT1, DNMT3a and DNMT3b mRNA levels in the endometrial tissue are shown in Fig. 2. DNMT1 mRNA levels were significantly lower in the mid-secretory phase than in the other menstrual phases (Fig. 2A). DNMT3a mRNA levels were significantly lower in the secretory phase than in the proliferative phase (Fig. 2B), being lowest in the mid-secretory phase (Fig. 2B). The pattern of change in the DNMT3b mRNA levels was similar to that for DNMT3a, but the changes were not statistically significant (Fig. 2C).
Since mRNA levels of DNMT1, DNMT3a and DNMT3b in the endometrium were lower in the mid-secretory phase than in the proliferative phase, we examined whether DNMT mRNA expression is influenced by progesterone and estrogen. We therefore focused on ESC, which differentiate to decidualized stromal cells under the influence of progesterone and estrogen during the secretory phase. In order to induce decidualization in vitro, ESC were treated with MPA and estradiol for 17 days. As shown in Fig. 3A, mRNA expression of IGFBP-1, a specific marker of decidualization, was clearly induced by MPA and estradiol for 17 days. DNMT3a and DNMT3b mRNA levels were gradually decreased by MPA + estradiol and were significantly lower in the MPA + estradiol group than in the control group on days 8 and 17 after treatment (Fig. 3C and D). However, DNMT1 mRNA levels did not change during the treatment (Fig. 3B).
The present study showed changes in the level of DNMT mRNAs in the human endometrium during the menstrual cycle. DNMT3a mRNA levels were significantly lower in the secretory phase than in the proliferative phase, being lowest in the mid-secretory phase. The pattern of change in DNMT3b mRNA levels was similar to that for DNMT3a. Furthermore, we showed that DNMT3a and DNMT3b mRNA level in ESC was down-regulated by MPA and estrogen. DNMT3a and DNMT3b are responsible for de novo CpG methylation (Hsieh, 1999). CpG methylation of the promoter region inactivates gene expression (Nan et al., 1998; Ushijima and Okochi-Takada, 2005). These findings lead us to speculate that the down-regulation of DNMT3a and DNMT3b mRNAs may be associated with expression of the genes that are induced during decidualization. In fact, a great number of genes are up-regulated and some genes are newly expressed in the human endometrium undergoing decidualization (Popovici et al., 2000; Kao et al., 2002; Riesewijk et al., 2003; Ace and Okulicz, 2004; Mirkin et al., 2005). Further studies are needed to find out which genes are regulated by DNA methylation during decidualization.
In the present study, DNMT1 mRNA levels in the endometrial tissue were significantly lower in the mid-secretory than in the proliferative phase, whereas DNMT1 mRNA levels were not affected by MPA and estradiol in ESC undergoing decidualization. Therefore, the low levels of DNMT1 in the mid-secretory phase endometrium may reflect the levels in the endometrial epithelium rather than the levels in the ESC. DNMT1 is responsible for accurately replicating genomic DNA methylation patterns to maintain genome stability during cell division (Liu et al., 1998). Endometrial epithelial cells do not proliferate during the mid-secretory phase, which seems to be compatible with the decreased DNMT1 expression during the mid-secretory phase.
There seems to be a discrepancy in this study between mRNA and protein levels for DNMT, and this may be due to the different sensitivities of RT–PCR and immunohistochemistry.
Little information is available regarding the regulation of DNMT expression. Interestingly, the present study showed that DNMT3a and DNMT3b are under the regulation of female sex steroid hormones, suggesting that DNA methylation may be influenced by female sex steroid hormones. It has been reported that DNA methylation status can be altered by a variety of factors including steroids and vitamins (Shiota, 2004). On the other hand, DNA methylation affects estrogen receptors in endometria, mammary glands and myometrium (Lapidus et al., 1996; Iwase et al., 1999; Giacinti et al., 2006; Asada et al., 2008). These findings suggest a close relationship between DNA methylation and female sex steroid hormones. However, further study is needed to clarify the molecular mechanisms of regulation of DNMTs expression.
The phase-specific and transient changes in the DNMT mRNAs during the menstrual cycle may suggest that DNA methylation status is changeable during the menstrual cycle, which may lead to changes in transcription levels of some genes. This is supported by recent reports that DNMTs are involved in both methylation and demethylation of CpG dinucleotides in human cells with cyclical changes in DNA methylation status (Kangaspeska et al., 2008; Metivier et al., 2008).
This study showed that DNMT mRNA levels change in the human endometrium during the menstrual cycle and that DNMT3a and DNMT3b mRNAs in ESC can be regulated by female sex steroid hormones. These results suggest that DNMTs have regulatory functions on gene expression in the human endometrium. Further studies are needed to show a potential role of epigenetic regulation in gene expression that is associated with decidualization.
Y.Y.: conception and design, acquisition of data, analysis of data and drafting the article; H.A., L.L., I.T., R.M., K.T., T.T., A.M. and H.T.: acquisition of data and N.S.: conception and design, interpretation of data, drafting the article and final approval.
This work was supported in part by Grants-in-Aid 17791121, 18791158, 19791153 and 20591918 for Scientific Research from the Ministry of Education, Science, and Culture, Japan.
- gene expression
- cell nucleus
- dna modification methylases
- menstrual cycle
- reverse transcriptase polymerase chain reaction
- rna, messenger
- gonadal steroid hormones
- stromal cells
- medroxyprogesterone acetate
- dnmt3a gene
- european society of cardiology