Molecular Human Reproduction, Vol. 9, No. 11, 671-679,
November 2003
© 2003 European Society of Human Reproduction and Embryology
Article |
Large scale validation of human N-myc Downstream-Regulated Gene (NDRG)-1 expression in endometrium during the menstrual cycle
Submitted on June 6; accepted on July 14, 2003
1 MetrioGene Biosciences Inc., 6100 Royalmount Avenue, Montreal, Quebec, H4P 2R2, 2 Warnex Inc., 3885, Bld Industriel, Laval, Quebec H7L 4S3 and 3 Research center of CHUM, 1560, Sherbrooke E., Montreal, Quebec H2L 4M1, Canada
4 To whom correspondence should be addressed. e-mail: kshazand{at}metriogene.com
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
A major challenge in the comprehension of the endometrial transformations leading to the completion of each menstrual cycle in humans is in the identification of specific molecular pathways underlying these monthly turnovers. Towards this goal we compared, by the differential display technique, the relative expression of mRNA in endometrial biopsies harvested in individuals (n = 48) either at the proliferative or the secretory phase of the menstrual cycle. We isolated a cDNA fragment homologous to NDRG1 (N-myc Downstream-Regulated Gene-1) that is present in markedly higher amounts in the secretory phase. Northern blot analysis and quantitative real time PCR experiments confirmed this result in distinct cohorts of individuals (44 and 560 respectively). A closer examination of data showed that the highest mRNA levels were found during the range of 25–28 days of the uterine cycle. Consistent with the mRNA data, the temporal profile of the NDRG1 protein showed a 15-fold increase during the secretory phase, as demonstrated by using semi-quantitative dot blot analyses (n = 92). Immunohistochemical localization revealed that NDRG1 was expressed both in epithelial and stromal cells. This large scale validation of the NDRG1 mRNA and protein increase in endometrium during the secretory phase is consistent with its differentiation-related function described in other tissues and its potential involvement in the window of implantation of the human endometrium, as suggested by previous chip-based evidence.
Key words: endometrium/menstrual cycle/NDRG1
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The cyclic growth, differentiation and breakdown of the human endometrium represent perhaps the most dynamic example of steroid-driven tissue plasticity in the adult. The endometrial tissue is composed of epithelial cells and connective tissue called stroma. During the proliferative (P) phase of the menstrual cycle (days 1–14), different cell types in the endometrium coordinately undergo extensive growth in response to the synergistic action of estrogens, growth factors and cytokines that leads to ovulation (Wang and Chard, 1999; Bigsby, 2002). This temporal checkpoint initiates a global tissue maturation and differentiation process, taking place in the secretory (S) phase (days 14–28), mainly governed by progesterone. During the S-phase, the functionalis epithelium thickens, epithelial cells elongate and ciliogenesis takes places both in epithelial and surface cells (Campbell et al., 2000). This rapid steroid-mediated tissue development, which rivals that of many neoplasias, allows blastocyst nidation during a specific time-frame called the ‘window of implantation’ at days 20–24, at the progesterone secretion peak (Lessey et al., 1998; Giudice et al., 1999). In the absence of implantation and the continued progestational environment of pregnancy, the superficial functionalis layer of the endometrium loses its receptive state, undergoes degradation, and is expelled with menstrual blood flow (Salamonsen et al., 1999). Within a period of 5–6 days, the entire lining regenerates without scarring, constituting a remarkable example of controlled tissue remodelling.
A complex cascade of interplaying genes governs the entire menstrual cycle process. Many steroid-induced endometrial genes have been identified and their expression pattern during the menstrual cycle has been characterized in vivo using various approaches (Rodgers et al., 1994; Apparao et al., 2001). Some endometrial transcripts are directly regulated by progesterone during the S-phase, e.g. glycodelin, insulin-like growth factor-1 (IGFBP-1) and tissue inhibitor of metalloproteinase-3 (TIMP-3) (Dong et al., 2002; Tseng et al., 2002). Given the spectacular impact of progesterone on tissue remodelling during the S-phase and its downstream consequences on critical processes such as fertility, identification and characterization of molecular intermediates of this specific cascade are of great interest. To this end, we carried out a comparative gene expression profiling study between the P- and S-phases of the menstrual cycle in order to identify new subsets of genes modulated during the peak of progesterone, occurring between days 17–25 (Tabibzadeh et al., 1998; Nikas et al., 2000). We chose to use a screening approach, differential display (DD), which allows for analysis of individual, and simultaneously several, samples. We focus herein on the identification and characterization of a cDNA fragment, namely DD5, that was present in much higher concentrations in S-phase samples. Nucleotide sequence of DD5 revealed that this transcript was identical to that of N-myc Down-Regulated Gene-1 (NDRG1), also termed RTP, CAP 43, rit42, TDD, or PROXY-1 in the literature. This observation was confirmed by northern blot analysis and a validation was performed on a large cohort of samples by real time PCR, allowing a more precise profiling of the gene. Levels of the NDRG1 protein were also shown to be modulated during the menstrual cycle, in epithelial as well as stromal cell types. Finally, mapping of the NDRG1 protein was performed on endometrial tissues by immunohistochemistry. This study constitutes a comprehensive temporal analysis of NDRG1 steady state mRNA and protein levels in endometrium collected from a large population of normally ovulating women.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Patient recruitment
Women who provided endometrial biopsies for this study were undergoing laparoscopy for various medical indications such as tubal ligation, tubal reanastomosis, hysterectomy and invasive surgery. The inclusion criteria of the study were the following: pre-menopausal age; not currently menstruating; menstrual cycles between 21 and 35 days in length; no acute salpingitis; no prior HIV diagnosis; no prior hepatitis B or C infection; not currently pregnant or in the last 3 months; not currently breastfeeding; no current use of a compound selected from the group consisting of GnRH agonists, progestins, Danazol, and oral contraceptives, or in the last 3 months; no current use of intrauterine device or in the last 3 months.
The participants in the study were subdivided into P- and S-phase groups based on the phase of the menstrual cycle indicated by the date of their last menstruation and confirmed by histological examination from a section of formalin-fixed endometrial tissue (Fox, 1995). Subjects ranged between 29 and 38 years of age. No statistical bias due to age or primary medical indication for surgery was observed between the P- and S-phase groups (data not shown). Different subgroups were used for DD and northern experiments, while all samples were analysed by real time PCR. This study was approved by PROCREA Biosciences Ethics Review Board and Internal Review Board of each participating clinical institution. Endometrial biopsies were obtained from women who gave informed written consent.
Isolation and preparation of RNA from biopsies
Endometrial biopsies (n = 560) were performed with Pipelle catheters under sterile conditions, from the uterine fundus. A portion of each biopsy was processed for histological confirmation, and the remainder was immediately transferred in RNAlaterTM (Ambion, Inc., USA), maintained on ice during transport from the hospital to our laboratory, and then stored at –80°C until RNA extraction. Total cellular RNA was isolated directly from endometrial biopsies using RNeasy® Maxi Kit (Qiagen Inc., Canada) according to the manufacturer’s protocol. Briefly, 300–900 mg of biopsy tissue was rinsed in phosphate-buffered saline, pH 7.4, then homogenized using an electric homogenizer. The concentration of the purified RNA was determined by spectrophotometry (OD260). For further analyses, equal RNA loading and integrity was confirmed by showing consistent intensities of 28S and 18S rRNA bands on RNAse-free agarose gel electrophoresis.
Differential display
DD was performed according to the protocol of Liang and Pardee (1998) using total RNA samples, extracted from 24 P- and 24 S-phase endometrial biopsies. Briefly, first strand cDNA synthesis was performed using 200 ng RNA in the presence of 0.01 mol/l dithiothreitol (DTT), 20 µmol/l dNTP, 0.2 µmol/l of anchor primers H-T11M (where M is G or A or C; Genset Corp, USA) and 200 IU of M-MLV reverse transcriptase (RT) in the appropriate reaction buffer (Invitrogen Life Technologies, Canada). RT reactions were performed in a total volume of 20 µl at 65°C for 5 min, followed by 37°C for 1 h, and 5 min at 75°C. S-phase-specific DD PCR products were isolated by subjecting 2 µl of the RT reaction to PCR amplification in the appropriate buffer containing 2 µmol/l dNTP, 1 IU rTaq DNA Polymerase and [
33P]dATP (2500 Ci/mmol) (all from Amersham Pharmacia Biotech) and 0.2 µmol/l of anchor and arbitrary primers (Genset Corp). A total of 18 different combinations of primer pairs (three anchor primers and six arbitrary primers) were used for the PCR reactions Amplifications were performed in a Stratagene Robocycler and involved 40 cycles of 94°C (1 min), 40°C (2 min) and 72°C (1 min) with a final extension of 7 min at 72°C.
Equivalent amount of DNA samples (based on levels of GAPDH expression determined by RT–PCR using specific primers, data not shown) were electrophoresed on a 6% denaturing polyacrylamide gel, and the dried gels were exposed to film. Differentially modulated bands (between P- and S-phase samples) were excised from the gel and DNA was reamplified according to the same PCR reactions described above and cloned using the TA Cloning® Kit (Invitrogen Life Technologies) according to the manufacturer’s instructions. Clones containing the desired fragments were identified by colony-PCR, at least three colonies per cDNA candidate were selected and validated by reverse northern blotting to decrease as much as possible the number of false positive clones. Briefly, this consisted of spotting PCR-amplified DNA fragments purified from the DD gels onto two identical nylon membranes (Boehringer Ingelheim Ltd, Canada) and then hybridized with two 32P-labelled DD PCR reactions, one from each phase, as probes. Still differential signals were kept in the study for the next step. Again, three clones per candidate with the expected differential pattern were selected for nucleotide sequence determination, performed using the ABI Prism Rhodamine cycle sequencing kitTM and run on an automated ABI Prism 310TM Sequencer (Applied Biosystems Canada). Nucleotide sequence alignments were performed with GenbankTM through the BLAST software available at the National Center for Biotechnology Information (NCBI) website (www.ncbi.nih.gov).
Northern blot analysis
Northern blot analyses were performed according to standard protocols (Sambrook et al., 2001). Briefly, 10 µg total RNA samples were resolved on formaldehyde gel and transferred onto Hybond N+® membranes (Amersham Pharmacia Biotech) according to the manufacturer’s instructions. RNA samples were immobilized on the membranes by UV irradiation and blots were hybridized overnight at 42°C in ULTRAhyb (Ambion, Inc.) with specific probes corresponding to cDNA fragments of NDRG1 coding regions (GenBank accession no. NM006096; nucleotide region: 2726–2939) and 18S rRNA (GenBank accession no. X03205; nucleotide region: 613–980) as a loading control. Identity of the amplified cDNA probes was assured by nucleotide sequence determination and fragments were then cloned in pCR®2.1 (TA Cloning® Kit; Invitrogen Life Technologies) and 32P-labelled using a random primer labelling kit (Strip-EZ DNA, Ambion, Inc). Hybridization signals were visualized by phosphorimaging, and quantification of the bands was performed using Gel-Pro® Analyser (version 4.0 software). Background-adjusted band intensities were normalized based on the intensity of 18S rRNA.
Quantitative real time RT–PCR
Total cellular RNA, previously isolated from endometrial biopsies as described above, was used as a template for the production of cDNA. Briefly, 2 µg of RNA was digested in a total of 20 µl with 1 IU DNAse I (Amplification Grade, Invitrogen Life Technologies) in 20 mmol/l Tris–HCl (pH 8.4), 2 mmol/l MgCl2, 50 mmol/l KCl at room temperature for 5 min, then at 65°C for 10 min. The reaction was terminated with 2.5 mmol/l EDTA. DNAse-treated RNA was then reverse-transcribed into cDNA in a total of 50 µl using 400 IU M-MLV reverse transcriptase in the presence of 100 ng random primers, 10 mmol/l DTT, 0.5 mmol/l (final concentration) of each of dGTP, dATP, dTTP and dCTP (all from Invitrogen Life Technologies). Following incubation at 37°C for 1 h, heating to 70°C in the presence of 1 mmol/l EDTA for 15 min terminated the reaction.
A relative standard curve (SC) for real time PCR, representing six 6-fold dilutions of a pool of cDNA, was used as a common set of samples that linked the 16 experimental PCR plates together and permitted the overall analysis of 560 samples (see Statistical data analysis below). Preparation and utilization of this SC as a quality control of the efficiency of amplification of the PCR plate is described elsewhere (Merlen et al., submitted for publication).
Real time PCR reactions were performed using TaqMan® chemistry for NDRG1 and SYBR Green chemistry for CK8 (Applied Biosystems Canada). PCR primers and fluorogenic probe were designed using Primer Express, version 1.5 (Applied Biosystems Canada) to amplify unique sequences in each gene of interest and are presented in Table I.
|
Fluorogenic probe was labelled with the reporter dye 6-carboxyfluorescein (6'-FAM) at the 5' end and with the quencher dye 6-carboxytetramethylrodamine (TAMRA) at the 3' end. In order to exclude genomic DNA contamination during the amplification of NDRG1, the amplicons spanned at least one exon–exon boundary. The CK8 gene has one single exon, making a similar strategy impossible.
Large scale population study of real time PCR reactions involved eight 96-well plates per phase of the menstrual cycle, whereby each plate contained duplicates of each dilution of the SC, duplicates of a no-template control, and duplicates of the 560 endometrial samples.
Quantitative–competitive (QC) algorithm and statistical data analysis
Details of the algorithm used for real time PCR data analysis are described elsewhere (Merlen et al., submitted for publication). Cycle threshold (CT) values were directly used as a measure of gene expression, without conversion in relative copy numbers. Final results (
CT) were transformed from exponential into linear space through 2(–CT) calculation method (ABI user bulletin 2, 1997). CT value is inversely proportional to the initial concentration of a given mRNA, since the higher the level of expression, the lower the number of PCR cycles needed to reach the threshold value of detection of the fluorescent signal.
All CT values >35 were excluded from the study because of the accuracy limits of the PCR machine. Briefly, data from each plate is analysed for experimental errors on duplicates and for detection of ‘bad runs’ (Tukey’s mean-difference plot). In a second step, all plates were assembled, standard curves as well as sample data were compiled in two series of data and analysed in parallel for inter-plate variations and potential statistical bias in the population distribution.
Quality-controlled data were then submitted to statistical comparison tests according to their distribution. In this report, all distributions were normal, allowing the use of Student’s t-tests for the statistical analysis. SPSS and Excel softwares were used for statistical analysis. For all analyses, P < 0.0001 was considered significant. Box plot graphs show the median (line within the box), quartiles (50% of the samples are located within the box), outliers (open cirlces) and extreme values (*). Outliers are between 1.5 box length and 3 box lengths from the box. Extremes are >3 box lengths from each extremity of the box. Whisker caps represent the maximum/minimum values of the population.
Isolation and preparation of proteins from endometrial biopsies
Total cellular protein extracts were isolated directly from endometrial biopsies (n = 92). Briefly, 200–300 mg of biopsy tissue was homogenized at room temperature using an electric homogenizer in the presence of 3 ml of lysis buffer consisting of 20 mmol/l Tris–HCl (pH 7.5), 0.1 mol/l NaCl, 2% sodium dodecyl sulphate, 5 mmol/l EDTA, 0.5 µg/ml leupeptin (Boehringer Ingelheim Ltd), 2 µg/ml aprotinin, and 200 µg/ml phenylmethylsulphonylfluoride (both from Sigma–Aldrich, Canada), lysed for 10 min at 100°C and centrifuged for 20 min at 14 000 g. Protein determination of supernatants was made using the DC Protein AssayTM (Bio-Rad Laboratories Ltd, Canada) according to the manufacturer’s instructions.
Protein dot blotting
Protein dot blot experiments were performed on protein extracts obtained from biopsies (n = 32, P-phase; n = 60 S-phase). Samples were spotted in duplicate using 2.5 and 5 µg of protein in order to avoid saturated signals. Spotting was performed on HybondTM ECLTM nitrocellulose membranes (Amersham Pharmacia Biotech). The membranes were blocked in 2.5% dry milk powder in TST buffer (10 mmol/l Tris–HCl pH 7.5 containing 100 mmol/l NaCl and 0.1% Tween-20), and incubated in a 1:2000 dilution (in 5% milk–TST) of either the rabbit polyclonal anti-NDRG1 antibody (Skuld-Tech) or a goat polyclonal anti-actin antibody (Santa Cruz Biotechnology) for 1 h. The anti-NDRG1 antibody used in this study was raised against a synthetic peptide mimicking the Ni/Cu2+ binding C-terminal sequence of the protein specific for the NDRG1 transcript (Piquemal et al., 1999; Zoroddu et al., 2001). Membranes were then incubated with a 1:2000 dilution of either a horseradish peroxidase (HRP)-conjugated goat anti-rabbit polyclonal antibody (Pharmingen) or a HRP-conjugated mouse anti-goat polyclonal antibody (Santa Cruz Biotechnology). After 1 h incubation with the secondary antibodies, membranes were washed and revealed by ECLTM chemiluminescent system according to the manufacturer’s protocol (Amersham Pharmacia Biotech). Membranes were exposed to Biomax MS films (Kodak). Multiple exposures of the films at variable time-points were done in order to capture an appropriate unsaturated signal intensity of the proteins of interest, and the most accurate ones were selected for densitometry analysis via Molecular Analyst software (Bio-Rad Laboratories). Incubations of the membranes with either non-immune rabbit IgG (Skuld-tech) or secondary antibodies alone were performed as negative controls; no signal was detected in either case (results not shown).
To diminish experimental variation, mean values were calculated between duplicate spots for those in which the coefficient of variation was 
20%; this threshold value of 20% was selected based on a 95% confidence interval criteria. A SC generated by a pool of biopsy protein extract, representative of both phases of the menstrual cycle, and serially diluted in duplicate spots containing 28, 14, 7, 3.5, 1.75 and 0.875 µg of protein, was spotted on each blot (see schematized membrane setup in Figure 4A). This SC was used for two purposes: (i) to transform OD units to µg of protein using the curve regression (see representative examples in Figure 4B), and (ii) to determine the linear zone of OD readings for each autoradiogram. By this means, only OD readings within the range of 0–20 units were used for our analyses and were found globally to represent 95% of our measurements. Finally, quantification of expression levels was performed on absolute amounts of proteins quantified as described in the previous section.
|
Immunohistochemistry
For these experiments, S-phase endometrial biopsies were snap-frozen at –80°C upon isolation. Slides with serial cryosections were fixed in acetone for 3 min at 4°C and then blocked in 10% normal goat serum/PBS for 5 min at 25°C (Dako Diagnostics Canada), washed in PBS, and incubated 2 h at 25°C with either a rabbit polyclonal anti-NDRG1 antibody (Piquemal et al., 1999; Skuld-Tech, Université Montpelier II, France); a mouse anti-human epithelial antigen (clone Ber-EP4 from Dako) or a mouse anti-vimentin (Dako). All antibodies were diluted in 10% normal goat serum/PBS. Slides were washed with PBS prior to incubation (2 h) with the appropriate ALEXA fluor® 546-labelled secondary antibodies (Molecular Probes, USA), and counterstaining with DAPI III (Vysis, USA). Secondary antibodies used alone as negative controls did not yield any staining (results not shown).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Identification of a DD signal (an up-regulated DD fragment) in S-phase biopsies
Total RNA isolated from individual (n = 24 in each phase) endometrial biopsies were subjected to DD assays in order to identify gene products that would discriminate between the P- and S-phases of the menstrual cycle. A representative DD autoradiogram of PCR products showing the presence of DD5 as a clear differential signal in a subset of six proliferative and six secretory phase biopsies is presented in Figure 1. Complementary DNA sequence determination performed on three independent clones revealed that DD5 was homologous to the 3'end of NDRG1 (GenBank accession NM006096).
|
Confirmation of the DD observation by northern blotting
To confirm the DD RT–PCR results, we first measured mRNA levels of NDRG1 throughout the menstrual cycle using northern blot analyses. Total RNA samples were isolated from 44 biopsies equally distributed in the P- and S-phases. The experiment was performed with a different cohort from that used for DD. RNA samples were sequentially hybridized with either NDRG1 or 18S rRNA probes (see Figure 2A). The P- and S-phase populations had a median dating (± SD) of 12 ± 2 and 24 ± 1 days respectively. Results from the densitometry analysis are presented in box plots in Figure 2B. Comparable mRNA levels between P- and S-phases were observed for 18S rRNA (1.12-fold ratio) resulting in a non-significant t-test comparison (P = 0.46), allowing its use as a loading control for the comparison of S- and P-phase endometrial RNA samples. By using 18S rRNA as a normalizer, the NDRG1 transcript is shown to be present in significantly higher amounts (P < 0.0001) by 5.2-fold in the S-phase.
|
Validation of NDRG1 S-phase-specific increase by real time PCR
Real time RT–PCR was used on a large scale population of endometrial biopsy samples (n = 560, 280 in each phase) to validate NDRG1 modulation throughout the menstrual cycle. For this purpose, we have previously developed a quality control procedure (see Materials and methods section) enabling us to assemble data from samples run on several 96-well plates. PCR data from NDRG1 mRNA are presented in Figure 3A.
|
To assess the cellular heterogeneity of endometrial tissue samples, we measured in parallel mRNA levels of an epithelial marker, cytokeratin 8 (CK8). Results showed similar levels of mRNA in the two populations for this gene, suggesting certain homogeneity in the cell content of the biopsy samples (Figure 3B). In addition, we used CK8 as a normalizer of the NDRG1 data, since the high abundancy of 18S rRNA (northern normalizer) made its use impossible in the PCR-based approach. CK8 PCR data were also submitted to the QC process, and normalization of NDRG1 data was finally carried out for each sample (Figure 3C). Higher NDRG1 mRNA levels in the S-phase is reflected by lower CT values. Fold modulation between the S- and P-phase populations was calculated according to 2(–
CT), where CT = (mean CT S-phase) – (mean CT P-phase) (ABI’s User Bulletin 2, 1997). NDRG1 mRNA levels are significantly higher in the S-phase, displaying a fold modulation of x3.5, when normalized with CK8 (P < 000.1, Figure 3C). With a high number of samples, it became possible to subdivide groups according to the day of the menstrual cycle in which each biopsy sample had been collected. This allowed us to monitor more precisely NDRG1 mRNA concentrations. Since no accurate histological dating is possible for the P-phase, only the S-phase samples were subdivided in groups of 5 days for the two first (15–19 and 20–24) and of 4 days (25–28) for the third sub-category of the S-phase. Such an analysis based on CT mean values of each group (Figure 3D) showed that this specific mRNA increased along the S-phase, with highest levels reached in the last sub-category (25–28 days) just before the onset of menses.
Western blot analysis of NDRG1 protein expression in endometrial biopsy samples
To complete NDRG1 profiling in the endometrium, protein levels were examined in biopsies isolated throughout the menstrual cycle.
We analysed the temporal levels of the NDRG1 protein during the menstrual cycle using the same anti-NDRG1 polyclonal antibody. This antibody is directed against the specific Ni/Cu+2 binding C-terminal sequence of NDRG1 (Piquemal et al., 1999) and recognizes a single protein of the expected 43 kDa mass in Western blot analysis of protein extracts of endometrial biopsies (data not shown). The population used in the semi-quantitative dot blot study, i.e. 32 individuals in the P-phase and 60 in the S-phase, has a median dating value (± SD) of 13 ± 6 and 21 ± 4 days respectively. Performing such a study using standard 96-well equipment required the establishment of a quantitative dot blot platform allowing inter-blot comparison. Our approach, in an enzyme-linked immunosorbent assay-like detection set-up (see Figure 4A for a schematized illustration), used two sets (2.5 and 5 µg) of duplicate samples, which were immunoblotted along with a serial dilution set of samples used as reference in duplicate points, generating a SC for each membrane (three representative SC regressions are shown in Figure 4B). Figure 4C shows a representative subset of immunodetected dot blots in which 59 biopsy samples were analysed for NDRG1 quantification. The box plot graph in Figure 4D summarizes NDRG1 protein quantification in S- and P-phase biopsy samples (2.5 µg spots) using transformed OD readings. Numbers on the top of each box represent the number of samples matching all QC criteria and used for the evaluation of the mean values and fold modulations. Two analyses were performed using data obtained with 2.5 and 5 µg spots and similar results were found for NDRG1 with both series of results (data not shown), revealing a statistically significant increase (P < 000.1) from P- to S-phase of 19-fold between the mean values of the two sample groups compared.
Protein localization in the endometrial tissue
Serial cryosections of S-phase endometrial biopsies were processed for immunohistochemistry using antibodies specific for endometrial glandular and stromal cells as localization controls (Ber-EP4 and anti-vimentin, respectively; Tabibzadeh, 1991) as well as the anti-NDRG1 polyclonal antibody. Figure 5 shows the results of two independent immunostainings (slides 1 and 2). Ber-EP4 staining is intense in epithelial cells of endometrial glands (note that identical structures on each slide are indicated by ‘G’) and lack of staining for the stroma (indicated by ‘S’). A reverse pattern is observed in the next section, when using anti-vimentin antibody as a stromal cell marker. Alternatively, both cell types were stained using the anti-NDRG1, suggesting that this protein is ubiquitously expressed in the endometrium, although a higher staining for epithelial cell type is visible.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
This report represents the first comprehensive study of human NDRG1 temporal expression profile, both at steady state mRNA and protein levels, in the endometrium throughout the menstrual cycle. Given the significant change in mRNA as well as protein steady state levels around ovulation (day 14), and the fact that at this temporal window radical hormonal changes take place in endometrium, these results strongly suggest a de-novo sex steroid-dependent induction of expression. We found a significant S/P modulation of NDRG1 mRNA of 5.2-fold (P < 0.0001) using northern blot analysis when normalized with 18S rRNA, and S/P modulation of 3.5-fold (P < 000.1) using real time PCR when normalized with CK8. The tissue staining with the anti-NDRG1 antibody suggests the presence of higher levels of this protein in epithelial cells. Nevertheless, the fact that the CK8 levels are quite stable between the two phases strongly supports the idea of an actual up-regulation of expression rather than a larger number of epithelial cells in the S-phase.
These data are consistent with two other gene profiling studies in the human endometrium using different experimental designs and analysis algorithms, and reporting on 3.2- and 3.5-fold S/P modulation of NDRG1 (RTP) transcript using Affymetrix oligoarrays (Kao et al., 2002; Borthwick et al., 2003).
NDRG1 belongs to the recently identified NDRG family that contains four members, namely NDRG1–4, encoded by four distinct genes (Okuda and Kondoh, 1999; Zhou et al., 1998; Qu et al., 2002). NDRG1 was shown to respond to various inducers such as homocysteine in human endothelial (Kokame et al., 1996) and colon cells (Van et al., 1997), to hypoxia in several cell types (Salnikow et al., 2000b), and to nickel and cobalt mimicking hypoxia in lung cells (Zhou et al., 1998; Salnikow et al., 2000a). As a potential differentiation marker, this gene is down-regulated in transformed cell lines, as compared with their normal counterparts (Park et al., 2000; Cangul et al., 2002; Qu et al., 2002), and its expression was shown to be modulated both in vitro and in vivo by N-myc, P53 and hypoxia-inducible factor (HIF)- (Kurdistani et al., 1998; Okuda and Kondoh, 1999). More specifically, steroid hormones were described among potential inducers of this gene. Estrogen and androgen responsiveness have been reported in breast and prostate cancer cell lines; for example, anti-estrogen ICI182 induces NDRG1 mRNA (Piquemal et al., 1999), whereas testosterone and dihydrotestosterone down-regulate its expression (Lin and Chang, 1997).
No clear function of NDRG1 has been demonstrated in the tissues where this gene is expressed and identification of its endogenous regulators in the human endometrium transcends the goal of the current investigation. Nevertheless, it has been shown that distinct signalling pathways modulate its transcription, mRNA stability, translation and post-translational modification, such as phosphorylation (Agarwala et al., 2000). Given the numerous reports on hypoxia-mediated up-regulation of NDRG1 in various cell lines and specifically in homocysteine-injured endothelium (Kokame et al., 1996), a role for this protein in protection from ischaemic cell damage was proposed. Ischaemic cellular stress leads to induction of HIF-1 (Semenza, 2000), a transcription factor that coordinates the up-regulation of numerous genes involved in cell survival mechanisms and angiogenesis (Sharkey et al., 2000; Semenza, 2002). Transcription of NDRG1 was shown to be dependent on HIF-1 activity (Salnikow et al., 2000b). Involvement of hypoxia-inducible genes in terminal differentiation/angiogenesis of the endometrium has been poorly explored to date. For instance, no study has yet documented in vivo the temporal regulation of HIF-1 in the differentiating human endometrium. However, the fact that expression of NDRG1 protein is maximal in the well-differentiated S-phase endometrium and reduced during the proliferative phase of this tissue is compatible with a growth inhibitory role observed in other systems (Kurdistani et al., 1998). It would also be of great interest to elucidate the transcriptional impact of estrogens and progesterone on the NDRG1 promoter and to investigate a potential synergistic effect of growth factors with sex steroid hormones.
In summary, this report represents a comprehensive study of the human NDRG1 temporal profile, both at mRNA and protein levels, in the endometrium throughout the menstrual cycle.
|
Acknowledgements |
|---|
The authors would like to thank Dr Ann-Muriel Steff, Martin Pagé, Isabelle Thibault and Julie Poliquin-Fleury and all the clinical network team including technicians and nurses for their excellent work, as well as P.Croteau for his helpful input of the statistical data analyses. The authors are also grateful to all the clinicians for their collaboration in the patient recruitment process. This study would not have been possible without the generous contribution of all women who have participated in the study. This work was sponsored by the private company Procrea/MetrioGene. At the time the research work reported in this paper was carried out, all the authors were employees of this company.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
1 ABI’s User Bulletin 2 (1997) Relative quantitation of gene expression. In User Bulletin series from ABI Prism 7700 Sequence detection system, Applied Biosystems.
Agarwala, K.L., Kokame, K., Kato, H. and Miyata, T. (2000) Phosphorylation of RTP, an ER stress-responsive cytoplasmic protein. Biochem. Biophys. Res. Commun., 272, 641–647.[CrossRef][ISI][Medline]
Apparao, K.B., Murray, M.J., Fritz, M.A., Meyer, W.R., Chambers, A.F., Truong, P.R. and Lessey, B.A. (2001) Osteopontin and its receptor alphavbeta(3) integrin are coexpressed in the human endometrium during the menstrual cycle but regulated differentially. J. Clin. Endocrinol. Metab., 86, 4991–5000.
Bigsby, R.M. (2002) Control of growth and differentiation of the endometrium: the role of tissue interactions. Ann. NY Acad. Sci., 955, 110–117.
Borthwick, J.M., Charnock-Jones, D.S., Tom, B.D., Hull, M.L., Teirney, R., Phillips, S.C. and Smith, S.K. (2003) Determination of the transcript profile of human endometrium. Mol. Hum. Reprod., 9, 19–33.
Campbell, S., Larsen, J., Seif, M.W., Allen, T.D., Knox, F., Jones, C.J. and Aplin, J.D. (2000) Mosaic characteristics of human endometrial epithelium in vitro: analysis of secretory markers and cell surface ultrastructure. Mol. Hum. Reprod., 61, 41–49.
Cangul, H., Salnikow, K., Yee, H., Zagzag, D., Commes, T. and Costa, M. (2002) Enhanced expression of a novel protein in human cancer cells: a potential aid to cancer diagnosis. Cell Biol. Toxicol., 18, 87–96.[CrossRef][ISI][Medline]
Dong, J.C., Dong, H., Campana, A. and Bischof, P. (2002) Matrix metalloproteinases and their specific tissue inhibitors in menstruation. Reproduction, 123, 621–631.[Abstract]
Fox, H. (1995) Haines and Taylor Obstetrical and Gynaecological Pathology, 4th edn. Churchill Livingstone, NewYork, pp. 810–834.
Giudice, L.C. (1999) Genes associated with embryonic attachment and implantation and the role of progesterone. J. Reprod. Med., 44, 165–171.[ISI][Medline]
Kao, L.C., Tulac, S., Lobo, S., Imani, B., Yang, J.P., Germeyer, A., Osteen, K., Taylor, R.N., Lessey, B.A. and Giudice, L.C. (2002) Global gene profiling in human endometrium during the window of implantation. Endocrinology, 143, 2119–2138.
Kokame, K., Kato, H. and Miyata, T. (1996) Homocysteine-respondent genes in vascular endothelial cells identified by differential display analysis: GRP78/BiP and novel genes. J. Biol. Chem., 271, 29659–29665.
Kurdistani, S.K., Arizti, P., Reimer, C.L., Sugrue, M.M., Aaronson, S.A. and Lee, S.W. (1998) Inhibition of tumor cell growth by RTP/rit42 and its responsiveness to p53 and DNA damage. Cancer Res., 58, 4439–4444.
Lessey, B.A. and Arnold, J.T. (1998) Paracrine signaling in the endometrium: integrins and the establishment of uterine receptivity. J. Reprod. Immunol., 39, 105–116.[CrossRef][ISI][Medline]
Liang, P. and Pardee, A.B. (1998) Differential display. A general protocol. Mol. Biotechnol., 10, 261–267.[ISI][Medline]
Lin, T.M. and Chang, C. (1997) Cloning and characterization of TDD5, an androgen target gene that is differentially repressed by testosterone and dihydrotestosterone. Proc. Natl Acad. Sci. USA, 94, 4988–4993.
Nikas, G. (2000) Endometrial receptivity: changes in cell-surface morphology. Semin. Reprod. Med., 18, 229–235.[CrossRef][ISI][Medline]
Okuda, T. and Kondoh, H. (1999) Identification of new genes ndr2 and ndr3 which are related to Ndr1/RTP/Drg1 but show distinct tissue specificity and response to N-myc. Biochem. Biophys. Res. Commun., 266, 208–215.[CrossRef][ISI][Medline]
Park, H., Adams, M.A., Lachat, P., Bosman, F., Pang, S.C. and Graham, C.H. (2000) Hypoxia induces the expression of a 43-kDa protein (PROXY-1) in normal and malignant cells. Biochem. Biophys. Res. Commun., 276, 321–328.[CrossRef][ISI][Medline]
Piquemal, D., Joulia, D., Balaguer, P., Basset, A., Marti, J. and Commes, T. (1999) Differential expression of the RTP/Drg1/Ndr1 gene product in proliferating and growth arrested cells. Biochim. Biophys Acta, 1450, 364–363.[Medline]
Qu, X., Zhai, Y., Wei, H., Zhang, C., Xing, G., Yu, Y. and He, F. (2002) Characterization and expression of three novel differentiation-related genes belong to the human NDRG gene family. Mol. Cell. Biochem., 229, 35–44.[CrossRef][ISI][Medline]
Rodgers, W.H., Matrisian, L.M., Giudice, L.C., Dsupin, B., Cannon, P., Svitek, C., Gorstein, F. and Osteen, K.G. (1994) Patterns of matrix metalloproteinase expression in cycling endometrium imply differential functions and regulation by steroid hormones. J. Clin. Invest., 94, 946–953.[ISI][Medline]
Salamonsen, L.A., Kovacs, G.T. and Findlay, J.K. (1999) Current concepts of the mechanisms of menstruation. Baillières Best Pract. Res. Clin. Obstet. Gynecol., 13, 161–179.[CrossRef][Medline]
Salnikow, K., Blagosklonny, M.V., Ryan, H., Johnson, R. and Costa, M. (2000a) Carcinogenic nickel induces genes involved with hypoxic stress. Cancer Res., 60, 38–41.
Salnikow, K., Costa, M., Figg, W.D. and Blagosklonny, M.V. (2000b) Hyperinducibility of hypoxia-responsive genes without p53/p21-dependent checkpoint in aggressive prostate cancer. Cancer Res., 60, 5630–5634.
Sambrook, J., Russell, D.W., Irwin, N. and Janssen, K.A. (2001) Molecular Cloning: A Laboratory Manual, 3rd edn, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Semenza, G. (2000) Surviving ischemia: adaptive responses mediated by hypoxia-inducible factor 1. J. Clin. Invest., 106, 809–812.[ISI][Medline]
Semenza, G. (2002) Signal transduction to hypoxia-inducible factor 1. Biochem. Pharmacol., 64, 993–998.[CrossRef][ISI][Medline]
Sharkey, A.M., Day, K., McPherson, A., Malik, S., Licence, D., Smith, S.K. and Charnock-Jones, D.S. (2000) Vascular endothelial growth factor expression in human endometrium is regulated by hypoxia. J. Clin. Endocrinol. Metab., 85, 402–409.
Tabibzadeh, S. (1991) Distinct subsets of stromal cells confined to unique microenvironments in human endometrium throughout the menstrual cycle. Am. J. Reprod. Immunol., 26, 5–10.[Medline]
Tabibzadeh, S., Lessey, B. and Satyaswaroop, P.G. (1998) Temporal and site-specific expression of transforming growth factor-beta4 in human endometrium. Mol. Hum. Reprod., 4, 595–602.
Tseng, L. and Mazella, J. (2002) Endometrial cell specific gene activation during implantation and early pregnancy. Front Biosci., 7, d1566–1574.[ISI][Medline]
VanBelzen, N., Dinjens, W.N.M., Diesveld, M.P.G., Groen, N.A, van der Made, A.C.J., Nozawa, Y., Vliestra, R., Trapman, J. and Bosman, F.T. (1997) A novel gene which is up-regulated during colon epithelial cell differentiation and down-regulated in colorectal neoplasms. Lab. Invest., 77, 85–92.[ISI][Medline]
Wang, H.S. and Chard, T. (1999) IGFs and IGF-binding proteins in the regulation of human ovarian and endometrial function. J. Endocrinol., 161, 1–13.[CrossRef][ISI][Medline]
Zhou, D., Salnikow, K. and Costa, M. (1998) Cap43, a novel gene specifically induced by Ni2+ compounds. Cancer Res., 58, 2182–2189.
Zoroddu, M.A., Kowalik-Jankowska, T., Kozlowski, H., Salnikow, K. and Costa, M. (2001) Ni(II) and Cu(II) binding with a 14-aminoacid sequence of Cap43 protein, TRSRSHTSEGTRSR. J. Inorg. Biochem., 84, 47–54.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. P. Ellen, Q. Ke, P. Zhang, and M. Costa NDRG1, a growth and cancer related gene: regulation of gene expression and function in normal and disease states Carcinogenesis, January 1, 2008; 29(1): 2 - 8. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Stein, E. K. Thomas, B. Herzog, M. D. Westfall, J. V. Rocheleau, R. S. Jackson II, M. Wang, and P. Liang NDRG1 Is Necessary for p53-dependent Apoptosis J. Biol. Chem., November 19, 2004; 279(47): 48930 - 48940. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||