| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Division of Endocrinology,
b Department of Obstetrics and Gynecology,
c Department of Physiology, University of Massachusetts Medical School, Worcester, Massachusetts 01655
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
female reproductive tract, gene regulation, menstrual cycle, progesterone, uterus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Tissues such as the primate endometrium are complex heterogeneous structures whose components are difficult or impossible to study in isolation [5]. In addition, different cells or cell types within a complex target tissue such as the endometrium can respond dissimilarly when exposed to the same hormonal milieu. For example, immunohistochemical techniques have been used to demonstrate that progesterone (P) inhibition of nuclear estrogen receptor is most pronounced in the upper regions of the endometrium while strong positive staining of glandular epithelia in the deep basalis is retained [6, 7–12]. Stromal cells in the upper region of the endometrium are also more rapidly affected by P. In addition, there are striking differences in proliferation between the two major morphological units of the endometrium. During P dominance (midsecretory phase), proliferation is primarily confined to the basalis during both natural menstrual cycles and artificial menstrual cycles in the rhesus monkey [6, 13–15]. These studies and others support the concept that the primate endometrium contains distinct microenvironments that can respond differently to the same hormonal milieu.
LCM can be used to overcome limitations of traditional means of analysis and allow the application of molecular methods of analysis on specific cell types within specific morphological units in the tissue of interest. This approach is particularly appropriate for studies on differential gene expression in the endometrium. Here, we describe our preparation and assessment of suitable material for molecular analysis. We used this approach to identify cell type- and region-specific gene expression in the rhesus endometrium with differential display reverse transcriptase polymerase chain reaction (DDRT-PCR).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mature female rhesus monkeys (Macaca mulatta) were housed in individual cages and ovariectomized at least 2 mo prior to initiation of artificial menstrual cycles. The protocols for creation of artificial cycles have been described in detail elsewhere [6, 16, 17]. Silastic implants containing estrogen (E) or P were placed s.c. in the intrascapular area during ketamine anesthesia (10 mg/kg). Removal of implants was also performed under ketamine anesthesia.
The following protocol for placement or removal of the implants was used to create adequate secretory phases. Basal E levels were maintained with a single 3.0-cm implant throughout the cycle. The E surge was created by sequential insertion of three 3.0-cm E implants on Days 10–12, followed by their removal on Day 13. Each of two P implants 3.0 cm in length were inserted sequentially on Day 13 and Day 16, followed by their sequential removal on Day 23 and Day 25 to simulate the luteal phase. Tissue harvesting was performed prior to insertion or removal of implants. All protocols used in these studies were approved by the Institutional Animal Care and Use Committee.
Tissue Collection and Preparation
Tissue was collected by endometriectomy at hysterotomy as previously described [6, 18]. Endometrial biopsy tissue was oriented in a small aluminum foil cup and frozen immediately in Tissue Tek OCT embedding compound. All samples were subsequently stored at -80°C prior to further processing. Cryostat (-25°C) sections (6 µm) were placed on untreated plain glass slides and immediately fixed in 70% ethanol prior to LCM.
Slides were stained with hematoxylin and eosin using the following sequential solutions (10–30 sec each): Mayer hematoxylin, distilled water, bluing reagent, 70% ethanol, 95% ethanol, and eosin Y. Slides were then dehydrated with two 10-sec washes using 95% ethanol and two 10-sec washes using 100% ethanol. Slides were then placed in xylene twice for at least 5 min and then dried in a vacuum desiccator for 15–20 min prior to LCM.
Glandular epithelial cells (G) or stromal cells (S) from either the basalis (B) or functionalis (F) regions of the endometrium were harvested using a PixCell II LCM System (Arcturus Engineering, Mountain View, CA). The 15- and 30-µm-diameter beams were used. Amplitude and pulse duration ranged from 35 to 50 mW and 3 to 5 msec, respectively. Tissue was collected on TF-100 caps (Arcturus) containing transfer film from two or three sections for each sample. Tissue samples from three different animals were pooled for subsequent analysis.
Synthesis and Amplification of cDNA Populations
RNA extraction was performed using the RNAqueous-4PCR Kit (Ambion, Austin, TX). The caps were placed in microcentrifuge tubes (Brinkman Instruments, Westbury, NY) with RNAqueous lysis/binding solution, which contains guanidinium thiocyanate. After vortexing and centrifuging the tubes, the caps were removed and RNA was isolated using the RNAqueous protocol. The amount of RNA recovered was insufficient to quantitate by conventional means. The DNase I treatment that is part of this protocol effectively removes genomic DNA [19, 20]. In addition, PCR analysis of samples in the absence of reverse transcriptase resulted in the absence of detectable bands (data not shown).
The Superscript Choice System (Life Technologies, Rockville, MD), was used for first strand cDNA synthesis using a mixture of both oligo(dT) and random hexamer primers according to the manufacturer's protocol. Second strand cDNA synthesis and adaptor ligation with EcoRI (NotI) adapters were performed using the same kit.
Complementary DNA populations were purified in Qiaquick spin columns (Qiagen, Valencia, CA) and amplified in 100 µl PCRs containing 0.5 µm LINK-CUA primer, 0.25 mM dNTPs, 1.5 mM MgCl2, 1x buffer, and 2 units Taq polymerase in a thermal cycler (94°C, 1 min; 50°C, 1 min; 72°C, 2 min) for 30 cycles. The LINK-CUA primer (5'-CUACUACUACUAAATTCGCGGCCGCGTCGAC-3') is complementary to the EcoRI adaptor [21]. Following amplification of the cDNA populations, 2% (2 µl or 40 ng) was used as template in PCRs with primers specific for human 18S rRNA (Ambion) and glyceraldehyde-3-phosphate dehydrogenase (G3PDH; Clontech, Palo Alto, CA). The results obtained with 18S rRNA were used to normalize the cDNA populations.
In addition to cDNA smears, the relative size range of our cDNA populations was also estimated using the above primers for both 18S rRNA (product, 324 base pairs [bp]) and G3PDH (product, 983 bp). Our preliminary data suggest that the relative size range of a cDNA smear can influence the subsequent detection of these housekeeping genes.
DDRT-PCR, Cloning, and Sequencing
DDRT-PCR was performed using the RNAimage kit (GenHunter, Nashville, TN). Two nanograms of cDNA was amplified in 20-µl reactions containing 1x buffer, 2.0 µM dNTPs, 20 Ci/mmole alpha-[33P]ATP, 0.2 µM HT11A primer (5'-AAGCTTTTTTTTTTTA-3'), 0.2 µM H-AP1 primer (5'-AAGCTTGATTGCC-3'), and 0.05 units Taq polymerase (Qiagen). Reactions were carried out in a PTC-200 thermal cycler (MJ Research, Watertown, MA) at 94°C for 1 min, 40°C for 2 min, and 72°C for 1 min for 40 cycles and analyzed by denaturing PAGE, omitting the fixing stage. The autoradiogram and gel were aligned by needle punctures, and individual bands were carefully excised from the gel with a razor blade. Gel slices attached to filter paper were eluted by boiling in 100 µl water for 10 min and spun to remove debris, and the supernatant was precipitated with glycogen. DNA fragments thus isolated were reamplified as described above except with 250 µM dNTPs in the absence of radiolabel. Products were directly cloned into the plasmid vector pCR 2.1-TOPO (Invitrogen, Carlsbad, CA) and sequenced (UMass Medical School Nucleic Acid Facility). Homology searches were performed against GenBank entries using BLAST programs (NCBI).
PCR Analysis
Once the sequence of the DDRT-PCR fragments was determined, specific primers were designed and used in PCR reactions with the four cDNA populations (BS, BG, FS, and FG) as template. Approximately 2% of the amplified cDNA populations (2 µl or 40 ng) was analyzed in 100-µl reactions containing 0.5 µM fragment-specific primers, 0.25 mM dNTPs, 1.5 mM MgCl2, 1x buffer, and 2 units Taq polymerase in a thermal cycler (94°C, 1 min; 48–55°C, 1 min; 72°C, 2 min). PCRs were carried out for 24–45 cycles to determine the linear range of amplification for each primer set. Products were analyzed by agarose gel electrophoresis, and comparative evaluation was performed by densitometric analysis (QuantiScan, Biosoft Inc., Ferguson, MO) of photographed gels. Primers for the gene fragments were selected for cloning and sequencing: FS-1, 5'-CTACAGCACCCTGGTCACCT-3' and 5'-CACAGATGCTGCTCCTTCAG-3' (product, 202 bp); BS-1, 5'-TCATATGGGACACTGCTGGT-3' and 5'-TGCCATAGAAATTCGGTCATC-3' (product, 110 bp); FG-1, 5'-TTCACGTTTGTTGCAGAAGC-3' and 5'-CAAGCAAGGGAAGTCTCAGG-3' (product, 187 bp); BG-1, 5'-TCAGAGGGAATGCTTCCAGT-3' and 5'-CCACCAATCCCACAGAAATC-3' (product, 199 bp); F1, 5'-CAGGAGACCATACAGGGTGC-3' and 5'-TTACTTTGGTGGCCTGCTTC-3' (product, 110 bp); and S1, 5'CAAGCCAGAGCCTTGAAAAG-3' and 5'-GGGGTCAGGGTATGGAGTTT-3' (product, 110 bp).
Restriction Enzyme Analysis
PCR fragments were purified on QIAquick spin columns (Qiagen). Two micrograms of DNA were digested with an appropriate enzyme (New England BioLabs, Beverly, MA). Digestions were performed with 1x buffer in a volume of 20 µl at 37°C for 16 h. The resulting fragments were separated on a 2% agarose gel in parallel with unrestricted sample.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Because of the small amount of material harvested by LCM, we adopted an amplification protocol for our cDNA populations that we had used previously [22, 23]. Although cDNA smears could not be observed in most preparations prior to amplification, cDNA smears could be observed in all preparations following amplification (data not shown). We estimate that after only one round of amplification our original cDNA amount has increased approximately 70-fold based on the change in intensity of a visible unamplified 18S rRNA band versus the subsequently amplified population.
We next used these cDNA populations with DDRT-PCR to determine whether we could identify putative cell type- or region-specific gene expression in the rhesus monkey endometrium during an adequate secretory phase. Figure 2 shows portions of a DDRT-PCR gel that contained bands (fragments) that appeared initially to show cell type- or region-specific gene expression. Six fragments were identified for further analysis (Fig. 2, A and B), four fragments that were amplified in a cell type-specific manner (FS-1, BS-1, FG-1, and BG-1) and two fragments that appeared to be region specific (F1 and S1). The corresponding bands were excised, eluted, cloned, and sequenced.
|
Because false positives can occur with the use of DDRT-PCR, we designed PCR primers from our sequence data to confirm the regulation of the six selected fragments. PCR analysis of our four cDNA populations confirmed the localization of three fragments (F1, BG-1, and FS-1). F1 was selectively expressed at high levels only in the functionalis (both FS and FG) of the endometrium (Fig. 3A). BG-1 and FS-1 showed region specificity and were highly expressed in a given cell type (Fig. 3, C and E, respectively).
|
Further confirmation of the authenticity of the three fragments was obtained by restriction enzyme digestion (Fig. 4). Restriction enzymes were chosen to give different fragment lengths based on their nucleotide sequence, and fragments were separated on a 2% agarose gel next to a DNA marker ladder (lane 7). Undigested FS-1 (lane 1, 202 bp) digested with BpmI gave the expected fragments of 128 bp and 74 bp (lane 2). The 74-bp fragment comigrates with a nonspecific band also seen in the undigested sample (lane 1). Digestion of BG-1 (lane 3, 199 bp) with XbaI resulted in the expected fragments of 123 bp and 76 bp (lane 4). Undigested F1 (110 bp, lane 5) was restricted by EcoNI to give the expected fragments of 94 bp and 16 bp (lane 6; 16-bp fragment is too small to be visible).
|
Nucleic acid sequence data identified F1 as human leukotriene B4 (167-nucleotide fragment with 93% identity) [24, 25]. BG-1 showed a 350-nucleotide homology (98%) to bacterial artificial chromosome clone AC007558 for which only the sequence is known, i.e., no gene loci have been assigned to date. FS-1 exhibited 483-nucleotide homology (94%) to hypothetical protein FLJ12436 (NM_024661), based on a 908-amino acid predicted coding region. Although S1 and BS-1 did not display the localization and specificity expected, each showed high nucleic acid sequence homology to known human mRNAs: S1 had 183-nucleotide homology (99%) with human protein phosphatase 3 (formerly 2B), and BS-1 had 238-nucleotide homology (94%) with human histamine N-methyltransferase (HNMT) [26]. FG-1 showed no significant homology with current entries in GenBank.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
One of the objectives of this study was to assess the quality and potential usefulness of endometrial tissue harvested in this manner for subsequent gene expression studies. There are numerous steps in the preparation of suitable genetic material from laser microdissected tissue, any one of which could compromise the quality of a sample [1]. Because of the time and effort required for an analysis of gene expression using this approach, it is useful to have some guide to the relative quality of a sample. Tissue limitations for our laboratory and others may not allow traditional means of analysis, e.g., RNA integrity or relative size of the cDNA population determined by agarose gel electrophoresis.
In an effort to overcome this drawback, we used adapter-specific primer amplification to allow us to visualize a cDNA smear. This approach coupled with the detection of an appropriate housekeeping gene(s) can serve as a useful guide to estimate sample quality for those investigators faced with limited tissue or cells. This approach also provides enough material (cDNA) from a single round of amplification to allow many comparative studies on gene expression to be performed.
A second objective of this study was to use these endometrial cell type- and region-specific cDNA populations to identify differential gene expression patterns. We used DDRT-PCR analysis and selected six fragments that showed a putative cell type-specific or region-specific expression. Although DDRT-PCR is a potentially powerful and important tool, the appearance of false positives can be a drawback [27]. After cloning and sequencing, specific primers for each of the fragments were designed and used to verify their expression patterns. Three fragments were shown to be false positives, underscoring the importance of confirming putative gene regulatory patterns initially identified by DDRT-PCR.
Two of these false positives showed strong homologies with a known human gene. The putative S1 fragment had 95% homology with human protein phosphatase 3 (formerly 2B). In eukaryotes, four major types of protein serine/threonine phosphatases (PP1, PP2A, PP2B, and PP2C) have been identified, and they play a number of important roles in cellular processes, including cell cycle regulation, growth factor signal transduction pathways, glycolysis, and other metabolic processes [28, 29]. One protein phosphatase with very broad substrate specificity, PP2A, is highly upregulated in the rhesus monkey endometrium during the midsecretory phase [30]. The substrate specificity of PP2B is, however, much more restricted than that of PP2A, and comprises proteins that regulate other protein kinases and phosphatases [29]. The putative BS-1 fragment had 93% identity with the human HNMT gene [26]. N-methylation by this enzyme results in the inactivation of histamine [31]. Histamine has been proposed to play an important role in implantation in rodents [32], and the expression of this gene during the primate secretory phase suggests a mechanism whereby the activity of histamine is modulated.
The three fragments identified in LCM samples showing the expected regulatory pattern were F1 (highly expressed in the glands and stroma of the functionalis), BG-1 (highly expressed in the glands of the basalis), and FS-1 (highly expressed in the stroma of the functionalis). Although BG-1 and FS-1 are currently uncharacterized gene fragments, F1 had 93% homology to a known gene, the human leukotriene B4 (LTB4) receptor [24, 25].
LTB4 is one of the most potent chemoattractant mediators, acting mainly on neutrophils but also on related granulocytes, macrophages, and endothelial cells. LTB4 activates inflammatory cells by binding to its cell surface receptors BLTR1 and BLTR2 and has been implicated in a number of inflammatory diseases [24]. Leukotriene levels are elevated in the endometrium of women with primary dysmenorrhrea and endometriosis [33]. Levels of LTB4 also increase in the rat uterus during the peri-implantation phase, implicating a role for this cytokine in the receptivity of the uterus for implantation [34]. The expression of this gene was localized to the endometrial functionalis, the target for blastocyst invasion/implantation. This is the first time a BLTR2 receptor ortholog (F1) has been shown to be expressed in the endometrium.
In this study, cDNA populations were prepared from limited amounts of LCM rhesus monkey endometrium. These cDNA populations were used to identify cell type- and region-specific gene expression patterns in the endometrium. The expression of these genes (known or uncharacterized) is likely to play a role in endometrial function during the primate secretory phase. Future studies will expand on these efforts to identify gene expression patterns that are hormonally and temporally regulated in the different microenvironments of the endometrium in the rhesus monkey.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: William C. Okulicz, Department of OB/GYN and Physiology, UMass Medical School, 55 Lake Ave. N, Worcester, MA 01655. FAX: 508 856 5933; william.okulicz{at}umassmed.edu
Received: 28 February 2002.
First decision: 27 March 2002.
Accepted: 25 April 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G B Godbole, D N Modi, and C P Puri Regulation of homeobox A10 expression in the primate endometrium by progesterone and embryonic stimuli Reproduction, September 1, 2007; 134(3): 513 - 523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. C. Okulicz and C. I. Ace Temporal Regulation of Gene Expression During the Expected Window of Receptivity in the Rhesus Monkey Endometrium Biol Reprod, November 1, 2003; 69(5): 1593 - 1599. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
