Molecular Human Reproduction, Vol. 9, No. 9, 541-549,
September 2003
© 2003 European Society of Human Reproduction and Embryology
Article |
Fibroids display an anti-angiogenic gene expression profile when compared with adjacent myometrium
Submitted on March 17, 2003; resubmitted on April 2, 2003. accepted on April 8, 2003
1 Centre for Women’s Health Research, Monash University Department of Obstetrics & Gynaecology, 246 Clayton Rd., Melbourne, 2 Victorian Bioinformatics Consortium, Monash University, Melbourne and 3 CSIRO, Mathematical and Information Sciences, Melbourne, Australia
4 To whom correspondence should be addressed. e-mail: gareth.weston{at}med.monash.edu.au
 |
ABSTRACT |
|---|
 Top ABSTRACT IntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
The aetiology of uterine fibroids remains unknown, despite causing significant gynaecological morbidity. Fibroids have a reduced microvascular density when compared with adjacent myometrial tissue. The aim of this study was to identify genes with differential expression between fibroid and adjacent normal myometrium, particularly genes with a role in angiogenesis. Total RNA was extracted from fibroid/myometrium pairs from 12 hysterectomy specimens, and used to perform 24 cDNA microarrays. There were 10 500 genes screened on each microarray for differential expression. Analysis of expression data was carried out using multiple t-tests, as well as a novel class prediction algorithm (GeneRaVETM). The differential gene expression of selected genes was confirmed by quantitative ‘real time’ RT–PCR. Selected genes with a role in angiogenesis were further analysed for expression in isolated cell populations of endothelial cells (fibroid and myometrium) and smooth muscle cells (fibroid and myometrium), to see if their expression was confined to particular cell types. Twenty-five genes with differential gene expression between fibroid and myometrium were identified. Insulin-like growth factor-2, endothelin A receptor, connective tissue growth factor (CTGF), cysteine-rich angiogenic inducer 61 (CYR61) and collagen 4
2 (COL4A2) were confirmed by RT–PCR. CTGF and CYR61, both angiogenesis promoters, were reduced in expression relative to myometrium. COL4A2, the precursor for an angiogenesis inhibitor, canstatin, was increased relative to myometrium. These three genes display an anti-angiogenic expression profile in fibroids relative to myometrium. These findings may explain the reduced microvascular density seen in fibroids relative to myometrium. Key words: angiogenesis/COL4A2/CTGF/CYR61/fibroid/microarray
|
Introduction |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Leiomyomata, or uterine fibroids, are the most common solid tumour to afflict women during their reproductive years (Vollenhoven, 1998). Fibroids are the most frequent indication for hysterectomy in Australia and the US (Treloar et al., 1999; Farquhar and Steiner, 2002), making them a significant and costly gynaecological problem.
Many differences have been observed between fibroids and myometrium in an attempt to explain fibroid growth, including different levels of peptide growth factors, chromosomal aberrations and subsequent gene changes, and different levels of sex steroid hormone receptors (reviewed by Anderson, 1998). Despite these observations, the precise aetiology of uterine fibroids has proven difficult to elucidate.
One interesting histopathological feature of fibroids is the difference in their vasculature relative to the adjacent myometrium. Reports on the vasculature of fibroids have been scant in the literature, despite the fact that uterine artery embolization via the femoral artery has been used for >20 years to treat fibroids (Abbott et al., 2002). Fibroids have recently been shown to have a smaller vascular area and microvessel density than adjacent myometrium (Casey et al., 2000; Poncelet et al., 2002). This is intriguing, as solid tumours have long been postulated to require increased microvessel density in order to support a growth in tumour mass (Folkman, 1998). Fibroids may behave differently because they are benign tumours. Nonetheless, although a reduced vascular density has been observed in fibroids compared with adjacent myometrium, the factor or factors responsible for this remain unknown.
DNA microarrays are one of an increasing number of tools available for screening for gene expression changes in up to several thousand genes at once (Lander, 1999). DNA microarrays have advantages over older techniques such as differential display (Pambuccian et al., 2002; Wu et al., 2002) as they provide quantitative as well as qualitative data. One of the major hurdles in the use of microarrays has been the complexity of the statistical analysis of the large datasets generated (Smyth et al., 2002). Several software programs to assist in the analysis of microarray data have become commercially available, at prices increasingly within the reach of the academic sector.
Both differential display (Pambuccian et al., 2002; Wu et al., 2002) and microarrays (Tsibris et al., 2002) have been used successfully to examine gene expression differences between fibroid and myometrium. The aim of our study was to examine gene expression changes between fibroids and adjacent myometrium from hysterectomy specimens, using cDNA microarrays. Our hypothesis was that the reduced vasculature in the fibroid, when compared with the adjacent myometrium, would be reflected in differences in transcriptional activity in genes with a role in angiogenesis. Therefore, we wanted to examine changes in angiogenesis-related genes, in particular, between fibroid and adjacent myometrium.
|
Materials and methods |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Sample collection
Ethical approval for the study was obtained from Southern Health Human Research and Ethics Committee B, and informed consent was obtained from all patients.
Samples of fibroid and adjacent normal myometrium were excised from a total of 28 uteri (mean age 44.3 years, range 33–54) at hysterectomy, and immediately frozen on dry ice. Fibroids sampled were >3 cm in diameter and from the inner third of the myometrium. Tissue sampling from the fibroids was random, without systematically favouring the inner or outer portion of the fibroids. The myometrium sampled was from the inner third of the myometrium. Twelve of the paired samples were used for the microarrays, 18 for the quantitative RT–PCR, and six for tissue culture, with some degree of overlap between the three groups where there was sufficient tissue collected. All hysterectomy specimens were from pre-menopausal women undergoing operations for fibroids. Tissue samples were only taken from women who had not received exogenous hormones for the previous 3 months.
RNA extraction
Thin shavings of the frozen specimens were homogenized in Trizol reagent (Invitrogen Life Technologies, Australia) according to the manufacturer’s protocol. Total RNA was precipitated from the aqueous phase of the Trizol preparation with an equal volume of 100% ethanol, and then run through a Qiagen RNeasy column (Qiagen, Germany) according to the manufacturer’s protocol for clean-up of the RNA. After resuspension in RNase-free water, RNA was further treated with an overnight ethanol precipitation as previously described (Weston et al., 2002). The final concentration of total RNA was 
5 µg/µl, required for the cDNA labelling process of the microarray experiments.
cDNA microarrays
The total RNA samples were used to prepare fluorescent-labelled cDNA for probing with glass microarray slides spotted with 10 500 human cDNA sequences (prepared in-house at the Peter MacCallum Cancer Institute Microarray Facility, Melbourne, Australia; the complete list of genes on the array is available at www.CCGPM.org). 60–90 µg of total RNA was used for each microarray, with an equal amount of reference total RNA used as the control. A direct-labelling procedure was used for creating fluorescent-labelled cDNA. The labelling and hybridization of the slides has been described previously (Weston et al., 2002).
A total of 24 microarrays were performed. In the first 12 arrays, individual fibroid RNA specimens were labelled with the experimental fluorescent dye Cy5 (cat. no. PA55201; Amersham Bioscience, UK). These were run against an aliquot of reference myometrial RNA labelled with Cy3 dye (cat. no. PA53201; Amersham). The reference myometrial RNA was formed from groups of six myometrial samples. In the second group of 12 arrays, individual myometrial RNA specimens were labelled with the experimental dye, and run against a reference pool of fibroid RNA specimens (again, two groups of six fibroid samples were used to form the reference RNA). This experimental design incorporated a dye-swap (Sterrenburg et al., 2002), to ensure that genes identified were up- or down-regulated regardless of which of the two dyes was used to label the RNA. The design also provided quantitative gene expression data on each of the 24 separate biological samples (12 pairs of fibroid and myometrium samples).
Slides were scanned using a dual UV-laser GSI Luminomics scanner (Packard Bioscience, USA) and fluorescent intensity analysed using the program Quantarray (Packard Bioscience). Images were formed by superimposing the Cy3 and Cy5 images for each slide using the Scanalyze software (M.Eisen, Brown Lab, Stanford, USA). Two of the arrays with myometrial RNA labelled with the Cy5 dye were excluded, due to poor image quality.
Normalization of data from the remaining 22 arrays was carried out using Genespring (Silicon Genetics, USA). Raw Quantarray data were log-transformed, and subsequent intensity-dependent normalization achieved by use of a Lowess fit. The Lowess fit adjusts for differential labelling efficiency of the two fluorescent dyes. Normalized values <0 were set to 0, and data for individual genes were rejected if the background-subtracted control value was <10.0. The Cy5 (experiment) fluorescent intensity was divided by the Cy3 (reference) intensity for each gene to give a gene expression level relative to reference for each of the 10 500 genes. Statistical analysis was carried out as described below.
Quantitative RT–PCR
One µg total RNA from each sample to be analysed by RT–PCR was DNase-treated using the DNA-free kit (Ambion, USA) according to the manufacturer’s protocol, then reverse-transcribed at 42°C for 1 h in a total volume of 20 µl (2.5 µl anchor primer oligodT15; Roche, Australia), 2 µl 10 mmol/l dNTP (Roche), 4 µl 5xRT buffer (Roche), 0.5 µl RNasin (Promega, USA), 2 µl dithiothreitol (Promega), 0.2 µl AMV reverse transcriptase (Roche), total RNA and sterile H2O to make up the volume). Two µl of a 1:10 dilution of the RT product was used as a template for the subsequent PCR.
Roche LightCycler was used to perform quantitative PCR. For a detailed overview of the principles of the technique see Drummond et al. (2000). Reagents for the LightCycler PCR were obtained from Roche. The primer sequences used are shown in Table I. Primer concentrations were 0.5 µmol/l. Each set of primers was optimized for magnesium concentration, annealing temperature and extension time as shown in Table II. Relative levels of mRNA expression for each of the genes tested were determined by measurement against a specific cDNA standard. All results were normalized against â-actin expression to correct for differences in concentration of the starting template.
|
|
Tissue culture and RNA samples of separate cell populations from fibroid and myometrium
Total RNA was extracted from myometrial and fibroid microvascular endothelial cells (MEC) and smooth muscle cells (SMC) cultured from a further set of six myometrium–fibroid tissue pairs as previously described (Gargett et al., 2002). These cultures were from passages 1–3. They were also >98% pure for each cell type as determined by flow cytometric analysis for CD31 and alpha smooth muscle actin for MEC and SMC respectively as detailed previously (Gargett et al., 2002).
Statistical analysis
Genespring was used to identify genes with statistically significant differential expression by performing t-tests with multiple test correction using the Benjamini–Hochberg false discovery rate (Benjamini and Hochberg, 1995). Significance tests were performed on Lowess-normalized data (see above). Genes with P < 0.05 after multiple test correction were considered significant if they were differentially expressed by a factor of 
1.5 in 80% of the replicates. The 1.5-fold threshold was chosen arbitrarily to focus on genes likely to have biologically significant as well as statistically significant gene expression changes.
The microarray data were examined via a second technique, components from a suite of statistical algorithms developed by the Bioinformatics Technology Group (CSIRO, Mathematical and Information Sciences), called GeneRaVETM, to identify differentially expressed genes. GeneRaVETM may be characterized as Bayesian stochastic variable selection. The validation strategy of GeneRaVETM uses permutation distributions and cross-validation. This strategy has been validated on public domain microarray data sets such as the B-cell lymphoma data of Alizadeh et al. (2000) and the prostate cancer data of Luo et al. (2001). In one of its formulations, GeneRaVE is able to obtain parsimonious sets of solution genes as sample classifiers. The solution sets may be validated by the magnitude of the proportion of misclassifications generated by the set, by subsequent permutation tests, and by solution stability tests such as leave-out-one-sample-at-a-time procedures.
For the data of the current paper, GeneRaVETM was first run using every microarray with the full complement of probe genes. The solution genes identified by GeneRaVETM in this step were placed in the initial gene solution set. Next, in a leave-out-one-sample-at-time procedure, GeneRaVETM was run using all arrays except for the first array, then using all arrays except the second and so on, finishing with running GeneRaVETM on all arrays except for the last array. At each run, a solution set of genes was obtained. This procedure provided a test of solution stability and had the potential to identify other solution genes. GeneRaVETM was then reapplied to the microarray data with the microarray data of all the solution set genes eliminated. Although repeated applications of GeneRaVETM in this fashion has the potential to generate many solution genes for group classification, it was decided to repeat this procedure until 25–30 solution genes were obtained. This procedure generated a ‘league table’ for ranking all the solution genes obtained by successive reapplications of GeneRaVETM.
Following the statistical analyses outlined, a list of genes (Table III) on the microarray with a known role in angiogenesis was created, following a literature search. The list of genes was examined to see if their expression varied between fibroid and adjacent myometrium.
|
Quantitative gene expression data from the LightCycler were analysed by either paired t-tests (fibroid versus myometrium pairs), or by one-way analysis of variance (matched sets of four cell populations).
|
Results |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
cDNA microarrays
A total of 25 genes was identified as being differentially expressed between fibroid and myometrium by at least one of the two statistical analysis methods (Table IV). Of the genes identified, 11 displayed a lower expression in fibroids than in the adjacent myometrium, while 14 showed higher expression than the adjacent myometrium. Of the 25 genes identified, 10 (40%) had previously been reported (Table IV).
|
Genespring identified 11 genes, five of which (45.5%) had been previously reported, while GeneRaVETM identified 23 genes, with nine (39%) previously reported. The Genespring list had only two genes not identified by the GeneRaVETM analysis, while the GeneRaVETM list had 14 extra genes.
In a further analysis, we divided fibroid and myometrial specimens based on the stage of the menstrual cycle (proliferative or secretory phase). We found no significant differences in gene expression between proliferative and secretory phase fibroid or myometrium. There were only half as many samples in each of the groups when divided on the basis of the phase of the menstrual cycle, reducing the power of this further analysis to detect gene expression changes.
Angiogenesis-associated genes by microarray
Of the genes identified from the analysis, three have been implicated in angiogenesis. CYR61 (cysteine-rich angiogenic inducer 61) (Babic et al., 1998) and CTGF (connective tissue growth factor) (Babic et al., 1999) are both angiogenesis promoters which were down-regulated in fibroids compared with myometrium in our arrays. Collagen 4á2 (COL4A2) is a precursor for the angiogenesis inhibitor, canstatin (Kamphaus et al., 2000), and was up-regulated in fibroid compared with adjacent myometrium.
We went on to examine the expression of angiogenesis promoters and angiogenesis inhibitors present in the group of 10 500 genes present on our microarrays (Table III) in pairs of fibroid and adjacent myometrium. We compared their expression profile to that of the three angiogenesis modulators above. There did not appear to be any significant differences in the expression profile of any of the angiogenesis promoters (see Figure 1A,B) or inhibitors (see Figure 1C,D) between fibroid and myometrium, other than the three genes already identified: CYR61, CTGF and COL4A2.
|
The expression pattern of the three angiogenesis-related genes (CYR61, CTGF and COL4A2) are represented graphically in three dimensions in Figure 2. This triad of angiogenesis-related genes displays an anti-angiogenic profile in fibroids when compared with adjacent myometrium (reduced expression of CYR61 and CTGF, and increased expression of COL4A2).
|
Quantitative RT–PCR for confirmation of selected genes identified by microarray
Confirmation of selected genes identified in our analysis was performed using ‘real time’ quantitative RT–PCR. Insulin-like growth factor-2 (IGF-2) and endothelin A receptor (ET-A R) were assessed (Figure 3) in addition to CYR61, CTGF and COL4A2 (Figure 4). For all five of the genes, the gene expression microarray result was confirmed. RT–PCR performed on 18 pairs (fibroid and myometrium from the same uterus) produced statistically significant gene expression differences by paired t-test. The specific levels of significance for each of the genes tested were IGF-2 (P = 0.0027), ET-A R (P = 0.0287), CYR61 (P = 0.0089), CTGF (P = 0.0307), and COL4A2 (P = 0.002). The fold-change in gene expression was in all cases underestimated by the microarray.
|
P < 0.05; *P < 0.01). Graphs on the left are line graphs linking paired specimens, those on the right are box-and-whisker plots.
|
There was considerable variation amongst the different fibroids in gene expression relative to their adjacent myometrium as measured by RT–PCR. CYR61 mRNA was down-regulated by
1.5-fold in fibroids compared with adjacent myometrium in 13/18 pairs, but in 5/18 pairs CYR61 mRNA levels in fibroids were equivalent to myometrium expression (<1.5-fold different in expression). In no fibroid–myometrium pairs was CYR61 expression up-regulated by 1.5-fold in the fibroid tissue. CTGF showed more variation than CYR61 in differential gene expression between different pairs. In 11/18 pairs, CTGF expression was down-regulated 1.5-fold in fibroids, but in 4/18 pairs, there was actually up-regulation 1.5-fold in fibroids, and in 3/18 CTGF expression was equivalent (<1.5-fold difference). COL4A2 also showed variation on an individual basis, despite the statistically significant overall trend to increased expression in fibroids. Eleven of the 18 pairs showed up-regulation in fibroids by 1.5-fold, but 6/18 showed equivalent expression (<1.5-fold difference between fibroid and myometrium expression) and 1/18 had down-regulation in fibroid, by >1.5-fold. Examining the gene expression changes of all five genes according to the phase of the menstrual cycle of the samples (proliferative or secretory phase), again we could find no statistically significant gene expression changes. Six of the 18 samples were proliferative phase, six were secretory, and on six samples we had no information on cycle phase. Only CYR61 showed any trend to alteration in gene expression across the cycle. There was a trend to increased expression in the myometrium in the proliferative phase of the cycle (mean in proliferative phase 1.88 relative units, mean in secretory phase 0.69 relative units, with P = 0.084 by t-test). This trend was not evident in the fibroid (mean in proliferative phase 0.51 relative units, mean in secretory phase 0.36 relative units, P = 0.29 by t-test)
CYR61, CTGF and COL4A2 expression in separate cell populations
Total RNA was obtained from cultured myometrial microvascular endothelial cells (MMEC), fibroid myometrial microvascular endothelial cells (FMEC), myometrial smooth muscle cells (MSMC) and fibroid smooth muscle cells (FSMC) from another six hysterectomy specimens. The separate cell populations were assessed for their mRNA levels of CYR61, CTGF and COL4A2 by quantitative RT–PCR (Figure 5), to determine the cell types expressing these three angiogenesis modulators. All four cell types expressed each of the three angiogenesis modulators in vitro, and the differences in levels of mRNA expression for all three genes was not significant. However, the highest expression of the angiogenesis promoters CYR61 and CTGF was in MMEC.
|
|
Discussion |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
This is the largest set of gene expression microarrays of fibroids to be reported in the literature to date, and identifies a total of 25 differentially expressed genes between fibroid and myometrium—15 of them not previously reported. It is the first report in the microarray literature to employ a novel new statistical algorithm, GeneRaVETM, for identifying genes with different expression patterns in two populations. This work identifies for the first time a set of three angiogenesis-related genes—CYR61, CTGF and COL4A2—with an anti-angiogenic gene expression profile in fibroids when compared with adjacent normal myometrium.
The only microarray study published to date (Tsibris et al. 2002) used nine paired hysterectomy specimens in a set of nine arrays, and identified 145 genes on the basis of fold-changes in gene expression between fibroids and adjacent myometrium. The smaller number of genes identified in our study was due in part to the stringency of the analysis methods employed, as well as to the inherently high variability of fibroids. Using statistical tests for significance with multiple test correction reduces the number of genes identified, but increases the true positive rate of the genes that are identified (Smyth et al., 2002). Many array studies in the reproductive literature currently use fold-changes in gene expression as a threshold when identifying genes (Tsibris et al., 2002; Weston et al., 2002), often identifying very large gene sets. The advantage from an increase in number of genes is likely to be offset by an increase in the false positive rate, making subsequent confirmation by other experimental methods essential.
The CCN gene family was named after the first three known members: CTGF, NOVH (nephroblastoma overexpressed gene) and CYR61 (Bork et al., 1993). Members of this gene family encode secreted proteins which modulate the actions of growth factors, and affect cell migration and extracellular matrix production (Brigstock, 1999). Both CTGF and CYR61 are promoters of angiogenesis (Babic et al., 1998; 1999). CYR61 has been shown to induce neovascularization in vivo, and chemotaxis of microvascular endothelial cells in vitro (Babic et al., 1998). CTGF has been shown to have multiple actions on human microvascular endothelial cells in vitro which stimulate angiogenesis, including endothelial cell proliferation, adhesion, migration, as well as reducing apoptosis and promoting endothelial cell survival (Babic et al., 1999).
This is the first study to show altered expression of the angiogenic inducer CTGF in fibroids. We identified CTGF as a differentially expressed gene with reduced expression in fibroid compared with myometrium. This was confirmed on subsequent quantitative PCR. The only other study to examine CTGF in human fibroids found CTGF protein confined to the endothelial cells, but did not attempt to quantify differences in expression (Uzumcu et al., 2000).
CYR61 has been reported in one study to be reduced in fibroid compared with adjacent myometrium (Sampath et al., 2001). Our work confirms these results. However, we extended our analysis to examine gene expression in separate cell populations, and found that microvascular endothelial cells as well as smooth muscle cells express CYR61 mRNA. This is in contrast to the study by Sampath et al., which did not detect CYR61 mRNA in microvascular endothelial cells by in-situ hybridization. Whether the increased CTGF and CYR61 gene expression in myometrium is from the smooth muscle cell or endothelial cell population is uncertain, as our quantitative RT–PCR on separated cell types in culture showed no significant differences (Figure 5). If the endothelial cells are the source of the difference in expression, it is possible that the differences are due to the influence of the different microenvironments of the microvascular endothelial cells within the fibroid (surrounded by an extensive extracellular matrix) compared with the myometrium.
COL4A2 gene encodes collagen 4á2, a basement membrane component associated ubiquitously with blood vessels. Canstatin, a 24 kDa fragment of collagen 4á2, has been shown to be an angiogenesis inhibitor (Kamphaus et al., 2000). In this study, we have shown for the first time that the COL4A2 gene is overexpressed in fibroid compared with myometrium. This may indicate that there is greater production and breakdown of collagen 4á2 in fibroids, generating more of the angiogenesis inhibitor canstatin in fibroids than in myometrium. At present, however, it is unknown if canstatin is produced in fibroids.
Our data show reduced expression of two recently identified angiogenesis promoters, and overexpression of the precursor of an angiogenesis inhibitor. The expression profiles of these three genes display an anti-angiogenic profile which may explain in part the reduced microvessel density observed in fibroids compared with myometrium (Casey et al., 2000).
Although this study has concentrated on CTGF and CYR61 with regard to their ability to promote angiogenesis, these two factors are known to have multiple biological actions (Brigstock, 1999). Through their actions on cell cycle regulation, cell adhesion and migration and production of extracellular matrix, they play a role in the growth of a variety of tumours, including breast cancer (Brigstock, 1999). In inflammation and wound repair, they are also postulated to play an important role. Paradoxically in a tumour such as a fibroid, characterized by increased extracellular matrix, CTGF is underexpressed in fibroid compared with adjacent myometrium. CTGF is overexpressed in a large number of pathological conditions characterized by fibrosis, including skin conditions such as scleroderma and keloid (Igarashi et al., 1996), and diseases of the lung such as pulmonary fibrosis (Phan and Kunkel, 1992). The underexpression of CTGF in fibroids despite the increased extracellular matrix highlights the current gaps in our understanding of the CCN family of proteins and the precise role they play in fibroid growth.
IGF-2 and ET-A R were two other genes identified in our arrays and confirmed by quantitative PCR. Overexpression of IGF-2 in fibroids compared with myometrium has been previously reported (Vollenhoven et al., 1993). Our study confirmed this. The overexpression of endothelin-A receptor in fibroids relative to myometrium has also been previously reported (Pekonen et al., 1994).
In this study, we utilized two statistical algorithms for identifying genes. One of those algorithms, Genespring, identified 11 genes by t-tests with multiple test corrections. Genespring is a powerful tool for visualizing gene expression data from DNA microarrays, and has been used to analyse microarray data from human endometrium (Kao et al., 2002). We also used a novel class prediction algorithm called GeneRaVETM, which identifies parsimonious gene sets that define different groups of data. It should be noted that the two statistical methods work by different statistical approaches. However, despite this, all but two of the genes from the Genespring list also appeared in the GeneRaVETM list. Five of the genes we identified from the gene lists were subsequently all confirmed by quantitative RT–PCR.
Fibroids have a reduced microvessel density compared with adjacent myometrium. This study identifies three genes with altered expression in fibroids which may explain this reduction in microvascular density. How the vessel density relates to fibroid tumour growth and progression is uncertain. It may be that the secreted proteins with dysregulated expression in fibroids may affect fibroid tumour growth via modulation of the stimulus of other growth factors such as bFGF (Brigstock et al., 1999). More work is required on members of the CCN family in fibroids to examine their potential role in fibroid growth, and in particular, in angiogenesis within fibroids.
|
Acknowledgements |
|---|
We would like to acknowledge the technical assistance of the staff of the microarray facility at the Peter MacCallum Cancer Institute. Nancy Taylor and Nikki Sam collected the fibroid and myometrial tissue used in this study. We acknowledge the assistance of Dr Mark Lawrence, Professor David Healy and Dr Elizabeth Farrell in obtaining tissue samples. P.A.W.Rogers’ salary is paid by the National Health & Medical Research Council of Australia under Fellowship grant 143805. G.Weston is supported by National Health & Medical Research Council of Australia clinical postgraduate scholarship 008202.
|
REFERENCES |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Abbott, J.A. and Garry, R. (2002) The surgical management of menorrhagia. Hum. Reprod. Update, 8, 68–78.
Adler, S.G., Kang, S., Feld, S., Cha, D.R., Barba, L., Striker, L., Striker, G., Riser, B.L., LaPage, J. and Nast, C.C. (2001) Glomerular mRNAs in human type 1 diabetes: Biochemical evidence for microalbuminuria as a manifestation of diabetic nephropathy. Kidney Int., 60, 2330–2336.[CrossRef][ISI][Medline]
Alizadeh, A.A., Eisen, M.B., Davis, R.E., Ma, C., Lossos, I.S., Rosenwald, A., Boldrick, J.C., Sabet, H., Tran, T., Yu, X. et al. (2000) Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature, 403, 503–511.[CrossRef][Medline]
Anderson, J. (1998) Factors in fibroid growth. Baillière’s Clin. Obstet. Gynecol., 12, 225–243.[ISI][Medline]
Babic, A.M., Kireeva, M.L., Kolesnikova, T.V. and Lau, L.F. (1998) CYR61, a product of a growth factor-inducible immediate early gene, promotes angiogenesis and tumor growth. Proc. Natl Acad. Sci. USA, 95, 6355–6360.
Babic, A.M., Chen, C.C. and Lau, L.F. (1999) Fisp12/mouse connective tissue growth factor mediates endothelial cell adhesion and migration through integrin alphavbeta3, promotes endothelial cell survival, and induces angiogenesis in vivo. Mol. Cell. Biol., 19, 2958–2966.
Bauer, M., Wilkens, H., Langer, F., Schneider, S.O., Lausberg, H. and Scafers, H. (2002) Selective upregulation of endothelin B receptor gene expression in severe pulmonary hypertension. Circulation, 105, 1034–1036.
Benjamini, Y. and Hochberg, Y. (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B, 57, 289–300.
Bork, P. (1993) The modular architecture of a new family of growth regulators related to connective tissue growth factor. FEBS Lett., 327, 125–130.[CrossRef][ISI][Medline]
Brigstock, D.R. (1999) The connective tissue growth factor/cysteine-rich 61/nephroblastoma overexpressed (CCN) family. Endocr. Rev., 20, 189–206.
Casey, R., Rogers, P.A. and Vollenhoven, B.J. (2000) An immunohistochemical analysis of fibroid vasculature. Hum. Reprod., 15, 1469–1475.
Drummond, A.E., Dyson, M., Thean, E., Groome, N.P., Robertson, D.W. and Findlay, J.K. (2000) Temporal and hormonal regulation of inhibin protein and subunit mRNA expression by post-natal and immature rat ovaries. J. Endocrinol., 166, 339–354.[Abstract]
Farquhar, C.M. and Steiner, C.A. (2002) Hysterectomy rates in the United States 1990–1997. Obstet. Gynecol., 99, 229–234.
Folkman, J. (1998) Is tissue mass regulated by vascular endothelial cells? Prostate as the first evidence. Endocrinology, 139, 441–442.
Gargett, C.E., Bucak, K., Zaitseva, M., Chu, S., Taylor, N., Fuller, P.J. and Rogers, P.A. (2002) Estrogen receptor-alpha and -beta expression in microvascular endothelial cells and smooth muscle cells of myometrium and leiomyoma. Mol. Hum. Reprod., 8, 770–775.
Igarashi, A., Nashiro, K., Kikuchi, K., Sato, S., Ihn, H., Fujimoto, M., Grotendorst, G.R. and Takehara, K. (1996) Connective tissue growth factor gene expression in tissue sections from localized scleroderma, keloid and other fibrotic skin disorders. J. Invest. Dermatol., 106, 729–733.[CrossRef][ISI][Medline]
Kamphaus, G.D., Colorado, P.C., Panka, D.J., Hopfer, H., Ramchandran, R., Torre, A., Maeshima, Y., Mier, J.W., Sukhatme, V.P. and Kalluri, R. (2000) Canstatin, a novel matrix-derived inhibitor of angiogenesis and tumor growth. J. Biol. Chem., 275, 1209–1215.
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.
Lander, E.S. (1999) Array of hope. Nature Genet., 21, 3–4.[CrossRef][ISI][Medline]
Lighten, A.D., Hardy, K., Winston, R.M.L. and Moore, G.E. (1997) Expression of mRNA for the insulin-like growth factors and their receptors in human preimplantation Embryos. Mol. Reprod. Dev., 47, 134–139.[CrossRef][ISI][Medline]
Lessl, M., Klotzbuecher, M., Schoen, S., Reles, A., Stockemann, K. and Fuhrmann, U. (1997) Comparative messenger ribonucleic acid analysis of immediate early genes and sex steroid receptors in human leiomyoma and healthy myometrium. J. Clin. Endocrinol. Metab., 82, 2596–600.
Luo, J., Duggan, D.J., Chen, Y., Sauvageot, J., Ewing, C.M., Bittner, M., Trent, J.M. and Isaacs, W.B. (2001) Human prostate cancer and benign prostatic hyperplasia: molecular dissection by gene expression profiling. Cancer Res., 61, 4683–4688.
Pambuccian, C.A., Oprea, G.M. and Lakatua, D.J. (2002) Reduced expression of early growth response-1 gene in leiomyoma as identified by mRNA differential display. Gynecol. Oncol., 84, 431–436.[CrossRef][ISI][Medline]
Pekonen, F., Nyman, T. and Rutanen, E.M. (1994) Differential expression of mRNAs for endothelin-related proteins in human endometrium, myometrium and leiomyoma. Mol. Cell. Endocrinol., 103, 165–170.[CrossRef][ISI][Medline]
Phan, S.H. and Kunkel, S.L. (1992) Lung cytokine expression in bleomycin-induced pulmonary fibrosis. Exp. Lung Res., 18, 29–43.[ISI][Medline]
Poncelet, C., Madelenat, P., Feldmann, G., Walker, F. and Darai, E. (2002) Expression of von Willebrand’s factor, CD34, CD31, and vascular endothelial growth factor in uterine leiomyomas. Fertil. Steril., 78, 581–586.[CrossRef][ISI][Medline]
Sampath, D., Zhu, Y., Winneker, R.C. and Zhang, Z. (2001) Aberrant expression of Cyr61, a member of the CCN (CTGF/Cyr61/Cef10/NOVH) family, and dysregulation by 17 beta-estradiol and basic fibroblast growth factor in human uterine leiomyomas. J. Clin. Endocrinol. Metab., 86, 1707–1715.
Sasaki, M., Dharia, A., Oh, B.R., Tanaka, Y., Fujimoto, S. and Dahiya, R. (2001) Progesterone receptor B gene inactivation and CpG hypermethylation in human uterine endometrial cancer. Cancer Res., 61, 97–102.
Smyth, G.K., Yang, Y.H. and Speed, T. (2002) Statistical issues in cDNA microarray data analysis. In Brownstein, M.J. and Khodursky, A.B. (eds), Functional Genomics: Methods and Protocols, Methods in Molecular Biology series, Humana Press, Totowa, NJ.
Sterrenburg, E., Turk, R., Boer, J.M., Van Ommen, G.B. and Den Dunnen, J.T. (2002) A common reference for cDNA microarray hybridizations. Nucleic Acids Res., 30, E116-6.
Treloar, S.A., Do, K.A., O’Connor, V.M., O’Connor, D.T., Yeo, M.A. and Martin, N.G. (1999) Predictors of hysterectomy: an Australian study. Am. J. Obstet. Gynecol., 180, 945–954.[CrossRef][ISI][Medline]
Tsibris, J.C., Segars, J., Coppola, D., Mane, S., Wilbanks, G.D., O’Brien, W.F. and Spellacy, W.N. (2002) Insights from gene arrays on the development and growth regulation of uterine leiomyomata. Fertil. Steril., 78, 114–121.[CrossRef][ISI][Medline]
Uzumcu, M., Al Homsi, M.F., Ball, D.K., Coskun, S., Jarousi, K., Hollanders, J.M.G. and Brigstock, D.R. (2000) Localization of connective tissue growth factor in human uterine tissues. Mol. Hum. Reprod., 6, 1093–1098.
Vollenhoven, B. (1998) Introduction: the epidemiology of uterine leiomyomas. Baillières Clin. Obstet. Gynecol., 12, 169–176.[CrossRef][ISI][Medline]
Vollenhoven, B.J., Herington, A.C. and Healy, D.L. (1993) Messenger ribonucleic acid expression of the insulin-like growth factors and their binding proteins in uterine fibroids and myometrium. J. Clin. Endocrinol. Metab., 76, 1106–1110.[Abstract]
Weston, G., Haviv, I. and Rogers, P.A.W. (2002) Microarray analysis of VEGF-responsive genes in myometrial endothelial cells. Mol. Hum. Reprod., 8, 855–863.
Wu, X., Blanck, A., Norstedt, G., Sahlin, L. and Flores-Morales, A. (2002) Identification of genes with higher expression in human uterine leiomyomas than in the corresponding myometrium. Mol. Hum. Reprod., 8, 246–254.
Xie, D., Nakachi, K., Wang, H., Elashoff, R. and Koeffler, H.P. (2001) Elevated levels of connective tissue growth factor, WISP-1 and CYR61 in primary breast cancers associated with more advanced features. Cancer Res., 61, 8917–8923.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. A. Behera, L. Feng, B. Yonish, W. Catherino, S.-H. Jung, and P. C. Leppert Thrombospondin-1 and Thrombospondin-2 mRNA and TSP-1 and TSP-2 Protein Expression in Uterine Fibroids and Correlation to the Genes COL1A1 and COL3A1 and to the Collagen Cross-link Hydroxyproline Reproductive Sciences, December 1, 2007; 14(8_suppl): 63 - 76. [Abstract] [PDF]
|
|
|
|
 |
|
 H. Ishikawa, M. Shozu, M. Okada, M. Inukai, B. Zhang, K. Kato, T. Kasai, and M. Inoue Early growth response gene-1 plays a pivotal role in down-regulation of a cohort of genes in uterine leiomyoma J. Mol. Endocrinol., November 1, 2007; 39(5): 333 - 341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Zaitseva, B. J. Vollenhoven, and P. A.W. Rogers Retinoic acid pathway genes show significantly altered expression in uterine fibroids when compared with normal myometrium Mol. Hum. Reprod., August 1, 2007; 13(8): 577 - 585. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Aitken, A. Khaund, S.A. Hamid, D. Millan, and S. Campbell The normal human myometrium has a vascular spatial gradient absent in small fibroids Hum. Reprod., October 1, 2006; 21(10): 2669 - 2678. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. De Falco, S. Staibano, F. P. D'Armiento, M. Mascolo, G. Salvatore, A. Busiello, I. F. Carbone, F. Pollio, and A. Di Lieto Preoperative Treatment of Uterine Leiomyomas: Clinical Findings and Expression of Transforming Growth Factor-{beta}3 and Connective Tissue Growth Factor Reproductive Sciences, May 1, 2006; 13(4): 297 - 303. [Abstract] [PDF]
|
|
|
|
|
|
X. Luo, L. Ding, and N. Chegini CCNs, fibulin-1C and S100A4 expression in leiomyoma and myometrium: inverse association with TGF-{beta} and regulation by TGF-{beta} in leiomyoma and myometrial smooth muscle cells Mol. Hum. Reprod., April 1, 2006; 12(4): 245 - 256. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Zaitseva, B. J. Vollenhoven, and P. A.W. Rogers In vitro culture significantly alters gene expression profiles and reduces differences between myometrial and fibroid smooth muscle cells Mol. Hum. Reprod., March 1, 2006; 12(3): 187 - 207. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-J. Wei, L. Chiriboga, A. A. Arslan, J. Melamed, H. Yee, and K. Mittal Ethnic differences in expression of the dysregulated proteins in uterine leiomyomata Hum. Reprod., January 1, 2006; 21(1): 57 - 67. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hussein-Fikret, P. J. Fuller, and C. E. Gargett Expression of Steroid Receptor Coactivators in Cultured Cells From Paired Myometrial and Fibroid Tissues Reproductive Sciences, September 1, 2005; 12(6): 445 - 451. [Abstract] [PDF]
|
|
|
|
|
|
E. Levens, X. Luo, L. Ding, R. S. Williams, and N. Chegini Fibromodulin is expressed in leiomyoma and myometrium and regulated by gonadotropin-releasing hormone analogue therapy and TGF-{beta} through Smad and MAPK-mediated signalling Mol. Hum. Reprod., July 1, 2005; 11(7): 489 - 494. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.D. Swartz, C.A. Afshari, L. Yu, K.E. Hall, and D. Dixon Estrogen-induced changes in IGF-I, Myb family and MAP kinase pathway genes in human uterine leiomyoma and normal uterine smooth muscle cell lines Mol. Hum. Reprod., June 1, 2005; 11(6): 441 - 450. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. A. Arslan, L. I. Gold, K. Mittal, T.-C. Suen, I. Belitskaya-Levy, M.-S. Tang, and P. Toniolo Gene expression studies provide clues to the pathogenesis of uterine leiomyoma: new evidence and a systematic review Hum. Reprod., April 1, 2005; 20(4): 852 - 863. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Luo, L. Ding, J. Xu, R. S. Williams, and N. Chegini Leiomyoma and Myometrial Gene Expression Profiles and Their Responses to Gonadotropin-Releasing Hormone Analog Therapy Endocrinology, March 1, 2005; 146(3): 1074 - 1096. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bailey, A. J Tyson-Capper, K. Gilmore, S. C Robson, and G N. Europe-Finner Identification of human myometrial target genes of the cAMP pathway: the role of cAMP-response element binding (CREB) and modulator (CREM{alpha} and CREM{tau}2{alpha}) proteins J. Mol. Endocrinol., February 1, 2005; 34(1): 1 - 17. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. P. Ponnampalam, G. C. Weston, A. C. Trajstman, B. Susil, and P. A.W. Rogers Molecular classification of human endometrial cycle stages by transcriptional profiling Mol. Hum. Reprod., December 1, 2004; 10(12): 879 - 893. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|