Hum. Reprod. Advance Access originally published online on September 19, 2005
Human Reproduction 2006 21(1):57-67; doi:10.1093/humrep/dei309
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ethnic differences in expression of the dysregulated proteins in uterine leiomyomata
1 Department of Pathology and 2 Department of Obstetrics and Gynecology, New York University School of Medicine, 560 First Avenue, New York, NY 10016, USA
3 To whom correspondence should be addressed at: Department of Pathology, New York University School of Medicine, Bellevue Hospital, NB4W1, 462 First Avenue, New York, NY 10016, USA. E-mail: weij03{at}med.nyu.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionAcknowledgementReferences |
|---|
BACKGROUND: Black ethnicity is one of the risk factors for uterine leiomyomata (ULM). Little is known about the ethnic differences in leiomyoma-associated gene products in women with uterine leiomyomata. METHODS: A total of 120 hysterectomies with ULM were collected from black, Asian, Hispanic and white women (30 cases from each group). Twenty-two gene products were selected for the study. The expressions of the selected dysregulated gene products were measured by the semiquantification and the immunoscores were normalized by matched myometrium. RESULTS: The relative expressions of progesterone receptor A (PR-A) (up-regulation), retinoid acid receptor
(down-regulation), and retinoid X receptor (RXR) (no change) in leiomyomata compared to normal myometrium in black women were significantly different compared to other ethnic groups (P < 0.05). About one-third of ULM from black women subclustered together in association with a group of up-regulated gene products. Many other gene products, including local growth factors, insulin-like growth factor (IGF)-signalling proteins, and cell proliferation markers, were dysregulated in ULM but showed non-significant differences between the ethnic groups. CONCLUSIONS: There are substantial differences of the sex steroid receptors and other nuclear receptors between black women and other ethnic groups. Based on tissue microarray data, there are at least two broad groups of leiomyomata presented by the dysregulation of different groups of gene products. One is dominated by up-regulation of amplified in breast cancer 1, CD24, hamartin, human mobility group gene 2, IGF2, PR-A and RXR, and the other is characterized by up-regulation of epithelial growth factor receptor, down-regulation of hamartin, PR-A and tuberin.
Key words: ethnicity/immunohistochemistry/leiomyoma/nuclear receptors/tissue microarray
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementReferences |
|---|
Uterine leiomyomata (ULM) are a common health problem for women. A recent study showed that a lifelong risk to develop leiomyoma is 70% in white and 80% in black women (Baird et al., 2003
). The extent of the health care problem is seen by 
200,000 hysterectomies due to ULM in the United States annually (Varol et al., 2001). During the past two decades, some progress has been made towards the understanding of the pathogenesis of the disease from the clinical, genetic, epidemiological and molecular perspectives (for review, see Flake et al., 2003). However, the causes of ULM remain largely unknown.
Observations from the clinical and epidemiological studies showed that ULM in black women tend to develop earlier and become larger, more numerous and more symptomatic, as compared to white women (Kjerulff et al., 1996). The higher incidence and greater morbidity of leiomyomata are also reflected in higher rates of hysterectomy for this condition in black women (Wilcox et al., 1994). Genetic susceptibility or environmental factors may play a role in the differences in ULM presentation between black and white women (Flake et al., 2003). Very few studies, however, have investigated these factors across ethnic groups. A large cohort study, comparing the incidence of leiomyomata in four ethnic groups of black, Hispanic, Asian and white Americans, indicated that: (i) black women have the highest incidence of leiomyomata compared to other ethnic groups; and (ii) after controlling for age, body mass index, birth history, alcohol consumption, and oral contraceptive use, these risk factors cannot explain the excessive rate of uterine leiomyoma among pre-menopausal black women (Marshall et al., 1997).
Understanding the reasons for such ethnic disparity may provide clues about aetiology and pathogenesis of ULM. One study found that a high fat diet is associated with significantly higher levels of estrogen in black women than in white women (Woods et al., 1996). The use of injectable progesterone, however, appears to be a protective factor and is inversely associated with the ULM risk in African-Americans (Wise et al., 2004). However, two studies failed to show significant differences in the expression of estrogen and progesterone receptors in the myometrium between black and white women (Sadan et al., 1988; Amant et al., 2003).
Transcription profiling (Tsibris et al., 2002; Ahn et al., 2003; Catherino et al., 2003; Chegini et al., 2003; Skubitz and Skubitz, 2003; Wang et al., 2003; Weston et al., 2003; Hoffman et al., 2004; Quade et al., 2004; Arslan et al., 2005) provides a useful molecular tool for identification of dysregulated genes and their products in ULM. Previously, we applied tissue microarray (TMA) for examining the dysregulated gene products in ULM (Wei et al., 2005a) and found that it is a practical and reliable way to characterize the expressed proteins in ULM.
In the current study, we applied immunohistochemistry with high-density tissue TMA to identify the ethnic differences in the dysregulated gene products between black, Asian, Hispanic and white women with ULM. The gene products examined in the study were primarily selected from the previously published gene chip data and also from other studies of ULM (Tsibris et al., 2002; Ahn et al., 2003; Catherino et al., 2003; Chegini et al., 2003; Skubitz and Skubitz, 2003; Wang et al., 2003; Weston et al., 2003; Hoffman et al., 2004; Quade et al., 2004; Arslan et al., 2005; Wei et al., 2005a). The purpose of the study is to further determine whether the selected gene products, which dysregulated in most leiomyomata, are differentially expressed among different ethnic groups. Particularly, we intended to search for the contributing factors among the selected dysregulated gene products in black women with ULM.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementReferences |
|---|
Patients and specimens
After reviewing 850 consecutive hysterectomies for ULM for a period of 4 years (2000–2003) in the New York University Medical Center, we identified those patients who underwent hysterectomy for a symptomatic ULM and who had a well-documented clinical information, including ethnicity. ULM coinciding with any malignancy in the reproductive organs were excluded from the study. In all, 360 patients met these criteria and were eligible for inclusion in the study. We then randomly selected 30 cases for each of the following ethnicities: black, Asian, Hispanics and white women. The sample size of 30 cases per ethnic group was derived from the power analysis using nQuery Adviser Statistical Software based on the results of our previous study (Wei et al., 2005b
), which showed that significant differences for most dysregulated gene products can be detected with sample size of 13 to 22 subjects per group with 80% of power at the 0.05 alpha level. A total of 120 cases was included in the final analysis. The general clinical information for each ethnic group is summarized in Table I.
|
The information on ethnicity was self-reported. We classified patients as ‘black’ if they indicated African-American but non-Hispanic; patients as ‘Hispanic’ if they indicated Hispanic and not African-American; patients as ‘Asian’ if they indicated Asian (in our study all Asians were Chinese-American); patients as ‘white’ if they indicated as white or Caucasian, but not Hispanic; patients as ‘other’ if they were not among the four ethnic/racial categories mentioned above. We did not include the patients defined as ‘other’ in this study.
The study was approved by the Institutional Board of Research Associates of New York University School of Medicine.
TMA
All tissue sections were reviewed and cellular areas of tumour were selected for TMA. Five 0.6 mm tissue cores were collected from each case, including two cores of each from the outer third (M1) and the inner third (M2) myometrium in posterior uterine wall and three cores of each from two to three tumour sections. The technical details and reliability data were previously described (Wei et al., 2005b). In brief, 600 tissue cores from 120 ULM with matched myometrium were arrayed into two recipient paraffin blocks. The tissue cores were arrayed consecutively according to increasing accession numbers to ensure a random distribution of cases and thereby avoid clustering according to race. Once the original dataset was created, the cases were de-identified through deletion of the accession numbers and replacement by random numbers for the rest of the study. Tissues from human breast cancer and normal myometrium which had been previously fixed for different times (4 h, 8 h, 24 h, 2 days and 3 days) were added in the blocks as control samples.
Immunomarkers
Antibodies against the following protein markers were selected for this study: (i) insulin-like growth factor and associated proteins: insulin-like growth factor 2 (IGF2), insulin-like growth factor receptor 
(IGF1-R), hamartin and tuberin; (ii) other growth factors: epithelial growth factor receptor (EGFR), platelet-derived growth factor (PDGF); (iii) steroid nuclear receptors and co-factors: estrogen receptor (ER), progesterone receptor A (PR-A), glucocorticoid receptor (GCR), retinoid acid receptor (RAR), retinoid X receptor (RXR), peroxisome proliferator-activated receptor-
(PPAR), steroid receptor co-factor 1 (SRC1) and amplified in breast cancer 1 (AIB1); (iv) cell proliferation and survival markers: Ki-67, cyclin D1, p27 and BCL-2; (v) other ULM-associated proteins: CD24, human mobility group gene (HMGA2), Factor VIII, and smooth muscle actin (SMA). SMA is used as control for immunostain. Twenty-two protein markers were examined in this study. The vendors and working conditions for the immunomarkers are summarized in Table II. The detailed protocols and conditions for each antibody followed the manufacturers’ recommendations and were previously described (Wei et al., 2005a).
|
Immunohistochemistry
The paraffin-embedded tissue array blocks were sectioned at 4 µm. Antigen retrieval was performed by either heat-induced epitope retrieval or by proteolytic enzyme digestion as previously described (Wei et al., 2005a). All immunohistochemical staining was performed on a Ventana Nexus automated system (Tucson, Arizona, USA). In brief, following antigen retrieval, endogenous peroxidase activity was blocked with 3% hydrogen peroxide. Primary antibodies were detected using standard biotinylated anti-mouse, anti-goat or anti-rabbit secondary antibodies. This was followed by application of streptavidin–horseradish peroxidase conjugate. The complex was visualized by the enzymatic oxidation of DAB (3,3'-diaminobenzidine tetrahydrochloride) substrate and enhanced with copper sulphate. Slides were counterstained with haematoxylin and mounted with permanent media.
Data analysis and scoring for immunodensity
Stained TMA slides were graded jointly by two pathologists using a visual semiquanficitation method (optical density of the immunoreactivity) depending on the staining characteristics of the antibody. One score system for immunointensity was used for the markers (AIB1, CD24, EGFR, hamartin, HMGA2, IGF2, IGF1R, SMA and SRC1), showing a diffuse immunoreactivity in tumours and the matched myometrium. A two-score system for immunointensity (II) and immunopositivity (IP) was used for those markers (BCL-2, ER, GCR, PDGF, PR, RAR, RXR and Tuberin) that were immunoreactive for only portion of the tested cells. The semiquantification for intensity was scored on a scale of: 0, negative; 1, weak; 2, moderate; 3, strong; 4, very strong. The semiquantification for percentage of immunopositive cells was scored on a scale of 1 (1–10%), 2 (11–50%), 3 (51–80%) and 4 (>80%). The combination scores for the two-score system was obtained by multiplying the scores for immuno-intensity by the score for passivity (IIxIP), using a method modified from Nisolle et al. (1999). Ki-67 and Factor VIII were scored by the percentage of immunopositive cells and vessels. The net gain or loss of immunoscore was calculated for a specific immunomarker from each individual ULM in comparison with matched myometrium (net value = tumour immunoscore – myometrial immunoscore). Each net value gave positive or negative scores. This produces net immunoscore values that range from –16 to +16.
Statistical analyses
Analysis of the data distribution using Shapiro–Wilk W-test of normality had shown that the differences in immunoscores between ULM and myometrium were not normally distributed for most of the markers. Therefore, for overall differences in continuous variables between four ethnic groups, we used the non-parametric Kruskal–Wallis test, which is appropriate for comparison of three or more unmatched groups, with Tukey–Kramer correction for multiple comparisons. We also conducted pair-wise group comparison using non-parametric Wilcoxon rank sum test. For comparison of categorical variables between ethnic groups we used 
2-test. Statistical analyses were carried out using JMP 4.0 (SAS Institute Inc. Cary, NC, USA) statistical software.
Unsupervised hierarchal cluster analysis
Net immunoscores were processed into a pre-cluster format using the TMA deconvoluter (Eisen et al., 1998). Data was clustered using Cluster and the output visualized with Treeview, software originally designed for analysing cDNA microarray data (Liu et al., 2002). A detailed application of the cluster analysis in our TMA was described in a previous study (Wei et al., 2005b). In brief, net immunoscores for markers scored by the standard scoring system (see above) were normalized for clustering by multiplying each net score by its absolute value. The standard and two-component datasets were combined and deconvolved. The deconvolved data were pre-filtered to eliminate samples that had insufficient data (<80% threshold), as well to eliminate scores in which the absolute, maximum minus minimum value was <0.50. Distances were calculated using an un-centred Pearson correlation metric, and then clustered by the average linkage method (Eisen et al., 1998).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementReferences |
|---|
General information
A total of 120 cases was selected from the patients who underwent hysterectomies for symptomatic uterine leiomyomata. The mean age was 46.8 years (range 37–59 years). The mean tumour size was 6.8 cm (range 1–21 cm). Overall, there was no statistical difference in the patients’ age and tumour sizes among the ethnic groups. However, in the pair-wise analysis, black and Hispanic women tended to be younger at hysterectomy (mean ages 45.97 and 45.6 years respectively) than white and Asian women (mean ages 48.2 and 47.4 years respectively, P < 0.02) (Figure 1). On average, Hispanic women tended to have larger tumours in size and weight than other ethnic groups but these differences were not statistically significant (Table I). However, the ethnic differences in immunoscores were not significantly affected after control for age of surgery (<50 and
50 years).
|
Myometrium
Gene expression in the normal myometrium varies during different phases of endometrial cycle (Kovacs et al., 2003) and may vary in different anatomic (spatial) regions (Noe et al., 1999). To minimize the score errors in matched myometrium due to these two variations, we first compared the ratio of the endometrial phases in each group. As illustrated in Table I, the ratio of proliferative/secretory/inactive endometrium, as well as the incidence of adenomyosis were similar between ethnic group. Considering that the outer myometrium is of non-paramesonephric origin and the inner myometrium is of paramesonephric origin (Noe et al., 1999), we collected two myometrial tissue cores, one from the outer third (M1) and another one from the inner third (M2) at the posterior myometrium in all cases. The immunoreactivity scores from M1 and M2 were determined separately. The mean immunoreactivity scores of M1 and M2 for 120 cases were then calculated. The inner myometrium showed a significantly higher expression of the sex steroid hormone receptors (P < 0.001) than outer myometrium (Figure 2). The proliferative rate was higher in the inner layer as well M1 = 1.61 0.57, M2 = 1.92 0.74. All other markers demonstrated a similar pattern as estrogen and progesterone receptors, but only some of them reach the statistical significance (data not shown). Therefore, in the following analysis, the scores for the normal myometrium were taken from the mean value of the inner and the outer myometrium for all markers.
|
Differential expression of the selected dysregulated gene products in ULM from the different ethnicities
Steroid hormone nuclear receptors and co-factors
Expression of ER was detectable in all cases, including the normal myometrium and ULM. An up-regulation of ER in ULM in comparison with myometrium was found in nearly 50% of the cases (59/120), no changes in 27.5% (33/120) and down-regulation in 23.4% (28/120). Up-regulation of ER was present in two peaks in the age groups of <40 and >50 years. Up-regulation of ER was at the lowest level in the late forties (Figure 3A). There was minimal difference of ER in myometrium among the different tumour sizes. An inverse association of tumour sizes with the level of ER expression in tumour tissue was observed (Figure 3B). The average immunoscores of ER in myometrium and ULM were quite different between the four ethnic groups. Black women had the highest scores of ER in both myometrium and leiomyoma (2.5 and 3.0), followed by the Hispanic and the Asian, and the white women had the lowest score of ER (1.4 and 1.9) (Figure 3C). The differences in ER between black and white was statistically significant both for ULM (P < 0.05) and myometrium (P < 0.04). However, comparison of the changes in expression level of ER in ULM against myometrium showed no significant differences among the ethnic groups (Table III, Figure 3C).
|
|
The overall immunoscores for PR-A was twice as high as that of ER. Up-regulation of PR-A was identified in 52.5% (63/120) of ULM, no change in 15.8% (19/120) and down-regulation in 24.2% (29/120). The remaining nine cases had no data due to missing tissue sections. The average scores of PR-A in myometrium were very similar among all racial groups (ranging from 5.1 to 5.4). However, black women showed much a higher level of PR-A in ULM, with an average value of 7.9 in comparison to the mean value of 5.8 in the other three races (P < 0.007). Black women had a significant up-regulation of PR-A in ULM (median, +2.8), followed by white women (median, +1.8). There were minimal changes of PR-A in Hispanic (median, 0) and Asian women (median, 0) (P = 0.02) (Table III, Figure 3D).
Expression of RAR was detectable in almost all ULM samples. Up-regulation of RAR was identified in 43.3% of cases (52/120), no change in 13.3% (16/120), and down-regulation in 34.2% (41/120). Eleven cases had no data. Among the cases with a down-regulation of RAR, >70% of them (29/41) were from black and Hispanic women. The overall expression of RAR showed a down-regulation in black, no change in Hispanic and an up-regulation in both Asian and white women (P < 0.02) (Table III).
There was a constant level of RXR expression (mean score of 3.7) in myometrium with a minimal difference between ethnic groups. Compared to the myometrium, up-regulation of RXR in ULM was present in 68 cases accounting for 56.7% of all cases, whereas down-regulation or no changes in RXR were observed in the equal numbers of the patients (26 cases each). Up-regulation of RXR was present in ULM from white, Asian and Hispanic women with the median gain values of +0.8, +1.3 and +0.5 respectively. Black women had a minimal change of RXR in ULM in comparison with matched myometrium with a median value of 0 (Table III).
PPAR was tested but was non-informative due to the weak signalling detected by immunohistochemistry.
A total of 70% of leiomyoma cases (84/120) displayed a down-regulation of GCR in comparison with myometrium. Down-regulation of GCR was evenly distributed in all four ethnic groups. There were no significant differences in the GCR expression between the four ethnic groups (Table III).
AIB1 was up-regulated in 59.2% of ULM patients (71/120). Only 12 cases showed down-regulation. The scores for up-regulation of AIB1 were the lowest in Hispanic women (median +1.0) and the highest in black and Asian women (both with median change of +2.0). However, the difference in AIB1 expression between the ethnic groups was not statistically significant.
SCR1 showed an up-regulation in 66 leiomyomas (55%) and a down-regulation in 18 leiomyomas (15%). The average immunoreactivity scores for SRC1 in myometrium were very similar among the four ethnic groups ranging from 2.4 to 2.6. The up-regulation of SRC1 was present in black, Asian and white women, but not in Hispanic women (Table III).
Insulin-like growth factors and associated proteins
IGF2 was diffusely expressed in both myometrium and ULM. Seventy-nine ULM (65.8%) showed up-regulation of IGF2 compared to the myometrium. IGF2 was down-regulated in only eight ULM. The average expression of IGF2 in myometrium was significantly higher in black and Hispanic women (mean scores: 5.3 and 5.4 respectively) than Asian and white women (mean scores: 3.9 and 4.0 respectively). However, the increased levels of IGF2 in ULM were similar in all ethnic groups and this led to a smaller relative up-regulation of IGF2 in black and Hispanic women compared to Asian and white women (Table III).
IGF1R showed a minimal up-regulation (median: 0) in ULM on average. Only 40.8% (49/120) of ULM showed IGF1R up-regulation, while nearly 50% of the cases showed no change of IGF1R. There were no significant differences in IGF1R expression among the four ethnic groups.
Tuberin was weakly to moderately expressed in the myometrium with a patchy cytoplasmic staining pattern. We scored tuberin by combining the intensity and percentage of the immunoreactivity. A total of 63 cases showed varying degrees of down-regulation (52.5%). Among them, 11 cases showed an absence of tuberin in ULM. There was no significant difference in tuberin expression in both myometrium and ULM among the four ethnic groups (Table III).
Hamartin was moderately expressed in the myometrium with a mean immunoscore of 1.3. Up-regulation of hamartin was noted in 41% of ULM, 23% of ULM showed down-regulation and 28% had no change of hamartin. No ethnic differences in hamartin expression were noted (Table III).
Proliferation, cell survival and other tumorigenic factors
Nearly 50% of the cases showed an up-regulation of BCL-2 in ULM (57/120). Immunoreactivity for BCL-2 in the myometrium was higher in black women than in other races, while similar levels of BCL-2 expression in ULM were noted in all races. Such expression pattern accounted for a smaller relative increase in BCL-2 in black women in comparison with other ethnic groups. The relative differences in BCL-2 expression between black and other ethnic groups were borderline statistically significant (P < 0.07) (Table III).
Ki-67 was scored by the percentage of immunopositive cells in both myometrium and ULM. Increased Ki-67 immunoreactivity in ULM was found in 62.5% of the cases (75/120). There was no significant difference in Ki-67 immunoreactive cell counts from ULM and matched myometrium between ethnic groups (Table III).
More than 50% of patients (61/120) showed an up-regulation of HMGA2. The ratios of immunopositive cases in each ethnic group were quite similar (black: 15; Asian: 14; Hispanic: 16; and white: 16). The median values of HMGA2 in myometrium and ULM were similar and no significant differences were noted between the four ethnic groups (Table III).
CD24 was up-regulated in 68 cases (56.7%). Only three cases showed slight down-regulation of CD24. Our results suggest that CD24 is a reliable ULM-associated marker. CD24 was detectable in most normal myometrium samples with a mean value of 1.04. The mean value in leiomyomas was 1.57. CD24 was up-regulated in a total of 22 cases of Hispanic women (73.3%) in comparison to 15–18 cases from the other ethnic groups. This difference was also reflected by higher mean and median values of the up-regulation of CD24 in Hispanic women, as compared to other ethnic groups (P < 0.02) (Table III).
PDGF was almost not detectable in normal myometrium. In contrast, 28 ULM (23.3%) were immunopositive (although mostly in focal tumour cells) for PDGF. Among the PDGF-positive cases, more were from white women (9/28) (Table III).
EGFR was up-regulated in 45 ULM (38%). In contrast, 49 ULM (41%) showed no change of EGFR expression between ULM and matched myometrium. A small proportion of ULM (16%) revealed a down-regulation of EGFR. There were no overall significant differences in EGFR expression among ethnic groups (Table III).
Factor VIII was studied to examine the vascularity of ULM in comparison to the myometrium. We found that ULM had significantly fewer vessels than the matched myometrium. The decrease in the number of vessels showed a similar pattern in all ethnic groups (Table III). There was no association between the vascular reduction in ULM and patients’ age, endometrial phase, tumour size and presence or absence of adenomyosis. The immunoreactivities for p27 and Cyclin D1 were either negative or too weak to be scored. No further analysis of these two markers was performed.
Dendrogram and cluster analysis
For the cluster analysis, we first calculated the net gain or loss of the immunoscore for a specific immunomarker from each individual leiomyoma in comparison with matched myometrium (net value = tumour immunoscore – myometrial immunoscore). The immunoscores were then processed into pre-cluster format using the TMA deconvoluter. In an unsupervised cluster analysis, the markers presented two broad groups in 120 cases, centred on CD24 (up-regulation) and GCR (down-regulation) (Figure 4). The distribution pattern is similar to our previous study in 60 ULM (Wei et al., 2005a), showing 48% of ULM were clustered together with up-regulation of PR-A, hamartin, IGF2, IFG1R, CD24, HMGA2, AIB1 and SRC1 (group I). All remaining cases showed less up-regulation or minimal changes of the examined proteins, except for EGFR. EGFR was the dominant marker up-regulated in most of the remaining cases accompanied by down-regulation of tuberin, hamartin and PR-A (group II), while showing minimal changes in cases with up-regulation of the other examined markers (Figure 4).
|
In the unsupervised cluster analysis, group I ULM from black women were sub-clustered together (Figure 4), while the other three races tended to be equally and randomly distributed. In contrast, in group II ULM, black and white women tended to aggregate together.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementReferences |
|---|
In the past few years, some progress has been made in characterizing the molecular differences between ULM and normal myometrium. This process is greatly facilitated by the data from transcriptional profiling of ULM. More than a hundred dysregulated genes were identified in ULM using gene array studies (Tsibris et al., 2002
; Ahn et al., 2003; Catherino et al., 2003; Chegini et al., 2003; Skubitz and Skubitz, 2003; Wang et al., 2003; Weston et al., 2003; Hoffman et al., 2004; Quade et al., 2004; Arslan et al., 2005). Some 20 dysregulated genes have been consistently detected in ULM by at least three different laboratories (Arslan et al., 2005). Further study of these dysregulated genes and their products is important for the further characterization of the ULM pathogenesis and growth.
Analysis of the dysregulated gene products in women with ULM from different ethnic groups provides one possible route to further dissect the contribution and dysfunction of these genes and their products in ULM. Since black women with ULM tend to have larger tumours and greater morbidity than other ethnic groups (Marshall et al., 1997), the rationale of this study was to examine if the dysregulated gene profiles in ULM result in different protein expression between black, Asian, Hispanic and white women. The gene products considered for this study were found to be dysregulated in ULM and included steroid hormone receptors (ER, PR-A, GCR) (Tsibris et al., 2002), retinoic acid receptors (RAR, RXR) (Gamage et al., 2000; Tsibris et al., 2002; Wei et al., 2005a), PPAR (Houston et al., 2003), nuclear receptor co-factors (AIB1, SRC1) (Wei et al., 2005a), IGF-signalling factors (IGF2, IGF1R) (Tsibris et al., 2002; Ahn et al., 2003; Skubitz and Skubitz, 2003; Wang et al., 2003; Weston et al., 2003; Hoffman et al., 2004; Quade et al., 2004a; Martin Chaves et al., 2004), tuberin, hamartin (Wei et al., 2005a), and other tumorigenic factors (HMGA2) (Gross et al., 2003), CD24 (Catherino et al., 2003), BCL-2 (Kovacs et al., 2003), EGFR (Dixon et al., 2000), PDGF (Di Lieto et al., 2002). Most of the antibodies to these gene products have been tested by us in a previous study of ULM (Wei et al., 2005a). As summarized in Figure 5, we observed a wide range in the levels of dysregulation of the selected gene products in ULM. Even for well-characterized proteins, only a certain proportion of the examined tumours showed a dysregulation (either up- or down-regulations). Apparently, patient’s age, tumour size, and sex steroid hormone status (Wei et al., 2005b) are among the factors that affect the expression levels of the dysregulated genes and their products in ULM. The markers with higher numbers of patients with dysregulation of gene products, such as IGF2, CD24, HMGA2, GCR and tuberin, appear to have a minimal difference between ethnic groups (Figure 5).
|
Dysregulation of steroid hormones and their receptors are thought of as the primary factors in determining ULM growth. The mean expression value of ER in both normal myometrium and ULM was higher in black women compared to other ethnic groups (Figure 2). PR-A gene product is significantly higher in black women than in Asians and Hispanics, and it is also higher in black than in white women but does not reach statistical significance (Figure 3). Our results of ER/PR protein levels detected by immunohistochemistry are different from the study by Amant et al. (2003), who used the binding assay and did not find a difference in ER and PR expression in the myometrium between black and white women. ER and PR quantification can be affected by many factors, including method of tissue preservation, tissue location (spatial region of the myometrium) and tumour size (we found that the larger myomas had much lower levels of ER/PR than the smaller ones) (Wei et al., 2005b). Differentially expressed ER and PR could be the crucial factors determining the differences in ULM progression between different ethnic groups.
The difference between the down-regulation or lower levels of the steroid nuclear receptors RAR and RXR in black women in comparison with their up-regulation in other ethnic groups is another important finding of this study. It has cast attention on the role of retinoic acid, RAR and RXR in tumour development with potential implications for ULM therapy and prevention, as previously suggested (Tsibris et al., 1999). The mechanism of RAR and RXR dysregulation in leiomyoma may be mediated by estrogen (Tsibris et al., 1999). Several studies from the transcription profiling of ULM showed the dysregulation of a group of genes related to the retinoic acid metabolism (Duester, 1996) including a down-regulation of alcohol and dehydrogenase 1 (ADH1) and aldehyde dehydrogenase 1 (ALDH1), and an up-regulation of retinoid acid binding protein II and retinoid acid dehydrogenase (Tsibris et al., 2002; Ahn et al., 2003; Catherino et al., 2003; Skubitz and Skubitz, 2003; Wang et al., 2003; Quade et al., 2004; Arslan et al., 2005). One can presume that up-regulation of RAR and RXR may be associated with down-regulation of the enzymes for a dysfunction of retinoic acid metabolism. The role of RAR and RXR in the determination of ethnic differences and in the development of leiomyomata requires additional studies.
The advantage of using high throughput TMA technology is that tumour tissues from hundreds of patients can be analysed in a reasonable time, and without immense labour and reagent requirement. More importantly, it can identify some dysregulated gene products that may not be detectable using gene chip analysis. For example, ER, PR-A, RAR and EGFR were overall up-regulated by measurement of the mean values. However, there were nearly equal numbers of ULM showing either up- or down-regulation of these gene products (Figure 5). EGFR showed different signatures in its up- and down-regulation (Figure 4). In addition, the functional proteins in the IGF signalling pathway due to the phosphorylation status can only be detected by immunohistochemistry, but not by mRNA.
Dysfunction of the IGF signalling pathway may be another important reason for a rapid and large tumour growth. Up-regulation of IGF2 in ULM was identified at mRNA and protein levels (Tsibris et al., 2002; Ahn et al., 2003; Catherino et al., 2003; Chegini et al., 2003; Skubitz and Skubitz, 2003; Wang et al., 2003; Weston et al., 2003; Hoffman et al., 2004; Quade et al., 2004; Arslan et al., 2005). There is growing evidence demonstrating the association of the dysregulation of the IGF system with many human neoplasms (Pollak, 2004). We previously examined the expression of IGF1 and IGF2 in 60 patients with ULM by the TMA (Wei et al., 2005a). IGF were up-regulated in a large number of tumours in comparison to the matched myometrium (Dixon et al., 2000; Wei et al., 2005a). In addition, we detected dysregulation of PI3K, p-AKT and BCL-2 in ULM (Wei et al., 2005a).
We found a reduction of tuberin in a leiomyomata group and tuberin has been shown to play a role in the IGF pathway (Gao and Pan, 2001; Potter et al., 2001). The proteins selected from the IGF signalling pathway in this study were dysregulated in large numbers of ULM. However, based on our data, dysfunction of the IGF pathway appears not to be responsible for the ethnic differences in ULM.
The calculated mean scores of the dysregulated gene products in each ethnic group represent a global dysregulation of the tumorigenic genes in these patients. Comparison of all dysregulated genes and their products in specific patients as well as the ethnic differences will be useful for further evaluation of the relationship between the ethnicity and abnormal gene expression.
We previously reported two broad groups of leiomyomata based on the analysis of the transcriptional profile in a large number of patients with leiomyomata (Wei et al., 2005a). Our dataset from these additional 120 ULM showed again two patterns of the dysregulated gene products in ULM. The first pattern (group I) is characterized by up-regulation of several gene products, including AIB1, CD24, hamartin, HMGA2, IGF2, PR-A and RXR. In contrast, the second pattern (group II) is characterized by up-regulation of EGFR, down-regulation of hamartin, PR-A and tuberin (Figure 4). These findings indicate that there are at least two different types of ULM based on TMA profiling from two independent cohort studies (Wei et al., 2005a).
The findings by comparison of mean differences in a group of the selected gene products between four different ethnic groups provide clues for the further characterization of potential molecular candidates that may contribute to the racial differences in development of ULM. Our attempts in searching for molecular markers for the racial differences in ULM are supported by the increasing knowledge from the gene chip data, advances in technologies for TMA and availability of the diverse patient population. As mentioned earlier, currently identified environmental factors that are associated with ULM cannot explain the racial differences for the tumour incidence and morbidity, particularly the greater disease burden in black women. Therefore, establishing a study strategy to examine large numbers of the dysregulated genes or gene products between ethnic groups will be a promising future direction to better understand and potentially develop new strategies for prevention of ULM.
|
Acknowledgement |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementReferences |
|---|
We thank Dr Mizuguchi for providing anti-tuberin and anti-hamartin antibodies. The work is presented in part at ‘Advances in Uterine Leiomyoma Research: 2nd NIH International Congress’.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementReferences |
|---|
Ahn, WS, Kim, KW Bae SM, Yoon JH, Lee JM, Namkoong SE, Kim JH, Kim CK, Lee YJ and Kim YW (2003) Targeted cellular process profiling approach for uterine leiomyoma using cDNA microarray, proteomics and gene ontology analysis. Int J Exp Pathol 846,267–279.
Amant F, Huys E, Geurts-Moespot A, Lindeque BG, Vergote I, Sweep F and Schoenmakers EF (2003) Ethnic variations in uterine leiomyoma biology are not caused by differences in myometrial estrogen receptor alpha levels. J Soc Gynecol Invest 102,105–109.
Arslan AA, Gold LI, Mittal K, Suen TC, Belitskaya-Levy I, Tang MS and Toniolo P (2005) Gene expression studies provide clues to the pathogenesis of uterine leiomyoma: new evidence and a systematic review. Hum Reprod 204,852–863.
Catherino WH, Prupas C, Tsibris JC, Leppert PC, Payson M, Nieman LK and Segars JH (2003) Strategy for elucidating differentially expressed genes in leiomyomata identified by microarray technology. Fertil Steril 802,282–290.
Chegini N, Verala J, Luo X, Xu J and Williams RS (2003) Gene expression profile of leiomyoma and myometrium and the effect of gonadotropin releasing hormone analogue therapy. J Soc Gynecol Invest 103,161–171.
Baird DD, Dunson DB, Hill MC, Cousins D and Schectman JM (2003) High cumulative incidence of uterine leiomyoma in black and white women: ultrasound evidence. Am J Obstet Gynecol 1881,100–107.
Di Lieto A, De Rosa G, De Falco M, Iannotti F, Staibano S, Pollio F, Scaramellino M and Salvatore G (2002) Relationship between platelet-derived growth factor expression in leiomyomas and uterine volume changes after gonadotropin-releasing hormone agonist treatment. Hum Pathol 332,220–224.
Dixon D, He H and Haseman JK (2000) Immunohistochemical localization of growth factors and their receptors in uterine leiomyomas and matched myometrium. Environ Health Perspect 108(Suppl 5)795–802.
Duester G (1996) Involvement of alcohol dehydrogenase, short-chain dehydrogenase/reductase, ldehyde dehydrogenase and cytochrome P450 in the control of retinoid signaling by activation of retinoic acid synthesis. Biochemistry 3538,12221–12227.
Eisen MB, Spellman PT, Brown PO and Botstein D (1998) Cluster analysis and display of genome-wide expression patterns. Proc Natl Acad Sci USA 95,14863–14868.
Flake GP, Andersen J and Dixon D (2003) Etiology and pathogenesis of uterine leiomyomas: a review. Environ Health Perspect 1118,1037–1054.
Gamage SD, Bischoff ED, Burroughs KD, Lamph WW, Gottardis MM, Walker CL and Fuchs-Young R (2000) Efficacy of LGD1069 (Targretin), a retinoid X receptor-selective ligand, for treatment of uterine leiomyoma. J Pharmacol Exp Ther 295,677–681.
Gao X and Pan D (2001) TSC1 and TSC2 tumor suppressors antagonize insulin signaling in cell growth. Genes Dev 15,1383–1392.
Gross, KL, Neskey DM, Manchanda N, Weremowicz S, Kleinman MS, Nowak RA, Ligon AH, Rogalla P, Drechsler K, Bullerdiek J and Morton CC (2003) HMGA2 expression in uterine leiomyomata and myometrium: quantitative analysis and tissue culture studies. Genes Chromosomes Cancer 38,68–79.[CrossRef][Web of Science][Medline]
Hoffman PJ, Milliken DB, Gregg LC, Davis RR and Gregg JP (2004) Molecular characterization of uterine fibroids and its implication for underlying mechanisms of pathogenesis. Fertil Steril 82,639–649.[CrossRef][Web of Science][Medline]
Houston KD, Copland JA, Broaddus RR, Gottardis MM, Fischer SM and Walker CL (2003) Inhibition of proliferation and estrogen receptor signaling by peroxisome proliferator-activated receptor gamma ligands in uterine leiomyoma. Cancer Res 636,1221–1227.
Kjerulff KH, Langenberg P, Seidman JD, Stolley PD and Guzinski GM (1996) Uterine leiomyomas. Racial differences in severity, symptoms and age at diagnosis. J Reprod Med 41,483–490.[Web of Science][Medline]
Kovacs KA, Lengyel F, Kornyei JL, Vertes Z, Szabo I, Sumegi B and Vertes M (2003) Differential expression of Akt/protein kinase B, Bcl-2 and Bax proteins in human leiomyoma and myometrium. J Steroid Biochem Mol Biol 87,233–240.[CrossRef][Web of Science][Medline]
Liu CL, Prapong W, Natkunam Y, Alizadeh A, Montgomery K, Gilks CB and van de Rijn M (2002) Software tools for high-throughput analysis and archiving of immunohistochemistry staining data obtained with tissue microarrays. Am J Pathol 161,1557–1565.
Marshall LM, Spiegelman D, Barbieri RL, Goldman MB, Manson JE, Colditz GA, Willett WC and Hunter DJ (1997) Variation in the incidence of uterine leiomyoma among premenopausal women by age and race. Obstet Gynecol 90,967–973.[CrossRef][Web of Science][Medline]
Martin Chaves EB, Brum IS, Stoll J, Capp E and Corleta HE (2004) Insulin-like growth factor 1 receptor mRNA expression and autophosphorylation in human myometrium and leiomyoma. Gynecol Obstet Invest 57,210–213.[CrossRef][Web of Science][Medline]
Mizuguchi M, Kato M, Yamanouchi H, Ikeda K and Takashima S (1996) Loss of tuberin from cerebral tissues with tuberous sclerosis and astrocytoma. Ann Neurol 40,941–4.
Nisolle M, Gillerot S, Casanas-Roux F, Squifflet J, Berliere M and Donnez J (1999) Immunohistochemical study of the proliferation index, oestrogen receptors and progesterone receptors A and B in leiomyomata and normal myometrium during the menstrual cycle and under gonadotrophin-releasing hormone agonist therapy. Hum Reprod 14,2844–2850.
Noe M, Kunz G, Herbertz M, Mall G and Leyendecker G (1999) The cyclic pattern of the immunocytochemical expression of oestrogen and progesterone receptors in human myometrial and endometrial layers: characterization of the endometrial–subendometrial unit. Hum Reprod 14,190–197.
Pollak MN (2004) Insulin-like growth factors and neoplasia. Novartis Found Symp 262,84–98; discussion 98–107,265–268.
Potter CJ, Huang H and Xu T (2001) Drosophila Tsc1 functions with Tsc2 to antagonize insulin signaling in regulating cell growth, cell proliferation and organ size. Cell 105,357–368.[CrossRef][Web of Science][Medline]
Quade BJ, Wang TY, Sornberger K, Cin PD, Mutter GL and Morton CC (2004) Molecular pathogenesis of uterine smooth muscle tumors from transcriptional profiling. Genes Chromosomes Cancer 40,97–108.
Sadan O, van Iddekinge B, Savage N, Robinson M and Zakut H (1988) Ethnic variation in estrogen and progesterone receptor concentration in leiomyoma and normal myometrium. Gynecol Endocrinol 2,275–282.[Web of Science][Medline]
Skubitz KM and Skubitz AP (2003) Differential gene expression in uterine leiomyoma. J Lab Clin Med 141,297–308.[CrossRef][Web of Science][Medline]
Tsibris JC, Porter KB, Jazayeri A, Tzimas G, Nau H, Huang H, Kuparadze K, Porter GW, O’Brien WF and Spellacy WN (1999) Human uterine leiomyomata express higher levels of peroxisome proliferator-activated receptor gamma, retinoid X receptor alpha and all-trans retinoic acid than myometrium. Cancer Res 59,5737–5744.
Tsibris JC, Segars J, Coppola D, Mane S, Wilbanks GD, O’Brien WF and Spellacy WN (2002) Insights from gene arrays on the development and growth regulation of uterine leiomyomata. Fertil Steril 78,114–121.[CrossRef][Web of Science][Medline]
Varol N, Healey M, Tang P, Sheehan P, Maher P and Hill D (2001) Ten-year review of hysterectomy morbidity and mortality: can we change direction? Aust NZ J Obstet Gynaecol 41,295–302.[Web of Science][Medline]
Wang H, Mahadevappa M, Yamamoto K, Wen Y, Chen B, Warrington JA and Polan ML (2003) Distinctive proliferative phase differences in gene expression in human myometrium and leiomyomata. Fertil Steril 80,266–276.[Web of Science][Medline]
Wei J, Chiriboga L, Mizuguchi M, Yee H and Mittal K (2005a) Expression profile of tuberin and some potential tumorigenic factors in 60 patients with uterine leiomyomata. Mod Pathol 18,179–188.[CrossRef][Web of Science][Medline]
Wei J, Chiriboga L and MK (2005b) Expression profile of the tumorigenic factors in association with tumor size and sex steroid hormone status in uterine leiomyomata. Fertil Steril 84,474–484.
Weston G, Trajstman AC, Gargett CE, Manuelpillai U, Vollenhoven BJ and Rogers PA (2003) Fibroids display an anti-angiogenic gene expression profile when compared with adjacent myometrium. Mol Hum Reprod 9,541–549.
Wilcox LS, Koonin LM, Pokras R, Strauss LT, Xia Z and Peterson HB (1994) Hysterectomy in the United States, 1988–1990. Obstet Gynecol 83,549–555.[Web of Science][Medline]
Wise LA, Palmer JR, Harlow BL, Spiegelman D, Stewart EA, Adams-Campbell LL and Rosenberg L (2004a) Reproductive factors, hormonal contraception and risk of uterine leiomyomata in African-American women: a prospective study. Am J Epidemiol 159,113–123.
Woods MN, Barnett JB, Spiegelman D, Trail N, Hertzmark E, Longcope C and Gorbach SL (1996) Hormone levels during dietary changes in premenopausal African-American women. J Natl Cancer Inst 88,1369–1374.
Submitted on April 14, 2005; resubmitted on August 5, 2005; accepted on August 15, 2005.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. D. Peddada, S. K. Laughlin, K. Miner, J.-P. Guyon, K. Haneke, H. L. Vahdat, R. C. Semelka, A. Kowalik, D. Armao, B. Davis, et al. Growth of uterine leiomyomata among premenopausal black and white women PNAS, December 16, 2008; 105(50): 19887 - 19892. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Peng, J. Laser, G. Shi, K. Mittal, J. Melamed, P. Lee, and J.-J. Wei Antiproliferative Effects by Let-7 Repression of High-Mobility Group A2 in Uterine Leiomyoma Mol. Cancer Res., April 1, 2008; 6(4): 663 - 673. [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]
|
|
|
|
|
|
Y. Zhao, Y. Wen, M. L. Polan, J. Qiao, and B. H. Chen Increased expression of latent TGF-{beta} binding protein-1 and fibrillin-1 in human uterine leiomyomata Mol. Hum. Reprod., May 1, 2007; 13(5): 343 - 349. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||