- Email: [email protected]
Allelic imbalances in endometrial stromal neoplasms: frequent genetic alterations in the nontumorous normal-appearing endometrial and myometrial tissues
Gynecologic Oncology 95 (2004) 662 – 671 www.elsevier.com/locate/ygyno
Allelic imbalances in endometrial stromal neoplasms: frequent genetic alterations in the nontumorous normal-appearing endometrial and myometrial tissues Farid Moinfara,*, Marie-Luise Kremsera, Yan Gao Manb, Kurt Zatloukala, Fattaneh A. Tavassolic, Helmut Denka b
a Institute of Pathology, Medical University of Graz, Austria Department of Gynecologic and Breast Pathology, Armed Forces Institute of Pathology, Washington, DC 20306, USA c Department of Pathology, Yale University, New Haven, CT 06520, USA
Received 3 June 2004 Available online 29 September 2004 This paper is dedicated to Mrs. Lore Saldow and Mr. Udo Saldow
Abstract Objectives. Endometrial stromal sarcoma (ESS) is among the rarest primary malignant tumors of the uterus. The aim of this study was to examine the possibility of loss of heterozygosity (LOH) and microsatellite instability (MIS) in different tissue components of ESS. Methods and materials. Using PCR, we examined DNA extracts from microdissected tissues of 27 uterus samples containing malignant stromal cells of ESS (20 low grade and 3 high grade sarcomas), benign tumor cells of endometrial stromal nodules (ESN, 4 cases) as well as tumor-free myometrial and endometrial tissues close to and distant from the tumors. Normal cervical tissues (epithelial cells, stroma cells) were also microdissected and analyzed. Fifteen polymorphic DNA markers (chromosomes 2p, 3p, 5q, 10q, 11q, 13q, and 17p) were tested to identify possible genetic alterations. Samples from 10 women with prolapsed uteri without any histopathologic abnormalities were also selected as controls. Results. While no genetic alterations could be identified in 12 (44.5%) ESS cases, 15 (55.5%) revealed LOH with at least one polymorphic DNA marker. LOH were found in 3 (100%) high-grade sarcomas, 10 (50%) low-grade ESS, and 2 (50%) benign ESN. Although LOH was found more often in the neoplastic stromal cells, several cases showed concurrent and independent LOH in the tumorfree myometrial or endometrial tissues either close to or distant from the tumors. The most common genetic abnormality (LOH) was observed at PTEN, a tumor suppressor gene located on chromosome 10q. No tumor was associated with microsatellite instability (MSI). The control group without any histologic abnormalities did not show LOH or MSI. Conclusions. The frequent occurrence of LOH and the lack of MSI suggest that loss of function(s) of tumor suppressor genes and not mismatch repair deficiency plays a key role in the pathogenesis of endometrial stromal neoplasms. The concurrent and independent occurrence of LOH in the stromal tumor cells and the tumor-free and normal-appearing myometrial and endometrial tissues strongly support the concept of genetic alterations in microenvironmental tissues and the interaction(s) between different tissue components in the development and progression of endometrial stromal neoplasms. D 2004 Elsevier Inc. All rights reserved. Keywords: Endometrial stromal sarcoma; Loss of heterozygosity; Genetic alterations in microenvironmental tissues; PTEN
Introduction * Corresponding author. Institute of Pathology, Medical University of Graz, Auenbruggerplatz 25, A-8036, Graz, Austria. Fax: +43 316 385 3432. E-mail address: [email protected] (F. Moinfar). 0090-8258/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2004.08.016
Endometrial stromal tumors are among the rarest uterine neoplasms. In most studies, uterine mesenchymal tumors comprise less than 5% of all primary uterine
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
malignant neoplasms [1,2] with endometrial stromal tumors accounting for less than 10% thereof [1,2]. According to the nomenclature of Norris and Taylor [3], most investigators divide these uterine stromal tumors into three groups: (i) endometrial stromal nodule (ESN), a well-circumscribed, expansible benign neoplasm consisting of uniform cells resembling the stromal cells of normal proliferative-phase endometrium; (ii) low-grade endometrial stromal sarcoma (ESS, low grade), an infiltrative tumor with mesenchymal cells cytologically identical to those seen in the ESN but associated with low aggressive behavior, and (iii) high-grade stromal sarcoma, a very aggressive invasive tumor composed of highly atypical and mitotically active cells that may focally show some morphological resemblance to endometrial stromal cells. However, since high-grade endometrial sarcomas usually lack specific differentiation and often bear no histological resemblance to endometrial stroma, it has been proposed that they be designated undifferentiated endometrial sarcoma (UES) [4]. Due to the rarity of endometrial stromal neoplasms, little is known about their epidemiology, pathogenesis, and molecular biology. Previous cytogenetic studies revealed clonal complex and heterogeneous chromosomal abnormalities in ESS including structural rearrangements of chromosomes 1, 3, 6, 7, 13, 14, 15, 17, 19, 20, and 21 [5–8]. Most of these studies, however, dealt with a few number of cases with ESS. A few recent studies [9,10] have also demonstrated that fusion of two zinc finger genes (JAZF1 and JJAZ1) by translocation t(7;17) is present in most low-grade ESS. In the present study, we analyzed a larger number of endometrial stromal tumors consisting of 4 ESN, 20 low-grade ESS, and 3 highgrade or undifferentiated sarcomas (UES). The aim of our study was to (i) examine possible loss of heterozygosity (LOH) and microsatellite instability (MSI) in endometrial stromal neoplasms and (ii) see whether genetic alterations in cases with sarcomas are also identifiable in tumor-free and normal-appearing endometrial and myometrial tissues.
Material and methods Samples from 27 women with endometrial stromal neoplasms consisting of 4 ESN, 20 low-grade ESS, and 3 undifferentiated endometrial sarcoma (UES) were selected from the files of the Armed Forces Institute of Pathology (AFIP), Washington, DC, and the Institute of Pathology, University of Graz, Austria. Endometrial stromal nodules (ESN) were sharply circumscribed without infiltration into the myometrium. All cases with endometrial sarcomas displayed extensive myometrial infiltration. The distinction between low- and high-grade sarcomas was made based on morphological criteria using the degree of cytologic atypia (mild vs. severe nuclear atypia) and
663
number of mitotic figures (less vs. more than 10 mitotic figures per 10 high-power microscopic fields) and tumor cell necrosis. Cases with highly anaplastic and mitotically active malignant mesenchymal cells that exhibited at least some (focal) morphological resemblance to endometrial stromal cells were considered as undifferentiated endometrial sarcomas. In addition, 10 uteri with elongatio colli (uterus prolaps) were selected as a control group. These cases were not associated with any recognizable endometrial or myometrial histopathologic abnormalities. In each case, the neoplastic tissue, tumor-free, and normal-appearing myometrium and endometrium close to and at a distance (at least 15-mm distance) from the tumor margin, normal-appearing ectocervical (squamous) epithelium, and cervical stromal cells were manually microdissected. Tumor-free adnexal tissues (ovary, fallopian tubes) or tumor-free lymph nodes were also microdissected to serve as an internal control. Microdissection and DNA extraction were carried out as previously described [11,12]. To avoid artifactual patterns (LOH or MSI artefacts), a minimum of 10.0 ng template DNA content was required before performing PCR amplification. Using PCR, we examined DNA extracts from the microdissected tissues with 15 polymorphic DNA markers on chromosomes 2p, 3p, 5q, 10q, 11q, 13q, and 17p, which are known for a high frequency of allelic imbalances in uterine leiomyosarcomas [13], endometrial adenocarcinomas (endometrioid type) [14]), and soft tissue sarcomas [15]. Gene Amp PCR kits, Taq DNA polymerase, and DNA size markers were obtained from Qiagen (Hilden, Germany). Fluorescent-labeled polymorphic DNA markers including BAT 26 (2p16), D3S1300 (3p14.2-21.1), TGF-
Table 1 LOH with at least one polymorphic marker in different tissue components Case number
ESS (ESN)
3 4 (ESN) 8 9 10 13 14 15 (ESN) 17 19 20 21 22 24 27
. . . o
.
o o
.
o o o
. . . .
MCT o o o o
. . . .
o
.
o o o o o
NM
NE
.
. .
o o
.
o o o
.
o o o o o o o
o o o o o o o o
NCE
NCS
o o o o o o o o o o
o o o o o o o o
.
o
.
.
.
o o o o
o o o o
o o o o
.: loss of heterozygosity (LOH). o: retention of heterozygosity (ROH). ESS: endometrial stromal sarcoma (including undifferentiated sarcoma); ESN: benign endometrial stromal nodule; MCT: myometrium close to the tumor; NM: normal myometrium; NE: normal endometrium; NCE: normal cervical squamous epithelium; NCS: normal cervical stroma.
664
Table 2 Distribution of LOH (.) with 15 polymorphic DNA markers BAT 26
D3S1300
TGF-BIIR
D5S107
D10S185
D10S221
D10S610
D10S2491
D11S1311
WT1
D13S153
D14S267
D17S579
D17S785
TP53.15
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
n n n n n n x n n n o x n n n n o o o n n n n n o n n
o o o
n o o n o n n n n n n n n n n n o o n n o n n n n n n
o o
x o o x o o o o o
o o o o o o x o o
n o
o x o x o o o o
o n o o o o x o o o o o o o o o x o o
o o o n o o x o o o o o o o o o o o o o
o o o o o o x o o o o o o o o o o o o o o o x o o o o
o o o o o o o o o o o o o
o o o o o o n o o x o o o o o o
o o o o o o n o o x o o o o
o o o o o o x n n o o o o o o o o o o o o
.
o o x
.
o x o x o x x o o x o x x o o o o o x
.
o o o n
. o o o o o o o o x o x o o o o o o o o
.
o o o x o o o o o o o o o
.
o o o
.
o o o o o o o o
. .
o o o o o o
. .
o n o o x o o o
.
o
.
o o o o x o o o o o o o
.
o o o o o o o o o o o o o o
.
o o o
. .: loss of heterozygosity (LOH); o: retention of heterozygosity (ROH); n: homozygosity; x: noninformative.
.
o
.
o o o o o
o o o o o o
.
. .
o o o o o x
.
o o o o o
.
o o o o o o o x o o
.
o o x o o o o x o o o o
.
o o o o o
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
Case number
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671 Table 3 Occurrence of LOH in different tissue components of endometrial stromal neoplasms Case number
Diagnosis
DNA marker
Location of alteration
3 3 4 4 8 8 9 10 10 10 13 14 15 15 15 17 19 20 20 21 22 22 22 24 24 27
LESS LESS ESN ESN LESS LESS LESS UES UES UES UES LESS ESN ESN ESN LESS UES LESS LESS LESS LESS LESS LESS LESS LESS LESS
D5S107 D10S610 D3S1300 D10S610 D3S1300 D5S107 D10S2491 D10S185 D10S221 D10S185 D10S610 D14S267 D17S785 D14S267 D10S610 D17S579 D10S221 D10S221 D11S1311 WT1 D14S267 D11S1311 TP53 D10S185 D10S2491 D10S221
ESS NE & NM ESN NE ESS ESS NM UES MCT MCT MCT MCT ESN MCT MCT & NM NSC MCT NE & NCE NSC ESS ESS ESS ESS ESS ESS ESS
LESS: low-grade ESS; UES: undifferentiated endometrial sarcoma; ESN: endometrial stromal nodule; MCT; myometrium close to the tumor; NE: normal endometrium (distant from the tumor); NM: normal myometrium (distant from the tumor); NSC: normal stroma of the cervix; NCE: normal cervical epithelium.
beta IIR (3p22), D5S107 (5q11.2-13.3), D10S221 (10q2526), D10S610 (10q25), D10S185 (10q 23-24), D10S2491 (10q23), D10S1765 (10q22-23), WT1 (11q13), D11S1311 (11q21-23.2), D13S153 (13q14.1-14.3), D14S267 (14q32), D17S785 (17q25), TP 53.15 (17p13.1) were purchased from MWG Biotech-AG (Mqnchen, Germany). PCR amplification was carried out in a thermal cycler (Eppendorf Master Cycler Gradient, Wesseling-Berzdorf, Germany) at the following settings: after denaturation at 928C for 12 min, samples were amplified for to 40–45 cycles at 928C, 51–628C, and 728C each for 1 min, with a final extension at 728C for 7 min. Amplified PCR products were subjected to electrophoresis on 7% polyacrylamide gels (Amersham Biosciences, New York, U.S.A), and the signals detected with a DNA sequencer (ALF Express, Amersham Biosciences) according to the manufacturer’s instruction. The data were automatically collected and analyzed using the Fragment Manager Software (Amersham Biosciences). To judge the presence or absence of LOH, two criteria, namely band appearance and allelic ratio, were used. The normal DNA samples (derived from lymph nodes, normal ovary, or fallopian tubes) were used to determine whether the samples were homozygous (one band or one peak only) or heterozygous (two bands or
665
peaks). In a heterozygous sample, LOH was identified when one of the two alleles was completely absent or when the ratio of alleles was clearly different from that of the constitutional DNA. The size of two alleles was determined according to the peaks of greatest height. The values for peak area of the two alleles in the paired normal control (N) and tumor (N) samples were used to determine allele loss. The ratio of alleles was calculated for each normal control and tumor sample, and then the tumor ratio was divided by the normal ratio (T1:T2/N1:N2). A ratio of less than 0.5 was considered as LOH [16]. Microsatellite instability (MSI) was defined as the presence of additional bands (peaks) or abnormal shifting of bands (peaks) compared to normal control tissue. After recognition of LOH at the PTEN locus, we examined all cases that were associated with LOH (10 cases) for the possibility of mutations of the exons of the PTEN gene. PCR-based mutational analysis was performed using primers derived from intronic sequences to amplify each of the nine PTEN exons [17,18]. PCR amplification was performed in 60-Al reaction volumes containing 40 ng genomic DNA, 1.5–3 mM MgCl2, 10 mM Tris–HCl (pH 9.2), 75 mM KCl, 100 AM each of dGTP, dATP, dTTP, and dCTP, 0.5–1 AM of each primer, and 1.5 units of Taq polymerase (Qiagen). All exons were PCR amplified as follows: 40 cycles consisting of 1 min at 958C; 1 min at 598C for exon 1, 518C for exons 2, 4, 9; 528C for exon 3; or 548C for exons 6, 7, 8; and 568C for exon 5; and 1 min at 728C, followed by a single 7-min extension at 728C. PCR fragments were purified with QIAquick PCR Purification Kit (Qiagen) according to the manufacturer’s instructions. For each exon, a negative control without DNA template was processed in parallel and sequenced to eliminate the possibility of contamination. Sequencing was performed directly on PCR products with both primers using the CEQDTCS-Quick Start Kit from Beckman Coulter (New York, USA). The products were analyzed on a Ceq 8000 Genetic Analysis System (Beckman Coulter). The data were aligned with original sequences obtained from the Genome Database (Johns Hopkins University, http://gdbwww.gdb.org/gdb).
Results Endometrial stromal nodule (ESN) Loss of heterozygosity (LOH) in ESN was observed in two out of four cases. One case showed LOH in the neoplastic stromal cells (D3S1300) as well as in nontumorous, histologically normal-appearing endometrium (D10S610). The second case revealed LOH in the tumor cells (D17S785), in myometrium close to the tumor (D14S267, D10S610), and in normal myometrium distant from the tumor (D10S610) (Tables 1 and 3). Retention of heterozygosity (ROH) in a case with ESN is shown in Fig.
666
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
chymal cells. All three cases, however, revealed LOH (D10S221, D10S185, D10S610) in nontumorous myometrium in proximity to the malignant tumors (Table 3). Figs. 5 and 6 show LOH in two cases of UES. No MSI could be identified in undifferentiated sarcomas. Mutational analysis of the PTEN gene All 10 cases that were associated with LOH of the PTEN gene were examined for mutations of nine exons of the PTEN gene. However, none of the cases was associated with mutation.
Discussion While some previous studies have shown LOH in uterine leiomyosarcomas and carcinosarcomas [13,14], the possibility of LOH or MSI in endometrial stromal neo-
Fig. 1. Retention of heterozygosity (ROH) of D3S1300 in an endometrial stromal nodule. Stromal nodule (SN), myometrium close to the tumor (MCT), normal endometrium (NE), normal cervical epithelium (NCE).
1. None of the examined ESN cases was associated with MSI. Low-grade endometrial stromal sarcoma (ESS, low grade) Ten of 20 low-grade ESS revealed LOH with at least one polymorphic marker. LOH was found in different areas including neoplastic mesenchymal cells (D5S107, D3S1300, D11S1311, D14S267, TP53, WT1, D10S185, D10S2491, D10S221) (Figs. 2 and 3), normal endometrium (Fig. 4) (D10S610, D10S221), normal myometrium (D10S610, D10S2491) (Fig. 4), and myometrium close to the tumor (D14S267). Tables 1–3 show the distribution of LOH in different tissue compartments. In two cases, LOH (D17S579, D11S1311) were also identified in normal cervical stromal cells distant from the malignant tumors. Furthermore, one case showed LOH (D10S221) exclusively in both normal cervical and normal endometrial tissues (Table 3). None of the cases revealed MSI. Undifferentiated endometrial sarcoma (UES) LOH was identified in all three cases examined. One case showed LOH (D10S185) associated with malignant mesen-
Fig. 2. Loss of heterozygosity (LOH) using polymorphic marker D14S267. Normal endometrium (NM), endometrial stromal sarcoma (ESS), myometrium close to the tumor (MCT). While one of the alleles is lost (LOH) in ESS, samples from non-neoplastic tissues of this case do not show alterations.
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
667
Ten out of 15 stromal tumors (66.5%) that were associated with LOH showed genetic changes (LOH) at the PTEN locus; three of them revealed LOH in the neoplastic stromal cells. With regard to the PTEN gene alteration, however, LOH was observed in the surrounding nontumorous myometrial or endometrial tissues in 7 out of 10 cases (70%). Consequently, the results of our study show that the alteration of the PTEN locus occurs more frequently in adjacent nontumorous tissues than that of neoplastic stromal cells. The tumor suppressor gene PTEN (phosphatase and tensin homologue deleted on chromosome 10) is also known as MMAC (mutated in multiple advanced cancers). It is one of the tumor suppressor genes that is commonly inactivated in cancer. Somatic mutation, deletion, and/or epigenetic inactivation of PTEN has been reported in human endometrial [17,18], brain [19], thyroid [20], breast [21], prostate [22], and skin cancers [23,24]. Previous studies demonstrated that PTEN has, at least, two major biochemical functions [23,24]: It has lipid phosphatase and protein phosphatase activity. The lipid phosphatase activity down-
Fig. 3. LOH at focus D10S2491 in endometrial stromal sarcoma (ESS, low grade) but not in samples from normal endometrium (NE) or normal myometrium (NM).
plasms has yet not been investigated. Several previous studies [5–8] have shown complex cytogenetic abnormalities in endometrial stromal sarcomas. A few recent studies [9,10] have also demonstrated a nonrandom translocation t(7;17) with fusion of two zinc finger genes (JAZF1, JJAZ1) in benign and malignant endometrial stromal neoplasms. Our study deals with a larger number of these very rare uterine mesenchymal tumors, investigating allelic imbalances such as LOH and MSI in the tumor cells as well as in the nontumorous surrounding uterine tissues. This study shows that LOH is a frequent event in endometrial stromal neoplasms. All three cases of undifferentiated endometrial sarcoma revealed LOH with at least one polymorphic DNA marker. Moreover, 50% of the examined cases with low-grade ESS exhibited LOH with at least one polymorphic marker. LOH was also identified in two out of four ESN. Our study is the first to show frequent occurrence of LOH not only in neoplastic stromal cells but also in tumorfree surrounding myometrial and endometrial tissues either close to or distant from the tumor (Tables 1 and 3). Indeed, 10 out of 15 benign and malignant stromal tumors (66.5%) showed LOH in normal-appearing and tumor-free endometrial or myometrial tissues (Fig. 7). Although in some cases, the observed genetic changes occurred simultaneously in ESS and surrounding endometrial or myometrial tissues, several tumors showed independent genetic alterations in different tissue components (Table 1).
Fig. 4. Retention of heterozygosity (ROH) in endometrial stromal sarcoma (ESS, low grade) and myometrium close the tumor (MCT) by using DNA polymorphic marker D10S610. Note that this focus is lost (LOH) in normal endometrium (NE) and normal myometrium (NM).
668
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
but no mutations were found in the genomic sequence of PTEN exons 5 to 9 in any of the fibrosarcomas [26]. The results of our study showed loss of one allele at the PTEN locus in 37% of the cases. Sequencing of the exons 1 to 9 in the present study, however, did not detect a mutation. One might consider several possible explanations for the occurrence of LOH of the PTEN gene without apparent inactivation of the retained allele. One possibility that should be taken into account is that another tumor suppressor gene is closely linked to PTEN and represents the actual target. In some human tumors, the closely positioned genes MINPP1 and BMPR1A have been found to be mutated in a subset of PTEN mutation-negative tumors [27,28]. A second possibility may be that the retained PTEN allele is inactivated by a mechanism other than mutation of the coding sequence, such as promoter mutation or epigenetic silencing. Promotor methylation of PTEN has been reported in human endometrial [29] and prostate carcinomas [30] and may be responsible for PTEN inactivation. A third possibility to be considered is that whenever there is loss of one PTEN allele, the activity of the remaining allele may be insufficient to
Fig. 5. LOH at D10S221 identified only in myometrium close to the tumor (MCT). Note the retention of heterozygosity (ROH) in endometrial stromal sarcoma (undifferentiated endometrial sarcoma), and in other tissue components such as normal myometrium (NM) and normal endometrium (NE).
regulates AKT activity by negative regulation of the growthfactor-induced phosphatidylinositol-3 kinase (PI3K) pathway, and up-regulates proapoptotic mechanisms such as caspases. The protein phosphatase activity of PTEN is involved in the inhibition of focal adhesion, in cell spreading and migration, as well as in the inhibition of growthfactor-stimulated MAPK signaling. Recent studies have suggested that PTEN blocks hyaluronic-acid-induced matrix metalloproteinase-9 secretion and invasion through its protein phosphatase activity [19]. Therefore, the combined effects of the PTEN lipid and protein phosphatase activities may result in aberrant cell growth and escape from apoptosis, as well as abnormal cell spreading and migration of neoplastic cells [23]. While LOH at the PTEN locus frequently occurs in certain types of sarcomas, mutation of PTEN is a very rare event [25]. Indeed, a recent study showed complete loss of one allele of PTEN in more than 60% of DMBA-induced fibrosarcomas. DNA sequencing was performed to investigate whether the remaining PTEN allele was inactivated
Fig. 6. Using polymorphic marker D10S185, LOH is observed in myometrium close to the tumor (MCT) and in endometrial stromal sarcoma (undifferentiated sarcoma). No genetic abnormality in normal myometrium (NM) or in normal cervical epithelium (NCE).
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
669
Fig. 7. Using DNA polymorphic marker D5S107, LOH is identified in myometrium close to the tumor (MCT), normal cervical stroma (NCS), and normal cervical squamous epithelium (NCE). Note the retention of hetrozygosity in endometrial stromal sarcoma (ESS), normal endometrium (NE), and normal myometrium (NM).
prevent tumor formation in a dose-dependent manner. Recent studies have suggested that PTEN might be haploinsufficient [31–33]. In fact, haploinsufficiency might explain why there are frequent allelic imbalances in the PTEN region, whereas actual PTEN mutations seem to be very rare, at least in certain types of tumors such as prostate and breast carcinomas [34]. The second most common locus altered in endometrial stromal neoplasms was D14S267 (located at 14q32). In two cases, these changes were only found in the myometrium close to the tumor. In one case, LOH was identified in the stromal cells of a low-grade ESS unassociated with genetic changes in the surrounding nontumorous myometrial or endometrial tissues. The locus D14S267 harbors a putative tumor suppressor gene commonly lost in a variety of malignant tumors such as uterine leimyosarcoma [13], colorectal [35], ovarian [36], as well as esophageal carcinomas [37]. Two of 18 informative cases in our study showed LOH at D3S 1300. This locus harbors the tumor suppressor gene FHIT that encompasses FRA3B, the most common fragile site in the human genome. LOH at D3S1300 occurred in benign neoplastic stromal cells of ESN in one case and in stromal cells of low-grade ESS in another. Several studies have shown LOH at the FHIT gene in different solid human tumors including colorectal [38], breast [11,12], endometrial [39], and uterine cervical [40] carcinomas. Other tumor suppressor genes that were found in our study to be rarely altered include P53 (tumor cells of one
low grade ESS), WT1 (tumor cells of one low grade ESS), and BRCA1 (neoplastic cells of an ESN). These tumor suppressor genes have been shown to play important roles for the development of a variety of malignant tumors including ovarian carcinomas [41] and osteosarcomas [42]. In contrast to endometrial stromal tumors, none of the 10 prolapsed uteri without histopathologic abnormalities revealed genetic changes as revealed by LOH or MSI. These findings support the hypothesis that endometrial stromal sarcoma is not just a localized genetic disease of the endometrium but rather represents a bmicroenvironmentalQ genetic alteration occurring in different tissue components of the uterus. Our findings suggest that a tissue field of somatic genetic alterations (LOH but not MSI) precedes the histopathological phenotypic changes of endometrial stromal sarcoma. Loss of functions of certain tumor suppressor genes such as PTEN in surrounding nontumorous uterine tissues could influence and facilitate tumor proliferation, cell spreading, and invasion of malignant endometrial stromal cells. The observed genetic alterations in the surrounding nontumorous uterine tissues and possible dynamic interaction(s) between neoplastic stromal cells and non-neoplastic endometrial and/or myometrial tissues could be crucial for the tumorigenesis, differentiation, and biological behavior of uterine stromal tumors. Interestingly, in a previous study, we observed concurrent and independent genetic alterations (LOH) in the epithelial (cancerous) and stromal (mesenchymal) cells of mammary
670
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
carcinomas [12]. Genetic abnormalities were found in the tumor-free stromal cells either close to invasive or noninvasive ductal carcinomas or even in the morphologically normal-appearing stromal cells distant from noninvasive or invasive carcinomas. The results of our previous study favored the concept of reciprocal stromal–epithelial interaction in mammary tumorigenesis. In summary, the present study is the first to examine allelic imbalances (LOH, MSI) in a larger number of endometrial stromal neoplasms. While LOH is a common event, MSI could not be identified in these neoplasms. The frequent occurrence of LOH and the lack of MSI suggest that loss of functions of tumor suppressor genes and not mismatch repair insufficiency play a key role in the pathogenesis of endometrial stromal neoplasms. More importantly, LOH occurs concurrently and independently in neoplastic stromal cells and in non-neoplastic surrounding endometrial/or myometrial tissues, raising the intriguing possibility of dynamic and reciprocal interaction (bcrosstalkQ) between different tissue components of the uterus. The interactions between cancer cells and their micro- and macroenviroment may promote tumor growth and differentiation. Investigation of this process might provide new insights into the mechanisms of tumorigenesis and could also lead to new therapeutic strategies.
[10]
[11]
[12]
[13]
[14]
[15] [16]
[17]
[18]
[19]
Acknowledgment This study was supported by a Lore Saldow research fund.
[20]
[21]
References [22] [1] Koss LG, Spiro RH, Brunschwig A. Endometrial stromal sarcoma. Surg Gynecol Obstet 1965;121:531 – 7. [2] Aaro LA, Symmonds RE, Dockertly MB. Sarcoma of the uterus. A clinical and pathologic study of 117 cases. Am J Obstet Gynecol 1966;94:101 – 9. [3] Norris HJ, Taylor HB. Mesenchymal tumors of the uterus. I. A clinical and pathological study of 53 endometrial stromal tumors. Cancer 1966;19:755 – 66. [4] Tavassoli FA, Devilee P. World Health Organization classification of tumours. Pathology & Genetics of Tumors of the Breast and Female Genital Organs. Lyon7 IARCP Press; 2003. [5] Sreekantaiah C, Li FP, Weidner N, Sandberg A. An endometrial stromal sarcoma with clonal cytogenetic abnormalities. Cancer Genet Cytogenet 1991;55:163 – 6. [6] Dal Cin P, Aly MS, De Wever I, Moerman P, Van Den Berghe H. Endometrial stromal sarcoma t(7;17)(p15–21;q12–21) is a common non-random chromosome change. Cancer Genet Cytogenet 1992; 63:43 – 6. [7] Laxman R, Currie JL, Kurmann RJ, Dudzinski M, Griffin CA. Cytogenetic profile of uterine sarcomas. Cancer 1993;71:1283 – 8. [8] Cheung AN, Tin VP, Ngan HY, Chung LP, Khoo US. Interphase cytogenetic study of endometrial stromal sarcoma by chromosome in situ hybridization. Mod Pathol 1996;9:910 – 8. [9] Koontz JI, Soreng AL, Nucci M, Kuo FC, Pauwels P, Van den Berghe H, Cin PD, Fletcher JA, Sklar J. Frequent fusion of the JAZF1 and
[23] [24]
[25]
[26]
[27]
[28]
[29]
[30]
JJAZ1 genes in endometrial stromal tumors. Proc Natl Acad Sci 2001;98:6348 – 53. Huang HY, Ladanyi M, Soslow RA. Molecular detection of JAZF1 and JJAZ1 gene fusion in endometrial stromal neoplasms with classic and variant histology: evidence for genetic heterogeneity. Am J Surg Pathol 2004;28:224 – 32. Moinfar F, Man YG, Bratthauer GL, Ratschek M, Tavassoli FA. Genetic abnormalities in mammary ductal intraepithelial neoplasia-flat type (bclinging ductal carcinoma in situQ): a simulator of normal mammary epithelium. Cancer 2000;88:2072 – 81. Moinfar F, Man YG, Arnould L, Bratthauer GL, Ratschek M, Tavassoli FA. Concurrent and independent genetic alterations in the stromal and epithelial cells of mammary carcinoma: implications for tumorigenesis. Cancer Res 2000;60:2562 – 6. Zhai YL, Nikaido T, Orii A, Horiuchi A, Toki T, Fujii S. Frequent occurrence of loss of heterozygosity among tumor suppressor genes in uterine leiomyosarcoma. Gynecol Oncol 1999;75:453 – 9. Tritz D, Pieretti M, Turner S, Powell D. Loss of heterozygosity in usual and special variant carcinomas of the endometrium. Hum Pathol 1997;28:607 – 12. Katsuyasu S, Masaki O, Miki H. Microsatellite alterations in various sarcomas in Japanese patients. J Orthoo Sci 1999;4:223 – 30. Cawkwell L, Bell SM, Lewis FA, Dixon MF, Taylor GR, Quirke P. Rapid detection of allele loss in colorectal tumors using microsatellites and fluorescent DNA technology. Br J Cancer 1993;67:1262 – 7. Tashiro H, Blazes MS, Wu R, et al. Mutation in PTEN are frequent in endometrial carcinoma but rare in other common gynaecological malignancies. Cancer Res 1997;57:3935 – 40. Bussaglia E, Rio E, Matias-Guiu X, Prat J. PTEN mutations in endometrial carcinomas: a molecular and clinicopathologic analysis of 38 cases. Hum Pathol 2000;31:312 – 7. Park MJ, Kim MS, Park IC, et al. PTEN suppresses hyaluronic acidinduced matrix metalloproteinase-9 expression in U87MG glioblastoma cells through focal adhesion kinase dephosphorylation. Cancer Res 2002;62:6318 – 22. Eng C. Role of PTEn, a lipid phosphatase upsteam effector of protein kinase B, in epithelial thyroid carcinogenesis. Ann N Y Acad Sci 2002;968:213 – 21. Kurose K, Gilly K, Matsumoto S, Watson PH, Zhou XP, Eng C. Frequent somatic mutations in PTEN and TP53 are mutually exclusive in the stroma of the breast carcinomas. Net Genet 2002;32:355 – 7. Deocampo ND, Huang H, Tindall DJ. The role of PTEN in the progression and survival of prostate cancer. Minerva Endocrinol 2003;28:145 – 53. WU H, Goel V, Haluska FG. PTEN signalling pathways in melanoma. Oncogene 2002;22:3113 – 22. Stahl JM, Cheung M, Sharma A, Trivedi NR, Shanmugam S, Robertson GP. Loss of PTEN promotes development in malignant melanoma. Cancer Res 2003;63:2881 – 90. Saito T, Oda Y, Kawaguchik T, et al. PTEN/MMAC1 gene mutation is a rare event in soft tissue sarcomas without specific balanced translocations. Int J Cancer 2003;104:175 – 8. Sjfling A, Samuelson E, Adamovic T, Behboudi A, Rfme D, Levan G. Recurrent allelic imbalance at the rat pten locus in DMBAinduce3d fibrosarcoma. Genes Chromosomes Cancer 2003;36:70 – 9. Gimm O, Chi H, Dahia PL, et al. Somatic mutation and germline variants of MINPP1, a phosphatase gene located in proximity to PTEN on 10q23.3, in follicular thyroid carcinomas. J Clin Endocrinol 2001;86:1801 – 5. Zhou XP, Woodford-Richens K, Lehtonen R, et al. Germline mutations in BMPR1A/ALK3 cause a subset of cases of juvenile polyposis syndrome and of Cowden and Bannayan–Riley–Ruvalcaba syndromes. Am J Hum Genet 2001;69:704 – 11. Salvesen HB, McDonald N, Ryan A, et al. PTEN methylation is associated with advanced stage and microsatellite instability in endometrial carcinoma. Int J Cancer 2001;91:22 – 6. Whang YE, Wu X, Suzuki H, et al. Inactivation of the tumor
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
[31]
[32]
[33] [34] [35]
[36]
suppressor PTEN/MMMAC1 in advanced human prostate cancer through loss of expression. Proc Natl Acad Sci 1998;95:5246 – 50. Di Cristofano A, Pesce B, Cordon-Cardo C, Pandolfi P. PTEN is essential for embryonic development and tumor suppression. Nat Genet 1998;19:348 – 55. Kwabi-Addo B, Giri D, Schmidt K, et al. Haploinsufficiency of the Pten tumor suppressor gene promotes prostate cancer progression. Proc Natl Acad Sci 2001;98:11563 – 8. Dong JT. Chromosomal deletions and tumor suppressor genes in prostate cancer. Cancer Metastasis Rev 2001;20:173 – 93. Di Cristofano A, Pandolfi PP. The multiple roles of PTEN in tumor suppression. Cell 2002;100:387 – 90. Bando T, Kato Y, Ihara Y, Yamagishi F, Tsukada K, Isobe M. Loss of heterozygosity of 14q32 in colorectal carcinoma. Cancer Genet Cytogenet 1999;111:161 – 5. Bandera CA, Takahashi H, Behbakht K, et al. Deletion mapping of two potential chromosome 14 tumor suppressor gene loci in ovarian carcinoma. Cancer Res 1997;57:513 – 5.
671
[37] Ihara Y, Kato Y, Bando T, et al. Allelic imbalance of 14q32 in esophageal carcinoma. Cancer Genet Cytogenet 2002;135:177 – 81. [38] Petursdottir TE, Hafsteinsdottir SH, Jonasson JG, et al. Loss of heterozygosity at the FHIT gene in different solid human tumors and its association with survival in colorectal cancer patients. Anticancer Res 2002;22:3205 – 12. [39] Ozaki K, Enomoto T, Yoshino K, et al. Fhit alterations in endometrial carcinoma and hyperplasia. Int J Cancer 2000;85:306 – 12. [40] Acevedo CM, Henriquez M, Emmert-Buck MR, Chuagui RF. Loss of heterozygosity on chromosome arms 3p and 6q in microdissected adenocarcinomas of the uterine cervix and adenocarcinoma in situ. Cancer 2002;94:793 – 802. [41] Quezdo MM, Moskaluk CA, Bryant B, Mills SE, Merino MJ. Incidence of loss of heterozygosity at p53 and BRCA1 loci in serous surface carcinoma. Hum Pathol 1999;30:203 – 7. [42] Sztan M, Papai Z, Szendroi M, Looij A, Olah E. Allelic losses from chromosome 17 in human osteosarcomas. Pathol Oncol Res 1997;3:115 – 20.
Allelic imbalances in endometrial stromal neoplasms: frequent genetic alterations in the nontumorous normal-appearing endometrial and myometrial tissues Farid Moinfara,*, Marie-Luise Kremsera, Yan Gao Manb, Kurt Zatloukala, Fattaneh A. Tavassolic, Helmut Denka b
a Institute of Pathology, Medical University of Graz, Austria Department of Gynecologic and Breast Pathology, Armed Forces Institute of Pathology, Washington, DC 20306, USA c Department of Pathology, Yale University, New Haven, CT 06520, USA
Received 3 June 2004 Available online 29 September 2004 This paper is dedicated to Mrs. Lore Saldow and Mr. Udo Saldow
Abstract Objectives. Endometrial stromal sarcoma (ESS) is among the rarest primary malignant tumors of the uterus. The aim of this study was to examine the possibility of loss of heterozygosity (LOH) and microsatellite instability (MIS) in different tissue components of ESS. Methods and materials. Using PCR, we examined DNA extracts from microdissected tissues of 27 uterus samples containing malignant stromal cells of ESS (20 low grade and 3 high grade sarcomas), benign tumor cells of endometrial stromal nodules (ESN, 4 cases) as well as tumor-free myometrial and endometrial tissues close to and distant from the tumors. Normal cervical tissues (epithelial cells, stroma cells) were also microdissected and analyzed. Fifteen polymorphic DNA markers (chromosomes 2p, 3p, 5q, 10q, 11q, 13q, and 17p) were tested to identify possible genetic alterations. Samples from 10 women with prolapsed uteri without any histopathologic abnormalities were also selected as controls. Results. While no genetic alterations could be identified in 12 (44.5%) ESS cases, 15 (55.5%) revealed LOH with at least one polymorphic DNA marker. LOH were found in 3 (100%) high-grade sarcomas, 10 (50%) low-grade ESS, and 2 (50%) benign ESN. Although LOH was found more often in the neoplastic stromal cells, several cases showed concurrent and independent LOH in the tumorfree myometrial or endometrial tissues either close to or distant from the tumors. The most common genetic abnormality (LOH) was observed at PTEN, a tumor suppressor gene located on chromosome 10q. No tumor was associated with microsatellite instability (MSI). The control group without any histologic abnormalities did not show LOH or MSI. Conclusions. The frequent occurrence of LOH and the lack of MSI suggest that loss of function(s) of tumor suppressor genes and not mismatch repair deficiency plays a key role in the pathogenesis of endometrial stromal neoplasms. The concurrent and independent occurrence of LOH in the stromal tumor cells and the tumor-free and normal-appearing myometrial and endometrial tissues strongly support the concept of genetic alterations in microenvironmental tissues and the interaction(s) between different tissue components in the development and progression of endometrial stromal neoplasms. D 2004 Elsevier Inc. All rights reserved. Keywords: Endometrial stromal sarcoma; Loss of heterozygosity; Genetic alterations in microenvironmental tissues; PTEN
Introduction * Corresponding author. Institute of Pathology, Medical University of Graz, Auenbruggerplatz 25, A-8036, Graz, Austria. Fax: +43 316 385 3432. E-mail address: [email protected] (F. Moinfar). 0090-8258/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2004.08.016
Endometrial stromal tumors are among the rarest uterine neoplasms. In most studies, uterine mesenchymal tumors comprise less than 5% of all primary uterine
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
malignant neoplasms [1,2] with endometrial stromal tumors accounting for less than 10% thereof [1,2]. According to the nomenclature of Norris and Taylor [3], most investigators divide these uterine stromal tumors into three groups: (i) endometrial stromal nodule (ESN), a well-circumscribed, expansible benign neoplasm consisting of uniform cells resembling the stromal cells of normal proliferative-phase endometrium; (ii) low-grade endometrial stromal sarcoma (ESS, low grade), an infiltrative tumor with mesenchymal cells cytologically identical to those seen in the ESN but associated with low aggressive behavior, and (iii) high-grade stromal sarcoma, a very aggressive invasive tumor composed of highly atypical and mitotically active cells that may focally show some morphological resemblance to endometrial stromal cells. However, since high-grade endometrial sarcomas usually lack specific differentiation and often bear no histological resemblance to endometrial stroma, it has been proposed that they be designated undifferentiated endometrial sarcoma (UES) [4]. Due to the rarity of endometrial stromal neoplasms, little is known about their epidemiology, pathogenesis, and molecular biology. Previous cytogenetic studies revealed clonal complex and heterogeneous chromosomal abnormalities in ESS including structural rearrangements of chromosomes 1, 3, 6, 7, 13, 14, 15, 17, 19, 20, and 21 [5–8]. Most of these studies, however, dealt with a few number of cases with ESS. A few recent studies [9,10] have also demonstrated that fusion of two zinc finger genes (JAZF1 and JJAZ1) by translocation t(7;17) is present in most low-grade ESS. In the present study, we analyzed a larger number of endometrial stromal tumors consisting of 4 ESN, 20 low-grade ESS, and 3 highgrade or undifferentiated sarcomas (UES). The aim of our study was to (i) examine possible loss of heterozygosity (LOH) and microsatellite instability (MSI) in endometrial stromal neoplasms and (ii) see whether genetic alterations in cases with sarcomas are also identifiable in tumor-free and normal-appearing endometrial and myometrial tissues.
Material and methods Samples from 27 women with endometrial stromal neoplasms consisting of 4 ESN, 20 low-grade ESS, and 3 undifferentiated endometrial sarcoma (UES) were selected from the files of the Armed Forces Institute of Pathology (AFIP), Washington, DC, and the Institute of Pathology, University of Graz, Austria. Endometrial stromal nodules (ESN) were sharply circumscribed without infiltration into the myometrium. All cases with endometrial sarcomas displayed extensive myometrial infiltration. The distinction between low- and high-grade sarcomas was made based on morphological criteria using the degree of cytologic atypia (mild vs. severe nuclear atypia) and
663
number of mitotic figures (less vs. more than 10 mitotic figures per 10 high-power microscopic fields) and tumor cell necrosis. Cases with highly anaplastic and mitotically active malignant mesenchymal cells that exhibited at least some (focal) morphological resemblance to endometrial stromal cells were considered as undifferentiated endometrial sarcomas. In addition, 10 uteri with elongatio colli (uterus prolaps) were selected as a control group. These cases were not associated with any recognizable endometrial or myometrial histopathologic abnormalities. In each case, the neoplastic tissue, tumor-free, and normal-appearing myometrium and endometrium close to and at a distance (at least 15-mm distance) from the tumor margin, normal-appearing ectocervical (squamous) epithelium, and cervical stromal cells were manually microdissected. Tumor-free adnexal tissues (ovary, fallopian tubes) or tumor-free lymph nodes were also microdissected to serve as an internal control. Microdissection and DNA extraction were carried out as previously described [11,12]. To avoid artifactual patterns (LOH or MSI artefacts), a minimum of 10.0 ng template DNA content was required before performing PCR amplification. Using PCR, we examined DNA extracts from the microdissected tissues with 15 polymorphic DNA markers on chromosomes 2p, 3p, 5q, 10q, 11q, 13q, and 17p, which are known for a high frequency of allelic imbalances in uterine leiomyosarcomas [13], endometrial adenocarcinomas (endometrioid type) [14]), and soft tissue sarcomas [15]. Gene Amp PCR kits, Taq DNA polymerase, and DNA size markers were obtained from Qiagen (Hilden, Germany). Fluorescent-labeled polymorphic DNA markers including BAT 26 (2p16), D3S1300 (3p14.2-21.1), TGF-
Table 1 LOH with at least one polymorphic marker in different tissue components Case number
ESS (ESN)
3 4 (ESN) 8 9 10 13 14 15 (ESN) 17 19 20 21 22 24 27
. . . o
.
o o
.
o o o
. . . .
MCT o o o o
. . . .
o
.
o o o o o
NM
NE
.
. .
o o
.
o o o
.
o o o o o o o
o o o o o o o o
NCE
NCS
o o o o o o o o o o
o o o o o o o o
.
o
.
.
.
o o o o
o o o o
o o o o
.: loss of heterozygosity (LOH). o: retention of heterozygosity (ROH). ESS: endometrial stromal sarcoma (including undifferentiated sarcoma); ESN: benign endometrial stromal nodule; MCT: myometrium close to the tumor; NM: normal myometrium; NE: normal endometrium; NCE: normal cervical squamous epithelium; NCS: normal cervical stroma.
664
Table 2 Distribution of LOH (.) with 15 polymorphic DNA markers BAT 26
D3S1300
TGF-BIIR
D5S107
D10S185
D10S221
D10S610
D10S2491
D11S1311
WT1
D13S153
D14S267
D17S579
D17S785
TP53.15
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
n n n n n n x n n n o x n n n n o o o n n n n n o n n
o o o
n o o n o n n n n n n n n n n n o o n n o n n n n n n
o o
x o o x o o o o o
o o o o o o x o o
n o
o x o x o o o o
o n o o o o x o o o o o o o o o x o o
o o o n o o x o o o o o o o o o o o o o
o o o o o o x o o o o o o o o o o o o o o o x o o o o
o o o o o o o o o o o o o
o o o o o o n o o x o o o o o o
o o o o o o n o o x o o o o
o o o o o o x n n o o o o o o o o o o o o
.
o o x
.
o x o x o x x o o x o x x o o o o o x
.
o o o n
. o o o o o o o o x o x o o o o o o o o
.
o o o x o o o o o o o o o
.
o o o
.
o o o o o o o o
. .
o o o o o o
. .
o n o o x o o o
.
o
.
o o o o x o o o o o o o
.
o o o o o o o o o o o o o o
.
o o o
. .: loss of heterozygosity (LOH); o: retention of heterozygosity (ROH); n: homozygosity; x: noninformative.
.
o
.
o o o o o
o o o o o o
.
. .
o o o o o x
.
o o o o o
.
o o o o o o o x o o
.
o o x o o o o x o o o o
.
o o o o o
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
Case number
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671 Table 3 Occurrence of LOH in different tissue components of endometrial stromal neoplasms Case number
Diagnosis
DNA marker
Location of alteration
3 3 4 4 8 8 9 10 10 10 13 14 15 15 15 17 19 20 20 21 22 22 22 24 24 27
LESS LESS ESN ESN LESS LESS LESS UES UES UES UES LESS ESN ESN ESN LESS UES LESS LESS LESS LESS LESS LESS LESS LESS LESS
D5S107 D10S610 D3S1300 D10S610 D3S1300 D5S107 D10S2491 D10S185 D10S221 D10S185 D10S610 D14S267 D17S785 D14S267 D10S610 D17S579 D10S221 D10S221 D11S1311 WT1 D14S267 D11S1311 TP53 D10S185 D10S2491 D10S221
ESS NE & NM ESN NE ESS ESS NM UES MCT MCT MCT MCT ESN MCT MCT & NM NSC MCT NE & NCE NSC ESS ESS ESS ESS ESS ESS ESS
LESS: low-grade ESS; UES: undifferentiated endometrial sarcoma; ESN: endometrial stromal nodule; MCT; myometrium close to the tumor; NE: normal endometrium (distant from the tumor); NM: normal myometrium (distant from the tumor); NSC: normal stroma of the cervix; NCE: normal cervical epithelium.
beta IIR (3p22), D5S107 (5q11.2-13.3), D10S221 (10q2526), D10S610 (10q25), D10S185 (10q 23-24), D10S2491 (10q23), D10S1765 (10q22-23), WT1 (11q13), D11S1311 (11q21-23.2), D13S153 (13q14.1-14.3), D14S267 (14q32), D17S785 (17q25), TP 53.15 (17p13.1) were purchased from MWG Biotech-AG (Mqnchen, Germany). PCR amplification was carried out in a thermal cycler (Eppendorf Master Cycler Gradient, Wesseling-Berzdorf, Germany) at the following settings: after denaturation at 928C for 12 min, samples were amplified for to 40–45 cycles at 928C, 51–628C, and 728C each for 1 min, with a final extension at 728C for 7 min. Amplified PCR products were subjected to electrophoresis on 7% polyacrylamide gels (Amersham Biosciences, New York, U.S.A), and the signals detected with a DNA sequencer (ALF Express, Amersham Biosciences) according to the manufacturer’s instruction. The data were automatically collected and analyzed using the Fragment Manager Software (Amersham Biosciences). To judge the presence or absence of LOH, two criteria, namely band appearance and allelic ratio, were used. The normal DNA samples (derived from lymph nodes, normal ovary, or fallopian tubes) were used to determine whether the samples were homozygous (one band or one peak only) or heterozygous (two bands or
665
peaks). In a heterozygous sample, LOH was identified when one of the two alleles was completely absent or when the ratio of alleles was clearly different from that of the constitutional DNA. The size of two alleles was determined according to the peaks of greatest height. The values for peak area of the two alleles in the paired normal control (N) and tumor (N) samples were used to determine allele loss. The ratio of alleles was calculated for each normal control and tumor sample, and then the tumor ratio was divided by the normal ratio (T1:T2/N1:N2). A ratio of less than 0.5 was considered as LOH [16]. Microsatellite instability (MSI) was defined as the presence of additional bands (peaks) or abnormal shifting of bands (peaks) compared to normal control tissue. After recognition of LOH at the PTEN locus, we examined all cases that were associated with LOH (10 cases) for the possibility of mutations of the exons of the PTEN gene. PCR-based mutational analysis was performed using primers derived from intronic sequences to amplify each of the nine PTEN exons [17,18]. PCR amplification was performed in 60-Al reaction volumes containing 40 ng genomic DNA, 1.5–3 mM MgCl2, 10 mM Tris–HCl (pH 9.2), 75 mM KCl, 100 AM each of dGTP, dATP, dTTP, and dCTP, 0.5–1 AM of each primer, and 1.5 units of Taq polymerase (Qiagen). All exons were PCR amplified as follows: 40 cycles consisting of 1 min at 958C; 1 min at 598C for exon 1, 518C for exons 2, 4, 9; 528C for exon 3; or 548C for exons 6, 7, 8; and 568C for exon 5; and 1 min at 728C, followed by a single 7-min extension at 728C. PCR fragments were purified with QIAquick PCR Purification Kit (Qiagen) according to the manufacturer’s instructions. For each exon, a negative control without DNA template was processed in parallel and sequenced to eliminate the possibility of contamination. Sequencing was performed directly on PCR products with both primers using the CEQDTCS-Quick Start Kit from Beckman Coulter (New York, USA). The products were analyzed on a Ceq 8000 Genetic Analysis System (Beckman Coulter). The data were aligned with original sequences obtained from the Genome Database (Johns Hopkins University, http://gdbwww.gdb.org/gdb).
Results Endometrial stromal nodule (ESN) Loss of heterozygosity (LOH) in ESN was observed in two out of four cases. One case showed LOH in the neoplastic stromal cells (D3S1300) as well as in nontumorous, histologically normal-appearing endometrium (D10S610). The second case revealed LOH in the tumor cells (D17S785), in myometrium close to the tumor (D14S267, D10S610), and in normal myometrium distant from the tumor (D10S610) (Tables 1 and 3). Retention of heterozygosity (ROH) in a case with ESN is shown in Fig.
666
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
chymal cells. All three cases, however, revealed LOH (D10S221, D10S185, D10S610) in nontumorous myometrium in proximity to the malignant tumors (Table 3). Figs. 5 and 6 show LOH in two cases of UES. No MSI could be identified in undifferentiated sarcomas. Mutational analysis of the PTEN gene All 10 cases that were associated with LOH of the PTEN gene were examined for mutations of nine exons of the PTEN gene. However, none of the cases was associated with mutation.
Discussion While some previous studies have shown LOH in uterine leiomyosarcomas and carcinosarcomas [13,14], the possibility of LOH or MSI in endometrial stromal neo-
Fig. 1. Retention of heterozygosity (ROH) of D3S1300 in an endometrial stromal nodule. Stromal nodule (SN), myometrium close to the tumor (MCT), normal endometrium (NE), normal cervical epithelium (NCE).
1. None of the examined ESN cases was associated with MSI. Low-grade endometrial stromal sarcoma (ESS, low grade) Ten of 20 low-grade ESS revealed LOH with at least one polymorphic marker. LOH was found in different areas including neoplastic mesenchymal cells (D5S107, D3S1300, D11S1311, D14S267, TP53, WT1, D10S185, D10S2491, D10S221) (Figs. 2 and 3), normal endometrium (Fig. 4) (D10S610, D10S221), normal myometrium (D10S610, D10S2491) (Fig. 4), and myometrium close to the tumor (D14S267). Tables 1–3 show the distribution of LOH in different tissue compartments. In two cases, LOH (D17S579, D11S1311) were also identified in normal cervical stromal cells distant from the malignant tumors. Furthermore, one case showed LOH (D10S221) exclusively in both normal cervical and normal endometrial tissues (Table 3). None of the cases revealed MSI. Undifferentiated endometrial sarcoma (UES) LOH was identified in all three cases examined. One case showed LOH (D10S185) associated with malignant mesen-
Fig. 2. Loss of heterozygosity (LOH) using polymorphic marker D14S267. Normal endometrium (NM), endometrial stromal sarcoma (ESS), myometrium close to the tumor (MCT). While one of the alleles is lost (LOH) in ESS, samples from non-neoplastic tissues of this case do not show alterations.
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
667
Ten out of 15 stromal tumors (66.5%) that were associated with LOH showed genetic changes (LOH) at the PTEN locus; three of them revealed LOH in the neoplastic stromal cells. With regard to the PTEN gene alteration, however, LOH was observed in the surrounding nontumorous myometrial or endometrial tissues in 7 out of 10 cases (70%). Consequently, the results of our study show that the alteration of the PTEN locus occurs more frequently in adjacent nontumorous tissues than that of neoplastic stromal cells. The tumor suppressor gene PTEN (phosphatase and tensin homologue deleted on chromosome 10) is also known as MMAC (mutated in multiple advanced cancers). It is one of the tumor suppressor genes that is commonly inactivated in cancer. Somatic mutation, deletion, and/or epigenetic inactivation of PTEN has been reported in human endometrial [17,18], brain [19], thyroid [20], breast [21], prostate [22], and skin cancers [23,24]. Previous studies demonstrated that PTEN has, at least, two major biochemical functions [23,24]: It has lipid phosphatase and protein phosphatase activity. The lipid phosphatase activity down-
Fig. 3. LOH at focus D10S2491 in endometrial stromal sarcoma (ESS, low grade) but not in samples from normal endometrium (NE) or normal myometrium (NM).
plasms has yet not been investigated. Several previous studies [5–8] have shown complex cytogenetic abnormalities in endometrial stromal sarcomas. A few recent studies [9,10] have also demonstrated a nonrandom translocation t(7;17) with fusion of two zinc finger genes (JAZF1, JJAZ1) in benign and malignant endometrial stromal neoplasms. Our study deals with a larger number of these very rare uterine mesenchymal tumors, investigating allelic imbalances such as LOH and MSI in the tumor cells as well as in the nontumorous surrounding uterine tissues. This study shows that LOH is a frequent event in endometrial stromal neoplasms. All three cases of undifferentiated endometrial sarcoma revealed LOH with at least one polymorphic DNA marker. Moreover, 50% of the examined cases with low-grade ESS exhibited LOH with at least one polymorphic marker. LOH was also identified in two out of four ESN. Our study is the first to show frequent occurrence of LOH not only in neoplastic stromal cells but also in tumorfree surrounding myometrial and endometrial tissues either close to or distant from the tumor (Tables 1 and 3). Indeed, 10 out of 15 benign and malignant stromal tumors (66.5%) showed LOH in normal-appearing and tumor-free endometrial or myometrial tissues (Fig. 7). Although in some cases, the observed genetic changes occurred simultaneously in ESS and surrounding endometrial or myometrial tissues, several tumors showed independent genetic alterations in different tissue components (Table 1).
Fig. 4. Retention of heterozygosity (ROH) in endometrial stromal sarcoma (ESS, low grade) and myometrium close the tumor (MCT) by using DNA polymorphic marker D10S610. Note that this focus is lost (LOH) in normal endometrium (NE) and normal myometrium (NM).
668
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
but no mutations were found in the genomic sequence of PTEN exons 5 to 9 in any of the fibrosarcomas [26]. The results of our study showed loss of one allele at the PTEN locus in 37% of the cases. Sequencing of the exons 1 to 9 in the present study, however, did not detect a mutation. One might consider several possible explanations for the occurrence of LOH of the PTEN gene without apparent inactivation of the retained allele. One possibility that should be taken into account is that another tumor suppressor gene is closely linked to PTEN and represents the actual target. In some human tumors, the closely positioned genes MINPP1 and BMPR1A have been found to be mutated in a subset of PTEN mutation-negative tumors [27,28]. A second possibility may be that the retained PTEN allele is inactivated by a mechanism other than mutation of the coding sequence, such as promoter mutation or epigenetic silencing. Promotor methylation of PTEN has been reported in human endometrial [29] and prostate carcinomas [30] and may be responsible for PTEN inactivation. A third possibility to be considered is that whenever there is loss of one PTEN allele, the activity of the remaining allele may be insufficient to
Fig. 5. LOH at D10S221 identified only in myometrium close to the tumor (MCT). Note the retention of heterozygosity (ROH) in endometrial stromal sarcoma (undifferentiated endometrial sarcoma), and in other tissue components such as normal myometrium (NM) and normal endometrium (NE).
regulates AKT activity by negative regulation of the growthfactor-induced phosphatidylinositol-3 kinase (PI3K) pathway, and up-regulates proapoptotic mechanisms such as caspases. The protein phosphatase activity of PTEN is involved in the inhibition of focal adhesion, in cell spreading and migration, as well as in the inhibition of growthfactor-stimulated MAPK signaling. Recent studies have suggested that PTEN blocks hyaluronic-acid-induced matrix metalloproteinase-9 secretion and invasion through its protein phosphatase activity [19]. Therefore, the combined effects of the PTEN lipid and protein phosphatase activities may result in aberrant cell growth and escape from apoptosis, as well as abnormal cell spreading and migration of neoplastic cells [23]. While LOH at the PTEN locus frequently occurs in certain types of sarcomas, mutation of PTEN is a very rare event [25]. Indeed, a recent study showed complete loss of one allele of PTEN in more than 60% of DMBA-induced fibrosarcomas. DNA sequencing was performed to investigate whether the remaining PTEN allele was inactivated
Fig. 6. Using polymorphic marker D10S185, LOH is observed in myometrium close to the tumor (MCT) and in endometrial stromal sarcoma (undifferentiated sarcoma). No genetic abnormality in normal myometrium (NM) or in normal cervical epithelium (NCE).
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
669
Fig. 7. Using DNA polymorphic marker D5S107, LOH is identified in myometrium close to the tumor (MCT), normal cervical stroma (NCS), and normal cervical squamous epithelium (NCE). Note the retention of hetrozygosity in endometrial stromal sarcoma (ESS), normal endometrium (NE), and normal myometrium (NM).
prevent tumor formation in a dose-dependent manner. Recent studies have suggested that PTEN might be haploinsufficient [31–33]. In fact, haploinsufficiency might explain why there are frequent allelic imbalances in the PTEN region, whereas actual PTEN mutations seem to be very rare, at least in certain types of tumors such as prostate and breast carcinomas [34]. The second most common locus altered in endometrial stromal neoplasms was D14S267 (located at 14q32). In two cases, these changes were only found in the myometrium close to the tumor. In one case, LOH was identified in the stromal cells of a low-grade ESS unassociated with genetic changes in the surrounding nontumorous myometrial or endometrial tissues. The locus D14S267 harbors a putative tumor suppressor gene commonly lost in a variety of malignant tumors such as uterine leimyosarcoma [13], colorectal [35], ovarian [36], as well as esophageal carcinomas [37]. Two of 18 informative cases in our study showed LOH at D3S 1300. This locus harbors the tumor suppressor gene FHIT that encompasses FRA3B, the most common fragile site in the human genome. LOH at D3S1300 occurred in benign neoplastic stromal cells of ESN in one case and in stromal cells of low-grade ESS in another. Several studies have shown LOH at the FHIT gene in different solid human tumors including colorectal [38], breast [11,12], endometrial [39], and uterine cervical [40] carcinomas. Other tumor suppressor genes that were found in our study to be rarely altered include P53 (tumor cells of one
low grade ESS), WT1 (tumor cells of one low grade ESS), and BRCA1 (neoplastic cells of an ESN). These tumor suppressor genes have been shown to play important roles for the development of a variety of malignant tumors including ovarian carcinomas [41] and osteosarcomas [42]. In contrast to endometrial stromal tumors, none of the 10 prolapsed uteri without histopathologic abnormalities revealed genetic changes as revealed by LOH or MSI. These findings support the hypothesis that endometrial stromal sarcoma is not just a localized genetic disease of the endometrium but rather represents a bmicroenvironmentalQ genetic alteration occurring in different tissue components of the uterus. Our findings suggest that a tissue field of somatic genetic alterations (LOH but not MSI) precedes the histopathological phenotypic changes of endometrial stromal sarcoma. Loss of functions of certain tumor suppressor genes such as PTEN in surrounding nontumorous uterine tissues could influence and facilitate tumor proliferation, cell spreading, and invasion of malignant endometrial stromal cells. The observed genetic alterations in the surrounding nontumorous uterine tissues and possible dynamic interaction(s) between neoplastic stromal cells and non-neoplastic endometrial and/or myometrial tissues could be crucial for the tumorigenesis, differentiation, and biological behavior of uterine stromal tumors. Interestingly, in a previous study, we observed concurrent and independent genetic alterations (LOH) in the epithelial (cancerous) and stromal (mesenchymal) cells of mammary
670
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
carcinomas [12]. Genetic abnormalities were found in the tumor-free stromal cells either close to invasive or noninvasive ductal carcinomas or even in the morphologically normal-appearing stromal cells distant from noninvasive or invasive carcinomas. The results of our previous study favored the concept of reciprocal stromal–epithelial interaction in mammary tumorigenesis. In summary, the present study is the first to examine allelic imbalances (LOH, MSI) in a larger number of endometrial stromal neoplasms. While LOH is a common event, MSI could not be identified in these neoplasms. The frequent occurrence of LOH and the lack of MSI suggest that loss of functions of tumor suppressor genes and not mismatch repair insufficiency play a key role in the pathogenesis of endometrial stromal neoplasms. More importantly, LOH occurs concurrently and independently in neoplastic stromal cells and in non-neoplastic surrounding endometrial/or myometrial tissues, raising the intriguing possibility of dynamic and reciprocal interaction (bcrosstalkQ) between different tissue components of the uterus. The interactions between cancer cells and their micro- and macroenviroment may promote tumor growth and differentiation. Investigation of this process might provide new insights into the mechanisms of tumorigenesis and could also lead to new therapeutic strategies.
[10]
[11]
[12]
[13]
[14]
[15] [16]
[17]
[18]
[19]
Acknowledgment This study was supported by a Lore Saldow research fund.
[20]
[21]
References [22] [1] Koss LG, Spiro RH, Brunschwig A. Endometrial stromal sarcoma. Surg Gynecol Obstet 1965;121:531 – 7. [2] Aaro LA, Symmonds RE, Dockertly MB. Sarcoma of the uterus. A clinical and pathologic study of 117 cases. Am J Obstet Gynecol 1966;94:101 – 9. [3] Norris HJ, Taylor HB. Mesenchymal tumors of the uterus. I. A clinical and pathological study of 53 endometrial stromal tumors. Cancer 1966;19:755 – 66. [4] Tavassoli FA, Devilee P. World Health Organization classification of tumours. Pathology & Genetics of Tumors of the Breast and Female Genital Organs. Lyon7 IARCP Press; 2003. [5] Sreekantaiah C, Li FP, Weidner N, Sandberg A. An endometrial stromal sarcoma with clonal cytogenetic abnormalities. Cancer Genet Cytogenet 1991;55:163 – 6. [6] Dal Cin P, Aly MS, De Wever I, Moerman P, Van Den Berghe H. Endometrial stromal sarcoma t(7;17)(p15–21;q12–21) is a common non-random chromosome change. Cancer Genet Cytogenet 1992; 63:43 – 6. [7] Laxman R, Currie JL, Kurmann RJ, Dudzinski M, Griffin CA. Cytogenetic profile of uterine sarcomas. Cancer 1993;71:1283 – 8. [8] Cheung AN, Tin VP, Ngan HY, Chung LP, Khoo US. Interphase cytogenetic study of endometrial stromal sarcoma by chromosome in situ hybridization. Mod Pathol 1996;9:910 – 8. [9] Koontz JI, Soreng AL, Nucci M, Kuo FC, Pauwels P, Van den Berghe H, Cin PD, Fletcher JA, Sklar J. Frequent fusion of the JAZF1 and
[23] [24]
[25]
[26]
[27]
[28]
[29]
[30]
JJAZ1 genes in endometrial stromal tumors. Proc Natl Acad Sci 2001;98:6348 – 53. Huang HY, Ladanyi M, Soslow RA. Molecular detection of JAZF1 and JJAZ1 gene fusion in endometrial stromal neoplasms with classic and variant histology: evidence for genetic heterogeneity. Am J Surg Pathol 2004;28:224 – 32. Moinfar F, Man YG, Bratthauer GL, Ratschek M, Tavassoli FA. Genetic abnormalities in mammary ductal intraepithelial neoplasia-flat type (bclinging ductal carcinoma in situQ): a simulator of normal mammary epithelium. Cancer 2000;88:2072 – 81. Moinfar F, Man YG, Arnould L, Bratthauer GL, Ratschek M, Tavassoli FA. Concurrent and independent genetic alterations in the stromal and epithelial cells of mammary carcinoma: implications for tumorigenesis. Cancer Res 2000;60:2562 – 6. Zhai YL, Nikaido T, Orii A, Horiuchi A, Toki T, Fujii S. Frequent occurrence of loss of heterozygosity among tumor suppressor genes in uterine leiomyosarcoma. Gynecol Oncol 1999;75:453 – 9. Tritz D, Pieretti M, Turner S, Powell D. Loss of heterozygosity in usual and special variant carcinomas of the endometrium. Hum Pathol 1997;28:607 – 12. Katsuyasu S, Masaki O, Miki H. Microsatellite alterations in various sarcomas in Japanese patients. J Orthoo Sci 1999;4:223 – 30. Cawkwell L, Bell SM, Lewis FA, Dixon MF, Taylor GR, Quirke P. Rapid detection of allele loss in colorectal tumors using microsatellites and fluorescent DNA technology. Br J Cancer 1993;67:1262 – 7. Tashiro H, Blazes MS, Wu R, et al. Mutation in PTEN are frequent in endometrial carcinoma but rare in other common gynaecological malignancies. Cancer Res 1997;57:3935 – 40. Bussaglia E, Rio E, Matias-Guiu X, Prat J. PTEN mutations in endometrial carcinomas: a molecular and clinicopathologic analysis of 38 cases. Hum Pathol 2000;31:312 – 7. Park MJ, Kim MS, Park IC, et al. PTEN suppresses hyaluronic acidinduced matrix metalloproteinase-9 expression in U87MG glioblastoma cells through focal adhesion kinase dephosphorylation. Cancer Res 2002;62:6318 – 22. Eng C. Role of PTEn, a lipid phosphatase upsteam effector of protein kinase B, in epithelial thyroid carcinogenesis. Ann N Y Acad Sci 2002;968:213 – 21. Kurose K, Gilly K, Matsumoto S, Watson PH, Zhou XP, Eng C. Frequent somatic mutations in PTEN and TP53 are mutually exclusive in the stroma of the breast carcinomas. Net Genet 2002;32:355 – 7. Deocampo ND, Huang H, Tindall DJ. The role of PTEN in the progression and survival of prostate cancer. Minerva Endocrinol 2003;28:145 – 53. WU H, Goel V, Haluska FG. PTEN signalling pathways in melanoma. Oncogene 2002;22:3113 – 22. Stahl JM, Cheung M, Sharma A, Trivedi NR, Shanmugam S, Robertson GP. Loss of PTEN promotes development in malignant melanoma. Cancer Res 2003;63:2881 – 90. Saito T, Oda Y, Kawaguchik T, et al. PTEN/MMAC1 gene mutation is a rare event in soft tissue sarcomas without specific balanced translocations. Int J Cancer 2003;104:175 – 8. Sjfling A, Samuelson E, Adamovic T, Behboudi A, Rfme D, Levan G. Recurrent allelic imbalance at the rat pten locus in DMBAinduce3d fibrosarcoma. Genes Chromosomes Cancer 2003;36:70 – 9. Gimm O, Chi H, Dahia PL, et al. Somatic mutation and germline variants of MINPP1, a phosphatase gene located in proximity to PTEN on 10q23.3, in follicular thyroid carcinomas. J Clin Endocrinol 2001;86:1801 – 5. Zhou XP, Woodford-Richens K, Lehtonen R, et al. Germline mutations in BMPR1A/ALK3 cause a subset of cases of juvenile polyposis syndrome and of Cowden and Bannayan–Riley–Ruvalcaba syndromes. Am J Hum Genet 2001;69:704 – 11. Salvesen HB, McDonald N, Ryan A, et al. PTEN methylation is associated with advanced stage and microsatellite instability in endometrial carcinoma. Int J Cancer 2001;91:22 – 6. Whang YE, Wu X, Suzuki H, et al. Inactivation of the tumor
F. Moinfar et al. / Gynecologic Oncology 95 (2004) 662–671
[31]
[32]
[33] [34] [35]
[36]
suppressor PTEN/MMMAC1 in advanced human prostate cancer through loss of expression. Proc Natl Acad Sci 1998;95:5246 – 50. Di Cristofano A, Pesce B, Cordon-Cardo C, Pandolfi P. PTEN is essential for embryonic development and tumor suppression. Nat Genet 1998;19:348 – 55. Kwabi-Addo B, Giri D, Schmidt K, et al. Haploinsufficiency of the Pten tumor suppressor gene promotes prostate cancer progression. Proc Natl Acad Sci 2001;98:11563 – 8. Dong JT. Chromosomal deletions and tumor suppressor genes in prostate cancer. Cancer Metastasis Rev 2001;20:173 – 93. Di Cristofano A, Pandolfi PP. The multiple roles of PTEN in tumor suppression. Cell 2002;100:387 – 90. Bando T, Kato Y, Ihara Y, Yamagishi F, Tsukada K, Isobe M. Loss of heterozygosity of 14q32 in colorectal carcinoma. Cancer Genet Cytogenet 1999;111:161 – 5. Bandera CA, Takahashi H, Behbakht K, et al. Deletion mapping of two potential chromosome 14 tumor suppressor gene loci in ovarian carcinoma. Cancer Res 1997;57:513 – 5.
671
[37] Ihara Y, Kato Y, Bando T, et al. Allelic imbalance of 14q32 in esophageal carcinoma. Cancer Genet Cytogenet 2002;135:177 – 81. [38] Petursdottir TE, Hafsteinsdottir SH, Jonasson JG, et al. Loss of heterozygosity at the FHIT gene in different solid human tumors and its association with survival in colorectal cancer patients. Anticancer Res 2002;22:3205 – 12. [39] Ozaki K, Enomoto T, Yoshino K, et al. Fhit alterations in endometrial carcinoma and hyperplasia. Int J Cancer 2000;85:306 – 12. [40] Acevedo CM, Henriquez M, Emmert-Buck MR, Chuagui RF. Loss of heterozygosity on chromosome arms 3p and 6q in microdissected adenocarcinomas of the uterine cervix and adenocarcinoma in situ. Cancer 2002;94:793 – 802. [41] Quezdo MM, Moskaluk CA, Bryant B, Mills SE, Merino MJ. Incidence of loss of heterozygosity at p53 and BRCA1 loci in serous surface carcinoma. Hum Pathol 1999;30:203 – 7. [42] Sztan M, Papai Z, Szendroi M, Looij A, Olah E. Allelic losses from chromosome 17 in human osteosarcomas. Pathol Oncol Res 1997;3:115 – 20.