Increased expression of avian erythroblastosis virus E26 oncogene homolog 1 in World Health Organization grade 1 meningiomas is associated with an elevated risk of recurrence and is correlated with the expression of its target genes matrix metalloproteinase-2 and MMP-9
The transcription factor avian erythroblastosis virus E26 (V-Ets) oncogene homolog 1 (Ets-1) is involved in tumor development and progression through the transcriptional regulation of several matrix-degrading enzyme systems, including matrix metalloproteinases (MMPs). It has been demonstrated that the MMPs are expressed strongly in high-grade meningiomas. To determine the biologic significance of Ets-1 in the progression of benign meningiomas, the authors investigated the expressions of Ets-1 and its target genes MMP-2 and MMP-9 in primary and recurrent, Grade 1 meningiomas.
The expression levels of Ets-1, MMP-2, and MMP-9 were examined by immunohistochemistry in 70 Grade 1 meningiomas, including 36 primary tumors without recurrence after 5 years of follow-up and 17 pairs of primary tumors and subsequent recurrences.
The results demonstrated higher expression of Ets-1, MMP-2, and MMP-9 proteins in meningiomas with subsequent recurrences compared with meningiomas from patients who had no recurrences (P < .001). In addition, Ets-1 expression was correlated with the expression of both MMP-2 and MMP-9.
Meningiomas are classified into 3 grades according to the World Health Organization (WHO).1 The majority of meningiomas are benign, slowly expansive, growing tumors with a favorable clinical outcome. One of the most important factors that determines prognosis and the likelihood of meningioma recurrence is histologic grade, which relies on features like high mitotic rate, necrosis, polymorphism, prominent nucleoli, and increased proliferation index.1, 2 However, considerable variation remains within the WHO grades. Some of these tumors without atypical features recur after complete surgical removal, whereas others do not. Several new biologic markers, including alkaline phosphatase, human telomerase reverse transcriptase protein, and hepatocyte growth factor and its receptor, met protooncogene, have been investigated in recent years.3–5 Another possible marker that has prognostic potential for patients with meningioma is the family of matrix metalloproteinases (MMPs). MMP expression in meningiomas has been investigated in recent years, and the results have shown that several MMPs (e.g., MMP-2, MMP-9, MMP-11, and MMP-12) occur in meningiomas.6–9 MMPs degrade extracellular matrix (ECM) components, such as collagen and proteoglycans, in physiologic and pathologic processes, including embryogenesis, tissue remodeling, tumor development, and invasion.10, 11 Invasion into the neighboring anatomic structures is among the factors that are associated with a high rate of recurrence in meningiomas.1, 12, 13 The activity of MMPs is regulated tightly at the levels of transcription, activation of secreted MMP proenzymes by proteolytic cleavage, and inhibition by their natural inhibitors.10, 11 An imbalance between MMPs and their natural inhibitors, tissue inhibitor of matrix metalloproteinases (TIMPs), can lead to excessive ECM degradation and is considered responsible for the aggressive behavior of tumors.10 Both the down-regulation of TIMPs and the up-regulation of MMPs can lead to an imbalance between the MMPs and their inhibitors. The promoter regions of the MMP genes often contain binding sites for the E26 transformation specific sequences (Ets) transcriptional factors. The Ets proteins are involved in controlling expression of the many genes that are responsible for numerous biologic functions, such as cell differentiation, development, labor, ECM regulation, and apoptosis.14–16 V-Ets oncogene homolog 1 (Ets-1), the first member of the Ets oncogene family, has been identified in the avian erythroblastosis virus E26 and has been shown to contribute to tumor development and progression.17, 18 The current study was conducted in order to investigate the expression of the transcription factor Ets-1 and its target genes MMP-2 and MMP-9 in relation to the recurrence of intracranial Grade 1 meningiomas.
MATERIALS AND METHODS
Seventy grade 1 meningiomas (according to the histologic criteria of the WHO) obtained from 53 patients (39 females and 14 males) were investigated (Table 1). No patient had received adjuvant therapy. All samples were fixed in formalin and embedded in paraffin. Tumors were classified into 3 groups. Group A included 36 grade 1 meningiomas that had not recurred during 5 years of follow-up period. Group B included of 17 grade 1 meningiomas with known recurrences. A second surgical procedure after patients showed radiologic and symptomatic evidence of disease progression was accepted as disease recurrence. Group C included 17 the recurrent grade 1 meningiomas from the 17 Group B primary tumors (Table 1).
Table 1. Summary of Patient Characteristics and Immunoreactive Scores for the Expression of E26 Avian Erythroblastosis Virus Oncogene Homolog 1, Matrix Metalloproteinase-2, and Matrix Metalloproteinase-9
Group A: Meningiomas without recurrences (n = 36 patients)
Group B: Meningiomas with subsequent recurrences (n = 17 patients)
A polyclonal rabbit antibody (1:350 dilution; Santa Cruz Biotechnology, Santa Cruz, CA) was used to detect Ets-1 immunoreactivity. The specificity of the antisera has been demonstrated previously.19 The other antibodies used were mouse monoclonal antibodies against MMP-2 (1:50 dilution; Oncogene, San Diego, CA20) and against MMP-9 (1:100 dilution; Oncogene21).
Tissue sections were deparaffinized in xylene and rehydrated in graded ethanol. Heat-induced epitope retrieval (600-watt microwave treatment for 20 minutes in 10 mM sodium citrate buffer [pH 6.0]) was employed prior to staining. No further advanced treatment was required. Immunostaining was performed using the DAKO Techmate™ 500 autostainer (DAKO Ltd., Ely, U.K.). Incubation time for the primary antibody was 45 minutes. Bound antibody was detected by using the labeled streptavidin-biotin method with biotinylated antimouse or antirabbit goat antibody (DAKO Ltd.) and diaminobenzidine (DAKO Ltd.) as a visualizing agent. Negative control preparations involved omitting the primary antibodies and replacing them with antibody diluent (DAKO Ltd.). Tissue sections from breast carcinomas were used as a positive control. Stained slides were scored by 2 blinded, independent observers according to the method described by Reiner et al. for both intensity (negative to strongly positive) and proportion tumor cells stained (0%, 1–9%, 10–50%, 51–80%, or >81%). Integer values were assigned from 0 to 3 for the intensity score and from 0 to 4 for the percentage of positive cells. These values were added to provide a single immunoreactive score (IRS).22 In tumors with nonhomogenous staining intensity, the most representative intensity was assigned. If there was interobserver discrepancy, then the mean values were assigned.
The Kruskal–Wallis test was used for nonparametric multiple comparisons. Further comparisons between specific groups were performed by using the nonparametric Mann–Whitney test (2-tailed). The Spearman test was used to determine correlations between the expression of Ets-1 and the respective MMPs. Chi-square tests or Fisher exact tests were used for categorical variables.
For all tests, a P value <.05 denoted significance. The statistical software package used was SPSS (version 11.0; SPSS Inc., Chicago, IL).
To determine the topographic distribution of Ets-1, MMP-2, and MMP-9 expression, double immunofluorescence was performed. Tissue sections were deparaffinized in xylene and rehydrated in graded ethanol. After heat-induced epitope retrieval (600-watt microwave treatment for 20 minutes in 10 mM sodium citrate buffer [pH 6.0]), the slides were rinsed in phosphate- buffered saline (PBS) (pH 7.4). The same antibodies that were used for immunohistochemistry were used for immunofluorescence. Before incubation with the primary antibody, each section was blocked with 10% normal goat serum in PBS for 20 minutes. Sections were incubated with antibody against Ets-1 for 1 hour at room temperature, washed with PBS, and incubated with indocarbocyanine (Cy3)-conjugated goat-antirabbit antibody (1:100 dilution; Jackson Immunoresearch Laboratories, Philadelphia, PA) for 1 hour at room temperature. After washing with PBS, the sections were incubated with a second primary antibody against MMP-2 or MMP-9 for 1 hour at room temperature, washed in PBS, and incubated with Alexa Fluor™ 488-conjugated goat-antimouse antibody (1:100 dilution; Molecular Probes, Eugene, OR). After washing with PBS, tissue sections were mounted and observed under a fluorescent microscope. The wave lengths of Cy3 and Alexa Fluor™ 488 correspond to red and green fluorescence, respectively.
In the current study, we used immunohistochemistry to analyze the expression of the transcription factor Ets-1 and its target genes MMP-2 and MMP-9 in Grade 1 meningiomas in relation to tumor recurrence. We also compared expression levels in primary tumors with expression levels in recurrent tumors. Patient data are summarized in Table 1. Recurrence in meningiomas was independent of age and gender (Table 1). There were 22 benign meningiomas without subsequent recurrences (Group A) had a hemispheric location (61%), i.e., convex and falx/parasagittal, and there were 14 tumors (39%) located at the base of the skull, including the olfactory grove, sphenoid wing, parasellar/suprasellar sites, and the cerebellopontine angle. The locations meningiomas from patients who developed subsequent recurrences (Group B) were hemispheric in 10 tumors (59%) and in the skull base in 7 tumors (41%) (Table 1). There was no difference between the groups with regard to tumor location (P = .5). Three patients from Group B and 4 patients from Group A underwent subtotal removal of tumors, whereas 14 patients from Group B and 32 patients from Group A underwent macroscopic total surgery (Table 1). The recurrence ratio for patients who underwent subtotal surgery (42%; 3 of 7 patients) was greater than that for the patients who underwent complete surgery (30%; 14 of 46 patients), although the difference did not reach statistical significance.
The expression levels of Ets-1, MMP-2, and MMP-9 are listed in Table 1. Ets-1 immunoreactivity was observed within the nuclei. Ets-1 expression in benign meningiomas without subsequent recurrences (Group A) was significantly lower than in primary benign meningiomas with subsequent recurrences (Group B; P < .001) (Fig. 1A,B). Ets-1 expression in meningiomas from Group A also was significantly lower than in the recurrent tumors (Group C; P < .001). There was no significant difference in expression between Group B and Group C (P = .5). Figure 2A shows the immunoscoring of Ets-1. Expression of MMP-2 and MMP-9 was observed within the cytoplasm. MMP-2 and MMP-9 were up-regulated significantly in Groups B and C compared with Group A (P < .001) (Figs. 1, 2B,C). There were no significant differences in MMP-2 expression (P = .9) or MMP-9 expression (P = .1) between Group B and Group C (Fig. 2B,C). Ets-1 expression was correlated with both MMP-2 expression and MMP-9 expression (P < .001) (Fig. 2D,E). Double immunofluorescence also essentially revealed the coexpression of Ets-1 and either MMP-2 or MMP-9 (Fig. 3).
For the current study, we investigated microscopic invasion of tumor cells into the neighboring structures, such as brain, dura mater, and bone. In Group A, 10 of 36 meningiomas (28%) revealed microscopic infiltration, whereas 13 of 17 (77%) meningiomas with subsequent recurrences showed infiltration into the neighboring structures (P < .01). Five tumors from Group A invaded dura mater, 3 tumors invaded dura mater and bone, and, in 2 meningiomas, groups of tumor cells invaded brain. Seven tumors in Group B had invasion within dura mater, 3 had invasion within dura mater and bone, and 3 showed invasion of groups of tumor cells into brain tissue (Table 1).
Group A consisted of 22 meningothelial meningiomas, 8 transitional meningiomas, and 6 fibroblastic meningiomas; and, in Group B, there were 8 meningothelial meningiomas, 5 transitional meningiomas, and 4 fibroblastic meningiomas (Table 2). When the tumors were analyzed according to subtype, the meningothelial meningiomas had a recurrence ratio of 27% (8 of 30 tumors), the transitional meningiomas had a recurrence ratio of 38% (5 of 13 tumors), and the fibroblastic meningiomas had a recurrence ratio of 40% (4 of 10 tumors; P = .6) (Table 2). There were no significant differences in the expression of Ets-1, MMP-2, and MMP-9 between the histologic subtypes (P > .05) (Table 2).
Table 2. Summary of Subtypes of Group A and Group B
The effect of Ets-1 on progression and recurrence has been demonstrated for several tumor entities.23–26 The role of Ets-1 in tumor progression is related mainly to transcriptional activation of ECM-degrading enzymes, such as MMPs.18 In vitro studies have shown that Ets-1 mediates adhesion, migration, and invasion by the transcriptional regulation of several MMPs.27, 28 The possible modulating effect of Ets-1 on ECM in meningiomas has been demonstrated by observation of its increased expression together with its target gene, urokinase-type plasminogen activator, in malignant and invasive meningiomas.29 In the current study, we demonstrated increased Ets-1 expression in WHO grade 1 meningiomas that had subsequent recurrences (Group B) compared with a control group that consisted of WHO grade 1 meningiomas that had no recurrences within 5 years of follow-up (Group A). It is noteworthy that there were no differences between Group A and Group B with regard to grade, tumor location, age, gender, or frequency of histologic subtypes. We suggest that Ets-1 expression may have value for predicting the risk of recurrence in inconspicuous, WHO grade 1 meningiomas.
In line with the regulatory action of Ets-1 on MMPs, we demonstrated significantly higher expression of MMP-2 and MMP-9 in WHO grade 1 meningiomas that had subsequent recurrences compared with their nonrecurrent counterparts. Tumor invasion requires degradation of ECM, in which MMPs play a critical role.10, 11 Associations between MMPs and aggressive behavior or metastatic spread have been reported in a variety of cancers, such as carcinomas of the breast, colon, and lung.30–32 It has been demonstrated that elevated expression of MMPs is associated with a poor prognosis in patients with brain tumors.33, 34 Several studies already have investigated the role of MMPs in meningioma infiltration and progression.7–9, 35, 36 High MMP-11 expression has been observed in atypical meningiomas that infiltrated the surrounding tissues and subsequently recurred.9 An association between meningioma invasion or recurrence and expression levels of MMP-2 and MMP-9 has been described.7, 35 However, in those studies, the recurrent or aggressive meningiomas were graded predominantly as atypical meningiomas, WHO grade 2 or as anaplastic meningiomas, WHO grade 3. In the current study, we were able to demonstrate significantly higher expression levels of MMP-2 and MMP-9 in grade 1 meningiomas that had subsequent recurrences compared with meningiomas that did not recurr within 5 years of follow-up. Thus, our findings indicate a predictive value of MMP-2 and MMP-9 expression for the assessment of future risk of recurrence in patients with grade 1 meningiomas. We detected increased invasion of meningioma cells in the WHO grade 1 meningiomas that had high expression levels of Ets-1 and MMP along with a consecutive recurrence compared with tumors that did not recur during the observation period. Ets-1 has been implicated previously in the prognosis of several tumors because it up-regulates MMPs.37–39 We also observed not only a correlation between the expression of these proteins but also, essentially, the same cellular distribution of Ets-1 and these MMPs. In the light previous data and our current results, we do not assume that the tight correlations between strong Ets-1 expression and strong MMP expression are independent events.
The extent of tumor resection is of crucial importance for recurrence in meningiomas. A fraction of meningiomas in unfavorable localizations cannot be removed completely. There was no difference in tumor location between the groups in our series. The recurrence ratio of subtotally resected meningiomas (42%) was higher than that of completely removed tumors (30%), but the difference was not significant. Many studies have shown that meningiomas may recur after radical excision, and subtotally resected meningiomas do not always recur.13, 40–42 In 1 study, there was no significant difference in the rate recurrence between completely and partially resected meningiomas.43 Thus, it seems likely that factors other than the extent of surgical resection appear to predict the outcome of patients with meningiomas. One of these factors is the infiltration of meningioma cells into neighboring structures.12, 13 We suggest that the up-regulation of Ets-1 leads to increased expression of MMP-2 and MMP-9, thus causing excessive ECM degradation, which represents a suitable microenvironment for tumor cells to disseminate into adjacent structures. Conversely, the re maining meningioma cells that have increased expression levels of Ets-1 and MMPs may be as numerous as the cells without those features, but they may differ in their potential to modify ECM and, thereby, to promote their growth. Thus, qualitative analyses of Ets-1 and MMP expression may yield more predictive data for recurrence of WHO grade 1 meningiomas than the assessment of the quantity of tumor cells left behind after resection. Given further support by additional studies, our results may open a discussion about a regimen of observation and treatment for patients with WHO grade 1 meningioma who have increased expression of Ets-1 and MMPs.
In conclusion, the current results suggest that increased expression of Ets-1, MMP-2, and MMP-9 may predict an increased risk of recurrence in patients with Grade 1 meningiomas. Such analyses may identify patients who are in need of closer follow-up or more aggressive therapy.