Neurol India Home 

Year : 2015  |  Volume : 63  |  Issue : 3  |  Page : 399--404

P53 stratification reveals the prognostic utility of matrix metalloproteinase-9 protein expression in glioblastoma

Arun H Shastry1, Balaram Thota2, Arivazhagan Arimappamagan3, Vani Santosh4,  
1 Department of Clinical Neurosciences, National Institute of Mental Health and Neurosciences, Bengaluru, Karnataka, India
2 Department of Neuropathology, National Institute of Mental Health and Neurosciences, Bengaluru, Karnataka, India
3 Department of Neurosurgery, National Institute of Mental Health and Neurosciences, Bengaluru, Karnataka, India
4 Department of Clinical Neurosciences; Department of Neuropathology, National Institute of Mental Health and Neurosciences, Bengaluru, Karnataka, India

Correspondence Address:
Vani Santosh
Department of Neuropathology, National Institute of Mental Health and Neurosciences, Bengaluru, Karnataka


Background: Despite the conventional acceptance of the matrix metalloproteinases (MMP)-2 and MMP-9, as markers of invasion in glioblastoma (GBM), there is no large body of evidence supporting their role as prognostic markers. Since the co-expression of MMPs with p53 was noted to be prognostic in other cancers, we evaluated the protein expression of MMP-2 and MMP-9 in GBM and explored their prognostic relevance with respect to p53 expression. Materials and Methods: Tumor tissues from a uniformly treated cohort of 132 GBM patients were examined for MMP-2, MMP-9, and p53 protein expression by immunohistochemistry (IHC). Survival analyses were performed by Cox-regression and Kaplan-Meier (KM) survival analysis. P53 IHC-based stratification of all GBM cases was performed, and subgroup-specific expression of MMP-2 and MMP-9 was correlated with survival. Results: MMP-2 and MMP-9 were expressed in p53 positive as well as p53 negative GBM tumors. MMP-2 and MMP-9 protein expressions had no correlation with prognosis. MMP-9 expression, however, emerged as a strong independent predictor of poor survival in p53 positive GBMs on both Cox-regression analysis (P = 0.036) and KM survival analysis (P = 0.008). Further, even on multivariate analysis, MMP-9 remained strongly associated with poor prognosis (P = 0.010). Conclusions: MMP-9 expression strongly associates with poor prognosis in p53 positive GBMs, but the absence of such correlation in p53 negative GBMs, skews the overall relation of this molecule with prognosis. The study highlights that the dual positivity of MMP-9 and p53 is of prognostic relevance in GBM.

How to cite this article:
Shastry AH, Thota B, Arimappamagan A, Santosh V. P53 stratification reveals the prognostic utility of matrix metalloproteinase-9 protein expression in glioblastoma.Neurol India 2015;63:399-404

How to cite this URL:
Shastry AH, Thota B, Arimappamagan A, Santosh V. P53 stratification reveals the prognostic utility of matrix metalloproteinase-9 protein expression in glioblastoma. Neurol India [serial online] 2015 [cited 2022 Oct 6 ];63:399-404
Available from:

Full Text


Invasion by the cancer cells into the normal tissue has been well-recognized as a hallmark of cancer. [1] In addition to the tumor cellular events such as the loss of cell-cell adhesion molecules, the property of invasion has been attributed to the modulation of extracellular matrix by matrix degrading enzymes like the matrix metalloproteinases (MMPs). [1] In fact, the notion of MMP mediated invasion of cancer cells has been the driving force behind MMP research in cancer biology. [2] Following the demonstration of the role of MMPs in facilitating cancer invasion, MMP inhibitors came into the scenario. Till date, more than 50 MMP inhibitors have been experimented in clinical trials; however, these have failed to show a response in terms of patient prognosis. [2] This has been attributed to various aspects ranging from sub-optimal design of the study to the complexity of action of the MMPs in cancer. [2]

Matrix metalloproteinases -2 and -9 (gelatinase A and B) are zinc-dependent endopeptidases belonging to the MMP family. These have also been shown to play a key functional role in invasion and angiogenesis in glioblastoma (GBM), the most common and aggressive primary neoplasm of the brain. [3],[4] A few studies have examined the expression of MMP-2 and MMP-9 in gliomas where a high expression of these has been identified in astrocytic tumors as opposed to oligodendroglial tumors. Further, their expression in astrocytic tumors showed no relation to the histological grade. [5],[6],[7] An absence of prognostic relation of these molecules in astrocytomas was also noted. [5] However, there are no large-scale clinical validations in GBMs. The function of MMP-2 in GBM has also been subjected to debate ever since its role in attenuating GBM tumor growth was demonstrated. [8] These suggested a possible context-dependent role of MMPs, and expose the lacunae in our current understanding of the biology of these molecules in GBM. This is further compounded by the significant heterogeneity of the GBM tissues existing at the cytopathological, transcriptional, and genomic levels. [9]

The regulation of the MMP expression by p53 has been shown in various contexts such as, a differential regulation of MMP-13 by wild-type and mutant p53, [10] wild-type p53 mediated inhibition of MMP-1 [11],[12] and MMP-9. [13],[14] Furthermore, p53 expression has been shown to promote MMP-2 expression, which is enhanced following irradiation. [15],[16] Interestingly, in ovarian cancer, parallel expression of MMP-2 and p53 has shown to enhance cellular invasion and metastasis. [17] These have led to an intriguing possibility of a biological interplay between p53 and MMPs in determining the tumor outcome.

In the current study, we have evaluated the expression pattern of MMP-2 and MMP-9 and assessed their prognostic value in a cohort of uniformly treated adult patients with newly diagnosed GBM.

 Materials and Methods

Patient and tissue samples

The clinical data and tissue samples were obtained following institutional ethical clearance and informed patient consent from patients who underwent surgery at the two clinical centers (National Institute of Mental Health and Neurosciences and Sri Sathya Sai Institute of Higher Medical Sciences, Bangalore, India). GBM samples (n = 132) were selected from a clinical cohort of patients who were newly diagnosed with GBM and had undergone maximal safe resection of the tumor as assessed by postoperative magnetic resonance imaging (MRI) scan and had a postoperative Karnofsky's Performance Score (KPS) ≥70. The cohort received uniform treatment, which included radiotherapy (total dose = 59.4 Gy), along with concomitant chemotherapy with temozolomide (100 mg/day, daily for 45 days) and cyclical chemotherapy with temozolomide (150 mg/sq m body surface area for 5 days every 28 days). The patients were regularly followed up clinically and with MRI. The overall survival was defined as the duration between surgery and death of the patient due to disease.


All cases were subjected to histopathology and were reviewed by a neuropathologist (V.S), and the diagnosis of GBM was confirmed [Figure 1]a. Paraffin sections (4 μm) from the above-mentioned tissue were collected on silane-coated slides, and the protein expression of p53, MMP-2, and MMP-9 was assessed by immunohistochemistry (IHC). Antigen retrieval was done by heat treatment at 850W in citrate buffer (for p53) or at 700W in Tris-EDTA buffer (for MMP-2 and MMP-9). After the initial processing steps, sections were incubated overnight with primary antibody at 4°C. The primary antibody for p53 (Mouse Monoclonal DO-7; Biogenex, USA) was used at a dilution of 1:200; MMP-2 (Mouse monoclonal MMP-2 {4D3}; Santacruz Biotechnology, USA) was used at a dilution of 1:20 and MMP-9 (Mouse monoclonal MMP-9 [2C3]; Santacruz Biotechnology, USA) was used at a dilution of 1:25. This was followed by incubation with secondary antibody (QD440-XAK, Biogenex). 3, 3'- Diaminobenzidine (Sigma-Aldrich) was used as the chromogenic substrate. Neurofibroma tumor tissue and liver tissue served as positive controls for MMP-2 and MMP-9, respectively. GBM tumors showing high p53 expression in our previous study served as positive controls. [18] A positive and a negative control (slide in which the primary antibody is omitted) was included with each batch of staining.{Figure 1}

The nuclear expression pattern of p53 in gliomas is well-documented and tumor tissues with 20% or more cells demonstrating strong staining were labeled as p53 positive GBM as in our previous study. [18] For MMP-2 and MMP-9, each stained slide was subjected to a visual semi quantitative analysis for the intensity and extent of immuno-reactivity. The percentage of cells showing definitive staining was expressed as the labeling index (LI). The LI across all tissue specimens for both MMP-2 and MMP-9, was tabulated.

Statistical analysis

SPSS 15.0 statistical software (SPSS, Inc., Chicago, IL) was used for analysis. P < 0.05 was considered significant.

Survival analysis

For correlation of protein expression with survival in all GBMs, the Cox-regression analysis was employed. The prognostic significance of MMP-2 and MMP-9 was then assessed following p53 stratification of GBM cases. The clinical parameters like the extent of surgical resection and postoperative KPS were standardized as part of the inclusion criteria into the study; hence, the only clinical variable included for analyses was patient age. The parameters that stood significant on a univariate analysis were subjected to multivariate analysis employing Cox-regression models.

For the purpose of clinical utility, the protein expression was dichotomized into low and high MMP expressing tumors with a median cut-off value of LI at 20%. Log-rank tests for significance were performed, and the Kaplan-Meier (KM) curves were generated. Results were reported using the P value and the estimated hazard ratio (HR) with their 95% confidence intervals (CI).



p53 expression was localized to the nucleus [Figure 1]b, and 46.2% (61/132) of the cases were p53 positive while 53.8% (71/132) were negative for p53 expression. The tumor cells in both p53 positive and negative GBMs showed cytoplasmic staining for MMP-2 [Figure 1]c and MMP-9 [Figure 1]d. The nuclear rim positivity was noted for MMP-9 in majority of the cases. The vascular endothelial cells also showed a positive staining for both MMP-2 and MMP-9.

 Survival analysis

The GBM patient cohort consisted of 132 patients, with the maximum follow-up period of 85 months. The median survival of patients of the whole group was 15 months. On evaluating for influence of various variables on survival, we noted that patient age (HR: 1.029; P = 0.004; CI: 1.009-1.049) was associated with shorter survival in GBM. p53, MMP-2, and MMP-9 expression showed no correlation with survival in the whole cohort of GBM in the present study. We then stratified the patients based on the p53 expression status and performed subgroup analyses which revealed some unique findings [Table 1]. In patients with GBM lacking p53 protein expression, none of the studied proteins (MMP-2 and MMP-9) showed a correlation with survival. Interestingly, in p53 positive GBMs, MMP-9 positivity demonstrated statistically significant association (P = 0.036) with survival on Cox-regression analysis. Further on multivariate analysis for age and MMP-9 expression, MMP-9 protein expression showed a strong independent association with poor survival (P = 0.010, HR = 1.023; 95% CI = 1.005-1.040); also the patient age continued to be associated with poor prognosis (P = 0.004, HR = 1.032, 95% CI = 1.010-1.040) in p53 positive GBM cases.{Table 1}

The GBM cases were dichotomized into low and high MMP-9 expressing tumors, at a median cut-off value of LI at 20% and the distribution of cases have been depicted in [Table 2]. Log-rank tests for significance and KM survival curves were plotted for all GBMs [Figure 2]a and following p53 stratification [Figure 2]b and c. Interestingly, we noted that high MMP-9 expression was strongly associated with poor prognosis (P = 0.008) only in p53 positive GBM [Figure 2]c. The lack of association of MMP-9 with survival in p53 negative GBM depicted by the overlapping of the KM survival curves [Figure 2]b was also a noteworthy observation.{Figure 2}{Table 2}


The classical clinical validation of potential targets in various cancers requires a demonstration of a survival correlation or a grade-specific expression; however these studies may be skewed due to underlying heterogeneity. Heterogeneity among various GBM samples is well-established [9],[19],[20],[21] and such heterogeneity plays a major role in hindering the identification of potential targets.

The earliest identification of such heterogeneity was the classification of all GBMs into primary and secondary GBMs. Primary GBMs arise without evidence of a preexisting low-grade tumor and represent the most frequent presentation while secondary GBMs develop by progression from low grade (Grade II/III) gliomas. Further studies pointed toward the existence of clinically relevant subgroups of GBM with distinct molecular genetic characterizations, although the exact classification has varied across the literature. [19],[20],[21] The prognostic significance of these different clinical and molecular parameters is being extensively studied. However, the well-established prognostic indicators for GBM (following maximal safe surgical resection and radio-chemotherapy) only include age, performance status, and MGMT methylation status, [22],[23] which may not be adequate to explain the diverse biology and behavior of the disease. Thus, identification of newer prognostic factors is important for planning and selecting treatment for patients with these tumors.

Various molecules which are considered to form a part in the pathways in the pathogenesis of GBM are under study for their influence on survival. In the present study, we have evaluated the prognostic relevance of p53, MMP-2, and MMP-9 in GBM.

The tumor suppressor gene p53 and its protein product have been widely investigated in many human tumors including GBMs. The p53 gene encodes for a nuclear protein involved in the regulation of the cell cycle and in cellular proliferation. Because the wild-type protein has a short half-life, IHC analysis detects mutant or abnormally stabilized p53. In gliomas, p53 immuno positivity has been recently shown as a moderately sensitive and highly specific marker to predict TP53 mutations, [24] enabling utilization of p53 IHC as a surrogate to identify p53 mutation in routine practice. Despite p53 mutation being traditionally identified with secondary GBMs, many studies including one from our group, have failed to demonstrate p53 as a prognostic marker in GBM. [18],[23],[25]

The MMP-2 and MMP-9 molecules have been shown to play a key functional role in GBM invasion and angiogenesis. [3],[4] There are also rich data available on the biological/functional roles played by these gelatinases in GBM, where these have been demonstrated to play a role in tumor growth, invasion, angiogenesis, and inhibition of apoptosis. [26],[27],[28],[29] Despite all these, there is no well-accepted consensus on the clinical significance of these molecules in terms of patient prognosis. In our study, we have shown that these molecules fail to show any prognostic relevance in GBMs when considered as a single entity. In the background of the genetic heterogeneity and multifactorial etiopathogenesis of the tumor, it is likely that some of the molecules might influence the behavior of the tumor based on the presence of upstream mutations/genetic changes. Previous reports on p53 mediated regulation of MMP molecules, [10],[11],[12],[13],[14],[15],[16] and a demonstration of co-expression of p53 and MMP being associated with worse prognosis in ovarian cancer, [17] prompted the evaluation of their relation in GBM in our study. All GBM cases were then stratified into two groups, p53 positive GBM, and p53 negative GBM. The prognostic relevance of MMP-2 and MMP-9 were then evaluated in both the subgroups. The present study documented that MMP-9 emerged as significant predictor of survival in p53 positive GBM, both on univariate and multivariate analysis.

Our study has generated certain novel findings and newer hypothesis in terms of glioma biology. The present study demonstrates how certain molecules which fail to show an association with prognosis in terms of survival do so following stratification in one subgroup but not in the other. This also leads us to envisage a potential role of subgroup-specific therapeutic targets, such as a possible role of MMP inhibitors in p53 positive GBMs. The current study provides an impetus to clinically address tumor heterogeneity in terms of routine pathological investigation like IHC and identify molecules of prognostic and possible therapeutic relevance.


Matrix metalloproteinase-9 protein expression pattern uniquely associates with a poor prognosis in p53 positive GBMs, but there is a notable absence of such correlation in p53 negative GBMs. This skews its overall relation with survival in GBM. MMP-9 has thus emerged as potential therapeutic targets in p53 positive GBM. Understanding their biological role in context-dependent manner is currently highly desirable for its validation as a therapeutic target. The current study highlights that the stratification of GBM into subgroups, based on p53 status, can provide an opportunity to identify newer therapeutic targets. Such stratification can also recognize the patient population, which is most likely to benefit from targeted MMP inhibitor therapy.


We acknowledge Prof. M.R.S Rao, Prof. Paturu Kondaiah, Prof. Kumaravel Somasundaram, Dr. A.S. Hegde, other project investigators and project assistants involved in NIMITLI-CSIR (Neuro-Oncology) project. Arun H Shastry is supported by Indian Council of Medical Research, Government of India, under the MD-PhD/TSS-2, medical scientist training program.


1Hanahan D, Weinberg RA. Hallmarks of cancer: The next generation. Cell 2011;144:646-74.
2Kessenbrock K, Plaks V, Werb Z. Matrix metalloproteinases: Regulators of the tumor microenvironment. Cell 2010;141:52-67.
3Deryugina EI, Bourdon MA, Luo GX, Reisfeld RA, Strongin A. Matrix metalloproteinase-2 activation modulates glioma cell migration. J Cell Sci 1997;110 (Pt 19):2473-82.
4Esteve PO, Tremblay P, Houde M, St-Pierre Y, Mandeville R. In vitro expression of MMP-2 and MMP-9 in glioma cells following exposure to inflammatory mediators. Biochim Biophys Acta 1998;1403:85-96.
5Kunishio K, Okada M, Matsumoto Y, Nagao S. Matrix metalloproteinase-2 and -9 expression in astrocytic tumors. Brain Tumor Pathol 2003;20:39-45.
6Munaut C, Noël A, Hougrand O, Foidart JM, Boniver J, Deprez M. Vascular endothelial growth factor expression correlates with matrix metalloproteinases MT1-MMP, MMP-2 and MMP-9 in human glioblastomas. Int J Cancer 2003;106:848-55.
7Thorns V, Walter GF, Thorns C. Expression of MMP-2, MMP-7, MMP-9, MMP-10 and MMP-11 in human astrocytic and oligodendroglial gliomas. Anticancer Res 2003;23:3937-44.
8Tremblay P, Beaudet MJ, Tremblay E, Rueda N, Thomas T, Vallières L. Matrix metalloproteinase 2 attenuates brain tumour growth, while promoting macrophage recruitment and vascular repair. J Pathol 2011;224:222-33.
9Furnari FB, Fenton T, Bachoo RM, Mukasa A, Stommel JM, Stegh A, et al. Malignant astrocytic glioma: Genetics, biology, and paths to treatment. Genes Dev 2007;21:2683-710.
10Sun Y, Cheung JM, Martel-Pelletier J, Pelletier JP, Wenger L, Altman RD, et al. Wild type and mutant p53 differentially regulate the gene expression of human collagenase-3 (hMMP-13). J Biol Chem 2000;275:11327-32.
11Sun Y, Sun Y, Wenger L, Rutter JL, Brinckerhoff CE, Cheung HS. p53 down-regulates human matrix metalloproteinase-1 (collagenase-1) gene expression. J Biol Chem 1999;274:11535-40.
12Sun Y, Zeng XR, Wenger L, Firestein GS, Cheung HS. P53 down-regulates matrix metalloproteinase-1 by targeting the communications between AP-1 and the basal transcription complex. J Cell Biochem 2004;92:258-69.
13Cohen M, Wuillemin C, Irion O, Bischof P. Regulation of MMP-9 by p53 in first trimester cytotrophoblastic cells. Hum Reprod 2008;23:2273-81.
14Liu J, Zhan M, Hannay JA, Das P, Bolshakov SV, Kotilingam D, et al. Wild-type p53 inhibits nuclear factor-kappa B-induced matrix metalloproteinase-9 promoter activation: Implications for soft tissue sarcoma growth and metastasis. Mol Cancer Res 2006;4:803-10.
15Wang JL, Sun Y, Wu S. Gamma-irradiation induces matrix metalloproteinase II expression in a p53-dependent manner. Mol Carcinog 2000;27:252-8.
16He M, Dong C, Ren R, Yuan D, Xie Y, Pan Y, et al. Radiation enhances the invasiveness of irradiated and nonirradiated bystander hepatoma cells through a VEGF-MMP2 pathway initiated by p53. Radiat Res 2013;180:389-97.
17Grelewski PG, Bar JK. The role of p53 protein and MMP-2 tumor/stromal cells expression on progressive growth of ovarian neoplasms. Cancer Invest 2013;31:472-9.
18Srividya MR, Thota B, Arivazhagan A, Thennarasu K, Balasubramaniam A, Chandramouli BA, et al. Age-dependent prognostic effects of EGFR/p53 alterations in glioblastoma: Study on a prospective cohort of 140 uniformly treated adult patients. J Clin Pathol 2010;63:687-91.
19Verhaak RG, Hoadley KA, Purdom E, Wang V, Qi Y, Wilkerson MD, et al. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell 2010;17:98-110.
20Huse JT, Phillips HS, Brennan CW. Molecular subclassification of diffuse gliomas: Seeing order in the chaos. Glia 2011;59:1190-9.
21Kim YW, Koul D, Kim SH, Lucio-Eterovic AK, Freire PR, Yao J, et al. Identification of prognostic gene signatures of glioblastoma: A study based on TCGA data analysis. Neuro Oncol 2013;15:829-39.
22Hegi ME, Diserens AC, Gorlia T, Hamou MF, de Tribolet N, Weller M, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 2005;352:997-1003.
23Weller M, Felsberg J, Hartmann C, Berger H, Steinbach JP, Schramm J, et al. Molecular predictors of progression-free and overall survival in patients with newly diagnosed glioblastoma: A prospective translational study of the German Glioma Network. J Clin Oncol 2009;27:5743-50.
24Takami H, Yoshida A, Fukushima S, Arita H, Matsushita Y, Nakamura T, et al. Revisiting TP53 Mutations and Immunohistochemistry - A Comparative Study in 157 Diffuse Gliomas. Brain Pathol 2015;25:256-65.
25Levidou G, El-Habr E, Saetta AA, Bamias C, Katsouyanni K, Patsouris E, et al. P53 immunoexpression as a prognostic marker for human astrocytomas: A meta-analysis and review of the literature. J Neurooncol 2010;100:363-71.
26Badiga AV, Chetty C, Kesanakurti D, Are D, Gujrati M, Klopfenstein JD, et al. MMP-2 siRNA inhibits radiation-enhanced invasiveness in glioma cells. PLoS One 2011;6:e20614.
27Gondi CS, Talluri L, Dinh DH, Gujrati M, Rao JS. RNAi-mediated downregulation of MMP-2 activates the extrinsic apoptotic pathway in human glioma xenograft cells. Int J Oncol 2009;35:851-9.
28Kesanakurti D, Chetty C, Bhoopathi P, Lakka SS, Gorantla B, Tsung AJ, et al. Suppression of MMP-2 attenuates TNF-á induced NF-êB activation and leads to JNK mediated cell death in glioma. PLoS One 2011;6:e19341.
29Lakka SS, Gondi CS, Yanamandra N, Olivero WC, Dinh DH, Gujrati M, et al. Inhibition of cathepsin B and MMP-9 gene expression in glioblastoma cell line via RNA interference reduces tumor cell invasion, tumor growth and angiogenesis. Oncogene 2004;23:4681-9.