【作者】 吕中强；

【导师】 张庆俊；

【作者基本信息】 河北医科大学 ， 外科学， 2007， 博士

【摘要】 目的:基质金属蛋白酶(Matrix Metalloproteinases,MMPs)是高度保守的依赖于Zn2+的蛋白水解酶家族,是肿瘤发生、侵袭和转移过程中起关键作用的酶之一。研究表明,多数MMPs在正常组织中表达量很低,而在肿瘤组织中,各种类型MMPs的表达均有不同程度的增高。MMPs不仅参与肿瘤的侵袭和转移过程,也与肿瘤发生发展的许多因素有关,包括细胞的生长、凋亡、血管生成和免疫监视等。MMPs基因启动子区的单核苷酸多态性(Single Nucleotide Polymorphism, SNPs)可能影响其基因的转录水平和蛋白质的表达,从而影响肿瘤等疾病的发生、发展及预后。脑胶质瘤(Glioma)是最常见的原发性脑肿瘤,尽管近年来在影像诊断、神经外科技术、放射治疗、化疗等方面已取得较大进展,该肿瘤的生存率并无明显提高。星形细胞瘤(Astrocytoma)是脑胶质瘤中最常见的病理类型,其病因尚未明确,近年研究提示遗传因素可能在星形细胞瘤的发生中起重要作用。鉴于有报道显示MMPs在星形细胞瘤中的表达增高,我们推测,其启动子区的基因多态性可能通过改变转录活性而影响星形细胞瘤的发生。最近,McCready等的研究显示,多形性胶质母细胞瘤组织中的MMP-1 2G/1G SNP可能在胶质母细胞瘤形成和侵袭中发挥作用,其他MMPs基因多态性与神经胶质细胞瘤的关系国内外尚未见报道。另外,MMPs基因启动子区多态性与肿瘤组织中MMPs蛋白表达及mRNA转录水平的关系尚不清楚。为此,本研究拟通过基于我院手术和病理证实的病例-对照研究,分析MMP-1、MMP-3、MMP-7和MMP-9基因启动子区基因多态性与成人脑星形细胞瘤易感性的关系,同时观察星形细胞瘤中MMP-1、3、7、9的蛋白及mRNA的表达与星形细胞瘤发生、发展的关系,特别是个体对星形细胞瘤的遗传易感性及其发生、发展的分子机制,为星形细胞瘤的预防和治疗提供理论依据。我们分三部分进行了研究。第一部分MMP-1、3、7、9基因启动子区多态性与成人星形细胞瘤的关联研究方法采用基于医院的病例对照方法。收集236例成人脑星形细胞瘤病例的手术标本,经组织学诊断后,以蛋白酶K消化、酚-氯仿抽提法提取肿瘤组织DNA。抽取366例健康对照外周静脉血5ml,以蛋白酶K消化-饱和氯化钠盐析法提取白细胞DNA。MMP-1 -1607 2G/1G, MMP-3 -1171 5A/6A, MMP-7 -181A/G, MMP-9 -1562 C/T基因型检测应用聚合酶链反应-限制性片段长度多态性分析( Polymerase Chain Reaction-Restriction Fragment Length Polymorphism,PCR-RFLP)方法进行。应用卡方检验比较对照组及病例组基因型频率的观察值与预期值,进行Hardy-Weinberg平衡检验。病例组与对照组的性别、基因型和等位基因型分布比较均采用四格表卡方检验。以非条件Logistic回归法计算表示相对风险度的比值比(Odds Ratio,OR)及其95%的可信区间(Confidence Interval,CI),并经性别、年龄校正。结果1 MMP-1-1607 2G/1G SNP的等位基因型及基因型总体分布在星形细胞瘤患者组和正常对照组之间有显著性差异(分别为P=0.002和﹤0.001)。与2G/2G基因型相比,1G/1G基因型可显著降低该肿瘤的发病风险,经性别、年龄校正的OR值为0.58(95%CI为0.42～0.79),而2G/1G基因型对该肿瘤的易感性无显著影响(校正OR值为0.74,95%CI为0.51～1.08)。根据性别、发病年龄(≤45岁或>45岁)进行的分层分析显示了相似的结果。根据WHO病理分级进行的分层分析显示,1G/1G基因型均可显著降低II、III级肿瘤的发病风险,而对I、IV级肿瘤的发病风险无显著影响。2 MMP-3 -1171 6A/5A多态性等位基因型及基因型总体分布在肿瘤组和正常对照组之间无显著性差异(P>0.05)。根据性别、发病年龄(≤45岁或>45岁)、WHO病例分级进行的分层分析显示,肿瘤组和正常对照组之间的基因型分布均无显著性差异。与MMP-3 6A/6A基因型相比,在总体比较及分层分析中均未发现6A/5A或5A/5A基因型可改变星形细胞瘤的发病风险,经性别、年龄校正的OR值分别为1.09(95% CI=0.75～1.57)和0.95 (95% CI=0.54～1.69)。3 MMP-7 -181 A/G SNP的基因型及等位基因型分布在星形细胞瘤及正常对照组有显著性差异(P=0.001)。与MMP-7 A/A基因型相比,A/G及G/G基因型可显著增加星形细胞瘤的发病风险,经性别、年龄较正的OR值分别为1.69及2.77(95% CI=1.01～2.84及1.27～6.02)。根据发病年龄进行的分层分析显示,与MMP-7 A/A基因型相比,G/G基因型可使45岁以下个体发生星形细胞瘤的风险增加3倍左右(校正OR=3.16,95%CI=1.09～9.16),而对45岁以上个体对此肿瘤的发病风险无显著影响。根据性别的分层分析显示,与MMP-7 A/A基因型相比,携带G/G基因型的男性个体发生星形细胞瘤的风险可增加3倍以上(校正OR=3.24,95%CI=1.12～9.41)。根据WHO病理分级进行的分层分析显示,MMP-7 G/G基因型较A/A基因型携带者发生II～IV级星形细胞瘤的风险增加3倍左右,校正OR值分别为2.95 (95% CI = 1.21～7.17), 3.32 (95% CI = 1.15～9.61),和3.25 (95% CI = 1.10～9.62),且A/G基因型可增加II级星形细胞瘤的发病风险约2倍(校正OR=2.06, 95% CI = 1.05～4.05),而此MMP-7基因多态性对I级星形细胞瘤的发病风险无显著影响。4 MMP-9 -1562 C/T SNP的等位基因型及基因型总体分布在肿瘤组和正常对照组之间无显著性差异(P值分别为0.926和0.818)。与C/C基因型相比,C/T+T/T基因型不能改变星形细胞瘤的发病风险(经性别、年龄校正OR值为1.09,95%CI为0.73～1.64)。根据性别、发病年龄、WHO病理分级进行的分层分析,也未发现C/T+T/T基因型与胶质瘤的发病风险相关。结论1总体分析及根据性别、发病年龄、病理分级进行的分层分析均显示,MMP-3 -1171 5A/6A多态性、MMP-9 -1562 C/T多态性不能改变星形细胞瘤的发病风险;2 MMP-1-1607 2G/1G多态性可影响星形细胞瘤的发病风险。与人群中占分布优势的2G/2G基因型相比,1G/1G基因型可显著降低该肿瘤的发病风险;3 MMP-7-181A/G多态性对星形细胞瘤发病风险有明显影响。与A/A基因型相比,携带G等位基因型(A/G和G/G基因型)可显著增加个体对星形细胞瘤的发病风险,此现象在45岁以下人群、男性个体中尤为明显。第二部分MMP-1、3、7、9蛋白表达与成人星形细胞瘤临床病理分级的关系方法收集上述星形细胞瘤标本中70例及8例脑外伤病人行手术内减压的脑组织石蜡包埋手术标本。光镜下常规病理形态学观察,确定星形细胞瘤的病理诊断及病理分级(WHO 1999)。组织标本MMP-1、3、7和9的蛋白表达以免疫组织化学方法检测,采用即用型S-P试剂盒进行。各组数据之间的相关性,采用Kendall’s tau-b等级相关分析,相关系数用tb值表示,P<0.05为显著性检验水准。不同组间蛋白表达阳性率比较用X2检验。结果1非肿瘤脑组织、I～IV级星形细胞瘤组织中MMP-1蛋白的阳性表达率分别为12.5%、75%、86%、90%、100%,各组间MMP-1蛋白的表达有显著性差异(X2=28.204,P<0.01)。MMP-1蛋白表达与星型胶质细胞瘤的恶性程度显著相关(tb=0.463,P<0.001)。随着胶质瘤病理分级增高,其MMP-1蛋白表达水平明显增加。2非肿瘤脑组织、I～IV级星形细胞瘤组织中MMP-3蛋白的阳性表达率分别为25%、33%、45.5%、35%、75%,各组间MMP-3的表达无显著性差异(X2=20.476,P=0.059)。然而,MMP-3蛋白表达与星型胶质细胞瘤的恶性程度显著相关(tb=0.276,P=0.005)。在胶质瘤组,随着胶质瘤病理分级增高,其MMP-3蛋白表达水平呈递增趋势。3非肿瘤脑组织、I～IV级星形细胞瘤组织中MMP-7蛋白的阳性表达率分别为37.5%、50%、77.3%、75%、100%,各组间MMP-7的表达有显著性差异(X2=29.779,P=0.003)。MMP-7蛋白表达与星型胶质细胞瘤的恶性程度显著相关(tb=0.416,P<0.001)。随着胶质瘤病理分级增高,其MMP-7蛋白表达水平明显增加。4非肿瘤脑组织、I～IV级星形细胞瘤组织中MMP-9的阳性表达率分别为0%、66.7%、81.8%、80%、100%,各组间MMP-9的表达有显著性差异(X2=44.038,P<0.001)。MMP-9蛋白表达与星型胶质细胞瘤的恶性程度显著相关(tb=0.427,P<0.001)。随着胶质瘤病理分级增高,其MMP-9蛋白表达水平明显增加。结论正常脑组织、I～IV级星形细胞瘤组织中MMP-1、3、7和9蛋白的表达与星型细胞瘤的恶性程度显著相关,可能为判断该肿瘤的恶性程度、侵袭能力及预后的良好分子标志物。第三部分星形细胞瘤中MMP-1、3、7、9 mRNA表达及相关性探讨方法取星形细胞瘤新鲜组织39例和正常新鲜脑组织12例,所有组织在离体30分钟内置冻存管中,并立即置于液氮罐中,-80℃冷冻保存备用。以一步法RNA抽提试剂盒提取组织细胞总RNA。以反转录法获得cDNA第一链。采用TaqMan探针技术,以荧光实时定量聚合酶链式反应法(Real-time Quantitative Polymerase Chain Reaction,RQ-PCR),检测组织中MMPs和内参GAPDH mRNA的含量。以MMPs和内参GAPDH mRNA含量的比值(均数±标准差)作为评价MMPs表达水平的指标。采用t检验,统计分析星形细胞瘤病理分级各组与正常脑组织MMPs mRNA表达水平的差异,以P<0.05为有显著性差异。结果1星形细胞瘤组织中MMP-9 mRNA的相对表达量明显高于非肿瘤脑组织(p=0.038)。低分化肿瘤组织(Ⅲ-Ⅳ级)MMP-9 mRNA表达量与非肿瘤脑组织及高分化肿瘤组织(Ⅰ-Ⅱ级)间有显著性差异(分别为p<0.001,p=0.001),而高分化肿瘤组织与非肿瘤脑组织间MMP-9 mRNA表达量无显著性差异(p=0.417)。2 MMP-1、MMP-3 mRNA相对表达量在低分化肿瘤组织(Ⅲ-Ⅳ级)与高分化肿瘤组织(Ⅰ-Ⅱ级)及非肿瘤脑组织中变化较大,之间无显著性差异(分别为p=0.26, p=0.419)。结论1低分化星形细胞瘤组织(Ⅲ-Ⅳ级)中MMP-9 mRNA表达量明显高于正常脑组织及高分化肿瘤组织。MMP-9 mRNA表达检测可用来判断星形细胞瘤的恶性程度及预后。2 MMP-1、MMP-3 mRNA表达量在非脑肿瘤组织和星形细胞瘤组织间无显著性差异,其检测对星形细胞瘤生物学行为的判断无显著意义。总之,本研究表明,MMP-1-1607 2G/1G和MMP-7-181A/G基因多态性可能与成人星形细胞瘤的发生有关,并可单独或与其他基因多态性联合作为筛选星形细胞瘤高危个体的分子标志物。MMP-1、3、7和9蛋白的高表达可作为判断星形细胞瘤疾病进展和不良预后的指标。MMP-9 mRNA检测则可用于星形细胞瘤恶性程度的判定。更多还原

【Abstract】 Objectives: Matrix metalloproteinases (MMPs), a family of highly conserved zinc-dependent proteolytic enzymes that degrade many different components of extracellular matrix (ECM) and regulate various cell behaviors, play important roles in tumor development, infiltration, and metastasis. It has been shown that MMPs expression is very low in normal tissues while elevates in tumor tissues with different extent. Growing evidence indicates that single nucleotide polymorphisms (SNPs) in promoters of MMP genes may result in variable transcription and expression of MMPs in different individuals, therefore, influence the development, progression, and outcome of tumors. Gliomas are the commonest primary brain tumors. Despite recent advances in diagnostic imaging, neurosurgical technique, radiation therapy, and chemotherapy, significant improvement in the survival of gliomas has not been achieved. Astrocytoma is the commonest type of brain gliomas and develops from a type of star-shaped cell called astrocyte. The etiology of astrocytoma is unknown. Although high-dose of radiotherapy and chemotherapy are supposed to be exogenous environmental causes, the accumulated data suggests that genetic background also play a role in the development of this tumor. Since it has been reported that the expression of MMP proteins may increase in astrocytoma tissues, we suspected that the SNPs in promoter region of MMP genes may alter susceptibility to astrocytoma via modifying the transcription activity of MMP genes. Recently, McCready et al. reported that the MMP-1 2G/1G SNP may play a role in the development and infiltration of glioblastoma, however, the association between the development of gloma and other MMP polymorphisms has not been reported so far. In addition, the relationship between the MMP polymorphisms and protein expression and mRNA transcription has not been clearly demonstrated. Hence, we conducted a hospital-based case control study on association between the promoter polymorphisms of MMP-1, 3, 7, 9 genes and susceptibility to adult astrocytoma. Additionally, the protein expression and mRNA transcription of the MMP-1, 3, 7, 9 were also evaluated in astrocytoma tumor tissues. This pilot study aims to investigate the relationship between the MMP-1, 3, 7, 9 genes and development and progression of astrocytoma, to help exploring the genetic liability as well as the molecular mechanism of the astrocytoma and provide candidate targets for prevention and control of this tumor type.Part I Polymorphisms in the MMP-1,3,7,9 promoters and susceptibility to adult astrocytomaMethodsThe association between the promoter polymorphisms of MMP-1, 3, 7, 9 genes and susceptibility to adult astrocytoma was tested in a hospital-based case control study. Tissues from 236 cases of adult astrocytoma patients were collected after pathologic diagnosis for extracting DNAs by proteinase K digestion followed by phenol-chloroform purification. Blood samples from 366 healthy subjects were collected for DNA extraction by proteinase K digestion followed by a salting out procedure. The MMP-1 -1607 2G/1G, MMP-3 -1171 5A/6A, MMP-7 -181A/G, and MMP-9 -1562 C/T genotypes were determined by polymerase chain reaction - restriction fragment length polymorphism (PCR-RFLP) assay. Hardy-Weinberg analysis was performed to compare the observed and expected genotype frequencies using Chi-square test. Comparison of the MMP genotype distribution among study groups was performed by means of two-sided contingency tables using Chi-square test or Fisher’s exact test. The odds ratio (OR) and 95% confidence interval (CI) were calculated using an unconditional logistic regression model and adjusted by age and sex accordingly.Results1 The overall distribution of the MMP-1 1607 2G/1G SNP allelotype and genotype among astrocytoma patients and healthy controls was significantly different (P=0.002 and P<0.001, respectively). Compared with 2G/2G, the 1G/1G genotype significantly decreased the risk of development of astrocytoma, the age and gender adjusted OR was 0.58 (95%CI=0.42~0.79), while the relationship between the 2G/1G genotype and astrocytoma was not observed (age and gender adjusted OR was 0.74, 95%CI=0.51~1.08). The similar results were obtained when stratified by gender and age at tumor diagnosis (≤45 Years or > 45 Years). When stratified according to the World Health Organization (WHO) astrocytoma classification, the 2G/2G genotype significantly reduced the risk of grade II and III astrocytoma, the age and gender adjusted OR was 0.61(95%CI=0.40~0.93) and 0.41(95%CI=0.20~0.85), respectively, while correlation between the MMP-1 SNP and susceptibility to grade I and IV astrocytoma was not demonstrated.2 The overall genotype and allelotype distribution among astrocytoma patients was not significantly different from that among healthy controls(P>0.05). Compared with 6A/6A, the most common genotype in the study population, both of the 5A/6A and 5A/5A genotype did not influence the risk of astrocytoma development, the age and gender adjusted OR was 1.09(95% CI=0.75~1.57) and 0.95 (95% CI=0.54~1.69), respectively. When stratified by gender, age at tumor diagnosis, and the WHO astrocytoma classification, association between the MMP-3 SNP and susceptibility to astrocytoma was also not observed.3 The overall distribution of the MMP-7 -181 A/G SNP genotypes among astrocytoma patients and healthy controls was significantly different (P = 0.001). Compared with the A/A genotype, the G/G genotype significantly increased the risk to the development of astrocytoma, the age and gender adjusted odds ratio was 2.77 (95% CI = 1.27~6.02), while the MMP-7 A/G genotype also marginally increased the risk of developing astrocytoma (the adjusted OR = 1.69, 95% CI = 1.01~2.84). Stratification analysis showed that the MMP-7 G/G genotype significantly increased the risk of astrocytoma among individuals younger than 45 years compared with the A/A genotype (adjusted OR = 3.16, 95% CI = 1.09~9.16), however, the G/G genotype did not modify the risk of the occurrence of astrocytoma among people older than 45 years (adjusted OR = 1.93, 95% CI = 0.61~6.20). In addition, stratification analysis indicated that the MMP-7 G/G genotype significantly increased the risk of astrocytoma among male subjects (adjusted OR = 3.24, 95% CI = 1.12~9.41). Stratification analysis according to the WHO pathological grading showed that of astrocytomas. A significant higher risk of developing grade II astrocytoma was observed among individuals harboring the MMP-7 A/G genotype compared with those harboring the A/A genotype (adjusted OR = 2.06, 95% CI = 1.05~4.05). In addition, the G/G genotype significantly increased the risk of developing grades II, III, and IV astrocytoma compared with the A/A genotype, the adjusted odds ratio was 2.95 (95% CI = 1.21~7.17), 3.32 (95% CI = 1.15~9.61), and 3.25 (95% CI = 1.10~9.62), respectively, whereas the distribution different among patients with grade I astrocytoma and healthy controls was not observed.4 The overall MMP-9 genotype and allelotype distribution among astrocytoma patients was not significantly different from that among healthy controls(P value was 0.92 and 0.818, respectively). Compared with the C/C genotype, genotypes with the -1562 T allele (C/T +T/T) did not show significant influence on the risk of astrocytoma development (age and gender adjusted OR= 1.09, 95% CI = 0.73~1.64). Stratification by gender, age, and the WHO astrocytoma classification also did not show an association between the MMP-9 SNP and susceptibility to astrocytoma.Conclusions1 The overall analysis and stratification analysis according to gender, age at diagnosis, and histological grade suggested that, the MMP-3 -1171 5A/6A and the MMP-9 -1562 C/T SNPs could not modify the risk of the adult astrocytoma development.2 The MMP-1-1607 2G/1G polymorphism significantly influenced the risk of adult astrocytoma. Compared with 2G/2G, the predominant genotype among the study population, individuals with the 1G/1G genotype significantly decreased the susceptibility to astrocytoma. 3 The MMP-7-181A/G polymorphism significantly influenced the risk of adult astrocytoma development. Individuals harboring the G allele (A/G and G/G genotypes) significantly increased the susceptibility to astrocytoma. This association was more significant among male and people younger than 45 years old.Part II: Relationship between expression of MMP-1,3,7,9 and clinical pathology in adult astrocytomaMethodsParaffin embedded blocks of tumor tissues from 70 pathologic diagnosed and graded astrocytoma cases and of non-tumor tissues from eight brain trauma patients were selected for detecting protein expression. MMP-1, 3, 7, 9 proteins were detected by immunohistochemistry (IHC) using the S-P staining kit. The status of the MMP expressions among histological grading groups was compared using Chi-square test. The correlation of the MMP expression with the pathologic grading was tested by the Kendall’s tau-b method with the significance level of 5%.Results1 The MMP-1 protein expression rate among non-tumor tissues and grade I～IV astrocytoma tissues was 12.5%, 75%, 86%, 90%, 100%, respectively. The MMP-1 expression among the above five groups was significantly different (X2=28.204,P<0.01). The MMP-1 expression was significantly correlated with the disease progression(tb=0.463,P<0.001). Thus, the MMP-1 expression significantly increased with the histological grading.2 The MMP-3 protein expression rate among non-tumor tissues and grade I～IV astrocytoma tissues was 25%, 33%, 45.5%, 35%, 75%, respectively. The MMP-3 expression among the above five groups was not significantly different (X2=20.476,P=0.059). However, the MMP-3 expression was significantly correlated with the disease progressionand significantly increased with the histological grading(tb=0.276,P=0.005).3 The MMP-7 protein expression rate among non-tumor tissues and grade I～IV astrocytoma tissues was 37.5%, 50%, 77.3%, 75%, 100%, respectively. The MMP-7 expression among the above five groups was not significantly different (X2=29.779,P=0.003). The MMP-7 expression was significantly correlated with the disease progression and significantly increased with the histological grading(tb=0.416,P<0.001).4 The MMP-9 protein expression rate among non-tumor tissues and grade I～IV astrocytoma tissues was 0%, 66.7%, 81.8%, 80%, 100%, respectively. The MMP-9 expression among the above five groups was not significantly different (X2=44.038,P<0.001). The MMP-9 expression was significantly correlated with the disease progression and significantly increased with the histological grading(tb=0.427,P<0.001).ConclusionsThe MMP-1,3,7,9 protein expression was significantly correlated to the tumor differentiation and may be useful molecular markers to predicate the prognosis of astrocytoma. Part III Expression and significance of MMP-1,3,7,9 mRNA in astrocytomaMethodsTumor tissues from 39 astrocytoma cases and non-tumor brain tissues from 12 brain trauma patients were frozen in liquid nitrogen in 30 minutes after resection and stored in a -80℃refrigerator. Total cellular RNA was extracted from tested tissues using the one-step RNA extraction kit. The first cDNA strand was generated by a reverse transcription system. The MMP mRNA from tumor and non-tumor brain tissues was quantitatively detected by the real-time fluorescence quantitative polymerase chain reaction (RQ-PCR) and TaqMan probe technology, with GAPDH gene as an internal reference.The MMPs mRNA expression level was calculated by the ratio of MMPs to GAPDH and expressed as mean±standard deviation. Student’s T test was used to compare the mRNA expression between the pathologic groups with a significance level of 5%.Results1 The MMP-9 mRNA expression in tumor tissues of astrocytoma was significantly higher than that in non-tumor tissues (p=0.038). The MMP-9 mRNA expression in low differentiation tissues (pathologic gradeⅢ-Ⅳ) was significantly higher than that in both of the high differentiation tissues (pathologic gradeⅠ-Ⅱ) and non-tumor brain tissues ( P value was =0.001 and <0.001, respectively). However, the significant difference between the high differentiation group and non-tumor group was not observed.2 The MMP-1, 3 mRNA expression among low differentiation tissues (pathologic gradeⅢ-Ⅳ),high differentiation tissues (pathologic gradeⅠ-Ⅱ), and non-tumor brain tissues was not significantly different ( P value was 0.26 and 0.419, respectively).Conclusions1 The MMP-9 mRNA expression in low differentiation tissues (histological gradeⅢ-Ⅳ) was significantly higher than that in both of the high differentiation tissues (pathologic gradeⅠ-Ⅱ) and non-tumor brain tissues, suggesting that detection of the MMP-9 mRNA may be used to estimate the tumor histological differentiation and disease prognosis.2 The MMP-1 and MMP-3 mRNA expression among astrocytoma tissues was not significantly different from that in non-tumor brain tissues, indicating that detection of the MMP-1 and MMP-3 mRNA may be not applicable for estimating the biological behavior of astrocytoma.In summary, this pilot study showes that the polymorphisms in promoter regions of the MMP-1, 7 genes are involved in the carcinogenesis of adult astrocytoma and may be used as molecular markers, alone or together with other polymorphic genes, to screen individuals those at risk to develop astrocytoma. The high expression of the MMP-1,3,7,9 proteins can be used as candidate markers to estimate the progression and prognosis of adult astrocytoma. Additionally, detection of the MMP-9 mRNA may be used as an indicator of malignant characteristics of astrocytoma.更多还原