Hansel Donna E, Rahman Ayman, House Michael, Ashfaq Raheela, Berg Karin, Yeo Charles J, Maitra Anirban
Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA.
Clin Cancer Res. 2004 Sep 15;10(18 Pt 1):6152-8. doi: 10.1158/1078-0432.CCR-04-0285.
Pancreatic endocrine neoplasms are neoplastic proliferations of islet cells or islet cell precursors and are capable of secreting a variety of synthetic products, including insulin, glucagon, gastrin, and vasoactive intestinal peptide. The biological behavior of pancreatic endocrine neoplasms is often unpredictable, and there are few reliable histopathologic criteria reliably correlating with metastatic ability. We have used the Affymetrix U133 GeneChip set (HG_U133 A and B; Affymetrix; Santa Clara, CA) representing approximately 33,000 characterized transcripts to examine global gene expression profiles from well-differentiated nonmetastatic (n=5) and metastatic (n=7) pancreatic endocrine neoplasms to determine molecular markers that predict disease progression. Microarray hybridization data were normalized using the GeneLogic GeneExpress Software System to identify differentially up- and down-regulated genes in metastatic versus nonmetastatic pancreatic endocrine neoplasms. Using a 3-fold change in gene expression as a threshold, we have identified 65 overexpressed and 57 underexpressed genes in metastatic pancreatic endocrine neoplasms as compared with nonmetastatic pancreatic endocrine neoplasms. Several classes of genes, including growth factors and growth factor-related molecules (IGFBP1, IGFBP3, and MET), developmental factors (TBX3 and MEIS2), cytoskeletal factors (beta 1 tubulin and ACTN2), cholesterol homeostasis mediators (LRP5, SLC27A2, and RXRG), intracellular signaling pathway mediators (DYRK1A, PKIB, and AK2), methyltransferases (MGMT and GAMT), and DNA repair and regulatory molecules (CHEK1 and ZNF198), were identified as differentially over- or underexpressed via this method. Immunohistochemical validation of microarray data were performed for two overexpressed genes, namely, the met proto-oncogene (MET) and insulin-like growth factor binding protein 3 (IGFBP3) with tissue microarrays of nonmetastatic (n=24) and metastatic (n=15) pancreatic endocrine neoplasms. Increased expression of IGFBP3 was confirmed in metastatic versus nonmetastatic pancreatic endocrine neoplasms (12 of 15, 80% versus 10 of 24, 42%), as well as in lymph node (6 of 7, 86%) and liver (9 of 9, 100%) metastases. Similarly, overexpression of MET was confirmed in metastatic versus nonmetastatic pancreatic endocrine neoplasms (5 of 15, 33% versus 4 of 24, 17%), as well as in lymph node metastases (4 of 7, 57%) and liver metastases (5 of 9, 56%). The majority of genes that demonstrated altered expression has not been previously identified as differentially expressed in metastatic pancreatic endocrine neoplasm lesions and may therefore represent newly identified molecules in the progression of these lesions.
胰腺内分泌肿瘤是胰岛细胞或胰岛细胞前体的肿瘤性增殖，能够分泌多种合成产物，包括胰岛素、胰高血糖素、胃泌素和血管活性肠肽。胰腺内分泌肿瘤的生物学行为往往不可预测，几乎没有可靠的组织病理学标准能与转移能力可靠相关。我们使用了代表约33000个已表征转录本的Affymetrix U133基因芯片组（HG_U133 A和B；Affymetrix；加利福尼亚州圣克拉拉）来检测高分化非转移性（n = 5）和转移性（n = 7）胰腺内分泌肿瘤的整体基因表达谱，以确定预测疾病进展的分子标志物。使用GeneLogic GeneExpress软件系统对微阵列杂交数据进行标准化，以识别转移性与非转移性胰腺内分泌肿瘤中差异上调和下调的基因。以基因表达3倍变化为阈值，我们已确定与非转移性胰腺内分泌肿瘤相比，转移性胰腺内分泌肿瘤中有65个基因过表达，57个基因低表达。通过这种方法鉴定出几类基因存在差异过表达或低表达，包括生长因子和生长因子相关分子（IGFBP1、IGFBP3和MET）、发育因子（TBX3和MEIS2）、细胞骨架因子（β1微管蛋白和ACTN2）、胆固醇稳态调节因子（LRP5、SLC27A2和RXRG）、细胞内信号通路调节因子（DYRK1A、PKIB和AK2）、甲基转移酶（MGMT和GAMT）以及DNA修复和调节分子（CHEK1和ZNF198）。使用非转移性（n = 24）和转移性（n = 15）胰腺内分泌肿瘤的组织微阵列对微阵列数据进行免疫组织化学验证，验证两个过表达基因，即原癌基因（MET）和胰岛素样生长因子结合蛋白3（IGFBP3）。与非转移性胰腺内分泌肿瘤相比，IGFBP3在转移性胰腺内分泌肿瘤中的表达增加得到证实（15例中的12例，80% 对24例中的10例, 42%），在淋巴结转移（7例中的6例，8）和肝转移（9例中的9例，100%）中也是如此。同样，与非转移性胰腺内分泌肿瘤相比，MET在转移性胰腺内分泌肿瘤中的过表达得到证实（15例中的5例，33% 对24例中的4例，17%），在淋巴结转移（7例中的4例，57%）和肝转移（9例中的5例，56%）中也是如此。大多数表达改变的基因此前未被确定在转移性胰腺内分泌肿瘤病变中存在差异表达, 因此可能代表这些病变进展过程中新发现的分子。
