C-jun NH2-terminal kinase 2α2 promotes the tumorigenicity of human glioblastoma cells

• Published in: Cancer research (Chicago, Ill.) Vol. 66; no. 20; pp. 10024 - 10031
• Main Authors: JIAN CUI, HAN, Shuang-Yin, CONGLI WANG, WANWEN SU, HARSHYNE, Larry, HOLGADO-MADRUGA, Marina, WONG, Albert J
• Format: Journal Article
• Language: English
• Published: Philadelphia, PA American Association for Cancer Research 15.10.2006
• Subjects: Antineoplastic agents; Biological and medical sciences; Medical sciences; Neurology; Pharmacology. Drug treatments; Tumors; Tumors of the nervous system. Phacomatoses; Human; Malignant glioma; Nervous system diseases; Enzyme; Toxicity; Central nervous system disease; Glioblastoma; Mitogen-activated protein kinase; Tumorigenicity; Malignant tumor; In vitro
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/0008-5472.can-06-0136

Abstract c-Jun NH2-terminal kinases (JNK) are members of the mitogen-activated protein kinase family and have been implicated in the formation of several human tumors, especially gliomas. We have previously shown that a 55 kDa JNK isoform is constitutively active in 86% of human brain tumors and then showed that it is specifically a JNK2 isoform and likely to be either JNK2α2 or JNK2β2. Notably, we found that only JNK2 isoforms possess intrinsic autophosphorylation activity and that JNK2α2 has the strongest activity. In the present study, we have further explored the contribution of JNK2 isoforms to brain tumor formation. Analysis of mRNA expression by reverse transcription-PCR revealed that JNK2α2 is expressed in 91% (10 of 11) of glioblastoma tumors, whereas JNK2β2 is found in only 27% (3 of 11) of tumors. Both JNK2α2 and JNK2β2 mRNAs are expressed in normal brain (3 of 3). Using an antibody specific for JNK2α isoforms, we verified that JNK2α2 protein is expressed in 88.2% (15 of 17) of glioblastomas, but, interestingly, no JNK2α2 protein was found in six normal brain samples. To evaluate biological function, we transfected U87MG cells with green fluorescent protein–tagged versions of JNK1α1, JNK2α2, and JNK2α2APF (a dominant-negative mutant), and derived cell lines with stable expression. Each cell line was evaluated for various tumorigenic variables including cellular growth, soft agar colony formation, and tumor formation in athymic nude mice. In each assay, JNK2α2 was found to be the most effective in promoting that phenotype. To identify effectors specifically affected by JNK2α2, we analyzed gene expression. Gene profiling showed several genes whose expression was specifically up-regulated by JNK2α2 but down-regulated by JNK2α2APF, among which eukaryotic translation initiation factor 4E (eIF4E) shows the greatest change. Because AKT acts on eIF4E, we also examined AKT activation. Unexpectedly, we found that JNK2α2 could specifically activate AKT. Our data provides evidence that JNK2α2 is the major active JNK isoform and is involved in the promotion of proliferation and growth of human glioblastoma tumors through specific activation of AKT and overexpression of eIF4E. (Cancer Res 2006; 66(20): 10024-31)