Selective CREB-dependent cyclin expression mediated by the PI3K and MAPK pathways supports glioma cell proliferation.

The cyclic-AMP response element binding (CREB) protein has been shown to have a pivotal role in cell survival and cell proliferation. Transgenic rodent models have revealed a role for CREB in higher-order brain functions, such as memory and drug addiction behaviors. CREB overexpression in transgenic animals imparts oncogenic properties on cells in various tissues, and aberrant CREB expression is associated with tumours. It is the central position of CREB, downstream from key developmental and growth signalling pathways, which gives CREB this ability to influence a spectrum of cellular activities, such as cell survival, growth and differentiation, in both normal and cancer cells. We show that CREB is highly expressed and constitutively activated in patient glioma tissue and that this activation closely correlates with tumour grade. The mechanism by which CREB regulates glioblastoma (GBM) tumour cell proliferation involves activities downstream from both the mitogen-activated protein kinase and phosphoinositide 3-kinase (PI3K) pathways that then modulate the expression of three key cell cycle factors, cyclin B, D and proliferating cell nuclear antigen (PCNA). Cyclin D1 is highly CREB-dependent, whereas cyclin B1 and PCNA are co-regulated by both CREB-dependent and -independent mechanisms. The precise regulatory network involved appears to differ depending on the tumour-suppressor phosphatase and tensin homolog status of the GBM cells, which in turn allows CREB to regulate the activity of the PI3K itself. Given that CREB sits at the hub of key cancer cell signalling pathways, understanding the role of glioma-specific CREB function may lead to improved novel combinatorial anti-tumour therapies, which can complement existing PI3K-specific drugs undergoing early phase clinical trials.

Stargazin and AMPA receptor membrane expression is increased in the somatosensory cortex of Genetic Absence Epilepsy Rats from Strasbourg.

Absence-like seizures in the Genetic Absence Epilepsy Rats from Strasbourg (GAERS) model are believed to arise in hyperexcitable somatosensory cortical neurons, however the cellular basis of this increased excitability remains unknown. We have previously shown that expression of the Transmembrane AMPA receptor Regulatory Protein (TARP), stargazin, is elevated in the somatosensory cortex of GAERS. TARPs are critical regulators of the trafficking and function of AMPA receptors. Here we examine the developmental expression of stargazin and the impact this may have on AMPA receptor trafficking in the GAERS model. We show that elevated stargazin in GAERS is associated with an increase in AMPA receptor proteins, GluA1 and GluA2 in the somatosensory cortex plasma membrane of adult epileptic GAERS. Elevated stargazin expression is not seen in the epileptic WAG/Rij rat, which is a genetically distinct but phenotypically similar rat model also manifesting absence seizures, indicating that the changes seen in GAERS are unlikely to be a secondary consequence of the seizures. In juvenile (6 week old) GAERS, at the age when seizures are just starting to be expressed, there is elevated stargazin mRNA, but not protein expression for stargazin or the AMPA receptor subunits. In neonatal (7 day old) pre-epileptic GAERS there was no alteration in stargazin mRNA expression in any brain region examined. These data demonstrate that stargazin and AMPA receptor membrane targeting is altered in GAERS, potentially contributing to hyperexcitability in somatosensory cortex, with a developmental time course that would suggest a pathophysiological role in the epilepsy phenotype.

Genetic absence epilepsy rats from Strasbourg have increased corticothalamic expression of stargazin.

Stargazin is membrane bound protein involved in trafficking, synapse anchoring and biophysical modulation of AMPA receptors. A quantitative trait locus in chromosome 7 containing the stargazin gene has been identified as controlling the frequency and duration of absence seizures in the Genetic Absence Epilepsy Rats from Strasbourg (GAERS). Furthermore, mutations in this gene result in the Stargazer mouse that displays an absence epilepsy phenotype. GAERS stargazin mRNA expression is increased 1.8 fold in the somatosensory cortex and by 1.3 fold in the thalamus. The changes were present before and after the onset of absence seizures indicating that increases are not a secondary consequence of the seizures. Stargazin protein expression was also significantly increased in the somatosensory cortex after the onset of spontaneous seizures. The results are of significant importance beyond the GAERS model, as they are the first to show that an increase in stargazin expression may be pro-epileptic.

Differential coding potential of ADAM22 mRNAs.

ADAM22 is one of three catalytically inactive ADAM family members highly expressed in the brain. Preliminary functional studies suggest possible roles in epilepsy and myelination. We report an additional eight new splice variants of human ADAM22. Analysis of the altered splicing patterns of ADAM22 mRNAs in glioma allows us to suggest alternate splicing patterns in normal brain compared to glioma may represent differential use of exon 32. We also report diversity in the 5' leader sequences of ADAM22 mRNAs as a consequence of alternate transcriptional initiation sites. ADAM22 has an additional transcriptional initiation element producing transcripts lacking the exon 1 sequence including the signal peptide. Variable transcriptional initiation in exon 1 produces a range of ADAM22 5' leader sequence lengths, all of which are significantly longer than those described in NCBI reference sequences. Longer 5' leader sequences contain a second upstream AUG codon which acts to inhibit ADAM22 translation.

Synergy between Apc min and an activated ras mutation is sufficient to induce colon carcinomas.

Colon carcinomas appear to arise from the cumulative effect of mutations to several genes (APC, DCC, p53, ras, hMLH1, and hMSH2). By using novel colonic epithelial cell lines derived from the Immorto mouse, named the YAMC (young adult mouse colon) cell line, and an Immorto-Min mouse hybrid, named the IMCE (Immorto-Min colonic epithelial) cell line, carrying the Apc min mutation, we investigated the effect of an activated v-Ha-ras gene on tumor progression. The YAMC and IMCE cell lines are normal colonic epithelial cell lines which are conditionally immortalized by virtue of expression of a temperature-sensitive simian virus 40 (SV40) large T antigen. Under conditions which permit expression of a functional SV40 large T antigen (33 degrees C plus gamma interferon), neither the YAMC nor the IMCE cell line grows in soft agar or is tumorigenic in nude mice. In vitro, when the SV40 large T antigen is inactivated (39 degrees C without gamma interferon), the cells stop proliferating and die. By infecting the YAMC and IMCE cell lines with a replication-defective psi2-v-Ha-ras virus, we derived cell lines which overexpress the v-Ha-ras gene (YAMC-Ras and IMCE-Ras). In contrast to the parental cell lines, under conditions in which the SV40 large T antigen is inactive, both the YAMC-Ras and IMCE-Ras cell lines continue to proliferate. Initally YAMC-Ras cells do not form tumors; however, tumors are visible after 90 days of incubation. IMCE-Ras cells form colonies in soft agar under both permissive and nonpermissive culture conditions. Furthermore, IMCE-Ras cells form tumors in nude mice within 3 weeks. The phenotype of the IMCE-Ras cell line thus clearly demonstrates that a defective Apc allele and an activated ras gene are sufficient to transform normal colonic epithelial cells and render them tumorigenic.

Asparagine 26, glutamic acid 31, valine 45, and tyrosine 64 of Ras proteins are required for their oncogenicity.

Ras and Rap1 proteins are related GTP-dependent signal transducers which require Gly-12, the effector domain (residues 32-40), and Ala-59 for stimulation of their GTPase activities by GAP1 and GAP3, respectively. The replacement of Gly-12 by Val or Ala-59 by Thr potentiates the Ras oncogenicity and Rap1A antioncogenicity. However, the mutations in the effector domain, in particular the replacement of Thr-35 by Ala, abolish both Ras oncogenicity and Rap1A antioncogenicity, indicating that the effector domain is involved in interactions of these signal transducers with their targets as well as the GAPs. In this paper, we demonstrate that (i) replacement of Tyr-64 of the Ha-Ras protein or Phe-64 of the Rap1A protein by Glu or other non-hydrophobic amino acids reduces their intrinsic GTPase activities and abolishes their stimulation by GAP1 or GAP3, respectively, (ii) replacement of Tyr-64 by Gly and other non-hydrophobic amino acids results in complete loss of the oncogenicity of the v-Ha-Ras protein, indicating that the hydrophobic residue 64, in addition to the known effector domain, is essential for the Ras protein to interact with its target as well as GAP1. In addition we have found that Asn-26, Glu-31, and Val-45 of the v-Ha-Ras protein are required for its oncogenicity. Replacement of the Ras residues at either positions 26, 31, or 45 by the corresponding Rap1A residues abolishes the Ras oncogenicity.

The residues of Ras and Rap proteins that determine their GAP specificities.

The oncogenic transformation of a normal fibroblast by mutated Ras genes can be reversed by overexpression of a Ras-related gene called Rap1A (or Krev1). Both Ras and Rap1A proteins are G proteins and appear to serve as signal transducers only in the GTP-bound form. Therefore, GAP1 and GAP3, which stimulate the intrinsic GTPase activities of normal Ras and Rap1A proteins, respectively, serve as attenuators of their signal transducing activities. In this paper, we describe the enzymatic properties of several mutated Rap1A and chimeric Ras/Rap1A (or -1B) proteins which lead to the following conclusions: (i) the GAP3-dependent activation of both Rap1A and -1B GTPases requires Gly12, but neither Thr61 nor Gln63; (ii) residues 64 to 70 of the Rap1 GTPases are sufficient to determine their specificities for GAP3; and (iii) residues 61 to 65 of the Ras GTPases are sufficient for determining their specificities for GAP1. Thus, the domains of the Ras or Rap1 proteins that determine whether their signals are attenuated by GAP1 or GAP3 are distinct from the N-terminal domain (residues 21 to 54) that determines whether their signals are oncogenic or antioncogenic. The Arg12 mutant of chimeric HaRas(1-54)/Rap1A(55-184) protein has been previously reported to be oncogenic (Zhang, K., Noda, M., Vass, W. C., Papageorge, A.G., and Lowy, D.R. (1990) Science 249, 162-165). In this paper, we show that the Val12 mutant of chimeric HaRas(1-54)/Rap1B(55-184) protein is also oncogenic, suggesting that the C-terminal geranylgeranylation of the Rap 1B protein can replace functionally the C-terminal farnesylation of the Ras protein to allow the G protein to be oncogenic.