A reporter assay for target validation in primary neuronal cultures.

The deposition of beta-amyloid peptides (Abeta42 and Abeta40) in neuritic plaques is one of the hallmarks of Alzheimer's disease (AD), and genes modulating their brain levels and neuronal effects could result in future disease modifying therapies. The causal association of candidate targets with AD is of paramount importance in current drug discovery, as a lack of efficacy of many candidate drugs is often due to inadequate validation of their pharmacological target. In Alzheimer's as well as in other neurodegenerative diseases, in vitro target validation is hampered by the difficulty of transfecting primary neuronal cultures and assaying the effects of genes on neuronal viability. Here we describe a rapid, sensitive and simple reporter-based assay for the validation of genes putatively associated with Abeta-mediated neurotoxicity, which can in principle be extended to the validation of targets in the context of other neuronal insults. The assay is suitable for the generation of robust and reproducible data in primary neuronal cultures allowing the dissection at a molecular level of complex pathways activated by the toxic insult in a cellular context that more closely represents the real disease situation.

A novel target for the Wilms' tumour suppressor protein (WT1) is bound by a unique combination of zinc fingers.

All isoforms of the Wilms' tumour suppressor protein, WT1, contain four consecutive zinc fingers which facilitate DNA binding. The predominant WT1 transcript contains a 9 base pair insertion resulting in an additional three amino acids, lysine-threonine-serine (KTS), between zinc fingers 3 and 4. WT1 zinc fingers 2, 3 and 4 are highly homologous to the zinc fingers of the early growth response gene, EGR1. However, only WT1--KTS is capable of binding an EGR1 consensus site. In contrast, the previously described genomic fragment, +P5 (D1S3309E), is bound by both WT1--KTS and WT1 + KTS. In this study, the region within + P5 to which both WT1 -- KTS and WT1 + KTS bind was defined as 5'-GGAGAGGGAGGATC-3'. EGR1 did not bind + P5. By creating zinc finger deletions, we demonstrate that zinc finger 1, but not zinc finger 4, is critical for + P5 binding; whereas zinc finger 4, but not 1, is necessary for the binding of WT1 target sites within EGR1, PDGF A chain and IGF2 promoters. Thus, zinc finger usage can vary with target and + P5 may represent a novel type of WT1 binding site, the physiological relevance of which must be investigated.

Human growth-differentiation factor 3 (hGDF3): developmental regulation in human teratocarcinoma cell lines and expression in primary testicular germ cell tumours.

We describe the cloning and initial characterization of a novel cDNA from human embryonal carcinoma (EC) cells. This cDNA, which we named human growth differentiation factor 3 (hGDF3), encodes the homologue of mouse GDF3, a TGFbeta superfamily member belonging to the Growth/Differentiation Factors. We have analysed the expression of hGDF3 in human embryonal carcinoma cell lines and in primary testicular germ cell tumours of adolescents and adults (TGCTs). Expression of hGDF3 in human EC cell lines is stem cell-specific, is down-regulated upon RA-mediated differentiation and is increased upon culture of the cells in the presence of activin A. In TGCTs, hGDF3 expression is low in seminomas, while expression in non-seminomas is readily detectable and appears to be associated with the EC and yolk sac components in the tumours. We have also mapped the hGDF3 locus to the short arm of human chromosome 12, a region consistently overrepresented in human testicular germ cell tumours. Thus, hGDF3 represents an embryonal carcinoma stem cell-associated marker both in vitro and in vivo.

Cloning and characterization of the human phosphoinositide-specific phospholipase C-beta 1 (PLC beta 1).

Phospholipase C-beta (PLC beta) catalyses the generation of inositol 1,4,5-trisphosphate (IP(3)) and diacylglycerol (DAG) from phosphatidylinositol 4,5-bisphosphate (IP(2)), a key step in the intracellular transduction of a large number of extracellular signals, including neurotransmitters and hormones modulating diverse developmental and functional aspects of the mammalian central nervous system. Four mammalian isozymes are known (PLC beta 1-4), which differ in their function and expression patterns in vivo. We have characterized the human PLC beta 1 genomic locus (PLC beta 1), cloned two distinct PLC beta 1 cDNAs (PLC beta 1a and b) and analysed their respective expression patterns in a comprehensive panel of human tissues using quantitative TaqMan technology. The two cDNAs derive from transcripts generated through alternative splicing at their 3' end, and are predicted to encode for PLC beta 1 isoforms differing at their carboxy-terminus. The human PLC beta 1 isoforms are co-expressed in the same tissues with a distinctly CNS-specific profile of expression. Quantitative differences in PLC beta 1 isoform expression levels are observed in some tissues. Transient expression of epitope-tagged versions of the two isoforms followed by immunofluorescence revealed localization of the proteins to the cytoplasm and the inner side of the cell membrane. Finally, we characterized the structure of the PLC beta 1 locus and confirmed its mapping to human chromosome 20.

An early increase in the disialoganglioside GD3 contributes to the development of neuronal apoptosis in culture.

We induced apoptosis in primary cultures of cerebellar granule neurons by switching the growing medium into a medium containing lower concentrations of K(+) (5 or 10 mM instead of 25 mM) or, alternatively, by addition of staurosporine. The apoptotic phenotype was always preceded by an early increase in the intracellular levels of the disialoganglioside GD3, which peaked at 2-6 h and returned back to normal at 12 h. GD3 synthase, the enzyme that forms GD3 from the monosialoganglioside GM3, was also induced at early times after the induction of apoptosis in granule cells. Immunofluorescent staining showed that GD3 increased in neuronal cell bodies and neurites, but was never localized in cell nuclei. In cultures switched into a low K(+)-containing medium, exogenously applied GD3, but not the disialoganglioside GD1a, accelerated the development of neuronal apoptosis. In contrast, the antisense-induced knock-down of GD3 synthase was protective against granule cell death induced by lowering extracellular K(+) from 25 to 10 - but not 5 - mM. These results demonstrate that an early and transient increase in GD3 synthesis is one of the factors that contribute to the induction of neuronal apoptosis in culture.

Two sides of the same coin: Wnt signaling in neurodegeneration and neuro-oncology.

Wnts function through the activation of at least three intracellular signal transduction pathways, of which the canonical beta-catenin mediated pathway is the best understood. Aberrant canonical Wnt signaling has been involved in both neurodegeneration and cancer. An impairment of Wnt signals appears to be associated with aspects of neurodegenerative pathologies while overactivation of Wnt signaling is a common theme in several types of human tumors. Therefore, although therapeutic approaches aimed at modulating Wnt signaling in neurodegenerative and hyperproliferative diseases might impinge on the same molecular mechanisms, different pharmacological outcomes are required. Here we review recent developments on the understanding of the role of Wnt signaling in Alzheimer's disease and CNS tumors, and identify possible avenues for therapeutic intervention within a complex and multi-faceted signaling pathway.

Bone morphogenetic proteins (BMPs) induce epithelial differentiation of NT2D1 human embryonal carcinoma cells.

Human embryonal carcinoma (EC) cells represent the stem cells of testicular germ cell tumours (TGCTs) and are morphologically, antigenically and functionally related to the stem cells of early mammalian embryos. Despite the large capacity for differentiation displayed by TGCT stem cells, little is known of the factors controlling their developmental potency. We have analyzed the differentiation elicited in NT2D1 human embryonal carcinoma (EC) cells by Bone Morphogenetic Proteins (BMPs) and compared it with that elicited by retinoic acid (RA). We have found that while RA induced expression of neuronal, endodermal and epithelial markers in NT2D1 human EC cells, treatment with BMPs resulted in a predominantly epithelial phenotype. We also provide evidence to suggest that at least some of the effects elicited by RA in human EC cells might be mediated through RA-induced expression of BMP-7. Thus BMPs may play an important role in specifying the type of differentiation arising from human multipotent stem cells. The manipulation of BMP signalling in human embryonic multipotent stem cells may therefore prove a useful approach in attempts to generate specific differentiated cell types in vitro, and loss of the malignant and/or transformed phenotype.

A luciferase-reporter vector with blue-white selection for rapid subcloning and mutational analysis of eukaryotic promoters.

A luciferase-reporter plasmid vector has been constructed for the rapid subcloning and analysis of eukaryotic promoters. Identification of recombinants, when using this vector, is greatly facilitated by the retention of blue-white selection (alpha-complementation).

Repression of promoters for the mouse insulin-like growth factor II-encoding gene (Igf-2) by products of the Wilms' tumour suppressor gene wt1.

We have investigated actions of the Wilms' tumour suppressor zinc finger transcription factor, WT1, on promoters of the mouse insulin-like growth factor II-encoding gene (Igf-2). Two variant forms of WT1 repressed the two major Igf-2 promoters (P2 and P3) in transient transfection assays. WT1-binding sites were characterised in both these promoters and in the transcribed region downstream from P2, exon 2. In each of these regions, there was a pair of WT1-binding sites, and mutational analysis of the exon-2 sites indicated that both were required for full repression. Cooperative binding of WT1 to these sites might explain the dominant-negative mutations of WT1 observed in some Wilms' tumours and Denys-Drash syndrome cases.

Imipramine treatment up-regulates the expression and function of mGlu2/3 metabotropic glutamate receptors in the rat hippocampus.

We examined the effect of a chronic imipramine treatment (10 mg/kg, i.p., once daily for 21 days) on the expression and function of metabotropic glutamate (mGlu) receptors in discrete regions of the rat brain. Chronic imipiramine treatment up-regulated the expression of mGlu2/3 receptor proteins in the hippocampus, nucleus accumbens, cerebral cortex and corpus striatum. Expression of mGlu1a receptor protein was increased exclusively in the hippocampus, whereas no changes in the expression of mGlu4 and mGlu5 receptors or Homer-1a protein were detected. Using hippocampal slices, we examined the stimulation of polyphosphoinositide (PI) hydrolysis induced by mGlu receptor agonists in control and imipramine-treated rats. Imipramine treatment amplified the PI response to the non subtype-selective mGlu receptor agonist, 1S,3R-aminocyclopentane-1,3-dicarboxylated (1S,3R-ACPD) in both hippocampal and cortical slices, but failed to affect the response to the selective mGlu1/5 receptor agonist, S-3,5-dihydroxyphenylglycine (DHPG). Amplification was restored when DHPG was combined with the selective mGlu2/3 receptor agonist, LY379268. In addition, 1S,3R-ACPD-stimulated PI hydrolysis was no longer enhanced in imipramine-treated rats when the mGlu2/3 component of the PI response was abrogated by the antagonist, LY341495. In contrast, the ability of LY379268 to inhibit forskolin-stimulated cAMP formation was reduced in hippocampal slices of rats chronically treated with imipramine. Taken together, these results suggest that neuroadaptive changes in the expression and function of mGlu2/3 receptors occur in response to chronic antidepressants.

Endogenous activation of group-II metabotropic glutamate receptors inhibits the hypothalamic-pituitary-adrenocortical axis.

Systemic injection of the mGlu2/3 receptor antagonist, LY341495 (1 mg/kg, i.p.), increased plasma corticosterone in mice to an extent similar to that induced by the despair test. Treatment with the mGlu2/3 receptor agonist, LY379268 (1 mg/kg, i.p.), or the non-competitive mGlu5 receptor antagonist, MPEP (5 mg/kg, i.p.), failed to induce significant changes in corticosterone levels. Searching for a site of action of LY341495, we examined the expression of mGlu receptor subtypes in the various anatomical regions of the mouse hypothalamic-pituitary-adrenal (HPA) axis. Only mGlu5 and -7 receptor mRNAs were detected in the adrenal gland by RT-PCR, whereas mGlu -1, -3, -4, -5, -7 and -8 receptor mRNAs were detected in the anterior pituitary. All transcripts (with the exception of mGlu5 and mGlu6 receptor mRNAs) were detected in the hypothalamus. However, Western blot analysis showed the presence of mGlu2/3 receptor proteins only in the hypothalamus and not in the anterior pituitary. This was consistent with functional data showing that LY341495 (0.1 and 1 microM) failed to affect ACTH secretion from isolated mouse anterior pituitaries. Moving from these observations, we examined whether LY341495 could activate the HPA axis by inhibiting mGlu2/3 receptors at hypothalamic level. We measured the release of corticotropin releasing hormone (CRH) in isolated mouse hypothalami incubated in the presence of subtype-selective mGlu receptor agonists or antagonists. Among all the drugs we have tested, only LY341495 was able to increase CRH secretion. With high concentrations of LY341495 (1 microM) this increase was similar to that induced by 50 mM K(+). The action of LY341495 was prevented by the combined application of the mGlu2/3 receptor agonist, LY379268. We conclude that group-II mGlu receptors tonically regulate the HPA axis by controlling CRH secretion at hypothalamic level.

Relationship between learning, stress and hippocampal brain-derived neurotrophic factor.

Brain-derived neurotrophic factor (BDNF) expression in the hippocampus is reduced in response to acute, as well as repeated immobilization stress. This effect might be mediated by corticosterone, because corticosterone administration is known to reduce hippocampal BDNF. However, rats subjected to a learning paradigm showed an increased BDNF expression in the hippocampus despite the high corticosterone levels found during the test. To dissect the relative contributions of learning and stress to the overall changes in BDNF levels we set up an experimental model in which two groups of rats received the same amount of stress, but only one group had the possibility to learn how to avoid it. Using this model, we now report that learning and stress exert an opposite modulation on BDNF levels in the hippocampus, and that the increasing effect of learning predominates over the decreasing effect of stress.

Activation of Fas receptor is required for the increased formation of the disialoganglioside GD3 in cultured cerebellar granule cells committed to apoptotic death.

Apoptosis was induced in cultured cerebellar granule cells by lowering extracellular K+ concentrations (usually from 25 to 10 mM). The apoptotic phenotype was preceded by an early and transient increase in the intracellular levels of the disialoganglioside, GD3, which behaves as a putative pro-apoptotic factor. We examined whether activation of Fas receptor mediates the increase in GD3 formation in granule cells committed to die. Degenerating granule cells showed increased expression of both Fas receptor and its ligand (Fas-L), at times that coincided with the increase in GD3 levels and the induction of GD3 synthase mRNA. Addition of neutralizing anti-Fas-L antibodies reduced the extent of 'low-K+'-induced apoptosis and abolished the increase in GD3 levels and GD3 synthase mRNA. Similar reductions were observed in cultures prepared from gld or lpr mice, which harbor loss-of-function mutations of Fas-L and Fas receptor, respectively. In addition, exogenous application of soluble Fas-L further enhanced both the increase in GD3 formation and cell death in cultured granule cells switched from 25 into 10 mM K+. We conclude that activation of Fas receptor is entirely responsible for the increase in GD3 levels and contributes to the development of apoptosis by trophic deprivation in cultured cerebellar granule cells.

Transactivation of mouse insulin-like growth factor II (IGF-II) gene promoters by the AP-1 complex.

The mouse insulin-like growth factor II gene (Igf2) is transcribed from three promoters (P1, P2 and P3), and is expressed in a tissue-specific and developmentally regulated fashion; however, little information is available on the transcription factors controlling Igf2 expression. The AP-1 complex is a transcription factor involved in the regulation of a variety of genes, including those encoding certain growth factors. We show that Igf2 P3 is transactivated by AP-1 in a transient expression assay, and that this effect is mediated through two non-consensus AP-1 binding sites characterised by DNA-protein interaction studies. Mutational analysis indicates these sites are required for AP-1 responsiveness and full promoter activity.

Wilms' tumour-suppressor protein isoforms have opposite effects on Igf2 expression in primary embryonic cells, independently of p53 genotype.

The p53 protein has been proposed as a modulator of the Wilms' tumour-suppressor protein (WT1) transcriptional regulation activity. To investigate this putative p53 role, the promoter P3 of the mouse insulin-like growth factor II gene (Igf2) was used as a target for WT1 regulation in primary cell cultures derived from p53 wild-type (p53+/+) and knock-out (p53-/-) mouse embryos. In these cells, the WT1 transcriptional activity was observed to be independent of p53 genotype. Furthermore, the two WT1 zinc finger (ZF) isoforms were for the first time found to have opposite effects on gene expression from a single promoter in the same cell type, WT1[-KTS] activating Igf2 P3, whereas WT1[+KTS] repressed its activity. In addition, we have mapped the WT1 binding sites and investigated the effect on WT1 binding activity of individual ZF deletions and Denys-Drash syndrome point mutations to this target.

Bone morphogenetic proteins and retinoic acid induce human endogenous retrovirus HERV-K expression in NT2D1 human embryonal carcinoma cells.

Expression of the HERV-K human endogenous retrovirus is very low in normal and tumor tissue, but is readily detected in testicular germ cell tumors (TGCT). NT2D1 human embryonal carcinoma cells represent in vitro models for the stem cells of TGCT, and can be differentiated by treatment with bone morphogenetic proteins (BMP) or retinoic acid (RA). In a search for BMP target genes in NT2D1 cells, HERV-K was identified as an early BMP and RA target. It was shown that HERV-K expression was induced upon treatment of NT2D1 cells with BMP or with RA, but not with activin or transforming growth factor (TGF)-beta. Induction of HERV-K expression was rapid but transient, with transcripts becoming undetectable in differentiated NT2D1 cultures. Thus NT2D1 cells provide a suitable in vitro system for the study of the factors controlling HERV-K expression during cellular differentiation, which may play a role in HERV-K expression in TGCT.

Identification, gene structure, and expression of human frizzled-3 (FZD3).

We report the identification, genomic structure, chromosomal localization, and expression analysis of human frizzled-3 (FZD3), a 7-transmembrane receptor belonging to the frizzled family. The cDNA obtained from adult human brain shows 91% identity at the nucleotide level and 98% at the amino acid level to mouse frizzled-3 (fzd3). The FZD3 locus is located on chromosome 8p21, spans 48 Kb and its coding sequence is distributed in 6 exons intercalated by 5 introns. FZD3 is expressed in all analyzed human tissues, with quantitatively higher expression in the CNS and in urogenital structures.

Cloning and analysis of the mouse follistatin promoter.

Follistatin is a secreted protein, which functions as an antagonist of different members of the TGF-beta superfamily, including activin and bone morphogenetic proteins. Expression of follistatin is tightly regulated during mouse development both spatially and temporally. In order to study the regulation of follistatin expression in the mouse embryo we have cloned and analyzed part of the 5' flanking region of the murine follistatin gene. Primer extension and RNase protection assays demonstrate that the murine follistatin promoter region has at least three distinct transcription initiation sites, which are each preceded by a TATA box. All of the transcription initiation sites are located within the first 500 bp upstream of the translational start site. Sequence analysis of this 500 bp region revealed several consensus binding sites for transcription factors including AP-1, Brachyury-T, CREB, Sp1, AP-2 and Tcf. To test whether the 5' region displays promoter activity, we transfected various 5' flanking region deletion constructs into F9 embryonal carcinoma (EC) cells and into P19 EC cells. In these two cell lines a region of only 262 bp upstream of the translation start site could drivereporter expression in a manner that reflects endogenous mRNA expression.

RNA binding by the Wilms tumor suppressor zinc finger proteins.

The Wilms tumor suppressor gene WT1 is implicated in the ontogeny of genito-urinary abnormalities, including Denys-Drash syndrome and Wilms tumor of the kidney. WT1 encodes Kruppel-type zinc finger proteins that can regulate the expression of several growth-related genes, apparently by binding to specific DNA sites located within 5' untranslated leader regions as well as 5' promoter sequences. Both WT1 and a closely related early growth response factor, EGR1, can bind the same DNA sequences from the mouse gene encoding insulin-like growth factor 2 (Igf-2). We report that WT1, but not EGR1, can bind specific Igf-2 exonic RNA sequences, and that the zinc fingers are required for this interaction. WT1 zinc finger 1, which is not represented in EGR1, plays a more significant role in RNA binding than zinc finger 4, which does have a counterpart in EGR1. Furthermore, the normal subnuclear localization of WT1 proteins is shown to be RNase, but not DNase, sensitive. Therefore, WT1 might, like the Kruppel-type zinc finger protein TFIIIA, regulate gene expression by both transcriptional and posttranscriptional mechanisms.

L-Acetylcarnitine induces analgesia by selectively up-regulating mGlu2 metabotropic glutamate receptors.

L-Acetylcarnitine (LAC, 100 mg/kg, s.c.), a drug commonly used for the treatment of painful neuropathies, substantially reduced mechanical allodynia in rats subjected to monolateral chronic constriction injury (CCI) of the sciatic nerve and also attenuated acute thermal pain in intact rats. In both cases, induction of analgesia required repeated injections of LAC, suggesting that the drug induces plastic changes within the nociceptive pathway. In both CCI- and sham-operated rats, a 24-day treatment with LAC increased the expression of metabotropic glutamate (mGlu) receptors 2 and 3 in the lumbar segment of the spinal cord, without changing the expression of mGlu1a or -5 receptors. A similar up-regulation of mGlu2/3 receptors was detected in the dorsal horns and dorsal root ganglia of intact rats treated with LAC for 5-7 days, a time sufficient for the induction of thermal analgesia. Immunohistochemical analysis showed that LAC treatment enhanced mGlu2/3 immunoreactivity in the inner part of lamina II and in laminae III and IV of the spinal cord. An increased mGlu2/3 receptor expression was also observed in the cerebral cortex but not in the hippocampus or cerebellum of LAC-treated animals. Reverse transcription-polymerase chain reaction combined with Northern blot analysis showed that repeated LAC injections selectively induced mGlu2 mRNA in the dorsal horns and cerebral cortex (but not in the hippocampus). mGlu3 mRNA levels did not change in any brain region of LAC-treated animals. To examine whether the selective up-regulation of mGlu2 receptors had any role in LAC-induced analgesia, we have used the novel compound LY 341495, which is a potent and systemically active mGlu2/3 receptor antagonist. LAC-induced analgesia was largely reduced 45 to 75 min after a single injection of LY 341495 (1 mg/kg, i.p.) in both CCI rats tested for mechanical allodynia and intact rats tested for thermal pain. We conclude that LAC produces analgesia against chronic pain produced not only by peripheral nerve injury but also by acute pain in intact animals and that LAC-induced analgesia is associated with and causally related to a selective up-regulation of mGlu2 receptors. This offers the first example of a selective induction of mGlu2 receptors and discloses a novel mechanism for drug-induced analgesia.