Oculocerebrorenal Lowe syndrome is a rare X-linked disorder characterized by bilateral cataract, mental retardation and renal Fanconi syndrome. The Lowe syndrome protein Ocrl1 is a PIP2 ...5-phosphatase, primarily localized to the trans-Golgi network (TGN), which ‘loss of function’ mutations result in PIP2 accumulation in patient's cells. Although PIP2 is involved in many cell functions including signalling, vesicle trafficking and actin polymerization, it has been difficult so far to decipher molecular/cellular mechanisms responsible for Lowe syndrome phenotype. We have recently shown that, through its C-terminal RhoGAP domain, Ocrl1 forms a stable complex with Rac GTPase within the cell. In line with this finding, we report here that upon epidermal growth factor induced Rac activation in COS-7 cells, a fraction of Ocrl1 translocates from TGN to plasma membrane and concentrates in membrane ruffles. In order to investigate the functionality of Ocrl1 in plasma membrane, we have analysed PIP2 distribution in human dermal fibroblasts (HDFs) from Lowe patients versus control HDFs. As revealed by both immunodetection and green fluorescent protein–PH binding, PIP2 was found strikingly to accumulate in PDGF induced ruffles in Lowe HDFs when compared with control. This suggests that Ocrl1 is active as a PIP2 5-phosphatase in Rac induced membrane ruffles. Cellular properties such as cell migration and establishment of cell–cell contacts, which depend on ruffling and lamellipodia formation, should be further investigated to understand the pathophysiology of Lowe syndrome.
OCRL mutations are associated with both Lowe syndrome and Dent-2 disease, two rare X-linked conditions. Lowe syndrome is an oculo-cerebro-renal disorder, whereas Dent-2 patients mainly present renal ...proximal tubulopathy. Loss of OCRL-1, a phosphoinositide-5-phosphatase, leads in Lowe patients' fibroblasts to phosphatidylinositol-4,5-bisphosphate (PI(4,5)P2) accumulation, with defects in F-actin network, α-actinin distribution and ciliogenesis, whereas fibroblasts of Dent-2 patients are still uncharacterized. To search for mechanisms linked to clinical variability observed between these two OCRL mutation-associated pathologies, we compared dermal fibroblasts from independent patients, four affected by Dent-2 disease and six with Lowe syndrome. For the first time, we describe that Dent-2 fibroblasts with OCRL loss-of-function (LOF) mutations exhibit decrease in actin stress fibers, appearance of punctate α-actinin signals and alteration in primary cilia formation. Interestingly, we quantified these phenotypes as clearly intermediate between Lowe and control fibroblasts, thus suggesting that levels of these defects correlate with clinical variations observed between patients with OCRL mutations. In addition, we show that Lowe and Dent-2 fibroblasts display similar PI(4,5)P2 accumulation levels. Finally, we analyzed INPP5B, a paralogous gene already reported to exhibit functional redundancy with OCRL, and report neither differences in its expression at RNA or protein levels, nor specific allelic variations between fibroblasts of patients. Altogether, we describe here differential phenotypes between fibroblasts from Lowe and Dent-2 patients, both associated with OCRL LOF mutations, we exclude direct roles of PI(4,5)P2 and INPP5B in this phenotypic variability and we underline potential key alterations leading to ocular and neurological clinical features in Lowe syndrome.
The oculocerebrorenal syndrome of Lowe (OCRL) is a rare X-linked disorder characterized by severe mental retardation, congenital cataracts and renal Fanconi syndrome. OCRL1 protein is a ...phosphatidylinositol 4,5-bisphosphate 5-phosphatase with a C-terminal RhoGAP domain. Considering the pleiotropic cellular functions of Rho GTPases (Rho, Rac and Cdc42) and their dysregulation in several forms of mental retardation, we have investigated the so far unexplored function of the RhoGAP domain of OCRL1. Activated Rac GTPase was found to stably associate with the OCRL1 RhoGAP domain in vitro and to co-immunoprecipitate with endogenous OCRL1. Contrasting with other GAPs, OCRL1 RhoGAP exhibited a significant interaction with GDP bound Rac in vitro. As compared to Rac, other Rho GTPases tested showed reduced (Cdc42) or no binding (RhoA, RhoG) to OCRL1 RhoGAP. Immunofluorescence studies in HEK and COS7 cells and Golgi perturbation assays with Brefeldin A demonstrated that a fraction of endogenous Rac co-localizes with OCRL1 and γ-adaptin in the trans-Golgi network. The OCRL1 RhoGAP domain showed low Rac GAP activity in vitro, and when expressed in Swiss 3T3 cells induced specific inhibition of RacGTP dependent ruffles, consistent with OCRL1 being an active RacGAP. OCRL1 appears to be a bifunctional protein which, in addition to its PIP2 5-phosphatase activity, binds to Rac GTPase. This novel property may play a role in localizing OCRL1 to the trans-Golgi network. Moreover, loss of OCRL1 RhoGAP and the resulting alteration in Rho pathways may contribute to mental retardation in Lowe syndrome, as illustrated in other forms of X-linked mental retardation.
Intrauterine growth retardation and postnatal acute diabetes result from insulin deficiency in double homozygous null mutants for Ins1 and Ins2 (Duvillié B, et al., Proc. Natl. Acad. Sci. USA ...94:5137-5140, 1997). The characterization of single homozygous null mutants for Ins1 or Ins2 is described here. Neither kind of mutant mice was diabetic. Immunocytochemical analysis of the islets showed normal distribution of the endocrine cells producing insulin, glucagon, somatostatin, or pancreatic polypeptide. Analysis of the expression of the functional insulin gene in Ins1-/- or Ins2-/- mice revealed a dramatic increase of Ins1 transcripts in Ins2-/- mutants. This compensatory response was quantitatively reflected by total pancreatic insulin content similar for both types of mutants and wild-type mice. Moreover, both mutants had normal plasma insulin levels and normal glucose tolerance tests. The determination of beta-cell mass by morphometry indicated beta-cell hyperplasia in the mutant mice. The beta-cell mass in Ins2-/- mice was increased almost threefold, which accounts for the increase of Ins1 transcripts in Ins2-/-mutants. This study thus contributes to evaluate the potential of increasing the beta-cell mass to compensate for low insulin production.
Transgenic mouse lines carrying the human insulin gene were previously shown to express it in pancreas but not in other tissues. The present study reports evidence that the expression of the ...transgene is restricted to a single category of cells. Immunofluorescence staining of frozen pancreas sections showed that the human C-peptide was present in pancreatic islets only, and more precisely in the B cells of the islets. Human insulin transcripts were initiated correctly in mouse pancreas at the same site as in human pancreas. Three different transgenic lines with different insertion sites and various copy numbers of the human insulin transgene had the same high levels of the transgene transcripts corresponding to a well-balanced contribution in insulin gene expression.
The expression of a number of genes encoding key players in insulin signalling and action, including insulin, insulin receptor (IR), downstream signalling molecules such as insulin receptor ...substrate-1 (IRS-1) and IRS-2, glucose transporters (GLUT4, GLUT2) and important metabolic enzymes such as glucokinase, has now been altered in transgenic or knockout mice. Such mice presented with phenotypes ranging from mild defects, revealing complementarity between key molecules or pathways, to severe diabetes with ketoacidosis and early postnatal death. Insulin action could also be improved by overproduction of proteins acting at regulatory steps. The development of diabetes by combining mutations, which alone do not lead to major metabolic alterations, validated the 'diabetogenes' concept of non-insulin-dependent diabetes mellitus. Genes encoding insulin-like growth factors (IGF-I and IGF-II) and their type I receptor (IGF-IR) have also been disrupted. It appears that although IR and IGF-IR are both capable of metabolic and mitogenic signalling, they are not fully redundant. However, IR could replace IGF-IR if efficiently activated by IGF-II. Studies with cell lines lacking IR or IGF-IR lend support to such conclusions. Concerning the issues of specificity and redundancy, studies with cell lines derived from IRS-1-deficient mice showed that IRS-1 and IRS-2 are also not completely interchangeable.
Expression of the human insulin gene was examined in transgenic mouse lines carrying the gene with various lengths of DNA sequences 5' to the transcription start site (+1). Expression of the ...transgene was demonstrated by 1) the presence of human C-peptide in urine, 2) the presence of specific transcripts in pancreas, but not in other tissues, 3) the specific immunofluorescence staining of pancreatic islets for human C-peptide, and 4) the synthesis and accumulation of human (pro)insulin in isolated islets. Deletions in the injected DNA fragment of sequences upstream from positions -353, -258, and -168 allowed correct initiation of the transcripts and cell specificity of expression, while quantitative expression gradually decreased. Deletion to -58 completely abolished the expression of the gene. The amount of human product that in mice harboring the longest fragment contributes up to 50% of the total insulin does not alter the normal proportion of mice insulins I and II. These results suggest that expression of the human insulin gene in vivo results from the cooperation of several cis-regulatory elements present in the various deleted fragments. With none of the deletions used, expression of the transgene was observed in cell types other than beta-islet cells.
A mini-human insulin gene and four derivatives mutated at several regions potentially involved in the regulation of gene expression were used to generate transgenic mouse lines. The effect of these ...mutations on the efficiency of gene expression and cell specificity was studied using three approaches: (1) Northern blot analysis using total RNA from pancreas and other organs, (2) radioimmunoassay to detect the human C-peptide in urine samples, and (3) immunocytochemistry of pancreas sections to examine whether expression of the transgene was still specifically expressed in
β-cells. Mutation of the
cis-acting elements located between −238 and −206 (GCII and CTII motifs) resulted in a strong decrease of gene expression in the pancreas of transgenic mice, but it did not lead to complete extinction of the transgene expression. This region alone (−255/−202), when linked to the minimal Herpes simplex virus thymidine kinase gene
(tk) promoter, failed to activate chloramphenicol acetyltransferase (CAT) gene expression in transfected insulinoma cells, while it was activated by the equivalent region of the rat insulin I gene. On the contrary, mutation of the DNA motifs located between −109 and −75 (GCI and CTI) or between −323 and −297 (CTIII) did not significantly affect the level of the human insulin gene expression in transgenic mice. Replacement of the insulin promoter (−58/+1) by the
tk promoter did not alter its level of expression in transgenic mice. In all instances, expression of the different transgenes remained localized in the islet
β-cells. Altogether, these results indicate that the GCII-CTII motif is an important regulatory element for efficient expression of the human insulin gene in vivo, although it alone does not allow gene expression as it would require the association of other elements.
Full text
Available for:
IJS, IMTLJ, KILJ, KISLJ, NUK, SBCE, SBJE, UL, UM, UPUK
PDF
