Search

The strength of Ca2+ signaling is a hallmark of T cell activation, yet the role of Ca2+ homeostasis in developing T cells before expressing a mature T cell receptor is poorly understood. We aimed to unveil specific functions of the two plasma membrane Ca2+ ATPases expressed in T cells, PMCA1 and PMCA4. On a transcriptional and protein level we found that PMCA4 was expressed at low levels in CD4−CD8− double negative (DN) thymocytes and was even downregulated in subsequent stages while PMCA1 was present throughout development and upregulated in CD4+CD8+ double positive (DP) thymocytes. Mice with a targeted deletion of Pmca1 in DN3 thymocytes had an almost complete block of DP thymocyte development with an accumulation of DN4 thymocytes but severely reduced numbers of CD8+ immature single positive (ISP) thymocytes. The DN4 thymocytes of these mice showed strongly elevated basal cytosolic Ca2+ levels and a pre-mature CD5 expression, but in contrast to the DP thymocytes they were only mildly prone to apoptosis. Surprisingly, mice with a germline deletion of Pmca4 did not show any signs of altered progression through the developmental thymocyte stages, nor altered Ca2+ homeostasis throughout this process. PMCA1 is, therefore, non-redundant in keeping cellular Ca2+ levels low in the early thymocyte development required for the DN to DP transition.

Abstract The outcome of T cell activation is determined by mechanisms that balance Ca2+ influx and clearance. Here we report that murine CD4 T cells lacking Neuroplastin (Nptn −/−), an immunoglobulin superfamily protein, display elevated cytosolic Ca2+ and impaired post-stimulation Ca2+ clearance, along with increased nuclear levels of NFAT transcription factor and enhanced T cell receptor-induced cytokine production. On the molecular level, we identified plasma membrane Ca2+ ATPases (PMCAs) as the main interaction partners of Neuroplastin. PMCA levels were reduced by over 70% in Nptn −/− T cells, suggesting an explanation for altered Ca2+ handling. Supporting this, Ca2+ extrusion was impaired while Ca2+ levels in internal stores were increased. T cells heterozygous for PMCA1 mimicked the phenotype of Nptn −/− T cells. Consistent with sustained Ca2+ levels, differentiation of Nptn −/− T helper cells was biased towards the Th1 versus Th2 subset. Our study thus establishes Neuroplastin-PMCA modules as important regulators of T cell activation.

Polyamines are important regulators of basal cellular functions but also subserve highly specific tasks in the mammalian brain. With this respect, polyamines and the synthesizing and degrading enzymes are clearly differentially distributed in neurons versus glial cells and also in different brain areas. The synthesis of the diamine putrescine may be driven via two different pathways. In the "classical" pathway urea and carbon dioxide are removed from arginine by arginase and ornithine decarboxylase. The alternative pathway, first removing carbon dioxide by arginine decarboxlyase and then urea by agmatinase, may serve the same purpose. Furthermore, the intermediate product of the alternative pathway, agmatine, is an endogenous ligand for imidazoline receptors and may serve as a neurotransmitter. In order to evaluate and compare the expression patterns of the two gate keeper enzymes arginase and arginine decarboxylase, we generated polyclonal, monospecific antibodies against arginase-1 and arginine decarboxylase. Using these tools, we immunocytochemically screened the rat brain and compared the expression patterns of both enzymes in several brain areas on the regional, cellular and subcellular level. In contrast to other enzymes of the polyamine pathway, arginine decarboxylase and arginase are both constitutively and widely expressed in rat brain neurons. In cerebral cortex and hippocampus, principal neurons and putative interneurons were clearly labeled for both enzymes. Labeling, however, was strikingly different in these neurons with respect to the subcellular localization of the enzymes. While with antibodies against arginine decarboxylase the immunosignal was distributed throughout the cytoplasm, arginase-like immunoreactivity was preferentially localized to Golgi stacks. Given the apparent congruence of arginase and arginine decarboxylase distribution with respect to certain cell populations, it seems likely that the synthesis of agmatine rather than putrescine may be the main purpose of the alternative pathway of polyamine synthesis, while the classical pathway supplies putrescine and spermidine/spermine in these neurons.

Src family kinases (SFK) control multiple processes during brain development and function. We show here that the phosphoprotein associated with glycosphigolipid-enriched microdomains (PAG)/Csk binding protein (Cbp) modulates SFK activity in the brain. The timing and localization of PAG expression overlap with Fyn and Src, both of which we find associated to PAG. We demonstrate in newborn (P1) mice that PAG negatively regulates Src family kinases (SFK). P1 Pag1(-/-) mouse brains show decreased recruitment of Csk into lipid rafts, reduced phosphorylation of the inhibitory tyrosines within SFKs, and an increase in SFK activity of >/ = 50%. While in brain of P1 mice, PAG and Csk are highly and ubiquitously expressed, little Csk is found in adult brain suggesting altered modes of SFK regulation. In adult brain Pag1-deficiency has no effect upon Csk-distribution or inhibitory tyrosine phosphorylation, but kinase activity is now reduced (-20-30%), pointing to the development of a compensatory mechanism that may involve PSD93. The distribution of the Csk-homologous kinase CHK is not altered. Importantly, since the activities of Fyn and Src are decreased in adult Pag1(-/-) mice, thus presenting the reversed phenotype of P1, this provides the first in vivo evidence for a Csk-independent positive regulatory function for PAG in the brain.

Neuroplastin (np) 55 and 65 are immunoglobulin superfamily members that arise by alternative splicing of the same gene and have been implicated in long-term activity-dependent synaptic plasticity. Both biochemical and immunocytochemical data suggest that np55 is the predominant isoform (>95% of total neuroplastin) in cerebellum. Neuroplastin immunoreactivity is concentrated in the molecular layer and synaptic glomeruli in the granule cell layer. Expression in the molecular layer appears to be postsynaptic. First, neuroplastin is associated with Purkinje cell dendrites in two mouse granuloprival cerebellar mutants, disabled and cerebellar deficient folia. Second, in an acid sphingomyelinase knockout mouse with widespread protein trafficking defects, neuroplastin accumulates in the Purkinje cell somata. Finally, primary cerebellar cultures show neuroplastin expression in Purkinje cell dendrites and somata lacking normal histotypic organization and synaptic connections, and high-magnification views indicate a preferential association with dendritic spines. In the molecular layer, differences in neuroplastin expression levels present as a parasagittal array of stripes that alternates with that revealed by the expression of another compartmentation antigen, zebrin II/aldolase c. Neuroplastin immunoreactivity is first detected weakly at postnatal day 3 (P3) in the anterior lobe vermis. By P5, parasagittal stripes are already apparent in the immature molecular layer. At this stage, punctate deposits are also localised at the perimeter of the Purkinje cell perikarya; these are no longer detected by P15. The data suggest a role for neuroplastins in the development and maintenance of normal synaptic connections in the cerebellum. J. Comp. Neurol. 462:286–301, 2003. © 2003 Wiley-Liss, Inc. [ABSTRACT FROM AUTHOR]