Search

Cbln1 is the prototype of a family (Cbln1-Cbln4) of secreted glycoproteins and is essential for normal synapse structure and function in cerebellum by bridging presynaptic Nrxn to postsynaptic Grid2. Here we report the effects of glycosylation on the in vitro receptor binding properties of Cblns. Cbln1, 2 and 4 harbor two N-linked glycosylation sites, one at the N-terminus is in a region implicated in Nrxn binding and the second is in the C1q domain, a region involved in Grid2 binding. Mutation (asparagine to glutamine) of the N-terminal site, increased neurexin binding whereas mutation of the C1q site markedly increased Grid2 binding. These mutations did not influence subunit composition of Cbln trimeric complexes (mediated through the C1q domain) nor their assembly into hexamers (mediated by the N-terminal region). Therefore, glycosylation likely masks the receptor binding interfaces of Cblns. As Cbln4 has undetectable Grid2 binding in vitro we assessed whether transgenic expression of wild type Cbln4 or its glycosylation mutants rescued the Cbln1-null phenotype in vivo. Cbln4 partially rescued and both glycosylation mutants completely rescued ataxia in cbln1-null mice. Thus Cbln4 has intrinsic Grid2 binding that is attenuated by glycosylation, and glycosylation mutants exhibit gain of function in vivo.

PEP-19 is a 7.6 kDa neuronally expressed polypeptide that contains a single calmodulin-binding IQ motif. The calmodulin-binding activity of several neuronal IQ motif proteins is regulated by phosphorylation of a conserved serine. We propose that the serine residue within the IQ motif of PEP-19 is phosphorylated, and that phosphorylation modifies the activity of PEP-19. Camstatin, a functionally active 25-residue fragment of PEP-19's IQ motif, binds calmodulin and inhibits neuronal nitric oxide synthase. A truncated camstatin-in which the IQ motif serine is the only phosphorylatable residue-was screened against 42 different kinases. Truncated camstatin is selectively phosphorylated by four isoforms of protein kinase C. Furthermore, treatment of full-length PEP-19 with PKCgamma catalyzes phosphorylation of the same serine residue. Fluorescent anisotropy shows that phosphorylation of camstatin inhibits its binding to calmodulin. NMR solution structures indicate that both camstatin and phospho-camstatin exist in similar dynamic turn-like conformations. This suggests that camstatin's greater affinity for calmodulin is due not to a change in the conformation of the phospho-peptide, but rather, to a disruption of hydrophobic interactions between phospho-camstatin and calmodulin caused by the presence of the hydrophilic phosphate group. The H(alpha) chemical shifts and the circular dichroism spectra of the camstatins are consistent with those of "nascent helices". We submit that PEP-19 is a PKC substrate, and that the phosphorylation state of PEP-19 may play a role in the modulation of calmodulin-dependent signaling.

Apaf-1 is a mammalian homolog of CED-4 that regulates cell death by participating in a ternary complex with cytochrome c, and procaspase-9. In the case of CED-4, two splice variants exist. The smaller (CED-4S) is proapoptotic while the larger (CED-4L) contains a short in-frame insert and is anti-apoptotic. We cloned a murine variant of apaf-1, termed apaf-1L, which contains an eleven amino acid insert similar to a recently described human apaf-1L clone. apaf-1 and apaf-1L have similar distributions in adult and fetal tissues, although apaf-1L transcripts are more abundant. Apaf-1L, undergoes homomerization and heteromerization with Apaf-1 in yeast. Apaf-1L also binds to caspase-9 and a dominant-negative isoform of caspase-9. Unlike CED-4, neither Apaf-1 variant was lethal in yeast. However, both Apaf-1 and Apaf-1L elicit cell death when cotransfected with caspase-9 into 293 EBNA cells. Although Apaf-1L was more potent than Apaf-1, their biological properties were qualitatively similar.

The spontaneous autosomal recessive mouse mutation, Purkinje cell degeneration (pcd), was first identified through its ataxic behavior. Since its discovery in the 1970s, the strain has undergone extensive investigation, although another quarter century elapsed until the mutant gene (agtpbp1 a.k.a. Nna1) underlying the pcd phenotype was identified. As Nna1 was initially discovered as a gene induced in motor neurons following axotomy the finding that its loss leads to selective neuronal degeneration points to a novel and unexpected common molecular mechanism contributing to the apparently opposing processes of degeneration and regeneration. The elucidation of this mechanism may of course have significant implications for an array of neurological disorders. Here we will first review the principle features of the pcd phenotype and then discuss the functional implications of more recent findings emanating from the characterization of Nna1, the protein that is lost in pcd. We also provide new data on the genetic dissection of the cell death pathways operative in pcd(3J) mice, proving that granule cell death and Purkinje cell death in these mice have distinct molecular bases. We also provide new information on the structure of mouse Nna1 as well as Nna1 protein levels in pcd(3J) mice.

Apaf-1 is a mammalian homolog of CED-4 that regulates cell death by participating in a ternary complex with cytochrome c, and procaspase-9. In the case of CED-4, two splice variants exist. The smaller (CED-4S) is proapoptotic while the larger (CED-4L) contains a short in-frame insert and is anti-apoptotic. We cloned a murine variant of apaf-1, termed apaf-1L, which contains an eleven amino acid insert similar to a recently described human apaf-1L clone. apaf-1 and apaf-1L have similar distributions in adult and fetal tissues, although apaf-1L transcripts are more abundant. Apaf-1L, undergoes homomerization and heteromerization with Apaf-1 in yeast. Apaf-1L also binds to caspase-9 and a dominant-negative isoform of caspase-9. Unlike CED-4, neither Apaf-1 variant was lethal in yeast. However, both Apaf-1 and Apaf-1L elicit cell death when cotransfected with caspase-9 into 293 EBNA cells. Although Apaf-1L was more potent than Apaf-1, their biological properties were qualitatively similar.

Peptide 19-immunoreactivity (PEP 19-ir) was examined in the dorsal root ganglion (DRG) and the mesencephalic trigeminal tract nucleus (Mes5) of the adult rat. Thirty-eight percent of DRG cells were immunoreactive (ir) for PEP 19. These neurons were small to large and measured 167-4583 micron2 (mean+/-S.D.=2048+/-913 micron2). Seventy-five percent of DRG cells2000 micron2 and 15% of those1000 micron2 exhibited PEP 19-ir. Thirty-six percent of DRG neurons in the range 1000-2000 micron2 showed the ir. In the Mes5, 87% of primary neurons were ir for this peptide. Muscle spindles in the soleus and masseter muscles contained PEP 19-ir spiral axon terminals. Double immunofluorescence methods revealed the co-expression of PEP 19 and calcium binding proteins. Eighty-six percent of parvalbumin-ir neurons exhibited PEP 19-ir. Conversely, 60% of PEP 19-ir neurons showed parvalbumin-ir. The cell size analysis revealed that 55% of PEP 19-ir neurons600 micron2 showed parvalbumin-ir and that all PEP 19-ir neurons600 micron2 were devoid of it. Ninety percent of PEP 19-ir DRG neurons showed S100-ir, whereas 60% of S100-ir ones co-expressed PEP 19-ir. In the Mes5, virtually all PEP 19-ir primary neurons exhibited parvalbumin-ir. The co-expression of PEP 19 and S100 could not be observed in the nucleus. The present study indicates that PEP 19-ir neurons which co-express parvalbumin-ir are proprioceptors in the spinal and the trigeminal systems. PEP 19-ir small DRG neurons without S100-ir are probably exteroceptors and may include unmyelinated nociceptors.

Hepatic leukemia factor (HLF) is a bZIP transcription factor related to the CES-2 protein, which controls apoptosis of the NSM serotoninergic neurons in Caenorhabditis elegans. Ectopic expression of HLF as an E2A-HLF fusion protein in t(17;19)-positive human pro-B cell acute lymphoblastic leukemias is believed to promote malignancy by interfering with apoptosis. While HLF has been linked to malignancies of the lymphoid system, it is not normally expressed in these cells. Rather, HLF transcripts are detected in the liver, kidney, lung and adult nervous system by Northern blotting. Despite the links to cell death, little is known of the distribution or function of HLF in the adult and developing mammalian nervous system. Therefore, we cloned mouse Hlf and studied its expression by in situ hybridization. During embryonic brain development, Hlf expression was restricted to the anterior pituitary and meninges. By early postnatal life, Hlf was highly expressed in somatosensory cortex, thalamic nuclei, and structures arising from ectodermal placodes. Subsequently, Hlf expression increased in the central nervous system and was found throughout the brain by adulthood. In the developing pituitary gland, Hlf was highly expressed in the rostral tip of the anterior lobe. This pattern is similar to that of Tef, an Hlf-related bZIP protein. However, while Tef is expressed in the anterior pituitary of the adult mouse, Hlf was detected in both the anterior and posterior pituitary. Hlf expression was not associated with cells undergoing programmed cell death in the nervous system. Hlf expression increased markedly with synaptogenesis and was coincident with barrel formation revealed by cytochrome oxidase staining. Together, these data suggest that Hlf plays a role in the function of differentiated neurons in the adult nervous system rather than programmed cell death.