aniimalsl-4 and in brain tissue homogenates5-8 indicate that the first two carbon atoms of these compounds are derived from serine while the remainder of the molecule is derived from palmitic acid or palmitaldehyde. Brady, Formica, and Koval6 reported that pyridoxal phosphate (PLP), Mnn++, nicotinamide, cytidine 5'-diphosphate (CDP)-choline, and several other cof actors stimulated the synthesis of sphingosine from serine in cell-free particulate preparations of rat brain. In subsequent experiments by these and other investigators, PLP and Mnn++ were routinely added to the reaction mixtures.9 Even though several PLP-dependent mechanisms for the condensation of serine with a 16-carbon derivative have been suggested,4-6 a requirement for this coenzyme in this specific reaction has never been clearly demonstrated. The very low rate of sphingolipid base synthesis in brain extracts has directed attention to the yeast Hansenula ciferri as a potential source of an enzyme system capable of synthesizing sphingolipids at a higher rate.'0 This yeast is able to produce large quantities of acetylated phytosphingosine and, to a lesser extent, acetylated dihydrosphingosine.'-14 In experiments with intact cells and radioactive precursors, Green, Kaneshiro, and Law'0 showed that phytosphingosine and dihydrosphingosine are derived from serine and palmitic acid, presumably by a pathway resembling that found in brain tissue. These investigators, however, were unable to demonstrate the formation of these lipids in vitro. More recently, Haskell and Snell'5 showed that the vitamin B6-deficient yeast, Hanseniaspora valbyensis, contained lowered ainounts of phytosphingosine, thus providing nutritional evidence for a role of this vitamin in sphingolipid biosynthesis. The postulated role for pyridoxal phosphate in the synthesis of sphingolipids is unusual for PLP-enzymes in that the serine carboxyl is replaced by a carbon chain derived from palmitate rather than by a hydrogen atom. We have therefore investigated the enzymatic aspects of this reaction, and present herein experiments that demonstrate synthesis of dihydrosphingosine from palmityl-CoA and serine by a cell-free particulate fraction of H. ciferri, and that PLP is required in the condensation reaction.

2′,3′‐Cyclic nucleotide 3′‐phosphodiesterase (CNP) is one of the earliest myelin‐related proteins to be specifically expressed in differentiating oligodendrocytes (ODCs) in the central nervous system (CNS) and is implicated in myelin biogenesis. CNP possesses an in vitro enzymatic activity, whose in vivo relevance remains to be defined, because substrates with 2′,3,‐cyclic termini have not yet been identified. To characterize CNP function better, we previously determined the structure of the CNP catalytic domain by NMR. Interestingly, the structure is remarkably similar to the plant cyclic nucleotide phosphodiesterase (CPDase) from A. thaliana and the bacterial 2′‐5′ RNA ligase from T. thermophilus, which are known to play roles in RNA metabolism. Here we show that CNP is an RNA‐binding protein. Furthermore, by using precipitation analyses, we demonstrate that CNP associates with poly(A)+ mRNAs in vivo and suppresses translation in vitro in a dose‐dependent manner. With SELEX, we isolated RNA aptamers that can suppress the inhibitory effect of CNP on translation. We also demonstrate that CNP1 can bridge an association between tubulin and RNA. These results suggest that CNP1 may regulate expression of mRNAs in ODCs of the CNS. © 2008 Wiley‐Liss, Inc.

Yeast and plant tRNA splicing entails discrete healing and sealing steps catalyzed by a tRNA ligase that converts the 2',3' cyclic phosphate and 5'-OH termini of the broken tRNA exons to 3'-OH/2'-PO4 and 5'-PO4 ends, respectively, then joins the ends to yield a 2'-PO4, 3'-5' phosphodiester splice junction. The junction 2'-PO4 is removed by a tRNA phosphotransferase, Tpt1. Animal cells have two potential tRNA repair pathways: a yeast-like system plus a distinctive mechanism, also present in archaea, in which the 2',3' cyclic phosphate and 5'-OH termini are ligated directly. Here we report that a mammalian 2',3' cyclic nucleotide phosphodiesterase (CNP) can perform the essential 3' end-healing steps of tRNA splicing in yeast and thereby complement growth of strains bearing lethal or temperature-sensitive mutations in the tRNA ligase 3' end-healing domain. Although this is the first evidence of an RNA processing function in vivo for the mammalian CNP protein, it seems unlikely that the yeast-like pathway is responsible for animal tRNA splicing, insofar as neither CNP nor Tpt1 is essential in mice.

Regeneration-induced CNPase homolog (RICH) is an axonal growth-associated protein, which is induced in teleost fish upon optical nerve injury. RICH consists of a highly acidic N-terminal domain, a catalytic domain with 2',3'-cyclic nucleotide 3'-phosphodiesterase (CNPase) activity and a C-terminal isoprenylation site. In vitro RICH and mammalian brain CNPase specifically catalyze the hydrolysis of 2',3'-cyclic nucleotides to produce 2'-nucleotides, but the physiologically relevant in vivo substrate remains unknown. Here, we report the NMR structure of the catalytic domain of goldfish RICH and describe its binding to CNPase inhibitors. The structure consists of a twisted nine-stranded antiparallel beta-sheet surrounded by alpha-helices on both sides. Despite significant local differences mostly arising from a seven-residue insert in the RICH sequence, the active site region is highly similar to that of human CNPase. Likewise, refinement of the catalytic domain of rat CNPase using residual dipolar couplings gave improved agreement with the published crystal structure. NMR titrations of RICH with inhibitors point to a similar catalytic mechanism for RICH and CNPase. The results suggest a functional importance for the evolutionarily conserved phosphodiesterase activity and hint of a link with pre-tRNA splicing.

Myelination of axons by oligodendrocytes enables rapid impulse propagation in the central nervous system. But long-term interactions between axons and their myelin sheaths are poorly understood. Here we show that Cnp1 , which encodes 2′,3′-cyclic nucleotide phosphodiesterase in oligodendrocytes, is essential for axonal survival but not for myelin assembly. In the absence of glial cyclic nucleotide phosphodiesterase, mice developed axonal swellings and neurodegeneration throughout the brain, leading to hydrocephalus and premature death. But, in contrast to previously studied myelin mutants, the ultrastructure, periodicity and physical stability of myelin were not altered in these mice. Genetically, the chief function of glia in supporting axonal integrity can thus be completely uncoupled from its function in maintaining compact myelin. Oligodendrocyte dysfunction, such as that in multiple sclerosis lesions, may suffice to cause secondary axonal loss.

This study reports the analysis of K+ channel activity in bovine periaxolemmal-myelin and white matter-derived clathrin-coated vesicles. Channel activity was evaluated by the fusion of membrane vesicles with phospholipid bilayers formed across a patch-clamp pipette. In periaxolemmal myelin spontaneous K+ channels were observed with amplitudes of 25–30, 45–55, and 80–100 pS, all of which exhibited mean open-times of 1–2 msec. The open state probability of the 50 pS channel in periaxolemmal-myelin was increased by 6-methyldihydro-pyran-2-one. Periaxolemmal-myelin K+ channel activity was regulated by Ca2+. Little or no change in activity was observed when Ca2+ was added to thecis side of the bilayer. Addition of 10 μM total Ca2+ also resulted in little change in K+ channel activity. However, at 80 μM total Ca2+ all K+ channel activity was suppressed along with the activation of a 100 pS Cl− channel. The K+ channel activity in periaxolemmal myelin was also regulated through a G-protein. Addition of GTPγS to thetrans side of the bilayer resulted in a restriction of activity to the 45–50 pS channel which was present at all holding potentials. Endocytic coated vesicles, form in part through G-protein mediated events; white matter coated vesicles were analyzed for G proteins and for K+ channel activity. These vesicles, which previous studies had shown are derived from periaxolemmal domains, were found to be enriched in the α subunits of G0, Gsα, and Giα and the low molecular weight G protein,ras. As with periaxolemmal-myelin treated with GTPγS, the vesicle membrane exhibited only the 50 pS channel. The channel was active at all holding potentials and had open times of 1–6 msec. Addition of GTPγS to the bilayer fused with vesicle membrane appeared to suppress this channel activity at low voltages yet induced a hyperactive state at holding potentials of 45 mV or greater. The vesicle 50 pS K+ channel was also activated by the 6-methyl-dihydropyron-2-one (20 μM).