No abstract

Brief periods of sleep loss have long-lasting consequences such as impaired memory consolidation. Structural changes in synaptic connectivity have been proposed as a substrate of memory storage. Here, we examine the impact of brief periods of sleep deprivation on dendritic structure. In mice, we find that five hours of sleep deprivation decreases dendritic spine numbers selectively in hippocampal area CA1 and increased activity of the filamentous actin severing protein cofilin. Recovery sleep normalizes these structural alterations. Suppression of cofilin function prevents spine loss, deficits in hippocampal synaptic plasticity, and impairments in long-term memory caused by sleep deprivation. The elevated cofilin activity is caused by cAMP-degrading phosphodiesterase-4A5 (PDE4A5), which hampers cAMP-PKA-LIMK signaling. Attenuating PDE4A5 function prevents changes in cAMP-PKA-LIMK-cofilin signaling and cognitive deficits associated with sleep deprivation. Our work demonstrates the necessity of an intact cAMP-PDE4-PKA-LIMK-cofilin activation-signaling pathway for sleep deprivation-induced memory disruption and reduction in hippocampal spine density.DOI: http://dx.doi.org/10.7554/eLife.13424.001

Full-length cDNA for the rat brain rolipram-sensitive cyclic AMP phosphodiesterase (PDE), RD1 was introduced into the expression vector pSVL. COS cells transfected with the recombinant vector pSVL-RD1 exhibited a 30-55% increase in homogenate PDE activity, which was abolished by rolipram (10 microM). Removal of the first 67 nucleotides of the RD1 cDNA yielded a truncated enzyme called Met26-RD1 which lacked the N-terminal first 25 amino acids. Whereas approx. 75% of RD1 activity was membrane-associated, Met26-RD1 activity was found exclusively in the cytosol fraction. Expression of RD1 nearly doubled membrane-associated PDE activity, while expression of Met26-RD1 increased cytosolic activity by approx. 30%. Membrane RD1 activity was found to be primarily associated with the plasma membrane, was not released by either high concentrations of NaCl or by a 'hypotonic shock' treatment, but was solubilized with low concentrations of Triton X-100. Phase separation of membrane components with Triton X-114 showed partition of RD1 into both the aqueous and detergent-rich phases, whereas Met26-RD1 partitioned exclusively into the aqueous phase. Both RD1 and Met26-RD1 specifically hydrolysed cyclic AMP; were unaffected by either Ca2+/calmodulin or by low cyclic GMP concentrations; exhibited linear Lineweaver-Burke plots with similar Km values for cyclic AMP (4 microM); both were potently and similarly inhibited by rolipram (Ki approx. 0.5 microM) and were similarly inhibited by cilostamide and 3-isobutyl-1-methylxanthine. Thermal inactivation, at 50 degrees C, showed that while the cytosolic-located fraction of RD1 (t0.5 approx. 3 min) and Met26-RD1 (t0.5 approx 3 min) were similarly thermolabile, membrane-bound RD1 was considerably more thermostable (t0.5 approx. 11 min). Treatment of both cytosolic RD1 and Met26-RD1 with Triton X-100 did not affect their thermostability, but solubilization of membrane RD1 activity with Triton X-100 markedly decreased its thermostability (t0.5 approx. 5 min). The N-terminal domain of RD1 appears not to influence either the substrate specificity or inhibitor sensitivity of this enzyme, but it does contain information which can allow RD1 to become plasma membrane-associated and thereby adopt a conformation which has enhanced thermostability.

Ser-13 and Ser-54 were shown to provide the sole sites for the protein kinase A (PKA)-mediated phosphorylation of the human cAMP-specific phosphodiesterase isoform HSPDE4D3. The ability of PKA to phosphorylate and activate HSPDE4D3 was mimicked by replacing Ser-54 with either of the negatively charged amino acids, aspartate or glutamate, within the consensus motif of RRES54. The PDE4 selective inhibitor rolipram ¿4-[3-(cyclopentoxy)-4-methoxyphenyl]-2-pyrrolidone¿ inhibited both PKA-phosphorylated HSPDE4D3 and the Ser-54-->Asp mutant, with an IC50 value that was approximately 8-fold lower than that seen for the non-PKA-phosphorylated enzyme. Lower IC50 values for inhibition by rolipram were seen for a wide range of non-activated residue 54 mutants, except for those which had side-chains able to serve as hydrogen-bond donors, namely the Ser-54-->Thr, Ser-54-->Tyr and Ser-54-->Cys mutants. The Glu-53-->Ala mutant exhibited an activity comparable with that of the PKA phosphorylated native enzyme and the Ser-54-->Asp mutant but, in contrast to the native enzyme, was insensitive to activation by PKA, despite being more rapidly phosphorylated by this protein kinase. The activated Glu-53-->Ala mutant exhibited a sensitivity to inhibition by rolipram which was unchanged from that of the native enzyme. The double mutant, Arg-51-->Ala/Arg-52-->Ala, showed no change in either enzyme activity or rolipram inhibition from the native enzyme and was incapable of providing a substrate for PKA phosphorylation at Ser-54. No difference in inhibition by dipyridamole was seen for the native enzyme and the Ser-54-->Asp and Ser-54-->Ala mutants. A model is proposed which envisages that phosphorylation by PKA triggers at least two distinct conformational changes in HSPDE4D3; one of these gives rise to enzyme activation and another enhances sensitivity to inhibition by rolipram. Activation of HSPDE4D3 by PKA-mediated phosphorylation is suggested to involve disruption of an ion-pair interaction involving the negatively charged Glu-53. The increase in susceptibility to inhibition by rolipram upon PKA-mediated phosphorylation is suggested to involve the disruption of a hydrogen-bond involving the side-chain hydroxy group of Ser-54.

COS-7 cells were transfected with a plasmid encoding a putative splice variant of PDE4A cyclic AMP-specific phosphodiesterase, RPDE-6 (RNPDE4A5). This led to the expression of a novel, cyclic AMP-specific, rolipram-inhibited phosphodiesterase activity. In such transfected cells a novel approximately 109 kDa species was recognized by anti-peptide sera raised against a dodecapeptide whose sequence is found at the extreme C-terminus of both RPDE-6 and another PDE4A splice variant. RD1 (RNPDE4A1A). RPDE-6 activity and immunoreactivity was found distributed between both pellet (approximately 25%) and cytosol (approximately 75%) fractions of transfected COS-7 cells. Soluble and pellet RPDE-6 activities exhibited similar low Km values for cyclic AMP (approximately 2.4 microM) and were both inhibited by low concentrations of rolipram, with IC50 values for the soluble activity being lower (approximately 0.16 microM) than for the pellet activity (approximately 1.2 microM). Pellet RPDE-6 was resistant to release by either high NaCl concentrations or the detergent Triton X-100. Probing brain homogenates with the anti-(C-terminal peptide) sera identified two immunoreactive species, namely an approximately 79 kDa species reflecting RD1 and an approximately 109 kDa species that co-migrated with the immunoreactive species seen in COS cells transfected to express RPDE-6. The approximately 109 kDa species was found distributed between both the low-speed (P1) and high-speed (P2) pellet fractions as well as the cytosol fractions derived from both brain and RPDE-6-transfected COS cells. In contrast, RD1 was found exclusively in the P2 fraction. Phosphodiesterase (PDE) activity immuno-precipitated by these antisera from brain cytosol had the characteristics of COS cell-expressed RPDE-6 with KmcyclicAMP approximately 3.7 microM and IC50rolipram approximately 0.12 microM. The distribution of PDE activity immunoprecipitated from the cytosol of various brain regions paralleled that seen for the distribution of the approximately 109 kDa immunoreactive species. It is suggested that the 109 kDa species identified in brain cytosol and pellet fractions is the native form of RPDE-6. The PDE4A splice variants, RD1 and RPDE-6, were shown to have distinct patterns of expression among various brain regions. PDE4A and PDE4B activities appear to provide the major source of PDE4 activity in brain membranes, whereas the cytosolic PDE4 activity is suggested to reflect predominantly the activity of the PDE4D family. Alternative splicing of the PDE4A gene confers distinct N-terminal domains on RPDE-6 and RD1, which attenuates the Vmax. of these enzymes and defines their distinct subcellular distribution pattern.