Kinetics and regulation of two catalytic subunits of cAMP-dependent protein kinase from Aplysia californica

Catalytic subunit of protein kinase A caged at the activating phosphothreonine

Caged reagents are photoactivatable molecules with applications in biological research. While a great deal of work has been carried out on small caged molecules, less has been done on caged macromolecules, such as proteins. Caged proteins would be especially useful in signal transduction research. Since most proteins involved in cell signaling are regulated by phosphorylation, a means to cage phosphorylated proteins would be generally applicable. Here we show that the catalytic subunit of protein kinase A can be activated by thiophosphorylation at Thr-197. The modified protein can then be caged with 4-hydroxyphenacyl bromide to yield a derivative with a specific catalytic activity that is reduced by approximately 17-fold. Upon photolysis at near UV wavelengths, an approximately 15-fold increase in activity is observed, representing an approximately 85-90% yield of uncaged product with a quantum yield phi(P) = 0.21. Because protein kinases belong to a superfamily with structurally related catalytic domains, the protein chemistry demonstrated here should be applicable to a wide range of signaling proteins.

Differential transcriptional regulation by the alpha- and gamma-catalytic subunit isoforms of cAMP-dependent protein kinase

The C gamma and C alpha isoforms of the cAMP-dependent protein kinase (PKA) share 83% identity including all critical catalytic and substrate-binding residues defined to date. Compared to C alpha, C gamma has a different substrate specificity and a selective pseudosubstrate specificity, exhibiting inhibition by regulatory subunits, but not by the protein kinase inhibitor. In these studies, C gamma-mediated gene transcription regulation was compared with that of C alpha in four cell lines using transient transfection/dual luciferase assays. As compared to C gamma, C alpha more efficiently activated a cAMP-response element (CRE)-regulated fragment of the human alpha-glycoprotein hormone promoter which was coupled to a firefly luciferase reporter gene (pGH alpha-fluc). This occurred in Cos7, Y1, and Kin8 adrenal cells by 23-, 6.5-, and 1.4-fold, respectively. In contrast, C gamma, but not C alpha, activated the Sp1RE-regulated herpes simplex virus thymidine kinase promoter which was coupled to a Renilla luciferase reporter (pTK-rluc). In Sp1-deficient Sf9 cells, pGH alpha-fluc expression was maintained for both isoforms, but cotransfection with an Sp1 expression plasmid was necessary and sufficient for activation of pTK-rluc expression by C gamma. In all cell lines, cotransfection with a PDK1 expression plasmid enhanced the transcriptional activation of both C alpha and C gamma (1.5- to 3-fold), while a catalytically inactive PDK1 mutant (PDK.KD) did not. These results suggest that both C alpha and C gamma can activate CRE-responsive genes; however, C alpha does so with better efficiency than C gamma. In contrast to C alpha, C gamma activates transcription of genes containing pTK-like Sp1RE sites. Activation of different C subunit isoforms can provide a means to diversify cAMP-mediated transcription, possibly affecting cell phenotype.

Two novel brain-specific splice variants of the murine Cbeta gene of cAMP-dependent protein kinase

We have previously characterized two murine cAMP-dependent protein kinase catalytic subunit genes, Calpha and Cbeta1. Targeted disruption of the Cbeta1 promoter revealed two splice variants of the Cbeta catalytic subunit gene (designated Cbeta2 and Cbeta3) that continue to be expressed. These variants arise from unique promoters and are brain-specific. Cbeta2 is expressed in several discrete areas in the limbic system. These include the lateral septum, the bed nucleus of the stria terminalis, the ventral medial hypothalamus, and the amygdala. In the neocortex, expression is highest in cortical areas such as the prefrontal and insular cortex that are associated limbic structures. By contrast, Cbeta1 is most highly expressed in the cortex and hippocampus and is also present in all non-neuronal tissues examined. Cbeta3 is expressed at very low levels with wide distribution throughout the brain. Both the Cbeta2 and Cbeta3 variants are enzymatically active and induce gene expression in transient transfections with a cAMP response element-reporter construct. This activity is inhibited by protein kinase A regulatory subunits, the protein kinase inhibitor, and the chemical inhibitor H-89. We also demonstrate that Cbeta1 is myristoylated at the amino terminus like the Calpha isoform, but neither Cbeta2 nor Cbeta3 is myristoylated. The discrete expression of Cbeta variants in the brain suggests specific functional roles in neuronal signaling.

In Aplysia sensory neurons, the neuropeptide SCPB and serotonin differ in efficacy both in modulating cellular properties and in activating adenylyl cyclase: implications for mechanisms underlying presynaptic facilitation

The facilitatory transmitters serotonin (5-HT) and the molluscan neuropeptides SCPA and SCPB both activate adenylyl cyclase in Aplysia mechanosensory neurons and produce multiple modulatory effects that contribute to increasing transmitter release from these cells. This enhancement of transmitter release from sensory neurons contributes to increased behavioral response during sensitization and classical conditioning in Aplysia. Recently, specific examples of modulation in these sensory neurons have been described that are more effectively initiated by 5-HT than by the SCPs. For example, in the present study, 5-HT produces 55% greater broadening of the normal sensory neuron action potential than did SCPB. These differences in the modulatory actions of the facilitatory transmitters have been interpreted as suggesting that 5-HT produces its modulatory effects at least partly via a cAMP-independent mechanism. However, we have found that the two types of facilitatory transmitters are not equally effective in activating adenylyl cyclase. In both whole CNS membranes and sensory neuron membranes, SCPB was less effective than 5-HT in stimulating adenylyl cyclase activity measured in steady state assays. Because electrophysiological experiments suggested that the response to the SCPs desensitizes rapidly, we further compared cyclase stimulation in perfused membrane assays that enable continuous monitoring of cyclase activity; however we observed that 5-HT was also more effective than SCPB in stimulating cyclase at the onset of transmitter exposure. We discuss the possibility that lower peak stimulation of cyclase by SCPB and a faster rate of desensitization could account for some of the differences between the SCPs and 5-HT in modulating sensory neurons.

Catalytic subunits of Aplysia neuronal cAMP-dependent protein kinase with two different N termini

Previously, two forms of cAMP-dependent protein kinase catalytic subunit generated by mutually exclusive use of two internal exon cassettes (A1 and A2) were demonstrated in Aplysia neurons. Here, it is shown that there also exist catalytic subunits with alternative N termini derived from two exons, N1 and N2, expressed in combination with either of the internal cassettes. Processed transcripts including N1 or N2 sequences are of about equal abundance in the nervous system, arise through alternative promoter use, and encode catalytically active polypeptides. The N2 amino acid sequence is 21 residues longer than the N1 sequence and is homologous to the nonmyristoylated N terminus of the TPK1 gene product, a yeast catalytic subunit homolog. These data support the view that cAMP-dependent protein kinase activity in Aplysia neurons is produced by a complex array of regulatory and catalytic subunits that generate multiple holoenzymes with a spectrum of properties.