Abstract: Thus, we have determined that SKIP is enriched at the mitochondria, where it can simultaneously anchor up to two type I PKA holoenzymes to phosphorylate ChChd3, another component of this protein assembly ( Fig. P1 C ). In conclusion, our study not only provides quantitative data on the stoichiometry of AKAP multi-PKA complexes inside cells, but also reveals a unique biological context for the anchoring protein SKIP as a mediator of PKA phosphorylation at the inner mitochondrial membrane ( Fig. P1 C ). Furthermore, as SKIP is present in the heart and mitochondria, and as the heart is an extremely energy-dependent tissue that relies heavily on the mitochondria for energy production, it will be interesting to investigate whether SKIP signaling complexes contribute to regulation of mitochondrial metabolic processes in the heart. In the final phase of this study, we focused on SKIP-binding partners of mitochondrial origin that may also be targets of PKA phosphorylation. Of particular interest was ChChd3, a component of the mitochondrial intermembrane space that is essential for maintaining integrity of the cristae, the folds in the mitochondrial inner membrane ( 5 ). Several biochemical and cell-based experiments indicate that SKIP and ChChd3 interact directly and that ChChd3 is efficiently phosphorylated by SKIP-anchored PKA. The single molecule pull-down (SIMPull) photobleaching assay is an exquisitely sensitive method for isolating and examining single molecules with minimal disruption of the protein complexes in which they may participate. Using this assay ( 4 ), we present quantitative data on the stoichiometry (e.g., the ratios) of the components of the PKA-AKAP complex within cells. This technique allowed us to derive two fundamental parameters of PKA anchoring that until now have been unattainable by conventional techniques. First, as shown by a fluorescent marker fused to RI, nearly 5% of the available RI associates with SKIP. Second, SKIP has the capacity to anchor 0, 1, or 2 molecules of RI dimer in situ and may exist in dynamic equilibrium between distinct occupancy states ( Fig. P1 A ). The distribution of SKIP and RI within the cell was evaluated by immunofluorescence confocal microscopy. For visualizing SKIP, fibroblast (COS-7) cells were genetically altered to express the SKIP protein fused to GFP. This fusion is useful because GFP emits a fluorescent signal that can allow the location of the SKIP protein to be monitored. In COS-7 cells, GFP-SKIP exhibited significant overlap with endogenous RI. The distribution of both proteins was similar and showed a pattern reminiscent of the mitochondria, the subcellular compartments associated with the majority of energy production in the cell. Antibodies against particular amino acid residues of the anchoring protein were generated to detect endogenous SKIP. The patterns of SKIP expression suggested that the protein was in two subcellular locations: a striated staining pattern reminiscent of Z bands, one of the crossbands discernible in cardiac muscle fibers, and a distribution that aligned with mitochondrial marker proteins ( Fig. P1 B , green). Live-cell imaging of SKIP-GFP and MitoTracker Orange, a fluorescent dye that stains mitochondria, in pulsatile adult heart muscle cells (cardiomyocytes), offered additional evidence that some of the anchoring protein is localized at the mitochondria. An additional experiment (mass spectrometry screen) provided more evidence for the mitochondrial localization of SKIP as type I PKA and members of a mitochondrial macromolecular complex consisting of the proteins ChChd3, ChChd6, SAM50, MTX1, and MTX2 were all identified as associating with SKIP. Finally, as mitochondria consist of an outer and inner membrane, subcellular fractionation of mouse hearts was performed to further localize SKIP. We observe that SKIP partitions to the intermembrane space and the matrix (region of the mitochondria enclosed by the inner membrane), thus localizing SKIP in a position to interact with the range of mitochondrial proteins identified from the mass spectrometry screen. Fig. P1. SKIP targets the type I PKA holoenzyme to mitochondria. ( A ) A model that depicts the RI anchoring sites on SKIP and the alternate stoichiometry of the SKIP∶PKA complexes. ( B ) The mitochondrial location of SKIP (green), RI (blue), and ChChd3 (red) in adult cardiomyocytes. ( C ) Cartoon depicting SKIP location and function in the intermembrane space of mitochondria. Within the cell, cAMP acts as a second messenger that relays signals initiated by other factors (e.g., hormones) to activate an enzyme known as protein kinase A (PKA). Upon activation, PKA adds a phosphate group to various proteins, thereby initiating a wide range of signaling events in mammalian cells. Prior to activation by cAMP, PKA exists as a holoenzyme consisting of two regulatory (R) subunits and two catalytic (C) subunits, with the R subunits preventing the C subunits from phosphorylating their targets. When cAMP levels are elevated, two cAMP molecules bind to each R subunit, thereby releasing the active C subunits. Where and when PKA becomes active has a profound influence on which cAMP responsive cellular events are propagated ( 1 ). The spatial and temporal organization of PKA requires a family of noncatalytic modulator proteins called A-kinase anchoring proteins (AKAPs). To date, 47 human AKAP genes have been identified as genes encoding proteins that direct PKA to defined locations within the cell. AKAPs interact with PKA via a helical region that fits into a docking and dimerization structure that is formed by the R subunits of each PKA holoenzyme ( 2 – 4 ). The majority of AKAPs sequester type II (RII) PKA subtypes, although several dual-function AKAPs can bind either type I (RI) or RII PKA. Here, we report that sphingosine kinase interacting protein (SKIP) is an AKAP that exclusively anchors the type I PKA holoenzyme, which affects how certain cAMP-mediated processes might be selectively regulated. Perhaps the most definitive proof of this finding comes from data showing that SKIP cannot anchor PKA in a mouse cell line that genetically lacks RI subunits. This genetic evidence augments additional studies showing that SKIP exclusively cofractionates (i.e., associates) with type I PKA subunits and that the two anchoring sequences on SKIP each have an extremely high affinity for RI. We also define certain amino acid residues in each SKIP helix as unique determinants of RI binding.

Abstract: Polypeptide growth factors activate common signal transduction pathways, yet they can induce transcription of different target genes. The mechanisms that control this specificity are not completely understood. Recently, we have described a fibroblast growth factor (FGF)-inducible response element, FiRE, on the syndecan-1 gene. In NIH 3T3 cells, the FiRE is activated by FGF-2 but not by several other growth factors, such as platelet-derived growth factor or epidermal growth factor, suggesting that FGF-2 activates signaling pathways that diverge from pathways activated by other growth factors. In this paper, we report that the activation of FiRE by FGF-2 requires protein kinase A (PKA) in NIH 3T3 cells. The PKA-specific inhibitor H-89 ( N -[2-( p -bromocinnamylamino)ethyl]-5-isoquinolinesulfonamide) blocked the FGF-2-induced activation of FiRE, the transcription of the syndecan-1 gene, and cell proliferation. Also, expression of a dominant-negative form of PKA inhibited the FGF-2-induced FiRE activation and the transcription of the syndecan-1 gene. The binding of activator protein-1 transcription-factor complexes, required for the activation of FiRE, was blocked by inhibition of PKA activity before FGF-2 treatment. In accordance with the growth factor specificity of FiRE, the activity of PKA was stimulated by FGF-2 but not by platelet-derived growth factor or epidermal growth factor. Furthermore, a portion of the PKA catalytic subunit pool was translocated to the nucleus by FGF-2. Noticeably, the total cellular cAMP concentration was not affected by FGF-2 stimulus. We propose that the FGF-2-selective transcriptional activation through FiRE is caused by the ability of FGF-2 to control PKA activity.