FIGURE 1.

Bioinformatics approach to identify new AKAPs. A, multiple sequence alignment of RIIBDs of several AKAPs. Amino acid residues in conserved positions are shaded in gray. The isoelectric points (pI) of the peptides are indicated. All sequences except murine (Mm) AKAP-Lbc, rat (Rn) AKAP18δ, and D. melanogaster (Dm) DAKAP550 are derived from human AKAPs. AKAP_IS is a peptide designed in silico (10, 52). B, flow chart for the identification of RII-binding domains. Depicted is the AKAP search pattern for screening of the Swiss-Prot Database for potential RII-binding proteins. For the data base search, the pI value range was limited to 3.0–6.4. Duplicates and peptides not accessible to synthesis were removed (Filter). The remaining peptides (2572) were spot-synthesized and subjected to RII overlay assay (supplemental Fig. 1). 829 peptides bound RII subunits. C, 829 peptides were synthesized on three membranes and subjected to RII overlay (Blocks 1–3; for sequences see supplemental Table 1). Signals were detected by autoradiography. Row A on each block contains the indicated control peptides, which are established RII-binding peptides or inactive versions thereof. Dashed boxes, RIIBDs from previously identified AKAPs. Solid box, peptide derived from the protein GSKIP. D, of the 829 RII-binding peptides, 27 representing putatively new AKAPs (and additionally 9 derived from previously described AKAPs) were spot-synthesized and subjected to RII overlay in the absence (control) or presence of the PKA anchoring disruptor peptide AKAP18δ-L314E or the inactive control peptide AKAP18δ-PP (10 μm each). Row A of each membrane contains the indicated control peptides; row B contains peptides derived from previously identified AKAPs; and rows C–E contain the putative RII-binding domains of the indicated proteins (given as abbreviations; for sequences see Table 1). Box, RII-binding domain of GSKIP. Displayed are representative membranes from at least three independent experiments.