1H NMR structural and functional characterisation of a cAMP-specific phosphodiesterase-4D5 (PDE4D5) N-terminal region peptide that disrupts PDE4D5 interaction with the signalling scaffold proteins, βarrestin and RACK1

K. John Smith, George S. Baillie, Eva I. Hyde, Xiang Li, Thomas M. Houslay, Angela McCahill, Allan J. Dunlop, Graeme B. Bolger, Enno Klussmann, David R. Adams, Miles D. Houslay

Research output: Contribution to journalArticle

45 Citations (Scopus)

Abstract

The unique 88 amino acid N-terminal region of cAMP-specific phosphodiesterase-4D5 (PDE4D5) contains overlapping binding sites conferring interaction with the signaling scaffold proteins, ßarrestin and RACK1. A 38-mer peptide, whose sequence reflected residues 12 through 49 of PDE4D5, encompasses the entire N-terminal RACK1 Interaction Domain (RAID1) together with a portion of the ßarrestin binding site. 1H NMR and CD analyses indicate that this region has propensity to form a helical structure. The leucine-rich hydrophobic grouping essential for RACK1 interaction forms a discrete hydrophobic ridge located along a single face of an amphipathic a-helix with Arg34 and Asn36, which also play important roles in RACK1 binding. The Asn22/Pro23/Trp24/Asn26 grouping, essential for RACK1 interaction, was located at the N-terminal head of the amphipathic helix that contained the hydrophobic ridge. RAID1 is thus provided by a distinct amphipathic helical structure. We suggest that the binding of PDE4D5 to the WD-repeat protein, RACK1, may occur in a manner akin to the helix-helix interaction shown for G? binding to the WD-repeat protein, Gß. A more extensive section of the PDE4D5 N-terminal sequence (Thr11-Ala85) is involved in ßarrestin binding. Several residues within the RAID1 helix contribute to this interaction however. We show here that these residues form a focused band around the centre of the RAID1 helix, generating a hydrophobic patch (from Leu29, Val30 and Leu33) flanked by polar/charged residues (Asn26, Glu27, Asp28, Arg34). The interaction with ßarrestin exploits a greater circumference on the RAID1 helix, and involves two residues (Glu27, Asp28) that do not contribute to RACK1 binding. In contrast, the interaction of RACK1 with RAID1 is extended over a greater length of the helix and includes Leu37/Leu38, which do not contribute to ßarrestin binding. A membrane-permeable, stearoylated Val12-Ser49 38-mer peptide disrupted the interaction of both ßarrestin and RACK1 with endogenous PDE4D5 in HEKB2 cells, whilst a cognate peptide with a Glu27Ala substitution selectively failed to disrupt PDE4D5/RACK1 interaction. The stearoylated Val12-Ser49 38-mer peptide enhanced the isoprenaline-stimulated PKA phosphorylation of the ß2-adrenergic receptors (ß2AR) and its activation of ERK, whilst the Glu27Ala peptide was ineffective in both these regards. © 2007 Elsevier Inc. All rights reserved.

Original languageEnglish
Pages (from-to)2612-2624
Number of pages13
JournalCellular Signalling
Volume19
Issue number12
DOIs
Publication statusPublished - Dec 2007

Keywords

  • β 2 -adrenergic receptors (β 2 AR)
  • βarrestin
  • Cyclic AMP
  • ERK
  • NMR structure
  • Peptide displacement
  • PKA
  • Protein-protein interaction
  • RACK1
  • Rolipram
  • Signalling scaffold
  • Spot immobilised peptide arrays

Fingerprint Dive into the research topics of '<sup>1</sup>H NMR structural and functional characterisation of a cAMP-specific phosphodiesterase-4D5 (PDE4D5) N-terminal region peptide that disrupts PDE4D5 interaction with the signalling scaffold proteins, βarrestin and RACK1'. Together they form a unique fingerprint.

  • Cite this