Structural anatomy of Protein Kinase C C1 domain interactions with diacylglycerol and other agonists

Abstract

Diacylglycerol (DAG) is a versatile lipid whose 1,2-sn-stereoisomer serves both as second messenger in signal transduction pathways that control vital cellular processes, and as metabolic precursor for downstream signaling lipids such as phosphatidic acid. Effector proteins translocate to available DAG pools in the membranes by using conserved homology 1 (C1) domains as DAG-sensing modules. Yet, how C1 domains recognize and capture DAG in the complex environment of a biological membrane has remained unresolved for the 40 years since the discovery of Protein Kinase C (PKC) as the first member of the DAG effector cohort. Herein, we report the high-resolution crystal structures of a C1 domain (C1B from PKCδ) complexed to DAG and to each of four potent PKC agonists that produce different biological readouts and that command intense therapeutic interest. This structural information details the mechanisms of stereospecific recognition of DAG by the C1 domains, the functional properties of the lipid-binding site, and the identities of the key residues required for the recognition and capture of DAG and exogenous agonists. Moreover, the structures of the five C1 domain complexes provide the high-resolution guides for the design of agents that modulate the activities of DAG effector proteins.

Fig. 1. Arrangement of protein chains, lipid, and detergent molecules in the unit cell and the asymmetric unit of the C1Bδ-DAG complex crystal (PDB ID: 7L92).

a The unit cell contains 72 DAG-complexed C1Bδ chains and 18/54 DAG/DPC molecules that peripherally associate with the protein surface. Structural Zn²⁺ ions of C1Bδ are shown as black spheres. b The asymmetric unit comprises 8 C1Bδ protein chains with 8 DAG molecules captured within a well-defined groove, and 2/6 peripheral DAG/DPC molecules. c Space-filling representation of the two distinct DAG/DPC micelles.

Fig. 2. Stereospecificity of DAG binding by C1Bδ.

a Backbone superposition of 8 DAG-complexed C1Bδ chains of the AU (cyan, PDB ID: 7L92) onto the structure of apo C1Bδ (sienna, PDB ID: 7KND). The sidechain of Trp252 reorients towards the tips of membrane-binding β12 and β34 loops upon DAG binding. DAG adopts one of the two distinct binding modes: “sn-1” (b) or “sn-2” (c). The formation of the C1Bδ-DAG complex in bicelles is reported by the chemical shift perturbations (CSPs) of the amide ¹⁵NH (d) and methyl ¹³CH₃ (e) groups of C1Bδ. Asterisks denote residues whose resonances are broadened by chemical exchange in the apo-state. The insets show the response of individual residues to DAG binding through the expansions of the ¹⁵N–¹H and ¹³C–¹H HSQC spectral overlays of apo and DAG-complexed C1Bδ. f ¹H–¹H_N Thr242 and Gly253 strips from the 3D ¹⁵N-edited NOESY-TROSY spectrum of the C1Bδ-DAG-bicelle complex. The protein-to-DAG NOE pattern is consistent with the distances observed in the “sn-1” mode (Chain 5, light purple) but not the “sn-2” mode (Chain 7, green). All distances are in Å and color-coded in the “sn-1” complex to match the labels in the spectrum; “w” denotes water protons. The medium-range NOE that would be characteristic of the “sn-2” complex is shown in red.

Fig. 3. Roles of C1Bδ loops in lipid binding.

a Polar backbone atoms and hydrophobic sidechains of DAG-interacting C1Bδ residues create a binding site whose properties are tailored to capture the amphiphilic DAG molecule. This is illustrated through the deconstruction of the “sn-1” binding mode into three tiers that accommodate the glycerol backbone (tier 1), the sn-1/2 ester groups (tier 2), and the acyl chain methylenes (tier 3). b Residue-specific lipid-to-protein PRE values of the amide protons, ¹H Γ₂, indicate that loop β34 is inserted deeper into the membrane than β12. The PRE value for Trp252 is that for the indole NHε group. Cross-peaks broadened beyond detection in paramagnetic bicelles are assigned an arbitrary value of 120 s⁻¹. His270 and Lys271 cross-peaks are exchange-broadened and therefore unsuitable for quantitative analysis (open circles). The PRE values were derived from ¹H_N transverse relaxation rate constants collected on a single sample in the absence (diamagnetic) and presence (paramagnetic) of 14-doxyl PC. The error was estimated using the r.m.s.d. of the base plane noise. The inset shows the top view of the “sn-1” mode C1Bδ-DAG complex color-coded according to the hydrophobicity.

Fig. 4. Peripheral DHPC molecules in the C1Bδ-ligand complexes.

DHPC molecules peripherally associate with the membrane-binding loop regions of C1Bδ complexed to a PDBu; b prostratin; c ingenol-3-angelate, d AJH-836 (one molecule per AU), and e AJH-836 (two molecules per AU). In e, one protein chain is ligand-free (color-coded green) and has three DHPC molecules, labeled 1 through 3 that cap the membrane-binding loop region. f The versatility of potential Trp252-lipid interactions, exemplified by the Trp252 sidechain from the ligand-free C1Bδ monomer (e green). In addition to non-polar contacts with the hydrophobic lipid moieties, the Trp sidechains can engage in H-bonding, cation–π, and CH–π interactions.

Fig. 5. Structures of the C1Bδ-ligand complexes reveal the interaction modes of PKC agonists.

a Chemical structures and polar groups involved in hydrogen-bonding interactions with C1Bδ of PKC agonists. The numbering of oxygen atoms follows the ALATIS system. b 3D structures of the complexes (PDB IDs from left to right: 7KNJ, 7LCB, 7KO6, and 7LF3) showing the ligand placement in the binding groove. The shape of the ligands’ hydrophobic cap, viewed from the top of the loop region, is outlined in maroon. The hydrophobic ridge that traverses the groove is marked with the maroon dashed line. c Ligand interactions with Thr242, Leu251, and Gly253 (underlined) that recapitulate the DAG hydrogen-bonding pattern are shown with red dashed lines. Blue dashed lines show ligand-specific hydrogen bonds, including the intra-ligand ones in PDBu and Prostratin. The depression created in loop β34 by Gly253 in the PDBu and Prostratin complexes accommodates DHPC molecules in the crystal.

Fig. 6. Structural analysis of C1Bδ-agonist complexes identifies three key oxygen-containing groups and the roles of conserved hydrophobic residues.

a Loop region of the backbone-superimposed C1Bδ complexes (pairwise r.m.s.d. <0.6 Å relative to chain 5 of the “sn-1” DAG complex). Oxygen-containing functional groups involved in the interactions with C1Bδ are highlighted by squares. b–e Pairwise comparison of the binding poses of b PDBu; c prostratin; d ingenol-3-angelate; and e AJH-836 relative to that of DAG in the binding groove. Hydrophobic sidechains that envelope the ligands and form the rim of the membrane-binding region are also shown. f Amino acid sequence of C1Bδ and the consensus sequence of DAG-sensitive C1 domains. g–i A subset of conserved hydrophobic residues that form a “cage”-like arrangement around the ligands, with a potential to form CH–π interactions in addition to the apolar contacts. The rotameric flip of Trp252, illustrated using DAG (g) and PDBu (h) complexes, is essential for creating a contiguous hydrophobic surface. i The Trp252 sidechain remains in its apo-state rotameric conformation in the C1Bδ-P13A complex that was crystallized in the absence of lipids/detergents. j Top view of the C1Bδ-P13A loop region showing the contribution of the C12-OH group to the hydrophilic character of the P13A “cap”.