Structural and mechanistic aspects of carotenoid cleavage dioxygenases (CCDs)

Abstract

Carotenoid cleavage dioxygenases (CCDs) comprise a superfamily of mononuclear non-heme iron proteins that catalyze the oxygenolytic fission of alkene bonds in carotenoids to generate apocarotenoid products. Some of these enzymes exhibit additional activities such as carbon skeleton rearrangement and trans-cis isomerization. The group also includes a subfamily of enzymes that split the interphenyl alkene bond in molecules such as resveratrol and lignostilbene. CCDs are involved in numerous biological processes ranging from production of light-sensing chromophores to degradation of lignin derivatives in pulping waste sludge. These enzymes exhibit unique features that distinguish them from other families of non-heme iron enzymes. The distinctive properties and biological importance of CCDs have stimulated interest in their modes of catalysis. Recent structural, spectroscopic, and computational studies have helped clarify mechanistic aspects of CCD catalysis. Here, we review these findings emphasizing common and unique properties of CCDs that enable their variable substrate specificity and regioselectivity. This article is part of a Special Issue entitled Carotenoids recent advances in cell and molecular biology edited by Johannes von Lintig and Loredana Quadro.

Keywords: Beta-propeller; Dioxetane; Monotopic membrane protein; Nitric oxide; Non-heme iron; Resveratrol.

Figure 1.. Phylogeny and cleavage specificity of CCDs.

A) Unrooted phylogenic tree of CCD protein sequences discussed in the main text. The tree was computed using MrBayes [102] based on a sequence alignment generated in Clustal Omega [103]. All bipartitions had posterior probabilities >90%. The scale represents average number of substitutions. Nos – Nostoc, Syn – Synechocystis, Pb – Pseudomonas brassicacearum. B) Cleavage site selectivity for CCDs and substrates discussed in the main text.

Figure 2.. Sequence alignment and topology of ACO, VP14, and CAO1.

A) Sequence alignment and corresponding secondary structures determined by X-ray crystallography. The conserved iron-binding His and Glu residues are marked with red and orange triangles, respectively. The black triangle indicates residues in close proximity to the iron that appear to occlude in many CCDs, although not in VP14 due to the smaller Ala side chain. B) Topology diagrams for ACO [19] (left), VP14 [53] (middle), and CAO1 [43] (right). The blades are color-matched to the secondary structure elements shown in panel A.

Figure 3.. Crystal structure of ACO.

Cartoon representation showing the seven bladed beta-propeller structure capped by a cluster of helical segments that form a dome housing the active site [19]. The iron cofactor (brown sphere) is directly coordinated by four His residues, three of which are stabilized by hydrogen bonding with conserved Glu residues. All structural images were generated using PyMOL. This figure was adapted with permission from [33]. Copyright John Wiley & Sons, Ltd.

Figure 4.. Active site of ACO.

A) Structure of the ACO active site [19]. The iron cofactor is shown as an orange sphere. An iron-bound solvent atom is shown as a red sphere. B) Tunnels leading to the iron center from the protein exterior. Residues thought to mediate membrane binding are shown as brown sticks. Substrate is presumably extracted from the membrane through the tunnel surrounded by the membrane-binding residues. Tunnels were generated using the MOLEonline web interface [104].

Figure 5.. Proposed catalytic mechanism of ACO.

Only the scissile double bond and directly connected carbons are shown for simplicity. The mechanism is based on both computational and experimental results as described in the main text.

Figure 6.. Active site of VP14.

A) Structure of the VP14 active site [53]. The iron cofactor is shown as an orange sphere. Iron-bound solvent and dioxygen are shown as red and gold spheres, respectively. B) Tunnels leading to the iron center from the protein exterior. Residues thought to mediate membrane binding are shown as brown sticks. Substrate is presumably extracted from the membrane through the tunnel surrounded by the membrane-binding residues.

Figure 7.. Active site of CAO1.

A) Structure of the CAO1 active site [43]. The iron cofactor is shown as an orange sphere. An iron-bound solvent atom is shown as a red sphere. B) A single tunnel leads to the iron center from the protein exterior. CAO1 and other stilbene-cleaving CCDs lack a hydrophobic patch that could mediate membrane binding.

Figure 8.. Structure of CAO1 in complex with piceatannol.

The structure was obtained by co-crystallizing piceatannol with Co-substituted CAO1, which is catalytically inactive [43]. The green mesh represents omit Fo-Fc electron density. Note the interaction of the 4-hydroxy group with Lys164 and Tyr133, both of which are important for catalytic function. This figure was adapted with permission from [33]. Copyright John Wiley & Sons, Ltd.

Figure 9.. Proposed catalytic mechanism of stilbene-cleaving CCDs.

The mechanism is based on both computational and experimental results as described in the main text.

Figure 10.. Sequence alignment between bovine RPE65 and mouse BCO2.

Secondary structure for RPE65 determined by crystallography is shown above the alignment. Residues thought to mediate membrane binding of these proteins are marked by horizontal orange brackets. The region mediating dimer formation in RPE65 is marked by a horizontal blue bracket. The conserved metal-binding His and Glu residues are marked by red and orange triangles, respectively, while the occluding Val residue is marked by a black triangle.

Figure 11.. Homology model of mouse BCO2.

The model was generated by the SwissModel server [105] using the crystallographic coordinates of RPE65 [91]. Residues that likely mediate membrane affinity in this protein are marked by pale orange spheres. The loop that may mediate dimer formation is colored blue. Key active site residues are shown as sticks and the iron cofactor is shown as a dark orange sphere.