Evolutionary changes in chlorophyllide a oxygenase (CAO) structure contribute to the acquisition of a new light-harvesting complex in micromonas

Abstract

Chlorophyll b is found in photosynthetic prokaryotes and primary and secondary endosymbionts, although their light-harvesting systems are quite different. Chlorophyll b is synthesized from chlorophyll a by chlorophyllide a oxygenase (CAO), which is a Rieske-mononuclear iron oxygenase. Comparison of the amino acid sequences of CAO among photosynthetic organisms elucidated changes in the domain structures of CAO during evolution. However, the evolutionary relationship between the light-harvesting system and the domain structure of CAO remains unclear. To elucidate this relationship, we investigated the CAO structure and the pigment composition of chlorophyll-protein complexes in the prasinophyte Micromonas. The Micromonas CAO is composed of two genes, MpCAO1 and MpCAO2, that possess Rieske and mononuclear iron-binding motifs, respectively. Only when both genes were introduced into the chlorophyll b-less Arabidopsis mutant (ch1-1) was chlorophyll b accumulated, indicating that cooperation between the two subunits is required to synthesize chlorophyll b. Although Micromonas has a characteristic light-harvesting system in which chlorophyll b is incorporated into the core antennas of reaction centers, chlorophyll b was also incorporated into the core antennas of reaction centers of the Arabidopsis transformants that contained the two Micromonas CAO proteins. Based on these results, we discuss the evolutionary relationship between the structures of CAO and light-harvesting systems.

Keywords: CAO; Chlorophyll; Chloroplast; Enzymes; Evolution; Photosynthesis; Photosynthetic Pigments; Photosystem.

FIGURE 1.

Comparison of the domain structure of the CAO protein among photosynthetic organisms. A, comparison of amino acid sequences of the Rieske center in CAO proteins. B, comparison of amino acid sequences of the mononuclear iron-binding motif in CAO proteins. In both panels, identical residues are shown in white type on a black background. C, domain structure of CAO proteins. Rieske center (Fe-S) and mononuclear iron (Fe)-binding motif are in the C domain. AtCAO, A. thaliana CAO; CrCAO, C. reinhardtii CAO; MpCAO1, M. pusilla CAO (Fe-S); MpCAO2, M. pusilla CAO (Fe); PhCAO, P. hollandica CAO. D, schematic drawing of core/peripheral antenna complexes of Arabidopsis, Chlamydomonas, Micromonas, and Prochlorothrix.

FIGURE 2.

Multiple amino acid sequence alignment of CAO proteins. Identical residues are shown in white type on a black background. The AtCAO sequences corresponding to the A domain, B domain, and C domain are shown. Asterisks and closed squares show binding sites of Rieske center and mononuclear iron-binding motif, respectively.

FIGURE 3.

Sequences of synthetic genes of MpCAO1 and MpCAO2.

FIGURE 4.

Generation of MpCAO overexpression of Arabidopsis lines. A, phenotypes of wild-type and transgenic lines. 1, wild type; 2, MpCAO1+MpCAO2–1; 3, MpCAO1+MpCAO2–2; 4, ch1-1; 5, MpCAO1; and 6, MpCAO2. B, confirmation of transgenic lines by PCR. C, confirmation of transgenic lines by immunoblot analysis. The MpCAO1-FLAG and MpCAO2-HA proteins were detected using anti-FLAG or anti-HA antibodies, respectively. Lane 1, wild type; lane 2, ch1-1; lane 3, MpCAO1; lane 4, MpCAO2; lane 5, MpCAO1+MpCAO2–1; lane 6, MpCAO1+MpCAO2–2.

FIGURE 5.

Co-immunoprecipitation of MpCAO1-FLAG and MpCAO2-HA proteins. The solubilized thylakoid membrane proteins from wild type or MpCAO1+MpCAO2 plants were co-immunoprecipitated with the anti-FLAG antibody (FLAG) or anti-HA antibody (HA) and subjected to SDS-PAGE. Detection of MpCAO1-FLAG and MpCAO2-HA proteins was performed by immunoblot analysis with anti-FLAG antibody (A) or anti-HA antibody (B), respectively. Lane 1, wild type; lane 2, MpCAO1+MpCAO2.

FIGURE 6.

Analysis of the oligomeric states of the CAO proteins. A, two-dimensional BN/SDS-PAGE followed by immunoblotting indicated that MpCAO1 and MpCAO2 proteins accumulated in monomeric, heterodimeric (86 kDa), and higher molecular weight forms in MpCAO1+MpCAO2 plant. B, two-dimensional BN/SDS-PAGE followed by immunoblotting indicated that BCFLAG proteins accumulated in monomeric (43 kDa), trimeric (126 kDa), and higher molecular weight forms in the BCFLAG overexpression plants. The BCFLAG proteins were detected using an anti-FLAG antibody or anti-CAO antibody. C, two-dimensional BN/SDS-PAGE followed by immunoblotting indicated that recombinant AtCAO protein formed higher molecular weight protein complexes in E. coli. The AtCAO proteins were detected using an anti-His antibody. D, two-dimensional BN/SDS-PAGE followed by immunoblotting indicated that PhCAO formed monomeric (41 kDa), trimeric (123 kDa, and higher molecular weight complexes. The PhCAO proteins were detected using an anti-CAO antibody.

FIGURE 7.

Multiple sequence alignment of AtCAO, dicamba monooxygenases (ddmC), and 2-oxoquinoline 8-monooxygenase (oxoO) proteins. Identical residues are shown in white type on a black background. Red asterisks and red circles show binding sites of Rieske center and mononuclear iron-binding motif, respectively.

FIGURE 8.

Separation of chlorophyll-protein complexes. Thylakoid membranes of M. pusilla (A), A. thaliana (WT) (B, C, and E), and MpCAO1+MpCAO2 plants (B, D, and E) were prepared. Chlorophyll-protein complexes of those plants were separated by native green gel electrophoresis (A and B). The identity of the chlorophyll-protein complexes corresponding to the bands are presented on the left side of the gel. CP1*, CP1 (PsaA/B)-LHCI complexes; CPa, core complexes of PSII (CP43/CP47); LHCII¹ and LHCII³, monomeric and trimeric LHCII, respectively; FC, free Chl. In addition, chlorophyll-protein complexes of WT and MpCAO1+MpCAO2 plants were also analyzed by two-dimensional SDS/SDS-PAGE, followed by silver staining (C and D). The amounts of PsaA/B, CP47, and Lhcb2 proteins were estimated by SDS-PAGE, followed by immunoblot analysis with their specific antibodies (E).

FIGURE 9.

Evolution of CAO structure. A, prochlorophyte CAOs have only the C domain (catalytic domain) and form a trimer. CAOs in green algae and land plants have A, B, and C domains and form a trimer. Mamiellales Micromonas CAOs are composed of two proteins with either a Rieske (Fe-S)- or mononuclear iron (Fe)-binding motifs. Neither protein has an A or B domain. They interact with each other to form a dimer. B, in each case, electrons must be transferred to a mononuclear iron (Fe) from a Rieske center (Fe-S) across the protein-protein interface.