Presence of free myxol and 4-hydroxymyxol and absence of myxol glycosides in Anabaena variabilis ATCC 29413, and proposal of a biosynthetic pathway of carotenoids

We identified the molecular structures of all carotenoids in Anabaena variabilis ATCC 29413 (= IAM M-204). The major carotenoids were beta-carotene, echinenone and canthaxanthin. Myxol glycosides were absent, while free forms of myxol and 4-hydroxymyxol were present. The 4-hydroxyl group of the latter was a mixture of (4R) and (4S) configurations, which is a rare mixture in carotenoids. Thus, this strain was the first cyanobacterium found to have free myxol and not myxol glycosides, and seemed to lack the gene for or activity of glycosyl transferase. In another strain of A. variabilis IAM M-3 (= PCC 7118), we recently identified (3R,2'S)-myxol 2'-fucoside and (3S,2'S)-4-ketomyxol 2'-fucoside, and hence the strain ATCC 29413 might be useful for investigating the characteristics of myxol glycosides in cyanobacteria. Based on the identification of the carotenoids and the completion of the entire nucleotide sequence of the genome in A. variabilis ATCC 29413, we proposed a biosynthetic pathway of the carotenoids and the corresponding genes and enzymes. The homologous genes were searched by sequence homology only from the functionally confirmed genes.     

2,2'-beta-hydroxylase (CrtG) is involved in carotenogenesis of both nostoxanthin and 2-hydroxymyxol 2'-fucoside in Thermosynechococcus elongatus strain BP-1

We identified the molecular structures, including the stereochemistry, of all carotenoids in Thermosynechococcus elongatus strain BP-1. The major carotenoid was beta-carotene, and its hydroxyl derivatives of (3R)-beta-cryptoxanthin, (3R,3'R)-zeaxanthin, (2R,3R,3'R)-caloxanthin and (2R,3R,2'R,3'R)-nostoxanthin were also identified. The myxol glycosides were identified as (3R,2'S)-myxol 2'-fucoside and (2R,3R,2'S)-2-hydroxymyxol 2'-fucoside. 2-Hydroxymyxol 2'-fucoside is a novel carotenoid, and similar carotenoids of 4-hydroxymyxol glycosides were previously named aphanizophyll. Ketocarotenoids, such as echinenone and 4-ketomyxol, which are unique carotenoids in cyanobacteria, were absent, and genes coding for both beta-carotene ketolases, crtO and crtW, were absent in the genome. From a homology search, the Tlr1917 amino acid sequence was found to be 41% identical to 2,2'- beta-hydroxylase (CrtG) from Brevundimonas sp. SD212, which produces nostoxanthin from zeaxanthin. In the crtG disruptant mutant, 2-hydroxymyxol 2'-fucoside, caloxanthin and nostoxanthin were absent, and the levels of both myxol 2'-fucoside and zeaxanthin were higher. Therefore, the gene has a CrtG function for both myxol to 2-hydroxymyxol and zeaxanthin to nostoxanthin. This is the first functional identification of CrtG in cyanobacteria. We also investigated the distribution of crtG-like genes, and 2-hydroxymyxol and/or nostoxanthin, in cyanobacteria. Based on the identification of the carotenoids and the completion of the entire nucleotide sequence of the genome in T. elongatus, we propose a biosynthetic pathway of the carotenoids and the corresponding genes and enzymes.     

The cyanobacterium Gloeobacter violaceus PCC 7421 uses bacterial-type phytoene desaturase in carotenoid biosynthesis

Carotenoid composition and its biosynthetic pathway in the cyanobacterium Gloeobacter violaceus PCC 7421 were investigated. beta-Carotene and (2S,2'S)-oscillol 2,2'-di(alpha-L-fucoside), and echinenone were major and minor carotenoids, respectively. We identified two unique genes for carotenoid biosynthesis using in vivo functional complementation experiments. In Gloeobacter, a bacterial-type phytoene desaturase (CrtI), rather than plant-type desaturases (CrtP and CrtQ), produced lycopene. This is the first demonstration of an oxygenic photosynthetic organism utilizing bacterial-type phytoene desaturase. We also revealed that echinenone synthesis is catalyzed by CrtW rather than CrtO. These findings indicated that Gloeobacter retains ancestral properties of carotenoid biosynthesis.     

Substrate specificities and availability of fucosyltransferase and beta-carotene hydroxylase for myxol 2'-fucoside synthesis in Anabaena sp. strain PCC 7120 compared with Synechocystis sp. strain PCC 6803

To elucidate the biosynthetic pathways of carotenoids, especially myxol 2'-glycosides, in cyanobacteria, Anabaena sp. strain PCC 7120 (also known as Nostoc sp. strain PCC 7120) and Synechocystis sp. strain PCC 6803 deletion mutants lacking selected proposed carotenoid biosynthesis enzymes and GDP-fucose synthase (WcaG), which is required for myxol 2'-fucoside production, were analyzed. The carotenoids in these mutants were identified using high-performance liquid chromatography, field desorption mass spectrometry, and (1)H nuclear magnetic resonance. The wcaG (all4826) deletion mutant of Anabaena sp. strain PCC 7120 produced myxol 2'-rhamnoside and 4-ketomyxol 2'-rhamnoside as polar carotenoids instead of the myxol 2'-fucoside and 4-ketomyxol 2'-fucoside produced by the wild type. Deletion of the corresponding gene in Synechocystis sp. strain PCC 6803 (sll1213; 79% amino acid sequence identity with the Anabaena sp. strain PCC 7120 gene product) produced free myxol instead of the myxol 2'-dimethyl-fucoside produced by the wild type. Free myxol might correspond to the unknown component observed previously in the same mutant (H. E. Mohamed, A. M. L. van de Meene, R. W. Roberson, and W. F. J. Vermaas, J. Bacteriol. 187:6883-6892, 2005). These results indicate that in Anabaena sp. strain PCC 7120, but not in Synechocystis sp. strain PCC 6803, rhamnose can be substituted for fucose in myxol glycoside. The beta-carotene hydroxylase orthologue (CrtR, Alr4009) of Anabaena sp. strain PCC 7120 catalyzed the transformation of deoxymyxol and deoxymyxol 2'-fucoside to myxol and myxol 2'-fucoside, respectively, but not the beta-carotene-to-zeaxanthin reaction, whereas CrtR from Synechocystis sp. strain PCC 6803 catalyzed both reactions. Thus, the substrate specificities or substrate availabilities of both fucosyltransferase and CrtR were different in these species. The biosynthetic pathways of carotenoids in Anabaena sp. strain PCC 7120 are discussed.     

Myxol and 4-ketomyxol 2'-fucosides, not rhamnosides, from Anabaena sp. PCC 7120 and Nostoc punctiforme PCC 73102, and proposal for the biosynthetic pathway of carotenoids

We identified the molecular structures of carotenoids in some Anabaena and Nostoc species. The myxoxanthophyll and ketomyxoxanthophyll in Anabaena (also known as Nostoc) sp. PCC 7120, Anabaena variabilis IAM M-3, Nostoc punctiforme PCC 73102 and Nostoc sp. HK-01 were (3R,2'S)-myxol 2'-fucoside and (3S,2'S)-4-ketomyxol 2'-fucoside, respectively. The glycoside moiety of the pigments was fucose, not rhamnose. The major carotenoids were beta-carotene and echinenone, and the minor ones were beta-cryptoxanthin, zeaxanthin, canthaxanthin and 3'-hydroxyechinenone. Based on the identification of the carotenoids and the completion of the entire nucleotide sequence of the genome in Anabaena sp. PCC 7120 and N. punctiforme PCC 73102, we proposed a biosynthetic pathway for the carotenoids and the corresponding genes and enzymes. Since only zeta-carotene desaturase (CrtQ) from Anabaena sp. PCC 7120 and beta-carotene ketolase (CrtW) from N. punctiforme PCC 73102 have been functionally identified, the other genes were searched by sequence homology only from the functionally confirmed genes. Finally, we investigated the phylogenetic relationships among some Anabaena and Nostoc species, including some newly isolated species.     

Myxoxanthophyll in Synechocystis sp. PCC 6803 is myxol 2'-dimethyl-fucoside, (3R,2'S)-myxol 2'-(2,4-di-O-methyl-alpha-L-fucoside), not rhamnoside

We identified the molecular structures of the carotenoids in Synechocystis sp. PCC 6803. Myxoxanthophyll in this cyanobacterium was myxol 2'-dimethyl-fucoside, (3R,2'S)-myxol 2'-(2,4-di-O-methyl-alpha-L-fucoside). The sugar moiety of the pigment was not rhamnose but dimethylated fucose, which has not been reported in carotenoid glycosides. The other carotenoids were beta-carotene, (3R,3'R)-zeaxanthin, echinenone, (3'R)-3'-hydroxyechinenone and deoxymyxol 2'-dimethyl-fucoside, (2'S)-deoxymyxol 2'-(2,4-di-O-methyl-alpha-L-fucoside). Generally, the group of polar carotenoids in cyanobacteria is referred to as myxoxanthophyll, and the structure is considered to be myxol 2'-rhamnoside. Since the name myxoxanthophyll can not specify the sugar moiety and the identification of the sugar moiety is unfeasible in many cyanobacteria, we propose the following naming convention: when the sugar moiety is unknown, the name is myxol glycoside, when known, as in the case of rhamnose and alpha-L-fucose, they should be named myxol 2'-rhamnoside and myxol 2'-alpha-L-fucoside, respectively.     

Mutational and functional analysis of the beta-carotene ketolase involved in the production of canthaxanthin and astaxanthin

Biosynthesis of the commercial carotenoids canthaxanthin and astaxanthin requires beta-carotene ketolase. The functional importance of the conserved amino acid residues of this enzyme from Paracoccus sp. strain N81106 (formerly classified as Agrobacterium aurantiacum) was analyzed by alanine-scanning mutagenesis. Mutations in the three highly conserved histidine motifs involved in iron coordination abolished its ability to catalyze the formation of ketocarotenoids. This supports the hypothesis that the CrtW ketolase belongs to the family of iron-dependent integral membrane proteins. Most of the mutations generated at other highly conserved residues resulted in partial activity. All partially active mutants showed a higher amount of adonixanthin accumulation than did the wild type when expressed in Escherichia coli cells harboring the zeaxanthin biosynthetic gene cluster. Some of the partially active mutants also produced a significant amount of echinenone when expressed in cells producing beta-carotene. In fact, expression of a mutant carrying D117A resulted in the accumulation of echinenone as the predominant carotenoid. These observations indicate that partial inactivation of the CrtW ketolase can often lead to the production of monoketolated intermediates. In order to improve the conversion rate of astaxanthin catalyzed by the CrtW ketolase, a color screening system was developed. Three randomly generated mutants, carrying L175M, M99V, and M99I, were identified to have improved activity. These mutants are potentially useful in pathway engineering for the production of astaxanthin.     

Improvement of a CrtO-type of beta-carotene ketolase for canthaxanthin production in Methylomonas sp

Two types of non-homologous beta-carotene ketolases (CrtW and CrtO) were previously described. We report improvement of a CrtO-type of beta-carotene ketolase for canthaxanthin production in a methylotrophic bacterium, Methylomonas sp. 16a, which could use the C1 substrate (methane or methanol) as sole carbon and energy source. The crtO gene from Rhodococcus erythropolis was improved for canthaxanthin production in an E. coli strain engineered to produce high titer carotenoids by error-prone PCR mutagenesis followed by in vitro recombination. The best mutants from protein engineering could produce approximately 90% of total carotenoids as canthaxanthin in the high titer E. coli strain compared to approximately 20% canthaxanthin produced by the starting gene. Canthaxanthin production in Methylomonas was also significantly improved to approximately 50% of total carotenoids by the mutant genes. Further improvement of canthaxanthin production to approximately 93% in Methylomonas was achieved by increased expression of the best mutant gene. Some mutations were found in many of the improved genes, suggesting that these sites, and possibly the regions around these sites, were important for improving the crtO's activity for canthaxanthin production.     

Characterization of cyanobacterial carotenoid ketolase CrtW and hydroxylase CrtR by complementation analysis in Escherichia coli

The pathway from beta-carotene to astaxanthin is a crucial step in the synthesis of astaxanthin, a red antioxidative ketocarotenoid that confers beneficial effects on human health. Two enzymes, a beta-carotene ketolase (carotenoid 4,4'-oxygenase) and a beta-carotene hydroxylase (carotenoid 3,3'-hydroxylase), are involved in this pathway. Cyanobacteria are known to utilize the carotenoid ketolase CrtW and/or CrtO, and the carotenoid hydroxylase CrtR. Here, we compared the catalytic functions of CrtW ketolases, which originated from Gloeobacter violaceus PCC 7421, Anabaena (also known as Nostoc) sp. PCC 7120 and Nostoc punctiforme PCC 73102, and CrtR from Synechocystis sp. PCC 6803, Anabaena sp. PCC 7120 and Anabaena variabilis ATCC 29413 by complementation analysis using recombinant Escherichia coli cells that synthesized various carotenoid substrates. The results demonstrated that the CrtW proteins derived from Anabaena sp. PCC 7120 as well as N. punctiforme PCC 73102 (CrtW148) can convert not only beta-carotene but also zeaxanthin into their 4,4'-ketolated products, canthaxanthin and astaxanthin, respectively. In contrast, the Anabaena CrtR enzymes were very poor in accepting either beta-carotene or canthaxanthin as substrates. By comparison, the Synechocystis sp. PCC 6803 CrtR converted beta-carotene into zeaxanthin efficiently. We could assign the catalytic functions of the gene products involved in ketocarotenoid biosynthetic pathways in Synechocystis sp. PCC 6803, Anabaena sp. PCC 7120 and N. punctiforme PCC 73102, based on the present and previous findings. This explains why these cyanobacteria cannot produce astaxanthin and why only Synechocystis sp. PCC 6803 can produce zeaxanthin.     

Characterization of two β-carotene ketolases,CrtO and CrtW,by complementation analysis in <Emphasis Type="Italic">Escherichia coli</Emphasis>

The pathways from β-carotene to astaxanthin are crucial key steps for producing astaxanthin, one of industrially useful carotenoids, in heterologous hosts. Two β-carotene ketolases (β-carotene 4,4′-oxygenase), CrtO and CrtW, with different structure are known up to the present. In this paper, we compared the catalytic functions of a CrtO ketolase that was obtained from a marine bacterium Rhodococcus erythropolis strain PR4, CrtO derived from cyanobacterium Synechosistis sp. PCC6803, and CrtW derived from a marine bacterium Brevundimonas sp. SD212, by complementation analysis in Escherichia coli expressing the known crt genes. Results strongly suggested that a CrtO-type ketolase was unable to synthesize astaxanthin from zeaxanthin, i.e., only a CrtW-type ketolase could accept 3-hydroxy-β-ionone ring as the substrate. Their catalytic efficiency for synthesizing canthaxanthin from β-carotene was also examined. The results obtained up to the present clearly suggest that the bacterial crtW and crtZ genes are a combination of the most promising gene candidates for developing recombinant hosts that produce astaxanthin as the predominant carotenoid.     

Unique Carotenoids in the Terrestrial Cyanobacterium <Emphasis Type="Italic">Nostoc commune</Emphasis> NIES-24: 2-Hydroxymyxol 2′-Fucoside,Nostoxanthin and Canthaxanthin

Cyanobacteria produce some carotenoids. We identified the molecular structures, including the stereochemistry, of all the carotenoids in the terrestrial cyanobacterium, Nostoc commune NIES-24 (IAM M-13). The major carotenoid was β-carotene. Its hydroxyl derivatives were (3R)-β-cryptoxanthin, (3R,3′R)-zeaxanthin, (2R,3R,3′R)-caloxanthin, and (2R,3R,2′R,3′R)-nostoxanthin, and its keto derivatives were echinenone and canthaxanthin. The unique myxol glycosides were (3R,2′S)-myxol 2′-fucoside and (2R,3R,2′S)-2-hydroxymyxol 2′-fucoside. This is only the second species found to contain 2-hydroxymyxol. We propose possible carotenogenesis pathways based on our identification of the carotenoids: the hydroxyl pathway produced nostoxanthin via zeaxanthin from β-carotene, the keto pathway produced canthaxanthin from β-carotene, and the myxol pathway produced 2-hydroxymyxol 2′-fucoside via myxol 2′-fucoside. This cyanobacterium was found to contain many kinds of carotenoids and also displayed many carotenogenesis pathways, while other cyanobacteria lack some carotenoids and a part of carotenogenesis pathways compared with this cyanobacterium.     

Functional implications of pigments bound to a cyanobacterial cytochrome b6f complex

A highly purified cytochrome b(6)f complex from the cyanobacterium Synechocystis sp. PCC 6803 selectively binds one chlorophyll a and one carotenoid in analogy to the recent published structure from two other b(6)f complexes. The unknown function of these pigments was elucidated by spectroscopy and site-directed mutagenesis. Low-temperature redox difference spectroscopy showed red shifts in the chlorophyll and carotenoid spectra upon reduction of cytochrome b(6), which indicates coupling of these pigments with the heme groups and thereby with the electron transport. This is supported by the correlated kinetics of these redox reactions and also by the distinct orientation of the chlorophyll molecule with respect to the heme cofactors as shown by linear dichroism spectroscopy. The specific role of the carotenoid echinenone for the cytochrome b(6)f complex of Synechocystis 6803 was elucidated by a mutant lacking the last step of echinenone biosynthesis. The isolated mutant complex preferentially contained a carotenoid with 0, 1 or 2 hydroxyl groups (most likely 9-cis isomers of beta-carotene, a monohydroxy carotenoid and zeaxanthin, respectively) instead. This indicates a substantial role of the carotenoid - possibly for strucure and assembly - and a specificity of its binding site which is different from those in most other oxygenic photosynthetic organisms. In summary, both pigments are probably involved in the structure, but may also contribute to the dynamics of the cytochrome b(6)f complex.     

Ketocarotenoid formation in transgenic potato

Potato has been genetically engineered for the production of commercially important ketocarotenoids including astaxanthin (3,3'-dihydroxy 4,4'-diketo-beta-carotene). To support the formation of 3-hydroxylated and 4-ketolated beta-carotene, a transgenic potato line accumulating zeaxanthin due to inactivated zeaxanthin epoxidase was co-transformed with the crtO beta-carotene ketolase gene from the cyanobacterium Synechocystis under a constitutive promoter. Plants were generated which exhibited expression of this gene, resulting in an accumulation of echinenone, 3'-hydroxyechinenone, and 4-ketozeaxanthin in leaves, as well as 3'-hydroxyechinenone, 4-ketozeaxanthin together with astaxanthin in the tuber. The amount of ketocarotenoids formed represent approximately 10-12% of total carotenoids in leaves and tubers. Negative effects on photosynthesis due to the presence of the ketocarotenoids in leaves could be excluded by the determination of variable fluorescence.     

Subcellular Localization of Carotenoid Biosynthesis in Synechocystis sp. PCC 6803

The biosynthesis pathway of carotenoids in cyanobacteria is partly described. However, the subcellular localization of individual steps is so far unknown. Carotenoid analysis of different membrane subfractions in Synechocystis sp. PCC6803 shows that “light” plasma membranes have a high carotenoid/protein ratio, when compared to “heavier” plasma membranes or thylakoids. The localization of CrtQ and CrtO, two well-defined carotenoid synthesis pathway enzymes in Synechocystis, was studied by epitope tagging and western blots. Both enzymes are locally more abundant in plasma membranes than in thylakoids, implying that the plasma membrane has higher synthesis rates of β-carotene precursor molecules and echinenone.     

Non‐endogenous ketocarotenoid accumulation in engineered Synechocystis sp. PCC 6803

The cyanobacterium Synechocystis sp. PCC 6803 is a model species commonly employed for biotechnological applications. It is naturally able to accumulate zeaxanthin (Zea) and echinenone (Ech), but not astaxanthin (Asx), which is the highest value carotenoid produced by microalgae, with a wide range of applications in pharmaceutical, cosmetics, food and feed industries. With the aim of finding an alternative and sustainable biological source for the production of Asx and other valuable hydroxylated and ketolated intermediates, the carotenoid biosynthetic pathway of Synechocystis sp. PCC 6803 has been engineered by introducing the 4,4′ β‐carotene oxygenase (CrtW) and 3,3′ β‐carotene hydroxylase (CrtZ) genes from Brevundimonas sp. SD‐212 under the control of a temperature‐inducible promoter. The expression of exogenous CrtZ led to an increased accumulation of Zea at the expense of Ech, while the expression of exogenous CrtW promoted the production of non‐endogenous canthaxanthin and an increase in the Ech content with a concomitant strong reduction of β‐carotene (β‐car). When both Brevundimonas sp. SD‐212 genes were coexpressed, significant amounts of non‐endogenous Asx were obtained accompanied by a strong decrease in β‐car content. Asx accumulation was higher (approximately 50% of total carotenoids) when CrtZ was cloned upstream of CrtW, but still significant (approximately 30%) when the position of genes was inverted. Therefore, the engineered strains constitute a useful tool for investigating the ketocarotenoid biosynthetic pathway in cyanobacteria and an excellent starting point for further optimisation and industrial exploitation of these organisms for the production of added‐value compounds.     

Transgenic carrot plants accumulating ketocarotenoids show tolerance to UV and oxidative stresses

Ketocarotenoids are strong antioxidant compounds which accumulate in salmon, shrimp, crustaceans and algae, but are rarely found naturally in higher plants. In this study, we engineered constitutive expression of an algal beta-carotene ketolase gene (bkt) in carrot plants to produce a number of ketocarotenoids from beta-carotene. These included astaxanthin, adonirubin, canthaxanthin, echinenone, adonixanthin and beta-cryptoxanthin. Leaves accumulated up to 56mug/g total ketocarotenoids and contained higher beta-carotene levels but lower levels of alpha-carotene and lutein. The photosynthetic capacity of transgenic plants was not significantly altered by these changes. However, when high-expressing transgenic plants were exposed to UV-B irradiation, they grew significantly better than the wild-type controls. Similarly, leaf tissues exposed to various oxidative stresses including treatment with H(2)O(2) and methyl viologen showed less injury and retained higher levels of chlorophyll a+b. Total carotenoid extracts from transgenic leaves had higher antioxidant and free-radical scavenging activity in vitro compared to control leaves. Transgenic tissues also accumulated lower amounts of H(2)O(2) following exposure to oxidative stresses, suggesting that free radical and reactive oxygen species were quenched by the ketocarotenoids.     

Functions of chromophore-containing domain in the large linker L<Subscript>CM</Subscript>- polypeptide of phycobilisome

In the large linker ArcE polypeptide of the phycobilisome (PBS) from the cyanobacterium Synechocystis sp. PCC 6803, the chromophore-containing 26-kDa domain was deleted with consequent disturbance of the main PBS functions. Phycobilisomes in mutant cells staying in contact with photosystem I cannot transfer energy to the photosystem II. Under the bright light conditions, the interaction of PBSs with the photoprotective orange carotenoid protein in the mutant was lost and the implementation of transition states 1 and 2 of the pigment apparatus was significantly reduced.     

PetC1 is the major Rieske iron-sulfur protein in the cytochrome b6f complex of Synechocystis sp. PCC 6803

Many of the completely sequenced cyanobacterial genomes contain a gene family that encodes for putative Rieske iron-sulfur proteins. The Rieske protein is one of the large subunits of the cytochrome bc-type complexes involved in respiratory and photosynthetic electron transfer. In contrast to all other subunits of this complex that are encoded by single genes, the genome of the cyanobacterium Synechocystis PCC 6803 contains three petC genes, all encoding potential Rieske subunits. Most interestingly, any of the petC genes can be deleted individually without altering the Synechocystis phenotype dramatically. In contrast, double deletion experiments revealed that petC1 and petC2 cannot be deleted in combination, whereas petC3 can be deleted together with any of the other two petC genes. Further results suggest a different physiological function for each of the Rieske proteins. Whereas PetC2 can partly replace the dominating Rieske isoform PetC1, PetC3 is unable to functionally replace either PetC1 or PetC2 and may have a special function involving a special donor with a lower redox potential than plastoquinone. A predominant role of PetC1, which is (partly) different from PetC2, is suggested by the mutational analysis and a detailed characterization of the electron transfer reactions in the mutant strains.