Oxygenic Phototrophs Need ζ-Carotene Isomerase (Z-ISO) for Carotene Synthesis: Functional Analysis in Arthrospira and Euglena (2024)

Article history

Abstract

Introduction

Carotenoids are synthesized in all phototrophic organisms and many non-phototrophic organisms, including bacteria, archaea, algae, fungi and land plants. In phototrophic organisms, they play important roles in light-harvesting, protection of excess light, assembly of pigment–protein complexes and stabilization of lipid membranes (Takaichi 2013).

The carotenoid biosynthesis pathway has been studied in various organisms. The first step is the condensation of two molecules of geranylgeranyl diphosphate (GGPP) to produce phytoene, catalyzed by phytoene synthase (PSY, CrtB) in all organisms. Phytoene is converted to lycopene via a series of desaturation and isomerization events; the enzymes involved in this reaction vary depending on the species. Most bacteria and fungi use a single-enzyme phytoene desaturase (PDS, CrtI), whereas land plants require four enzymes—PDS (CrtP), ζ-carotene desaturase (ZDS, CrtQ), ζ-carotene isomerase (Z-ISO) and cis-carotene isomerase (CrtISO, CrtH)—to catalyze desaturation and isomerization reactions (Takaichi and Mochimaru 2007, Wurtzel 2019). The gene encoding Z-ISO has been functionally identified in only two angiosperms, Arabidopsis thaliana and Zea mays (Li etal. 2007, Chen etal. 2010), and has been little studied in other organisms.

Arthrospira, also known as ‘Spirulina’, is a filamentous, non-N₂-fixing cyanobacterium that comprises large amounts of protein, polyunsaturated fatty acids, vitamins, minerals and valuable pigments such as phycocyanin, chlorophyll and carotenoids (Spolaore etal. 2006). This cyanobacterium is commercially produced for food and feed as a food coloring and additive. The complete genome sequences of six species of Arthrospira, including Arthrospira platensis NIES-39 (Fujisawa etal. 2010), have been determined, and the identification and manipulation of genes involved in producing important gene products in Arthrospira is expected. Carotenoids have been previously identified in A. platensis NIES-39 (Sugiyama etal. 2017): the major carotenoids were β-carotene and zeaxanthin, and the minor ones were myxol 2′-methylpentoside, oscillol 2,2′-dimethylpentoside, 3′-hydroxyechinenone and β-cryptoxanthin. We have previously identified lycopene β-cyclase (CruA) and β-carotene hydroxylase (CrtR) genes of A. platensis NIES-39 by functional complementation (Sugiyama etal. 2017).

Euglenophyta, a division of eukaryotic unicellular phytoflagellates, is widely distributed in freshwater habitats, and has been proposed as a feedstock to produce biodiesel and various valuable compounds such as β-1,3-glucan known as paramylon (Gissibl etal. 2019). Although Euglenophyta members comprise β-carotene, zeaxanthin, diatoxanthin, diadinoxanthin and 9′-cis-neoxanthin, little is known about the corresponding genes related to carotenoid synthesis in this microalgal phylum. In Euglena gracilis, one of the best studied Euglenophyta, some genes required for carotenoid biosynthesis have been functionally identified: the GGPP synthase gene (crtE), the PSY gene (crtB), the PDS genes (crtP1 and crtP2), the ZDS gene (crtQ) and the β-carotene hydroxylase gene (CYP97H1) (Kato etal. 2016, 2019; Tamaki etal. 2019).

In this study, Z-ISO genes were isolated from the cyanobacterium A. platensis NIES-39 and eukaryotic microalga E. gracilis, and characterized by functional complementation analysis in transformed Escherichia coli, which comprise 9,15,9′-tri-cis-ζ-carotene, respectively. This is the first functional analysis of Z-ISO as a ζ-carotene isomerase in cyanobacterial and eukaryotic microalgal species. The absence of this gene in the genomes of anoxygenic phototrophs, such as green sulfur bacteria and Chloracidobacterium, was confirmed.

Results and Discussion

Isolation of the ζ-carotene isomerase genes from A. platensis and E. gracilis

To identify ζ-carotene isomerase (Z-ISO) genes, a homology search of the genome sequence of A. platensis NIES-39 (Fujisawa etal. 2010) was performed, using the amino acid sequence of the Z-ISO in A. thaliana [NCBI reference sequences (RefSeq) accession no. NP_563879.1] as a query (Table1). One candidate Z-ISO gene [DNA Data Bank of Japan (DDBJ) accession no. BAI90028] was found and isolated from A. platensis NIES-39. We also found candidate Z-ISO genes in other Arthrospira species, Arthrospira sp. PCC 8005, A. platensis C1 (PCC 9438), A. platensis strain Paraca, A. platensis YZ and A. maxima CS-328 (Table1).

We also obtained the nucleotide sequence of the candidate Z-ISO gene in E. gracilis from the transcriptome data (Yoshida etal. 2016). The cDNA sequence of EgZ-ISO comprised a spliced-leader sequence, 5′-TTTTTTTTCG-3′, transferred to the 5′ end of transcripts of E. gracilis (Tessier etal. 1991), indicating that the corresponding sequence codes for the full-length cDNA of this gene. The cDNA for EgZ-ISO was isolated from E. gracilis by reverse transcription-PCR (RT-PCR). The sequencing of EgZ-ISO cDNA was submitted to the DDBJ under accession number LC380019 (protein ID: BBD75421).

The alignment of the deduced amino acid sequences of Z-ISO from A. platensis, E. gracilis and A. thaliana is shown in Fig.1. The deduced amino acid sequences of Z-ISO from A. platensis and E. gracilis were highly conserved relative to that of the functional Z-ISO from A. thaliana (58% and 62% identities, respectively), except for N-terminal region, implying a potential functional similarity between these proteins. The ChloroP method (Emanuelsson etal. 2000) also predicted a putative transit cleavage site at residues 58 and 126 for Arabidopsis (Chen et al. 2010) and Euglena Z-ISO, respectively (Fig.1). The TMHMM program (http://www.cbs.dtu.dk/services/TMHMM/) predicted two transmembrane domains, which are found in typical plastid-targeted proteins in E. gracilis (Durnford and Gray 2006), in the N-terminal region of EgZ-ISO, suggesting the plastid localization of the EgZ-ISO protein (Supplementary Fig. S1).

Functional analysis of Z-ISOs from Arthrospira and Euglena

First, the function of Z-ISO from A. platensis and E. gracilis was confirmed using a heterologous E. coli expression system. The E. coli strain carrying the plasmid pET-EBP that comprises the crtE and crtB genes from the soil bacterium Pantoea ananatis (previously, Erwinia uredovora), and the crtP gene from A. platensis cultured in the darkness, showed three HPLC peaks with characteristic spectrums of ζ-carotene isomers, which was consistent with earlier reports (Breitenbach and Sandmann 2005) (Fig.2A). The first HPLC peak showed a comparably high cis peak at 296 nm relative to the maximum peak at 399 nm in the visible region (see spectra; Fig.2, right), consistent with 9,15,9′-tri-cis-ζ-carotene (Breitenbach and Sandmann 2005). When cultured in light, the first HPLC peak was reduced and the second peaks was increased (Fig.2E), as photoisomerization of 15-cis double bond of 9,15,9′-tri-cis-ζ-carotene to form 9,9′-di-cis-ζ-carotene (Breitenbach and Sandmann 2005). When the Z-ISO gene of A. thaliana was expressed in the E. coli cells cultured in darkness, the first HPLC peak was reduced and the second peaks was markedly increased (Fig.2D). The elution profiles, absorption spectra and molecular masses of these peaks (Fig.2) were consistent with those previously reported for Z-ISO from A. thaliana (Chen etal. 2010). Consequently, the first, second and third peaks might reflect the presence of 9,15,9′-tri-cis-ζ-carotene, 9,9′-di-cis-ζ-carotene and all-trans-ζ-carotene, respectively.

When the Z-ISO genes of A. platensis and E. gracilis were similarly expressed in the E. coli strain in darkness, both peaks 2 and 3 were detected (Fig.2B, C). Therefore, both Z-ISO genes were able to isomerize 9,15,9′-tri-cis-ζ-carotene to 9,9′-di-cis-ζ-carotene, and all-trans-ζ-carotene in darkness, similar to the ability of Z-ISO of A. thaliana (Fig.2D). This is the first evidence that Z-ISO is a ζ-carotene isomerase in cyanobacterial and eukaryotic microalgal species. Furthermore, we have shown that Z-ISO is universally functional in oxygenic phototrophs including cyanobacteria, eukaryotic microalgae and land plants.

As aforementioned, when E. coli cells comprising pET-EBP and the empty vector (pACYCDuet-1) were cultured in darkness, the strains accumulated one major carotene, namely 9,15,9′-tri-cis-ζ-carotene (Fig.2A). On the contrary, when cultured in light, they accumulated both 9,9′-di-cis-ζ-carotene and all-trans-ζ-carotene (Fig.2E), as previously reported (Breitenbach and Sandmann 2005; Li etal. 2007). From these results, it appears that some isomerization is non-enzymically caused by light. Arabidopsis Z-ISO mutants showed the delayed greening phenotype (Chen etal. 2010). Also, both Z-ISO mutants in Arabidopsis and maize had reduced carotenoids and chlorophylls compared to the each wild-types, even grown under light conditions (Chen etal. 2010). Therefore, the Z-ISO activity is needed in photosynthetic organisms.

Phylogenetic distribution of Z-ISO

To investigate the phylogenetic distribution of Z-ISO, we performed a genome-wide survey of Z-ISO in phototrophs. We found that Z-ISO homologs were widely distributed in the genome of oxygenic phototrophs, such as cyanobacteria, rhodophyta, heterokontophyta, haptophyta, euglenophyta, chlorophyta and tracheophyta. All of which have oxygenic PSI- and PSII-type photosystems, as a single-copy gene (Table1).

To determine the phylogenetic distribution of Z-ISO, we generated the phylogenetic tree by alignment of Z-ISO amino acid sequences of oxygenic phototrophs (Fig.3). As a result, the Arthrospira Z-ISO was placed in the same clade as other cyanobacterial Z-ISO. EgZ-ISO was confirmed to be closely related to Z-ISO of species of chlorophyta such as Chlamydomonas reinhardtii and Chlorella variabilis. The neighboring clade comprises Z-ISO of tracheobiota members, specifically A. thaliana, Solanum lycopersicum, Z. mays and Oryza sativa. These results suggest that Z-ISO is conserved in oxygenic phototrophs, although the limited number of genomes was assessed in this study.

Z-ISO from green sulfur bacteria and Chloracidobacterium genomes

Homologous Z-ISO genes were searched in genomes of green sulfur bacteria (Chlorobaculum tepidum TLS, Chlorobaculum limnaeum strain DSM 1677, Chlorobaculum parvum NCIB 8327, Chlorobium sp. GBChlB, Chlorobium luteolum DSM 273, Chlorobium limicola DSM 245, Chlorobium phaeovibrioides DSM 265 and Chlorobium phaeobacteroides BS1) and Chloracidobacterium thermophilum strain B^T, which have PSI-type photosystem and produce lycopene by combination of crtP and crtQ as cyanobacteria, but we could not find the homologous Z-ISO genes. On the contrary, homologous crtH/crtISO genes encoding for cis-carotene isomerase were also found in both green sulfur bacteria and Chloracidobacterium. The absence of Z-ISO enzymes in green sulfur bacteria and Chloracidobacterium might be compensated by the function of light (Fig.2E).

We functionally identified the ζ-carotene isomerase genes from A. platensis and E. gracilis as homologs of the functional Z-ISO gene from A. thaliana. Our findings suggest that Z-ISO is widely distributed and functions in oxygenic phototrophs, including cyanobacteria, eukaryotic microalgae and land plants. In contrast, we could not find Z-ISO homologs in the genome of anoxygenic phototrophs, including green sulfur bacteria and Chloracidobacterium. Thus, Z-ISO is a functional component of carotene synthesis in oxygenic phototrophs, whereas our findings suggest that it is absent in anoxygenic phototrophs using the crtP and crtQ desaturation system.

Materials and Methods

Cloning of Z-ISO from A. platensis NIES-39

To find sequences homologous to the known ζ-carotene isomerase (Z-ISO) genes, a homology search of the genome sequence was performed using the amino acid sequence of Z-ISO in A. thaliana (RefSeq accession no. NP_563879.1) as a query. The Z-ISO genes were amplified by PCR from A. platensis NIES-39 genomic DNA as per Sugiyama etal. (2017). The following primers were used: 5′-TACATATGGCAGATCTCTTGGAGTTAACTTTATTATCCGT-3′ as a forward primer and 5′-TTACCAGACTCGAGGGTACCCTACCAGCCTACGTTACTTGT-3′ as a reverse primer. The PCR products were cloned into the pACYCDuet-1 vector (Merck Millipore) by an infusion cloning reaction, and sequenced. All infusion cloning reactions were performed using In-Fusion Cloning kits (Takara Bio), as per manufacturer’s instructions. Sequencing of these clones was carried using dideoxy chain-termination methods. The plasmid was designated pACYC-ApZISO.

Cloning of EgZ-ISO cDNA

For the identification of the Z-ISO gene in E. gracilis, TBLASTN analysis was carried out against a E. gracilis transcriptome data (Yoshida etal. 2016) using the amino acid sequence of Z-ISO in A. thaliana as a query. cDNA of EgZ-ISO was isolated from E. gracilis by RT-PCR, cloned into the pMD-20T vector, and sequenced as per Kato etal. (2016), with primers designed based on the transcriptome data (Yoshida etal. 2016). The following primers were used: 5′-TTCTCCATGCCTTATTATTC-3′ and 5′-GATACATGTTGGTACATATTG-3′. The plasmid was designated pMD-EgZISO.

For functional analysis, the EgZ-ISO cDNA was amplified by PCR using pMD-EgZISO as a template, and cloned into the pACYCDuet-1 vector by an infusion cloning reaction, and sequenced. The following primers were used: 5′-TACATATGGCAGATCTCATGCCTTATTATTCTTTGCCA-3′ and 5′-TTACCAGACTCGAGGTCACCACGGCAGGCTGTA-3′. The plasmid was designated pACYC-EgZISO.

Cloning of AtZ-ISO cDNA

Total RNA was isolated from the leaves of A. thaliana with RNeasy Plant Mini Kit (QIAGEN). First-strand cDNA was synthesized with SuperScript First-Strand Synthesis System for RT-PCR (Thermo Fisher Scientific) form total RNA. The Arabidopsis Z-ISO cDNA was isolated by RT-PCR. The following primers were used: 5′- TACATATGGCAGATCTCATGGCGGTTTATCATCTCCT -3′ as a forward primer and 5′- TTACCAGACTCGAGGGTACCTTACCAATGAAGCCTAAAGCT -3′ as a reverse primer. The PCR products were cloned into the pACYCDuet-1 vector by an infusion cloning reaction, and sequenced. The plasmid was designated pACYC-AtZISO.

Sequence analysis and phylogenetic analysis

A sequence similarity search was performed by BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Multiple sequence alignments of amino acid sequences were conducted using CLUSTALW (http://clustalw.ddbj.nig.ac.jp/) and presented using BOXSHADE (https://embnet.vital-it.ch/software/BOX_form.html). A phylogenetic tree was generated using MEGA X software (Kumar etal. 2018), using the Neighbor-Joining method with the bootstrap value obtained using 1000 replicates. Complete genome sequences of anoxygenic phototrophs were obtained from GenBank [C. limnaeum strain DSM 1677 (CP017305.1), C. parvum NCIB 8327 (CP001099.1), C. limicola DSM 245 (CP001097.1), C. phaeovibrioides DSM 265 (CP000607.1), C. phaeobacteroides BS1 (CP001101.1) and C. thermophilum strain B^T (CP002514.1, CP002515.1)] and an Ensembl genome [C. tepidum TLS (ASM698v1), C. sp. GBChlB (ASM72417v1) and C. luteolum DSM 273 (ASM1248v1)].

Functional analysis of Arthrospira Z-ISO and EgZ-ISO in E. coli cells

To test for functional complementation of Arthrospira Z-ISO and EgZ-ISO, the plasmid pET-EBP, conferring 9,15,9′-tri-cis-ζ-carotene accumulation, was first constructed (Breitenbach etal. 2001; Breitenbach and Sandmann 2005). The crtE and crtB genes of P. ananatis (previously, E. uredovora) and the Arthrospira crtP gene were amplified by PCR using the plasmid pACCRT-EB (Misawa etal. 1995; Sugiyama etal. 2017) and A. platensis NIES-39 genomic DNA, respectively, as templates. The following primers were used: 5′-CCAGGATCCGATTCGATGACGGTCTGCGCAAAAA-3′ as a forward primer and 5′-TGCGGCCGCAAGCTTCTAGAGCGGGCGCTGCCAGA-3′ as a reverse primer for crtE and crtB; and 5′-GCAGATCTGATGCGAGTTGCGATC-3′ as a forward primer and 5′-GGGTACCTTAGCTAGTCCTCAAGTCTTGT-3′ as a reverse primer for crtP. The crtE and crtB fragment was cloned into multi-cloning site 1 of pETDuet-1 vector (Merck Millipore), and sequenced. The crtP fragment was cloned into the MCS2 of the pETDuet-1 vector comprising the crtE and crtB genes. The plasmid was designated pET-EBP. Escherichia coli strain BL21 (DE3) (Thermo Fisher Scientific) carrying the plasmid pET-EBP was used as the host for synthesizing 9,15,9′-tri-cis-ζ-carotene. The plasmids pACYC-ApZISO, pACYC-EgZISO and pACYC-AtZISO were transformed into the 9,15,9′-tri-cis-ζ-carotene-accumulating E. coli. The co-transformed cells were pre-cultured at 25°C until the OD₆₀₀ reached 0.4–0.6 in Luria–Bertani broth comprising chloramphenicol (30 mg l⁻¹) and ampicillin (50 mg l⁻¹). The transformants were then treated with 0.5 mM isopropyl-β-D-thiogalactoside, followed by culture for 24 h at 25°C.

Extraction and analysis of carotenoids from E. coli cells

Pigments from E. coli cell pellets were extracted with cold acetone using a mixer for 5 min. After centrifugation, the supernatant fraction was collected and dried under N₂ flow. The dried residues were dissolved in methyl tert-butyl ether/methanol (7 : 3, v/v). HPLC analysis was performed on a Shimadzu Prominence series machine (Shimadzu) with a photodiode array detector (SPD-M20A, Shimadzu). A C₃₀ YMC column (250 × 4.6 mm, 5 μm) (YMC) was employed for the separation. The extract was eluted at a rate of 0.8 ml/min with solvent A (water/methanol, 5 : 95, v/v) for 2 min, followed by a linear gradient from solvent A to solvent B (methyl tert-butyl ether/methanol, 7 : 3, v/v) for 23 min and solvent B alone for 15 min. For mass analysis, an LTQ Orbitrap hybrid ion-trap/Fourier transform mass spectrometer (Thermo Fisher Scientific) with atmospheric pressure chemical ionization was used as per Sugiyama etal. (2017). The capillary temperature was set to 250°C, the atmospheric pressure chemical ionization vaporizer temperature was held at 375°C and the capillary voltage was optimized to 9 V. Screening was performed in full scan mode, covering the range from m/z 200 to 1000. Carotenoids were identified by comparison of their retention time, characteristic absorption spectra and mass spectra of ζ-carotene isomers with previously published data (Breitenbach and Sandmann 2005; Chen etal. 2010).

Funding

The Japan Society for the Promotion of Science [grant number 17K07945 to T.S.].

Acknowledgments

We thank Professor Norihiko Misawa (Ishikawa Prefectural University, Japan) for providing the pACCRT-EB plasmids.

Footnote: The nucleotide sequence of Z-ISO from Euglena reported in this paper has been submitted to DDBJ under accession number LC380019.

Disclosures

The authors have no conflicts of interest to declare.

References

© The Author(s) 2019. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oup.com

Issue Section:

Regular Papers

Email alerts

Citing articles via

More from Oxford Academic