Skip to main content
European Commission logo print header

Production of fungal carotenoids for healthy nutrition

Risultati finali

Ubiquinone (coenzyme Q) is an intermediary in redox reactions. Ubiquitous in living forms, widespread in the cells, ubiquinone is an essential component of the mitochondrial respiratory chain, where it acts as a carrier of hydrogen atoms. Ubiquinone contains a quinone ring and a terpenoid chain, which are synthesised separately. The hydrophobic terpenoid tail serves to anchor the coenzyme to membranes. Industry appreciates the beneficial effects of ubiquinone on health. It has antioxidant properties and protects against highly reactive chemicals and free radicals. It reduces blood pressure and prevents congestive heart failure. The amounts in the body varies in different organs, with highest concentrations in the heart, and normally declines with age. The Zygomycetes Blakeslea trispora and Phycomyces blakesleeanus accumulate ubiquinone, ß carotene, various different sterols, and other terpenoids, all of them produced via the mevalonate pathway. These fungi are used or could be used for the industrial production of these terpenoids, edible oil, chitosan, and various organic acids. By measuring the specific radioactivity of terpenoids made from radioactive mevalonate, leucine or acetate in the presence of excess glucose in wild types and mutant strains we have concluded that these fungi have separate subcellular compartments for the production of carotene, sterols and triacylglycerols. The terpenoid moiety of ubiquinone is synthesised in the same compartment as ergosterol. These compartments contain separate pools of all their common metabolites, beginning from acetyl-CoA. Mevalonate carbon atoms do not find their way back to general metabolism, i. e., these fungi lack the shunt pathway. The compartments are regulated independently. The very large variations in carotene content caused by many environmental and genetic changes are not accompanied by variations in the ubiquinone content, which is about 0,3 mg ubiquinone / g dry mass in both fungi. The ubiquinone content increases when the cultures grow on leucine or acetate as carbon sources and is not affected by illumination. The variations are modest: Blakeslea contains up to 0,8 and Phycomyces up to 0,6 mg ubiquinone/ g dry mass depending in the media composition. Phycomyces, but not Blakeslea, increases the production of ubiquinone in presence of oligomycin.
Two prominent genes of the carotenogenic pathway in Xanthophyllomyces encoding for pytoene synthase / lycopene cyclase and astaxanthin synthase were expressed and functionally characterized. Furthermore, an improvement of the catalytic efficiency of phytoene synthase by E.coli-mediated random mutagenesis was attempted. A promising candidate for the astaxanthin synthase gene is a novel monooxygenase of the cytochrome P450 3A subfamily type since its expression in a beta-carotene producing Xanthophyllomyces mutant led to the formation of astaxanthin. Therefore, the function of this monooxygenase to catalyse the introduction of 3-hydroxy groups and 4-keto groups into beta-carotene as analysed. First, the cDNA of this gene was expressed in E. coli with an established ß-carotene background together with a cDNA for a NADPH cytochrome P450 reductase from yeast. After supplementation of hemin ore aminolevulinic acid, ketolase activitity could be established but no hydroxylation observed. Since only small amounts of canthaxanthin are formed in E. coli, a possible subsequent hydroxylation step may be prevented by these low substrate level. In a next step, defined Xanthophyllomyces mutants were generated. Into a the beta-carotene accumulating mutant that produced astaxanthin upon transformation with the P450 gene, either a hydroxylase or ketolase gene was introduced. The ketolase transformant yielded high amounts of canthaxanthin as final product but no astaxanthin. In the hydroxylase mutant large amounts of zeaxanthin were found but again no astaxanthin. The conclusion drawn from these results indicate that the P450 gene encodes an enzyme which catalyzes two reactions: first ketolation of beta-carotene and then a hydroxylation step forming astaxanthin as the final product. Phytoene synthase and lycopene cyclase are encoded by a fusion gene. In this gene, the N-terminal lycopene cyclase region and the C-terminal phytoene synthase region is separated by a putative protease cleavage motif. In order to test the hypothesis that that the protein is post-translationally cleaved into two independent functional enzymes, truncated gene constructs were made and assayed in the E. coli complementation system. One protein expressed from amino acid 249 to 674 was active exclusively as a phytoene synthase. With another expressed enzyme starting from amino acid 1 to exactly 254, it was possible to obtain a protein capable of lycopene cyclization to beta-carotene. This result demonstrated that an individual lycopene cyclase cleaved off the phytoene synthase moity is enzymtically active. For a further biochemical characterization of the phytoene synthase/lycopene cyclase heterologous expression in substantial quantity is important. Expression of the full-length cDNA with a 6-His moiety for metal chelate affinity purification in E. coli was quite low due to 54 rare codons. Therefor, plasmids with t-RNAs for some of these codons were co-expressed. Finally, substantial expression of phytoene / -carotene synthase was achieved in the E. coli strain Rosetta which is capable to synthesize all the rare t-RNAs. However, the yield was comparably low not exceeding 3 to 4% of total protein. Since phytoene synthesis is the bottle-neck in the carotenogenic pathway in Xanthophyllomyces we mutagenized the phytoene synthase / lycopene cyclase gene by error prone PCR or with the mutator E. coli strain XL-1 red and screened for higher phytoene synthase activity or modified cyclization reaction. In several mutants higher phytoene synthesis capacity or a more pronounced cyclization reaction to bi-cyclic products than in the original transformant was obtained. All new phenotypes could be attributed to mutations in the promoter resulting in higher enzyme expression. Thus, no genetically improved phytoene synthase or lycopene cyclase was obtained. However, our results indicate that over-expression of both enzymes will increase carotenoid production and support formation and of bi-cyclic astaxanthin versus 3-HO-4-keto-torulene. Our finding that higher phytoene synthase activities leads to higher astaxanthin production let us look for physiological conditions that up-regulate carotenogenesis in Xanthophyllomyces.
Cloning and characterization of the carB and carRA genes from the beta-carotene overproducing strain B. trispora F-744 (GenBank AY176662). An 8788 bp DNA region contained two complete ORFs read in opposite directions corresponding to the carB and carRA genes. Two introns were predicted in carB and one in carRA. Intron splicing in carRA was proved by cDNA sequence. General transcription signals were detected at the 5-flanking region (TATA and CAAT boxes) and 3-flanking region (AATAAA polyadenylation signal) as were poorly conserved APE-like elements (consensus sequence GAANNTTGCC) involved in gene regulation in response to light. Three copies of the binding site (TTCTTTGTTY) for the transcription factor Ste11 (a �TR box�) were found with minor changes in the carB-carRA promoter region of B. trispora. A conserved sequence (GCXTGTATXTTATAXAAAXAAXAA) including the TATA-box is present upstream of the translation start codon of carB (position -78) and carRA (position -68) genes. RNA secondary structures that can act as potential transcriptional terminators were downstream of carB and carRA genes. Deduced amino acid sequences encoded by the carB and carRA genes of B. trispora F-744. The CarB deduced protein is similar in size, sequence, and signature pattern to other fungal phytoene dehydrogenases, with amino acid sequence similarity to phytoene dehydrogenases from M. circinelloides (81%), P. blakesleeanus (72%), Xanthophyllomyces dendrorhous (50%), N. crassa (49%), Fusarium fujikuroi (48%), and Cercospora nicotianae (47%). A consensus dinucleotide binding motif for the FAD superfamily was located at the N-terminus (IVVIGAGIGGTATAARLAREGFRVTVVEKNDFSGGRCSFIHHDGHRFDQG), and a bacterial-type phytoene dehydrogenase signature sequence, corresponding to a carotenoid-binding domain, was found at the C-terminus (NLFFVGASTHPGTGVPIVLAG). The deduced CarB protein has a highly hydrophobic region at the C-terminal end that is characteristic of membrane-associated proteins. The carRA deduced protein was similar to fungal proteins with lycopene cyclase/phytoene synthase activity, and had high similarity to lycopene cyclases/phytoene synthases from M. circinelloides (67%), P. blakesleeanus (55%), X. dendrorhous (31%), F. fujikuroi (29%), and N. crassa (28%). The CarRA protein had two different domains, as described for P. blakesleeanus, M. circinelloides, and X. dendrorhous: a hydrophobic transmembrane domain (R), located near to the 5'-end encoding lycopene cyclase (240 residues), and a hydrophilic domain, P, downstream that encodes phytoene synthase (362 residues). A putative protease cleavage site AQAILH (residues 241-246) that could split the polypeptide between residues 243 and 244 is present at the boundary between the two domains. The wild type carB and carRA genes from B. trispora (GenBank AY176663). The B. trispora wild type (NRRL2457) and mutant (F-744 beta-carotene overproducing) strains both had the same hybridization pattern in Southern blots, suggesting that carB and carRA are present at a single copy per genome. The carB and carRA mutant and wild type genes differ from each other by two nucleotides per gene (four total differences). The A1791C mutation in carB substitutes an arginine for the wild-type serine at position 528 (S528R). An A1958G transition mutation occurred in the carB stop codon, changing the wild-type TAA to TAG, but not affecting the protein sequence. The T497C mutation in the carRA gene changes the wild-type proline at position 143 to a serine (P143S), while the C568T mutation does not change the amino acid encoded (glycine at 166). Transcripts of the carB (1.8 kb) and carRA (1.9 kb) genes were the expected size and present at a constant level during the fermentation. Expression of the carB and carRA genes of B. trispora F-744 in M. circinelloides. Several transformants for each strain/plasmid combinations were grown on YNB plates. The mycelium of the MS8/pLeu4 and MS23/pEPM9 control transformants remained as white as the parental strains, while MS8/pALBT119 and MS23/pALBT120 had a weak yellowish coloration, suggesting the accumulation of carotenoids other than phytoene. Southern analyses of digested genomic DNA with probes from carRA and carB genes are consistent with autonomous replication, the absence of rearrangements, and the lack of hybridization with the endogenous carRP gene of M. circinelloides. The carotenoids accumulated by wild-type strains of M. circinelloides, mutants and transformants were analyzed by HPLC. At present, beta-carotene is produced commercially by chemical synthesis. We are now using a new industrial process with B. trispora to produce beta-carotene. Although neither gene number nor sequence in B. trispora can yet be altered by transformation, the carB and carRA sequences described here can be used to guide the development of overproducing strains that are altered either in these genes themselves or in other genes that affect their expression.
Membrane and globule formation was monitored by light and electron microscopy in Blakeslea. Light microscopy studies using Nomarski optics showed distended hyphal filaments within which coloured compounds accumulated. An EM study of globule formation in the mated strains of F986 and F921 in submerged culture revealed that at 48 h post inoculation, the lipid globules contained internal membranes similar to the prolamellar bodies of etioplasts. At the point of first appearance of colour, 3 days post inoculation, the lipid globules were filled with an extensive internal membrane system not dissimilar to the grana stacks of the thylakoid membrane. These membranous structures then disappeared as the globules increased in size later in development: Globules purified at 7 days post inoculation were devoid of internal membranes and had a greater size distribution. In Xanthophyllomyces, a large vacuole-like cavity was observed by electron microscopy. The vacuole-like cavity in the mutant strain was observed to occupy a larger proportion of the cell area compared with the wild type strain. No internal membrane systems were found. Protein and transcript levels were not determined. Globule associated proteins in Blakeslea were separated by SDS-PAGE. At least two proteins, and in some cases three, were found in lipid globules from the wild type strains F921 and F986, and from the mutants SB 38 and SB52. The proteins were less than 30 kDa and highly hydrophobic. The absence of a known genome hindered protein identification. Poor growth in submerged culture hindered large-scale globule isolation for the purification of sufficient quantities of protein for identification by tandem mass spectrometry and Edmon sequencing.
A transformation protocol allowing the generation of hygromycin-resistant mycelia of Blakeslea trispora has been developed based in the use of Agrobacterium tumefaciens containing a binary vector. Among five different Agrobacterium strains checked, best growth was exhibited by SK502 and GV2260. The first one was chosen for the transformation experiments. Samples of 10e5, 10e6 and 10e7 fresh spores from Blakeslea in 100 µl were mixed with 100µl of the Agrobacterium culture previously treated with 4'-hydroxy-3,5-dimethoxyacetophenone (acetosyringone). The strains SK561, SK562 and SK563 were obtained from SK502 upon introduction of the plasmid pBHT2, containing the HygR gene (strain SK561), or with derivatives of this plasmid containing the gene pyrG from either Phycomyces blakesleeanus (plasmid pIP1, strain S562) or from Mucor circinelloides (plasmid pIM1, strain SK563). Different Blakeslea wild type strains were used. Samples of the Blakeslea/Agrobacterium mixtures, previously incubated up to three days at temperatures from 22°C to 28°C on a medium containing acetosyringone, were transferred to selection plates supplemented with 0.5mM cefotaxime as an Agrobacterium-specific antibiotic, and 200mg/l hygromycin to select transformation. The medium also contained 2% DMSO to increase hygromycin sensitivity. Incubations were done on cellophane or filter paper laying on the agar medium. Putative transformants were obtained from strains F986 and CBS198.90, but no mycelia growing under selective conditions were obtained with other strains. Primary transformants were able to grow when transferred to the hygromycin/DMSO supplemented medium. These putative transformants were able to grow upon a new transfer to fresh selective medium, although some of them exhibited a slower growth rate. The borders of the colonies were more irregular than control plates without selection, indicating either a lack of stability or segregation derived from heterokaryosis. Those with the best growth were selected for DNA extraction and demonstration of the presence of the hygR gene. These putative transformants retained the ability to grow in the presence of hygromycin and southern blot analysis with the hygR probe showed the presence in at least two of them of a low size band in undigested samples, suggesting the presence of free DNA carrying the hygR gene. This result demonstrates that at least some of the strains are authentic transformants. However, the signals were lost upon subculturing, indicating lack of stability of the hygR DNA. Consequently, the transformants are not stably maintained.
Several Blakeslea mutants with enhanced ß-carotene production are available. Furthermore, mating of plus and minus strains highly stimulates carotenogenesis. In those mutants and matings up-regulation of enzyme levels and activities were investigated in correlation to the ß-carotene formation. At first, antisera were produced against HMGCoA reductase, the phytoene synthase part of CarRA, the lycopene cyclase part of CarRA (all from Phycomyces) and against IPP isomerase from yeast. After heterologous expression of appropriate segments of these enzymes, the proteins were isolated and used for immunization of rabbits. The partially purified antisera against HMG-CoA reductase and IPP isomerase were not sensitive enough for Blakeslea to detect both proteins either in crude extracts or enriched preparations. Upon subcellular fractioning a protein was detectable in the insoluble fraction of the mated mutants after ultracentrifugation of a size of the lycopene cyclase domain with the lycopene cyclase antiserum. With the phytoene synthase antiserum a corresponding soluble protein was detectable documenting that the bifunctional lycopene cyclase/phytoene synthase protein is post-translationally cleaved to an individual soluble phytoene synthase of 36 kDa and a membrane-bound lycopene cyclase of 30 kDa. The purified antisera against phytoene synthase allowed the detection and determination of this enzyme under different conditions. With matings, we could establish a kinetic for the formation of phytoene synthase. The levels of this enzyme increased from the mating on to day 4. Then, a steady decline was observed and disappearance of the phytoene synthase band at day 8. In parallel, ß-carotene accumulated over 4 days and afterwards stayed constant. These data are in support of phytoene synthase as a limiting enzyme of carotenogenesis in Blakeslea. Immuno determination of the phytoene synthase in Blakeslea mutants and matings demonstrated that the level of the synthase correlated with the amount of carotene formed. It is quite low in wild type, higher in carotenogenic mutants and increases about 10-fold in matings. In vitro assays of phytoene synthase resulted in considerably higher activity in mutants and matings compared to the wild type. A definite induction of the phytoene synthase / lycopene cyclase mRNA levels in the mated cultures upon mating and a tendency to decrease in the following days was observed. Transcription of the phytoene synthase / lycopene cyclase and phytoene desaturase genes of Blakeslea was also analyzed by Northern hybridization using RNA samples obtained from pilot-scale feed-batch fermentations for ß-carotene production. Steady-state levels of transcripts were found after 5 days. In Phycomyces, mRNA levels of the phytoene synthase / lycopene cyclase gene and phytoene desaturase gene were determined in the wild type, four overproducing mutants (carS, carD, carF and carScarF) and four structural mutants (carA, carR, carRA and carB). Cultures were grown for 24 h in the dark, in the presence or absence of two different chemicals (retinol and dimethyl phtalate) activating carotenogenesis, or in the light. After 24 h, transcription of these genes is similar in the wild type, in various mutants and in the wild type exposed to chemical activators. However, light strongly stimulates their mRNA levels in all strains tested, even in a mutant that shows very little increase in carotene synthesis after light exposure. The expression of two genes encoding HMGCoA synthase and reductase, two enzymes of the early terpenoid pathway, was also investigated. No activation of the mRNA levels was found for these genes either under chemical activation of the pathway or mutations in the regulatory genes carS, carF and carD. A minor increase of the mRNA levels was found in the light and after sexual stimulation. With increasing age of the Phycomyces culture, a decrease of the genes for phytoene synthase / lycopene cyclase and phytoene desaturase from day 3 on to day 7 was observed. In contrast, the mRNA levels of the HMGCoA synthase and reductase genes were constant or even slightly increased with age of the culture. In the course of this project we found physiological conditions that up-regulate carotenogenesis including astaxanthin production in Xanthophyllomyces. Low light intensity as well as oxygen supply increased total carotenoid formation considerably. A combination of both factors, resulted in a 4-fold total carotenoid increase. The astaxanthin yield under those cultivation conditions was even 5-fold higher attributed to a stronger oxygenase reaction. High oxygen up-regulated geranylgeranyl pyrophosphate synthase and phytoene desaturase messages whereas higher transcripts of phytoene synthase / lycopene cyclase and astaxanthin synthase accumulated under low oxygen. In each case, light was antagonistic.
A method was developed to facilitate the isolation of mutants with an increased lipid content, expected to result in a reduced cell density. The method is based in the centrifugation of spores from colonies surviving to a mutagenesis treatment in a ficoll solution of appropriate density. Simulation experiments with carotenoid overproducing mutants, previously shown to contain a higher lipid content, suggested the centrifugation at 4000 rpm in a 45% ficoll solution as the one suitable to separate wild type spores from low density spores. A search under these selecting conditions allowed the isolation of individual strains producing floating spores in further centrifugation tests, while the wild type spores in parallel experiments accumulated as a pellet in the bottom of the tubes. Some of the mutant strains exhibited morphological or pigmentation alterations. Carotenoid and lipid analyses were carried out with 19 candidate mutants. Most of them contained slight increases in carotenoid content, except one, that exhibited an strong increase (four-fold). Most of the mutants did not show major differences in lipid content, indicating that there are other cellular processes affecting cell density. New experiments were done under more stringent selection conditions (40% ficoll), requiring a lower density of the spores to float on the top of the solution. Two out of 13 floating strains isolated contained about 40% more lipids than the wild type. As in the previous set of experiments, most of the strains contained more carotenoids than the wild type. - Strains with improved carotene production have been obtained by mutation and heterokaryosis from two selected wild strains of opposite sex, F921 and F986. Following exposure of their spores to a chemical mutagen, high-carotene mutants were identified after a screening of about 200,000 colonies. The best mutant accumulated five times the amount produced by the wild type. Further increases in carotene content were obtained after a new round of mutagenesis in the best mutant. Highest production was achieved in intersexual heterokaryons with mutant nuclei of opposite sex. These contained up to 39 mg of beta-carotene or up to 15 mg of lycopene per g (dry mass) under standard laboratory conditions in which the original wild strains contained about 0.3 mg of beta-carotene per g dry mass. The production was shifted to lycopene in cultures incubated in the presence of nicotine and in lycopene-rich mutants derived from the wild strains. - Chemical analyses of a correlation between lipid and carotenoid by HPLC-PDA and GC were performed on submerged cultures of the Blakeslea strains F986 and F921 and their crosses at 3 and 7 days post inoculation. The mated cultures had elevated levels of lipids and carotenoids, which appeared earlier during development than the parental strains. The more than five-fold increase in the beta-carotene content of the mated strains in submerged culture did not correspond with an equally large increase in fatty acid content. Beta-carotene content was observed to increase during development whereas the fatty acid content did not. The mutants SB52 and SB38 accumulated high levels of carotenoids in surface cultures but did not have corresponding increases in fatty acids, as observed by TLC. In submerged culture, these strains did not pigment nor grow as well as the parental strains and were not used in subsequent analyses. Analyses were also performed on the low-density mutants M24 and M26 and compared to their wild type background F921. These mutants accumulated less carotenoids and fatty acids than the parental strain, indicating that fatty acid content was not responsible for the buoyant spore phenotype. No relationship between the biosynthesis and accumulation of carotenoids and lipids could be found in any analysis of Blakeslea. Investigations on inhibitors of carotenoid and fatty acid biosynthesis were therefore not performed. PUFA accumulation in the mycelia was analysed, and fatty acids identified by co-migration with known standards. Lipid globules were isolated and purified as part of sub cellular fractionation process. Carotenoids were found to accumulate within these organelles and internal membrane systems were observed during the early stages of development. Carotenoids from industrial Blakeslea biomass were visualised as crystalline deposits within the globules. Globule size appeared to correspond with increased carotenoid production in the mycelia though globule number did not seem to alter greatly. TLC analysis of lipids in the globules showed the presence of beta-carotene, fatty acids, diglycerides, monoglycerides and phospholipids. A cell-free system was established in Blakeslea strains F986, F921, SB52 and SB38. The incorporation of radiolabelled isotopes from MVA was observed in crude extracts but not in isolated lipid globules of 7-day post inoculation cultures.
The formation of b-carotene is highly induced when the two mating types of B. trispora mate. Molecular genetic analysis of carotenogenesis in Blakeslea trispora is being studied using cDNA AFLP transcript profiling technique. The cDNA AFLP technology has been set up in order to identify genes that are involved in this process. One of the important steps in this technique is the direct determination of the nucleotide sequence from the AFLP patterns. The common technique was modified to give good result for fungal gene expression. Numerous (fragments of) genes are in the process of being identified and their sequnce analysis and gene identification of genes that are potentially involved in the induced carotenogenesis process of B. trispora using bioinformatics techniques is under way.
A gene (ast) from Xanthophyllomyces encoding a cytochrome P450, described in a F. Hoffman La Roche-owned patent (EP 1 035 206 A1). The gene product was able to convert b-carotene to astaxanthin after the introduction of the gene in a b-carotene accumulating mutant of X. dendrorhous. Based on information from patent literature, the ast gene of Xanthophyllomyces strain CBS 6938 (our model strain), including the promoter and terminator regions, was isolated by PCR techniques. Furthermore, an ast cDNA copy was isolated from a Xanthophyllomyces cDNA library, also by PCR. Expression of the ast gene in a b-carotene accumulating Xanthophyllomyces strain (PR-1-104) restored astaxanthin production, indicating that the gene is functionally expressed. The ast gene was disrupted by insertional inactivation. A fragment of the ast gene was cloned into a X. dendrorhous vector that contained a G418 selection marker. The linearized vector was transferred to X. dendrorhous by means of electroporation. By over-expression of the homologous carotenogenic ast gene in strain(s) CBS 6938 (and PR-1-104), an increase in astaxanthin content was observed. Recently, strains have been constructed that over-express a combination of two carotenogenic genes in order to further increase the astaxanthin (or other carotenoid) content. These genes are crtE, crtYB, crtI and ast. Comparison of carotenoid levels, especially intermediates, of wild and crt gene knock-out strains, showed no increase of these intermediates in the knock-out strains. This indicates that the end product astaxanthin (or its precursor b-carotene) does not exert a feed back regulatory effect on the carotenogenic enzymes.

È in corso la ricerca di dati su OpenAIRE...

Si è verificato un errore durante la ricerca dei dati su OpenAIRE

Nessun risultato disponibile