WO1992022648A1

WO1992022648A1 - A method of generating high carotenoid-producing microorganisms, microorganisms obtained by the method and a process for producing carotenoid-containing cells or cell parts or purified carotenoid

Info

Publication number: WO1992022648A1
Application number: PCT/DK1992/000186
Authority: WO
Inventors: Ingrid Stampe Villadsen
Original assignee: Individual
Current assignee: Individual
Priority date: 1991-06-14
Filing date: 1992-06-15
Publication date: 1992-12-23
Anticipated expiration: 1993-12-14
Also published as: DK115191D0; CA2111477A1; AU1985192A

Abstract

High carotenoid-producing, in particular astaxanthin-producing, microorganisms are generated by treating cells of a carotenoid-producing microorganism strain with a mutagen, culturing the mutagenized cells in the presence of a phytoene dehydrogenase inhibitor to reduce the level of carotenoid production of all the cells, and selecting a mutant cell or colony showing increased carotenoid coloring compared to the background. In particular, the carotenoid-producing microorganism is Phaffia rhodozyma, and the cells are grown to the late stationary phase before they are treated with the mutagen. The obtained microorganisms can be grown to obtain a biomass containing the carotenoid in an amount of at least 1600, preferably at least 1900 νg/g dry weight. When culturing a carotenoid-producing microorganism to produce carotenoid-containing microbial cells or purified carotenoid, particularly high yields of carotenoid can be obtained by growing the cells to the haploid phase before they are harvested or further treated. Still higher yields may be obtained by using a culture medium containing peroxisome-inducing substances.

Description

A method of generating high carotenoid-producing microorganisms, microorganisms obtained by the method and a process for producing carotenoid-containing cells or cell parts or purified carotenoid

This invention concerns a method of generating high carotenoid-producing, in particular astaxanthin-producing, microorganisms by treating cells of a carotenoid-producing microorganism strain with a mutagen, culturing the mutagenized cells and selecting a mutant colony showing increased content of carotenoid.

The invention also concerns the high carotenoid-producing, in particular astaxanthin-producing microorganisms generated by this method, and especially some deposited strains of the yeast Phaffia rhodozyma.

Further, the invention concerns a process for producing carotenoid-containing, in particular astaxanthin- containing, microbial cells or cell parts or purified carotenoid, in particular astaxanthin, by cultivation of the high carotenoid-producing microorganisms of the invention.

BACKGROUND OF THE INVENTION

Carotenoids

It is known that the red colour of the meat of anadromous fish such as salmon or sea trout is due to red pigments such as astaxanthin which is present in the feed consumed by the fish. In natural surroundings, the fish obtain their red colour from crustaceans or other astaxanthin- containing organisms, but when being bred in fish farms, the fish do not have access to these natural pigmentation sources and therefore do not obtain the attractive red colour unless red pigments are supplied in the feed.

Thus, astaxanthin isolated from crustacean wastes or produced synthetically as well as other synthetic red pigments such as cantaxanthin have been used as consti¬ tuents in fish feed. However, the use of cantaxanthin in animal feed is prohibited in certain countries, and the synthetic astaxanthin production as well as the process for isolating natural astaxanthin are rather expensive and often also subject to seasonal variations.

The carotenoids are a group of pigments, yellow to red in color, which are widely distributed in the plant and animal kingdoms. The color varies dependent on the lenghts of the chromophore and the type of oxygen-containing groups attached. Pigment formation is seldom in yeasts but is a characteristics of various species of the yeast genera Rhodotorula, Cryptococcus, Sporobolomyces and

Phaffia. Also certain bacteria, other fungi and unicellu¬ lar algae produce carotenoids.

The biosynthesis and the production of variuos carotenoids by different microorganisms has been rewieved recently

(Ref. 15, 16). The Rhodotorula rubra produces torulene and torularrhodin, Rhodotorula aurea and some species from the Mucorales produce 0-carotene, and the yeast Phaffia rhodo- zyma, the bacteria Mycobacterium lacticola and Brevibacte- rium 103, and the green alga Haematococcus pluvialis produce astaxanthin.

In prokaryotes, conserved enzyme reactions mediate the early reactions of carotenoid biosynthesis (Ref. 17) but whether the enzymes are conserved between eukaryotes and prokaryotes as well remain to be seen. However, the bio- synthesis seem to follow the same route in all organisms.

The carotenoids, like all terpenoids, are synthesized from hydroxymethylglutaryl-coenzyme A, which is first converted to mevalonic acid. The specific part of the pathway begins with the condensation of two molecules of geranylgeranyl pyrophosphate to form phytoene, a colourless carotene. Four dehydrogenations transform phytoene into lycopene, the pigment of red tomatoes. Two cyclisations convert lycopene to ^-carotene. In Phycomyces these reactions are carried out by an enzyme aggregate that contains four copies of a dehydrogenase and two copies of a cyclase. In Giberella fujikuroi the production of torulene requires a fifth dehydrogenation by the same enzyme; the final break and oxidation to neurosporaxanthin appear to be carried out in a single step. On the other hand, several dehydro- genases have been postulated for Ustilago violacea. The detection of some minor intermediates indicates that the order of the reactions is sometimes altered in these organisms, and may be different in others. Canthaxantin, a likely intermediate in the biosynthesis of astaxanthin in the crustaceans, was not detected in Phaffia rhodozyma.

An additional pathway leading to astaxanthin from 0- carotene was reported by Andrewes et al. (Ref. 7), who found the carotenoids of P. rhodozyma to be unusual, comprising echinenone, 3-hydroxyechinenone, phoenicoxan- thin and astaxanthin. Astaxanthin was found to be the major pigment synthesized by this yeast; in naturally occurring yeast it comprised approximately 85% of the carotenoid mixture.

Despite the common pathway for the biosynthesis of carotenoids, the regulation of the individual enzymes is not universal. Certain compounds stimulate carotenoid biosynthesis in some organisms, but inhibit it in other organisms .

The most important function of carotenoid pigments, is to protect the organisms against oxidative damage, e.g. the deleterious effect of singlet oxygen and free radicals (Ref. 14, 17). The protection of _>-carotene against radiation occurs by the absorption of energy in the blue region of the spectrum. Recent epidemiological and oncogenical studies suggest that normal and high levels of 0-carotene in the body may protect against cancer (Ref. 18, 19). 0-carotene also functions as a precursor of vitamin A in mammals.

The Genus Phaffia

The yeast Phaffia rhodozyma was found in Japan in 1967, in the exodeous fluids excreted from beach trees and named after H. Phaff who isolated this organism (Ref. 1). Phaffia rhodozyma was shown to be different from other pigmented yeasts in the ability to produce the carotenoid pigment, astaxanthin (3,3'-dihydroxy-0,0-carotene-4,4'- dione) and was, in addition, ascribed to a new genus (Ref. 6).

The genus, Phaffia, was placed in the Deuteromycotina described as having the unique characteristics of being a carotenoid-producing, fermentative yeast. Its ability to ferment sugars is in contrast to other carotenoid- synthesizing yeasts which are all strictly aerobic. Its special metabolic properties such as the ability to split urea by urease and its ability to assimiliate inositol are rare among ascomycetous yeasts and so is its ability to produce starchlike components. Another characteristic is the content of 48,3 mol-% G+C, which is the upper limit for ascomycetous yeasts, but lower than what is found in basidiomyceteous yeasts (Ref. 9). Phaffia reproduces by enteroblastic budding leading to a multilayered cell wall. The polysaccharides in the laminar cell wall contain a high proportion of α-l,3-glucan indicative of both ascomyceteous and basidiomyceteous yeasts (Ref. 8), but the nature of the cell wall has lead to the belief that Phaffia is an imperfect species of basidiomyceteous origin.

P. rhodozyma, the only known species in this genus, was hitherto thought to reproduce exclusively by budding, and attempts to identify a sexual cycle have up to this moment been unsuccesfull (Ref. 6).

Other characteristics of Phaffia include the absence of ballistoconidia, but the presence of chlamydospores. In connection with the present invention it has been determined that Phaffia also produces haploid spores of sexual origin.

STATE OF THE ART

In order to be able to use Phaffia rhodozyma as an effec¬ tive economical source af astaxanthin, strains that over¬ produce carotenoids are needed for commercial development.

From the international (PCT) patent application WO 88/08025 (Danisco Bioteknologi A/S, Denmark) it is known to produce a Phaffia rhodozyma yeast cell which, when grown on a specified medium in shake flask at 20-22 °C for 5 days, produces astaxanthin in an amount of at least 300 αg/g of yeast dry matter, by treating a wild type Phaffia rhodozyma culture with a mutagen and selecting a resulting mutant which, when grown under the above conditions, is capable of producing astaxanthin in an amount of at least 300 ng/g of yeast dry matter. Three mutant strains of Phaffia rhodozyma generated by this method have been deposited at the Centraalbureau voor Schimmelcultures in Baarn, Netherlands, and the best producing of these strains, CBS 215-88, is stated to produce 1340 μg total pigment/g yeast dry matter, of which 880 μg/g is astaxanthin, when cultured in shake flask under the specified conditions. After fed-batch fermenta¬ tion in a 4 m³ fermenter for 72 hours using corn steep solid and sucrose as carbohydrate sources, this strain contained 1360 μg total pigment/g yeast dry matter of which 1080 μg/g were astaxanthin.

Inhibitors of the redox pathway (Ref. 2 and the interna¬ tional (PCT) patent application WO 90/01552) and the end ring analog of 0-carotene, 0-ionone (Ref. 3) have been reported to be af great value in selecting yeasts with increased astaxanthin content; in both cases from wild type levels 150-350 μg/g dry weight up to 1000 μg/g dry weight.

The selection of mutant strains with carotenoid content up to about 1200 μg/g dry weight can be done by visual inspection.

However, with a carotenoid content of more than 1200 μg/g dry weight it is not possible by visual means to distin¬ guish between strains producing either 1200 or more μg carotenoid/g dry weight.

DESCRIPTION OF THE INVENTION

In order to obtain strains with an astaxanthin content higher than what has been obtained in the prior art, an inhibitor of carotenoid biosynthesis was sought which would reduce the color formation in a way permitting the selection of high carotenoid-producing strains due to changes in activities of enzymes of the biosynthetic pathway. Several inhibitors of the color production were tested for their ability to be applied as specific inhibitors of the astaxanthin biosynthesis. Among the compounds tested, 1000 ppm of piperonyl butoxide, 0,1 mM diphenylamine and 5 mM 3-amino-l,2,4-triazole, only diphenylamine proved partly successfull. Since both diphenylamine and thymol have been reported to inhibit the enzyme phytoene dehydrogenase in the fungus Phycomyces blakesleanus in the biosynthetic pathway to 0-carotene (Ref. 4), such phytoene dehydrogenase inhibitors are thought to be useful in selecting mutagenized strains with a high production of carotenoids. Also the herbicide flurtamone has been shown to interfere with phytoene dehydrogenase. Plating of astaxanthin-producing cells of Phaffia rhodozyma treated with the mutagen "ICR 170" or EMS on YM plates containing 0,1 mM thymol resulted in light yellow colonies, with a few out of approximately 10000, being more colored. These light orange to orange colonies were now selected visually, purified by repeated streaking on thymol-containing YM plates, three to six times, and were then determined to yield as high as 2000 μg/g dry weight in shake flasks and 3500 μg/g dry weight in fermenters.

Thus, one aspect of the present invention is a method of generating carotenoid-producing, in particular astaxanthin-producing, microorganisms with a carotenoid content of at least 1600 μg/g dry weight, preferably at least 1900 μg/g dry weight, when cultured in shake flask to the stationary phase, by treating cells of a carotenoid-producing microorganism strain with a mutagen, culturing the mutagenized cells, and selecting a mutant colony showing increased content of carotenoid, said method comprising the following steps: (a) treating the cells with the mutagen,

(b) culturing the mutagenized cells in the presence of a phytoene dehydrogenase inhibitor to reduce the level of carotenoid production of all the cells,

(c) selecting one or more cells or colonies showing increased carotenoid coloring compared to the back¬ ground,

(d) and, if necessary, purifying each selected cell or colony by repeated streaking on plates containing phytoene dehydrogenase inhibitor and selection of colonies showing increased carotenoid coloring.

Besides the visual selection of colonies showing increased carotenoid coloring it is also possible in the method according to the invention to select single highly colored cells by using quantitative flow cytometry/cell sorting as described in Ref. 23.

In contrast to what is described in the prior art, care was taken not to induce severe killing during mutageniza- tion, so killing curves were made in each single case in order to ensure a 50% survival and thus to avoid unwanted growth-inhibiting secondary mutations. Further, the mutagenization procedure was repeated using different mutagens. The rationale for selecting the mutagens was to change between two different types of mutagens; one type that induces transitions and thus a single base change or change of a single amino acid, (e.g. UV, EMS or hydroxylamine) followed by or alternating with the other type that induces frameshift mutations (e.g. "ICR 170") and thus causes gross alterations in the proteins (Ref. 20). In addition, hot spot mutations by frame shift mutagens are known to occur in regions of bent DNA, most often indicating regions of regulatory DNA sequences (Ref. 21).

In order to maximize the isolation of high astaxanthin- producing cells of Phaffia rhodozyma, a new selection scheme is suggested. All low carotenoid or astaxanthin containing cells were killed with a combination of a phytoene dehydrogenase inhibitor and a superoxide radical producer or with a combination of a phytoene dehydrogenase inhibitor and another inhibitor of the carotenoid biosynthetic pathway. This procedure permits only cells with a superhigh content of carotenoids to grow. It has been shown that a combination of 50 μM diphenylamine and 100 μM duroquinone kills more than 106 cells of the 0- carotene producing yeast Rhodotorula mucilaginosa (Ref. 5), but not that it can be used to select the one cell that overproduces 0-carotene.

In Phaffia the combination of 1 mM 0-ionone and 0,1 mM thymol kills more than 10^s cells. The use of 0-ionone alone kills more than 80% of the cells, whereas thymol does not exhibit any deathly effect on the cells. Incuba¬ tion of a mixture of white mutants of Phaffia and wild type astaxanthin containing cells in 0,1 mM 0-ionone results in a survival primarily of the astaxanthin containing cells.

The production of astaxanthin in high producing strains has been found to be greatly variable. Even after isola- ting single high-producing colonies, and after having purified them twice by streaking on agar plates, again selecting single high producing colonies, a high propor¬ tion of lighter colonies appeared. The high producing strains deposited in connection with this invention there- fore have been streaked on agar plates as many as six times. These strains are now stable and exhibit only 3% lighter colonies on plates containing thymol. In contrast to this, a strain, DBT 415, generated according to WO 88/08025, which was obtained from Danisco Bioteknologi A/S and used as the starting strain in the following Example 3, is grossly unstable exhibiting about 30% lighter colonies on plates containing thymol. Thus, inhibitors of phytoene dehydrogenase are also a potent means of isolating stable strains for commercial production of astaxanthin.

The reason for the unstability remains unknown; many factors, some of which are discovered in connection with this invention may influence the stability. One is that mutagenizations have been performed in the exponential growth phase, where the cells are diploid, and therefore it is difficult to isolate mutants at all - another is that the astaxanthin biosynthetic enzymes may reside in a plasmid, unstable by nature, or in the genome of the mitochondria, the number of copies of which vary with the growth phase, or that the astaxanthin in the cells are bound to mitochondria or to peroxisomes, and therefore depends on their synthesis for its location.

Identification of a sexual life cycle of Phaffia rhodozyma

When Phaffia is grown in shake flasks containing conven¬ tional YM medium at 21 °C, two to three types of cells can be distinguished, depending on the stage of growth. In the early exponential growth phase, cells are reproduced by the formation of buds with a relatively broad basis, like the cells seen in the description of the genus (Ref. 9).

In the stationary phase, however, small cells, presumably the previously mentioned chlamydospores, are the dominant type, constituting 50-80% of the total amount of cells. When the culture was examined at the breakpoint between exponential and stationary growth phase (OD _cn--. = 12-17), a few % of asci-like cells, each containing 2-4 round bodies, were observed. The conventional spore stain, methyl green, used to stain the sexual spores, the asci, of the yeast Saccharomyces cerevisiae, where the ascus spores turn green while vegetative cells are stained red, were also used to distinguish Phaffia spores. Here also, the big budding cells from the early exponential growth phase were colored primarily red, whereas the small cells appearing in the stationary phase were colored green, indicating the presence of sexual spores in the stationary growth phase.

The sizes of the two types of cells were: Big cells 10.7 X 10 μm and the small cells 7.4-5.0 X 4.1 μm, corresponding to volumes of 460 μm³ and 98-66 μm³, respectively.

For the final demonstration of a sexual life cycle in Phaffia a defined complete medium that contains urea and not ammonia as the easily accessible nitrogen source was used. When the cells start being depleted for nitrogen (urea) around a cell density of 2 x 10 cells per ml, conjugation will occur followed by a sporulation (Example 4). In ammonia-containing medium sporulation is less prominent, mainly diploid cells are observed. Another requirement for sporulation is the presence of all amino acids and the bases uracil and adenine. Without these latter bases the cells may conjugate, but not sporulate.

The final proof of a sexual cycle came from an analysis on a SKATRON flow cytometer on populations of cells fixed in different stages af the whole life cycle or growth deve¬ lopment of Phaffia. The cells were stained with DAPI, a conventional stain for DNA (Ref. 10) and the cell size (rather the scatter of individual cells) as well as the content of DNA was determined in each single cell. Due to the very thick cell wall found in Phaffia 10 x normal amounts of DAPI were used.

Fig. 1 demonstrates the analysis of a sample taken in the early stationary growth phase. Some cells with a DNA content of N, as well as cells with a DNA content of 4N can be seen. Treating the cells with mutanase, as demonstrated in the following, to open the cells, resulted in the appearence of more cells with a content of N, together with the appearence of a new type of cells detected with the forward scatter.

Thus, the small cells are contained within the big cells, they have a DNA content of 4N and must be haploid, the big cells are diploid (2N) and have replicated the DNA in their genome to 4N before they undergo a meiosis and each cell then gives rise to four external spores, each with a haploid DNA content. I have found that the content of astaxanthin in Phaffia is highest in the late stationary phase, and so these haploid cells must contain the pigment.

In conclusion, the genus Phaffia rhodozyma most obviously belongs to the basidiomyceteous yeasts, due to the appearence of external spores liberated from sexual cells undergoing meiosis, in accordance with previous beliefs.

As mutagenization of cells containing only one genome is far more effective than mutagenization of diploid cells with two genomes, it is preferable in the method of the invention that the cells are grown to the haploid phase before they are treated with the mutagen. As demonstrated above, for Phaffia rhodozyma this haploid phase occurs in the late stationary phase of growth in defined medium. Likewise, the production of astaxanthin in Phaffia rhodozyma is dependent on the sporulation. The contents of carotenoids when growing Phaffia rhodozyma DSM 6559 under the conditions stated in the following Example 4, measured as described in Ref. 25, was 360 μg/g dry weight in the whole exponential growth phase. In the early stationary phase and during the sporulation itself the content of carotenoids increased to 666 μg/g dry weight. Two days later the haploid cells contained 1100 μg/g dry weight, and two days later again (totalling 7 days of growth) 2100 μg/g dry weight. Further, by stimulation with 0.5 % DL- mevalonic acid lactone the production of carotenoids was increased to 2800 μg/g dry weight. In contrast, when grown in YM medium the content of carotenoids maximally increased to 1200 μg/g dry weight, and this could not be stimulated with DL-mevalonic acid lactone. Alle experiments were performed in conventional shake flasks.

Thus, as the content of asthaxanthin in Phaffia rhodozyma is highest in the late stationary phase, it is preferable, when producing astaxanthin-containing cells of Phaffia rhodozyma by the process of the invention, to grow the cells to the later stationary phase before harvesting.

Subcellular location of astaxanthin in Phaffia rhodozyma

A method of removing the thick cell wall of Phaffia rhodozyma has been found and used to identify the particulate nature of astaxanthin-containing structures in the cell.

In order to perform several experiments all widely used in genetics and molecular biology, like e.g. the recovery of subcellular components like nuclei, identification of the chromosome sizes, protoplasting, and isolation of possible plasmids, a method to remove the thick cell wall of Phaffia is needed.

Several enzymes that digest the cell wall were analyzed for their ability to digest the cell wall of Phaffia rhodozyma. The enzyme zymolyase, widely used to degrade the cell wall in Saccharomyces cerevisiae did not attack the cell wall of Phaffia; but the enzyme mutanase, "NOVOZYME 234", was able to make wholes in the cell wall. The lytic system worked most efficiently at pH 5-6 and at 30 °C. The cell wall of 10⁶ cells of Phaffia could be degraded in 30 min, whereas the previous method using another enzyme took 24 h at 37 °C. The efficiency of the enzyme can be magnified, and the digestion time shortened considerably, when the enzyme is applied together with the redox agent thioglycol.

Mutanase was used to prove the existence of a sexual cycle as well as to fractionate the subcellular components of Phaffia in order to determine the subcellular location of the pigment astaxanthin.

The method used for purification of nuclei (Ref. 12) was used with several modifications: Gram amounts of cells were used, the cells were broken with mutanase, (50 g of cells and 2.3 g of mutanase), and the protoplasts were disintegrated with a Potter-Elvehjelm homogenisator at maximum speed for 3 min. No absolute purification of the astaxanthin-containing subcellular particles were obtained using this method, but an immense enrichment of the particles containing the pigment was obtained in the supernatant after the precipitation of the nuclei. The fraction heavily enriched for astaxanthin contained raspberrylike structures, that could be pelleted easily, thus establishing the particulate nature of astaxanthin- containing structures in Phaffia. These structures do not resemble mitochondria or nuclei. However, they appear be peroxisomes. Thus it is important to consider fermentations on peroxisome-inducing medium (Ref. 13), as the sole carbon source when using Phaffia rhodozyma for a commercial production of astaxanthin. The peroxisome-inducing substance for use in such medium may for example be selected from the group consisting of unsaturated fatty acids having at least 14 carbon atoms, D-alanine, alkylamines and mixtures thereof, optionelly in combination with lower alcohols.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a graph showing the content of DNA in cells of Phaffia rhodozyma at 0D.,_₀ 18.5 after 4 days of growth in YM medium. DNA was stained as in Ref. 10 except that 10 μg of DAPI/ml was used. N indicates the genomic DNA content (haploid), whereas 4N indicates the presence of cells with 4 x genomic DNA content.

Figure 2 is a graph illustrating the growth of Phaffia rhodozyma in a minimal salt medium with glucose as the carbon source and urea as the nitrogen source. The optical density was measured in a 1 cm cuvette using a Zeiss PMQII spectrophotometer.

Figure 3 shows photos of: (a) Two haploid cells having conjugated and budded to form af bigger diploid cell, (b) One cell containing (c) three nuclei when stained with DAPI. Magnifications, a: 1400x; b,c: lOOOx.

Figure 4 shows photos of diploid cells in sporulation: (a) A diploid cell simultaneously producing two haploid spores, (b) A diploid cell producing four haploid spores. Magnification, 3500x. Figure 5 is a schematic illustration of the sexual life cycle of Phaffia rhodozyma. In the early exponential growth phase the cells are budded and haploid (N). Cells of opposite mating types make contact and conjugate to form diploid cells (2N). The conjugated pairs form diploid buds (2N) that eventually sporulate to form four haploid external spores (N).

EXAMPLES

In Examples 1-3 below, the following general procedure was followed:

EMS mutagenesis: The strain to be mutagenized was grown in 50 ml of YM medium, in shake flasks for two days. From the exponential growing culture, around 3 X 10^β cells were harvested by centrifugation 5000 RPM for 5 min. The pellet was resuspended in 5 ml 10 mM sodium phosphate pH 7.0, 30 μl of EMS was added, and the cells were incubated for 10- 30 minutes at room temperature while shaking. The reaction was stopped by adding 25 ml 1% NaCl. The cells were washed twice in the above buffer and resuspended in YM medium. Further selection was made as described below.

"ICR 170" mutagenesis: The strain to be mutagenized was grown in 50 ml of YM medium, in shake flasks for two days. From the exponentially growing culture, around 3 X 10' cells were harvested by centrifugation 5000 RPM for 5 min. The pellet was resuspended in 5 ml freshly prepared solution of 100 μg/ml "ICT 170" in 0.1 M sodium phosphate pH 7.0, and the cells were incubated for one hour at room temperature while shaking. The reaction was stopped by adding 25 ml of 1% NaCl, the cells were washed twice in the above buffer and resuspended in YM medium. Further selection was made as described below. For survival measurements treated as well as untreated cells were diluted and plated on YM plates to ensure a survival of about 50% in each case.

For selection: 1: Visual screening: Treated cells were diluted and spread on about 40 YM plates, ensuring about 100-300 colonies per plate. 2: Inhibitor screening: Treated cells were diluted and spread on about 40 YM plates containing 0.1 mM thymol, ensuring about 100-300 colonies per plate. Colonies that were more orange than background were selected, and purified further by streaking up to six times on the same plates.

EXAMPLE 1

Using Phaffia rhodozyma strain CBS 5908 and employing EMS treatment to a survival of about 50%, the content of carotenoids was increased from 350 μg/g to 1529 μg/g dry weight by visual screening, obtaining a strain termed IVP 104. This strain was treated twice with "ICR 170" and a strain termed IVP 106 having 1710 μg/g dry weight was selected. The strain IVP 106 was finally treated again with EMS, the mutagenized cells were plated on thymol- containing plates, and a strain termed IVP 107 was selected containing 1952 μg/g dry weight when cultured in shake flasks. IVP 107 is deposited in Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Mascheroder Weg IB, D-3300 Braunschweig, Germany, under the accession number: DSM 6560.

EMS

"ICR 170" "ICR 170"

l EMS, then thymol

EXAMPLE 2

The above-mentioned strain IVP 104 was mutagenized with "ICR 170" as above, but higher producing strains were selected directly on thymol-containing plates. As a result a strain termed IVP 112 containing 2050 μg carotenoids/g dry weight was obtained. IVP 112 is deposited in Deutsche Sammlung von Mikroorganismen und Zellkulturen under the accession number: DSM 6561.

. "ICR 170", then thymol

EXAMPLE 3

The strain DBT 415 obtained from Danisco Bioteknologi A/S, Raffinaderivej, 2300 Copenhagen S, Denmark, and origina¬ ting from Phaffia rhodozyma strain ATCC 24261 was mutage- nized with "ICR 170", and higher producing colonies were selected on thymol-containing plates. A strain termed IVP 14 was selected as containing 1962 μg/g carotenoids dry weight. IVP 14 is deposited in Deutsche Sammlung von Mikroorganismen und Zellkulturen under the accession number: DSM 6559.

* "ICR 170", then thymol

EXAMPLE 4

The sexual cycle in Phaffia occurs almost synchronously when the cells are grown in a defined medium supplemented with amino acids and bases, with glucose as the C-source and urea as the N-source.

Defined complete medium for growth of Phaffia rhodozyma (DCMP): One liter of medium contains the following components:

0.1 g urea

20 g glucose

0.875 g H₂P0₄

0.125 g K₂HP0₄

0.5 g MgS0₄.7H₂0 0.1 g NaCl

0.05 g CaCl₂.6H₂0

70 μg ZnSO. . 7H₂0

40 μg thiamine 15 μg biotin ©

40 μg pyridoxine 40 μg pantothenic acid 20 μg riboflavin 40 μg niacin

20 mg adenine 20 mg alanine 20 mg arginine 100 mg aspartic acid 20 mg cysteine

100 mg glutamic acid 10 mg guanine 20 mg histidine 20 mg inositol 30 mg isoleucine 60 mg leucine 30 mg lysine 20 mg m thionine 20 mg para-aminobenzoic acid 50 mg phenylalanine 20 mg proline 375 mg serine 200 mg threonine 20 mg tryptophan 30 mg tyrosine 40 mg uracil 150 mg valine

A small inoculum of cells of DSM strain 6559 grown in a defined medium is inoculated into 50 ml of the defined complete medium described above in a 250 ml shake flask so that a minium of 6 generations of growth is allowed before

OD 4 _Λ5_cn0 nm reaches 0,1 (measured in a Zeiss PMQII spectrophotometer). The cells are grown at 21 °C while shaking at 170 RPM. In exponentially growing cultures till a cell concentration of around 2 x 10 cells single cells budding in the typical manner described for Phaffia are seen. From OD 0.5 to 1.0 the cells pair, only to conjugate later. After fusion a larger diploid bud appears around OD 1.5 (FIG. 3a). In stationary phase occuring from OD 3 to 6 the diploid cells sporulate to excrete four external haploid spores mainly formed at one end of the cell (FIG. 4). After OD 6 the culture contains predominantly single haploid cells. Staining the DNA in a sporulating culture with DAPI revealed the occurrence of three or four nuclei in one cell seen in FIG. 3b,c, one nucleus here being present in the bud.

The whole sexual life cycle is schematized in Fig. 5.

REFERENCES

1. Phaff, H.J., et al.: "A comparative study of the yeast flora associated with trees on the Japanese islands and on the west coast of North America", pp. 759-774 in Proc. IV IFS Fermentation Technology Today, Kyoto, Society of Fermentation Technology Osaka (1972).

2. Gil-Hwan An et al.. Applied and Environmental Microbiology, Jan. 1989, pp. 116-124.

3. Lewis, M.J. et al.. Applied and Environmental Microbiology, Sept. 1990, pp. 2944-2945.

4. Bejarano, E.R. , and Cerda-Olmedo, E., Phytochemistry, Vol. 23, No. 6, pp. 1623-1626 (1989).

5. Moore, M.M., et al., Archives of Biochemistry and Biophysics, Vol. 270, No. 2, May 1, pp. 419-431 (1989).

6. Miller, M.W., et al., International Journal of Systematic Bacteriology, Apr. 1976, pp. 286-291.

7. Andrewes, A.G., et al., Phytochemistry, Vol. 15, pp. 1003-1007 (1976).

8. The Yeasts. Biology of Yeast. Vol I, Second Edition, Academic Press 1987, Eds. A.H. Rose, J.S. Harrison. Chap. 2. Classification of Yeasts, pp. 5-61, N.J.W. Kreger-van Ri .

9. Yeasts. Characterization and Identification. Bornatt, Payne & Yarrow, Cambridge Univ. Press 1983. 10. Williamson, D.H., and Fennell, D.J., Visualization of Yeast mitochondrial DNA with the fluorescent stain "DAPI". Methods Enzymol. 56, pp. 728-733 (1978).

11. Molecular Biology of the Gene, Vol. I, 4th Edition (1987), The Benjamin/Cummings Publishing Company, Inc., Chapter 18, "Yeasts as the E. coli of Eucaryotic Cells", pp. 550-594.

12. Methods in Cell Biology, Vol. XII, Acad. Press 1975, Chapter 6, John H. Duffus: "The Isolation of Yeast Nuclei and Methods to Study Their Properties".

13. The Yeasts. Yeast Organelles. Vol. IV, Second Edition, M. Veenhuis and W. Harder: Microbodies, pp. 610-653. Academic Press 1991, Eds: A.H. Rose and J.S. Harrison.

14. Ami Ben-Amotz and Mordhay Avron, Trends in Biotechnology, Vol. 8, No. 5, May 1990, pp. 121-125.

15. Armstrong, G.A., et al., Proc. Natl. Acad. Sci, USA, Vol. 87, pp. 9975-9979. December 1990.

16. Biotechnology of Vitamins, Pigments and Growth

Factors, Ed. Vandamme, E.J., Elsevier Appl. Biotech. Series 1989.

17. Kurashige, M., et al., Physiol. Chem. Phys. & Med. NMR. (1990) 22, pp. 27-38.

18. Ziegler, R.G. (1989), J. Nutr. 119, pp. 116-122.

19. Bendich, A., and Olson, J.A., The FASEB Journal, 1989, 3, pp. 1927-1932. 20. Sherman, F., et al.: Laboratory Course Manual for Methods in Yeast Genetics, Cold Spring Harbor Laboratory 1986.

21. Goodenough, U., Genetics, Third Edition, Holt Saunders Japan (1984).

22. Hampsey, D.M., et al., Molecular and Cellular Biology, Dec. 1986, pp. 4425-4432.

23. Sandmann, G. , et al., Plant Physiol. (1990) 9_4, pp. 476-478.

24. Gil-Hwan An et al. , Biotechnology, Vol. 9, Jan. 1991, PP. 70-73.

25. Sedmak, J.J., Weerasinghe, D.K. and Jolly, S.C.: Extraction and Quantitation of Astaxanthin from Phaffia rhodozyma. Biotechnology Techniques Vol. 4, No. 2, pp. 107-112 (1990); received as revised 24th January.

Claims

P a t e n t C l a i m s :

1. A method of generating carotenoid-producing, in particular astaxanthin-producing, microorganisms with a carotenoid content of at least 1600 μg/g dry weight, preferably at least 1900 μg/g dry weight, when cultured in shake flask to the stationary phase, by treating cells of a carotenoid-producing microorganism strain with a mutagen, culturing the mutagenized cells, and selecting a mutant colony showing increased content of carotenoid, said method comprising the following steps:

(a) treating the cells with the mutagen,

2. A method according to claim 1, wherein the carotenoid- producing microorganism is selected from the group con¬ sisting of unicellular algae, fungi of the Mucorales order and of the genus Neurospora, yeasts of the genera Rhodo¬ torula, Cryptococcus, Sporobolomyces and Phaffia, and bacteria of the genera Mycobacterium and Brevibacterium.

3. A method according to claim 1 or 2, wherein the carotenoid-producing microorganism is mainly producing astaxanthin.

4. A method according to any of the preceding claims, wherein the carotenoid-producing microorganism is a yeast of the species Phaffia rhodozyma, a bacterium of the species Mycobacterium lacticola or Brevibacterium 103 or a green alga of the species Haematococcus pluvialis.

5. A method according to any of the preceding claims, wherein the cells of the carotenoid-producing micro¬ organism are grown to the haploid phase before they are treated with the mutagen.

6. A method according to claim 5, wherein the carotenoid- producing microorganism is Phaffia rhodozyma, and the cells are grown to the late stationary phase before they are treated with the mutagen.

7. A method according to any of the preceding claims, wherein the mutagen is selected from the group consisting of two different types of mutagens, one that induces transitions such as a single base change, and another that induces frameshift mutations.

8. A method according to any of the preceding claims, wherein the cells are treated with the mutagen to a survival rate of 1 to 90%, preferably 10 to 60%, most preferably 40-50%.

9. A method according to any of the preceding claims wherein the phytoene dehydrogenase inhibitor is diphenyl¬ amine, flurtamone or thymol.

10. A method according to any of the preceding claims, wherein a substance toxic to cells of low carotenoid content is used together with the phytoene dehydrogenase inhibitor.

11. A method according to claim 10, wherein the substance toxic to cells of low carotenoid content is selected from the group consisting of superoxide radical generators and alcohols.

12. A method according to claim 11, wherein the substance toxic to cells of low carotenoid content is duroquinone (tetramethyl-p-benzoquinone).

13. A method according to claim 10, wherein the substance toxic to cells of low carotenoid content is another inhibitor of the carotenoid biosynthetic pathway.

14. A method according to claim 13, wherein said inhibitor is selected from the group consisting of nicotine, 2-(4-chlorophenylthio)triethylamine (CPTA), imidazole, geraniol, veratrole and 0-ring containing carotenoids such as retinol and 0-ionone.

15. A method according to any of the preceding claims, wherein steps (a) to (c) and, optionally, step (d) are repeated one or more times, each time on the selected cell or colony from the previous cycle.

16. A method according to claim 15, wherein the carotenoid-producing microorganism is a strain of Phaffia rhodozyma, and at least two different types of mutagen are used in the different cycles.

17. A carotenoid-producing, in particular astaxanthin- producing, microorganism with a carotenoid content of at least 1600 μg/g dry weight, preferably at least 1900 μg/g dry weight, when cultured in shake flask to the stationary phase, generated by the method according to any of the preceding claims.

18. A microorganism according to claim 17 which is selected from the group consisting of the yeast strain deposited under the accession No. DSM 6559, the yeast strain deposited under the accession No. DSM 6560, and the yeast strain deposited under the accession No. DSM 6561.

19. A process for producing carotenoid-containing, in particular astaxanthin-containing, microbial cells or cell parts or purified carotenoid, in particular astaxanthin, by cultivation of a carotenoid-producing, in particular astaxanthin-producing, microorganism under aerobic conditions in a medium containing assimilable sources of carbon, nitrogen and inorganic salts as well as biotin or desthiobiotin at a temperature in the range of 15-26 °C, optionally followed by one or several of the following steps in arbitrary sequence:

- harvesting the cells from the culture,

- opening the cells and isolating the cell parts, homogenizing the cells to obtain a homogenate, - extracting the carotenoid from the cells, the cell parts or the homogenate, c h a r a c t e r i z e d by using a carotenoid-producing microorganism according to claim 17 or 18 and growing the cells thereof to obtain a biomass containing the carotenoid in an amount of at least 1600 μg/g dry weight.

20. A process for producing carotenoid-containing, in particular astaxanthin-containing, microbial cells or cell parts or purified carotenoid, in particular astaxanthin, by cultivation of a carotenoid-producing, in particular astaxanthin-producing, microorganism under aerobic conditions in a medium containing assimilable sources of carbon, nitrogen and inorganic salts as well as biotin or desthiobiotin at a temperature in the range of 15-26 °C, optionally followed by one or several of the following steps in arbitrary sequence:

- harvesting the cells from the culture,

- opening the cells and isolating the cell parts, homogenizing the cells to obtain a homogenate,

- extracting the carotenoid from the cells, the cell parts or the homogenate, c h a r a c t e r i z e d by growing the cells of the carotenoid-producing microorganism to the haploid phase before they are harvested or further treated.

21. A process according to claims 19 or 20, wherein the carotenoid-producing microorganism is a strain of Phaffia rhodozyma, and the cells thereof are grown to the late stationary phase.

22. A process for producing carotenoid-containing, in particular astaxanthin-containing, microbial cells or cell parts or purified carotenoid, in particular astaxanthin, by cultivation of a carotenoid-producing, in particular astaxanthin-producing, microorganism under aerobic conditions in a medium containing assimilable, sources of carbon, nitrogen and inorganic salts as well as biotin or desthiobiotin at a temperature in the range of 15-26 °C, optionally followed by one or several of the following steps in arbitrary sequence: - harvesting the cells from the culture, opening the cells and isolating the cell parts, homogenizing the cells to obtain a homogenate, extracting the carotenoid from the cells, the cell parts or the homogenate, c h a r a c t e r i z e d in that the medium further contains peroxisome-inducing substances.

23. A proces according to claim 22, wherein the peroxisome-inducing substance is selected from the group consisting of unsaturated fatty acids having at least 14 carbon atoms, D-alanine, alkylamines and mixtures thereof, optionally in combination with lower alcohols.