Recombinant materials for carotenoid production

TECHNICAL FIELD

[0001] The present invention relates to methods for transforming Phaffia yeast, transformed Phaffia strains, as well as recombinant DNA for use therein.

BACKGROUND OF THE INVENTION

[0002] Methods for transforming the yeast Phaffia rhodozyma have been disclosed in European patent application 0 590 707 A1. These methods involve incubation of protoplasts with DNA or incubation of Phaffia cells with DNA followed by lithium acetate treatment. The recombinant DNA used to transform Phaffia strains with either of these methods comprised a Phaffia actin gene promoter to drive expression of the selectable marker genes coding for resistance against G418 or phleomycin. The methods involve long PEG and lithium acetate incubation times and transformation frequencies are low. When protoplasts are used, the transformation frequency is dependent on the quality of the protoplast suspension, making the procedure less reliable.

[0003] Recently a method for transforming Phaffia strains has been reported by Adrio J. L. and Veiga M. (July 1995, Biotechnology Techniques Vol. 9, No. 7, pp. 509-512). With this method the transformation frequencies are in the range of 3 to 13 transformants per μg DNA, which is low. A further disadvantage of the method disclosed by these authors consists in increased doubling time of the transformed cells. The authors hypothesised that this may be due to interference of the autonomously replicating vector with chromosome replication.

[0004] Clearly, there is still a need for a reliable and efficient method of transforming Phaffia strains with foreign DNA. It is an objective of the present invention to provide methods and means to achieve this. It is a further objective of the invention to optimize expression of certain genes in Phaffia rhodozyma in order to make Phaffia a more suitable production host for certain valuable compounds.

SUMMARY OF THE INVENTION

[0005] The invention provides a method for obtaining a transformed Phaffia strain, comprising the steps of contacting cells or protoplasts of a Phaffia strain with recombinant DNA under conditions conducive to uptake thereof, said recombinant DNA comprising a transcription promoter and a downstream sequence to be expressed which is heterologous to said transcription promoter, in operable linkage therewith, identifying Phaffia rhodozyma cells or protoplasts having obtained the said recombinant DNA in expressible form, wherein the transcription promoter comprises a region that is found upstream of the open reading frame of a highly expressed Phaffia gene. According to a preferred embodiment of the invention said highly expressed Phaffia gene is a glycolytic pathway gene, more preferably the glycolytic pathway gene is coding for Glyceraldehyde-3-Phosphate Dehydrogenase (GAPDH). According to one aspect of the invention, said heterologous downstream sequence comprises an open reading frame coding for resistance against a selective agent, such as G418 or phleomycin.

[0006] Another preferred method according to the invention is one, wherein said recombinant DNA comprises further a transcription terminator downstream from the said DNA to be expressed, in operable linkage therewith, which transcription terminator comprises a region found downstream of the open reading frame of a Phaffia gene. It is still further preferred, that the recombinant DNA is in the form of linear DNA.

[0007] Another preferred embodiment comprises, in addition to the steps above, the step of providing an electropulse after contacting of Phaffia cells or protoplasts with DNA.

[0008] According to another embodiment the invention provides a transformed Phaffia strain capable of high-level expression of a heterologous DNA sequence, which strain is obtainable by a method according to the invention. Preferably, said Phaffia strain contains at least 10 copies of the said recombinant DNA integrated into its genome, such as a chromosome, particularly in the ribosomal DNA locus of said chromosome.

[0009] The invention also provides recombinant DNA comprising a transcription promoter and a heterologous downstream sequence to be expressed, in operable linkage therewith, wherein the transcription promoter comprises a region found upstream of the open reading frame of a highly expressed Phaffia gene, preferably a glycolytic pathway gene, more preferably a gene coding for Glyceraldehyde-3-Phosphate Dehydrogenase.

[0010] Also provided is recombinant DNA according to the invention, wherein the heterologous downstream sequence comprises an open reading frame coding for reduced sensitivity against a selective agent, preferably G418 or phleomycin. Said recombinant DNA preferably comprises further a transcription terminator downstream from the said heterologous DNA sequence to be expressed, in operable linkage therewith.

[0011] Further aspects of the invention concern a microorganism harbouring recombinant DNA according to the invention, preferably Phaffia strains, more preferably Phaffia rhodozyma strains, as well as cultures thereof.

[0012] According to still other preferred embodiments isolated DNA fragments are provided comprising a Phaffia GAPDH-gene, or a fragment thereof, as well as the use of such a fragment for making a recombinant DNA construct. According to one embodiment of this aspect said fragment is a regulatory region located upstream or downstream of the open reading frame coding for GAPDH, and it is used in conjunction with a heterologous sequence to be expressed under the control thereof.

[0013] The invention according to yet another aspect, provides a method for producing a protein or a pigment by culturing a Phaffia strain under conditions conducive to the production of said protein or pigment, wherein the Phaffia strain is a transformed Phaffia strain according to the invention.

[0014] According to another aspect of the invention, a method for obtaining a transformed Phaffia strain, comprising the steps of

[0015] contacting cells or protoplasts of a Phaffia strain with recombinant DNA under conditions conducive to uptake thereof,

[0016] said recombinant DNA comprising a transcription promoter and a downstream sequence to be expressed in operable linkage therewith,

[0017] identifying Phaffia rhodozyma cells or protoplasts having obtained the said recombinant DNA in expressible form,

[0018] wherein the downstream sequence to be expressed comprises an isolated DNA sequence coding for an enzyme involved in the carotenoid biosynthetic pathway of Phaffia rhodozyma. Preferably, said enzyme has an activity selected from geranylgeranyl pyrophosphate synthase (crtE), phytoene synthase (crtB), phytoene desaturase (crtI) and lycopene cyclase (crtY), more preferably an enzyme having an amino acid sequence selected from the one represented by SEQIDNO: 13, SEQIDNO: 15, SEQIDNO: 17 and SEQIDNO: 19. According to a further embodiment, the transcription promoter is heterologous to said isolated DNA sequence, such as a glycolytic pathway gene in Phaffia. Especially preferred according to this embodiment is the Glyceraldehyde-3-Phosphate Dehydrogenase (GAPDH) gene promoter.

[0019] Also provided is a transformed Phaffia strain obtainable by a method according to the invention and capable of expressing, preferably over-expressing the DNA sequence encoding an enzyme involved in the carotenoid biosynthesis pathway gene.

[0020] The invention is also embodied in recombinant DNA comprising an isolated DNA sequence according to the invention, preferably in the form of a vector.

[0021] Also claimed is the use of such a vector to transform a host, such as a Phaffia strain.

[0022] A host obtainable by transformation, optionally of an ancestor, using a method according to any one of claims 1 to 5, wherein said host is preferably capable of over-expressing DNA according to the invention.

[0023] According to a further embodiment a method is provided for expressing an enzyme involved in the carotenoid biosynthesis pathway, by culturing a host according to the invention under conditions conducive to the production of said enzyme. Also provided is a method for producing a carotenoid by cultivating a host according to the invention under conditions conducive to the production of carotenoid.

[0024] The following figures further illustrate the invention.

DESCRIPTION OF THE FIGURES

[0025]
FIG. 1. Mapping of the restriction sites around the Phaffia rhodozyma GAPDH gene. Ethidium bromide stained 0.8% agarose gel (A) and Southern blot of chromosomal DNA (B) and cosmid pPRGDHcos1 (C) digested with several restriction enzymes and hybridized with the 300-bp PCR fragment of the Phaffia rhodozyma GAPDH gene. Lane 1, DNA×KpnI; 2, ×PstI; 3, ×SmaI; 4, ×SphI; L, lambda DNA digested with BstEII; 5, ×SstI; 6, ×XBaI and 7, ×XhoI.

[0026] The blot was hybridized in 6×SSC, 5× Denhardt's, 0.1% SDS, 100 ng/ml herring sperm DNA at 65° C. and washed with 0.1×SSC/0.1% SDS at 65° C. Exposure time of the autoradiogram was 16 h for the cosmid and 48 h from the blot containing the chromosomal DNA.

[0027]
FIG. 2. The organisation of two subclones; pPRGDH3 and derivative (A) and pPRGDH6 and derivatives (B) containing (a part of) the GAPDH gene of Phaffia rhodozyma. The PCR probe is indicated by a solid box. The direction and extent of the sequence determination is indicated by arrows.

[0028] solid boxes: GAPDH coding sequence

[0029] open box: 5′ upstream and promoter region of GAPDH

[0030] open box: 3′ non-coding Phaffia rhodozyma GAPDH sequence

[0031] solid line: GAPDH intron

[0032] hatched box: Poly-linker containing sites for different restriction enzymes

[0033] dotted line: deleted fragments

[0034]
FIG. 3. Cloning diagram of Phaffia transformation vector; pPR2.

[0035] solid box: 5′ upstream and promoter sequence of GAPDH

[0036] hatched box: G418

[0037] solid line: pUC19

[0038] open box: ribosomal DNA of Phaffia rhodozyma

[0039] Only restriction sites used for cloning are indicated.

[0040]
FIG. 4. Construction of pPR2T from pPR2T.

[0041] Solid box (BamHI-HindIII fragment): GAPDH transcription terminator from Phaffia. All other boxes and lines are as in FIG. 3. Only relevant details have been depicted.

[0042]
FIG. 5. Detailed physical map of pGB-Ph9. bps=basepairs; rDNA ribosomal DNA locus of Phaffia; act.pro 2=actin transcription promoter; act.15′ non-translated and aminoterminal region of the open reading frame; NON COD.=non-coding region downstream of G418-gene;

[0043]
FIG. 6. Detailed physical map of pPR2. GPDHpro=GAPDH transcription promoter region from Phaffia. Other acronyms as in FIG. 5.

[0044]
FIG. 7. Detailed physical map of pPR2T. Tgdh=GAPDH transcription terminator of Phaffia. All other acronyms as in FIGS. 5 and 6.

[0045]
FIG. 8. Overview of the carotenoid biosynthetic pathway of Erwinia uredovora.

[0046]
FIG. 9. Representation of cDNA fragments and a restriction enzyme map of the plasmids pPRcrtE (A); pPRcrtB (B), pPRcrtI (C) and pPRcrtY (B).

DETAILED DESCRIPTION OF THE INVENTION

[0047] The invention provides in generalised terms a method for obtaining a transformed Phaffia strain, comprising the steps of

[0048] contacting cells or protoplasts of a Phaffia strain with recombinant DNA under conditions conducive to uptake thereof,

[0049] said recombinant DNA comprising a transcription promoter and a downstream sequence to be expressed which is heterologous to said transcription promoter, in operable linkage therewith,

[0050] identifying Phaffia rhodozyma cells or protoplasts having obtained the said recombinant DNA in expressible form,

[0051] wherein the transcription promoter comprises, a region that is found upstream of the open reading frame of a highly expressed Phaffia gene.

[0052] In order to illustrate the various ways of practicing the invention, some embodiments will be high-lighted and the meaning or scope of certain phrases will be elucidated.

[0053] The meaning of the expression recombinant DNA is well known in the art of genetic modification, meaning that a DNA molecule is provided, single or double stranded, either linear or circular, nicked or otherwise, characterised by the joining of at least two fragments of different origin. Such joining is usually, but not necessarily done in vitro. Thus, within the ambit of the claim are molecules which comprise DNA from different organisms or different genes of the same organism, or even different regions of the same gene, provided the regions are not adjacent in nature. The recombinant DNA according to the invention is characterised by a transcription promoter found upstream of an open reading frame of a highly expressed Phaffia gene, fused to a heterologous DNA sequence. With heterologous is meant ‘not naturally adjacent’. Thus the heterologous DNA sequence may be from a different organisms, a different gene from the same organism, or even of the same gene as the is promoter, provided that the downstream sequence has been modified, usually in vitro. Such modification may be an insertion, deletion or substitution, affecting the encoded protein and/or its entrance into the secretory pathway, and/or its post-translational processing, and/or its codon usage.

[0054] The strong transcription promoter according to the invention must be in operable linkage with the heterologous downstream sequence in order to allow the transcriptional and translational machinery to recognise the starting signals. The regions upstream of open reading frames of highly expressed Phaffia genes contain TATA-like structures which are positioned at 26 to about 40 nucleotides upstream of the cap-site; the latter roughly corresponds with the transcriptional start site. Thus in order to allow transcription of the heterologous downstream sequence to start at the right location similar distances are to be respected. It is common knowledge, however, that there is a certain tolerance in the location of the TATA-signal relative to the transcription start site. Typically, mRNAs of the eukaryotic type contain a 5′-untranslated leader sequence (5′-utl), which is the region spanning the transcription start site to the start of translation; this region may vary from 30 to more than 200 nucleotides. Neither the length nor the origin of the 5′-utl is very critical; preferably it will be between 30 and 200 nucleotides. It may be from the same gene as the promoter, or it may be from the gene coding for the heterologous protein. It is well known that eukaryotic genes contain signals for the termination of transcription and/or polyadenylation, downstream of the open reading frame. The location of the termination signal is variable, but will typically be between 10 and 200 nucleotides downstream from the translational stop site (the end of the open reading frame), more usually between 30 and 100 nucleotides downstream from the translational stop site. Although the choice of the transcription terminator is not critical, it is found, that the when the terminator is selected from a region downstream of a Phaffia gene, preferably of a highly expressed Phaffia gene, more preferably from the GAPDH-encoding gene, the level of expression, as well as the frequency of transformation is improved.

[0055] It was found that significant numbers of clones were obtained which could grow on very high G418 concentrations (up to, and over, 1 mg/ml). Transcription promoters according to the invention are said to be from highly expressed genes, when, they can serve to allow growth of transformed Phaffia cells, when linked to a G418 resistance gene as disclosed in the Examples, in the presence of at least 200 μg/ml, preferably more than 400, even more preferably more than 600, still more preferably more than 800 μg/ml of G418 in the growth medium. Examples of such promoters are, in addition to the promoter upstream from the GAPDH-gene in Phaffia, the promoters from Phaffia genes which are homologous to highly expressed genes from other yeasts, such as Pichia, Saccharomyrces, Kluyveromvces, or fungi, such as Trichoderma, Aspergillus, and the like. Promoters which fulfill the requirements according to the invention, may be isolated from genomic DNA using molecular biological techniques which are, as such, all available to the person skilled in the art. The present invention provides a novel strategy for isolating strong promoters from Phaffia as follows. A cDNA-library is made from Phaffia mRNA, using known methods. Then for a number of clones with a cDNA insert, the DNA fragment (which represents the cDNA complement of the expressed mRNA) is sequenced. As a rule all fragments represent expressed genes from Phaffia. Moreover, genes that are abundantly expressed (such as the glycolytic promoters) are overrepresented in the mRNA population. Thus, the number of DNA-fragments to be sequenced in order to find a highly expressed gene, is limited to less than 100, probably even less than 50. The sequencing as such is routine, and should not take more than a couple of weeks. The nucleotide sequences obtained from this limited number of fragments, is subsequently compared to the known sequences stored in electronic databases such as EMBL or Geneseq. If a fragment shows homology of more than 50% over a given length (preferably more than 100 basepairs) the fragment is likely to represent the Phaffia equivalent of the gene found in the electronic database. In yeasts other than Phaffia, a number of highly expressed genes have been identified. These genes include the glycolytic pathway genes, phosphoglucoisomerase, phosphofructokinase, phosphotrioseisomerase, phosphoglucomutase, enolase, pyruvate kinase, alcohol dehydrogenase genes (EP 120 551, EP 0 164 556; Rosenberg S. et al., 1990, Meth. Enzymol.: 185, 341-351; Tuite M. F. 1982, EMBO J. 1, 603-608; Price V. et al., 1990, Meth. Enzymol. 185, 308-318) and the galactose regulon (Johnston, S. A. et al., 1987, Cell 50, 143-146). Accordingly, those Phaffia cDNA fragments that are significantly homologous to the highly expressed yeast genes (more than 40%, preferably more than 50% identity in a best match comparison over a range of more than 50, preferably more than 100 nucleotides) should be used to screen a genomic library from Phaffia, to find the corresponding gene. Employing this method, 14 higly expressed mRNAs from Phaffia rhodozyma have been copied into DNA, sequenced, and their (putative) open reading frames compared to a nucleic acid and amino amino acid sequence databases. It turned out that 13 out of these fourteen cDNAs coded for ribosomal protein genes, of which one coded simultaneously to ubiquitin; one cDNA codes for a glucose-repressed gene. The isolation of the genes and the promoters usually found upstream of the coding regions of these genes is now underway, and it is anticipated that each of these transcription promoters may advantageously be used to express heterologous genes, such as carotenoid biosynthesis genes. Among the genes and transcription promoters especially preferred according to this invention are the promoter found upstream of the ubiquitin-ribosomal 40S protein corresponding to the cDNA represented in SEQIDNO:10, the glucose-repressed cDNA represented in SEQIDNO:26, the 40S ribosomal protein S27 encoding cDNA represented in SEQIDNO:28, the 60S ribosomal protein P1α encodings cDNA represented by SEQIDNO:30, the 60S ribosomal protein L37e encoding cDNA represented in SEQIDNO:32, the 60S ribosomal protein L27a encoding cDNA represented in SEQIDNO:34, the 60S ribosomal protein L25 encoding cDNA represented in SEQIDNO:36, the 60S ribosomal protein P2 encoding cDNA represented in SEQIDNO:38, the 40S ribosomal protein S17A/B encoding cDNA represented in SEQIDNO:40, the 40S ribosomal protein S31 encoding cDNA represented in SEQIDNO:42, the 40S ribosomal protein S10 encoding cDNA represented in SEQIDNO:44, the 60S ribosomal protein L37A encoding cDNA represented in SEQIDNO:46, the 60S ribosomal protein L34 encoding cDNA represented in SEQIDNO:48, or the 40S ribosomal protein S16 encoding cDNA represented in SEQIDNO:50.

[0056] Promoters from these or other highly expressed genes can be picked up by the method according to the invention using only routine skills of (a) making a cDNA library on mRNA isolated from a Phaffia strain grown under desired conditions, (b) determining (part of) the nucleotide sequence of the (partial) cDNAs obtained in step (a), (c) comparing the obtained sequence data in step (b) to known sequence data, such as that stored in electronic databases, (d) cloning putative promoter fragments of the gene located either directly upstream of the open reading frame or directly upstream of the transcription start site of the gene corresponding to the expressed cDNA, and (e) verifying whether promoter sequences have been obtained by expressing a suitable marker, such as the G418 resistance gene, or a suitable non-selectable “reporter” sequence downstream from a fragment obtained in (d), transforming the DNA into a Phaffia rhodozyma strain and determining the level of expression of the marker gene or reporter sequence of transformants. A transcriptional promoter is said to be of a highly expressed gene if it is capable of making Phaffia rhodozyma cells transformed with a DNA construct comprising the said promoter linked uptream of the G418 resistance marker resistant to G418 in concentrations exceeding 200 μg per liter culture medium, preferably at least 400, more prefereably more than 600 μg/l. Especially preferred promoters are those conferring resistance against more than 800 μg/ml G418 in the growth medium.

[0057] Optionally, the transcriptional start site may be determined of the gene corresponding to the cDNA corresponding to a highly expressed gene, prior to cloning the putative promoter sequences; this may serve to locate the transcriptional initiation site more precisely, and moreover, helps to determine the length of the 5′-non-translated leader of the gene. To determine the location of the transcription start site, reverse primer extension, or classical SI-mapping may be performed, based on the knowledge of the cDNA sequence. Thus the exact location of the transcription promoter can be determined without undue burden, and the isolation of a fragment upstream of the transcription start site and containing the promoter, from a hybridising genomic clone (for example a phage or cosmid) is routine. Cloning the putative promoter fragment in front (upstream) of the coding region of, for example the G418-resistance gene, and transforming the gene cassette to Phaffia in order to evaluate the level of G418 resistance, and hence the level of expression of the G418-resistance gene as a consequence of the presence of the promoter is routine.

[0058] In a manner essentially as described for the isolation of other strong promoters, above, a transcription terminator may be isolated, with the proviso, that the terminator is located downstream from the open reading frame. The transcription stop site can be determined using procedures which are essentially the same as for the determination of the transcription start site. All these procedures are well known to those of skill in the art. A useful handbook is Nucleic Acid Hybridisation, Edited by B. D. Hames & S. J. Higgins, IRL Press Ltd., 1985; or Sambrook, sub. However, it is not critical that the transcription terminator is isolated from a highly expressed Phaffia gene, as long as it is from an expressed gene.

[0059] Using recombinant DNA according to the invention wherein the open reading frame codes for reduced sensitivity against G418, a transformation frequency was obtained up to 160 transformants per μg of linear DNA, at a G418 concentration in the medium of 40 μg/ml.

[0060] About 10 to 20 times as much transformed colonies were obtained with the vector according to the invention (pPR2) than with the prior art vector pGB-Ph9, disclosed in EP 0 590 707 A1 (see Table 2; in the experiment of Example 7, the improvement is even more striking).

[0061] The method according to the invention calls for conditions conducive to uptake of the recombinant DNA. Such conditions have been disclosed in EP 509 707. They include but are not limited to the preparation of protoplasts using standard procedures known to those of skill in the art, and subsequent incubation with the recombinant DNA. Alternatively, Phaffia cells may be incubated overnight in the presence of LiAc and recombinant DNA. Still further alternative methods involve the use of particle acceleration. According to a preferred embodiment, the conditions conducive to uptake involve electroporation of recombinant DNA into Phaffia cells, such as described by Faber et al., (1994, Current Genetics 25, 305-310). Especially preferred conditions comprise electroporation, wherein the recombinant DNA comprises Phaffia ribosomal DNA, said recombinant DNA being in the linear form, most preferably by cleaving said recombinant DNA in the said ribosomal region. Still further preferred conditions, comprise the use of recombinant DNA in amounts of between 1 and 10 μg per 108 cells, more preferably about 5 μg recombinant DNA is used per 2×108 cells, which are cultivated for 16 h at 21° C.

[0062] Once cells have been transformed according to the method, identification of transformed cells may take place using any suitable technique. Thus, identification may be done by hybridisation techniques, DNA amplification techniques such a polymerase chain reaction using primers based on the recombinant DNA used, and the like. A preferred method of identifying transformed cells is one which employs selection for the recombinant DNA that comprises a gene coding for reduced sensitivity against a selective agent. A useful selective agent is G418, hygromycin, phleomycin and amdS. Genes that code for reduced sensitivity against these selective agents are well known in the art. The open reading frames of these genes may be used as the heterologous downstream sequence according to the invention, allowing selective enrichment of transformed cells, prior to identification of transformed cells. Once transformed cells have been identified they may used for further manipulation, or used directly in the production of valuable compounds, preferably in large scale fermentors.

[0063] It will be clear, that a very efficient method for transforming Phaffia strains has been disclosed. Moreover, not only the frequency of transformation is high, the expression levels of the transforming DNA is very high as well, as is illustrated by the exceptionally high resistance against G418 of the transformed Phaffia cells when the open reading frame of the G418-resistance gene was fused to a promoter according to the invention when compared to the G418 resistance gene under control of the actin promoter in pGB-Ph9. It is concluded, therefore, that the GAPDH-promoter is a high-level transcriptional promoter that can be suitably used in conjunction with any heterologous DNA sequence, in order to reach high expression levels thereof in Phaffia strains.

[0064] It will be clear that the availability of new expression tools, in the form of the recombinant DNA according to the invention, creates a wealth of possibilities for producing new and valuable biomolecules in Phaffia.

[0065] Preferably, the downstream sequence comprises an open reading frame coding for proteins of interest. For example genes already present in Phaffia, such as those involved in the carotenoid pathway, may be manipulated by cloning them under control of the high-level promoters according to the invention. Increased expression may change the accumulation of intermediates and/or end-products or change the pathway of β-carotene, cantaxanthin, astaxanthin and the like. The overexpression of the crtB gene from Erwinia uredovora will likely increase astaxanthin levels, as the product of this gene is is involved in the rate limiting step. The expression of a protein of interest may also give rise to xanthophylls not known to be naturally produced in Phaffia, such as zeaxanthin. An open reading frame that may be suitably employed in such a method includes but is not limited to the one encoding the protein producing zeaxanthin (crtz gene) obtained from Erwinia uredovora (Misawa et al. 1990. J. Bacteriol. 172 6704-6712). Other carotenoid synthesis genes can be obtained for example from Flavobacterium (a gram-positive bacterium), Synechococcus (a cyanobacterium) or Chlamydomonas or Dunaliella (algae). Obviously, carotenoid synthesis genes of a Phaffia strain, once the genes have been isolated and cloned, are suitably cloned into a recombinant DNA according to the invention and used to modify the carotenoid content of Phaffia strains. Examples of cloned carotenoid genes that can suitably be overexpressed in Phaffia, are those mentioned in FIG. 8. Particularly useful is crtE from Phycomyces blakesleanus, encoding Geranylgeranyl Diphosphate Synthase, and crtB, encoding phytoene synthase, as this step appears to be the rate-limiting step in carotenoid synthesis in Thermus thermophylus (Hoshino T. et al., 1994, Journal of Fermentation and Bioengineerin 77 No. 4, 423-424). Especially preferred sources to isolate carotenoid biosynthetic genes or cDNAs from are the fungi Neurospora crassa, Blakeslea trispora. Other yeasts shown to possess cross-hybrising species of carotenoid biosynthetic genes are Cystofylobasidium, e.g. bisporidii and capitatum.

[0066] Carotenoid biosynthesis genes have also been identified in plants; these plant cDNAs or genes from plants may be used as well. Optionally, the codon usage of the Phaffia genes or cDNAs may be adapted to the preferred use in the host organism.

[0067] Of special interest according to the present invention, are the DNA sequences coding for four different enzymes in the carotenoid biosynthesis pathway of Phaffia rhodozyma, represented in the sequence listing. It will be clear to those having ordinary skill in the art, that once these DNA sequences have been made available it will be possible to bring about slight modifications to the DNA sequence without modifying the amino acid sequence. Such modifications are possible due to the degeneracy of the genetic code. Such modifications are encompassed in the present invention. However, also modifications in the coding sequences are envisaged that create modifications in the amino acid sequence of the enzyme. It is well known to those of skill in the art that minor modifications are perfectly permissible in terms of enzymatic acitivty. Most changes, such as delections, additions or amino acid substitutions do not affect enzymatic acitivity, at least not dramatically. Such variants as comprise one or more amino acid deletions, additions or substitutions can readily be tested using the complementation test disclosed in the specification. The skilled person is also familiar with the term “conservative amino acid substitutions”, meaning substitutions of amino acids by similar amino acids residing in the same group. The skilled person is also familiar with the term “allelic variant”, meaning naturally occurring variants of one particular enzyme. These conservative substitutions and allelic enzyme variants do not depart from the invention.

[0068] As stated, at the DNA level considerable variation is acceptable. Although the invention discloses four DNA sequences, as represented in SEQIDNO: 12, SEQIDNO: 14, SEQIDNO: 16, SEQIDNO: 18, SEQIDNO: 20, or SEQIDNO: 22, in detail also isocoding variants of the DNA sequence represented in SEQIDNO: 12, SEQIDNO: 14, SEQIDNO: 16, SEQIDNO: 18, SEQIDNO: 20, or SEQIDNO: 22, are encompassed by the present invention. Those of skill in the art would have no difficulty in adapting the nucleic acid sequence in order to optimize codon usage in a host other than P. rhodozyma. Those of skill in the art would know how to isolate allelic variants of a DNA sequence as represented in SEQIDNO: 12, SEQIDNO: 14, SEQIDNO: 16, SEQIDNO: 18, SEQIDNO: 20, or SEQIDNO: 22 from related Phaffia strains. Such allelic variants clearly do not deviate from the present invention.

[0069] Furthermore, using the DNA sequences disclosed in the sequence listing, notably SEQIDNO: 12, SEQIDNO: 14, SEQIDNO: 16 or SEQIDNO: 18, as a probe, it will be possible to isolate corresponding genes form other strains, or other microbial species, or even more remote eukaryotic species if desired, provided that there is enough sequence homology, to detect the same using hybridisation or amplification techniques known in the art.

[0070] Typically, procedures to obtain similar DNA fragments involve the screening of bacteria or bacteriophage plaques transformed with recombinant plasmids containing DNA fragments from an organism known or expected to produce enzymes according to the invention. After in situ replication of the DNA, the DNA is released from the cells or plaques, and immobilised onto filters (generally nitro-cellulose). The filters may then be screened for complementary DNA fragments using a labeled nucleic acid probe based on any of the sequences represented in the sequence listing. Dependent on whether or not the organism to be screened for is distantly or closely related, the hybridisation and washing conditions should be adapted in order to pick up true positives and reduce the amount of false positives. A typical procedure for the hybridisation of filter-immobilised DNA is described in Chapter 5, Table 3, pp. 120 and 121 in: Nucleic acid hybridisation—a practical approach, B. D. Hames & S. J. Higgins Eds., 1985, IRL Press, Oxford). Although the optimal conditions are usually determined empirically, a few useful rules of thumb can be given for closely and less closely related sequences.

[0071] In order to identify DNA fragments very closely related to the probe, the hybridisation is performed as described in Table 3 of Hames & Higgins, supra, (the essentials of which are reproduced below) with a final washing step at high stringency in 0.1*SET buffer (20 times SET=3M NaCl, 20 mM EDTA, 0.4 M Tris-HCl, pH 7.8), 0.1% SDS at 680 Celsius).

[0072] To identify sequences with limited homology to the probe the procedure to be followed is as in Table 3 of Hames & Higgins, supra, but with reduced temperature of hybridisation and washing. A final wash at 2*SET buffer, 50° C. for example should allow the identification of sequences having about 75% homology. As is well known to the person having ordinary skill in the art, the exact relationship between homology and hybridisaiion conditions depend on the length of the probe, the base composition (% of G+C) and the distribution of the mismatches; a random distribution has a stronger decreasing effect on Tm then a non-random or clustered pattern of mismatches.

[0073] The essentials of the procedure described in Table 3, Chapter 5 of Hames & Higgins are as follows:

[0074] (1) prehybridisation of the filters in the absence of probe, (2) hybridisation at a temperature between 50 and 68° C. in between 0.1 and 4*SET buffer (depending on the stringency), 10*Denhardt's solution (100*Denhardt's solution contains 2% bovine serum albumin, 2% Ficoll, 2% polyvinylpyrrolidone), 0.1% SDS, 0.1% sodiumpyrophosphate, 50 μg/ml salmon sperm DNA (from a stock obtainable by dissolving 1 mg/ml of salmon sperm DNA, sonicated to a length of 200 to 500 bp, allowed to stand in a water bath for 20 min., and diluted with water to a final concentration of 1 mg/ml); hybridisation time is not too critical and may be anywhere between 1 and 24 hours, preferably about 16 hours (o/n); the probe is typically labeled by nick-translation using 32P as radioactive label to a specific activity of between 5*107 and 5*108 c.p.m./μg; (3) (repeated) washing of the filter with 3*SET, 0.1% SDS, 0.1% sodiumpyrophosphate at 68° C. at a temperature between 50° C. and 68° C. (dependent on the stringency desired), repeated washing while lowering the SET concentration to 0.1%., wash once for 20 min. in 4*SET at room temperature, drying filters on 3MM paper, exposure of filters to X-ray film in a cassette at −70° C. for between 1 hour and 96 hours, and developing the film.

[0075] Generally, volumina of prehybridisation and hybridisation mixes should be kept at a minimum. All “wet” steps may be carried out in little sealed bags in a pre-heated water bath.

[0076] The above procedure serves to define the DNA fragments said to hybridise according to the invention. Obviously, numerous modifications may be made to the procedure to identify and isolate DNA fragments according to the invention. It is to be understood, that the DNA fragments so obtained fall under the terms of the claims whenever they can be detected following the above procedure, irrespective of whether they have actually been identified and/or isolated using this procedure.

[0077] Numerous protocols, which can suitably be used to identify and isolate DNA fragments according to the invention, have been described in the literature and in handbooks, including the quoted Hames & Higgins, supra).

[0078] With the advent of new DNA amplification techniques, such as direct or inverted PCR, it is also possible to clone DNA fragments in vitro once sequences of the coding region are known.

[0079] Also encompassed by the claims is a DNA sequence capable, when bound to nitrocellulose filter and after incubation under hybridising conditions and subsequent washing, of specifically hybridising to a radio-labelled DNA fragment having the sequence represented in SEQIDNO: 12, SEQIDNO: 14, SEQIDNO: 16 or SEQIDNO: 18, as detectable by autoradiography of the filter after incubation and washing, wherein said incubation under hybridising conditions and subsequent washing is performed by incubating the filter-bound DNA at a temperature of at least 50° C., preferably at least 55° C., more preferably at least 60° C. in the presence of a solution of the said radio-labeled DNA in 0.3 M NaCl, 40 mM Tris-HCl, 2 mM EDTA, 0.1% SDS, pH 7.8 for at least one hour, whereafter the filter is washed at least twice for about 20 minutes in 0.3 M NaCl, 40 mM Tris-HCl, 2 mM EDTA, 0.1% SDS, pH 7.8, at a temperature of 50° C., preferably at least 55° C., more preferably at least 60° C., prior to autoradiography.

[0080] The heterologous DNA sequence according to the invention may comprise any open reading frame coding for valuable proteins or their precursors, like pharmaceutical proteins such as human serum albumin, IL-3, insulin, factor VIII, tPA, EPO, α-interferon, and the like, detergent enzymes, such as proteases and lipases and the like, cell wall degrading enzymes, such as xylanases, pectinases, cellulases, glucanases, polygalacturonases, and the like, and other enzymes which may be useful as additives for food or feed (e.g. chymosin, phytases, phospholipases, and the like). Such genes may be expressed for the purpose of recovering the protein in question prior to subsequent use, but sometimes this may not be necessary as the protein may be added to a product or process in an unpurified form, for example as a culture filtrate or encapsulated inside the Phaffia cells.

[0081] The yeast cells containing the carotenoids can be used as such or in dried form as additives to animal feed. Furthermore, the yeasts can be mixed with other compounds such as proteins, carbohydrates or oils.

[0082] Valuable substances, such as proteins or pigments produced by virtue of the recombinant DNA of the invention may be extracted. Carotenoids can also be isolated for example as described by Johnson et al. (Appl. Environm. Microbiol. 35: 1155-1159 (1978)).

[0083] Purified carotenoids can be used as colorants in food and/or feed. It is also possible to apply the carotenoids in cosmetics or in pharmaceutical compositions.

[0084] The heterologous downstream sequence may also comprise an open reading frame coding for reduced sensitivity against a selective agent. The open reading frame coding for an enzyme giving G418 resistance was used satisfactorily in the method according to the invention, but the invention is not limited to this selection marker. Other useful selection markers, such as the phleomycin resistance gene may be used, as disclosed in EP 590 707. Each of these genes is advantageously expressed under the control of a strong promoter according to the invention, such as the GAPDH-promoter.

[0085] The invention is now being illustrated in greater detail by the following non-limitative examples.

Experimental

[0086] Strains:

[0087]

E. coli
DH5α: supE44lacU169 (80lacZM15) hsdR17 recA1 endA1 gyrA96 thi-1 relA1

[0088]

E. coli
LE392: supE44 supF58 hsdR514 galK2 galT22 metB1 trpR55 lacY1

[0089]

P. rhodozyma
CBS6938

[0090] Plasmids:

[0091] pUC19 (Gibco BRL)

[0092] pTZ19R

[0093] PUC-G418

[0094] pGB-Ph9 (Gist-brocades)

[0095] pMT6 (1987, Breter H.-J., Gene 53, 181-190))

[0096] Media:

[0097] LB: 10 g/l bacto tryptone, 5 g/l yeast extract, 10 g/l NaCl. Plates; +20 g/l bacto agar. When appropriate 50 μg/ml ampicillin.

[0098] YePD: 10 g/l yeast extract, 20 g/l bacto peptone, 20 g/l glucose. Plates; +20 g/l bacto agar.

[0099] When appropriate 50 μg/ml Geneticin (G418).

[0100] Methods: All molecular cloning techniques were essentially carried out as described by Sambrook et al. in Molecular Cloning: a Laboratory Manual, 2nd Edition (1989; Cold Spring Harbor Laboratory Press).

[0101] Enzyme incubations were performed following instructions described by the manufacturer. These incubations include restriction enzyme digestion, dephosphorylation and ligation (Gibco BRL).

[0102] Isolation of chromosomal DNA from Phaffia rhodozyma as described in example 3 of patent Gist-brocades; EP 0 590 707 A1. Chromosomal DNA from K. lactis and S. cerevisiae was isolated as described by Cryer et al. (Methods in Cell Biology 12: 39, Prescott D. M. (ed.) Academic Press, New York).

[0103] Isolation of large (>0.5-kb) DNA fragments from agarose was performed using the Geneclean II Kit whereas small (<0.5-kb) and DNA fragments or fragments from PCR mixtures were isolated using Wizard™ DNA Clean-Up System (Promega).

[0104] Transformation of E. coli was performed according to the CaCl2 method described by Sambrook et al. Packaging of cosmid ligations and transfection to E. coli LE392 was carried out using the Packagene Lambda DNA Packaging System (Promega), following the Promega protocols.

[0105] Isolation of plasmid DNA from E. coli was performed using the QIAGEN (Westburg B.V. NL).

[0106] Transformation of Phaffia CBS6938 was done according to the method for H. polymorpha described by Faber et al., supra;

[0107] Inoculate 30 ml of YePD with 1 CBS6938 colony

[0108] Grow 1-2 days at 21° C., 300 rpm (pre-culture)

[0109] Inoculate 200 ml of YePD with pre-culture to OD600=between 0 and 1 (if above 1 dilute with water)

[0110] Grown o/n at 21° C., 300 rpm until OD600=1.2 (dilute before measuring)

[0111] Centrifuge at 5 min. 8000 rpm, room temperature. Remove supernatant thoroughly

[0112] Resuspend pellet in 25 ml 50 mM KPi pH 7.0, 25 mM DTT (freshly made)

[0113] Transfer suspension to a fresh sterile 30 ml centrifuge tube and incubate for 15 min. at room temperature

[0114] Centrifuge 5 min. at 8000 rpm 4° C., remove supernatant thoroughly

[0115] Resuspend pellet in 25 ml of ice cold STM (270 mM sucrose, 10 mM Tris pH 7.5, 1 mM MgCl2)

[0116] Centrifuge 5 min. at 8000 rpm, 4° C.

[0117] Repeat washing step

[0118] Resuspend cells in 0.5 ml of ice cold STM (3*109 cells/ml). Keep on ice!

[0119] Transfer 60 μl of cell suspension to pre-cooled Eppendorf tubes containing 5 μg transforming DNA (use precooled tips!), Keep on ice

[0120] Transfer Cell/DNA mix to precooled electroporation cuvettes (top to bottom)

[0121] Pulse: 1.5 kV, 400 Ω, 25 μF

[0122] Immediately add 0.5 ml of ice cold YePD. Transfer back to ep using a sterile Pasteur pipette

[0123] Incubate 2.5 hrs at 21° C.

[0124] Plate 100 μl onto YePD-plates containing 40 μg/ml G418

[0125] Incubate at 21° C. until colonies appear.

[0126] Pulsed Field Electrophoresis was performed using a GENE Navigator+accessories (Pharmacia). Conditions: 0.15*TBE, 450 V, pulse time 0.5 s, 1.2% agarose, run time 2 h.

[0127] Polymerase Chain Reaction (PCR) experiments were performed in mixtures having the following composition:

[0128] 5 ng of plasmid DNA or 1 μg chromosomal DNA

[0129] 0.5 μg of oligo nucleotides (5 μg degenerated oligo's in combination with chromosomal DNA)

[0130] 10 nm of each dNTP

[0131] 2.5 μm KCl

[0132] 0.5 μm Tris pH 8.0

[0133] 0.1 μm MgCl2

[0134] 0.5 μg gelatin

[0135] 1.3 U Taq polymerase (5 U in combination with chromosomal DNA) H2O was added to a total volume of 50 μl

[0136] Reactions were carried out in an automated thermal cycler (Perkin-Elmer). Conditions: 5 min. 95° C., followed by 25 repeated cycli: 2′ 94° C., 2′ 45° C.3′ 72° C. Ending; 10 min. 72° C.

[0137] Fusion PCR reactions were performed as described above, except that 2 DNA fragments with compatible ends were added as a template in equimolar amounts.

[0138] Oligo nucleotide sequences were as follows:

13005:CGGGATCCAA(A/G)CThACNGGNATGGC;(SEQIDNO: 1)3006:CGGGATCC(AIG)TAICC(C/AIG)(C/T)A(T/C)TC(AIG)TT(A/G)TC(AIG)TACCA;(SEQIDNO: 2)4206:GCGTGACTTCTGGCCAGCCACGATAGC;(SEQIDNO: 3)5126:TTCAATCCACATGATGGTAAGAGTGTTAGAGA;(SEQIDNO: 4)5127:CTTACCATCATGTGGATTGAACAAGATGGAT;(SEQIDNO: 5)5177:CCCAAGCTTCTCGAGGTACCTGGTGGGTGCATGTATGTAC;(SEQIDNO: 6)5137:CCAAGGCCTAAAACGGATCCCTCCAAACCC;(SEQIDNO: 7)5138:GCCAAGCTTCTCGAGCTTGATCAGATAAAGATAGAGAT;(SEQIDNO: 8)

EXAMPLE 1

[0139] G-418 Resistance of Phaffia Transformant G418-1

[0140] To determine the expression of the G418 resistance gene in pGB-Ph9, transformant G418-1 (EP 0 590 707 A1) was exposed to increasing concentrations of G418.

[0141] Two dilutions of a G418-1 culture were plated onto YepD agar containing 0-1000 μg/ml G418 (Table 1).

2TABLE 1Survival of Phaffia transformant G418-1 on YepD agarmedium containing increasing concentrations of G418.Phaffia G418-1Phaffia G418-1Dil. =Dil. =Phaffia (CBS6938)10−4(OD600 =10−5(OD600 =Dil. =[G418] μg/ml7)7)0(OD600 = 5)0>30074>300200>300700300>30061040021213050010206000007000008000009000001000000

[0142] At a concentration of 600 μg/ml G418 less than 1% of the plated cells survived. It can be concluded, that despite multicopy integration of pGB-Ph9, G418-1 shows a rather weak resistance to G418 (Scorer et al., 1994, Bio/Technology 12, p. 181 et seq., Jimenez and Davies, 1980, Nature 187 p. 869 et seq.), most probably due to a weak action of the Phaffia actin promoter in the plasmid. The results that the Phaffia actin promoter works poorly, prompted us to isolate promoter sequences of Phaffia with strong promoter activity.

EXAMPLE 2

[0143] Synthesis of Specific Probes of Glycolytic genes from Phaffia rhodozyma by PCR

[0144] The polymerase chain reaction (PCR) technique was used in an attempt to synthesize a homologous probe of the genes encoding glyceraldehyde-3-phosphate dehydrogenase (GAPDH), phosphoglycerate kinase (PGK) and the triose phosphate isomerase (TPI) of Phaffia rhodozyma.

[0145] A set of degenerated oligonucleotides was designed based on the conserved regions in the GAPDH-gene (Michels et al., 1986. EMBO J. 5: 1049-1056), PGK-gene (Osinga et al., 1985. EMBO J. 4: 3811-3817) and the TPI-gene (Swinkels et al., 1986. EMBO J. 5: 1291-1298).

[0146] All possible oligo combinations were used to synthesize a PCR-fragment with chromosomal DNA of Phaffia rhodozyma (strain CBS6938) as template. Chromosomal DNA of Saccharomyces cerevisiae and Kluyveromyces lactis as template was used to monitor the specificity of the amplification. The PCR was performed as described above, the PCR conditions were 1′ 95° C., 2′ annealing temperature (Ta), in 5′ from annealing temperature to 72° C., 2′ 72° C., for 5 cycli followed by 1′ 95° C., 2′ 55° C. and 2′ 72° C. for 25 cycli and another elongation step for 10′ 72° C. Three different Ta were used 40° C., 45° C. and 50° C.

[0147] Under these conditions, only one primer combination produced a fragment of the expected size on chromosomal DNA of Phaffia as template. Using the oligo combination no: 3005 and 3006 and a Ta of 45° C. a 0.3-kb fragment was found. Specifically, the GAPDH oligonucleotides correspond with amino acids 241-246 and 331-338 of the published S. cerevisiae sequence. (It was concluded that to isolate the promoters corresponding to the PGK- and TPI-genes from Phaffia, either further optimization of the PCR-conditions is required, or homologous primers should be used. Another alternative method for isolating high level promoters is disclosed in the detailed description, supra.

[0148] The amplified fragment was purified from the PCR reaction and was digested with BamHI and ligated into the dephosphorylated BamHI site of pTZ19R. The ligation mixture was transformed to competent E. coli DH5α cells prepared by the CaCl2-method and the cell were plated on LB-plates with 50 μg/ml Amp and 0.1 mM IPTG/50 μg/ml X-gal. Plasmid DNA was isolated from the white colonies. The pTZ19R clone with the right insert, called pPRGDHI, was subsequently used for sequence analysis of the insert.

[0149] The cloned sequence encoded for the carboxy terminal fragment of GAPDH of Phaffia as shown by comparison with the GAPDH-gene sequence of S. cerevisiae (Holland and Holland, 1979. J. of Biol. Chem. 254: 9839-9845).

EXAMPLE 3

Isolation of the GAPDH-Gene of Phaffia

[0150] To obtain the complete GAPDH-gene including expression signals the 0.3-kb BamHI fragment of pPRGDH1 was used to screen a cosmid library of Phaffia.

[0151] Preparation of the Vector for Cosmid Cloning.

[0152] Vector preparation was simplified, because of the presence of a double cos-site in pMT6. PMT6 was digested to completion with blunt end cutter PvuII to release the cos-sites. Digestion efficiency was checked by transformation to E. coli DH5α and found to be >99%.

[0153] The PvuII digested pMT6 was purified by phenol:chloroform extraction and ethanol precipitation and finally solved in 30 μl TE at a concentration of 2 μg/μl.

[0154] The vector was subsequently digested with cloning enzyme BamHI and the vector arms were purified as described above (“Experimental”).

[0155] Preparation of Target DNA

[0156] Isolation of genomic DNA of Phaffia strain CBS6938 was performed as described in the part named “Experimental”. The cosmid pMT6 containing inserts of 25-38-kb are most efficiently packaged. Therefore genomic DNA was subjected to partial digestion with the restriction enzyme Sau3A. Target DNA was incubated with different amounts of enzyme. Immediately after digestion the reactions were stopped by the extraction of DNA from the restriction mixture with phenol-chloroform. The DNA was precipitated by using the ethanol method and the pelleted DNA after centrifugation was dissolved in a small volume of TE. Contour clamped homogeneous electric field (CHEF) electrophoresis was used to estimate the concentration and size of the fragments (Dawkins, 1989, J. of Chromatography 492, pp. 615-639).

[0157] Construction of Genomic Cosmid Library.

[0158] Ligation of approximately 0.5 μg of vector arm DNA and 0.5 μg of target DNA was performed in a total volume of 10 μl in the presence of 5 mM ATP (to prevent blunt end ligation). Packaging in phage heads and transfection to E. coli LE 392 as described in Experimental. The primary library consisted of 7582 transfectants with an average insert of 28-kb as determined by restriction analysis. The library represents 3.5 times the genome with a probability of the presence of all genes in the library of 0.97 as calculated according to Sambrook (supra). For library amplification the transfectants were pooled by resuspending in 8 ml LB-broth. Additional 4.8 ml glycerol was added. The transfectants mixture was divided into 16 samples of 800 μl each and stored at −80° C. This amplified library consisted of 2.9*109 transfectants.

[0159] Screening of the Cosmid Library.

[0160] A 100 μl sample was taken from this library and further diluted (106) in LB-broth and 200 μl was plated onto 10 LB-plates containing ampicillin. The plates were incubated overnight at 37° C. Each plate contained 300-400 colonies and filters were prepared. These filters were screened with the GAPDH-probe using hybridization and washing conditions as described above (“Experimental”). After 16 hours exposure, 3 strong hybridization signals were found on the autoradiogram.

[0161] Cosmid DNA isolated from these positive colonies was called pPRGDHcos1, pPRGDHcos2 and pPRGDHcos3.

[0162] Chromosomal DNA isolated from Phaffia rhodozyma strain CBS 6938 and cosmid pPRGDHcos1 was digested with several restriction enzymes. The DNA fragments were separated, blotted and hybridized as described before. The autoradiograph was exposed for different time periods at −80° C. The film showed DNA fragments of different length digested by different restriction enzymes which hybridize with the GAPDH-probe (FIG. 1).

[0163] Furthermore, from Southern analysis of the genomic DNA of Phaffia using the GAPDH fragment as probe, it was concluded that the GAPDH-encoding gene is present as a single copy gene in Phaffia rhodozyma, whereas in Saccaromyces cerevisiae GAPDH is encoded by three closely related but unlinked genes (Boucherie et al, 1995. FEMS Microb. Letters 135:127-134).

[0164] Hybridizing fragments of pPRGDHcos1 for which a fragment of the same length in the chromosomal DNA digested with the same enzyme was found, were isolated from an agarose gel. The fragments were ligated into the corresponding sites in pUC19. The ligation mixtures were transformed to competent E. coli cells. The plasmids with a 3.3-kb SalI insert and a 5.5-kb EcoRI insert were called pPRGDH3 and pPRGDH6, respectively. The restriction map of pPRGDH3 and pPRGDH6 is shown in FIG. 2. Analysis of the sequence data of the insert in pPRGDH1 showed us that there was a HindIII site at the C-terminal part of the GAPDH-gene. From this data it was suggested that the insert in pPRGDH6 should contain the complete coding sequence of GAPDH including promoter and terminator sequences.

EXAMPLE 4

Characterization of the GAPDH-gene

[0165] In order to carry out sequence analysis without the need to synthesize a number of specific sequence primers a number of deletion constructs of plasmids pPRGDH3 and pPRGDH6 were made using convenient restriction sites in or near the putative coding region of GAPDH gene.

[0166] The plasmids were digested and after incubation a sample of the restriction mixture was analyzed by gel electrophoresis to monitor complete digestion. After extraction with phenol-chloroform the DNA was precipitated by ethanol. After incubation at −20° C. for 30′ the DNA is pelleted by centrifugation, dried and dissolved in a large volume (0.1 ng/μl) of TE. After ligation the mixtures were transformed to E. coli. Plasmid DNA isolated from these transformants was analyzed by restriction analysis to reveal the right constructs. In this way the deletion constructs pPRGDH3δHIII, pPRGDH6δBamHI, pPRGDH6δSstI and pPRGDH6δSalI (FIG. 1).

[0167] In addition to this, the 0.6-kb and 0.8-kb SstI fragments derived from pPRGDH6 were subcloned in the corresponding site of pUC19.

[0168] Sequence analysis was carried out using pUC/M13 forward and reverse primers (Promega). The sequencing stategy is shown in FIG. 2 (see arrows).

[0169] On the basis of homology with the GAPDH-gene sequence of S. cerevisiae (Holland and Holland, 1979. J. of Biol. Chem. 254: 9839-9845) and K. lactis (Shuster, 1990. Nucl. Acids Res. 18, 4271) and the known splice site concensus J. L. Woolford. 1989. Yeast 5: 439-457), the introns and the possible ATG start were postulated.

[0170] The GAPDH gene has 6 introns (FIG. 1) and encodes a polypeptide of 339 amino acids. This was completely unexpected considering the genomic organisation of the GAPDH genes of K. lactis and S. cerevisiae which have no introns and both consist of 332 amino acids. The homology on the amino acid level between the GAPDH gene of Phaffia and K. lactis and S. cerevisiae is 63% and 61%, respectively.

[0171] Most of the introns in the GAPDH gene are situated at the 5′ part of the gene. Except intron III all introns contain a conserved branch-site sequence 5′-CTPuAPy-3′ found for S. cerevisiae and S. pombe.

[0172] By computer analysis of the upstream sequence using PC-gene 2 putative eukaryotic promoter elements, TATA-box (position 249-263 in SEQIDNO: 111) and a number of putative Cap signal (between position 287 and 302 in SEQIDNO: 11) were identified.

EXAMPLE 5

Cloning of the GAPDH Promoter Fused to G4118 in pUCG418

[0173] In order to construct a transcription fusion between the GAPDH promoter and the gene is encoding G418 resistence the fusion PCR technique was used.

[0174] Using plasmid pPRGDH6 the GAPDH promoter could be amplified by standard PCR protocols (“Experimental”).

[0175] In the PCR mix pPRGDH6 and oligo's No. 5177 and 5126 (Sequences in “Experimental”) were used. A 416 bp DNA fragment was generated containing the entire GAPDH promoter sequence. In addition this fragment also contains a HindIII, XhoI and a KpnI restriction site at it's 5′end and 12 nt overlap with the 5′ end of the gene encoding G418 resistance.

[0176] The 217 bp portion of the 5′end of the G418 coding sequence was also amplified by PCR using pUC-G418 and oligo's 4206 and 5127. A 226 bp DNA fragment was obtained containing the 217 bp 5′end of G418 and having a 9 nucleotides overlap with the 3′end of the earlier generated GAPDH promoter fragment. It also contained a MscI site at it's 3end.

[0177] The PCR fragments were purified from the PCR mixture using the WIZARD Kit. Approximately 1 μg of the GAPDH promoter fragment and 1 μg of the G418 PCR fragment were used together with oligo's 5177 and 4206 in a fusion PCR experiment (Experimental). A 621 bp DNA fragment was generated, containing the GAPDH promoter directly fused to the 5′ portion of G418. After purification the DNA fragment was digested with MscI and KpnI. The 3.4 Kb MscI-KpnI fragment of pUC-G418, containing pUC sequences and the 3′ portion of G418, was used as a vector. The ligation mixture was transformed to competent E. coli DH5α cells. Transformant colonies containing the fusion PCR DNA inserted were identified by digestion with different restriction enzymes.

[0178] Thus, plasmid PPR1 was obtained, containing the GAPDH promoter directly fused to the G418 marker gene. Three pPR1 vectors isolated from independent transformants were used in further cloning experiments.

[0179] To target the plasmid, after transformation, to a specific integration site a 3.0-kb SstI fragment containing a part of the ribosomal DNA of Phaffia was cloned in pPR1. The ribosomal DNA fragment was isolated from an agarose gel after digestion with SstI of plasmid pGB-Ph11 (EP 590 707 A1). This fragment was ligated in the dephosphorylated SstI site of pPR1. The ligation mixture was transformed to competent E. coli cells. Plasmid DNA was isolated and using restriction analysis it was shown that several colonies contain the expected plasmid pPR2. The complete cloning strategy is shown in FIG. 3.

EXAMPLE 6

Transformation of Phaffia with DPR2

[0180] Transformation of Phaffia strain 6938 was performed using an electroporation procedure as previously described by Faber et al. (1994, Curr. Genet. 1994: 25,305-310) with the following modifications:

[0181] Electropulsing was performed using the Bio-rad Gene Pulser with Pulse Controller and with Bio-rad 2 mm cuvettes.

[0182] Phaffia was cultivated for 16 h at 21° C.

[0183] Per transformation 2×108 cells were used together with 5 μg of linearized vector. Linearization was done in the rDNA sequence using ClaI to enable integration at the rDNA locus in the Phaffia genome. Following the electric pulse (7.5 kV/cm, 400 Ω and 25 μF) 0.5 ml YePD medium was added to the cell/DNA mixture. The mixture was incubated for 2.5 h at 21° C. and subsequently spread on 5 selective YEDP agar plates containing 40 μg/ml G418.

[0184] As shown in Table 2 we were able to generate transformants with 115 transformants per jig DNA; the average transformation frequency was 50 transformants/μg pPR2 as judged over a number of experiments. Transformation of the closed circular form of pPR2 did not result in transformation suggesting that there is no autonomously replicating sequence present within the vector sequences. Using pPR2 a 10 to 50-fold increase in transformation frequency was found compared to a previous constructed transformation vector for Phaffia, called pGB-Ph9. In this latter vector a translation fusion was made between the 5′ part of the actin gene of Phaffia and G418.

[0185] In order to analyze the level of resistance of transformants the mixture or DNA/cells was plated onto selective plates containing different amounts of G418. Although the total number of transformants decreases with the increasing amounts of G418, we were still able to obtain a considerable number of transformants (table 3).

[0186] In another experiment 30 transformants obtained under standard selection conditions (40 μg/ml) were transfered to plates containing 50, 200 or 1000 μg/ml. After incubation of the plates at 21° C. for 4-5 days, 23 transformants out of 30 tested were able to grow on plates containing 200 μg/ml G418. One transformant was able to grow on plates containing upto and above 1000 μg/ml G418.

3TABLE 2Transformation frequency of pGB-Ph9 and pPR2Exp. 1Exp. 2—698pGB-Ph9xBglII467pPR2 cccn.dn.dpPR2(A)xClaI71456(B)639124(C)443153

[0187] Total number of transformants (>1 mm) in different transformation experiments after 4-5 days incubation.

4TABLE 3Comparison of G418 sensitivity as a result of two differentG418-resistance genes in pGB-Ph9 and pPR2concentrationNumber of transformantsG418 (μg/ml)pPR2xClaIpGB-Ph9xBglII (=pYac4)40480250346—60155—7061—80141—9072—10064—

[0188] Analysis of pPR2 Transformants.

[0189] To analyse the integration event and the number of integrated vector copies total genomic DNA from six independent transformants was isolated. Therefore these transformants were cultivated under selective conditions, i.e. YePD+50 μg/ml G418. Chromosomal DNA was digested with ClaI. The DNA fragments were separated by gel electrophoresis and transfered to nitrocellulose and the Southern blot was probed with Phaffia DNA.

[0190] Besides the rDNA band of 9.1 kb an additional band of 7.1 kb of similar fluorescing intensity was observed in the transformants. This band corresponds to the linearised form of pPR2. From the intensity of these bands it was concluded that the copy number was about 100-140 copies of pPR2. These results are similar to those observed for pGB-Ph9, ruling out that the improved G418-resistance is due to differences in copy number of integrated vectors alone. It is not known whether the multiple copy event is caused by multiple copy integration of pPR2 or by the amplification of a single copy in the rDNA or a combination of both events.

EXAMPLE 7

Construction of pPR2T by Cloning the GAPDH-Terminator into pPR2

[0191] Eukaryotic mRNAs contain modified terminal sequences, specificaly the 3′ terminal poly(A). As the prokaryotic gene encoding G418 resistance lacks eukaryotic termination signals, which might effect proper transcription termination and mRNA stability (1994, Raue, H. A., TIBTECH 12: 444-449), a part of the 3′ non-coding sequence of GAPDH was introduced.

[0192] To that end, a 307 bp fragment, consisting of 281 bp of the 3′ non-coding region of GAPDH and other additional cloning sequences, was amplified by PCR using the oligo's 5137 and 5138 (“Experimental”). The upstream oligo 5137 consists of the last 14 nucleotides of the coding and 17 nucleotides of the 3′ non-coding region of GAPDH. By base substitutions of the 5th (T->A) and 8th (T->C) nucleotide of the non-coding sequence a BamHI restriction, site was introduced. In addition this fragment contains a XhoI and a HindIII restriction site at its 3′ end.

[0193] The PCR fragment was purified from the PCR mixture using the WIZARD Purification Kit and digested with BamHI and HindIII. A 288 bp fragment was isolated and cloned into the corresponding sites of the previously constructed Phaffia transformation vector pPR2, yielding pPR2T.

[0194] Upon transformation of Phaffia, using G418 as selective agent, the transformation frequencies (number of transformants per μg of DNA) obtained with the improved construct pPR2T was approximately 5 to 10 times higher than the transformation frequency of pPR2 (i.e. without a Phaffia homologous transcription termination signal). The results of a typical experiment are given in Table 4.

5TABLE 4Transformation frequency at 50 μg/mlG418 for pGB-Ph9, pPR2 and pPR2TVectortransformantstransformants/μg DNApGB-Ph9 (ccc)——pGB-Ph9 (xBglII)601pPR2 (ccc)1—pPR2 (xClaI)3000-960050-160pPR2T (ccc)——pPR2T (xClaI)45600760pPR2T (xSfiI)108018

[0195] Phaffia cells transformed with pPR2T were tested for their ability to grow on high levels of G418. The level of G418 on which growth is still possible was taken as a measure of the expression level of the G418 resistance gene in transformants, as a result of the presence of the Phaffia promoter, and/or terminator. Preliminary results indicate that the number of transformants able to grow on high levels of G418 are significantly higher than without terminator.

[0196] In Summary

[0197] From the above results, it was concluded, that the presence of the GAPDH-promoter (pPR2) resulted in a considerable increase of the transformation frequency (from 1 to at least 50 per fig of DNA) when compared to the vector containing the actin-promoter (pGB-Ph9). These results are in line with the results obtained with the G418 sensitivity test (Table 3 and 4) which indicate superior expression levels under the control of the GAPDH promoter. The possibility that the difference in transformation frequency could be due solely to the difference in linearising the vectors, (BglII, ClaI and SfiI all cut inside the ribosomal DNA locus, but at different positions), was ruled out by comparison of pPR2(×SflI) with pGB-Ph9(xSfiI). The difference in transformation frequency between the two pPR2 and pGB-Ph9, linearised with SfiI is still considerable. However, it is concluded that the choice of the linearisation site does have effect on the transformation frequency; linearisation with ClaI is preferred.

[0198] The improvements obtained by using a high-level promoter, such as GAPDH, are irrespective of whether a homologous terminator is used (pPR2 (without homologous terminator) performs far better than pGB-Ph9, both in G418 sensitivity tests, as well as in terms of transformation frequency).

[0199] The presence of a homologous terminator results in both higher transformation frequencies and higher expression levels; this result is concluded to be independent of the promoter used. Preliminary results indicate that considerable improvements are obtained when the pGB-Ph9 construct is completed with a transcription terminator, such as the GAPDH-terminator used in pPR2T.

[0200] The following Examples illustrate the isolation of DNA encoding enzymes involved in the carotenoid biosynthesis pathway of Phaffia rhodozyma. These DNA sequences can suitably be used for a variety of purposes; for example to detect and isolate DNA sequences encoding similar enzymes in other organisms, such as yeast by routine hybridisation procedures, to isolate the transcription promoters and/or terminators, which can be used to construct expression vectors for both heterologous as well as homologous downstream sequences to be expressed. The DNA sequences encoding carotenoid biosynthesis genes can suitably be used to study the over-expression, either under the control of their own promoters or heterologous promoters, such as the glycolytic pathway promoters illustrated above. For example, transformation of Phaffia rhodozyma with carotenoid encoding DNA sequences according to the invention effectively results in amplification of the gene with respect to the wild-type situation, and as a consequence thereof to overexpression of the encoded enzyme.

[0201] Hence, the effect of over-expression of one or more genes encoding carotenoid biuosynthesis genes can thus be studied. It is envisaged that mutant Phaffia strains can be obtained producing higher amounts of valuable carotenoids, such as β-carotene, cantaxanthin, zeaxanthin and/or astaxanthin. Similarly, the DNA sequences encoding enzymes involved in the carotenoid biosynthesis pathway can be introduced into other hosts, such as bacteria, for example E. coli, yeasts, for example species of Saccharomyces, Kluyveromyces, Rhodosporidium, Candida, Yarrowia, Phycomyces, Hansenula, Picchia, fungi, such as Aspergillus, Fusarium, and plants such as carrot, tomato, and the like. The procedures of transformation and expression requirements are well known to persons skilled in these arts.

[0202] Strains:

[0203]

E. coli
XL-Blue-MRF′Δ(mcrA)183Δ(mcrCB-hsdSMR-mrr) 173 endA1 supE44 thi-1 recA1 gyrrA96 relA1 lac[F′ proAB laqqZΔM15 Tn10 (Tetr)]

[0204] ExAssist™ interference-resistant helper phage (StategeneR)

[0205]

P. rhodozyma
CBS6939 or

[0206]

P. rhodozyma
asta 1043-3

[0207] Plasmids Used for Cloning:

[0208] pUC19 Apr (Gibco BRL)

[0209] Uni-ZAP™ XR vector (lambda ZAPR II vector digested with EcoRI-XhoI, CIAP treated; StrategeneR)

[0210] Media:

[0211] LB: 10 g/l bacto tryptone, 5 g/l yeast extract, 10 g/l NaCl. Plates; +20 g/l bacto agar.

[0212] When appropriate 50-100 μg/ml ampicillin (Ap), 30 μg/ml chloramphenicol (Cm) and 1 mM isopropyl-1-thio-β-galactopyranoside (IPTG) was added.

[0213] YePD: 10 g/l yeast extract, 20 g/l bacto peptone, 20 g/l glucose. Plates; +20 g/l bacto agar.

[0214] All molecular cloning techniques were essentially carried out as described by Sambrook et al. in Molecular Cloning: a Laboratory Manual, 2nd Edition (1989; Cold Spring Harbor Laboratory Press). Transformation of E. coli was performed according to the CaCl2 method described by Sambrook et al.

[0215] Enzyme incubations were performed following instructions described by the manufacturer. These incubations include restriction enzyme digestion, dephosphoryiation and ligation (Gibco BRL). Isolation of plasmid DNA from E. coli was performed using the QIAGEN (Westburg B.V. NL).

[0216] For sequence analysis deletions constructs and oligonucleotides were made to sequence the complete sequence using a Taq DYE Primer Cycle Sequencing kit (Applied Biosystems).

EXAMPLE 8

Description of Plasmids

[0217] Plasmids (pACCAR25ΔcrtE, pACCAR25ΔcrtB, pACCRT-EIB, pACCAR16ΔcrtX and pACCAR25ΔcrtX), which contain different combinations of genes involved in the biosynthesis of carotenoid in Erwinia uredovora were gifts from Prof. Misawa; Kirin Brewery co., LTD.; Japan). The biosynthetic route of carotenoid synthesis in Erwinia uredovora is shown in FIG. 8.

[0218] In addition a derivative of pACCAR25ΔcrtX, designated pACCAR25ΔcrtXAcrtI, was made in our laboratory. By the introduction of a frameshift in the BamHI restriction site the crtI gene was inactivated. E. coli strains harboring this plasmid acummulate phytoene which can be monitored by the red phenotype of the colony.

[0219] All plasmids are derivatives of plasmid pACYC184 (Rose R E; Nucl. Acids Res. 16 (1988) 355), which contains a marker conferring chloramphenicol-resistance. Furthermore these plasmids and derivatives thereof contain a replication origin that is compatible to vectors such as pUC and pBluescript. Each plasmid contains a set of carotenoid biosynthetic genes of Erwinia uredovora mediating the formation of different carotenoid in E. coli. The complete list of plasmid used in this study is shown in Table 5.

6TABLE 5Summary of carotenoid producing E. coli strains used in this study.CAROTENOIDCOLORPLASMID:GENOTYPE:ACCUMULATED:PHENOTYPE:pACCAR25ΔcrtEcrtB; crtI; crtY;farnesylwhitecrtX;pyrophosphate/crtZisopentenylpyrophosphatepACCAR25ΔcrtBcrtE; crtI; crtY;geranylgeranylwhitecrtX;pyrophosphatecrtZpACCAR25ΔcrtXcrtE; crtB;phytoenewhiteΔcrtIcrtY; crtZpACCRT-EIBcrtE; crtB; crtIlycopeneredpACCAR16ΔcrtXcrtE; crtB; crtIβ-caroteneyellowcrtYpACCAR25ΔcrtXcrtE; crtB; crtI;zeaxanthinyellow/crtY;orangecrtZGenes encoding: crtE, geranylgeranyl pyrophosphate synthase; crtB, Phytoene synthase; crtI, phytoene desaturase; crtY, lycopene cyclase; crtX, β-carotene hydroxylase; crtZ, zeaxanthin glycosylase

EXAMPLE 9

Construction of cDNA Library of Phaffia rhodozyma

[0220] a) Isolation of Total RNA from Phaffia rhodozyma

[0221] All solutions were made in DEPC-treated distilled water and all equipments were soaked overnight in 0.1% DEPC and then autoclaved.

[0222] A 300 ml Erlemeyer containing 60 ml YePD culture medium was inoculated with Phaffia rhodozyma strain CBS69381/1043-3 from a preculture to a final OD600 of 0.1. This culture was incubated at 21° C. (300 rpm) until the OD600 had reached 3-4.

[0223] The cells were harvest by centrifugation (4° C., 8000 rpm, 5 min) and were resuspended in 12 ml of ice-cold extraction-buffer (0.1 M Tris-HCl, pH 7.5; 0.1 M LiCl; 0.1 mM EDTA). After centrifugation cells were resuspended in 2 ml of ice-cold extraction-buffer, 4 g of glassbeads (0.25 mm) and 2 ml phenol were added.

[0224] The mixture was vortexed 5 times at maximum speed for 30 s with 30 s cooling incubation intervals on ice.

[0225] The cell/glassbeads/phenol mixture was centrifuged (5 min, 15.300 rpm, 4° C.) and the aqueous phase (sup 1) was transferred to a fresh tube and was kept on ice.

[0226] The phenolic phase was retracted by adding an additional volume of 1 ml extraction buffer and 2 ml phenol.

[0227] After centrifugation (5 min, 15.300 rpm, 4° C.). the aquaous phase was transferred to sup 1 and extracted with an equal volume phenol:chloroform.

[0228] After centrifugation (5 min, 15.300 rpm, 4° C.), the aquaous phase was transferred to a fresh tube and 0.1 volume of 3 M NaAc; pH 5.5 and 2.5 volumes of EtOH was added to precipitate RNA (incubation overnight −20° C.).

[0229] The precipitate was collected by centrifugation (10 min, 15.300 rpm, 4° C.) and drained off excess liquid and the RNA pellet was washed with 70% icecold EtOH.

[0230] After removing excess liquid the RNA was resuspended in 200-800 μl DEPC-treated water. RNA was stored at −70° C. A 60 ml culture yielded 400-1500 μg total RNA. The integrity of total RNA was checked by formaldehyde RNA gel electrophoresis.

[0231] b) Selection of poly(A)+ RNA

[0232] Isolation of poly(A)+ from total RNA was carried out essential as described by Sambrook et al., 1989 (Molecular cloning, a laboratory manual, second edition) using the following solutions. All solutions were prepared in DEPC-treated water and autoclaved.

7RNA denaturation buffer:1 M NaCl; 18% (v/v) DMSO.Column-loading buffer (HEND):10 mM Hepes, pH 7.6; 1 mMEDTA; 0.5 M Na Cl; 9% (v/v) DMSO.Elution buffer (HE):10 mM Hepes, pH 7.6; 1 mM EDTA.

[0233] Oligo(dT)-cellulose Type 7 was supplied by Pharmacia Biotech. 0.1 g (dry weight) of oligo(dT)-cellulose was add to 1 ml HEND and the suspension was gently shaked for 1 h at 4° C. Total RNA (1.5 mg dissolved in 500 μl) and 1 ml 1 M NaCl; 18% (v/v) DMSO was heated to 65° C. for 5 min. Then 600 μl NaCl/DMSO was added to the RNA, mixed and placed on ice for 5 min. The poly(A)+ isolation was carried out be two cycles of purification. The final yield was about 45 μg poly(A)+ RNA.

[0234] c) cDNA Synthesis

[0235] cDNAs were synthesized from 7.5 μg poly(A)+-RNAs using the cDNA Synthesis Kit (#200401; StrategeneR). Synthesis was carried out according to the instruction manual with some minor modification.

[0236] SuperScript™ II RNase H− Reverse Transcriptase (Gibco BRL) was used in the first strand reaction instead of MMLV-RT.

[0237] The following reagents were add in a microcentrifuge:

[0238] 3 μl of poly(A)+ RNAs

[0239] 2 μl of linker-primer

[0240] 23.5 μl DMQ

[0241] Incubate 10 min 70° C., spin quickly in microcentrifuge and add,

[0242] 10 μl of 5× First Strand Buffer (provided by Gibco BRL)

[0243] 5 μl of 0.1 M DTT (provided by Gibco BRL)

[0244] 3 μl of first strand methyl nucleotide mixture

[0245] 1 μl of RNase Block Ribonuclease Inhibitor (40 U/μl)

[0246] Annealling of template and primers by incubation the mixture at 25° C. for 10 min followed by 2 min at 42° C. and finally add;

[0247] 2.5 μl SuperScript™ II RNase H− Reverse Transcriptase

[0248] First-strand reaction was carried out at 42° C. for 1 h.

[0249] Size fractionation was carried out using GenecleanR II kit (supplied BIO 101, Inc.). The volume of the cDNA mixture obtained after XhoI digestion was brought up by adding DMQ to a final volume of 200 μl. Three volumes of NaI was added and the microcentrifuge tube was placed on ice for 5 min. The pellet of glassmilk was washed three times using 500 μl New Wash. Finally the cDNA was eluted in 20 μl DMQ.

[0250] The yield of cDNA was about 1 μg using these conditions.

[0251] d) cDNA Cloning

[0252] cDNA library was constructed in the Uni-ZAP™ XR vector using 100 ng cDNAs. Ligation was performed two times overnight incubation at 12° C. The cDNA library was packaged using the PackageneR lambda DNA packaging system (Promega) according to the instruction manual. The calculated titer of the cDNA library was 3.5 106 pfu.

[0253] e) Mass Excission

[0254] Mass excision was carried out described in the protocol using derivatives of E. coli XL-Blue-MRF′ as acceptor strain (see Table 5). Dilution of cell mixtures were plated onto 145 mm LB agar plates containing ampicillin, chloramphenicol and IPTG, yielding 250-7000 colonies on each plate. The plates were incubatied overnight at 37° C. and further incubated one or two more days at room temperature.

EXAMPLE 10

[0255] Cloning of the Geranylgeranyl Pyrophosphate Synthase Gene (crtE) of Phaffia rhodozyma

[0256] a) Isolation of cDNA Clone

[0257] The entire library was excised into a farnesylpyrophosphate/isopentenyl pyrophosphate accumulating cells of E. coli XL-Blue-MRF, which carries the plasmid pACCAR25ΔcrtE (further indicated as XL-Blue-MRF′[pACCAR25ΔcrtE]). The screening for the crtE gene was based on the color of the transformants. Introduction of the crtB gene in a genetic background of XL-Blue-MRF′[pACCAR25ΔcrtE] would result in a restoration of the complete route for the biosynthesis of zeaxanthin-diglucoside, which could be monitored by the presence of a yellow/orange pigmented colony. About 8.000 colonies were spread on LB agar plates containing appropriate antibiotics and IPTG. One colonie was found to have changed to a yellow/orange color.

[0258] b) Characterization of Complementing cDNA Clone

[0259] These colonies were streaked on LB-ampicillin agar plates. Plasmid DNA was isolated from this yellow colonies and found to include a 1.85 kb fragment (FIG. 2A). The resulting plasmid, designated pPRcrtE, was used for retransformation experiments, (Table 6). Only the transformation of XL-Blue-MRF′[pACCAR25ΔcrtE] with pPRcrtE resulted in a white to yellow color change in phenotype. To test whether the color change was due to complemention and not caused by cDNA alone pPRcrtE was transformed into XL-Blue-MRF′. Selection of transformants on LB-ampicillin agar plate containing IPTG did not result in color changes of the colonies (Table 6). Therefore we tentatively concluded, that we have cloned a cDNA of P. rhodozyma encoding GPPP synthase which is involved in the conversion of IPP and FPP to GGPP.

8TABLE 6Color phenotype of carotenoid producing E. colistrains transformed with pPRcrtE.pUC19 (control)pPRcrtEXL-Blue-MRF'whitewhite(Ap, IPTG)XL-Blue-MRF'whiteyellow/orange[pACCAR25ΔcrtE](Ap, Cm, IPTG)XL-Blue-MRF'whitewhite[pACCAR25ΔcrtB](Ap, Cm, IPTG)Transformation: 10 ng of each plasmid was mixed to CaCl2 competent E. coli cells. Transforment cells were selected by plating 1/10 and 1/100 volume of the DNA/cell mixture on LB agar-medium containing the appropriate antibiotics (in brackets).

[0260] c) Sequence Analysis of cDNA Fragment

[0261] Plasmid pPRcrtE was used to determine the nucieotide sequence of the 1.85 kb cDNA.

[0262] The sequence comprised 1830 nucleotides and a 31 bp poly(A) tail. An open reading frame (ORF) of 375 amino acids was predicted. The nucleotide sequence and deduced amino acid sequence are shown as SEQIDNO: NO 14 and 15, respectively. A search in SWISS-PROT protein sequence data bases using the Blitz amino acid sequence aligrunent program indicated amino acid homology (52% in 132 aa overlap; Neurospora crassa) especially to the conserved domain I in geranylgeranyl-PPi synthase enzymes of different organisms (Botella et al., Eur. J. Biochem. (1995) 233; 238-248).

EXAMPLE 11

Cloning of the Phytoene Synthase Gene (crtB) of Phaffia rhodozyma

[0263] a) Isolation of cDNA Clone

[0264] The entire library was excised into a geranylgeranylpyrophosphate accumulating cells of E. coli XL-Blue-MRF′, which carries the plasmid pACCAR25ΔcrtB (further indicated as XL-Blue-MRF′[pACCAR25ΔcrtB]). The screening for the crtB gene was based on the color of the transformants. Introduction of the crtB gene in a genetic backgrpund of XL-Blue-MRF′[pACCAR25ΔcrtB] would result in a restoration of the complete route for the biosynthesis of zeaxanthin-diglucoside, which could be monitored by the presence of a yellow/orange pigmented colony.

[0265] About 25.000 colonies were incubated on LB agar plates containing appropriate antibiotics and IPTG. Three colonies were found to have changed to a yellow/orange color.

[0266] b) Characterization of Complementing cDNA Clone

[0267] These colonies were streaked on LB-ampicillin agar plates. Plasmid DNA, designated pPRcrtB1 to 3, was isolated from these yellow colonies and found to include a 2.5 kb fragment (FIG. 2B). One of the resulting plasmids, pPRcrtB1 was used for retransformation experiments (Table 7). Only the transformation of XL-Blue-MRF′[pACCAR25AcrtB] with pPRcrtB resulted in a white to yellow color change in phenotype. Therefore we tentative conclude that we have cloned a cDNA of P. rhodozyma encoding phytoene synthase which is involved in the conversion of 2 GGPP molecules via prephytoene pyrophosphate into phytoene.

9TABLE 7Color phenotype of carotenoid producing E. colistrains transformed with pPRcrtB.pUC19 (control)pPRcrtBXL-Blue-MRF'whitewhite(Ap, IPTG)XL-Blue-MRF'whiteyellow/orange[pACCAR25ΔcrtB(Ap, Cm, IPTG)XL-Blue-MRF'whitewhite[pACCAR25ΔcrtE(Ap, Cm, IPTG)Legend: see Table 6.

[0268] c) Sequence Analysis of cDNA Fragment.

[0269] Plasmid pPRcrtB2, which contains the longest cDNA insert, was used to determine the nucleotide sequence of the 2.5 kb cDNA. The sequence comprised 2483 nucleotides and a 20 bp poly(A) tail. An open reading frame (ORF) of 684 amino acids was predicted. The nucleotide sequence and deduced amino acid sequence are shown in SEQIDNOs: 12 and 13, respectively. A search in SWISS-PROT protein sequence data bases using the Blitz amino acid sequence alignment program Data indicated some amino acid homology (26% identity in 441 aa overlap of crtB gene of Neurospora crassa) with crtB genes of other organisms.

EXAMPLE 12

Cloning of the Phytoene Desaturase Gene (crtI) of Phaffia rhodozyma

[0270] a) Isolation of cDNA Clone

[0271] The entire library was excised into a phytoene accumulating cells of E. coli XL-Blue-MRF′, which carries the plasmid pACCAR25ΔcrtXAcrtl (further indicated as XL-Blue-MRF′[pACCAR25ΔcrtXAcrtf]). The screening for the crtI gene was based on the color of the transformants. Introduction of the crtI gene in a genetic background of XL-Blue-MRF′[pACCAR25ΔcrtXΔcrtI] would result in a restoration of the complete route for the biosynthesis of zeaxanthin, which could be monitored by the presence of a yellow/orange pigmented colony.

[0272] About 14.000 colonies were incubated on LB agar plates containing appropriate antibiotics and IPTG. Two colonies were found to have changed to a yellow/orange color.

[0273] b) Characterization of Complementing cDNA Clones

[0274] These colonies were streaked on LB-ampicillin agar plates. Plasmid DNA, designated pPRcrtI.1 and pPRcrtI.2, was isolated from these yellow colonies and found to include a 2.0 kb fragment (FIG. 2C). One of the resulting plasmids, pPRcrt1.1 was used for retransformation experiments (Table 8). Only the transformation of XL-Blue-MRF′[pACCAR25ΔcrtXΔcrtl] with pPRcrtI resulted in a white to yellow color change in phenotype. Therefore we tentative conclude that we have cloned a cDNA of P. rhodozyma encoding phytoene desaturase which is involved in the conversion of phytoene to lycopene.

10TABLE 8Color phenotype of carotenoid producing E. colistrains transformed with pPRcrtI.pUC19pPRcrtIXL-Blue-MRF'whitewhite(Ap, IPTG)XL-Blue-MRF'whiteyellow/orange[pACCAR25ΔcrtXΔcrtI(Ap, Cm, IPTG)XL-Blue-MRF'whitewhite[pACCAR25ΔcrtB(Ap, Cm, IPTG)Legend: see Table 6.

[0275] c) Sequence Analysis of cDNA Fragment

[0276] One of the plasmid pPRcrtI was used to determine the nucleotide sequence of the 2.0 kb cDNA. The sequence comprised 2038 nucleotides and a 20 bp poly(A) tail. An open reading frame (ORF) of 582 amino acids was predicted. The nucleotide sequence and deduced amino acid sequence are shown in SEQIDNOs: 16 and 17, respectively. A search in SWISS-PROT protein sequence data bases using the Blitz amino acid sequence alignment program Data indicated amino acid homology to phytoene desaturase gene of N. crassa (53% identity in 529 aa overlap).

EXAMPLE 13

Cloning of the Lycopene Cyclase Gene (crtY) of Phaffia rhodozyma

[0277] a) Isolation of cDNA Clone

[0278] The entire library was excised into a lycopene accumulating cells of E. coli XL-Blue-MRF′, which carries the plasmid pACCRT-EIB (further indicated as XL-Blue-MRF′[pACCRT-EIB]). The screening for the crtY gene was based on the color of the transformants. Introduction of the crtY gene in a genetic background of XL-Blue-MRF′[pACCRT-EIB] would result in a restoration of the complete route for the biosynthesis of β-carotene, which could be monitored by the presence of a yellow pigmented colony. About 8.000 colonies were incubated on LB agar plates containing appropriate antibiotics and IPTG. One colony was found to have changed to a yellow color.

[0279] b) Characterization of Complementing cDNA Clone

[0280] This colony was streaked on LB-ampicillin agar plates. Plasmid DNA was isolated from this yellow colony and found to include a 2.5 kb fragment (FIG. 2B). The resulting plasmid, designated pPRcrtY, was used for retransformation experiments (Table 9. Surprisingly, not only transformation of XL-Blue-MRF′[pACCRT-EIB] but also transformation of XL-Blue-MRF′[pACCAR25ΔcrtB] with pPRcrtY resulted in a red to yellow color change in phenotype.

11TABLE 9Color phenotype of carotenoid producing E. colistrains transformed with pPRcrtY.pUC19pPRcrtBXL-Blue-MRF'whitewhite(Ap, IPTG)XL-Blue-MRF'redyellow[pACCRT-EIB(Ap, Cm, IPTG)XL-Blue-MRF'redyellow[pACCAR25ΔcrtB(Ap, Cm, IPTG)Legend: see Table 6.

[0281] A second transformation experiment was carried out including the previously cloned cDNA of pPRcrtB. As shown in table 6 the cDNA previously (example 3) isolated as encoding phytoene synthase was able to complement the crtY deletion resulting in the biosynthesis of β-carotene in XL-Blue-MRF′[pACCRT-EIB].

[0282] Sequence analysis of the cDNA insert of pPRcrtY (SEQIDNOs: 18 and 19) showed that it was similar to the sequence of cDNA fragment of pPRcrtB.

[0283] From these data we tentative conclude that we have cloned a cDNA of P. rhodozyma encoding phytoene synthase and lycopene cyclase which is involved in the conversion of 2 GGPP molecules via prephytoene pyrophosphate into phytoene and lycopene to β-carotene, respectively. This is the first gene in a biosynthetic pathway of carotenoids synthesis that encodes two enzymatic activities.

12TABLE 10Color phenotype of carotenoid producing E. coli strains transformedwith different cDNAs of Phaffia rhodozyma (Ap, Cm, IPTG).pUC19pPRcrtEpPRcrtBpPRcrtYXL-Blue-whiteyellow/whitewhiteMRF′[PACCAR25ΔcrtE]orangeXL-Blue-whitewhiteyellow/yellow/MRF′[pACCAR25ΔcrtB]orangeorangeXL-Blue-redredyellowyellowMRF′[pACCRT-EIB]Legend: see Table 6

EXAMPLE 14

[0284] Cloning of the Isopentenyl Diphosphate (IPP) Isomerase Fene (idi) of Phaffia rhodozyma

[0285] a) Isolation of cDNA Clone

[0286] The entire Phaffia cDNA library was excised into lycopene accumulating cells of E. coli XL-Blue-MRF′, each carrying the plasmid pACCRT-EIB (further indicated as XL-Blue-MRF′[pACCRT-EIB]).

[0287] About 15.000 colonies were incubated on LB agar plates containing appropriate antibiotics and IPTG. One colony was found to have a dark red colour phenotype.

[0288] b) Characterization of Complementing cDNA Clone

[0289] This colony was streaked on LB-ampicillin agar plates. Plasmid DNA was isolated from this yellow colony and found to include a 1.1 kb fragment. The resulting plasmid, designated pPRcrtX, was used for retransformation experiments (Table 11).

[0290] All colonies of XL-Blue-MRF′[pACCAR-EIB] transformed with pPRcrtX had a dark red phenotype. From these data we tentatively concluded, that we have cloned a cDNA of P. rhodozyma expression of which results in an increased lycopene production in a genetically engineered E. coli strain.

13TABLE 11Color phenotype of carotenoid producing E. coli strainstransformed with pPRcrtX.pUC19pPRcrtXXL-Blue-MRF′whitewhite(Ap, IPTG)XL-Blue-reddark redMRF′[pACCRT-EIB(Ap, Cm, IPTG)Legend: see Table 6.

[0291] c) Sequence Analysis of cDNA Fragment

[0292] In order to resolve the nature of this gene the complete nucleotide sequence of the cDNA insert in pPRcrtX was determined. The nucleotide sequence consist of the 1144 bp. The sequence comprised 1126 nucleotides and a poly(A) tail of 18 nucleotides. An open reading frame (ORF) of 251 aminoacids with a molecular mass of 28.7 kDa was predicted. The nucleotide sequence and deduced amino acid sequence are shown in SEQIDNOs: 20 and 21, respectively.

[0293] A search in SWISS-PROT protein sequence data bases using the Blitz amino acid sequence alignment program Data indicated aminoacid homology to isopentenyldiphosphate (IPP) isomerase (idi) of S. cerevisiae (42.2% identity in 200 aminoacid overlap). IPP isomerase catalyzes an essential activation step in the isoprene biosynthetic pathway which synthesis the 5-carbon building block of carotenoids. In analogy to yeast the gene of Phaffia was called idiI. The cDNA clone carrying the genes was then called pPRidi.

EXAMPLE 15

Overexpression of the idi Gene of P. rhodozyma in a Carotenogenic E. coli

[0294] Lycopene accumulating cells of E. coli XL-Blue-MRF′, which carry the plasmid pACCRT-EIB (further indicated as XL-Blue-MRF′[pACCRT-EIB]) were transformed with pUC19 and pPRidi and transformants were selected on solified LB-medium containing Amp and Cm. The transformants, called XL-Blue-MRF′[pACCRT-EIB/pUC19 and [pACCRT-EIB/pPRidi], were cultivated in 30 ml LB-medium containing Amp, Cm and IPTG at 37° C. at 250 rpm for 16 h. From these cultures 1 ml was used for carotenoid extraction and analysis. After centrifugation the cell pellet was dissolved in 200 μl aceton and incubated at 65 IC for 30 minutes. Fifty μl of the cell-free aceton fraction was then used for high-performance liquid chromatography (HPLC) analysis. The column (chrompack cat. 28265; packing nucleosil 100C18) was developed with water-acetonitrile-2-propanol (from 0 to 45 minutes 9:10:81 and after 45 minutes 2:18:80) at a flow rate of 0.4 ml per minute and recorded with a photodiode array detector at 470+/−20 nm. Lycopene was shown to have a retention time of about 23 minutes under these conditions. The peak area was used as the relative lycopene production (mAu*s). The relative lycopene production was 395 and 1165 for, XL-Blue-MRF′[pACCRT-EIB/pUC19] and [pACCRT-EIB/pPRidi], respectively.

[0295] These data show the potentials of metabolic pathway engineering in Phaffia, as increased expression of the idi of Phaffia rhodozyma causes a 3-fold increase in carotenoid biosynthesis in E. coli.

[0296] This cDNA may be over-expressed in a transformed Phaffia cell with a view to enhance carotenoid and/or xanthophyll levels. The cDNA is suitably cloned under the control of a promoter active in Phaffia, such as a strong promoter according to his invention, for example a Phaffia glykolytic pathway promoter, such as the GAPDH-gene promoter disclosed herein, or a Phaffia ribosomal protein gene promoter according to the invention (vide sub). Optionally, the cDNA is cloned in front of a transcriptional terminator and/or polyadenylation site according to the invention, such as the GAPDH-gene terminator/polyadenylation site. The feasibility of this approach is illustrated in the next example, where the crtB gene from Erwinia uredovora is over-expressed in Phaffia rhodozyma by way of illustration.

EXAMPLE 16

[0297] Heterologous Expression of Carotenogenic Gene from Erwinia uredovora in Phaffia rhodozyma.

[0298] The coding sequence encoding phytoene synthase (crtB) of Erwinia uredovora (Misawa et al., 1990) was cloned between the promoter and terminator sequences of the gpd (GAPDH-gene) of Phaffia by fusion PCR. In two separate PCR reactions the promoter sequence of gpd and the coding sequence of crtB were amplified. The former sequence was amplified using the primers 5177 and 5128 and pPR8 as template. This latter vector is a derivative of the Phaffia transformation vector pPR2 in which the promoter sequence has been enlarged and the BglI restriction site has been removed. The promoter sequence of gpd was amplified by PCR using the primers 5226 and 5307 and plasmid pPRgpd6 as template. The amplified promoter fragment was isolated, digested with KpnI and BamHI and cloned in the KpnI-BglII fragment of vector pPR2, yielding pPR8. The coding sequence of crtB was amplified using the primers 5131 and 5134 and pACCRT-EIB as template. In a second fusion PCR reaction, using the primers 5177 and 5134, 1 μg of the amplified promoter and crtB coding region fragment used as template yielding the fusion product Pgpd-crtB. The terminator sequence was amplified under standard PCR conditions using the primers 5137 and 5138 and the plasmid pPRgdh6 as template. Primer 5137 contains at the 5′ end the last 11 nucleotides of the coding region of the crtB gene of E. uredovora and the first 16 nucleotides of the terminator sequence of gpd gene of P. rhodozyma. By a two basepair substitution a BamHI restriction site was introduced. The amplified fusion product (Pgpd-crtB) and the amplified terminator fragments were purified and digested with HindIII and BamHI and cloned in the dephosphorylated HindIII site of the cloning vector pMTL25. The vector with the construct Pgpd-crtB-Tgpd was named pPREXI.1.

[0299] The HindIII fragment containing the expression cassette Pgpd-crtB-Tgpd was isolated from pPREX1.1 and ligated in the dephosphorylated HindIII site of the Phaffia transformation vector pPR8. After transformation of the ligation mixture into E. coli a vector (pPR8crtB6.1) with the correct insert was chosen for Phaffia transformation experiments.

[0300] Phaffia strain CBS6938 was transformed with pPR8crtB6.1, carrying the expression cassette Pgpd-crtB-Tgpd, and transformants were selected on plates containing G418. The relative amount of astaxanthin per OD660 in three G418-resistant transformants and the wild-type Phaffia strains was determined by HPLC analysis (Table 12). For carotenoid isolation from Phaffia the method of DMSO/hexane extraction described by Sedmak et al., (1990; Biotechn. Techniq. 4, 107-112) was used.

14TABLE 12The relative astaxanthin production in a Phaffiatransformant carrying the crtB gene of E. uredovo-raRelative amountof astaxanthinStrain:(mAU*s/OD660)P. rhodozyma CBS6938448P. rhodozyma CBS6938[pPR8crtB6.1]#1626[pPR8crtB6.1]#2716[pPR8crtB6.1]#4726Primers used:5128:5′caactgcaatgatggtaagagtgttagag 3′5177:5′cccaagctttctcgaggtacctggtgggtgcatgtatgtac3′5131:5′taccatcatggcagttggfctcgaaaag3′5134:5′cccaagcttggatccgtctagagcgggcgctgcc3′5137:5′ccaaggcctaaacggatccctccaaacc 3′5138:5′gccaagcttctcgagcttgatcagataaagatagagat3′5307:5′gttgaagaagggatccttgtggatga 3′The gpd sequences are indicated in bold, the crtB sequences in italic, additional restriction sites for cloning are underlined and base substitution are indicated by double underlining.

EXAMPLE 17

Isolation and Characterization of the crtB Gene of Phaffia

[0301] It will also be possible to express the Phaffa rhodozyma gene corresponding to crtB and express it under the control of its own regulatory regions, or under the control of a promoter of a highly expressed gene according ot the invention. The Phaffia transformation procedure disclosed herein, invariably leads to stably integrated high copy numbers of the introduced DNA, and it is expected, that expression of the gene under the control of its own promoter will also lead to enhanced production of carotenoids, including astaxanthin. To illustrate the principle, a protocol is given for the cloning of the crtB genomic sequence, below.

[0302] To obtain the genomic crtB-gene including expression signals the 2.5 kb BamHI-XhoI fragment was isolated from the vector pPRcrtB and used as probe to screen a cosmid library of Phaffia.

[0303] The construction and screening of the library was carried out as described in Example 3 using the crtB gene as probe instead of the gapdh-gene.

[0304] After the rounds of hybridization, 2 colonies were identified giving a strong hybridization signal on the autoradiogram after exposure. Cosmid DNA isolated from these colonies was called pPRgcrtB#1.1 and pPRgcrtB#7, respectively.

[0305] Chromosomal DNA isolated from Phaffia rhodozyma strain CBS 6938 and cosmid pPRgcrtB#7 was digested with several restriction enzymes. The DNA fragments were separated, blotted and hybridized with a amino-terminal specific probe (0.45 kb XbaI fragment) of crtB under conditions as described before. After exposure, the autoradiogram showed DNA fragments of different length digested by different restriction enzymes which hybridized with the crtB probe. On the basis that no EcoRI site is present in the cDNA clone a EcoRI fragment of about 4.5 kb was chosen for subcloning experiments in order to determine the sequence in the promoter region and to establish the presence of intron sequences in the crtB gene. A similar sized hybridizing fragment was also found in the chromosomal DNA digested with EcoRI. The fragment was isolated from an agarose gel and ligated into the corresponding site of pUC19. The ligation mixture was transformed to competent E. coli cells. Plasmids with the correct insert in both orientations, named pPR10.1 and pPR10.2, were isolated from the transformants. Comparison of the restriction patterns of pPR10.1/pPR10.2 and pPRcrtB digested with XbaI gave an indication for the presence of one or more introns as the internal 2.0 kb XbaI fragment in the cDNA clone was found to be larger in the former vectors. The subclone pPR10.1 was used for sequence analysis of the promoter region and the structural gene by the so-called primer walking approach. The partial sequence of the insert in show in SEQIDNO: 22. Comparison of the cDNA and the genomic sequence revealed the presence of 4 introns.

EXAMPLE 18

Isolation of Promoter Sequences with High Expression Levels

[0306] This example illustrates the the feasibility of the “cDNA sequencing method” referred to in the detailed description, in order to obtain transcription promoters from highly expressed genes.

[0307] For the isolation and identification of transcription promoter sequences from Phaffia rhodozyma genes exhibiting high expression levels, the cDNA library of Phaffia rhodozyma was analyzed by the following procedure.

[0308] The cDNA library was plated on solified LB-medium containing Amp and 96 colonies were randomly picked for plasmid isolation. The purified plasmid was digested with XhoI and XbaI and loaded on a agarose gel. The size of the cDNA inserts varied from 0.5 to 3.0 kb. Subsequently, these plasmids were used as template for a single sequence reaction using the T3 primer. For 17 cDNA clones no sequence data were obtained. The sequences obtained were translated in all three reading frames. For each cDNA sequence the longest deduced amino acid sequences were compared with the SwissProt protein database at EBI using the Blitz program. For 18 deduced amino acid sequences no homology to known proteins was found whereas six amino acid sequences showed significant homology to hypothetical proteins. Fifty-five amino acid sequences were found to have significant homology to proteins for which the function is known. About 50% (38/79) were found to encode ribosomal proteins of which 12 full-length sequences were obtained.

15TABLE 13Overview of expressed cDNAs, encoded proteins andreference to the Sequence ListingcDNAcoding forSEQ ID NO:10ubiquitin-40S2411Glu-repr.gene261840S rib.prot S27283560S rib.prot P1α303860S rib.prot L37e324660S rib.prot L27a346460S rib.prot L25366860S rib.prot P2387340S rib.prot S17A/B407640S rib.prot S31427840s rib.prot S10448560S rib.prot L37A468760S rib.prot L34489560S rib.prot S1650

[0309] By sequence homology it was concluded that in Phaffia the 40S ribisomal protein S37 is fused to ubiquitin as is found in other organisms as well. The nucleotide sequences and deduced amino acid sequences of the full length cDNA clones are listed in the sequence listing. Six ribosomal proteins were represented in the random pool by more than one individual cDNA clone. The 40S ribosomal proteins S10 (SEQIDNO:44), S37 (+ubiquitin) (SEQIDNO:24) and S27 (SEQIDNO:28) were represented twice and 60S (acidic) ribosomal proteins P2 (SEQIDNO:38), L37 (SEQIDNO:46) and L25 (SEQIDNO:36) found three times. From these results we conclude, that these proteins are encoded by multiple genes or that these genes are highly expressed. Therefore isolation of these promoter sequences are new and promissing target sequences to isolate high level expression signals from Phaffia rhodozyma. Furthermore, a cDNA clone was isolated which showed 50% homology to an abundant glucose-repressible gene from Neurospora crassa (Curr. genet. 14: 545-551 (1988)). The nucleotide sequence and the deduced amino acid sequence is shown in SEQIDNO:26. One of the advantages of such a promoter sequence is that it can be used to separated growth (biomass accumulation) and gene expression (product accumulation) in large scale Phaffia fermentation.

[0310] For the isolation of the promoter sequences of interest (as outlined above) a fragment from the corresponding cDNA clone can be used as probe to screen the genomic library of Phaffia rhodozyma following the approach as described for the GAPDH-gene promoter (Example 3, supra). Based on the determined nucleotide sequence of the promoter, specific oligonucleotides can be designed to construct a transcription fusion between the promoter and any gene of interest by the fusion PCR technique, following the procedure as outlined in Example 5 (supra).

EXAMPLE 19

Isolation of Carotenogenic Genes by Heterologous Hybridization

[0311] For the identification and isolation of corresponding carotenoid biosynthetic pathway genes from organisms related to Phaffia rhodozyma heterologous hybridization experiments were carried out under conditions of moderate stringency. In these experiments chromosomal DNA from two carotenogenic fungi (Neurospora crassa and Blakeslea trispora) and the yeasts S. cerevisiae and three yeast and fungal species from the genus Cystofylobasidium was used. These three carotenogenic yeasts are, based on phylogenetic studies, the ones most related to P. rhodozyma.

[0312] Chromosomal DNA from the yeast species Cystofylobasidium infirmo-miniatum (CBS 323), C. bisporidii (CBS 6346) and C. capitatum (CBS 6358) was isolated according the method as developed for Phaffia rhodozyma, described in example 3 of European patent application 0 590 707 A1; the relevant portions of which herein incorporated by reference. Isolation of chromosomal DNA from the fungi Neurospora crassa and Blakeslea trispora was essentially carried as described by Kolar et al. (Gene, 62: 127-134), the relevant parts of which are herein incorporated by reference.

[0313] Chromosomal DNA (5 μg) of C. infirmo-miniatum, C. bisporidii, C. capitatum, S. cerevisiae, P. rhodozyma, N. crassa and B. trispora was digested using EcoRI. The DNA fragments were separated on a 0.8% agarose gel, blotted and hybridized using the following conditions.

[0314] Hybridization was carried out at two temperatures (50° C. and 55° C.) using four different 32P labelled Phaffia probes. The probes were made using random primed hexanucleotide labellings reactions using the XhoI-XbaI fragment(s) from the cDNA clones pPRcrtE, pPRcrtB, pPRcrtl and pPRidi as template. Hybridization was carried out o/n (16 h) at the indicated temperatures. After hybridization the filters were washed 2 times for 30 min. at the hybridization temperatures using a solution of 3*SSC; 0.1% SDS; 0.05% sodiumpyrophosphate. Films were developed after exposure of the filters to X-ray films in a cassette at −80° C. for 20 h.

[0315] Using the cDNA clone of crtE of P. rhodozyma faint signals were obtained for C. infirmo-miniatum, C. capitatum. Using the cDNA clone of crtB of P. rhodozyma strong signals were obtained to the high molecular weight portion of DNA from C. infirmo-miniatum and C. capitatum. Furthermore a strong signal was obtained in the lane loaded with digested chromosomal DNA from B. trispora. Only a faint signal was obtained for C. capitatum at 50° C. using the cDNA clone of crtl of P. rhodozyma. Using the cDNA clone of idi of P. rhodozyma faint signals were obtained with chromosomal DNA from C. infirmo-miniatum, C. bisporidii and C. capitatum at both temperatures. A strong signal was obtained in the lane loaded with digested chromosomal DNA from B. trispora.

[0316] We conclude, that carotenoid biosynthesis cDNAs or genes, or idi cDNAs or genes, can be isolated from other organisms, in particular from other yeast species by cross-hybridisation with the cDNA fragments coding for P. rhodozyma carotenoid biosynthesis enzymes, or isopentenyl pyrophosphate isomerase coding sequences respectively, using moderately stringent hybridisation and washing conditions (50° C. to 55° C., 3×SSC).

[0317] Deposited Microorganisms

[0318]

E. coli
containing pGB-Ph9 has been deposited at the Centraal Bureau voor Schimmelcultures, Oosterstraat 1, Baarn, The Netherlands, on Jun. 23, 1993, under accession number CBS 359.3.

[0319] The following strains have been deposited under the Budapest Treaty at the Centraal Bureau voor Schimmelcultures, Oosterstraat 1, Baam, The Netherlands, on Feb. 26, 1996:

16ID nr.Organismrelevant featureDeposit numberDS31855E. colicrtY of P. rhodozymaCBS 232.96DS31856E. colicrtI of P. rhodozymaCBS 233.96DS31857E. colicrtE of P. rhodozymaCBS 234.96DS31858E. colicrtB of P. rhodozymaCBS 235.96

[0320]

Number	Date	Country	Kind
95203620.0	Dec 1995	EP
96200943.7	Apr 1996	EP

	Number	Date	Country
Parent	09091725	Nov 1998	US
Child	10001192	Oct 2001	US

Recombinant materials for carotenoid production

Information

Publication Number

Date Filed

Date Published

Inventors

CPC

US Classifications

International Classifications

Abstract

Description

Claims

Priority Claims (2)

Continuations (1)